IV7 CORNELL UNIVERSITY. THE THE GIFT OF ROSWELL P. FLOWER FOR THE USE OF THE N. Y. STATE VETERINARY COLLEQB. 1897 p f^ ■V -^ PATHOGENIC MICROORGANISMS A PEACTICAL MANUAL FOR STUDENTS, PHYSICIANS AND HEALTH OFFICEES BY WILLIAM HALLOCK PARK, M.D. PBOFESSOR OF BACTEKIOLOGY AND HYGIENE, UNIVERSITY AND BELLEVUE HOSPITAL MEDICAL COLLEGE AND DIRECTOR OF THE BUREAU OF LABORATORIES OF THE DEPARTMENT OF HEALTH, NEW YORK CITY AND ANNA WESSELS WILLIAMS, M.D. ASSISTANT DIRECTOR OF THE BUREAU OF LABORATORIES OF THE DEPARTMENT OF HEALTH; CONSULTING PATHOLOGIST TO THE NEW YORK INFIRMARY FOR WOMEN AND CHILDREN Assisted by CHARLES KRUMWIEDE, Jr., M.D. ASSISTANT DIRECTOR OP THE BUREAU OF LABORATORIES; ASSISTANT PROFESSOR OF BACTERIOLOGY AND HYGIENE IN THE UNIVERSITY AND BELLEVUE HOSPITAL MEDICAL COLLEGE, NEW YORK CITY SIXTH EDITION, ENLARGED AND THOROUGHLY REVISED WITH 209 ENGRAVINGS AND 9 FULL-PAGE PLATES LEA & FEBIGEI^ NEW YORK AND PHILADELPHIA 1917 Copyright LEA & FEBIGER 1917 PREFACE TO THE SIXTH EDITION. The first edition of this book was called Bacteriology in Medicine and Surgery. It was written to make available for others the practical knowledge which had been acquired in the work of the bacteriological laboratories of the city of New York, and was intended more for medical practitioners than for medical students or laboratory workers. When the second edition had been exhausted the improvement in methods of cultivating and studying the protozoa had reached a point rendering it advantageous to include the animal as well as the vegetable germs. This was done and the title of the third edition was altered to conform with the text, which had been broadened to give in outline practically the whole field of pathogenic microorganisms. In the fifth edition the material was rearranged in order to bring more closely together all of the pathogenic organisms. Under this arrange- ment Part I deals with the general characteristics and methods of study of all the microorganisms considered (molds, yeasts, bacteria, and protozoa). Part II includes the study of the individual pathogenic microorganisms and their near relatives. Part III presents certain practical aspects of the subject under the title Applied Microbiology. In the present edition the practical application of serums and vaccines has been transferred also to Part III. The nine plates included in this edition have been arranged and drawn especially for this book. Owing to recent advances in the knowledge of microbiology we have added much new material and rewritten several parts of the book. The whole subject of immunity has been extensively revised. A number of recent media and stains have been added in their respective sections. The chapter on Intestinal Bacteria has been rewritten. The experience gained in the war with preventive measures against typhoid, paratyphoid, tetanus, and wound infections has been added. All of the anaerobes are brought together in one chapter and revised. The use of tuberculins in diagnosis and treatment has been revised by Dr. G. B. White. The chapter on Glanders was again revised by Dr. B. Van H. Anthony. The chapter on Complement-fixation was revised by M. A. Wilson, and that on Disinfection by W. C. Noble. The section on Ferments and Antiferments was written by Dr. R. L. Kahn. The IV PREFACE bibliography has been revised, with special reference to the needs of the student rather than to the complete recognition of contributions of investigators. The index and much of the proof-reading was done by C. Van Winkle. We are further indebted to our associates in the laboratory for aid in many different ways. Dr. Krumwiede's name has been added to the title page because of the valuable assistance he has given us. W. H. P. A. W. W. New York, 1917. ■ CONTENTS. PART I. PRINCIPLES OF MICROBIOLOGY. CHAPTER I. Introductory Historical Sketch 17 CHAPTER II. Classification and General Characteristics op Microorganisms . . 24 CHAPTER III. The Microscope and the Microscopic Examination op Microorganisms 66 CHAPTER IV. General Methods Used in the Cultivation op Microorganisms . . 87 CHAPTER V. The Use op Animals por Diagnostic and Test Purposes 127 CHAPTER VI. The Procuring and Handling of Material for Microbiological Examination prom Those Suffering prom Disease 130 CHAPTER VII. The Relation op Microorganisms to Disease 137 CHAPTER VIII. The Nature of the Substances Causing Disease: Microbal Poisons 151 yi CONTENTS CflAPTER IX. The Resistance of the Host to Mickobal Infection 157' CHAPTER X. Nature of the Protective Defences of the Body and Theib Manner OF Action. Phagocytosis. Ehblich's "Side-chain" and Other Theories ■ 164 CHAPTER XI. Complement-fixation: The Technic of the Test and its Practical Applications 186 CHAPTER XII. Agglutination and Precipitation of Microorganisms and Their Pro- teins ■ 202 CHAPTER XIII. Opsonins. Opsonic Index. Leukocyte Extract 217 CHAPTER XIV. Protein Hypersensitiveness or Anaphylaxis 227 PART II. PATHOGENIC MICROORGANISMS INDIVIDUALLY CONSIDERED. CHAPTER XV. The Pathogenic Molds (Hyphomycetes, Eumycetes) and Yeasts (Blastomycetes) „„q CHAPTER XVI. The Pyogenic Cocci 245 CONTENTS vii CHAPTER XVII. ii The Diplococcus of Pneumonia (Pnetjmococcus, Streptococcus Pneu- monia, Micrococcus Lancbolatus) 266 CHAPTER XVIII. Meningococcus or Micrococcus (Intracbllularis) Meningitidis, and '■ THE Relation of it and of Other Bacteria to Meningitis . . 277 CHAPTER XIX. The Gonococcus or Micrococcus GoNORRHEiE. Micrococcus Melitbnsis 284 CHAPTER XX. The Bacillus and the Bacteriology of Diphtheria 292 CHAPTER XXI. Intestinal Bacteria 316 CHAPTER XXII. The Colon-typhoid Group of Bacilli 323 CHAPTER XXIII. The Typhoid Bacillus 334 CHAPTER XXIV. Paratyphoid Group 351 CHAPTER XXV. Dysentery Group 356 CHAPTER XXVI. Bacillus Pyocyaneus (Bacillus of Green and of Blue Pus)., Bacillus Proteus (Vulgabis) 361 vni CONTENTS CHAPTER XXVII. The Bacillus and the Bacteriology of Tuberculosis 366 CHAPTER XXVIII. Other Acid-fast Bacilli: Bacillus of Leprosy, Bacillus of Rat Leprosy, Bacillus of Johne's Disease in Cattle, and the Group of Non-pathogenic Acid-fast Bacilli 394 CHAPTER XXIX. Glanders Bacillus (Bacillus Mallei). Bacillus Abortus (Bang) . . 399 CHAPTER XXX. The Group of Hemoglobinophilic Bacilli. Bordet-Gengou Bacillus 408 CHAPTER XXXI. Microorganisms Belonging to the Hemorrhagic Septicemia Group . 421 CHAPTER XXXII. The Anthrax Bacillus 426 CHAPTER XXXIII. Anaerobic Bacilli ^^^ CHAPTER XXXIV. The Cholera Spirillum (Cholera Vibrio) and Similar Varieties . . 455 CHAPTER XXXV. PATkoGi3NW3 Microorganisms Belonging to the Higher Bacteria (Trichomycetbs) . ,„„ 466 CHAPTER XXXVI. FiLTRABLB VIRUSES. DISEASES OF UNKNOWN EtIOLOGY .... 473 CONTENTS ix CHAPTER XXXVII. Flagellata 4g7 CHAPTER XXXVIII. Trypanosoma 494 CHAPTER XXXIX. Spirocheta and Allies 504 9 CHAPTER XL. BODO. POLYMASTIGIDA 518 CHAPTER XLI. Ameba 522 a CHAPTER XLII. Spokozoa. Ciliata 533 CHAPTER XLIII. The Malarial Organisms. Babesia 537 CHAPTER XLIV. - Smallpox (Variola) and Allied Diseases 553 &i CHAPTER XLV. Rabies. Yellow Fever 560 X CONTENTS PART III. APPLIED MICROBIOLOGY. CHAPTER XLVI. The Practical Application of Bacterial Vaccines 585 CHAPTER XLVII. The Practical Application of Serum Therapy 597 CHAPTER XLVIII. The Bacteriological Examination of Water, Air, and Soil. The Con- tamination and Purification of Water. The Disposal of Sewage 612 CHAPTER XLIX. The Bacteriology of Milk in its Relation to Disease 623 CHAPTER L. The Bacteriological Examination of Shellfish . 643 CHAPTER LI. The Soil Bacteria and their Functions. Sewage Bacteria. Bacteria IN Industries 648 CHAPTER LII. The Destruction of Bacteria by Chemicals. Practical Use of Dis- infectants 656 CHAPTER LIII. Practical Disinfection and Sterilization (House, Person, Instru- ments, AND Food). Sterilization op Milk for Feeding Infants . 670 PATHOGENIC MICE00RGANI8MS. PART I. PUmCIPLES OF MICROBIOLOGY. CHAPTER I. INTRODUCTORY HISTORICAL SICETCH. Although most of the more important discoveries in microbiology which place it on the footing of a science are of comparatively recent date, the foundations of its study were laid over two centuries ago. From that time the history of microorganisms has been intimately associated with that of medicine. Indeed, it is only through the inves- tigations into the life history of these minute forms that our present knowledge of the etiology, course, and prevention of the infectious diseases has been acquired. The prominent position which the study of microorganisms abeady holds toward medicine is, moreover, daily increasing in importance. Original discoveries are constantly adding to our knowledge of germ diseases, and the outlook is favorable for eventually obtaining, through serums, through attenuated cultures, or through the toxic substances produced by microorganisms themselves, means for immunizing against, if not of curing, an increased number of the specific infections. Even at present, bacterial products and pro- tective serums are used successfully as preventive or curative agents in several of the most prevalent infectious diseases. Om- knowledge concerning other microorganisms has enabled us largely to limit their dissemination and so to prevent disease. An acquaintance, therefore, with the main facts concerning these microorganisms is most necessary to the education of the modern physician. The vast majority of the known microorganisms which cause disease belong to the closely related groups of lowest plants and animals, i. e., the bacteria, the molds, the yeasts, and the protozoa. A few of 1 The correct form of this word is under discussion. According to derivation — microbe + logia — the better spelling would be microbology, but the form given above is the one accepted at present in the dictionaries. 2 18 INTRODUCTORY HISTORICAL SKETCH the pathogenic metazoa (some of the parasitic worms) in some of their stages are also microscopic; therefore microscopic methods of study are also appHcable to them; but since they are fully presented in works on clinical microscopy they are not given here. Before entering into a detailed consideration of the subject it may be interesting and instructive to review very briefly a few of the important steps which led to the development of the science, and upon which its foundation rest's, in which we shall see that the results obtained were gained only through long and laborious research a,nd after many obstacles were met and overcome by accurate observation and experiment. Probably the first authentic observations of living microscopic organisms of which there is any record are those of Kircher, in 1659. This original investigator demonstrated the presence in putrid meat, milk, vinegar, cheese, etc., of "minute living worms," but did not describe their form or character. Not long after this, in 1675, Leeuwenhoeck observed in rain-water, putrid infusions, and in his own and other saliva and diarrheal evacuations living, motile "animalcula" of most minute dimensions, which he described and illustrated by drawings. Leeuwenhoeck prac- tised the art of lens grinding, in which he eventually became so pro- ficient that he perfected a lens superior to any magnifying glass obtainable at that day, and with which he was enabled to see objects very much smaller than had ever been seen before. "With the greatest astonishment," he writes, "I observed distributed everywhere through the material which I was examining animalcules of the most microscopic size, which moved themselves about very energetically." The work of this observer is conspicuous for its purely objective char- acter and absence of speculation; and his descriptions and illustrations are done with remarkable clearness and accuracy, considering the imperfect optical instruments at his command. It was not until many years later, however, that any attempt was made to define the char- acters of these minute organisms and to classify them systematically. At that time all of the microscopic organisms seen were classed together as little animals. Indeed, all of the microorganisms first described at any length were probably protozoa, and only after further improvement of lenses and a more minute study of the organisms were bacterial forms gradually recognized as a separate class. The same skepticism that is seen in the acceptance of most new discoveries was displayed by doubters of the truth of these early reports of microscopic findings. Chief among the skeptics must be placed Linnaeus, who in the first edition of his Systema Natura (1735) absolutely denies the existence of Leeuwenhoeck's animalcula, though in the later editions he grudgingly admits them under the significant generic name of Chaos (Chaos proteus (Ameba), etc.). The first ideas of the structure of the protozoa were drawn from analogy. The early observers thought that each tiny organism possessed an internal structure made up of organs and tissues similar INTRODUCTORY HISTORICAL SKETCH 19 to those in metazoa. They could not conceive of motion without articulation, tendons, and muscles; nor of food absorption without an alimentary tract, and they were so impressed with the ideas of what they thought they ought to see that they were convinced that they really saw many of the complicated structures possessed by metazoa. For exarnple, the contractile vacuole, a characteristic pulsating vesicle of the protozoa, discovered by Joblot in 1754, was thought by many to be lungs, other vacuoles were said to be stomachs, the mouths were often seen and the rest of the alimentary tract was supplied from the imagination; the red pigment spots of many forms were interpreted as true eyes, etc. There were many opponents to these views, how- ever, and the idea of the cell being the unit of structure, which was advanced by Schleiden in 1838, helped determine the fact that protozoa were single cells with no definite organic structure. With the publications of Dujardin (1835-41) a correct idea of the structiural simplicity of the microorganisms gained ground. But for some time after, the controversy regarding the simple nature of pro- tozoa was strenuously carried on. It is a most instructive bit of history in research work, showing how the lack of minute observation, the exercise of a too vivid imagination, and the close reasoning from analogy may lead one astray, while the proper use of these functions may bring out the truth. KoUiker, Biitschli, Engelmann, and Hertwig, with many others (1870-80) finally demonstrated fully the unicellular nature of the protozoa. The most important characteristic of a protozoon, its life history, was first partially made out by Trembley in 1744-47. Biitschli helped determine the sexual activities of the members of this group, while Maupas (1889) was the first to demonstrate the conditions leading to their conjugation. From the earliest investigations into the life history and properties of germs, microorganisms have been thought to play an important part in the causation of infectious diseases. Shortly after the first investigations into this subject the opinion was advanced that puer- peral fever, measles, smallpox, typhus, pleurisy, epilepsy, gout, and many other diseases were due to contagion. In fact, so wide-spread became the belief in a causal relation of these minute organisms to disease that it soon amounted to a veritable craze, and all forms and kinds of diseases were said to be produced in this way upon no other foundation than that these organisms had been found in the mouth and intestinal contents of men and animals and in water. Among those who were specially conspicuous at this time for their advanced views on the germ theory of infectious diseases was Marcus Antonius Plenciz, a physician of Vienna. This acute observer, who published his views in 1762, maintained that not only were all infec- tious diseases caused by microorganisms, but that the infective mate- rial could be nothing else than a living organism. On these grounds he endeavored to explain the variations in the period of incubation of 20 INTRODUCTORY HISTORICAL SKETCH the different infectious diseases. He also insisted that there were special germs for each infectious disease by which the specific disease was produced. Plenciz believed, moreover, that these organisms were capable of multiplication in the body, and suggested the possibility of their being conveyed from place to place through the air. These views, it is true, were largely speculative, and rested upon insufficient experiment, but they were so plausible, and the arguments put forward in their support were so logical and convincing, that they continued to gain ground, in spite of considerable opposition and ridicule, and in many instances the conclusions reached have since been proved to be correct. The mode of infection, its unlimited devel- opment among large numbers of .individuals, and gradual spread over wide areas— the incubation, course of, and resulting immunity _ in recovery from infectious diseases — all pointed to a living organism as the probable cause. Among other distinguished men of the day whose observations exerted a most powerful influence upon the doctrine of infection, may be mentioned Henle. His writings {Pathological Investigations, 1840, and Text-book of Rational Pathology, 1853), in which he described the relation of microorganisms to infectious diseases, and defined the character and action of bacteria upon certain phases and symptoms of these affections, are remarkable for their clearness and precision. Origin of Microorganisms. — But, meanwhile, the question which most interested these investigators into the cause of infectious disease was: Whence are these microorganisms derived which were supposed to produce them? Were they the result of spontaneous generation due to vegetative changes in the substances in which the organisms were found, or were they reproduced from similar preexisting organisms — the so-called vitalistic theory? This question is intimately connected with the investigations into the origin and nature of fermentation and putrefaction. Spallanzani, in 1769, demonstrated that if putrescible infusions of organic matter were placed in hermetically sealed flasks and then boiled, the liquids were sterilized; neither were living organisms found in the solutions, nor did the infusions decompose; they remained unchanged for an indefinite period. The objection was raised to these experiments that the high tem- perature to which the liquids had been subjected so altered them that spontaneous generation could no longer take place. Spallanzani met the objection by cracking one of the flasks and allowing air to enter, when living organisms and decomposition again appeared in the boiled infusions. Another objection raised by the believers in spontaneous generation was that, in excluding the oxygen of the air by hermetically sealing the flasks, the essential condition for the development of fermentation which required free admission of this gas was interfered with. This objection was then met by Schulze, in 1836, by causing the air admitted to the boiled decomposable liquids to pass through strong sulphuric OklGIN OF MICROdRGANISMS 2l acid. Air thus robbed of its living organisms did not produce decora- position. Schwann, in 1839, obtained similar results in another way: he deprived of microorganisms the air admitted to his boiled liquids by passmg it through a tube which was heated to a temperature high enough to destroy germs. To this investigator is also due the credit of having discovered the specific cause — the yeast plant, or Saccharo- myces cerevisiw—oi alcoholic fermentation, the process by which sugar is decomposed into alcohol and carbonic acid. Again it was objected to these experiments that the heating of the air had perhaps brought about some chemical change which hindered the production of fermentation. Schroeder and von Dusch, in 1854, then showed that by a simple process of filtration, which has since proved of inestimable value in bacteriological work, the air can be mechanically freed from germs. By placing in the mouth of the flask containing the boiled solutions a loose plug of cotton, through which the air could freely circulate, it was found that all suspended micro- organisms could be excluded, and that air passed through such a filter, whether hot or cold, did not cause fermentation of boiled infusions. Similar results were obtained by Hoffmann, in 1860, and by Chevreul and Pasteiu-, in 1861, without a cotton filter, by drawing out the neck of the flask to a fine tube and turning it downward, leaving the mouth open. In this case the force of gravity prevents the suspended bacteria from ascending, as there is no current of air to carry them upward through the tube into the flask containing the boiled infusion. These facts have since been practically confirmed on a large scale in the preservation of food by the process of sterilization. Indeed, there is scarcely any biological problem which has been so satisfac- torily solved or in which such uniform results have been obtained; but all through the experiments of the earlier investigators irregu- larities were constantly appearing. Although in the large majority of cases it was found possible to keep boiled organic liquids sterile in flasks to which the oxygen of the air had free access, the question of spontaneous generation still remained unsettled, inasmuch as occa- sionally, even under the most careful precautions, decomposition did occur in such boiled liquids. This fact was explained by Pasteur, in 1860, by experiments showing that the temperature of boiling water was not sufiieient to destroy all living organisms, and that, especially in alkaline liquids, a higher temperature was required to insure sterilization. He showed, how- ever, that at a temperature of 110° to 112° C, which he obtained by boiling under a pressure of one and one-half atmospheres, all living organisms were invariably killed. Pasteur, at a later date (1865), demonstrated the fact that the organ- isms which resist boiling temperature are, in fact, reproductive bodies, which are now known as scores. In 1876 the development of spores was carefully investigated and explained by Ferdinand Cohn. He, and a little later Koch, showed 22 INTRODUCTORf HISTORICAL SKETCH that certain rod-shaped organisms possess the power of passing into a resting or spore stage, and when in this stage they are much less sus- ceptible to the injurious action of higher temperatures than in their normal vegetative condition. When and how life began no one is yet able to say. That sponta- neous generation may even be taking place now under unknown condi- tions is conceivable, but all such ideas are purely hypothetical and there is no evidence that under present conditions any of the known microorganisms have originated in any way except from a previous similar cell. Stimulated by the establishment of the fact, through Pasteur's investigations, that fermentation and putrefaction were due tQ the action of living organisms reproduced from similar preexisting forms, and that each form of fermentation was due to a special microorganism, the study of the causal relation of microorganisms to disease was taken up with renewed vigor. Reference has already been made to the opinions and hypotheses of the earlier observers as to the microbic origin of infectious diseases. The first positive grounds, however, for this doctrine, founded upon actual experiment, were the investigations into the cause of certain infectious diseases in insects and plants. Thus, Bassi, in 1837, demonstrated that a fatal infectious malady of the silkworm — pebrine — was due to a parasitic protozoon. Pasteur later devoted several years' study to an exhaustive investigation into the same subject; and in like manner Tulasse and Kiihne showed that certain specific affections in grains, in the potato, etc., were due to the invasion of parasites. Very soon after this it was demonstrated that microorganisms were probably the cause of certain infectious diseases in man and the higher animals. Davaine, a famous French physician, has the honor of hav- ing first demonstrated the causal relation of a microorganism to a specific infectious disease in man and animals. The anthrax bacillus was discovered in the blood of animals dying from this disease by Pollender in 1849 and by Davaine in 1850; but it was not until 1863 that the last-named observer demonstrated by inoculation experiments that the bacillus was the cause of anthrax. The next discoveries were those relating to wounds and the infections to which they are liable. Rindfleisch, in 1866, and Waldeyer and von Recklinghausen, in 1871, were the first to draw attention to the minute organisms occurring in the pyemic processes resulting from infected wounds, and occasionally following typhoid fever. Further investigations were made by Billroth, Fehleisen, and others, in erysipe- latous inflammations secondary to injury, who agreed that in these conditions microorganisms could almost always be detected in the lymph channels of the subcutaneous tissues. The brilliant results obtained by Lister, in 1863-1870, in the anti- septic treatment of wounds to prevent or inhibit the action of infective organisms, exerted a powerful influence on the doctrine of bacterial infection, causing it to be recognized far and wide and gradually lessen- ORIGIN OF MICROORGANISMS 23 ing the number of his opponents. Lister's methods were suggested to him by Pasteur's investigations on putrefaction. In 1877 Weigert and Ehrlich recommended the use of the aniline dyes as staining agents and thus made possible a more exact micro- scopic examination of microorganisms in cover-glass preparations. In the year 1880 Pasteur published his discovery of the bacillus of fowl cholera and his investigations upon the attenuation of the virus of anthi-ax and of fowl cholera, and upon protective inoculation against these diseases. In the meantime he showed that pure cultures might be obtained by the dilution method. Laveran, in the same year, announced the discovery of parasitic bodies, the first pathogenic protozoa described, in the blood of persons sick with malarial fever, and thus stimulated investigations upon the immensely important unicellular animal parasites. In 1881 Koch published his fundamental researches upon pathogenic bacteria. He introduced solid culture media (agar had already been used by Frau Hess) and the "plate method" for obtaining pure cultures, and showed how different organisms could be isolated, cultivated independently, and, by inoculation of pure cultures into susceptible animals, could be made, in many cases, to reproduce the specific disease of which they were the cause. To him more than to any other are due the methods which have enabled us to prove absolutely, in a broad sense, the permanence of bacterial varieties. It was in the course of this work that the Abbe system of substage condensing apparatus . was first used in bacteriology. Koch's postulates, namely, that (1) a specific organism always associated with a disease (2) when isolated in pure culture (3) and inoculated into a healthy susceptible animal always produces the disease, and (4) may be obtained again in pure culture, were long accepted as the only proof of the causal relationship of that organ- ism to the disease. Later it was learned that immunological reactions added greatly to our knowledge of the specificity of microbes in disease. In 1882 Pasteur pubhshed his first communication upon rabies. The method of treatment devised by him is still in general use. A little later came the investigations of Loffler and Roux upon the diphtheria bacillus and its toxins, and that of Kitasato upon tetanus. These researches paved the way for Behring's work on diphtheria anti- toxin, which in its turn stimulated investigation upon the whole sub- ject of immunity. The number of investigators rapidly increased as the importance of the earlier fundamental discoveries became apparent. Additions to the science of microbiology have been made from many sides and the practical application of the facts learned from these investigations is steadily increasing. The most important of these are given in the following pages. REFERENCES. Calkins: Protozoology, 1909, Lea & Febiger, New York and Philadelphia. Loffler, Feied: Vorlesuugen iiber die geschichtliohe Entwickelung der Lehre von den Bakter'ien, Leipzig, F. C. W. Vogel, 1887, with bibliography. Valleri, Radot: Life of Pasteur. CHAPTER II. CLASSIFICATION AND GENERAL CHARACTERISTICS OF MICROORGANISMS. In general the lowest forms of life are microscopic. Among these microscopic organisms are many which have in common the ability to produce disease in the higher animals and plants. Some of these disease-producing or pathogenic microorganisms are plants, others are animals, while some are difficult to classify. Only those that are of direct importance to man are considered in this work. CLASSIFICATION OF MICROORGANISMS. The classification of microorganisms is still in the transition stage. This is due chiefiy to the diflBculties encountered in studying the indi- vidual morphological characteristics of such minute bodies and in deter- mining their limits of variation. There is no one distinctive characteristic known which separates the lowest plants from the lowest animals. While the lowest micro- organisms of all — the bacteria — are usually classed as plants, their structure is so simple and their biological characteristics are so varied that their relationship to the vegetable kingdom is not clear-cut. In their possession of more or less rigid plasmolysible bodies, in the ten- dency of many to grow in filaments, and in the ability of some to use simple elements as food, they resemble plants; while in the motility of many, the non-possession by all of chlorophyl, and in the necessity of many for complex food, they resemble animals. There is a similar difficulty in definitely classifying many members of the other groups of closely related microorganisms, namely, the protozoa, the yeasts, and molds, and it has been suggested that under the name Protista a third kingdom be formed consisting of all of these lowest microorganisms. As a rule, in the more minute organisms genera are based upon morphological characteristics, and species upon biochemical, physiologi- cal, and pathogenic properties. This is due to the fact that the morphology of varieties may vary extremely under different conditions and that it may give no indication whatever of the relation of micro- organisms to disease and fermentation — the chief characteristic of importance to human beings. The properties of bacteria, for example, which are fairly constant under uniform conditions and which have been more or less used in systems of classification, are those of spore and capsule formation. CLASSIFICATION OF MICR06rGANISmS 25 motility (flagella formation), reaction to staining reagents; relation to temperature, to oxygen, and to other food material, and, finally, their relation to fermentation and disease. But any one of these properties, in many groups of microorganisms, such as the coli and the strepto- coccus groups, under certain conditions may so vary that, taking it as a basis for classification, an organisin might be dropped from the group with which it had been classified and be placed in an entirely different group. Thus, the power to produce spores or flagella may be held in abeyance for a time or may be totally lost; the relations to oxygen may be gradually altered, so that an anaerobic bacterium may grow in the presence of oxygen or the contrary may be the case; parasitic organisms may be so cultivated in the test-tube as to become saprophytic varieties, and those which when taken from their natural habitat have no power to grow in the living body may be made to acquire pathogenic properties, by causing the germs to develop in a series of animals of the same species whose resistance has been over- come by reducing their vitality through poisons or other means, or by giving enormous doses of germs at first. Permanence of Species. — It is true, however, that certain microorgan- isms retain some of their characteristics, so far as we know, for an indefinite time. While we cannot believe that the multitude of varie- ties which now exist have always existed, and we may theorize as to the part" played by the fluctuating variations just mentioned in pro- ducing permanent species, the fact observed in the few years during which bacteria have been studied is that most pathogenic species as observed in disease have remained practically unaltered. The diph- theria bacilli are the same today as when Lofiler discovered them in 1884, and the disease itself is evidently the same as history shows it to have been before the time of Christ. The same permanence of disease type is true for tuberculosis, smallpox, hydrophobia, leprosy, etc. Under practically unchanged conditions, therefore, such as exist in the bodies of men, most germs which have once become estab- lished as parasites continue to reproduce new generations which retain their peculiar (specific) characteristics. It is true that among the countless organisms developed some fail to hold the parasitic char- acteristics. These either continue as saprophytes or cease to exist. That new pathogenic disease varieties are coming into existence from time to time is, of course, a possibility, but not a certainty. Such true mutations have been from time to time reported, notably by Tenfold for the coli group, but whether or not the changed characteristics may be considered species characteristics cannot at present be decided. Our lack of more definite knowledge in regard to significance of these changes, as we have said, is the chief cause of other many unsatisfactory results from attempts at classification. At present a systematic effort is being made by the Society of American Bacteriologists to formulate rules for the adoption of a classification that will be accepted by all. Until such a universally 20 CHARACTERISTICS OF MICROORGANISMS accepted classification can be determined it seems wiser to note only such a broad, simple grouping as the one given below. Working Classification or Pathogenic Microokganisms in Outline Kingdom. Plants (fungi) Subkingdom. Molds (Hyphomycetes) Yeasts (Blastomycetes) Bacteria (Schizomycetes) Classes in which patho- genic species occur. Mycomycetes Phycomycetes Unclassified (Fungi Imperfecta) Oidia Saccharomycetes Cocci (Coccaceae) Bacilli (Bacteriac eae) SpiriUa (Spirillaoeae) Higher bacteria (Tricho- bacteria) Animals (Proto- zoa) Flagellates Flagellata (Mastogophora) Amebffi Rhizopoda (Sarcodina) Sporozoa Telosporidia, Neosporidia Ciliates Ciliata Genera in which chief patho- genic species occur. Aspergillus, Penicillium. Mucor. Microspora, Trichophyta, Sporo- tricha, Achoria. Oidium. Saccharomyces. Micrococcus, Diplococcus, Strep- tococcus, Tetracoccus. Baeaius (Bacterium). Spirillum[Spirocheta,Treponema]. Cladothrix, Nocardia (Strepto- thrix) . Cercomonas,Leptomonas,Herpet- omonas, Trypanosoma, Leish- mania, Trichomonas, Lamblia [Spirocheta, Treponema]. Entameba. Eimeria, Hemogregarinida, Pro- teosoma, Hemameba, Babesia, Rhinosporidium, Myxobolus, Nosema, Sarcocystis, etc. Balantidium. Unclassified : Ultramicroscopic organisms. Nomenclature. — It is well to call attention to the fact that in naming species, especially among the bacteria, the binomial law of nomenclature has been frequently violated. Such names as Bacillus coli communis should not be accepted; the- name Bacillus coli is sxiflScient as well as correct. GENERAL CHARACTERISTICS OF EACH GROUP OF MICROORGANISMS. Introduction. — ^The knowledge gained from the fact that each variety of microbe possible of cultivation may grow in distinctive ways upon so-called artificial culture media has been an immense aid in studying EXPLANATION OF PLATE I. Partly schematic. Rearranged and drawn by Williams from the indicated authors. Fig. 1. — ^Aspergillus glaucus. Fruiting hyphse growing from mycelium : A, conidiophore; B, sterigma; C, conidia; D, beginning perithecium; E, conidiophore bearing spores; F, perithecium containing rudiments in section ; G, ascus containing eight spores (De Bary) . Fig. 2. — PenicilUum, showing formation of conidia, A. Fig. 3. — Mucor mucedo: A, sporangium containing spores; B, spores liberated; C, chlamydospores ; D, E, F, stages in the formation of a zygospore. Fig. 4. — Oidium lactis. Fig. 5. — Sporotrichum schenki, showing formation of whorled spores on branched mycelium. Fig. 6. — Yeast from human infection in culture showing mycelium-like growth. Fig. 7. — Saccharomyces cerevisiae (Hansen): A, budding; B, spore formation. PLATE I MolM 2? the characteristics of the more minute forms, for the individual cell of most varieties is so minute that even the highest magnification we have may show httle if any morphological difference between organisms which produce distinctly different diseases, or between a pathogenic and a non-pathogenic form. There are, however, certain morphological and biological characteristics of the single cell which are pronounced, and we therefore study these before going on to the study of cultures, that is, of microorganisms in masses. Morphological descriptions should always be accompanied by a definite statement of the habitat — if from a culture, age of the growth, the medium from which it was obtained, and the temperature at which it was developed. MOLDS (HYPHOMYCETES, EUMYCETES). The molds (see Plate I) are closely related to the higher bacteria. Like them they grow in branching filaments or threads forming a mycelium, but these filaments or hyphse, as they are called, are more definite .than those of the trichobacteria, and they are multicellular; that is, the filaments are septate, in some varieties always (myco- mycetes) and in others when forming spores (phycomycetes). These two groups of Inolds have other differential characteristics, particularly in their fruiting organs. The mycomycetes are divided into two groups: (1) Ascomycetes, which form a spore sac, containing a definite number of spores, a multiple of two, usually eight (ascospores) ; (2) basidiomycetes, which form a definite spore-bearing cell called the basidium or conidiophore, from which four (usually) conidia or sterigmata protrude (mushrooms, etc.). The sexual fruiting organs are varied. Yeasts produce a definite sporangium containing a definite number of spores (two or a multiple of two), therefore they are closely related to the ascomycetes. The phycomycetes are characterized by the formation of no definite basidimn or ascus. The asexual fruiting organs are formed in three different ways: 1. Ends of hyphse swell and are shut off by septum, forming sporan- gium in which (a) swarm spores develop. These become free by rupture of sporangium wall, swim about, and then form a new plant; (b) spores form a cell wall while in sporangium. 2. Spores produced directly by constriction at end of certain hyphse. Such spores are called conidia and the hyphae conidiophores; each conidium is able to develop into a new plant. 3. Spores produced by separation of hyphse into short-walled seg- ments or oidia. When the spores are thick-walled they are called chlamydospores. Sexual reproduction occurs in one of two ways: 1. The ends of two similar hyphas become attached, then they segment from the rest of the hyphse and become surrounded by a 28 CHARACTERISTICS OF MICROORGANISMS thick wall. The partition in the centre dissolves and the protoplasm mixes. The resulting body is called a zygospore. It produces a new- plant. The members of this group are called zygomycetes. 2. The swollen ends of two dissimilar hyphse, one large (female) and one smaller (male), become attached, fuse, segment, and become surrounded by a thick wall. The body is called an oospore. The plants are called oomycetes. Fungi Imperfecti. — The molds that are not definitely classified are divided into two groups: 1. Known forms which do not fit in the above group. 2. Not fully known forms. To this group belong most of the molds causing himian disease. YEASTS (BLASTOMYCETES). These microorganisms have been for many centuries of the greatest importance in brewing and baking. They are not uncommonly pres- ent in the air and in cultures made from the throat. Certain recent experiments have shown that some varieties when injected are capable of producing tumor-like growths. Certain varieties are pathogenic for mice, and in recent years (since 1894) there have been reported a number of cases of human infection from yeasts. The position which the yeasts occupy in systematic biology has not, thus far, been accurately determined. In fact, it is still a question as to whether they constitute distinct fungi or whether they should be classed under the molds. The chief morphological characteristic of the yeasts is their frequent and marked method of reproduction by means of budding. For this reason these organisms go by the name of blastomycetes in contrast to the fission fungi, or schizomycetes, and the thread fungi, or hypho- mycetes. The fact was mentioned above that some yeasts form asco- spores and that they are therefore related to the ascomycetes. The soor fungus, which at one time grows in long threads, at another time (under certain conditions almost exclusively) multiplies by budding, is also a transition form. It is thus seen that no hard-and-fast line exists between yeasts and molds. The most important property of yeasts, though one not possessed by all to the same degree, is that of producing alcoholic fermentation. In brewing we distinguish between the yeasts that can be employed practically, "culture yeasts," and those which often act as disturbing factors, so-called "wild" yeasts. The enzyme causing this action is called zymase. The shape of most of the culture yeasts is oval or elliptic (Plate I, Fig. 7). Round or globular forms are more often met with among the wild species which usually excite only a slight degree of fermentation. They are known as "torula" forms. But sausage-shaped and thread forms are also met with. Pathogenic forms may also be round. BACTERIA 29 The individual yeast cells are strongly refractive, so that under the microscope at times they have almost the lustre of fat droplets. This is important because in examining fresh tissues the yeast cells may be hard to distinguish from fat droplets, often requiring the aid of certain reagents for their identification. The size of the individual yeast cells varies enormously, even in those of the same species or the same culture. In old colonies indi- viduals may be found no larger than cocci, 0.5m to 1m in diameter, while in other colonies, especially on the surface of a liquefied medium, giant yeast cells are found often attaining a diameter of 40m or more. In spite of these wide fluctuations, however, the various species are characterized by a fairly definite average in size and form. Each cell contains a more or less definite nucleus, which is demonstrated by the usual chromogenic stains. During the process of budding the nucleus of the cell moves toward the margin, where it divides. At this point the limiting membrane of the cell ruptures or usually a hernia-like protrusion develops which has the appearance of a button attached to the cell. The daughter-cell so formed rapidly increases in size and gradually assumes the shape and size of the mother-cell. A fact of the utmost importance for the propagation of the blasto- mycetes and continuation of the species is the formation of spores (Plate I, Fig. 7). In this also the cell nucleus takes part, dividing into several fragments, each of which becomes the centre of a new cell lying within the original cell. These new cells possess a firm mem- brane, a cell nucleus, and a little dense protoplasm. The number of spores developed in the yeast cells varies, but is constant for a given species. As a rule, one cell does not produce more than four endogenous spores, but species have been observed, e. g., ScMzosaccharomyces octosporm (Beijerinck), in which eight spores are found. Guilliermond has described conjugation in yeasts before the forma- tion of spores. The vitality of yeasts is truly enormous. Hansen as well as Lindner were able to obtain a growth from cultures twelve years old. Busse succeeded in getting a luxuriant growth from a dry potato culture seven and a half years old, which was almost as hard as bone. BACTERIA. Definition. — Bacteria may be defined as extremely minute, simple, unicellular microorganisms, which reproduce themselves with exceeding' rapidity, usually by transverse division, and grow without the aid of chlorophyl. They have no morphological nucleus, but contain nuclear material which is generally diffused throughout the cell body in the form of larger or smaller granules. Natural Habitat.— There are such wonderful differences in the con- ditions of life and nutrition which suit the different varieties, that bac- teria are found all over the known world. Wherever there is sufficient 30 CHARACTERISTICS OF MICROORGANISMS moisture, one form or another will find other conditions adequate for multiplication. Thus, we meet with bacterial life between 0° and 75° C. Some live only in the tissues of men, others in lower animals, a large number may grow in both man and lower animals, others still grow only in plants, but by far the greater number live in dead organic matter. For some free oxygen is necessary to life; for others it is a poison. Morphology (see Plates II and III).— The form and dimensions of bacterial cells at their stage of complete development must be distin- guished from those which they possess just after or just before they have divided. As a spherical cell develops preparatory to its division into two cells it becomes elongated and appears as a short oval rod; at the moment of its division, on the contrary, the transverse diameter of each of its two halves is greater than their long diameter. A short rod becomes in the same way, at the moment of its division, two cells, the long diameter of each of which may be even a trifle less than its short diameter, and thus they appear on superficial examination as spheres. Size. — ^The dimensions of the adult individual vary greatly in the different species as well as in members of the same species. The largest bacillus recorded is SOyu^ to GO/u long and 4/i to b/x wide (B. biitschlii, see Fig. 7). One of the smallest forms known {B. influenzas) has an average size of 0.5m x 0.2ij,. Some pathogenic organisms are so small (ultramicroscopic, see p. 70, also chapter on Filtrable Viruses) as to be invisible with any magnifi- cation which we now possess. We know of their existence either by the fact that they may be cultivated on artificial media, producing appear- ances of mass growth, and that such cultures when inoculated into susceptible animals cause the characteristic disease, or by the fact EXPLANATION OF PLATE II. Fig. 1. — lUustrating cocci single or in irregular groups (micrococcus, staphylococcus), micrococcus from air. X 1000. Fig. 2. — -Illustrating cocci in twos — Diplocooous pneumoniae from peritoneal exudate of rabbit. X 1000. Fig. 3. — Illustrating cocci in chains — Streptococcus pyogenes. X 1000. Fig. 4. — Illustrating cocci in fours — Micrococcus tetragenus from spleen of mouse. X 1000. Fig. S. — Illustrating cocci in packets — Sarcina lutea from air. X 1000. Fig. 6. — Illustrating large single bacilli: B. subtilis (hay bacillus). X 1000. Fig. 7. — Illustrating small bacilli; mostly in twos — -B. hoffmanni from human throat X 800. (Park.) Fig. 8. — Illustrating bacilli in chains — ^Anthrax bacillus from spleen of niouse. X 500. Fig. 9. — Illustrating bacilli in bunches — Typhoid bacillus from human spleen. X 500. Fig. 10. — Illustrating bacilli in threads — ^Anthrax bacilli from blood of frog. Fig. 11. — Spirillum undula, showing flagella. X 1000. Fig. 12. — Cholera spirilla, gelatin culture. X 1000. Fig. 13. — Large spirilla in chains from water. X 1000. Fig. 14. — Smaller spirilla in chains — Spirillum rubrum. X 1000. Fig. 15. — Streptothrix Candida — Broth culture (Zettnow, from KoUe and Wassermann) . Fig. 16. — Streptothrix hominis from sputum (Zettnow, from KoUe and Wassermann) . Unless otherwise indicated, the photographs are from Frankel and Pfeiffer. ' A M, or mioromillimeter, is if^-Jjjy of an inch. PLATE n T'PES OF BACTERIA I SPHERE (cocci OR COCCACE^ ) CO oo oocooo 8g —^ CX/T ,W» n CYLINDER (bacilli OR BACTERIACE/E ) !^. K^^^^^^ ^J t^^ ■^ Uf V * mSPIRAL (SPIRILLA OR SPIRILLACE/E ) £^ / C ^ 3 // 12 &.^ ,■7/ /J J-f rVHIGHER BACTERIA ( TRICHOBACTERIA ) BACTERIA 31 that the filtrates alone are infectious. Some filtrates are infectious after passing through the pores of the finest filter. Shape. — ^The basic forms of the single bacterial cells are threefold^ the sphere, the rod, and the segment of a spiral. Although under different conditions the type form of any one species may vary con- siderably, yet these three main divisions under similar conditions are constant, and, so far as we know, it is never possible, by any means, to bring about changes in the organisms that will result in the perma- nent conversion of the morphology of the members of one group into that of another — that is, micrococci always, under suitable conditions, produce micrococci, bacilli produce bacilli, and spirilla produce spirilla. As bacteria multiply, the cells produced from the parent cell have a greater or less tendency to remain attached. This is on account of the slimy envelope which is more or less developed in all bacteria. In some varieties this tendency is extremely slight, in others it is marked. This union may appear simply as an aggregation of separate bacteria or so close that the group appears as a single cell. According to the method of the cell division and the tenacity with which the cells hold together, there are different groupings of bacteria, which aid us in their differentiation and identification. Thus, in cocci we get the bacterial cell dividing into one, two, or three planes (Plate II, Figs. 1-5), while in bacilli and spirilla the division is generally in only one plane (Plate II, Figs. 6-9). 1. Spherical Form, or Coccus. — ^The size varies from about 0.15^ as minimum diameter to 3/x as maximum; the average size of the pathogenic cocci is O.S^i. The single elements are, at the moment of their complete development, so far as we can determine, practically spherical; but when seen in the process of multiplication through division the form is seldom that of a true sphere. Here we have elon- gated or lancet-shaped forms, as frequently seen in the diplococcus of pneumonia, or the opposite, as in the diplococcus of gonorrhea, where the cocci appear to be flattened against one another. Those cells which divide in one direction only and remain attached are found in pairs (diplococci) or in shorter or longer chains (streptococci). Those which divide in two directions, one at right angles to the other, form groups of foiu" (tetrads). Those which divide in three directions and cling together form packets in cubes (sarcinse). Those which divide in any axis form irregularly shaped, grape-like bunches (staphylococci). 2. Rod Form, or Bacillus. — The type of this group is the cylinder. The length of the fully developed cell is always greater than its breadth. The size of the cells of different varieties varies enormously: from a length of 30/i and a breadth of 4/^ to a length of 0.2/i and a breadth of 0.1/i. The largest bacilli met with in disease do not, however, usually develop over 3fi x Ifi, while the average is 2fi x O.Sm- Bacilli are roughly classed, according to their form, as slender when the ratio of the long to the transverse diameter is from 1 to 4 to 1 to 10, and as thick when the proportions of the long to the short diameter is approximately 1 to 2. The characteristic form of the bacillus has a straight axis, with 32 CHARACTERISTICS OF MICROORGANISMS uniform thickness throughout, and flat ends; but there are many exceptions to this typical form. Thus, frequently the motile bacteria have rounded ends, many of the more slender forms have the long axis, slightly bent; some few species, as for example the diphtheria bacilli, invariably produce many cells whose thickness is very unequal at different portions. Spore formation also causes an irregularity of the cell outline. The bacilli, except when they develop from spores or granules, divide only in the plane perpendicular to their long axis. A classi- fication, therefore, of bacilh according to their manner of grouping is much simpler than in the case of the cocci. We may thus have bacilli as isolated cells, as pairs (diplobacilli), or as longer or shorter chains (streptobacilli). 3. Spiral Form, or Spirillum. — The members of the third mor- phological group are spiral in shape, or only segments of a spiral. Here, too, we have large and small, slender and thick spirals. The twisting of the long axis, which here lies in two planes, is the chief characteristic of this group of bacteria. Under normal conditions the twisting is uniform throughout the entire length of the cell. The spirilla, like the bacilli, divide only in one direction. A single cell, a pair, or the union of two or more elements may thus present the appear- ance of a short segment of a spiral or a comma-shaped form, an S-shaped form, or a complete spiral or corkscrew-like form. EXPLANATION OF PLATE III. Partly schematic (Williams). Stained by Giemsa except where otherwise indicated. Figs. 1 to 4. — Anthrax bacilli at different stages of development. Fig. 1. — Shows chromatin-staining masses (red), in plastin-staining cells (light blue). Fig. 2. — Shows same organism after one hour on fresh medium, chromatic substance distributed throughout medium and whole organism taking deeper, more homogeneous stain. Fig. 3. — Organism showing two kinds of granules according to staining powers. Fig. 4. — Same organism placed on fresh medium extrudes some granules (waste- products) and redissolves others (chromatin granules). Figs. 5 to 9. — Similar granules in diphtheria bacilli at various stages of development. Fig. 10. — Metachromatic granules in B. pyocyaneus, Neisser stain. Fig. 11. — Metachromatic granules in Sarcinse, Neisser stain. Fig. 12. — Metachromatic granules in B. influenzae, Giemsa's stain. Fig. 13. — Metachromatic granules in gonocooci, Giemsa's stain. Fig. 14. — Pneumococcus, capsule stained by Hiss method. Fig. 15.— Pneumococcus mucosus (Streptococcus mucosus) by Welch stain for capsule. Fig. 16. — Rhinoscleroma bacillus by Hiss stain for capsules. Fig. 17. — Plasmolysis in cholera bacillus showing capsule. Figs. 18 to 21 inclusive. — Types of flagella by LofHer stain. Fig. 18. — Monotrioha, cholera spirillum. Fig. 19. — Amphitricha, water bacillus. Fig. 20. — Lophotricha, spirillum undula. Pig. 21. — Peritricha, typhoid bacillus. Fig. 22. — Formation and end germination of spores In anthrax. Fig. 23. — Lateral germination of spore in B. subtilis. Fig. 24. — Central germination of spores in B. alvei (Wilson). Fig. 25. — Type of spores: A, central; B, eccentric; C, end. Fig. 26. — Diphtheria bacilli in old cultures, Giemsa's stain. Fig. 27. — Plague bacilli in old cultures, stained by methylene blue. Fig. 28. — Influenza bacilli in old cultures, Giemsa's stain. PLATE HI STRUCTURE OF BACTERIA NUCLEAR MATERIAL AND CELL GRANULES i /> 7 3 9 10 -:^ // 12 •'•^^^■ 4» « W 13 * cS9% •H-liig ii il, .f. 14- CAPSULES ANO MEMBRANES 1 _ --ISS # /5 /5 FLAGELLA /C /7 Uf / r- S .m^- "\- 21 SPORES 22 ^'^ ^ES 25 CSZl r F G IRREGULAR FORMS IN OLD CULTURES (INVOLUTION FORMS] («^'^* A. W. WILLIAMS, DEL. O 9^ .' 27 «•-» % " tP 4 4 ■■*• BACTERIA 33 Structure of Bacterial CeUs.— When examined living in a hanging orop (see p. 71) under the microscope, bacteria appear usually as color- less retractive bodies with or without spores or other more highly retractive areas. It is only by the use of stains that we are able to see more of their structure. Capsule.— Special staining methods (see p. 78) show that many bacteria (some in\'estigators say all) under certain conditions possess a capsule (PMe III, Figs. 14-16), a gelatinous envelope which is sup- posed to be formed from the outer layer of the cell membrane. Some bacteria easily develop a much thicker capsule than others. Such forms are known as capsule bacteria. These generally produce a slimy growth on cultivation (e. g., B. mucosm). Fig. 1. — Bacillus pneumonisE (B. mucosus) stained for capsule by Huntoou's method X about 9nn. rTTuntnon ^ X about 900. (Huntoon.) Capsules develop best in animal tissues. In cultures, with a few exceptions, they require for their development special albuminous culture media, such as milk, blood serum, bronchial mucus, etc. In ordinary nutrient media or on potatoes the capsule may be visible in the first culture generations when grown from the body, but usually it shows very indistinctly, if at all. The capsule is distinguished by a diminished power of staining with ordinary aniline dyes; therefore, unless special staining methods are used, the bacteria may appear to be lying in a clear unstained area. With certain dyes the inner por- tion of the capsule stains, giving the bacteria an apparent greater diameter. The demonstration of the capsule is often of help in difPer- entiating between different but closely related bacteria; e. g., some forms of streptococcus and pneumococcus. Cell Membrane. — That all bacteria possess a cell membrane is shown (1) by special staining methods {e. g., flagella stains, see p. 80) and (2) by plasmolysis, demonstrated by placing the bacteria in a 1 per cent, solution of sodium chloride when the central portion (entoplasm?) contracts and separates in places from the membrane (see also p. 57, 3 34 CHARACTERISTICS OF MICROORGANISMS under Influence of Pressure). In some bacteria the membrane is slightly developed, while in others (e. fir., B. tuberculosis) it is well developed. It is different in composition from the membrane of. higher plants in not possessing cellulose. In some forms, however, a similar carbohydrate, hemicellulose, has been demonstrated. In cer- tain forms a substance related to chitin, found in the cyst walls of protozoa, has been found. Some observers consider the cell mem- brane merely a concentrated part of the cytoplasm, similar to the ectoplasm of higher cells. That it is closely related to the living part of the cell is shown by the connection of the organs of locomotion (flagella) with it (Plate III, Figs. 17 and 21). The Cell Substance. — The chief views in regard to the nature and the structure of the cell substance contained within the membrane may be summarized as follows: 1. Bacteria have a definite morphological, more or less centrally situated nucleus (Feinberg, Nakanischi, ShotteUus, Swellengrebel, and others). 2. Bacteria have no nucleus or differentiated nuclear material (Fischer, Migula, Massart, and others). 3. The whole organism, except the membrane which is a dehcate layer of C5rtoplasm, is a nucleus (BiitscUi, Lowit, Boni, and others). 4 . The nuclear material is in the form of distributed chromatin granules through- out the cytoplasm (Hertwig, Schaudinn, GuiUiermond, Zettnow, and others). 5. A variety of the fourth view is that bacteria possess both chief elements of a cell, namely, cytoplasm and karyoplasm, but that these are so finely mixed that they cannot be morphologically differentiated (Weigert, Mitro- phanow, Gotschlich). 6. The latest view advanced, which is a variation of the views 3, 4, and 5, is that the bacterial cell is a relatively simple body — a cytode in Haeckel's sense, or the plasson of Van Beneden — ^which possesses both chromatin and plastin, the relative amounts of these chief substances of a cell corresponding more to the amounts found in the nuclei of higher cells than in their cytoplasm (Ruzicka, Ambrdz). The last two authors call attention to the fact that both nucleus and cytoplasm in the higher cells are composed of a mixture of chromatin and plastin and that the chief difference between the two mixtures is one of amount and not of kind. Our own studies of the structure of bacteria lead us to agree with the views expressed in Nos. 4 and 6 of the above summary— that is, bacteria possess both chief elements of a cell, namely, chromatin and plastin, and according to the stage of growth and division (varying with species) the chromatin may be in the form of morphological granules or may be so finely divided and mixed with the plastin as to be indistinguishable from it. (See Plate III, Figs. 1, 2, 5, 6, 7.) Metachromatic Granules (Plate III, Figs. 1-13).— These granules appear in unstained bacteria as light-refracting, in stained preparations as deeply stained areas. They have a great affinity for dyes, and so stain readily and give up the stain with some difficulty. With complex stains they show a greater aflSnity than the rest of the bacillus for certain constituents of the stain — e. g., with polychromic methylene blue they take up more of the azure, thus appearing red as does the stained nuclein in morphological nuclei. In certain bacteria, such as BACTERIA 35 the diphtheria baciUi, they are especially well marked in young, vigor- ous cultures. Here they have diagnostic value. At least some of these granules are nuclear in character. Certain other granules which take up stains readily, and others still which absorb stains with difficulty, are of the nature of starch of fat or of other food products. Meyer has described some as being composed of volutin, a protein characterized by insolubility in alcohol and solu- bility in water, acids and alkalis. Certain saprophytic forms have sulphvu", others iron granules. Organs of Motility. — ^The outer surface of spherical bacteria is almost always smooth and devoid of appendages; but that of the rods and spirals is frequently provided with fine, hair-like appendages, orflagella, which are organs of motility (Plate III, Figs. 18-21). These flagella, either singly or in tufts, are sometimes distributed over the entire body of the cell, or they may appear only at one or both ends of the rod. The polar flagella appear on the bacteria shortly before division. The Fig. 2. — Bacilli showing one polar Fig. 3. — Bacilli showing multiple flagellum. flagella. flagella are believed to be formed from the outer cell layer (ectoplasm) or possibly from the capsule, though they have been described by certain authors as arising in endoplasmic granules. So far as we know, the flagella are the only means of locomotion possessed by the bacteria. They are not readily stained, as special staining agents are required for this purpose (see p. 80). The envelope of the bacteria, which usually remains unstained with the ordinary dyes, then becomes colored and more distinctly visible than is commonly the case. Occasionally, however, some portion of the envelope remains unstained, when the flagella present the appearance of being detached from the body of the bacteria by a narrow zone. In stained cultures of richly flagel- lated bacteria peculiar plaited masses sometimes are observed, con- sisting of flagella which have been detached and then matted together. Bacteria may lose their power of producing flagella for a series of gen- erations. Whether this power be permanently lost or not we do not know, 36 CHARACTERISTICS OF MICROORGANISMS Bacteria are named according to the number and position of the flagella they possess as follows: Monotricha (a single flagellum at one pole, e. g., cholera spirillum); amphitricha (a flagellum at each pole, e. g., many spu-illa); lophotricha (a tuft of flagella at one pole,' e. g., Spirillum undula); peritricha (flagella projecting from all parts of surface, e. g., B. alvei, B. typhosus, and others) (Plate III, Figs. 18-21). So far, in only a few bacteria (the largest spirilla) have flagella been demon- strated during life, and then only under special conditions (see K. Reichert for bibliography). We have, however, an organism belonging to the B. alvei group, which shows its flagella very distinctly during life when a small portion of the viscid growth in a liquefying LofSer blood-serum tube is transferred to a hanging mass of agar (p. 72) and examined under high magnification. The flagella on this organism may also -be seen with dark-field illumination. Reichert claims that all motile bacteria show their flagella by this method. Spores. — ^These important structures of the bacterial cell are described in detail under Physiological Characteristics of Bacteria. Physiological Characteristics of Bacteria. — ^The essential physiological activities of bacteria are: motility, growth, reproduction, and spore formation. Motiljity. — ^Many bacteria when examined under the microscope are seen to exhibit active movements in fluids. The movements are of a varying character, being described as rotary, undulatory, sinuous, etc. At one time they may be slow and sluggish, at another so rapid that any detailed observation is impossible. Some bacteria are very active in their movements, difl^erent individuals progressing rapidly in different directions, while with many it is difficult to say positively whether there is any actual motility or whether the organism shows only molecular movements — so-called " Brownian" movements, or pedesis — a dancing, trembling motion possessed by all finely divided organic particles. In order to decide definitely with regard to the motility of any bacterial preparation, it is well to make two hanging drops. To one, 5 per cent, of formalin is added, which, of course, kills the organism. If, now, the live culture shows motility, which is not shown by the killed culture, it is an indication that one is dealing with a motile culture. Very young cultures, of but three or four hours' development, in neutral nutrient bouillon should be examined at a temperatm-e suitable for their best growth. Not all species of bacteria which have flageUa exhibit at all times spontaneous movements; such movements may be absent in certain culture media and at too low or too high temperatures, or with an insufficient or excessive supply of oxygen; hence one should examine cultures under various conditions before deciding as to the non-motility of any organism. The highest speed of which an organism is capable has been approx- imately estimated with some forms, and the actual figures show an actual slow rate of movement, though, comparatively, when the size of the organism is considered, the movement is, rapid. Thus, the 1 Some investigators consider that every flagellum is essentially a tuft, composed of many small fibrils. BACTERIA 37 cholera spirillum may travel for a short time at the rate of 18 cm. per hour. Movement is influenced by many factors, such as chemicals (the oxygen in the air especially), heat, light, and electricity. The tactile property which enables microorganisms to take cognizance of various forces is known as taxis; when forces attract, the phenomenon is known as positive taxis and when they repel, it is called negative taxis. Chemo- taxis, or the effect of chemicals, is taken up in detail on page 59. Growth and Reproduction. — Under favorable conditions bacteria grow rapidly to a certain size, more or less constant for each species, and then divide by fission into approximately equal halves. The average time required for this cycle is twenty to thirty minutes. Probably in all species the nuclear material divides first. This is certainly the case in the group to which the B. diphtherioe belongs, where division of the nuclear granules may be observed in the living organism before the characteristic snapping of the cell body and where division into equal halves seldom occurs. Fio. 4. — Successive stages in division of B. diphtherise, showing relation of line of division to metachromatic granule. Continuous observation of living bacillus drawn without camera lucida. (Williams.) According to our observations on the living cell of members of this group, division takes place at a point occupied by a metachromatic granule (Fig. 4). Before division of the cell body the metachromatic granule, which appears to contain nuclear substance, elongates and shows a darker line at or near its centre. This seems to divide and form two lines,' each of which has at a point near the surface a very tiny, refractive granule, staining deeply with chromatin stains. Between these two lines the cell body suddenly divides with a snap, like the opening of a jack-knife, division beginning at the point between the two tiny granules, and the two new cells remain for a variable time attached at opposite points, thus giving the V-shaped forms. Kurth and Hill also called attention to division by snapping in members of the diph- theria-bacillus group, though neither recognized the relation between the position of the metachromatic granules and the point of division. The tiny granules are probably similar to the cell-partition granules described by various observers. It is not often that the favorable conditions mentioned above for the production of equal and rapid division obtain for any time, since even in pure cultures bacteria in their growth soon produce an envi- ronment unfavorable for further multiplication. Several factors help to make this enviroimient : (1) The using up of suitable food and moistiu-e; (2) the disintegration of food substances into various inju- rious products, such as acids, alkalis, ferments; (3) in mixed cultures the overgrowth of one or more varieties. As these unfavorable con- 38 CHARACTERISTICS OF MICROORGANISMS ditions are more or less constantly present, we seldom see such absolute symmetry in the growth and division of bacteria as is usually described. In fact, except under ideally favorable conditions (e. g., rapid suc- cessive transfers from young cultures on the most favorable food medium), we can never see absolutely equal fission among bacteria; and in some species, notably the diphtheria group, division is extremely irregular even in our usual twenty-four cultures on favorable media. Involution and Degeneration Forms. — ^It follows, from the conditions considered above, that, as cultures grow older or when media unfavor- able to equal division are used, the bacteria may show extremely irreg- ular forms, absolutely different from the young forms, such as long threads or filaments with irregular thickenings, coccus forms from bacilli and spirilla which have divided without increasing in length, bacillar forms from cocci which have grown without dividing, and apparently branched forms from many varieties of bacilli and spirilla. These have been called involution or degeneration forms. In our study of the so-called branched forms of the diphtheria bacillus we have observed the following interesting fact: Under certain conditions, marked apparent branching appears at a definite time in the age of the culture. The conditions are: slightly disturbed growth in pellicle on nutrient broth. When such peUicles are examined every day they are found to contain, on the sixth to the twelfth day, varying chiefly with the amount of disturbance, many large intensely staining forms with one to several apparent branches and many large metachromatic granules (Figs. 5 and 6). The facts that these forms were the only ones to show active growth and division when examined on a hanging mass of agar and that in such growth the metachro- matic granules seem to fuse (Fig. 6) before fission led us to suppose that these' forms represent a primitive sexual process, a sort of autogamy. Schaudinn (Fig. 7) has reported a prunitive conjugation (autogamy) and a relationship between the chromatin granules, or nuclear substance, and the spores in certain bacteria. Although elongation in the greater diameter and complete division at right angles to this is the rule for the majority of bacteria, there are certain groups, which, instead of becoming separated from each other as single cells, tend to produce an incomplete segmentation, the cells remaining together in masses, as the sarcinse, for example, which divide more or less regularly in three du-ections. The indentations upon these masses or cubes, which indicate the point of incomplete fission, give to these bundles of cells the appearance commonly ascribed to them — that of a bale of rags. The rod-shaped bacteria never divide longi- tudinally. Spore formation must be distinguished from vegetative reproduction. This is the process by which the organisms are enabled to enter a stage in which they resist deleterious influences to a much higher degree than is possible for them to do in the growing or vegetative condition. It is true that in all non-spore-bearing cultures a certain proportion of the bacteria are more resistant than the average. No marked differ- ence in protoplasm, however, has been noted in them other than the ability to stain more intensely and sometimes to show strong meta- chromatic areas. The difference between these and the less resistant BACTERIA 39 forms is not great. Some have believed that this resistance is due to certain bodies called arthrospores, which are abnormally large cells with, usually, a thickened cell wall and increased staining properties, formed as a rule in old cultures. Foulerton and others have described similar forms in some of the higher bacteria and consider them spores. (See under Nocardia.) Fig. 5. — B. diphtherias "No. 8" from 9 days' broth pellicle, showing many "branched" forms. Stained with carbol- fuchsin. X 1500 diameters. Fig. 6. — B. diphtherije "No. 8" from 10 days' broth pelhcle, showing longi- tudinal fusion and position of metachroma- tic granules. Stained with Loffler's methylene blue. X 2000 diameters. W i (1 «.'; Fig. 7. — Bacillus biltschlii: a to c, incomplete division of the cell; d to/, gradual collec- tion of chromatin granules at ends of cells; g to i, formation of end spores from these chromatin end masses. (After Schaudinn.) The true spores (endospores) of the lower bacteria are definite bodies. These are strongly refractile and glistening in appearance, oval or round in shape, and composed of concentrated protoplasm developed within the cell and surrounded by a very dense envelope (Plate III, Figs. 22-25). They are characterized by their power of resisting the injurious influences of heat, desiccation, and chemical disinfectants up to a certain limit (see chapter on Disinfection). Spores also stain with great diflBculty. (See page 79 for details.) 40 CHARACTERISTICS OF MICROORGANISMS The production of endospores in the different species of bacteria, though not identical in every instance, is very similar. The conditions under which they are produced in nature are supposed to be similar to those observed in artificial cultures, but they may not always be similar, hence we must not consider a bacterium a non-spore bearer because in the laboratory it has not been seen to form spores. Usually the formation of spores in any species is best observed in a streak culture on nutrient agar or potato, which should be kept at the temperature nearest the optimum for the growth of the organism to be examined. At the end of twelve, eighteen, twenty-four, thirty, thirty-six hours, etc., specimens of the culture are observed, first unstained in a hanging drop or on an agar mass, and then, if round or oval, highly refractile bodies are seen, stained for spores. Each bacillus, as a rule, produces but one spore, and more than two have never been observed. Motile bacteria usually come to a state of rest or immobility pre- vious to spore formation. Several species first become elongated. The anthrax bacillus does this, and a description of the method of its production of spores may serve as an illustration of the process in other bacteria (Plate III, Fig. 22, A-E). In the beginning, the protoplasm of the elongated filaments is homogeneous, but after a time it becomes turbid and finely granular. These fine granules are then replaced by a smaller number of coarser granules, the so-called sporogenous granules supposed to be chiefly nuclear in nature, which by coales- cence finally amalgamate into a spherical or oval refractive body. This is the spore. As soon as the process is completed there may appear between each two spores a delicate partition wall. For a time the spores are retained in a linear position by the cell membranes of the baciUi, but these are later dissolved or broken up and the spores are set free. Not all the cells that make the effort to form spores, as shown by the spherical bodies contained in them, bring these to maturity; indeed, many varieties, under certain cultural conditions, lose altogether their property of forming spores. The following are the most important spore types: (a) the spore lymg in the centre of the cell; which may be much distended in its central portion, giving it a spindle shape or Clostridium, e. g., Bacillus hutyriem; (b) the spore lying at the extremity of a cell much enlarged at that end— the so-called "head spore" or plectridium, e. g., the tetanus bacillus; (c) the spore lying eccentrically (Plate III, Fig. 25, According to Schaudinn and others, in certain spore-bearing bacteria the spore formation is part of a sexual-like process. (See under Reproduction.) The germination of spores takes place as follows (Plate III, Figs. 22-24) : By the absorption of water they become swollen and pale in color, losing their shining, refractive appearance. Later, a little protuberance is seen upon one side (equatorial or central germination) or at one extremity of the spore (polar germination), and this grows BACTERIA 41 out to form a rod which consists of soft-growing protoplasm enveloped in a membrane, which is formed of the endosporium or inner layer of the cellular envelope of the spore. The outer envelope, or exospor- ium, is either cast off when it may be seen in the vicinity of the newly formed rod; or it may he absorbed, as is often the case after central germination. The chief spore formers among the pathogenic bacteria are the anaerobes (tetanus, malignant edema, intestinal bacteria). Only one distinctly pathogenic aerobe produces endospores — the anthrax bacillus. The Higher Forms of Bacteria (see end of Section II).— Some forms grow out into true or false branching threads and thus make a group of organisms intermediate between bacteria and the molds. These have been called higher bacteria or trichobacteria (see table, p. 26). They show increased complexity of structure and function: (1) in forming irregularly segmented filaments composed of elements similar to those found in the lower forms and showing either true or false branching; (2) in developing certain portions of their substance into reproductive bodies from which the new individuals grow (Plate II, Figs. 15 and 16). The filaments seen sometimes among the lower forms have inde- pendent segments, which may easily separate and grow as tiny unicel- lular forms, while in the higher forms the filaments in their growth show a certain interdependence of their parts. For example, growth often occurs from only one end of the filament while the other becomes attached to some fixed object. The members of the higher bacteria which are pathogenic for man have as yet been incompletely studied and classified. The following divisions serve as an attempt at differentiation: 1. Leptothrix grows in stiff, almost straight threads, in which division processes are seldom or never observed, and no branching has been seen. 2. Cladothrix grows in threads which rapidly fragment and produce false branching, that is, the terminal cell remains partly attached, but is pushed to one side by further growth from the parent thread; thus a Y-shaped growth is produced. Bacillary characteristics appear in old cultures. 3. Actinomyces grows in threads with true branching. Spores have been observed. It is characterized by the radiating wreath-like forms which it alone produces in the living body. 4. Nocardia (Streptothrix) grows in threads which produce abundant true branching; later there is fragmentation and formation of conidia. Reproduction among the Higher Bacteria. — ^These forms increase in length for a time and then, at the free ends, or at intervals along the filaments, they produce small rounded cells, called conidia or spores, from which new individuals are formed. The terminal spores may be flagellated after their separation from the parent filament. The flagellated forms may resemble certain flagellata among the protozoa. 42 CHARACTERISTICS OF MICROOROANISMS PROTOZOA. Definition.— A protozoon (the lowest form of life classed in the animal kingdom) is a morphologically single-celled organism composed of protoplasm which is differentiated jnto cytoplasm and nucleus (or nuclear substance) both of which show many variations throughout the more or less comphcated life cycle that each individual undergoes. The protozoa are of higher grade than the bacteria because of their greater complexity in structure and life cycle (Plate IV). Their shape and size vary so widely that no general description will fit all types. Some forms are small enough to pass through a Berkefeld filter, while the largest varieties described are about f inch long. The Cytoplasm. — ^The cytoplasm consists of a mixture of substances, the most important of which belong to the proteins. It is more or less fluid, but, because of differences in the density and solubility of the several parts, it often presents an alveolar, linear, or granular appearance, which may come out clearly in fixed and stained specimens, but is usually not well seen in the living cells. Ectoplasm and Entoplasm. — Frequently the protozoon cytoplasm is' differentiated into a concentrated, viscid, more homogeneous, or hyaline outer layer called the ectoplasm and a more fluid granular central portion called the entoplasm. These two portions have different functions. The ectoplasm helps form the various organs (organelles) of motion, contraction, and prehension such as pseudopods (false feet), flagella (whip-like threads), cilia (hair filaments), suctorial tubules (through which food passes), and myonemes (contractile fibrils found in ciliates, gregarines, and a few flagellates). The entoplasm digests the food and contains the nucleus, as well as various granules and vacuoles. Some vacuoles serve as food digestors, and hence contain digestive ferments. The so-called contractile vacuoles which periodically fill and empty themselves may be considered as excretory organelles. Other substances are seen from time to time in the entoplasm, such as bacteria, red blood cells, fatty granular pigments, bubbles of gas, crystals, etc. The Nucleus. — ^The simplest morphological nucleus is a vesicular body which is differentiated from the cytoplasm by its essential con- stituent chromatin, so-called because it has a strong affinity for certain basic staining materials. Chromatin consists mostly of nuclein and appears in the form of smaller or larger granules, masses, or rods. Generally, the chromatin particles are mixed with a second less intensely staining substance with more of an affinity for acid stains called plastin or paranuclein, similar to the substance from which the true nucleolus of the metazoon cell seems to be formed. This sub- stance may appear in one or more distinct rounded bodies. Most of the chromatic substances of the nucleus in many protozoa are often massed together in an intensely staining ball-like body called the karyosome which undergoes various cyclic changes during the growth and development of the organism. The centrosome is generally PROTOZOA 43 imbedded in the karyosome. The chromatin and plastin lie imbedded in a third substance in the form of an achromatic network called linin, which is closely related to the cytoplasmic network. There may or may not be a definite nuclear membrane. Sometimes there is no defi- nitely structured nucleus, but the nuclear substance in the form of small chromatin masses or granules is distributed throughout the cytoplasm (the so-called "distributed nucleus") similar to that seen in bacteria. Somatic and Generative Chromatin. — Some chromatin substances of the cell have physiological properties different from others. At times substances which have only vegetative properties are active, forming the so-called somatic or trophic chromatin; at other times, substances appear during sexual activities called generative or sexual or idiochromatin, and from these the vegetative (somatic) chromatin for the new cells is again formed. In the ciliata both these chromatin elements are present as distinct morphological bodies during the entire Ufe of the organism, the somatic form in the macronucleus and the generative form in the micronucleus. Chromidia. — ^The chromatin elements, in the form of granules, small irreg- ular masses, threads, network, etc., which at certain stages pass from the nucleus into the cytoplasm, or which at times are, possibly, formed in the cytoplasm, were named ''Chromidien" by R. Hertwig, who in 1899 first described their appearance. Their function in generative processes was demonstrated in 1903 by Schaudinn. Dming their formation the nucleus may entirely disappear, so that morphologically the cell may be considered non-nuclear. At a definite time thereafter new typical nuclei may be formed from those chromidial substances. Locomotor Nucleus {Kinetic Nucleus). — In flagellates stiU another definite physiological chromatin is seen in the small body called the kinetic nucleus, wMch is either apart from or merged into a smaller body, the blepharoplast, forming the root of the flagellum. The kinetic nucleus is so called because it produces the locomotor apparatus. Both the kinetic and trophic nuclei may contain somatic and generative chromatin at the same time. The Centrosome. — This is a small body which is always present in meta- zoan cells, playing an important part in cell division, but it has not been demonstrated as a morphological, entity in many varieties of protozoa; part of the karyosome, however, may take its place, or there may always be a true centrosome within the karyosome. Whenever a centrosome appears in pro- tozoa, it has its origin ui the nucleus, resembling in this the kinetic nucleus and blepharoplast. AU these four bodies (centrosome, blepharoplast, kinetic nucleus, and karyosome) therefore may be considered as having a similar morphological origin. Physiological Characteristics of Protozoa. — In common with all other living organisms protozoa possess the characteristics of motility, nutrition, respiration, and reproduction. Motility. — ^All protozoa react in certain characteristic ways toward chemical, mechanical, and electrical stimuli. Many are affected by light, while probably none reacts to sound. They manifest the reaction usually by motion of some sort. Most animal parasites, especially the higher forms, exert a positive taxis for leukocytes, principally for the large mononuclears and the eosinophiles. This fact is made use of in clinical diagnosis. Nutrition. — Many protozoa, especially the pathogenic forms, absorb fluid food directly through the body wall; but the majority take in 44 CHARACTERISTICS OF MICROORGANISMS solid food, such as small animal or vegetable organisms and organic waste, some through more or less definite regions of the body, others through any part of the surface by extending pseudopodia and entirely surrounding the food object, forming a so-called gastric vacuole. After the food is digested the waste products are excreted, sometimes by osmosis, generally through special structures as the contractile vacuoles which regularly eject fluid substances to the outside of the organism. Respiration. — It is supposed that the contractile vacuole has a respi- ratory as well as an excretory function. The interchange of gases is always going on, if not through a contractile vacuole, then by osmosis through any part of the wall. Growth and Reproduction. — Under favorable conditions, new proto- plasm is constructed rapidly, and the mass increases faster than the surface which, according to Spencer, initiates cell division. The changes generally appear first in the nucleus. The simplest variety of reproduction is a two-celled fission which may be either longitudinal or transverse, either of which may be direct (amitotic) or indirect (mitotic). A modification of equal fission is the so-called budding which may be single or multiple. When growth occurs so that fission is for a time incomplete, one cytoplasm containing several nuclei which finally separate into as many daughter organisms, the process is called multiplicative reproduction, or brood formation or internal budding. In the most extreme cases of multiplicative reproduction as it occurs among sporozoa the mother-cell with its nucleus separates simul- taneously into large numbers of tiny daughter-cells. Such a process, when it occiu's without conjugation and encystment, is called schiz- ogony and the new cells are called merozoites. When such a multi- plicative division occurs (generally after fertilization) within a cyst, it is spoken of as sporogeny and the new cells are called sporo- zoites. Sexual Phenomena. — Sexual phenomena (syngamy) fundamentally similar to those seen in metazoa have been observed in all groups of protozoa studied. The reproduction by the usual division or budding is interrupted at certain times in the life history of each organism and individuals come together in such a way that their nuclei fuse, usually after having undergone characteristic reduction divisions. When the union is permanent, we speak of it as copulation and liken the process to that of the fecundation of the ovum by a spermatozoon. When the union is transient we call it conjugation. Here the two cells fuse for a time when the nuclei interchange protoplasm and then the cells separate and each one continues to grow and divide independently. When in a partly divided cell or in an apparently single cell, two nuclei, after undergoing reduction division, or its like, fuse, the process is called autogamy. The developmental cycle of a protozoon consists of all the changes which occur in its growth from one act of fertilization to another (Plate IV, Fig. III). Many protozoa carry on the sexual part of their life cycle in one host and the asexual part in another (e. g., malarial organisms). ■ PROTOZOA 45 Cyst Formation. — If protozoa do not get the required amount of water or air or suitable food, they cease their special movements, round out into more or less of a sphere and form a resisting membrane of chitin within which they may live for a long time, withstanding periods of desiccation, extreme heat and cold, and they may be blown about as dust until they find conditions again favorable for renewed growth; then water is absorbed, the cyst is ruptured and active life begins anew. In parasitic forms encystment plays an important part in the passage from the old host to the new. The majority of forms would not be able to exist outside of the body of the host without having some protective membrane. The cyst may be formed simply for protection from drought, etc., when it is called a hypnocyst, from which the organism may emerge in about the same form as when it encysted; or the cyst may precede reproduction by spore for- mation or simple division, when it is called a sporocyst. In either case it may consist of a simple wall or it may be formed of several walls to enable it to resist prolonged desiccation, when it is called a resting cyst. Characteristics of Each of the Four Groups of Protozoa.— Flagellata. — Flagellata are protozoa which move in the adult forms by one or several flagella or whip-like processes. If pseudopodia develop, they are transitory. Generally the flagella arise from the anterior part of the organism, and in motion the larger ones (primary flagella) are directed forward, while smaller ones (secondary flagella) are directed backward, acting as rudders. Certain flagellata possess a modification of their bodies in what is called the undulating membrane, which consists of a fluted protoplasmic process attached along one side of the organism, the free edge of which is prolonged as the flagellum. It has been shown that flagella are not simple protoplasmic processes, but that they have more or less of a framework of elastic fibers as well, hence their power in locomotion can be better understood. Except with special stains, which bring out these fibers, they appear homogeneous. The flagella arise from some definite place in the cytoplasm, some- times from a distinctly differentiated chromatic body which has been given various names, such as blepharoplast, kinetic nucleus or centro- some, sometimes near this from a basal granule, microsome, diplosome, or flagellum root, sometimes directly from the nucleus. The basal granules seem to be derived primarily from the kinetonucleus, and may be considered from a physiological stand-point as a part of the motor nuclei. The body of the flagellates is generally more or less elongated and, except in most primitive ones, is fixed in its outline. The latter charac- teristic is chiefly due to the fact that the organisms usually possess definite though deUcate membranes containing elastic fibrils. The cytoplasm is usually not differentiated into an ento- and ectoplasm. It often contains one to several contractile vacuoles, as well as food vacuoles, and there is frequently a definite opening or cytostome for the entrance of food. There are usually many granules and inclusions 46 CHARACTERISTICS OF MICROORGANISMS of various kinds scattered throughout the cytoplasm, and myoneme striations are seen in some forms. The nucleus, as a rule, situated anteriorly, varies much according to different species and to different stages of development. The flagellata multiply either in the free motile condition or after encystment. In the first case, as a general thing, they divide longi- tudinally. The basal granules divide with the nuclei and the flagella EXPLANATION OF PLATE IV. Partly schematic. Rearranged and drawn by Williams. All stained by Giemsa. I. Flagellates. Fig. 1. — Illustrating one flagellum. Leishmania: A, intracellular form; B, cultural forms. Fig. 2. — Illustrating undulating membranes: A, Trypanosoma lewisi; B, Trypanosoma brucei; C, Trypanosoma gambiense. Fig. 3. — Illustrating two flagella. Bode lacertae (after Prowazek). Fig. 4. — Illustrating four flagella. Trichomonas. II. AMEB.a!. Illustrating points considered differential in the two chief types of amebae (entamebEe) described as parasitic in human beings. Fig. 1. — Entameba coli, vegetative stage. Fig. 2. — Dividing nucleus. Fig. 3.. — Entameba coli cyst containing eight nuclei. Fig. 4. — Entameba histolytica, vegetative stage. Fig. 5. — Four-nucleate cyst. III. Spohozoa. A, description of Figs. 1 to 16. (After Schaudinn.) The life cycle of Eimeria schubergi. In 1, the sporozoites, becoming free by bursting the sporocysts, pass out through an aperture in the wall of the oocyst, and are ready to enter the epithelial cells of the host. 2 to 6 represent the asexual reproduction or schizogony, commencing with infection of an epithelial cell by a merozoite or a sporozoite; the merozoite after stage 6 may start again (5) at stage 2, as indicated by the arrows, or it, may go on to the formation of gametocytes (9 to 11). 9 to 11 represent the sexual generation, the line of development becoming split into two lines — male (tf) and female (?) — culminating in the highly differentiated gametes, which conjugate and become again a single line, shown in 12-14. The zygote thus formed goes on to the production of spores, 15 and 16. 2 and 3 represent epithelial cells showing penetration of a merozoite or a sporozoite and its change into a schizont; 4, the nucleus of the schizont divided into numerous daughter-nuclei; 6, segmentation of the schizont into numerous merozoites, about a central mass of residual proto- plasm, which in this figure is hidden by the merozoites; S, merozoites passing to reinfect host cell and repeat the process of schizogony; 7, 8, merozoites to be differentiated into male and female gametocytes; 9, the two gametocytes within a host cell; the microgametocyte (cf) has fine granulations; the macrogametocyte (2) has coarse granulations. 11, a female gametocyte undergoing maturation; 13, mature macrogamete, freed from the host cell, and sending a cone of reception toward an approaching microgamete. In 12 the nuclei of the last stage have become microgametes, each with two flagella. The free microgametes are swimming to find a macrogamete. 14, the zygote (fertilized macrogamete), surrounded by a tough membrane or oocyst, which allows no more microgametes to enter, and con- taining the female chromatin, which is taking the form of a spindle, and the male chromatin in a compact lump. 15, the nucleus of the zygote divided — the nuclei of the sporoblasts. In 16 the four sporoblasts become distinct, leaving a small quantity of residual protoplasm; each sporoblast has formed a membrane, the sporo- cyst. Within each sporocyst two sporozoites form about a sporal residuum. B, Babesia infecting red blood cells: 1, pear-shaped bodies; 2, dividing forms; 3, eight pear-shaped bodies in a cell; 4, irregular ring-like bodies; 5, large, irregular body; 6, body with a flagellum-like projection. IV. CiLIATES. Fig. 1. — Balantidium coli (after Hartmann) : A, adult form; B, C, dividing forms; D, conjugating forms, PLATE IV TYPES OF PROTOZOA I.FLAGELLATES-M0VIN6 BY FLA6ELLA ■0B *i. \\_ >• * # ^ \^^ iJ) .a^->i Y V )c IX 2 "3 n.AMEB/E - MOVING BY PSEUDOPODS J HSPOROZOA-INTRACELLULAR, PRODUCING MANY SPORES »«*"» 6 A. W. WILLIAMS, DEL. PROTOZOA 47 of the daughter organisms are usually formed anew. Multiple division is also observed. In the second case the flagellata may or may not conjugate before they encyst. Then they divide within the cyst. The sexual cycle varies much in different species. Isogamy has been noticed between fully grown individuals as well as between smaller forms. The union of different-sized forms, or anisogamy, has also been observed. Also autogamy is said not to be infrequent. It is claimed that certain of the flagellates pathogenic for man require a second host for the development of their sexual cycle. The flagellates are subdivided into several orders, in only two of which are forms found which are pathogenic for man. (See Part II.) The different types of pathogenic flagellates are shown in Plate IV, i, Figs. 1-4. Among the most important pathogenic forms are the trypanosomes. Trypanosoma. — ^The name trypanosoma (boring animal) was given by Gruby, in 1843, to certain free-swimming hemoflagellates found by him in the blood of frogs. Much later similar flagellates were found in the blood plasma of many different species of vertebrates and in the intestinal tract of several blood-sucking invertebrates. Some of the forms, including those found in man, are pathogenic. A number of the blood-sucking insects are carriers of the diseased species to healthy animals. Typical trypanosomes are characterized by a comparatively long, spirally twisted body, along one side of which is attached an undulating membrane having a cord-like edge that is continued forward as a free whip (flagellum). The flagellum arises near the posterior end of the organism in a small granule called the basal granule, which may be connected with the blepharoplast, a larger chromatin mass, called also the kinetonucleus because of its control over the motor apparatus. The nuclear apparatus consists of a macro- or trophonucleus, and of the above-mentioned kinetonucleus or blepharoplast, which last func- tions as a centrosome. The trophonucleus is usually situated near the middle of the organism; it is granular, thick, and egg-shaped, but varies somewhat in size and shape. The cytoplasm is faintly alveolar or granular, varying with age, environment, and possibly species. Toward the straight border of the cell the cytoplasm is more or less striated and in a few species definite myonemes are seen. Reproduction occurs usually by longitudinal, occasionally by multiple division. The 'life cycle is not well known. Though transmission occurs through the bites of various invertebrates, notably flies, the few sexual changes described as taking place in the intestines of some of these intermediate hosts have not been fully corroborated. That an intermediate host is not necessary for the continued life of at least one species of trypanosome seems to be proved by the fact of direct trans- mission of T. equiperdum from horse to horse through coitus. Leishmania. — ^Another important pathogenic flagellate is Leishmania. In humans this grows chiefly within large mononuclear cells. It shows its flagellated forms in cultures. (Plate IV, Fig. 1. See also Part II.) 48 CHARACTERISTICS OF MICROORGANISMS Amebida. — Under amebida (syn., amebse) we include forms composed of naked, simply constructed protoplasm having the power of produc- ing lobose pseudopodia which are used as organs of motion and of nutrition. The pseudopodia are protoplasmic processes which are projected in irregular succession from different parts of the surface of the cell, producing in this way an irregular motion. The form of the pseudo- podia varies considerably in the different species. For instance, there are broad, blunt processes or narrow, less blunted ones, and each may be short or long, single or slightly branched. The cytoplasm may or may not take a share in their formation. The forms, of course, vary within limits according to the condition of the medium in which the amebse are living. Movements are always called forth by some physi- cal or chemical excitant. When such an excitant is desirable for food the pseudopods flow around it, and it is subsequently absorbed in the cytoplasm of the organism. The members of this group may possess one nucleus or several. Ameba hinucleata has two nuclei in the young adult stage, and Pelo- myxa palvstris, living in the bottom ooze of ponds, has an enormous number of nuclei. A marked feature of the nuclear apparatus is the formation of chromidia which, as has already been noted, may play such an important part in sexual reproduction. Generally each ameba hs,s one contractile vacuole, but occasionally some are seen with several or with none. Saprophytic forms belonging to this order are common. They may be found wherever there are moisture and decaying vegetable matter. The pathogenic forms are not so frequent. Because of the possibility that the still unknown causes of certain diseases (see Rabies and Smallpox) are organisms related to this order, it is especially important to study both saprophytic and pathogenic varieties, since a knowledge of the former which are more easily studied may help us understand obscure points in the life history of the latter. Notwithstanding the common occurrence of saprophytic forms, the full life history of few of them has been worked out, and until the full cycle of development of any so-called ameba is known it is impossible to say whether that particular form belongs among rhizopoda or whether it is a developmental form of another group, as ameboid forms may occur at some time in the life history of all groups. It is quite possible that some of the organisms described as belonging to this order are really members of entirely different orders. For instance, it is known that the flagellate Trichomonas loses its flagella before copulation and crawls about by means of short blunt pseudopods as a typical ameba. Amebse reproduce by simple fission, by budding, and by brood for- mation. In the last case the reproduction is usually preceded by encyst- ment. Two forms have been described most frequently as parasitic in man. The differences between them are shown in Plate IV, ii. Figs. 1-5. PROTOZOA 49" The Sporozoa. — The sporozoa are a group of exclusively parasitic protozoa of very wide-spread occurrence, living in the cells, tissues, and cavities of animals of every class. Generally they are harmless, but some varieties may produce pathological changes and even fatal diseases severely epidemic. As their name indicates, they are all characterized by reproduction through spore formation, but they exhibit the utmost diversity of structural and developmental characteristics. As a rule, each species is parasitic on one kind of tissue of a particular species of host. They are generally taken into the system in the spore stage either (1) with the food of the host, (2) by the bites of insects, or (3) by inhalation. The spore membranes are dissolved by the fluids of the host, and thus one or more germs of sporozoites are set free to bore into the special cells of the host. Here they grow, some remaining permanently intra- cellular, others only in the young stages. The latter either pass differ- ent phases of their more or less complicated life history in different parts of the body of one and the same host or they pass some phases of their life cycle in the cells of an intermediate host. . The sporozoa vary widely in size as well as in other characteristics. Prom the smallest, several of which can be contained in a single blood cell, there are all gradations in size up to those that may be seen by the naked eye {Porospora gigantea, 16 mm.). Besides being characterized by the power to produce very many resisting spores, the sporozoa are also characterized by the fact that as a class they possess none of the special organs found in other pro- tozoa for ingesting or digesting solids. Many develop flagella during sexual phases or show ameboid movement during certain stages of their life cycle, but the flagella and pseudopodia are organs of loco- motion and not of nutrition. Food vacuoles or contractile vacuoles have not been found. The life cycle of a typical sporozoon is rearranged and condensed from Schaudinn in Plate IV, iii, Figs. 1-16. Ciliata. — ^The ciliata (Plate IV) belong to the most complex of the protozoa. They possess a definite entoplasm containing nuclei and food vacuoles, and a definite ectoplasm containing basal granules from which arise the cilia which give the group its name. They have organoid structures which receive the food, some having definite mouth openings, and definite places for excreting waste products. The food vacuoles may contain acid or alkaline digestive products. The nuclear material is differentiated into two forms, a large macronucleus and a much smaller micronucleus. The function of the macronucleus is sup- posed to be vegetative, and that of the micronucleus reproductive. The macronucleus varies in size and shape and is completely filled with an alveolar chromatin. The micronucleus also varies in size and shape, and except in reproductive phases is generally vesicular in structure, with the chromatin heaped in one mass. Division of the nuclei takes place by mitosis in the case of micronuclei, and by amitosis, as a rule, in the case of the macronuclei. Under conditions unfavorable for growth the ciliata may encyst. 4 50 CHARACTERISTICS OF MICROORGANISMS Conjugation is necessary to the life activity of these organisms. The phenomenon of conjugation in the ciliata has been well worked out. The micronuclei play the most important part, whereas the macro- nuclei simply break up and disappear in the protoplasmi. According to the arrangement of the cilia, the ciliata are divided into the four orders given in the general classification (see p. 26). Among these, the second, the order of the Heterotricha, interests us. In the Heterotricha the cilia are uniform over most of the body, while a specialized set fused into a series of firm vibratory plates is found about the mouth. Only one genus, Balantidium, has been observed in man (Plate IV; also Part II). CHEMICAL COMPOSITION OF MICROORGANISMS. Quantitatively considered, the bodies of microorganisms consist largely of water, salts (chiefly phosphorus, potassium, chlorine, calcium iron and sulphur), fats, and albuminous substances. There are also present, in smaller quantities, extractive substances soluble in alcohol and in ether. Special varieties contain unusual substances, as wax and hemicellulose in tubercle bacilli. Each variety, furthermore, yields protein substances peculiar to itself, as shown in the effects produced by animal inoculation. At present we know but little con- cerning the differentiation of these specific substances. This subject will be taken up in detail under Toxins, etc. According to Cramer, many bacteria contain amyloid substances which give a' blue reaction with iodine. True cellulose has not been found in bacteria by Vaughan or other workers, but large quantities of a gelatinous carbohydrate similar to hemicellulose have been obtained. Nuclein is found in all microorganisms. The nuclein bases — zanthin, guanin, and adenin — have been obtained in considerable amounts. Vaughan found no sodium chloride in his alcoholic extracts. There is a group of bacteria which contains large amounts of sulphiu- — viz., the Beggiatoa — and another group, the Cladothrix, is capable of separating ferric oxide from water containing iron. ^Microorganisms possess the capacity to a high degree of accom- modating their chemical composition to the variety of soil in which they are growing. Thus, B. prodigiosus when grown on potato con- tains 21.5 per cent, of dry residue and 2.7 per cent, of ash; when cul- tivated on turnips it contains 12.6 per cent, of dry residue and 1.3 per cent, of ash. Besides the concentration of the culture, its tem- perature and age also influence the amount of residue and ash pro- duced. Qualitatively, a variation is shown by the addition of peptone in the culture media which tends to increase the percentage of nitrog- enous matter in the microbe, or by the addition of glucose which decreases it. The chemical composition of the bodies of animal parasites is an almost unexplored field. The ectoplasm and the cyst sacs in general are made up principally of a substance called chitin. Glycogen has EFFECTS OF SURROUNDING FORCES UPON MICROORGANISMS 51 been isolated from many forms. Proteolytic enzymes and acid secre- tion in digestive vacuoles have been demonstrated. Microchemical Reactions.— To a certain degree the chemical com- position of the individual organism may be studied both in the living and in the dead individual by the addition of the testing substances to a hanging drop or to a spread of such organism and the examination of it under the microscope. Thus, fats have been demonstrated by staining with osmic acid, Sudan III, or Scharlach R., as well as by alcohol-ether extraction. Of special importance in this regard is the resistance which bacteria possess to diluted alkalies. Inasmuch as the majority of animal tissues are dissolved when treated with alkalies, this method has been adopted for rendering visible unstained bacteria in tissues. (See also Principles of Staining, p. 77). EFFECTS OF SURROUNDING FORCES UPON MICROORGANISMS. 1. Food. — ^Naturally, the eifect of food upon organisms is marked. Though the majority of pathogenic microorganisms grow easily on certain artificial foods (culture media), some of them, like most of the protozoa, we have not yet been able to cultivate outside of the body of their host. Those microorganisms which seem to depend entirely upon a living host for their existence are known as strict 'parasites; those which live only upon dead organic (a few on inorganic) sub- stances are called strict saprophytes; those which can lead a saprophytic existence, but which usually thrive only within living tissues, are called facultative saprophytes, while those that grow usually on dead material, but may grow within living tissues, are called facultative parasites. The strict saprophytes, which represent the large majority of all microorganisms are not only harmless to living organisms, but perform many exceedingly important functions in nature, such as the destruction of dead organic matter and its preparation for plant food through decomposition, putrefaction, and fermentation, while onr group (see below, the Nitrifying Bacteria) are constructive in their- activities. . The parasites, on the contrary, may be harmful invaders (pathogenic microorganisms) of the body tissues, exciting by their growth and products many forms of disease. (See chapter on Rela- tion of Microorganisms to Disease.) The substances essential for the majority of those forms which can be grown artificially are organic material as a source of carbon and nitrogen, an abundance of water, and certain salts. Either calcium or magnesium and sodium or potas- sium salts are usually required, also sulphur and phosphorus salts. Iron is demanded by a few varieties. The demands of microorganisms for food of a definite composition vary considerably. (See chapter on Cultivation of Microorganisms.) While it is true that very wide differences in relative composition and total concentration of food media may have slight effect upon the general development of a given organism, slight changes in com- 52 CHARACTERISTICS OF MICROORGANISMS position and reaction of the media often have a great effect upon morphology, rate of growth, motility, and specific products of growth. Reaction of Media. — The reaction of the media is of very great impor- tance. Most bacteria, particularly the pathogenic forms, grow best on those media that are slightly alkaline or neutral to litmus. Yeasts and molds grow best on a slightly acid medium. An amount of acid or alkali insufficient to prevent the development of bacteria may still s.ufEce to rob them of some of their most important functions, such as the production of poison. The different effect upon closely allied varieties of bacteria of a slight excess of acid or alkali is sometimes made use of in separating those which may be closely allied in many other respects. Influence of One Species upon the Growth of Another. — When one species of organism is grown in a food medium, that medium usually becomes less suitable for the growth of its kind and of other organisms. This is due partly to the impoverishment of the foodstuffs, but more to the production of chemical substances or enzymes. When differ- ent species are grown together, the antagonistic action of one upon the other may be shown from the beginning. Some species, however, have a cooperative or symbiotic action with other species. In nature, microorganisms usually occur in mixed cultures (e. g., in water, milk, intestinal contents of all animals), and here we may see antagonistic action in the prevalence of one species over others (e. g., the lactic acid formers in the intestines), or cooperative action in the equal and luxuriant growth of two or more species (e. g., pneumococcus and influenza bacillus in the lungs). Experimentally, the existence of antagonisms can be demonstrated by inoculating alternate streak cultures of various bacteria on gelatin or agar niedia. It is found that many species will grow not at all or only sparingly when in close proximity to some other species. This antagonism, however, is often only one-sided in character. Again, when gelatin or agar plates are planted with a mixture of two species of bacteria, it may be observed that only one of the two grows. A third method of making this experiment is simultaneously to inoculate the same liquid medium with two species, and then to examine them later, both microscopically and by making plate 'cultures; not infre- quently one species may take precedence over the other which after a time it may entirely overcome. The symbiotic or cooperative action of microorganisms may be demonstrated experimentally in the following examples: (a) Pneumococci, when grown together with a bacillus obtained from the throat, produce very large, succulent colonies. The influenza bacillus, which will not grow alone upon ordinary nutrient agar, will grow well there in the presence of certain other bacteria. Some anae- robic species grow even with the admission of air if only some aerobic species are present (tetanus bacilli with diphtheria bacilli). (b) Certain chemical effects, as, for instance, the decomposition of nitrates, cannot be produced by many species of bacteria alone, but only when two are associated. EFFECTS OF SURROUNDING FORCES UPON MICR06rGANISMS 53 2. Behavior towaxd Oxygen and other Gases.— The majority of microorganisms absolutely require free oxygen for their growth, but a considerable minority fail to grow unless it is excluded. This latter fact, noted first by Pasteur, led him to divide germs into aerobic and anaerobic forms. Between these two groups we have those that can grow both with and without oxygen. C rganisms that can grow under conditions other than the most favorable are called facultative organisms. (a) Aerobic Organisms. — Growth only in the presence of free oxygen. The slightest restriction of air inhibits development. Spore formation, especially, requires the free admission of air. (6) Anaerobic Organisms. — Growth and spore formation only on the total exclusion of free oxygen. Among this class of organisms are the bacillus of malignant edema, the tetanus bacillus, the bacillus of symp- tomatic anthrax, and many soil bacteria. Exposed to the action of oxygen, the vegetative forms of these bacteria are readily destroyed; the spores, on the contrary, are very resistant. Anaerobic germs being deprived of free oxygen — ^the chief source of energy used by the aerobic species to oxidize the nutritive substances in the culture media — are dependent for their oxygen upon decomposable substances, such as grape-sugar. (c) Facultative Anaerobic and Facultative Aerobic Organisms. — The greater number of aerobic germs, including most of the pathogenic species, are capable of withstanding, without being seriously affected, some restriction in the amount of oxygen admitted (facultative anaerobes), and some grow equally luxuriantly under both conditions. Life in the animal body, for example, as in the intestines, necessi- tates existence with diminished supply of oxygen. If in any given variety of bacteria the amount of oxygen present is unfavorable, there will be more or less restriction in some of the life processes of this variety, such as pigment and toxin production, spore formation, etc. Pigment formation almost always ceases with the exclusion of oxygen, but poisonous products of decomposition may be more abundantly produced. It has been observed not infrequently that certain species which on their isolation at first show more or less anaerobic development — that is, a preference to grow in the depth of an agar-stick cultm-e, for instance — after awhile seem to become markedly aerobic, growing abun- dantly on the surface of the medium (facultative aerobes) . Those organisms that grow best or grow exclusively when the oxygen is only partly removed are called micro-aerophilic organisms} Other Gases. — ^While all facultative organisms as well as strict anae- robes grow well in nitrogen and hydrogen, they behave very differently toward carbonic acid gas. A large number of these species do not grow at all, being completely inhibited in their development until oxygen is again admitted — for example, B. anthracis and B. subtilis and other allied species. It has been found in some species, as glanders ' Lyon, in Science, 1917, Ixv, 19, suggests that the word oligaerobic is better than the word micro-aerophilic. 54 CHARACTERISTICS OF. MICROORGANISMS and cholera, that the majority of the organisms are quickly killed by CO2, while few, such as staphylococci, offer a great resistance, render- ing impossible complete sterihzation by means of this gas. The streptococcus as well as the staphylococcus exhibits a scanty growth. A mixture of one-fourth air to three-fourths carbonic acid gas seems to have no injurious effect on bacteria which cannot grow in an atmos- phere of pure CO2. Under pressure CO2 is more effective (page 58). Sulphuretted hydrogen in large quantity is a strong bacterial poison. Even in small amounts it kills some bacteria. 3. Eifect of Temperature. — Some form of microbic life is possible within the limits of d° and 70° C. The maximum and minimum tem- peratiue for each individual species ordinarily lies from 10° to 30° C. apart, and the optimum covers about 5° C. Usually the temperature of the soil in which the germs are deposited is the controlling factor in deciding whether growth will or will not take place. Thus, nearly all parasitic microorganisms require for development a temperature near that of the body of their host, while many saprophj'tic forms grow best at temperatures lower than 37° C. Microbes when exposed to lower temperature than suffices for their growth, while having their activities decreased, may not be otherwise injured unless actually frozen for a certain time; when exposed to higher temperatures than allows of growth the life of the organism is more or less quickly destroyed. Sudden marked changes in temperature are detrimental. Microorganisms have been classified according to the temperatures at which they develop, as follows: Psychrophilic Microbes. — Minimum at 0° C, optimum at 15° to 20° C, maximum at about 30° C. To this class belong many of the water microorganisms, such as the phosphorescent bacteria in sea-water; and many molds and yeasts. Mesophilic Microbes. — Minimum at 5° to 25° C, optimimi about 37° C, maximum at about 43° C. To this class belong all pathogenic bacteria, most parasitic and many saprophytic forms. Thermophilic Microbes. — Minimum at 25° to 45° C, optimum at 50° to 55° C, maximum at 60° to 70° C. This class includes a number of soil bacteria which are almost exclusively spore-bearing bacilli. They are also found widely distributed in feces. By carefully elevating or reducing the temperature the limits within which a species will grow can be altered. Thus, the anthrax bacillus may be gradually made to accommodate itself to a temperatiu-e of 42° C, and pigeons, which are comparatively immune to anthrax, partly on account of their high body temperature (42° C), when inocu- lated with this anthrax succumb to the infection. Another culture accustomed to a temperature of 12° C. kills frogs kept at 12° C. We have cultivated a very virulent diphtheria bacillus so that it will grow at 43° C. and produce strong toxin. Effect of Low Temperature. — ^Temperatures even far under 0° C. are only slowly injurious to microorganisms, different species being affected with varying rapidity. This has been demonstrated by numerous EPPEdTS OP StJRROVNDiNG PORCES VPON MiCROdR&ANlSMS 55 experiments in which they have been exposed for weeks in a refriger- ating mixture at —18° C. If a culture of typhoid bacilli is frozen, about 50 to 70 per cent, of the organisms are killed at the time. At the end of one week not more than 10 per cent, survive, and at four weeks not over 1 per cent. After six months none survives. More resistant bacteria live longer and spores may survive in ice for years. Bacteria have even been subjected to a temperature of — 175°,C. by immersing them in liquid air kept in an open tube for two hours, and 15 to 80 per cent, were found still to grow when placed in favorable conditions. We found about 10 per cent, of typhoid bacilli alive after thirty minutes' exposure to this low temperature. Staphylococci were more resistant. Spores were scarcely affected at all. Efiect of High Temperatures. — Prolonged temperatures from 5° to 10° C. over the optimima affect microorganisms injuriously in several respects. For instance, varieties may be produced of diminished activity of growth, the virulence and the property of causing fermen- tation may be decreased, and the power of spore formation may be gradually lost. If the maximum temperature is exceeded, the organism dies. The thermal death-point for the psychrophilic species is about 37° C, for the mesophUic species about 45° to 55° C, and for the thermophilic species about 75° C. There are no non-spore-bearing bacteria, except possibly a few cocci, which when moist are able to withstand a tem- perature of 100° C. even for a few minutes. A long exposure to tem- peratures between 60° and 80° C. has the same result as a shorter one at the higher temperatures. Ten to thirty minutes' exposure to moist heat will at 60° C. kill the cholera spirillum, the streptococcus, the typhoid bacillus, and the gonococcus, and at 70° C. the staphylococcus, the latter being among the most resistant of the pathogenic organisms which show no spores. A much shorter exposure will kill a large percentage of any mass of these bacteria. Effect of Dry Heat. — ^When microorganisms in a desiccated condition are exposed to the action of heated dry air, the temperature required for their destruction is much above that required when they are in a moist condition or when they are exposed to the action of hot water or steam. A large number of pathogenic and non-pathogenic species are able occasionally to resist a temperature of over 100° C. dry heat for from ten minutes to one hour. In any large number of bacteria a few are always more resistant than the majority. A temperature of 120° to 130° C. dry heat maintained for one and a half hours will destroy all bacteria in the absence of spores. Resistance of Spores to Heat. — Spores possess a great power of resist- ance to both moist and dry heat. Dry heat is comparatively well borne, many bacterial spores resisting a temperature of over 130° C. for as long as three hours. Exposed to 150° C. for one hour, practi- cally all spores are killed. Moist heat at a temperature of 100° C, either boiling water or free-flowing steam, destroys the spores of most varieties of bacteria within fifteen minutes; certain pathogenic and 56 CHARACfERlSflCS OF MICR06rGANISMS ilon-pathogenic species, however, resist this temperature for hours. The spofes of a bacillus from the soil were destroyed after five and a half or six hours' exposure to streaming steam. They were destroyed, however, by exposure for twenty-^five minutes in steam at 113'' to 116^* C. and in two minutes at 127" C. The spores from tetanus bacilli may require twenty minutes' exposure to kill them. Spores in fatty media are more resistant to heat. The resistance of spores to moist heat i^ tested by suspending threads, upon which the spores have been dried, in boiling water or steam. The threads are removed from minute to minute and laid upon agar or in broth and kept at a suitable temperature for the germination of any living spores. 4. Influence of Light. — A large number of microorganisms are inhib- ited in growth by the action of bright daylight, more are affected by direct sunlight, and when the action of the sun's rays is prolonged they lose their power of developing when later placed in the dark. Some motile organisms move toward the point of greatest luminosity, others away from it. Light-seeking protozoa have green or yellow chromatophores, and usually at the anterior end a red pigment spot. The violet and blue rays are more active than other parts of the spectrum in determining motion. The susceptibility of bacteria to light may be tested, according to H. Buchner, by suspending a large number of bacteria in nutrient gelatin or agar and pouring the media while still fluid in Petri dishes, upon each of which has been pasted a strip of black paper on the side exposed to the light. The action of heat may be excluded by allowing the ray of light to pass through a layer of water or alum of several centi- meters' thickness. After the plates have been exposed to the light for one- half, one, one and a half, two hours, etc., they are taken into a dark room and allowed to stand at 20° to 35° C. a sufficient length of time to allow of growth, and then examined to see whether there are colonies anywhere except on the parts covered by the paper; when the colonies exposed to the light have been completely destroyed, there is lying in a clear sterile field a sharply defined region of the shape of the paper strip crowded with colonies. Protected by ordinary non-colored glass the sun's rays act very slowly. Only the ultraviolet, violet, and blue rays of the spectrum seem to possess bactericidal action; green light has very much less; red and yellow light, none at all. The action of light is apparently assisted by the admission of air; anaerobic species, like the tetanus bacillus, and facultative anaerobic species, such as the colon bacillus, are able to withstand quite well the action of intense, direct sunlight in the absence of oxygen, for four hours. According to Richardson and Dieudonne, the mechanism of the action of light may be at least partially explained by the fact that in agar plates exposed to light for a short time (even after ten minutes' exposure to direct sunlight) hydrogen peroxide (H2O2) is formed. This is demonstrated by exposing an agar plate half-covered with EPFEdfSOF SURROUNDING PORCES UPON MICRo6rGANISMS 5? black paper, upon which a weak sohition of iodide of starch is poured, and over this again a dilute solution of sulphate of iron; the side exposed to the light turns blue black. In gases containing no oxygen, hydrogen peroxide is not produced, and the light has no injurious effect. Access of oxygen also explains the effect which light produces on culture media which have been exposed to the action of sunlight, as standing in the sun for a time, when afterward used for inoculation. Some bacteria subsequently introduced into such media grow badly — far worse than in fresh culture media which are kept in the shade. Influence of Radium. — Radio-active fluids have a slight inhibiting eflfect on microbal growth, but nothing decided enough to be used for therapeutic purposes has been evolved up to the present time. Influence of X-rays. — These rays have a slight inhibiting effect on microorganisms when they are directly exposed to them. 5. Influence of Electricity. — ^The majority of the observations here- tofore made on this subject would seem to indicate that there is no direct action of the galvanic current on bacteria; but the effect of heat and the electrolytic changes in the culture liquid resulting from the electrolysis may destroy them. Protozoa may be contracted by moderate induction shocks and killed by strong ones. When a current of electricity is passed through a liquid medium most active protozoa swim with their long diameters in the direction of the lines of force to assemble behind the cathode. Most flagellates and a few ciliates, however, move toward the anode. The direction of motion has been shown by Dale to vary with the nature and concen- tration of the mediima. This whole question has been little studied. 6. Influence of Agitation. — Meltzer has shown that the vitality of bacteria is destroyed by protracted and violent shaking, which causes a disintegration of the cells. Many species are more quickly autolyzed after violent shaking. This fact is made use of in the production of bacterial vaccines. 7. Influence of Pressure. — Microorganisms in fluids which are sub- jected to great pressiue are for a time inhibited in their growth. When oxygen or nitrogen are used, the same moderate inhibition occurs. Osmosis. — Osmosis, due to differences of pressure between the medium and the microorganisms and to the permeability of the cell membrane for different substances is constantly occurring. Presum- ably the normal development of an organism takes place when the osmotic pressure within the cell is equal to (isotonic) that of its medium. When an organism is transferred to a new medium with an osmotic pressiu-e markedly different from that of the old one, decided changes in morphology may occur. If the difference is too great or the trans- fer is too sudden, death may result. If the new medium has a higher pressure, then water is abstracted from the cell and the protoplasm shrinks from its membrane. This is called plasmolysis. When the new medium has a lower pressure than the old, the cell may burst. This process is called plasmoptysis. 58 CHARACTERISTICS OP MICROORGANISMS Influence of Carbonic Acid under Pressure. — D'Arsonval and Charrin submitted a culture of B. pyoq/aneus to a pressure of fifty atmos- pheres under carbonic acid. At the end of four hours cultures could still be obtained, but the bacillus had lost its power of pigment pro- duction. A few colonies were developed after six hours' exposure to this pressure, but after twenty-four hours no development occurred. Other bacteria subjected to pressure have exhibited more resistance. We have subjected broth and milk containing typhoid, dysentery, diphtheria, arid colon bacilli to the gas under a pressure of 75 and 150 pounds. Within twenty-four hours 99 per cent, of those in the broth and 98 per cent, of those in the milk were destroyed. Within one week the broth was sterile and within four weeks the milk was sterile. Tubercle bacilli and staphylococci were much more resistant, but little effect was noticed in twenty-four hours. The results were the same whether the cultures were kept at 10° or 25° C. Bottled waters charged with carbonic acid are usually sterile. 8. EfEect of Drying. — For growth, microorganisms require much moisture. Want of water affects them in different ways. Upon dried culture media development soon ceases; but on media dried gradually at the room temperature (nutrient agar, gelatin, potato) they live often for a long time, even when there are no spores to account for their longevity. A shrunken residue of such cultures placed in bouillon has often been found, after a year or more, to yield living bacteria. The yeasts and molds are still more resistant. The question as to how long the non-spore-bearing forms are capable of retaining their vitality when dried on a cover-glass or silk threads has been variously answered. We know now that there are many factors which influence the retention of vitality; spores, of course, are more resistant than vegetative forms. The varying results sometimes reported by different observers may be explained by the fact that the conditions under which they were made were different, depending upon the desiccator used, the medium upon which the cultures were grown, and the use of silk threads or cover-glasses. In all these experiments, of course, it should be pre- viously determined that in spore-bearing species there are no spores present. Even when a dried culture lives for a long time the majority of the organisms die in a few hours after drying. We have found 1,500,000 colon bacilli to be reduced to 100,000 after three hours' drying. In tissues or exudates they resist drying much longer than when improtected. Encysted protozoa withstand long periods of desiccation. Most forms when dried quickly remain viable much longer than when dried slowly. Duration of Life in Pure Water. — When microorganisms which require much organic food for their development (and these include most of the pathogenic species) are placed in distilled water, they soon die- that is, within a few days. Their death is largely due to plasmoptysis. Even in sterilized well-water or surface water their life duration does PRODUCTS OP MWrOBAL GROWfH 59 not usually exceed eight to fourteen days, and they rarely multiply. Instances, however, of much more extended life under certain conditions are recorded. 9. Tactic EfEect of Chemicals.— Chemotaxis.— The deleterious effect of chemicals, especially those used as germicides, will be considered under Disinfection. Some chemical substances exert a peculiar attraction for micro- organisms, known as positive chemotaxis, while others repel them — ■ negative chemotaxis. Moreover, all varieties are not affected alike. Oxygen, for example, attracts aerobic and repels anaerobic bacteria, and for each variety there is a definite proportion of oxygen, which most strongly attracts. The chemotactic properties of substances are tested by pushing the open end of a fine capillary tube, filled with the substance to be tested, into the edge of a drop of fluid containing the organisms and examining under the microscope. We are able thus to watch the action of the microbes and note whether they crowd about the tube opening or are repelled from it. Among substances showing positive chemotaxis for nearly all microorganisms are peptone and urea, while among those showing negative chemotaxis are alcohol and many of the metallic salts. Such experiments are, of course, rough. The diffusion of the substances from the tube into the surrounding medium must play an extremely active role in the final result. PRODUCTS OF MICROBAL GROWTH. Microorganisms not only are acted upon by their surroundings, as has already been shown, but they themselves act, often markedly, upon these surroundings. We have spoken, under the Effect of Food (p. 51) of the great changes which may be produced in growths by slight changes in the food medium. So, many of the products, as noted below, are influenced to a greater or a less extent by environment. Production of Light. — ^Microorganisms which have the property of emitting light (photogens) are quite widely distributed in nature, par- ticularly in media rich in salt, as in sea-water. Many of these have been acciu-ately studied. The emission of light is a property of the living protoplasm of the organism, and is not usually due to the oxi- dation of any photogenic substance given off by them; at least only in two instances has such substance been claimed to have been isolated. While these organisms cannot emit light except during life, they can live without emitting light. They are best grown under free access of oxygen in a culture medium prepared by boiling fish in sea-water (or water containing 3 per cent, sea-salt), to which 1 per cent, peptone, 1 per cent, glycerin, and 0.5 per cent, asparagin are added. The power of emitting light is soon lost unless the organism is frequently trans- planted to fresh media. Thermic EfEects. — The production of heat by microorganisms does not attract attention in our usual cultures because of its slight amount, and even fermenting culture liquids with abundance of bacteria cause 60 CHARACtSRlSflCS OF MICROORGANISMS no sensation of warmth when touched by the hand. Careful tests, however, show that heat is produced. The increase of temperature in organic substances when stored in a moist condition, as tobacco, hay, manure, etc., is due, partly at least, to the action of bacteria. Chemical EfEects.— The chemical changes which take place in sub- stances as they are split up by microorganisms depend on the nature of the substances involved and the conditions under which they exist, and on the varieties of germs present. Chemists can as yet enumerate only some of the substances evolved, and describe, in but a few cases, the manner in which they were produced. The chemical activity may be divided into the following four types: (1) Production of sub- stances which help in some way the life of the cell. These substances may be secreted and retained within the cell, or liberated from it, e. g., ferments or enzymes ; true toxins. (2) Production of substances liberated by the bacteria as waste products. (3) Production of substances by the breaking down of the food media, e. g., putrefactive products, due largely to enzyme action. (4) The production of substances which help form the protoplasm of the bacterial cell itself. Fermentation. — Fermentation may be defined broadly as a chemical decomposition of an organic compound, induced by the life processes of living organisms (organized ferments or enzymes), or by chemical substances thrown off from the organisms (unorganized or chemical ferments or enzymes). It has been shown by Buchner and others that, in those cases of fermentation in which formerly it was believed the organized cell itself was necessarily concerned, ferments causing the same changes as the organized cells may be separated by various methods, such as crushing, filtering and so on. This brings fermenta- tion by unorganized and organized ferments very closely together, the one being a substance thrown off from the cell, the other a sub- stance ordinarily retained within the cell. The elaboration of both ceases with the death of the microorganisms producing them. Fer- mentation, therefore, requires the living agent or its enzyme. It furthermore demands the proper nutriment, temperatm-e, moisture, and the absence of deleterious substances. The enzyme itself is not markedly diminished in quantity after the fermentation ceases, though the process yields products that inhibit its action; hence fermentation ceases when the products are in excess, or when the nutriment is exhausted. That the process will often begin again after diluting the fermented medium, shows that the concentration of the harmful products plays an important part in the inhibitory action. The fact that the enzyme apparently does not bind itself to any of the end-products allies it to those chemical agents known as "katalyzers." Characteristics of Ferments or Enzymes. — Ferments are amorphous, non-dialyzable products of living protoplasm. They withstand mod- erate dry heat, but are usually destroyed in water solutions on exposure of ten to thirty minutes to a temperature of 60° to 70° C. They are injiffed by acids, but are resistant to all alkalis. They, even when present in the most minute quantities, are able partly to split up or CeHiaOe = 3C2H6O + 2CO2 Grape-sugar. 2 Alcohol. Or, 2 Carbon dioxide. CeHiaOe = , 2C3H6O3 3C2H402 Grape-sugar. ' 2 Lactic acid. 3 Acetic acid, PRODUCTS OF MICROBAL GROWTH 61 decompose complex organic compounds into simpler substances, and thus to render the foodstuff suitable for microbal growth. The enzymes may be grouped as sugar-splitting, inverting, fat- splitting, proteolytic, diastatic, rennin-like or lab enzymes, and oxydizing enzymes. Sugar-splitting Enzymes {Zymase Lactase, Maltase, etc.). — Many bacteria and yeasts are capable of splitting sugars, especially under anaerobi6 conditions. The action may be indicated by the following equations : Or, Fat-splitting Enzymes {Lipase) . — Little is known about this enzyme. It has industrial importance because of its action in rendering butter and other fats rancid. The Pr&ieolytic Enzymes. — ^The proteolytic enzymes which are somewhat analogous to trypsin — being capable of changing albumin- ous bodies into soluble and diffusible substances — are very widely distributed. The liquefaction of gelatin, which is chemically allied to albumin, is due to the presence of a proteolytic ferment, gelase. The production of proteolytic ferments by different cultures of the same variety of bacteria varies considerably — ^far more than is generally supposed. Bitter-tasting products of decomposition may be formed by cer- tain liquefying bacteria in media containing protein, as, for example, in milk. Diastatic Enzymes. — Diastatic ferments convert starch into sugar. This action is demonstrated by mixing starch paste with suitable cultm-es, then adding thymol and keeping the digestion for six to eight hours in the incubating oven; on the addition of Fehling's solution and heating, the reaction for sugar appears — ^the reddish-yellow precipitate due to the reduction of the copper. Inverting Enzymes. — Inverting ferments (that is, those which convert polysaccharides into monosaccharides) are of very frequent occurrence. Bacterial invertase is slightly less susceptible to heat than are some others ferments, and is produced in culture media free from protein. Rennin-like or Lab Enzymes. — Rennin-like ferments (substances having the power of coagulating milk and other liquid proteins with neutral reaction, independent of acids) are found not infrequently among bacteria. Oxydizing Enzymes. — Oxydizing enzymes or oxydases produce change by the addition of oxygen. A common example of this is the production of vinegar chiefly by B. aceti and B. pasteurianum. Alkaline Products and the Fermentation of Urea. — Aerobic bacteria always produce alkaline products from albuminous substances. Many species also produce acids from sugars, which explains the fact that 62 CHARACTERISTICS OF MICROORGANISMS neutral or slightly alkaline broth often becomes acid at first from the fermentation of the sugar contained in the meat used for making the media. When the sugar is used up, the reaction often becomes alka- line, as the production of alkalis continues. The substances produc- ing the alkalinity in cultures are chiefly ammonia, amine, and the ammonium bases. The conversion of urea into carbonate of ammonia by Micrococcus urece affords an example of the production of alkaline substances by bacteria: CO(NH2)2 + 2H2O = C03(NH4)2 Urea. 2 Water. Ammonium carbonate. Poisonous Products. — The poisonous products of microorganisms are considered in Chapter VII. Pigment Production. — Pigments have no known importance in con- nection with disease, but are of interest and have value in identifying bacteria. Very little is known of their chemical composition. Red and Yellow Pigments. — Of the twenty-seven red and yellow chromogenic bacteria studied by Schneider, almost all produce pig- ments soluble in alcohol and insoluble in water. The large majority of these pigments possess in common the property of being colored blue green by sulphuric acid and red or orange by a solution of potash. Though varying considerably in their chemical composition and in their spectra, they may be classified, for the most part, among that krge group of pigments common to both the animal and vegetable kingdoms known as lipochromes, and to which belong the pigments of fat, yolk of eggs, the carotin of carrots, turnips, etc. Violet Pigments. — Certain bacteria produce violet pigments, also insoluble in water and soluble in alcohol, but insoluble in ether, ben- zol, and chloroform. These are colored yellow when treated in a dry state with sulphuric acid, and emerald green with potash solution. Blue Pigments.— Blue pigments, such as the blue pyocyanin produced '/? by-B. pyocyanem; and the fluorescent pigment common to many so-called m. fluorescent bacteria (bacteriofluorescence) are examples. In cultures the pigment is at first blue; later, as the cultures become alkaline, it is green. Numerous investigations have been made to determine the cause of the variation in the chromogenic function of bacteria. All condi- tions which are unfavorable to the growth of the bacteria decrease the production of pigment, as cultivation in unsuitable media or at too low or too high a temperature, etc. The B. prodigiosus seldom makes pigment at 37° C, and when transplanted at this temperature, even into favorable media, the power of pigment production is grad- ually lost. B. pyocyanem does not produce pigment under anaerobic conditions. Occasionally colored and uncolored colonies of the same species of bacteria may be seen to occur side by side in one plate culture, as, for example, in the case of Staphylococcus pyogenes. PRODUCTS OF MICROS AL GROWTH 63 Reduction Processes. — The following processes depend wholly or in part upon the reducing action of nascent hydrogen: 1. Sulphuretted hydrogen (HaS). All bacteria, according to Petri and Maassen, possess the power of forming sulphuretted hydrogen, particularly in liquid culture media containing much peptone (5 to 10 per cent.); only a few bacteria form HjS in bouillon in the absence of peptone, while about 50 per cent, in media containing 1 per cent, peptone possess the property of converting sulphur into sulphuretted hydrogen, for which purpose is required the presence of nascent hydrogen. (For demonstration see Chapter IV.) 2. The reduction of blue litmus pigments, methylene blue, and indigo to colorless substances. The superficial layer of cultures in contact with the air shows often no reduction, only the deeper layers being affected. By agitation with access of air the colors may be again restored, but at the same time, if acid has been formed, the litmus pigment is turned red. 3. The reduction of nitrates to nitrites, ammonia, and free nitrogen. The first of these properties seems to pertain to a great many bacteria. Aromatic Products of Decomposition. — ^Many microbes produce aromatic substances as the result of their growth. The best known of these are indol, skatol, phenol, and tyrosin. Systematic investigations have only been made with regard to the occurrence of indol and phenol. Decomposition of Fats. — Pure melted butter is not a suitable culture medium for microbes. The rancidity of butter is brought about (1) as the result of a purely chemical decomposition of the butter by the oxygen of the air under the influence of sunlight, and (2) through the formation of lactic acid from the milk-sugar left in the butter. Fats are, however, attacked by bacteria with the consequent production of acid when mixed with gelatin and used as cultiu-e media. Putrefaction. — By putrefaction is understood in common parlance every kind of decomposition due to microbes which results in the pro- duction of malodorous substances. Scientifically considered, putre- faction depends upon the decomposition of albuminous substances, which are frequently first peptonized and then further decomposed. Typical putrefaction occurs only when oxygen is absent or scanty; the free passage of air through a culture of putrefactive bacteria — an event which does not take place in natural putrefaction — very much modifies the process; first, biologically, as the anaerobic bacteria are inhibited, and then by the action of the oxygen on the products or by-products of the aerobic and facultative anaerobic bacteria. As putrefactive products we have peptone, ammonia, and amines, leucin, tyrosin, and other amino substances; oxyfatty acids, indol, skatol, phenol, ptomains, toxins, and, finally, sulphuretted hydrogen, mercaptans, carbonic acid, hydrogen, and possibly marsh gas. Nitrifsdng Bacteria. — According to recent observations, nitrification is produced by a special group of bacteria, cultivated in the laboratory with difficulty, which do not grow on our usual cultiu"e media. From the investigations of Winogradsky it would appear that there are two 64 CHARACTERISTICS OF MICROORGANISMS common microorganisms present in the soil, one of which converts ammonia into nitrites and the other converts nitrites into nitrates. Conversion of Nitrous and Nitric Acids into Free Nitrogen. — This process is performed by a number of bacteria. The practical importance of these organisms is that by their action large quantities of nitrates in the soil, and especially in manure, may become lost as plant food by being converted into nitrogen. By the aid of certain root bacteria, which gain entrance to the roots of legumes and there produce nodular formations, the leguminous plants are enabled to assimilate nitrogen from the atmosphere. It is not known exactly how this assimilation of nitrogen occurs, but it is assumed that the zooglea-like bacteria, called bacteroids, constantly observed in the nodules, either alone or in a special degree, possess property of assimilating and combining nitrogen. It seems, moreover, to have been recently established that, independently of the assistance of the legumes, certain bacteria exist free in the soil, .which acciunu- late nitrogen by absorbing it from the air. These various nitrifying, denitrifying, and nitrogen-fixing bacteria are described in detail in the special chapter upon bacteria in nature. Formation of Acids from Carbohydrates. — Free acids are formed by many microbes in culture media containing some form of sugar or other fermentable carbohydrates, such as the alcohol mannite. The production of acid in ordinary bouillon takes place on account of the presence of sugar, which is usually found in small quantities in the meat. If after the sugar is consumed, not enough acid has been formed to kill the bacteria, the acid is neutralized gradually and in the end the reaction becomes less acid or even alkaline. Among the acids produced, the most important is lactic acid; also traces of formic acid, acetic acid, propionic acid, and butyric acid, and not infrequently some ethyl alcohol and aldehyde or acetone are formed. Occasionally no lactic acid is present. Formation of Gas from Carbohydrates and other Fermentable Substances of the Fatty Series. — The only gas produced in visible quantity in sugar-free culture media is nitrogen. If sugar is vigorously decomposed by bacteria, as long as piu-e lactic acid or acetic acid is produced there may be no development of gas, as, for instance, with the B. typhosus on grape-sugar, but frequently there is much gas developed, especially in the absence of air. About one-third of the acid-producing species also develop gas abundantly, this consisting chiefly of CO2, which is always mixed with H. Marsh-gas is seldom formed by bacteria, with the exception of those decomposing cellulose. (For demonstration see Chapter IV.) Formation of Acids from Alcohol and other Organic Acids.— It has long been known that B. aceti and allied bacteria convert dilute solu- tions of ethyl alcohol into acetic acid by oxidization: CHa + 02 = , CHs + mo. CH2OH COOH. PRODUCTS OF MICROBAL GROWTH 65 The higher alcohols — glycerin, dulcit, mannite, etc. — are also con- verted into acids. Finally, numerous results have been obtained from the conversion of the fatty acids and their salts into other fatty acids by bacteria. REFERENCES. Ambroz; Eutwickeluugszyklus des B. nitri n. sp., etc., Centralbl. f. Bakt., etc., I. Abt., orig., 1909, 51, 193 (with bibliography on structure and development of l)acteria). Buchner: Berichte d. Deutsch. chem. Gesellsch., xxx, 117-124 and 1110-1113. Buschke: Die SprosspUze in Kolle und Wassermann's Die Mikroorganismen, Jena, 2d edition, 1913. Calkins: The Protozoa, 1st edition. New York, 1901. Also article entitled The Protozoa, in Osier's Modern Medicine, Philadelphia, 1907, -vol. i; also Protozoology, New York and Philadelphia, 1909. Cramer: Arch. f. Hyg., xii to xxviii. Doflein: Lehrbuch der Protozoenkunde, 2d edition, Jena, 1909; Handbuch der pathogenen Mikroorganismen, Kolle und Wassermann, 2d edition, Jena, 1913. Jennings: Behavior in Lower Organisms, New York, Macmillan & Co., 1906. Lang: Protozoa in Vergleichende Anatomic der Wirbellosen Thiere, new edition, 1909. Lankester: Treatise on Zoology, 1st edition, London, Part I, first and second fascicles, 1909. Meyer: Flora, 1908, p. 95. Migula: System der Bakterien, Jena, 1897. Moore: The Pathology of Infectious Diseases of Animals, 4th edition, New York, 1916. Oppenheim: Die Fermente. u. ihre Wirkung, Leipzig, 1903. Penfold: Jour. Hyg., March, 1911, xi, 30-67. Petrtjschky: Die pathogenen Mychomyceten, in Kolle und Wassermann's Die Mikro- organismen, Jena, 2d edition, 1913. Plaut: Die Hyphenpilze in Kolle und Wassermann's Die Mikroorganismen, Jena, 2d edition, 1913. Rosenow: Tr. Chicago Path. Soc, July 1, 1913-, ix, 61; Jour. Inf. Dis., January, 1914, xiv, 1. Rtjzicka: Cytologic der sporenbildenden Bakterien, etc., Centralbl. f. Bakt., 1909, II. Abt., vol. xxvii. Schaudinn: Beitrage zur Kenntnis der Bakterien, etc.. Arch. f. Protistenk, 1902, i, 306,' and 1903, ii, 416. Vaughan: Herter Lectures, 1915. Zettnow: Romanowski's Farbung bei Bakterien , Zeitschr. f. Hyg., etc., 1899, xxx, 1; Centralbl. f. Bakt., 1900, I. Abt. xxvii, 803. CHAPTER III. THE MICROSCOPE AND THE MICROSCOPIC EXAMINATION OF MICROORGANISMS. THE MICROSCOPE. If lenses were capable of refracting all light equally, and bringing to a focus in one plane all rays proceeding from one plane in the object, the microscope would be a comparatively simple instrument. But simple lenses have several serious optical defects. 1. Spherical Aberration.^ — Points in the plane of the object are imaged on the curved surface of the spherical lens. This defect may be some- what diminished by combining convex and concave lenses, and by restricting the size of the field. Objectives corrected in this way are called aplanatic. 2. Chromatic Aberration. — This defect is due to the fact that the rays of light vary in their refraction according to their wave length (colors), e. g., the red rays have the longest focus and the violet the shortest. This is considerably corrected by combining planoconcave lenses of flint glass with biconvex lenses of crown glass — achromatic objectives. Still more of a correction is made by combining several different kinds of lenses with a lens of fluorite — apochromatic objectives. Monochromatic light may be employed and thus chromatic aberra- tion may be entirely avoided. 3. DifEraction. — Less luminous secondary images about the primary image, due to scratches or foreign particles or other defects may occa- sionally occur in the very best lenses. In order to understand fully the principle of the microscope, works on optics should be consulted. Different Parts of the Microscope (Figs. 8 and 9). — A complete instrument usually has four oculars, or eye-pieces (A) which are num- bered from 1 to 4, according to the amount of magnification which they yield. Nos. 2 and 4 are most useful for bacteriological work. The objective — ^the lens (5) at the distal end of the barrel — serves to give the main magnification of the object. For stained bacteria the xV achromatic oil-immersion lens is regularly employed; for photo- graphic purposes the apochromatic lenses are needed, although even here they are not indispensable. A yV lens may at times be useful, but hardly necessary; a No. 4 ocular and a yV lens give a magnification of about 1000 diameters (Fig. 10). For unstained bacteria we employ either the yV immersion or \ dry lens, according to the purpose for which we study the bacteria; for the examination of colonies when, as a rule, we do not wish to see individual bacteria but only the general THE MICROSCOPE 67 (Ffg^lir^ °^ ^^""^^ ^°''^'' ^^ ""'^ '^"'^' °^ """"^^ ^"''^'" magnification The stage (C)— the platform upon which the object rests-should be large enough to support the Petri plates if culture work is to be done, i he distance between the optical axis of the instrument and the pillar must be great enough to permit one to examine rather more than half the surface of the Petri dish without revolving it. The iris diaphragm -—Eye-lens "* __i. E _ Diaphragm > of the Ocular '- Field-lens .^-Draw Tube ,-Body Tube Draw Tube Diaphragm with Society Screw Coverglabs ,■?£££ Society Screw 'Mount ^^ Back-leus ,' j\Iiddle-lens ,- Front-lens ^- Working Distance of the Objective Fig. 8. — Microscope. Fig. 9. — Internal structure of the micro- scope. (D) opens and closes, and, like the iris of the eye, controls the amount of . light. Its opening is diminished or increased by moving a small arm underneath the stage. The reflector or mirror (E) placed beneath the stage serves to direct the light to the object to be examined. It has two surfaces — one concave and one plane. The concave surface must not be employed when the substage condenser is being used, otherwise the rays of light reaching the stage from the condenser will not be correctly focused. The concave surface may be used when unstained 68 MICROSCOPIC EXAMINATION OF MICROORGANISMS objects, such as colonies, or hanging drops are examined. At the same time the Abbe condenser should be lowered and the iris blender (D) regu- lated. The coarse adjustment (F) is the rack-and-pinion arrangement by which the barrel of the microscope can be quickly raised or lowered. It is used to bring the bacteria roughly into focus. If the bearings become loose, tighten the little screws at the back of the pinion box. Keep the teeth clean. If the bearings need oiling, use an acid-free lubricant, such as paraffin oil. The fine adjustment (G) serves to raise and lower the barrel very slowly and evenly, and is used for the exact study of the bacteria when high-power lenses are used. It is neces- sarily of limited range and delicate in its mechanism. If, when look- ing into the eye-piece, no change of focus is noticed by turning the micrometer head, or if the micrometer head ceases to turn, the adjust- ment has reached its limit. Eaise the barrel of the microscope by means of the coarse adjustment, then turn the micrometer back to Fig. 10. — Anthrax bacilli and blood cells. Fig. 11. — Colonies of diphtheria bacilli. X 1000 diameters. X 200 diameters. - bring the fine adjustment midway within its range. When the fine adjustment head stops, do not force it. For the microscopic study of microbes it is essential that we magnify the organisms as much as possible and still have their definition clear and, sharp. For this pur- pose the microscope should be provided with an oil-immersion system and a substage condensing apparatus. In using the oil-immersion lens a drop of oil (oil of cedar) of the same index of refraction as the glass is placed upon the face of the lens, to connect it with the cover- glass when the bacteria are in focus. There is thus no loss of light through deflection, as is the case in the dry system. If the lenses become dirty, they should be wiped gently with Japanese lens paper or a clean, soft, old-linen handkerchief. If necessary, breathe on the lens before wiping, and if this does not succeed, use a little xylol or chloroform. These substances are not to be used unless necessary. An immersion objective should always be cleaned immediately after THE MICROSCOPE 69 using. The objective should always be kept covered so as to prevent dust dropping in. Light. — ^The best light is obtained from white clouds or a blue sky with a northern exposure. Avoid direct sunlight. If necessary, use white shades to modify the sunlight. Artificial light has one advantage over daylight in that it is constant in quality and quantity. The Welsbach biu-ner and a whitened incandescent bulb give a good light. A blue glass between the artificial light and the lens is often of value. An eye shade may be helpful. Substage condensing apparatus {H) is a system of lenses situated beneath the central opening of the stage. It serves to condense the light passing from the reflector to the object in such a way that it is focussed upon the object, thus furnishing the greatest amount of luminosity. Between the condenser and the reflector is placed the iris diaphragm. Focussing. — ^Focus the body tube down by means of the coarse adjust- ment until the objective approaches very near to the cover-glass, being careful not to touch it. Then with the eye at the eye-piece focus up carefully with the coarse adjustment until the specimen comes plainly into view. Be careful not to pass this focal point. It is easily unnoticed if the light is too intense and the specimen thin and trans- parent. If the sliding tube coarse adjustment is used, focus carefully by giving the tube a spiral movement. When the object is brought fairly well into focus by means of the coarse adjustment, use the fine adjustment to focus on the partic- ular spot desired, for if this spot is in the centre of the field of the low power it should be somewhere in the field of the higher power. It is too much to ask of the maker that the lenses be made absolutely par- focal and centred. The delicacy of the centring can be appreciated when the magnification and the extremely small portion examined are considered. When the objectives are not thus fitted to the nose- piece, refocussing and again hunting up the object are necessary. In so doing we repeat the caution always to focus up before turning the nose-piece. When no revolving nose-piece is used, the change of objec- tives means the unscrewing of one and the screwing of the other into its place, and refocussing. The beginner should always use the low-power objectives and oculars first. The low-power objectives have longer working distances and are not so apt to be injiu-ed. They always show a larger portion of the specimen and thus give one a better idea of the general contour. After obtaining this general idea the higher powers can be used to bring out greater detail in any particular part. Generally speaking, it is best to use a high-power objective and low-power eye-piece in preference to a low-power objective and high-power eye-piece. In the latter case any imperfections in the objective are magnified unduly by the eye-piece, giving, as a rule, poor definition. Tube Length and Cover-glass. — All objectives are corrected to a certain tube length (160 mm. by most makers — Leitz, 170 mm.) and 70 MICROSCOPIC EXAMINATION OF MICROORGANISMS all objectives in fixed amounts of over 0.7 N. A. are corrected to a definite thickness of cover-glass as well (Zeiss, 0.15 mm., 0.2 mm.; Leitz, 0.17 mm.; Bausch & Lomb and Spencer, 0.18 mm.). These objectives give their best results only when used with the cover-glass and tube length for which they are corrected. As indicated in Fig. 9 the tube length extends from the eye-lens of the eye-piece to the end of the tube into which the objective or nose-piece is screwed. If a nose-piece is used, the draw tube must be correspondingly shortened. If the cover-glass is thinner than that for which the objective is cor- rected, the tube must be lengthened to obtain best results; if thicker, shortened. The more expensive objectives are provided with adjustable mounts by which the distances between the lens systems may be changed to compensate for difference of thickness of cover. They are successfully used only in the hands of an expert. One of them out of adjustment is worse than an ordinary objective. Dark-ground Illuinuiation and the Examination of Ultramicro- scopic Particles. — The apparatus constructed by Siedentopf and Zsigsmondy makes visible, even in solutions otherwise apparently homogeneous, very minute particles, which heretofore could not be seen even with the highest magnifications. Particles In (a milli- micron = one-millionth of a millimeter) are thus rendered visible. This increased power in microscopic analysis is made possible by intense (electric arc lamp) lateral focal illumination of the objects examined, and by shutting off the rays passing in the usual way through the tube of the microscope. The greater the difference between the refractive index of the objects coUoidally dissolved or otherwise held in suspension and the fluid which surrounds them, the brighter will be the appearance of the objects, and therefore the more readily visible. The microscopic field is dark; the objects which refract the light show as brightly illuminated, sharply defined pictures, in which the black margin corresponds to the contour of the object. The illuminated portion is surrounded by a fine dark zone, this in tm-n by alternate bright and dark zones, in which the illumination rapidly decreases. Reichert, of Vienna, has recently simplified this apparatus by devis- ing a new condenser. The light which illuminates the object has a greater refraction than the cone of light entering the objective which produces the image. Its advantages over the first method are: (1) It utiUzes the source of light better; (2) any dry objective can be used without alterations; (3) small particles are seen without the disturbing refraction rings. With this apparatus such living organisms as the Treponema pallidvm, and the fiagella on certain bacteria, which can scarcely be seen by ordinary microscopes on account of their low refractive indices, may be demonstrated with great clearness. The use of microphotography with ultraviolet light (according to A. Kohler) makes visible particles that cannot be seen by ordinary fight, because of the inability of the violet rays to pass through certain substances, e. g., chromatin. MICROSCOPIC METHODS 71 This method is said to increase 40,000 times our present limit of vision. The few discoveries claimed by these means for diseases of unknown origin have so far lacked sufficient corroboration to constitute them proved. Double Microscopes.— These have been devised by Metz and others for the piu-pose of making comparative studies of two objects. MICROSCOPIC METHODS. The direct microscopic examination of suspected substances for micro- organisms can be made either with or without staining. Unstained; the microbes are examined living in a hanging drop or on transparent solid media, under daylight, or better, artificial light, to note their number, their motility, their size, form, and spore formation, their general arrangement, their reactions to specific serums and to vital stains, etc. But for more exact study of their structure they must be stained in a dried film preparation on a glass slide or a cover-glass or when in tissues, in sections. Elimination of Foreign Organisms from Preparations. — Since germs are present in the air, in dust, in tap-water, on our bodies, clothes, and on all surrounding objects, it follows that when we begin to .examine substances for microbes the first requisite is, that the materials we use, such as staining fluids, cover-glasses, etc., should be practically free from organisms, both living and dead, otherwise we may not be -able to tell whether those we detect belong originally in the substances examined or only in the materials we have used in the investigation. Therefore, all solutions are filtered and all apparatus thoroughly cleaned and when necessary sterilized. Examination of Bacteria in the Hanging Drop. — For this examination special slides and methods are desirable. The slide used is one in. which there is ground out on one surface a hollow having a diameter of about J inch (Fig. 12). According to the pxirpose for which the hanging drop is to be studied, sterilization of the slide and cover-glass may or may not be necessary. IwiiiWPiiaiiiiwiiiiiiiias,;^^ iitj Fig. 12. — Hollow slide with cover-glass The technic of preparing and studying the hanging drop is as follows: The surface of the glass around the hollow in the slide is smeared with a little vaselin or other inert oil. This has for its purpose both the sticking of the cover-glass to the shde and the prevention of evaporation in the drop placed in the httle chamber, which is to be formed between the cover-glass when placed over the hoUow, and the slide. If the bacteria to be studied are in a fluid, we place a large platinum loopful upon the centre of the cover-glass and, to avoid drying, immediately invert it by means of a slender pair of forceps over the hollow in the slide, being very careful to have the drop over the centre of the cover-glass. The cover-glass is then pressed on the slide so as to spread the vaselin and make a perfect seal. 72 MICROSCOPIC EXAMINATION OF MICROORGANISMS If the bacteria are growing on solid media, or are obtained from thick pus or tissues from organs, they are mixed with a suitable amount of bouillon or sterile physiological salt solution' either before or after being placed upon the cover-glass. In studying living bacteria to determine only their grouping and motion we may use less magnification than for studying other charac-- teristics. In studying unstained bacteria and tissues we shut off as large a portion of the light with our diaphragm as is compatible with distinct vision, and thus favor contrasts which appear as lights and shadows, due to the differences in light transmission of the different materials under examination. It is necessary to remember that they are seen with difficulty, and that we are very apt, unless extremely careful in focussing, to allow the lens to go too far, and so come upon the cover-glass, break it, destroy our preparation, and, if examining pathogenic bacteria, infect the lens. This may be avoided by first finding the hanging drop with a low-power lens and then centring it. The edge of the drop is focussed more easily than the centre. The lens of higher magnification is now very gradually lowered, while at the same time gently moving the slide back and forth the slightest extent possible with the left hand. If any resistance is felt, the lens should be raised, for it has gone beyond the point of focus and is touch- ing the cover-glass. Hanging Mass or Hanging Block Cultures.— In order to study the morphology and manner of multiplication of individual microorganisms to better advantage than in the hanging drop, we have used hanging masses of agar, made by placing a large platinum loopful of melted agar on a sterile cover-glass and allowing it to harden, protected from dust. The organisms are placed on the free surface of this mass which is then inverted over a hollow slide and studied as in a hanging drop. Hill devised the following procedure: Melted nutrient agar is poured into a Petri dish to a depth of about i to I inch. When cool, a block is cut out about i inch square. The block is placed, under surface down, on a shde and protected from dust. A very dilute suspension of the growth to be exam- ined is then made in sterile bouillon and spread over the upper surface of the block. The slide and block are then put in the incubator for ten minutes to dry shghtly. A clean cover-slip is now' placed on the agar block in such a way as to avoid large air bubbles. The slide is then removed. With the aid of a platinum loop a drop or two of melted agar is run along each side of the block to fill any angles between it and the cover-glass. After drying in the incubator for five minutes it is placed over a hollow slide and sealed with paraffin. We consider the hanging-niass method better than that of the hanging block m many instances, because in the former method no pressure is exerted on the bacteria, and more oxygen is allowed them. Film Preparation (spread, smear).— Film preparation is made as follows: A very small amount of the blood, pus, discharges from mucous membranes, cultures from fluid media, or other material to be examined is removed, usually by means of a sterile swab or platinum loop, and 1 Physiological salt solution is usually 0.8 per cent. NaCl in distilled water. MICROSCOPIC METHODS 73 smeared undiluted in an even, thin film over a perfectly clean/ thin cover-glass or slide. From cultures on solid media, however, on account of the abundance of organisms in the material, a little of the growth is diluted by adding it to a small loopful of filtered or distilled water, which has been previously placed on the glass slide. It is best to add to the drop just enough of the culture to make a perceptible cloudiness. Blood films may be made either by the cover-glass or the slide method. To make a cover-glass preparation, two square, very thin (hence flexible) cover-glasses are cleaned. Holding one with thumb and index fingers by oppo- site corners, the tip of a drop of blood obtained .by needle puncture of finger or lobe of ear is made to touch the centre of the cover-glass, and the second clean cover-glass held similarly is allowed to fall upon the first one in such a manner that the corners do not coincide. The blood droplet spreads by capil- larity into a thin film, which is a sign to pull the two covers apart in the plane in which they he; good results depend upon cleanliness, rapidity, and success in sliding the two covers apart. To make a sUde film, the tip of the exuded blood drop is made to touch one slide near one end, and the edge of the second slide, held at an acute angle to the first one, is made to bisect the drop, which will spread at the point of con- tact by capOlarity across the slide. Upon pulling the second or spreading slide over the first slide, never changing the angle and applying gentle pressure, a thin layer of blood suitable for examination will be formed. A sHde made in this manner should be dried immediately by agitation in the air. It may then be fixed and stained in various ways. Milk films, after fixation, are cleared of fat by means of ether, xylol or alkaline solution.^ The film either is dried thoroughly in the air and then fixed with heat or any chemical fixative, or it may be placed in any of the fixatives while still moist. The usual fixatives are methyl alcohol, absolute ethyl alcohol, Zenker's solution, etc. (see p. 85). When fixed with heat, the glass is held by any one of the several kinds of forceps commonly used, and is passed three times by a rather slow movement through the Bunsen or alcohol flame. The film thus prepared is usually stained either by the simple addi- tion of a solution of an aniline dye, for from a few seconds to five minutes, 1 To render new cover-slips clean and free from grease, the method recommended by Gage is useful: Place in following solution over night: Bichromate of potash (K2Cr20i) 200 grams Water, tap or distilled 800 c.c. Sulphuric acid 1200 c.c. The bichromate is dissolved in the water by heating in an agate kettle; the sulphuric acid is added very slowly and carefully on account of great heat developed. After cool- ing, it is kept in a glass vessel. It may be used more than once. Glasses are removed the next morning and cleansed in running tap-water until the yellow color disappears. They are then placed in ammonia alcohol until used. When used, wipe with soft, clean linen or cotton cloth. If old cover-slips are used, boil first in 5 per cent, sodium carbonate solution. Another procedure is, after washing with soap and water and rinsing in water, to soak the cover-glasses in alcohol, then wipe with soft linen, then place in a Petri dish, and heat in the dry sterilizer for one hour at 200° C. to burn off fatty substances. The heating may be done by holding the cover-glass in the flame sufficiently to heat thoroughly with- out softening. A cover-glass is not clean when a drop of water spread over it does not remain evenly distributed, but gathers in droplets. 2 One-half or 1 per cent, sodium hydrate. 74 MICROSCOPIC EXAMINATION OF MICROORGANISMS or by one of the more complicated special stains described later. When the stain is to be hastened or made more intense, the dye is used warm. The cover-glass or slide, with the charged side uppermost, may either rest on the table or be held by some modification of Cornet's forceps. When the solution is to be warmed, the cover-glass may be floated, smeared side down, upon the fluid contained in a porcelain dish resting on a wire mat, supported oil a stand; or the solution may be poured on the glass which may then be held over the flame in the Cornet forceps. If a shde is used, it is simply inserted m the fluid or covered by it. The fluid both in the dish and on the slide should be carefully warmed so as to steam without actually boihng. The slide should be kept completely covered with fluid. After staining the film, the cover-glass or sUde is grasped in the forceps and thoroughly but gently washed in clean water and then dried, first between layers of filter paper and then in the air or high over a flame. If a Cover-glass has been used, a drop of balsam or water is placed on a glass slide and the cover-glass put upon it with the film side down. Films made on slides are usually unmounted. Cedar oil is added at the time of examination with the oil-immersion lens and washed off with xylol immediately after. Burn's Indian-ink Method of Demonstrating Bacteria. — In 1907, 1908, and 1909 Burri recommended the following method for isolating and studying single bacterial cells. A solution of India ink (fliissige Perl- tusche) in water 1 to 10 (better 1 to 4) is sterilized in test-tubes in the autoclave for fifteen minutes. A small drop of this ink is mixed care- fully with a drop of the fluid to be examined. If cultures from isolated cells are desired, the fluid should first be diluted so that a drop contains presumably a single organism; then drops of the mixture are placed in rows upon nutrient agar plates. If the bacteria are to be examined immediately, a drop of the mixture ^ink plus undiluted bacterial fluid) is allowed to dry upon a glass slide and then examined under an oil- immersion lens. The bacteria appear a briUiant white upon a dark field, particles of the ink surrounding the organisms like a capsule. This method is especially applicable for the demonstration of such - organisms as the Tr. pallidum which have poor staining qualities and a low index of refraction. Stains and Staining Methods for Microorganisms. — Protozoa stain in general as do animal cells. The protoplasm of mature bacteria reacts to stains much as does the nuclear chromatin of animal- cells, though the intensity of the staining varies somewhat with the con- dition of growth, such as the age, the species, the media, and so on. The best bacterial stains are the basic aniline dyes, which are com- pounds derived from the coal-tar product aniline (C6H5NH2).^ Aniline Dyes. — The aniline dyes which are employed for staining piu-poses are divided into two groups. In one the basic part of the molecule acts and the stains are spoken of as nuclear stains, since they color the nuclear chromatin of both cells and bacteria. In the other ' For a good description of the composition and action of the various stains, see A. B. Lee's Microtomist's Vade-Mecum, 7th edition, 1913. MICROSCOPIC METHODS 75 the staining act depends upon the acid part of the molecule, and the bacteria and cytoplasm of the higher cells are stained faintly. The stains in the latter group are used chiefly for contrast coloring. The basic dyes are usually employed as salts of hydrochloric acid, while the acid dyes occur as sodium or potassium salts. The following are the most commonly used basic aniline stains : Blue stains — ^methylene blue, thionin blue I (give the best differentiation; Red stains — basic fuchsin, safranin ) ^fficult to overstain). Brown stain — Bismarck brown (weak, may be used as counter-stain). Green stain — ^methyl green. Pink stain — eosin (weak; may be used as counter-stain). Violet stains — methyl violet, gentian violet, crystal violet (most intense stains; may overstain). These dyes are all more or less crystalline powders, and while some are definite chemical compounds, others are mixtures. For this reason various brands are met with on the market, and the exact duplication of stains is not always possible. Dyes should be obtained from rehable houses only. It is advisable to keep on hand stock saturated alcoholic solutions from which the staining solutions are made. These stock solutions are made by pouring into a bottle enough of the dye in substance to fill it to about one-quarter of its capacity. The bottle should then be filled with 95 per cent, ethyl alcohol, tightly corked, well shaken, and allowed to stand twenty-four hours. If at the end of this time all the staining material has been dissolved, more should be added, the bottle being again shaken and allowed to stand for another twenty-four hours. This must be repeated until a permanent sediment of undissolved coloring matter is seen upon the bottom of the bottle. This bottle wiU then be labelled "saturated alcoholic solution," of whatever dye has been employed. The dilution for use in staining is made by filling a small bottle three-fourths with distilled water, and then adding the concen- trated alcoholic solution of the dye', little by little, until one can just see through the solution. It is sometimes desirable to use a more concentrated solution with dyes such as methylene blue. Care must be taken that the color does not become too dense; usually about 1 part to 10 is sufficient. General Observations on the Principles of Staining Bacteria. — The staining of bacteria is not to be considered simply as a mechanical saturation of the cell body with the dye, in which the latter is dissolved in the plasma. It is rather a chemical combination between the dye substance and the plasma. ' The dependence of the staining process upon the solvent condition of the dye is shown in the following observations: 1. Entirely water-free, pure alcoholic dye solutions do not stain well. 2. Absolute alcohol does not decolorize bacteria, while diluted alcohol is an active decolorizing agent. The compound of dye substance and plasma is therefore insoluble in pure alcohol. 3. The more completely a dye is dissolved the weaker is its staining power. For this reason pure alcoholic solutions are inactive; and the so-called weak dye solutions to which strong dye solvents have been added are limited in their action on ceirtain bacteria in which the dye substance is closely united. This is the principle of Neisser's stain for diphtheria baciUi, viz., acetic-acid- methylene-blue solution. On the other hand, the addition of alkalis to the dye mixture renders the solvent action less complete and the staining power more intense. According to Michaels, however, in Loffler's methylene-blue solution the role of the alkali is pm-ely of a chemical nature, by which it converts the methylene blue into methylene azure (azure II). 76 MICROSCOPIC EXAMINATION OF MICROORGANISMS The dependence of the staining process upon the nature of the microbe is exhibited in the, following facts:. Certain microbes stain easily, others with difficulty. To the latter belong, for example, the tubercle bacUlus and lepra bacillus. Spores and fiagella also stain with difficulty. The easily stained objects require but a minimum of time to be immersed in a watery solution, while the others must be stained by special dyes with or without the aid of outside influences (heat, mordants, etc.) . The difficultly stained objects are at the same time not easily decolorized. The explanation of the resistance which these bacteria show to staining as well as to decolorizing agents is to be sought in two ways: either on the assump- tion that they possess a difficultly permeable or a resisting envelope, or that they have a special chemical constitution. The latter hypothesis holds good only, if at all, in regard to flagella and spores; while the assumption of the resisting envelope has reference more particularly to the tubercle bacillus, and is probably correct. The presence of fatty and waxy bodies in the envelope of these microorganisms is capable of demonstration. Moreover, after extrac- tion of these bodies by ether the tubercle bacillus loses its power of resisting acids, which peculiar resistance can also be artificially produced in other bacteria having normally no such resisting power. In many instances, doubtless, both of these causes, viz., resistant envelope and chemically different constitution, work together to produce the above-mentioned results. Elective staining properties, whereby certain species of organisms are exclu- sively or rapidly and intensely stained by certain dyes, have repeatedly been observed. Of the greatest practical importance in this respect is the Gram stain (see p. 78, and Chapter VI), used for the differential diagnosis of many species of bacteria; although a distinct classification of bacteria into those which are stained and those which are not stained by Gram's solution has been shown to be impracticable. There are some bacteria, however, which act uniformly toward Gram under most conditions; as, for example, the anthrax bacillus and the pyogenic cocci are always positive, the cholera and plague bacilli and goiiococci are always negative to Gram. Other species, again, are at one time stained and at another decolorized by Gram'; thus pyocyaneus is stained only in young individuals. Previous heating or extraction with ether does not prevent the action of Gram's stain, but treatment with acids or alkahes renders it impossible. Bacteria so treated, however, after one hour's immersion in Loffler's mordant regain their property of staining with Gram. As to the natiu-e of Gram's staining solution, it may be mentioned that only the pararosanihnes (gentian violet, methyl violet, and Victoria blue) are suit- able for the purpose, whereas the rosanihnes (fuchsin and methylene blue) give negative results. The reason for this is that the iodine compounds with the pararosaniUnes are fast colors, while those with the rosanihnes are unstable. These latter compounds when treated with alcohol break up into their con- stituents, the iodine is washed out, and the dye substance remaining in the tissues stain them uniformly; that is, without differentiation. But iodine- pararosaniUne compounds are not thus broken up and consequently stain those portions of the tissue more or less, according to the affinity which they have for the dye substance. The parts stained by Gram are thus distinguished from those stained violet, not only quantitatively, but qualitatively; it is not a gentian violet, but an iodine-pararosaniline staining which occurs. Mordants and Decolorizing Agents. — We have already noted that the protoplasm of unrelated microbes may respond differently to the several dyes. There is, however, seldom any difficulty in selecting a dye which will stain sufficiently to make microbal cells in pure cultures distinctly visible. When the microbes are imbedded in tissue or mixed in a film with blood or pus, it is frequently difficult to prevent the stain from so acting on the tissue or pus elements as to obscure the organises. MICROSCOPIC METHODS 77 Various methods are then employed to stain the germs more intensely than the tissues or to decolorize the tissue more than the organisms. Heating, the addition of alkali to the staining fluid and prolonging the action of the dyes increase the staining properties. We regulate these so as to give the best results. We also use mordants; that is, substances which fix the dye to the bacterial cell, such as aniline oil or solutions of carbolic acid on metallic salts. The decolorizing agents used chiefly are mineral acids, vegetable acids, diluted alcohols, various oils, and hot water. Formulae of the More Generally Used Staining Mixtures.— LorFLER's Methylene-blue Solution. — Saturated alcoholic solution of methylene blue, 30 c.c; caustic potash in a 0.01 per cent, solution, 100 c.c. (The alkali not only makes the cell more permeable, but also increases the staining power by liberating the free base from the dye.) Films stained two to five minutes, heated if more intense stain is desired. Sections stained one-quarter to several hours and decolorized until faint blue; contrast stain eosin; washed, dehydrated, cleared, and mounted (see p. 85). CABBOL-njCHSij^f , OR Zibhl-Neelsen Solution. — Distilled water, 100 c.c. ; carbolic acid (crystalline), 5 gm.; alcohol, 10 c.c; fuchsin, 1 gm.; or it may- be prepared by adding to 100 c.c. of a 5 per cent, watery solution of carbolic acid, 10 c.c. of a saturated alcoholic solution of fuchsin. The carbolic acid, like the alkali, favors the penetration of the stain. The last two methods, combined with heating, are used to stain spores and certain resistant bacteria as the tubercle bacilli and other "acid resisters," so that they retain their color when exposed to decolorizing agents (see below) . Carbol-gentian Violet. — One part of saturated alcoholic solution of gentian violet to 10 parts of a 2 to 5 per cent, solution of carbolic acid. Carbol-methylene blue, first used by Ktihne, consists of 1.5 gm. of methylene blue, 10 gm. of absolute alcohol, and 100 c.c. of a 5 per cent, solution of carbolic acid. Carbol-thionin consists of 10 parts of a saturated alcoholic solution of thionin and 100 parts of a 1 per cent, solution of carbolic acid. Polychrome Methylene Blub (Goldhorn).^-To prepare the stain dissolve 1 gm. lithium carbonate in 200 c.c. clean water and add 1 gm. methylene blue. Shake and dissolve. Pour into porcelain dish over water-bath, stirring fre- quently until blue color changes to a rich purple. Run through cotton in funnel; make up to 200 c.c. To 100 c.c. add 5 per cent, acetic acid until a faint pink is just visible on litmus paper above level of point discolored by the dye. Now add the remaining 100 c.c. of dye and allow to stand in open dish for forty-eight hours. Rim once more through cotton into clean bottle. It is not necessary to use distilled water, and satisfactory results are obtained with all the different forms of methylene blue tried. B-X (Griibler) is pref- erable. Fix the smear by immersion in commercial wood alcohol for fifteen to thirty seconds; wash well and stain for about ten to fifteen seconds in • polychrome; wash and stain for from fifteen to sixty seconds in 0.05 per cent. aqueous eosin. Wash again in water and dry in air without heat. Body of parasites blue; chromatin is red to purple. Results may be varied by using polychrome or eosin for different lengths of time. Admirable preparations may be obtained, even when there is pre- cipitation, by just rinsing the smear a little in 50 per cent, ethyl alcohol. This will remove any precipitation. Koch-Ehhlich Aniline Water Solution of Fuchsin or Gentian Violet is prepared as follows: To 98 c.c. of distilled water add 2 c.c. aniline oil, or, more roughly but with equally good results, pour a few cubic centimeters of saturated aniline oil into a test-tube, then add sufficient water nearly to fill it, In either cape th? jnixtures are thoroughly shaken and th?n filtered into a 78 MICROSCOPIC EXAMINATION OF MICRODROANISMS beaker through moistened filter paper until the filtrate is perfectly clear. To 75 CO. of the filtrate (aniline water) add 25 c.c. of the saturated alcoholic solution of either fuchsin, methylene blue, or gentian violet, or add the alcoholic solution until the aniline water becomes opaque and a film begins to form on the surface. Gbam's Stain. — Another differential method of staining which is employed is that known as Gram's method. In this method the object to be stained is floated on or covered with the aniline or carbolic gentian-violet solution described above. After remaining in this for a few minutes it is rinsed in water and then immersed in an iodine solution (Lugol's), composed of iodine, 1 gm.; potassium iodide, 2 gm.; distilled water, 300 c.c. In this it remains for from one to three minutes and is again rinsed in water. It is then placed in strong alcohol until most of the dye has been washed out. If the cover-glass as a whole still shows a violet color, it is again treated with the iodine solution, followed by alcohol, and this is continued until no trace of violet color is visible to the naked eye. It may then be washed in water and examined, or before examination it may be counter-stained for a few minutes by a weak solution of a contrasting dye, such as eosin, fuchsin, carmine, or Bismarck brown. This method is useful in demonstrating the capsule which is seen to surround some bacteria — particularly the pneumococcus — and also in differentiating between varieties of bacteria; for some do and others do not retain their stain when put in the iodine solution for a suitable time (see Chapter VI, for further remarks upon Gram's stain; see also p. 76). The Modifications of Gram's Stain are Many. — One only is given here. Nicolle's Modification. — Stain cold in carbol-gentian violet one minute; wash in tap- water; stain cold in the iodine mixture one minute; wash- in tap)- water; decolorize ten seconds in acetone (1 part) and alcohol (3 parts); wash in tap- water; counter-stain ten seconds in dilute carbol-fuchsin (1 to 10). Stain for B. Diphtherice. Neisser Stain. — The Neisser stain is carried out by placing the cover-slip smear of diphtheria or other bacilli in solution No. 1 for from two to three seconds, and then, after washing, in No. 2 for from three to five seconds. The bacilli will then appear either entirely brown or will show at one or both ends a dark blue, round body. With characteristic diphtheria baciUi, taken from a twelve to eighteen hours' growth on serum, nearly all will show the blue bodies (Fig. 116), while with pseudb types (Fig. 122), few will be seen. The solutions are as follows: No. 1. Alcohol (96 per cent.) 20 parts. Methylene blue (Qriibler) 1 part. Distilled water 950 parts. Acetic acid (glacial) SO parts. No. 2. Bismarck brown 1 part. Boiling distilled water 500 parts. Stain for B. PeHussis.—Toluidm blue, 5 grams; alcohol, 100 c.c. ; water, 500 c.c. Staining of Capsules.— Many methods of demonstrating the cap- sule have been devised. Three only will be given here. Welch's Glacial Acetic Acid Method is as follows:' (1) Cover the prepara- tion with glacial acetic acid for a few seconds; (2) drain off and replace with amhne gentian-violet solution; this is to be repeatedly added until all the acid IS replaced; (3) wash in 1 or 2 per cent, solution of sodium chloride and mount in the same. Do not use water at any stage. The capsule stains a pale violet. (See Plate III, Fig. 15.) Hiss's Copper Sulphate Method (Fig. 13). — The organisms are grown, jf possible, on ascitic fluid or serum media. If not, the organisms should be MICROSCOPIC METHODS 79 spread on the cover-glass mixed with a drop of serum, or better, with a drop of one of the diluted serum media. Dry in the air and fix by heat. The capsules are stained as follows: A 5 per cent, or 10 per cent, aqueous solution of gentian violet or fuehsin (5 c.c. satm-ated alcohoUc solution gentian violet to 95 c.c. distilled water) is used. This is placed on the dried and fixed cover-glass preparation and gently heated for a few seconds until steam arises. The dye is washed off with a 20 per cent, solution of copper sulphate (crystals). The preparation is then placed between filter paper and thoroughlv dried (Plate III, Figs. 14 and 10). Fig. 13. — Capsule stain by Hiss's method. Rhinoscleroma bacillus. X 1000. (Thro.) Huntoon's Method' (Fig. 1, p. 33). — ^A 3 per cent, solution of nutrose is cooked for 1 hour in an Arnold and tubed unfiltered after adding 0.5 per cent, carbohc. The organisms to be stained are mixed with a drop of nutrose, spread in thin film on glass shde and dried in air, not fixed. The stain is made up of 0.5 per cent, concentrated nitric acid, 1 per cent, of a 1 per cent, acetic acid solu- tion, 1 per cent, of alcoholic solution of basic fuehsin (or other stain), and 1 per cent, of carbol-fuchsin all in distiUed water. The stain is kept on the film for 30 seconds, washed in water, and dried. Staining Spores and Acid-fast Bacteria. — ^We have already rioted that during certain stages in the growth of a number of bacteria, spores are formed which refuse to take up color when the bacteria are stained in the ordinary manner. Special methods have been devised for causing the color to penetrate through the resistant spore membrane. In the simplest method a cover-sKp after having been prepared in the usual way is covered with Ziehl's carbol-fuchsin solution and held over the Bunsen flame until the fluid steams. This is continued for one or two minutes. It is then washed and dipped in a decolorizing acid solution, such as a 2 per cent, alcoholic solution of nitric acid, or a 1 per cent, solution of sulphuric acid in water, until all visible color has disappeared, then it is washed and dipped for one-half minute in a saturated watery solution of methylene blue. The bodies of the baciUi are blue and the spores red. This same method is also used for staining acid-fast bacilli. Sometimes the spores refuse to take the stain in this manner. We then can adopt Moeller's method, which is designed still further to favor the penetration of the coloring matter through the spore mem- brane. The prepared cover-slip is held for two minutes in chloroform, then washed off in water, and placed from one-half to three minutes in a 5 per cent, solution of chromic acid, again washed off in water, and now stained by car- > Personal communication 80 MICROSCOPIC EXAMINATION OF MICROORGANISMS bol-fuchsin, which is steamed for several minutes. The staining fluid is then washed off and the preparation decolorized in a 3 per cent, solution of hydro- chloric acid or a 5 per cent, solution of sulphuric acid. The preparation is finally stained for a minute in methylene-blue solution. The spores wUl be red and the body of the cells blue. The different spores vary greatly in the readiness with which they take up the dyes, and we have, therefore, to experi- ment with each variety as to the length of time it should be exposed to the maceration of the chromic acid. Even under the best conditions it is almost impossible to stain some spores. Spore Stain. — Himtoon has reported the following rapid and reliable method: 4 gm. acid fuchsin (Grlibler) dissolved in 50 c.c. 2 per cent, aqueous acetic acid. 2 gm. methylene blue (Grubler) dissolved in 50 c.c. 2 per cent, aqueous acetic acid. Mix the two solutions, shake and set aside for fifteen minutes. Heavy precipitate results. Filter mixture through weU-moistened filter paper. Use the filtrate for staining. Reddish-purple filtrate will keep for several weeks. Refilter if precipitate appears. To Stain. — Make a rather thick smear, preferably from an agar slant. Cover smear with dye and steam one minute. Wash in water. Film appears bright red. Dip slide in dilute solution of sodium carbonate (7 or 8 drops saturated solution in tumblerful of water) . When film turns blue, rinse immediately in water. Dry and examine. Spores are stained red, the body of the bacillus is stained blue. Plate III, Fig. 22, shows stained spores. The Hermann Stain foe Acid-east Bacilli. — ^A, crystal violet 3 per cent, solution in alcohol; B, ammonium carbonate 1 per cent, solution in water; mix 1 part of A with 3 parts of B just before using; steam three minutes, decol- orize with 10 per cent, nitric acid, wash in alcohol, and counter-stain in Bismarck brown. (See Plate VI.) Staining Flagella. — For the demonstration of flagella, which are possessed by all motile bacteria, we are indebted first to Loffler. The staining of bacterial flagella well is one of the most difficult of bacterio- logical procedures. In all methods, young (twelve-to-eighteen-hour) cultures of agar should be chosen. Enough of the culture to produce slight cloudiness is placed in a few cubic centimeters of filtered tap water in a test-tube. This may be used immediately, or allowed to stand in the thermostat at blood heat for from one to two hours to permit slight development. A tiny drop of this rather thin emulsion is allowed to spread with as little manipulation as possible over the cover-glass so that it may dry quickly. This latter point seems to be the important one, since slow drying allows the bacteria to shed their flagella. Bunge's modification of Loffler' s method is carried out as follows: Cover- glasses which have been most carefully cleaned are covered by a very thin smear. After drying in the air and passing three times through the flame, the smear is treated with a mordant solution, which is prepared as follows: To 3 parts of saturated watery solution of tannin add 1 part of a 25 per cent, solu- tion of ferric chloride. This mordant should be allowed to stand for several weeks before using. After preparing the cover-sUp with all precautions neces- sary to cleanliness, the filtered mordant is allowed to act cold for five minutes, after which it is warmed and then in one minute washed off. After drying, the smear is stained with the carbol-fuchsin or carbol-gentian violet solution, and then washed, dried, and mounted. (Plate III, Figs. 18-21.) Frequently the flagella appear well stained, but often the process has to be MICROSCOPIC METHODS 81 repeated a number of times. Overheating of the fihn prevents the staining of the flagella. The cell membrane may also show by this method. Van Ermengen's method gives good results. It is as follows: The films are placed for one hour at room temperature, or are heated for five minutes over a water-bath at 100° C. in the following solution: Solution A. Osmic acid, 2 per cent, solution ,1 part. Tannin, 10 to 25 per cent, solution 2 parts. Wash successively with water, absolute alcohol, and water, then place in the following solution for a few seconds: Solution B. 0.5 per cent, solution of AgNOa in distilled water. Without washing transfer them to a third solution: Solution C. Gallic acid 5 grams Tannin 3 grams Fused potassium acetate 10 grams Distilled water 350 c.c. After keeping in this for a few seconds, place again in Solution B until fihn begins to turn black. Then wash and examine. Eosin-methylene-blue Stains. — ^These are polychrome staining mixtures introduced by Nocht, Romanowsky and others chiefly for the staining of animal cells, but they are also useful in differentiating bacteria and other germs in tissues and exudates, especially in differentiating those organisms that take ordinary stains faintly, such as the spirochetes. They are fine differential stains for chromatin. Many modifications have been proposed. In a study of the essential constituents of the Romanowsky stain MacNeal says, both methylene azure and methylene violet are present and participate in the nuclear staining. The preparation of solu- tions directly from the pure dyes, methylene azur, methylene violet, methylene blue and eosin, has been recommended as the best manner of preparing these staining solutions as the proportion of the various constituents may be varied at will to obtain various kinds of differ- entiation. As a routine blood stain for study of leukocytes and staining of hematozoa, the following is recommended by MacNeal: Solution A. Methylene azur 0.3 Methylene violet (Bernthsen's, insoluble in water) 0.1 ^ Methylene blue 2.4 Methyl alcohol, pure 500.0 Solution B. Eosin, yellowish, water soluble 2,5 Methyl alcohol, pure 500.0 These solutions keep for at least a year. They are mixed in equal parts and diluted by the addition of 25 c.c. of methyl alcohol to each 100 c.c. of the mixture. This final mixture is employed in the same manner as Leishman's stain. It keeps for a few months. 6 82 MICROSCOPIC EXAMINATION OF MICROORGANISMS Giemsa's Method.— Smea,rs are fi^ed in neutralized methyl alcohol for one minute. . There are several variations of Giemsa's method. Two of them are given here : I. . Azur II — Eosin 3.0 grams Azur II 0.8 gram. Glycerin (Merck, chem. pure) 250.0 c.c. Methyl alcohol (chem. pure) 250.0 c.c. Both glycerin and alcohol are heated to 60° C. The dyes are put into the alcohol and the glycerin is added slowly, stirring. The mixture is allowed to stand at room tem- perature overnight, and after filtration is ready for use. The solution is prepared ready for use by Griibler, Leipzig. One drop of the stain to every cubic centimeter of distilled water, made alkaline by the previous addition of 1 drop of a 1 per cent, solution of potas- ■ sium carbonate to 10 c.c. of the water, is poured over the slide and allowed to stand for one-half to three hours. The longer time brings out the structure better^ and in twenty-four hours well-made smears are not overstained. After the stain is poured off, the smear is washed in running tap-water for one to three minutes, and dried with filter paper. If the smear is thick, the organisms may come out a little more clearly by dipping in 50 per cent, methyl alcohol before washing in water; then the washing need not be so thorough. By this method of staining, the cytoplasm of protozoa stains blue and the nuclear sub- stance a blue red or azur. Young bacteria usually take a dark purple stain, and their metachromatic granules an azur. II. 1.0 gm. azur I (Griibler) in 1000 c.c. distilled water. 0.8 gm. azur II (Griibler) in 1000 c.c. distilled water. 1.0 gm. French eosin in 100 c.c. distilled water. To mix for use. 9.0 c.c. azur I solution. 9.0 c.c. azur II solution. 0.15 CO. French eosin solution. 40.0 c.c. distilled water. Stain one and a half hours at incubator temperature. Wright's Stain. — One per cent, methylene blue (alcohol rectified), and 0.5 per cent, sodium carbonate are mixed and placed in a steam sterilizer for one hour. When cold, 0.1 per cent, solution of extra B-A eosin (500 c.c. eosin to 600 c.c. methylene-blue solution), is added until the mixture becomes purplish, and a finely granular black precipitate appears. This precipitate is filtered off and dried without being washed. A saturated solution of this is made in pure methyl alcohol. This is filtered and then diluted by adding to 40 c.c. of it 10 c.c. of methyl alcohol. In using, a few drops are placed on the fihn for a minute; then water is dropped on until a greenish irridesceiice appears. The stain then remains on for two minutes; then is washed off with distilled water, allowing a little to remain on until differentiation is complete. Dried with filter paper. Leishman's Stain. — This is a modification of Jenner's stain which is simply a solution of eosin and methylene blue in methyl alcohol. Instead of ordinary methylene blue, Leishman used the active constituents formed in this stain. Solution A. — To a 1 per cent, solution of medicinally pure methylene blue in distilled water add 0.5 per cent, sodium carbonate and heat at 65° C. for twelve hours, then allow it to stand ten days at room temperature. Solution B.— Eosin extra B-A (Griibler) 0.1 per cent, solution in distilled water. Mix Solution A and B in equal amounts and allow to stand six to twelve hours, stirring at intervals. Filter and wash the precipitate thoroughly. Collect, dry and powder it. 0.15 gram is dissolved in 100 c.c. of pure methyl alcohol to form the staining solution. It keeps perfectly for at least five MICROSCOPIC METHODS 83 months. To stain, cover the dried but unfixed film of blood with the staining solution. After thirty to sixty seconds add about an equal amount of distilled water. AUow this mixture to act for five minutes. Wash in distilled water for about one minute, examining the specimen mounted in water under the micro- scope. Blot, dry thoroughly, mount in balsam, or preserve the specimen as an unmounted film. Methods of StaMng Spirochetes. — Giemsa's method gives excellent results. Goldhom's Method, which also gives fine results, is a modification of Giemsa's. It is as follows: Dye; water, 200 cm.; lithium carbonate, 2 grams; methylene blue (Merck's medicinal or a similar preparation), 2 grams. This mixture is heated in a rice boiler with a moderate amount of heat until a rich polychrome has formed. This is determined by examining a sample against artificial light and noting the appearance of a distinctly red color. The solution is allowed to cool and the residue is removed by filtering through cotton. To one-half of this filtrate 5 per cent, acetic acid is gradually added until a strip of Utmus paper shows above the fine of discoloration a distinct acid reaction. The remaining haK of the dye is now added, so as to carry the reaction back to a low degree of alkalinity. A 0.5 per cent. French eosin solution is now added gradually, while the mixture is being stirred until a filtered sample shows a pale bluish color with slight fluorescence. The mixture is allowed to stand for one day and filtered. The precipitate is collected on a double filter paper and dried at a temperature not exceeding 40° C. It is then removed from the filter paper and dissolved in commercial wood alcohol. It is allowed to stand for one day in an open vessel and then filtered. To use the stain on smears sufficient dye to cover the smear is dropped on an unfixed preparation and allowed to remain for three or four seconds; the excess is then poured off. The slide is now introduced slowly into clean water with the film side down, is held there for four or five seconds and is then shaken in the water to wash off the excess of dye. It is then allowed to dry and is ready for examination. The paUidum stains violet. Silver Impregnation Method. — In Smears. — ^Until recently the demonstration in smears of the syphilis spirochete by the silver impregnation method, so suc- cessfully used by Levaditi in section, has been unsatisfactory. Stern, however, and Flexner corroborating him, have gotten beautiful results by the following simple method: (a) Air-dried in 37° incubator for some hours. (6) Ten per cent, aqueous silver nitrate for some hours (Flexner thinks three or four days' exposure better) in diffuse dayhght. (c) When the brownish color reaches a certain tone (easily recognized after experience) and when a metallic sheen develops, the slide is washed well in water, dried, and mounted. The blood cells are well preserved, they have a delicate dark brown contour, and contain fine light brown granules. The spirochetes are deep black on a pale brown and in places a colorless background. Other spirochetal organisms may be silvered by this method, but as they inay be (fifferentiated with greater difficulty than with Giemsa's stain, the latter should always be used as well. These organisms may also be demonstrated by the India-ink method (see p. 74). The flagella are brought out by LoSler's method or by the stain recommended by Goldhorn. In Sections. — Sections are prepared by the silver impregnation method of Levaditi (Levaditi and Manouehan, 1906) as follows: Fix small pieces of tissue 0.5 mm. in thickness for twenty-four to forty-eight hours in formahn, 10 per cent. Wash in 95 per cent, alcohol twelve to sixteen hours. Wash in distilled water until the pieces sink. Impregnate two or three hours at room temperature and four to six hours at 50° C. in the following fluid: Nitrate of silver, 1; pyri- dine, 10 (add just before using); distilled water, 100. Wash rapidly in 10 per cent, pyridine. Reduce the sUver by placing in tlie following mixture for several 84 MICROSCOPIC EXAMINATION OF MICROORGANISMS hours: Pyrogallic acid, 4; acetone, 10 (add just before using); pyridine, 15; dis- tilled water, 100. Harden in alcohol; xylol; paraffin. Levaditi's first method is longer but more rehable. Fix small pieces in formalin, 10 per cent. Harden in 95 per cent, alcohol. Wash in distilled water several minutes. Impregnate three to five days at 37° C. in 1.5 per cent, solution silver nitrate. Reduce twenty-four hours in: Pyrogallic acid, 4; formalin, 5; water, 100. Imbed in paraffin. By these methods the spirochetes appear densely black. Ross's Method of Examining a Large Quantity of Malarial Blood in One Film. — ^A large drop of blood (about 20 c.mm.) is placed on a glass shde and is shghtly spread over an area which can be covered by an ordinary cover- glass. This is allowed to dry in the air or it is warmed over a flame without heating it more than enough to fix the hemoglobin. The dry film is then covered with an aqueous solution of eosin (10 per cent.) and allowed to remain about fifteen minutes. This is then gently washed off and a weak alkaline methylene-blue solution is run over the film and left for a few seconds, when the preparation is again gently washed. After drjdng, it is ready for examination. The Eosin-methylene-blue Method Recommended by Mallory for Tissues may be Used for Smears as Follows. — The smears are fixed in Zenker's solution for one-half hour; after being rinsed in tap- water they are placed successively in 95 per cent, alcohol + iodine, one-quarter hour; 95 per cent, alcohol, one- half hour; absolute alcohol, one-half hour; eosin solution, twenty minutes, rinsed in tap-water; methylene-blue solution, fifteen minutes, differentiated in 95 per cent, alcohol from one to five minutes; and dried with filter paper. Staining Method for Negri Bodies {Williams's Modification of Van Giesen's Method). — Smears partially air-dried are fixed for ten seconds in neutral methyl alcohol to which 0.1 per cent, picric acid has been added. Excess of fixative removed by filter paper. Smears then stained in following solution : Saturated alcoholic solution fuchsin, 0.5 c.c; saturated alcohoHc solution methylene blue, 10 c.c. ; distilled water, 30 c.c. The stain is poured on the smear and held over the flame until it steams. The smear is then washed in tap-water and blotted with fine filter paper. The Negri bodies are magenta, the nerve cells blue, and the red blood cells yellow, or salmon color (Plate X, Fig. 1). This staining mixture may be kept in the ice-box for a long time. Heidenhain's Iron-hematoxylin Stain. — (a) Mordant and differentiating fluid: Iron oxyammonium sulphate, 2.5 grams; distilled water, 100 c.c. (6) Staining fluid: Hematoxylin, 1 gram; alcohol, 10 c.c; distilled water, 90 c.c. (To be kept in a red bottle and allowed to stand for about four weeks before using.) For use see chapter on Amebse. Preservation of Smears. — Dry stained preparations of bacteria keep indefinitely, but if mounted in Canada balsam, cedar oil, or dammar lac they tend gradually to fade, although many preparations may be preserved fOr many months or years. l3ry unstained spreads should be kept in the ice-box until stained. Examination of Microorganisms in Tissues.— Occasionally it is of impor- tance to examine the organisms as they occur in the tissues themselves. The tissues should be obtained s5on after death, so as to prevent as much as pos- sible postmortem changes, with consequent increase or decrease in the num- ber of microbes. Selected pieces of tissues can be frozen by ether or carbon dioxide and sections cut, but the best results are obtained when the material is imbedded in paraffin or in celloidin. Fixing and Hardening Tissues.— From properly selected portions small pieces, not larger than I inch by i inch, are removed and placed in one of the following fixatives : 1. Absolute Alcohol— Ahsohite alcohol for from four to eight hours, and longer if thicker. For the larger pieces it is better to change the alcohol after eight hours. The pieces of tissue should be, kept from falling to the bottom, as MICROSCOPIC METHODS 85 the higher layers of alcohol remain nearer absolute. If along with the micro- organisms one wishes to study the finer structure of the tissue, it is better to employ one of the other fixatives. 2. Formalin.~FoT fixing in formalin the tissue is put in 4 to 10 per cent, formalin solution for three to twenty-four hours, and then in successive strengths of alcohol. 3. Corrosive Sublimate. — Corrosive subhmate (saturated solution in 0.75 per cent, sodium chloride solution) is an excellent fixing agent. Dissolve the sublimate in the salt solution by heat, allow it to cool; the separation of crystals will show that saturation is complete. For pieces of tissue | inch in thickness four hours' immersion is sufficient; for larger, twenty-four hours may be necessary. They should then be placed in pieces of gauze and left in running water for from twelve to twenty-four hours, according to the size of the pieces, to wash out the excess of subhmate. 4. Svblimate Alcohol. — Hot subhmate (saturated) alcohol (50° C.) or satu- rated sublimate, to which 5 per cent, glacial acetic acid may be added. The preparation should remain in it a few seconds, then should be washed for one- half hour in 60 per cent, iodine-alcohol, and then placed in 70 per cent, alcohol. They may remain here for an indefinite time, until they are to be stained, when they are rinsed in distilled water and then p^laced in the staining fluid. 5. Osmic Add. — Two per cent, osmic acid (to be kept in a red glass with a ground-glass stopper). Moist smears are exposed to its fumes for a few seconds, small pieces for sections, four to eight hours, then carried through the various alcohols and xylol and mounted or imbedded in the usual way. 6. Hermann's Fluid. — ^A 1 per cent, solution platinum chloride, 15 c.c; a 2 per cent, solution osmic acid, 4 c.c; glacial acetic acid, 1 c.c. Moist spreads may be fixed for several minutes; very small pieces of tissue for twenty-four hours. 7. Zenker's Fluid. — ^Add to a solution of MuUer (potassium bichromate, 2 to 21 parts; sodium sulphate, 1 part; water, 100 parts) 5 per cent, of saturated sublimate solution and, when ready to use, 5 per cent, of glacial acetic acid. Moist spreads are fixed for one to five minutes, small pieces of tissue for three to twelve hours. They are then washed with water or put immediately into successive alcohols. To harden they are placed successively for twenty-four hours each in the following strengths of ethyl alcohol: 30 per cent., 69 per cent., and 90 per cent. Finally they are placed in absolute alcohol for twenty-four hours which dehy- drates them and they are then ready to be imbedded in paraffin. To imbed in paraffin, the pieces are put in (1) cedar oil imtil translucent; (2) cedar oU and paraflSn, equal parts, at 37° C. for two hours; (3) paraffin 52° C. two hours in each of two baths. They are then boxed ready for sec- tions. Sections are cut at 3 to 6m, and are dried at 36° C. for about twenty- four hours, protected from dust. Xylol may be used instead of cedar oil. The paraffin sections of tissue having been prepared and cut, they are ready for staining after the paraffin is removed. If all of the subhmate has not been removed by the water, the sections may be immersed in iodine-alcohol for ten minutes. Lofflee's Staining Method fob Sections. — The section is placed in Loffler's aUsahne methylene-blue solution for five to thirty minutes, decolor- ized for a few seconds in 1 per cent, acetic acid. It is then placed in absolute alcohol, xylol, and Canada balsam. The number of seconds during which the preparation remains in the acetic acid must be tested by trials. REFERENCES. BuBEi: Das Tuscheverfahren als Einfaches Mittel, etc., Jena, 1909. Giemsa: Deutsche med. Woch., 1905, xxxi, 1026. Hiss: Jour. Exp. Med., 1905, vi, 317 86 MICROSCOPIC EXAMINATION OF MICROORGANISMS Huntoon: Jour. Am. Med. Assn., May 2, 1914; Ixii, 18, 1397; also, Proc. Soe. Bacteriologists, Dec, 1916. Kohleb: Ztsoh: f. wiss. Mikros., 1904, xxi, 129. Lea: Microtomist's Vade-mecum, 1913, 7th ed. Leishman: Brit. Med. Jour., 1901, p. 635; 1902, p. 757. Levaditi and Manouelian: Compt. rend. Soc. biol., 1906, Ix, 134. Metz: Ztschr. f. wiss. Mikroskop., 1913, xxx, 188. Neisseb: Ztschr. f. hyg., 1897, xxiv, 443. Reichbkt: Jour. Roy. Micr. Soc, 1907, p. 364. Weight: Jour. Med. Res., 1902, ii (New Series), 138. MacNeal: Jour. Infect. Dis., 1906, iii, 412. Williams: Am. Jour. Pub. Hyg., 1908, xviii. No. 1. CHAPTER IV. GENERAL METHODS USED IN THE CULTIVATION OF MICROORGANISMS. CULTURE MEDIA, METHODS OF ISOLATION AND CULTIVATION, CULTURAL CHARACTERISTICS. The methods employed for the artificial cultivation of microorganisms are of fundamental importance. By their use we can obtain one variety growing apart from all others, namely, in pure culture. This pure culture may be planted on various media and the morphological, biochemical, and cultural characteristics studied for classification and identification. It is evident that all glassware and instruments used must be free from other microorganisms; that is, they must be sterile. Preparation of Glassware. — ^Various types of glassware, such as test-tubes, flasks, bottles, beakers, pipettes, etc., are used (see Figs. 14 to 22). New glass- ware, as a rule, only requires mechanical cleansing with soap and water, loosening adherent dirt with test-tube or bottle brushes. Old glassware con- taining cultures should be sterihzed either in the autoclave or in a covered boiler. In the latter case the tubes are covered with water, about 5 per cent, of washing soda or soap added, and boiled for one hour. Any solid medium present is melted and the glassware can then be washed according to the directions already given. After washing, drain and allow to dry. Neutralization of Glassware. — ^Where sUght changes of reaction are important the glassware should be neutraUzed after washing. Soak in 1 per cent, hydro- chloric acid for several hours or boil for one-half hour, wash free of acid, and rinse in distilled water. New glassware, especially the cheaper grades, is most likely to give off free alkali. The clean glassware is plugged loosely with ordinary non-absorbent cotton. The cotton should not be twisted in, as creases will form along the glass, leav- ing channels for contamination. Either fold into a plug or take a square of cotton, fold one corner, place a rod against the fold, and push the cotton into the neck of the container or tube. A sufiicient amoimt of cotton should pro- ject for handling and the plug should be just tight enough to allow one to lift the container by means of the plug. Several thicknesses of filter paper may be used to cover beakers and other wide-mouthed containers. Sterilization of Glassware. — ^AU glassware is sterihzed by dry heat. This is done after plugging. Some type of hot-air sterilizer is used (Fig. 23). Heat to 160° C. for one hour. Leave the sterilizer closed when the sterilization is finished, so that cooling is gradual or the glassware may crack. This heating not only sterihzes but sets the shape of the cotton plugs. CULTURE MEDIA. General Considerations. — Most microorganisms causing disease re- quire complex foodstuffs similar in constitution to those in the animal body. The general basis of media for these types is an extract or infusion of meat. To this may be added peptone and sodium 88 THJE CULTIVATION OF MICROORGANISMS chloride. Some may require uncoagulated proteins, such as serum or blood, or even fresh tissues. Carbohydrates may be added. The non- pathogenic microorganisms vary in their ability to grow on these more complex media, and some will only grow on simple media containing Fig. 14. — Test-tubes can be also used with- out lip. Average size, 6 x f inches. Fig. 15. — Beakers. inorganic salts. Media may be either fluid or solid. In the latter case there is added some jelly substance such as gelatin or agar or an albumin which is coagulated by heating. J5'C 1000 re Fig. 16. — Globe flask. Fig. 17. — Volumetric flask. Fig. 18. — Erlenmeyer flask. Certain technical methods are employed in the preparation of nearly all culture media, such as adjustment of reaction, clearing, filtering, etc. These details must be understood before the actual preparation of media is given. CULTURE MEDIA 89 Reaction of Media. — Titration and Adjustment. — A moderately alkaline reac- tion to litmus is satisfactory for the growth of most pathogenic microorganisms. For ordinary work the usual acid reaction of the mediar in preparation may be reduced by the addition of a 4 per cent, sodium hydrate solution until red litmus paper is turned slightly but distinctly blue and blue litmus paper remains the same color. Litmus, however, is not a deUcate indicator, and mmp-''^^'r:^ Fig. 19.— Type of bottle commonly used for dilutions, etc., or as substitutes for flasks. Fig. 20.— Blake bottle. Laid on its flat side it gives a large surface of broth or agar. Fig. 21.— Petri dish. Com- monest size is 10 c.c. in diameter. phenolphthalein is generally used, giving us more accurate knowledge of the reaction. The neutral points of the two indicators are different, so that media which are alkaline to litmus are still acid to phenolphthalein. The materials necessary for titration and adjustments of the reaction to phenolphthalein are normal and twentieth normal solutions of sodium hydrate and of hydrochloric acid, a 0.5 per cent, solution of phenolphthalein in 50 per cent, alcohol, burettes, casserole, and stirring rod. Fig. 22. — Types of fermentation tubes. Methods of Titration. — ^Two methods may be used: (a) room temperature and (6) boiling temperature. The former is the more accurate if the medium to be tested has previously been heated to the boiUng-point during its preparation. Under these circumstances with this method the reaction of a medium is set at a temperature more nearly approximating that at which the substance will be used, 37° C. of the incubator. It is essential, however, to use method b §0 THE CULTIVATION OF MICROORGANiSMS boiling temperature, when titrating meat juice which has not been heated above 50° C. (or up to the boiling-point) for the purposeof dissolving added peptone and salt. The boiling in the casserole is needed to approximate the later conditions when the medium is boiled and sterilized. This method of titration as given below is the standard method. It is often misapplied, however, when used for the titration of media already heated to the boiling-point. Another error in its results is due to the fact that the more phenolphthalein is heated, the less sensitive and accurate it is as an indicator. Technic. — Room temperature method: 5 c.c. of medium to be tested, 45 c.c. of distilled water^ and 1 c.c. of the phenolphthalein solution are mixed in a cas- serole. (When agar is titrated, the water should be warmed to about 30° C. before adding the hot agar.) If no pink color is present, the medium is acid. FiQ. 23. — Hot-air sterilizer. Lautenschlagetform. While the mixture sibeing stirred twentieth-normal solution of sodium hydrate (^0 NaOH) from a burette is run in until a delicate pink tinge is observed. This color should not disappear with stirring. (See page 91 for the calcula- tion for correction.) Boiling Temperature. — Standard method: 5 c.c. of medium to be tested and 45 c.c. of distilled Water are mixed in the casserole and boiled for one minute, then 1 c.c. of the phenolphthalein solution is added. If no color is present the medium is acid and while hot the twentieth-normal solution of sodium hydrate (^^ NaOH solution) is run in from a burette until a faint but distinct pink color appears. This color must remain on stirring, otherwise more alkali 1 Distilled water, freshly boiled and cooled, should be used for very accurate results, since the CO2 has been driven off by the boiling and is not reabsorbed within a few hours. CULTURE MEDIA 01 IS needed. l''rom the amount added we determine the acidity of the medium and estunate how much normal (?) solution of NaOH must be added to obtain the reaction desired, for example: Five c.c. required 2.4 c.c. of f^ NaOH to neutrahze, therefore 100 c.c. (twenty times as much) would require 2.4 c.c. of ? NaOH (twenty times as strong), in other words, the medium is 2.4 per cent, acid to phenolphthalein (,+2.4 per cent.). Assuming we desire a reaction of +1 per cent., we must then add 2.4 c.c. —1 c.c. or 1.4 c.c. of » NaOH to every 100 c.c. of medium or 14 c.c. to a hter. /'\ ^ — , =23 i24 ^25 Outflow 67 Sec 44.S DELIVERS 50 CC 20 °C Outflow 43 Set. Fig. 24 Graduated Volumetric pipette. pipette. Obtainable in various sizes and graduations Fig. 25 Burettes. Most convenient type has a blue line against a white background on back, giving sharper readings. Should, on the other hand, the inixture in the casserole show a pink color then the medium is alkaline and ^ HCl is used, heating as above, until only a faint pink color persists. If we use 0.5 c.c, then the medium is 0.5 per cent, alkaline (—0.5 per cent.), and it will require 0.5 c.c. of ^ HCl for every 100 c.c, or 5 c.c. for every hter to bring it to the neutral point. But we want the reaction to be +1 per cent, or 1 per cent, acid, we therefore add 5 c.c. plus 10 c.c. or 15 c.c. of ^ HCl to each liter. After the first correction of the reaction by the addition of normal alkali or acid the medium should be again titrated after further boiling or treatment in 92 THE CULTIVATION OP MICROORGANISMS the autoclave. Additional correction will usually be necessary, for each time meat infusion is boiled, or heated under ipressure, the reaction rises some- what in acidity. This may have to be repeated several times before the reac- tion desired is obtained. Various combinations, precipitation, etc., occur which make this necessary. After experience a suitable excess of normal soda solution may be added to allow for the rise in acidity which will occur by the time the final sterilization is ended. This rise is not so marked in media pre- pared with beef extract (Liebig's) as a basis. The reaction of media should always be adjusted before filtration, and after adjustment the medium must be heated for three to five minutes or precipita- tion will occur on subsequent sterilization, especially if sterilized in the auto- clave, due to the fact that the temperature is higher than that employed during the adjustment of the reaction. Clearing Media. — This is done by coagulation of an albumin, which, as it coagulates, enmeshes the fine particles. In certain methods the coagulation of the soluble albiunins in the watery extract of the meat clears the medium when heated. Under other conditions an albumin such as egg is added. If the medium is hot, it must first be cooled to below 60° C. One or two eggs are used for each liter. They are broken in a small pan and mixed with a small quantity of water by means of an egg-beater. This is then added to the medium, stirring thoroughly. The medium is then heated usually in the Arnold sterilizer or autoclave, to coagulate the egg albumens. Filtering Media. — For fluid media, paper or cotton is used. For media which sohdify on cooling, cotton is preferable. Where media are cleared, the coagulated albumins which settle on the paper or cotton act as a filtering medium. If paper is used, it should be folded or a corrugated funnel employed. If cotton is used, a spiral of wire or a perforated porcelain plate is placed in the funnel to support the cotton and two strips of cotton laid crosswise in the funnel, the torn ends extending up the side of the funnel. Before filtration the paper or cotton is wet with water so that any h fat in the medium will not pass through. In wet- ting the cotton, moisten the torn ends so they adhere to the glass and then carefully moisten the rest. The medium is poured into the funnel along a glass rod, which is helpful in holding the cotton in place when this is used. The filtrate is poured back until it comes through clear. In the case of media which sohdify on coohng, the funnel should be heated before adding the medium and the medium awaiting filtration kept hot. The use of a hot-water funnel to keep the funnel and its contents hot is a great advantage. Failing this, the flask and funnel may be placed in an Arnold sterilizer. Preparation of Media for Use.— After filtration the medium can be either placed in flasks for storage or placed in various containers for immediate use and sterilized. In filling, care should be taken not to wet the necks of the containers or the cotton plugs will stick. Test-tubes and other small con- tainers are best filled from a funnel with a stop- cock or fitted with rubber tube and glass tube with Fig. 26.— Hot-water funnel: pinch-cock. The amount in each container will a, point heat; 6, inlet for depend On the use to which it is to be put. For the Winter. ordinary 15 cm. (6-inch) test-tube a depth of media of about 3 to 5 cm. is sufficient. If plates are to be poured, at least/10 c.c. should be placed in a tube. If the medium is to be slanted in the tube, less is required. The slants may be made immediately after steri- hzation or the media melted and slanted as required. To make the slants the tubes should be laid in a row on a table with a glass rod or strip of wood to CULTURE MEDIA 93 raise the upper end. Care should be taken that the slant does not touch the cotton, and sufficient media be present in the bottom of the tube ("butt") so that on raising the tube the slant will not fall down. The medium should be well set before raising. Fig. 27 Fig. 28 Fig. 29 Types of filters. Fig. 30 Sterilization of Media. — Sterilization may be accomplished bv filtration through various types of filters which do not allow the passage of bacteria, by heat, and occasionally by chemicals, as chloro- form. Sterilization bj'' heat may be accom- plished in several ways — namely, intermittent (fractional) steriUzation at 100° C. or at 60° to 80° C, and sterilization by steam under pressure. Filtration. — This method is resorted to where the application of heat will inj ure the solution, as in the case of some sugars, or where the necessary amount of heat would cause coagulation, as in the case of tissue extracts or serum. Various types of filters made of unglazed porcelain or compressed diatomaceous earth are used. Types of these filters are the Berkefeld, Pasteur, Chamberland, and Doulton filters. The bacteria are held back because of the fineness of the pores. Various grades of fineness are procurable in some brands, and on this depends the rapidity of filtration. In any case the permeability of the filter should be tested before use by filtering a broth culture of some small micro- organism, such as Bacillus prodigiosm. Before filtration the fluid to be filtered should be rendered as clear as possible by filtration through paper or if necessary through paper pulp. The latter is prepared by soaking torn-up Fig. 31. — Filter pump for attach- ment to water faucet. 94 THE CULTIVATION OF MICROORGANISMS filter paper in water and then placing a layer in a porcelain filter, over a per- forated porcelain plate, draining off the water and packing the layer tight. The depth of the layer will depend on the fluid to be filtered, its density, and the finexiess of the particles to be removed. This should be fastened in the neck of a filtering flask and suction applied to hasten the filtration. Similar results may be obtained by filtration through sand. The filtration through, a filter candle (the size selected will depend on the amount to be filtered) is accelerated by the use of suction or pressure. Suction may be secured by the use of a filter pump attached to the hydrant or other types of suction apparatus. Pressure may be obtained by air under pressure from an installed system or by a hand pump and cylinder to equalize the pressure. A manometer is employed to- determine the pressure. The accompanying cuts show the method of setting up the apparatus (Figs. 27-31). Fig. 32 Figs. 32 and 33. — Arnold steam sterilizer. Fig. 33 Two types. The filter candle and aU attachments with which the fluid will come in con- tact after passage through the candle must be sterile. The glassware can be sterihzed in the hot-air oven and the filter candle and rubber connections can be sterilized by boiling for one hour or by steam under pressure. (See Auto- clave.) After use the filter candle should be freed of all soluble material, especially coagulated matter, by running through it an abundance of clear water. If used for infective matter, it can then be sterilized by boiling. In' any case the surface should be lightly scrubbed with a fine brush after use. A new filter should be cleansed before use by filtering clear water. It should then be placed in cold water and boiled thoroughly. After continued use the candles gradually become clogged. They can be renewed in some degree by careful heating to glowing in an oven. This is apt to produce fissures, hence the filter should be retested before use. Heat. — Intermittent {"Fractional") Sterilization. — This method has two divisions : 1. That of applying heat at a temperature' of 100° C. or close to it, as in the Arnold sterilizer, on three or more successive days, to media containing special CULTURE MEDIA 95 sugars, etc. The daily period of heating depends upon the size of the con- tamer; test-tubes of media require twenty minutes, whereas 1- to 2-Hter flasks require forty-five to sijcty minutes to allow heat to penetrate. Fig. 34. — Upright type. Fig. 35. — Horizontal type. Figs. 34 and 35. — Autoclaves may be heated by direct application of heat, or by steam under pressure, when available. 2. The application of temperatures of only 60° to 80° C. on successive days to such substances as blood serum, transudates from body cavities (e. g., ascitic fluid), etc. 96 THE CULTIVATION OF MICROORGANISMS The principle underlying fractional or intermittent sterilization is as follows: All bacteria when free from spores are killed by exposure for one or two hours to only 60° to 80° C. heat. If the material is left overnight at about 22° C, the spores, if present, will develop to bacteria and be destroyed by the second heating. Some of the bacteria may produce spores before the second heating, or some of the spores may be slow in developing and escape. For this reason a third heating, or with the low temperatures, five or six heatings on successive days may be necessary. When this method is applied to albuminous material before coagulation, such as sera, tissue extracts, etc., the temperature selected should be below the coagulation-point of the material, usually about 60° C. AH material to be treated in this manner should be as free from contamination as possible. The method has its best application to originally sterile material which, because of necessary manipulation may have become slightly contami- nated. The heating may be done in a water bath or in a water-jacket oven. Sterilization by Steam under Pressure. — This is done by means of an auto- clave. Various forms are available which may be heated by gas or by steam (Figs. 34 and 35). All air should be displaced by the steam before closing the vent. Too rapid reduction of pressure will cause the media to boil up and wet or even blow out the plugs. The time and pressure employed will depend on conditions. As a rule 15 pounds' pressure is employed and twenty minutes are sufficient for media in test-tubes. Flasks should be heated one-half to -one hour, depending on their size. Lower pressures are at times desirable where overheating is a factor, and the heating may be repeated on two suc- cessive days. The heating should be timed from the time the pressure desired is reached. The temperature will vary with the pressure thus: 5 pounds' pressure = temperature 108.8° C. 10 " " = " 115.6° C. 15 " " = " 121.3° C. Storage of Media. — Media after sterilization may deteriorate for two reasons, namely, contamination (especially by molds penetrating the cotton stoppers), and by drying. The former may be avoided by keeping the media at a low temperature, 40° to 45° F., in a dry refrigerator. Where media are used in small amounts they may be kept in flasks and tubed as needed. Evaporation of the tubed media can be lessened by dipping the plugs in paraffin or by the use of rubber caps made for this purpose. The stock flasks after removal from the steriUzer may be sealed with paraffin or sealing wax. The plug should be cut off and pushed in slightly and melted paraffin poured on it, care being taken that it does not run through the plug into the media. This can be avoided if the paraffin is nearly cold enough to set. Seahng wax may be applied in the same way or rubber caps or rubber tissue tied over the neck. If the stock flasks are capped, a small pledget of cotton moistened with bichloride of mercury solution may be placed between the plug and the cap to discourage the development of molds which may adhere during manipulation. The neck of the flask should be wiped free of the bichloride before pouring out the media. This precaution is specially useful for media stored in flasks at room temperature. The necks of all containers may be protected from dust by covering with paper before sterilization or inverting tumblers over the necks of flasks. COMPOSITION OF CULTURE MEDIA. Meat Infusion.— One pound of finely chopped meat usually beef or veal, is macerated with 1000 c.c. of water and placed in the ice-chest for eighteen to twenty-four hours, or it may be extracted by heating to a temperature not exceeding 50° C. for one hour. The infusion is then strained through cheese- cloth and the meat juice squeezed out by twisting the cloth or by means of a meat press. The fluid contains the soluble albumins, extractives, salts carbo- COMPOSITION OF CULTURE MEDIA 97 hydrates, and coloring matter of the meat. This forms the basis for the various media to be described. If this infusion is to be stored as such, it must be boiled to coagulate the albumins (which also clears it), filtered, placed in flasks, and sterilized. Many advise the neutralization to phenolphthalein^ before heating, on the ground that less of the foodstuff is precipitated than by heating the very acid infusion. As a substitute for the meat 2 to 5 grams of Liebig's extract of beef may be used for each liter of water. When media are made with this basis, they are spoken of as meat-extract media in contradistinction to meat-infusion media. Nutrient Bouillon.— /n/imom Bror.,TT^T,o -1 - ., 1 i.,T 1- ,. . . of mold colonies. forty-eight hours, and growths at 2U° C. not older than torty-eight AMORPHOUS, without visible dillcrcntiation in structure. NAPITORM, liquefaction with the form of a turnip. to seventy-two hours. To secure uniformity in cultures, in all eases ARBORESCENT, a branched, tree-hke growth. NITROGEN REQUIREMENTS, the necessary nitrogenous food. P/elin.inary cultivation shall be practi^sed as described ^ BEADED, in stab or stroke, disjointed or semiconlluent colonies t^i ■ ■ j » • i v, iV ♦ ■, , % n. f Keport of the Committee on .Standard Methods oi the Laboratory , ' , ,. " ' . This IS determined by adding to rnlrogen-free media the nitro- Section of the American Public Health Association, 190.5. along the lines of inoculation ^^^ compound to be tested. (3) The observation of cultural and biochemical features BRIEF, a few days, a week. /^t»at t?o^t7t\tt ^ki- *i ...i ^ r „ i shall cover a period of at least fifteen days and frequently longer, BRITTLE ..rowthdrv friable under the platinum needle OPALESCENT, resembling the color of an opal. j j^ jl ^e made according to the revised .Standard -Methods BRITTLE, growth dry, triable under the platinum needle. OPTIMUM TEMPERATURE, temperature at which growth is most above referred to. All media shall be made accordiOL- to the same BULLATE, growth rising in convex prominences, like a. blistered ^^^^.j Standard Methods. surface, T^r^xxTnTT^ ■ « ■ i u * ■ i .u r ■ -.u *■ . (4) Gelatin stab cultures shall be held for .six wi-i-ks to deter- p£j^j^jQj^£ j^ fluid bacterial growth forming either a continuous of r r . „ BUTYROUS, growth of a hutter-like consistency. an interrupted sheet oyer the fluid. """' ''a" AnirSonia and indol tests shall l,,- made a, cd „, len.h CHAINS, PEPTONIZED, said of curds dissolved bv trypsin. day, nitrite tests at end of fifth day. Short chains, composed of L' to 8 elements. PERSISTENT, many weeks, or months. „i, Titrate with ^, N.aOH, using phenolplitliaici,, as an indi- Long chaina, composed of more than 8 elements. t»t Tmiri-vc-T- n r *u tv. ""^ Cn,IATE, having fine, hair like extensions, lilce cilia. ^iy^'LZZ°\fuT r ,■ ,, • "^torniake titrations at same tunc from blank Tlu. difference Xr^T™,, ■ 1 T „ ■ 1 , ■ u 1 \ i ■ 1 . PSEUDOZOOGLE.S, clumps of bacteria, not dissolving readily in gives the amount of acid jiroduced CLOUDY, said of fluid cultures which do not contain pseudozooglea;. ■ . , ■ , ^ \- , , ■ The titration should be done after boiling to drixc ,,ff any COAGULATION., the separation of casein from whey in milk. This ''■'''"• ='"'='°S from imperfect separation, or more or less fusion p,-,^ ^^^^^^^ ^^ ^^^ ^^„^^^^ may take place quickly or slowly, and a,s the result either of the °^ '^e components, but not having the degree of compactness (7) Generic nomenclature shall L,.gin with the year 1«72 ■ ' . .,,11; . and gelatinization seen in zoogleai. (C ohn s first important paper) . lormation of an acid or of a lab terment. _..„.„„..„„ ■ , r , i- 1 .j n Species nomenclature shall begin with the year l.ssn (Koch's CONTOURED, an irregular, si.mothly undulating surface, like that PULVINATE, in the form of a cushion, decidedly convex. discovery of the pour plate method for the separation of organisms) , y', , PUNCTIFORM, very minute colonies, at the limit of natural vision (S) Chromogen,-sis shall be recorded in standard color icrms. or £L reiiei iiup. CONVEX surface, the segment of a circle, but Hattened. RAPID, developing in twenty-four to forty-eight hours. COPROPHYL, dung bact.-ria. RAISED, growth thick, with abrupt or terraced edges. TABLE I. CORIACEOUS, growth tough, leathery, not yielding to the platinum RHIZOID, growth of an irregular branched or root-like character, a NUMERICAL SYSTEM OF RECORDING THE SALIENT needle, as in B. mycoiJta. CHARACTERS OF AN ORGANISM. (GROUP NUMBER.) CRATERIFORM, round, dejiressed, due to the liquefaition of the RING, same as RIM, growth at the upper margin of a liquid culture. ^ , , , ,. ... 1 1 1 ^ .1 1 100 Endospores produced medium. adhering more or less closely to the glass. 2U0 Endospores not produce.! CRETACEOUS, growth opaque and white, chalky. REPAND, wrinkled. lU Aerobic (strict! CURLED, composed of parallel chains in wavy strands, as in anthrax gACCATE, liquefaction the shape of an elongated sac tubular, jf] SiaeroblnJtnc'n''''' colonies. cylindrical. I .........../. . Gelatin liquefied DLASTASIC ACTION, same as DIASTATIC, conversion of starch int., ^^ floating islands of bacteria, an interrupts pelli.le or bacteria -' , Gelatin not li.tuefie.l water-soluble substances by diastase. ...embrinc o i "^"j and gas^ f ron, dextrose . . , ., 1 1- . ■ 1 *■ membrane. 0.2..... Acid t\-ithout gas trom .l.-xtrose ECHINULATE, in agar stroke a growth along line ,il inoculation, ^^^^ ,,^niTins five or six days or more for development, 0.3 No acid from dextrose with toothed or pointed margins; in stab cultures growth beset _„„„_ ,■ w »■ , , 1 0.4 No growth with dextrose with pointed outgro,v ths SHORT, applied to time, a few days, a week. 01 ^,i^l ^„,i ^^3 f,„„j j^.^^^^ „„„„„„ 1 , 1 n r SPORANGIA, cells containing endospores. 0.02 ,\cid without gas from lact.:).se EFFUSE, growth thin, veily, unusually spreading. ™T..T.T™r. ,1 , r KK 1,1 r ,■ 1 03 No acid from lactose ENTIRE smooth havin.. a margin ,l,.stitute of teeth or notches. SPREADING, growth extending much beyond the hi 1 mocula- ^ ^^ j^^ ^^.^^.^j^ ^.j^,^ ,^^.,^^^ EROSE, border irregularly toothed «""■ '■ .'■ ■ ^^^ '^''^'1 millimetres or more. U.OOl Acid and gas froiu saccharose FILAMENTOUS, growth composed of long, irregularly placed or STRATIFORM, liquefying to the walls of the tube a, the ,op an.l 0.002 . Ac^ld .ithout gas f,-o.^^^^^ interwoven filaments ""'n proceeding downward horizontally. 004 No growth with saccharose FILIFORM ill str.,ke .,r St lb cultures a uniform growth along line THERMAL DEATH-POINT, the degree of heat re.iuired to kill 0.0001 Nitrates reduced with evolution of gas of inoculation ' yonng fluid cultures of an organism exposed f..r ten minutes '^ ^^ ■■;■■■ :,^|t*ra1S rTd*u?ed"wifhout gas formation FIMBRIATE, bonier fringe.l with slen.ler jiroces.ses, larger than (in thin-walled test-tubes of a diameter not exceeding 20 mm.) 0,00001 Fluorescent fil.,„o.^*s " in the thermal water-bath. The water must be kept agitated 0.00002 Violet chromogens niaments. QOOii'^ Rlue chromoirens FLOCCOSE, gn^wth .-.imposed of short .airve.l chains, variously so that the temperature shall be uniform durmg the exposure. o! 00004 :::::::: ICreen chromogens oriented TRANSIENT, a few days. 0.00005 Yellow chromogens FLOCCULENT, sahl of fluids which contain pseudozooglca-, .. ,., small TURBID, cloudy with flocculent particles; cloudy plus Hocculence. ooSI^? Red"chromogenl'^™ adherent masses of ba.teria of various sliapes and floating in UMBONATE, haling a button-like, raised centre. OOOOS ! .'..'..'-. Brown chromogens tlie culture Huid. UNDULATE, border wavy, with shallow sinuses. n oonm v"'' ','""™°^'^".*' FLUORESCENT, liaiing one color by traiisi,ntte.l light and another vj-RRUCOSE. growth wart-like, with wart-like prominences. 0000001 '..:'...:. Dms'iasie Ttlon on potato starch, strong by reflected light. VERMIFORM-CONTOURED, growth like a mass of worms or 000002 Diastasic action on potato starch, feeble GRAM'S STAIN a method of differential bleaching after gentian , 0.000003 Diastasic action on potato starch, absent ..^ Ii. 1 ■ i„* „*.. The ^ mark is to be given only intestinal coils. 0000001 Acid and gas from glycerin violet, methyl violet, etc. 1 he t mark is to peg y yjj^LOUS, groTith beset with hair-like extensions. 0.0000002 Acid without gas from glycerm when the bacteria are deep blue or remain blue after counter- .v,tn *i, ^1 v, , i, 1 1 n..^, „ 0.0000003 No acid from glycerin staining with Bismarck brown. VISCID, growth follows the needle when touched and withdrawn, 0000004 No growth wdth glycerin 1 A sediment on shaking rises as a coherent swirl. GRUMOSE, clotted. 7r»r»rTT7a? f,r-n-, <7Plitinnii^ mi^^p^ nf hnctprin nnp nf the moat tvnical -l^he Renas according to the system of Migula is given ita proper INFtJNDIBTJLIFORM, form of a funnel or inverted cone. ZOOGLE^. farm gelatinous ma^.^eb ot tjactena. one ol tfie most typical ^^^^ ^^. ^^ precedes the number thus: (7) ■rir^»irv^«^.r,,T™, .-. x L. c ri T V, o ofFpi^'t nf vprv tHiii tilms cxamples of which is the ^tTeptococcus mesenterioides ot sugar ^-?.l^';l''T' -°*her-of-pea I. The eff'* »'^;;;>"^ ■^^3- ^,,,, "^i Ln.co„ os--g-fA 1 NO CHANGE Wi ^iHW-.HEMOLYSiSfl COMPLEMENT ERYTHROCYTES \ =EfT£:-^"_= ^ Tf : = J HEMOLYSIN ' \ V- ----=^--- = = -"-~--'-~-"--~- ~-~--==J V J> III if DIAGRAM OF NEGATIVE COMPLEMENT FIXATION TEST Fig. 71 The chief reactions used in this test, as well as the whole test, may also be expressed by equations as follows: Complement-fixation Equations. I. Susceptible animal -|- specific antigen = Specific (microbal, hemolytic, etc.) amboceptor serum (Homologous serum) Fixation II. Microbal amboceptor -f- ( ™"'™ ^ I antigen -|- complement (fresh ^ Ipseudo- J guinea-pig serum) of comple- ment with no visible III. Hemolytic amboceptor -|- hemolytic antigen + complement ^ serum Hemolytic system or indicator. y reaction. Fixation of com- plement indicated by hemolysis. 188 ComPlSmSnT-fixA fl6M Test as Employed in the Diagnosis of Syphilis. Non-specific (lipo- ' Pseudo-antigen, which tropic) amboceptor may be acetone insol- C Complement in specific serum + ■ uble fraction of alco- 4- • (fresh guinea- (serum from syphi- holic extract of guinea- [ pig serum) . litic cases). pigs' hearts. / / , = No Hemolytic ambo- / hemolysis ceptor (serum from f Hemolytic antigen / (positive rabbit inoculated + (suspension of sheep ^/ diagnosis) with sheep red blood 1 red blood cells) . ■ cells). Sensitized red blood cells. Historical Note. — The complement-fixation test has been developed from the following classieal experiment of Bordet and Gengou (1901) : Typhoid bacteria (antigen), inactivated typhoid-immune serum (amboceptor), and normal serum (complement) were mixed in a test-tube; after an interval washed erythrocytes and inactivated homologous immune serum (hemolysin) were added. No hemolysis occurred, indicating that the complement had been fixed by the interaction of antigen and amboceptor, whereby the hemolytic system was left incomplete. Wassermann and others have applied this method in measuring the ambo- ceptor content of specific sera, the first practical apphcation of its use being in the diagnosis of sypMlis. Here it is known as the Wassermann reaction. As carried out in the diagnosis of syphilis the reaction was foimd to be due not to a specific antibody but to a lipoid. The complement-fixation may be applied to the identification of an unknown organism or other protein material that is tested as antigen against known immune serum, or it may be applied to the identification of unknown amboceptor in an immune serum that is tested against known antigen, inhibition of hemolysis showing that the antigen and immune serum are homologous. Complement-fixation is used therefore (1) in the diagnosis of various infectious diseases (for example, syphilis, gonococcus infections, and glanders), (2) in the differentiation of pro- teins, (3) in the standardization of some immune sera (for example, antigonococcus and antistreptococcus horse sera), and (4) in helping to establish the etiology of infectious diseases. Technic. — Preparation of Reagents. — ^The reaction of glassware and reagents has been found to have an important bearing on the accuracy "of the test, acidity or alkalinity giving rise to false reactions, positive or negative according to the degree of acidity or alkalinity. All new glass- ware should be neutralized by standing overnight in 1 per cent, hydro- chloric acid. Then, like previously neutraUzed glassware, it is washed in tap-water, thoroughly rinsed in hot distilled water, and sterilized. The distilled water and salt solution used in the preparation of reagents and in the performance of the test are tested for neutrality to phenol- phthalein. Physiological salt solution, 0.85 per cent, to 0.9 per cent., is used in blood-washing, in diluting other reagents, and in the performance of the test. Immune Serum. — ^The method of withdrawing immune serum from an animar depends on the animal and the purpose of bleeding. In ANTIGEN 189 the case of a horse, the bleeding is made from the jugular vein. Rabbits are bled from the marginal ear-vein if only a few cubic centimeters (less than 5 c.c.) of serum are required. If a large amount is required, the rabbit is etherized, tied on a board made for that purpose, and under aseptic precautions bled to death from the carotid artery. A cannula may be inserted in the artery and the blood allowed to flow into pne or more tubes. To obtain the maximum amount of serum, the tube of blood should be slanted at room temperature until coagulation has taken place. In obtaining blood for complement-fixation tests on human beings venous puncture is the most satisfactory method. The drawing of sufficient blood froni finger or ear-lobe is a tedious process and the blood cells are apt to be somewhat broken, so that the serum is tinged with hemoglobin and rendered unfit for testing. Blood is most easily obtained from the median basilic vein of the elbow. The arm should be rendered aseptic by the usual method of scrubbing with soap and water and the application of a 1 to 2000 bichloride of mercury pack for ten minutes. The site of puncture should then be rubbed with alcohol and ether. A hgature is placed above the elbow sufficiently tight to fill the vein, but not tight enough to impede the arterial circulation. A sterile needle is then introduced and 5 to 10 c.c. of blood allowed to flow into a sterile test- tube, which is corked and left slanted at room, temperature until the blood is firmly clotted. The tube should then be placed in the ice-box until serum has separated. If it is necessary to mail a blood specimen, serum only should be sent, as lysis of the blood cells would be caused by the heat and shaking to which the specimen would be exposed in transit. The serum should be mailed in a sealed ampoule or tightly corked tube. All immune serum is separated from the red blood cells and fibrin of the blood by centrifugalization before coagulation has taken place, or, better, by pipetting or poiu-ing from the clot after coagulation. Serum thus removed may be entirely freed from erythrocytes by centrifugaliza- tion. Serum should be removed from the clot before spontaneous lysis of- the blood cells occurs, as hemoglobin has the power of fixing com- plement in itself and a hemolyzed specimen of serum, one containing hemoglobin, is anticomplementary, i. e., it inhibits hemolysis without the presence of a specific antigen, hence cannot be tested for a specific ambo- ceptor. Sermn, both before and after removal from the clot, should be kept in a cool place, at a temperature not higher than 0° C, as the antibody content weakens more rapidly at a high temperatiu-e than at a low. For practical purposes serum is best preserved by freezing.^ The addition of a preservative is not advisable, as the accuracy of the test may thereby be invalidated. Contamination of an immune serum should be avoided, as it may result in an anticomplementary and non-specific action of the serum. All immune serum before use in tests should be inactivated, i. e., heated for one-half hour at 56° C, to destroy comple- ment and serum components that might give rise to a non-specific fixation. Antigen. — ^The method of preparing antigen depends on the nature of the test to be made. Each antigen is described in detail below. A • Serum is perfectly preserved by evaporating to dryness in a vacuum desiccator, but the procedure is complicated. 190 COMPLEMENT-FIXATION bacterial antigen may be prepared as in the original method of Bordet- Gengou, by suspending in physiological salt solution a twenty-four-hour agar growth of the bacterium, making a rather concentrated emulsion. Bacterial extracts give more specific results than emulsions unless the emulsions be made from the organisms dried in vacuo after being treated with alcohol and ether. Olitsky and Bernstein, among others, have shown that if antigens are made from cultures grown on serum media non- specific complement-fixing bodies may be found in them. The best method of preparation depends on the organism. Complement. — Not all fresh serum has the power of reactivating the serum of an alien species, but that of the guinea-pig has unusual power in this respect; hence guinea-pig serum is generally used for complement in hemolytic work. There is a wide variation in the activating power of guinea-pig serum, and also in its power of combining with antigen and immune serum in the complement-fixation reaction. It is advisable to use the pooled serum of at least three pigs, in order to obtain comple- ment of average activating and combining properties. Guinea-pigs that have been used for antitoxin tests and other purposes may be of use in complement-fixation work after a rest of three months, but their serum is less apt than that of unused pigs to be of normal activating power. In the complement-fixation test for glanders pigs that have been inoculated with horse serum must never be used, as substances are formed that cause a reaction with the horse serum that is being tested for glanders and the test is unsatisfactory; serum controls in themselves inhibit hemolysis and no reading can be made of a specific reaction. Gravid pigs should not be used for comple- ment, as their serum is apt to be weak in activating power. To obtain complement, guinea-pigs may be bled from the heart by aspiration and used again after several weeks' rest, or they may be bled to death. In this laboratory the pigs, after being stunned by a blow at the base of the skull, are bled from the throat^ into Petri dishes, which are left at room temperature until the serum begins to separate and then set in the ice-box for several hours. The serum is drawn off with a capillary pipette and centrifuged if not free from blood cells. Before being pooled, the serum from each pig should be tested separately for natural hemolysin, which is occasionally present, and for activating power. Serum containing natural hemolysin or serum of weak activating power should be discarded. Complement deteriorates rapidly if exposed to sunlight or to a warm temperature (over 70° F.) ; if kept in the ice- box (at a temperature below 17° C.) it is good for at least twenty-four hours. Complement may be preserved for several weeks by freezing. Its activating power is retained, but its capacity for being bound may weaken, so that it is not considered reliable. In our experience complement frozen for one week is as good as perfectly fresh comple- ment. Complement is used in a 10 per cent, dilution made with physiological salt solution (see titration tables below). 1 Care must be taken not to cut the esophagus, as the stomach contents might render the serum unfit for use in complement-fixation work. ANTIGEN 191 Erythrocytes from sheep, goat, man, ox, or other animal may be used, and they must be washed free from serum, with physiological salt solution. In our laboratory sheep's cells, washed six times, are used, in a 5 per cent, suspension. , A sheep is bled from the jugular vein into a sterile bottle or flask containing glass beads, and before the blood coagulates it is thoroughly shaken for the purpose of defibrination. Blood is washed in the following manner: In washing a small amount of blood it is convenient to put 2 to 4 c.c. in a 15- c.c. graduated centrifuge tube. The tube is filled with physiological salt solu- tion and a mixture of blood and sahne is effected by means of a pipette. The volume of saline should be three times that of the blood, otherwise the washing is not thorough. The blood is rapidly centrifugalized long enough for the ery- throcytes to fall to the bottom of the tube. The supernatant fluid is decanted or drawn off with a pipette attached to a pressure pump or fitted with a rubber bulb. Physiological salt solution is again mixed with the cells and the tube centrifugalized. This process is repeated until the blood cells are entirely free from senun. In most laboratories three washings are considered sufficient, but we have found that after even four washings the serum is not always com- pletely removed. Hence we make it a ride to wash six times. After the last washing the level of the erythrocytes is marked on the tube, before the cells have been disturbed by the removal of the supernatant fluid. The amount of blood per centrifuge tube, the speed and duration of centrifugalization should always be the same, at least for the last washing, in order that the packing of the cells be uniform day by day. Blood that has been drawn for over two days in summer or three days in winter is unsuitable for hemolytic work, as the resistance of the erythrocytes weakens on standing and hemolysis occurs rapidly, so that perfect balance of the hemolytic system cannot be obtained. In places where it is inconvenient to obtain sheep's cells, human erythrocytes are frequently used. An advantage in the human system arises from the fact that by its use one error in the test is avoided. Human serum frequently contains natural antisheep amboceptor and the excess of hemolysin introduced into the hemolytic system when such a serum is being tested may, according to some investigators, cause a weakening or loss of a positive reaction. In our experience this error rarely or never occurs in the Wassermann reaction, owing to oiu- practice of reading the reaction as soon as controls are completely hemolyzed; hence the use of a hemolytic system other than sheep or the absorption of the natural hemolysin from the serum is not necessary in Wassermann work. In tests with bacterial antigens where the fixation of comple- ment is less firm than in the Wassermann reaction the native antisheep amboceptor in human serum may be a source of error. Hemolysin is obtained by successive inoculations of a rabbit (or other experimental animal) with the type of red blood cell to be used in tests. Inoculations may be made intraperitoneally or intravenously. Owing to the frequency of abscesses after intraperitoneal inoculations the intravenous method is preferable. In this laboratory, rabbits are inoculated intravenously at intervals of two days with fresh, thoroughly washed sheep's cells in a 50 per cent, suspension in physiological salt solution, the doses being 2 c.c, 4 c.c, and 6 c.c. On the ninth day after 192 COMPLEMENT-PlXATlOt} the last inoculation the rabbit is bled to death from the carotid. The serum is put up in small bottles or ampoules and heated for one-half hour at 56° C. on three successive days, to destroy the complement and to insure sterilization. Hemolytic sera of high titre (1 to 3000) may be produced by this method. Standardization of Reagents. — ^An accurate daily standardization of the hemolytic system is very important for reliable results, and this may be accomplished by means of a hemolysin or a complement titra- tion. In the former, varying amounts of hemolysin are incubated with a constant amount of complement and erythrocyte suspension. In a complement titration the amounts of complement are varied and hemolysin is constant. In most laboratories it is considered preferable to follow a hemolysin titration. Preliminary titrations of new hemolytic immune serum are made in dilutions of 1 to 100, 1 to 1000, etc., to determine the dilution of hemolysin in which 0.05 c.c. is one unit. By a unit of hemolysin is meant the smallest amount that gives complete hemo- lysis of 0.1 c.c. of a 5 per cent, sheep's cell suspension in the presence of an excess of complement after one hour's incubation in a 37° C. water-bath. Once standardized the hemolysin is used in the same dilution every day. The technic of the titration is given below. The volume, like that of all our complement-fixation tests, is one-tenth the volume of the classical Wassermann — i. e., 0.5 c.c, instead of 5 c.c. After thorough shaking the titration is incubated one hour in the water-bath at 37° C, and examined frequently to determine the rapidity of the reaction, according to which from two to three units are used in the test. As hemolysin is stable and fresh, and erythrocytes from a healthy sheep vary httle in resistance, such a titration indicates sufficiently for practical work the value of the complement. If the complement is normally active and free from, hemolysin, the erythrocytes fresh and thoroughly washed, the saline solution isotonic and all reagents accurately diluted, hemolysis at the end of the hour's incubation is complete (Plate V, e) in the first six tubes, nearly complete (Plate V, d) in the seventh, partial (Plate V, c) in the eighth, slight in the ninth, and lacking (Plate V, a and b) in the last four tubes. Then since 0.05 c.c. is the unit of hemolysis, 0.1 c.c. is, in general, used to sensitize 0.1 c.c. of erythrocytes. The interpreta- tion of a hemolysin titration can be learned by experience only. Hemolysin Titration. 10 per cent. complement. c.c. 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.0 0.1 0.0 Number of tube. Hemolysin in standard dilution. 0.0. 1 0.1 2 0.09 3 0.08 4 0.07 5 0.06 6 0.05 7 0.04 8 0.03 9 0.02 10 0.01 11 0.1 12 0.0 1.3 0.0 5 per cent. erythrocyte 0.85 per cent. suspension. saline. 0.0. c.c. 0.1 0.0 .a 0.1 0.21 0,1 0.22 •So 0.1 0.23 Sl 0.1 0.24 is 0.1 0.25 Sts 0.1 0.26 §i 0.1 0.27 Si 0.1 0.1 0.1 0.1 0.28 0.29 0.3 0.3 i' 0.1 0.4 fl PLATE V a. Complete absence of hemolysis before settling of cells. b. Conciplete absence of hemolysis after settling of cells. c. Partial hem.olysis. d. Nearly complete hennolysis. . e. Complete hemolysis. ANTIGEN 193 Complement Titbation. Hemolysin in 5 per cent. Number of 10 per cent. standard erythrocyte 0.85 per cent. tube. complement, dilution. suspension. saline. 0.0. c.c. c.c. c.c. 1 0.1 0.05 0.1 0.25 (D 2 0.09 0.05 0.1 0.26 ^ 3 0.08 0.05 0.1 0.27 to 4 0.07 0.05 0.1 0.28 So 5 0.06 0.05 0.1 0.29 o t^ ^ CO 6 0.05 0.05 0.1 0.3 §^ • 7 0.04 0.05 0.1 0.31 °:S 8 0.03 0.05 0.1 0.32 °l 9 0.02 0.05 0.1 0.33 ■7-1 ^ 10 0.01 0.05 0.1 0.34 ^1 11 0.1 0.0 0.1 0.3 it 12 0.0 0.05 0.1 0.35 3 13 0.0 0.0 0.1 0.4 A Antigen Titbation. Number of 10 per cent. 0.85 per cent. SensiLizea erythrocyte tube. Immune serum Antigen complement ;. saline. suspension. c.c. c.c. c.c. CO. CO. 1 0.01 0.25 0.1 0.0 a 0.2 2 0.01 0.2 0.1 0.0 'Z^ 0.2 ^ 3 0.01 0.15 0.1 0.05 v^ 0.2 fld 4 0.01 0.1 0.1 0.1 M n 0.2 o 5 0.01 0.05 0.1 0.15 ^S'S 0.2 gco 6 0.01 0.025 0.1 0.2 ■It 0.2 7 0.0 0.4 0.1 0.0 §•? 0.2 §1 8 0.0 0.3 • 0.1 0.0 ° i 13 43 0.2 ^■f 9 0.0 0.2 0.1 0.05 l| 0.2 II 10 0.0 0.1 0.1 0.1 ^^ 0.2 11 0.0 0.05 0.1 0.15 3 S 0.2 § ^ 12 0.02 0.0 0,1 0.2 0.2 If an exact determination of complement value is desired the hemoly- sin titration may be followed by a complement titration as given above. The unit of complement is the smallest amount that completely hemolyzes 0.1 c.c. of 5 per cent, erythrocyte suspension sensitized by one unit of hemolysin. If the balance of the hemolytic system is to be obtained by varying complement instead of hemolysin, two units of hemolysin instead of one unit should be used in titrating complement. In our opinion more uniform and accurate results are secured by varying hemolysin than by varying complement. Antigen. — ^There are two requirements for a good antigen, a long range and specificity.^ The range of an antigen is determined by mixing varying amoimts of the antigen with a constant amount of a previously tested homologous immune serum, a constant amount of complement and a constant amount of sensitized erythrocyte suspension. The specificity of an antigen is determined by using heterologous instead of homologous immune serum in the titration. The technic employed in this laboratory in the titration of an antigen is given above. Readings are made after the erythrocytes have settled, and meantime the titration should be kept in the ice-box to prevent a continuation of hemolysis. An immediate reading may be made by centrifugalizing the tubes. If ' The range of an antigen is the difference between the anticomplementary dose (the smallest amount of antigen that is in itself inhibitory) and the minimum fixing dose, the antigen unit. IS^ = 194 COMPLEMENT-FIXA TION fixation is complete through 0.025 c.c. (tube 6) a 10 per cent, solution of the antigen should be titrated in the same manner to determine the unit of antigen, which is the smallest amount that, ivith two units of homolo- gous immune serum (or 0.01 c.c. of a himian serum which has given a H — I — I — |- leaction) , gives complete fixation of complement. The anti- complementary dose is the smallest amount of antigen that is in itself inhibitory. The longer the range of the antigen the greater is the probability of success in diagnosis. An amount of antigen that fixes complement completely with a serum of high antibody content may give incomplete or no fixation with a serum of low antibody content. Hence in making diagnostic tests, where the detection of even a small amount of antibody is desired, it is advisable to use much more than one unit of antigen. As the maximum amount of antigen that may he used with safety is one-fourth the anticomplementary dose an antigen of long range is necessary. If in the table just given inhibition is complete (Plate V, a and h) in the first five or six tubes, hemolysis is complete (Plate V, e) in tubes 8 to 12, and there is very sHght inhibition (Plate V, d) in tube 7 only, then 0.1 c.c. is the amount of antigen to be used in diagnostic tests. Antigen should, 'always he so diluted that 0.1 c.c. may he used. If one-fourth the anticomplementary dose gives complete fixation with a heterologous immune serum the antigen is non-specific and unsuitable for tests. Occasionally an antigen is lytic for erythrocytes. In such a case tubes containing the largest amount of antigen show more hemolysis than those containing less. The fixation curve instead of dropping (Fig. 72, a) first rises and then drops again (Fig. 72, h). If a lytic antigen is also anticomplementary and has a long fixation range it may he used, otherwise it should he discarded. Complete fixation -Incomplete fixation /: Complete fixation Incomplete fixation -^^ hemolysis hemolysis Number of tube. 1 2 3 4 5 6 Patient's serum. c.c. 0.02 0.01 0.04 0.02 0.0 0.0 Antigen in standard dilution. Fig. 72 Test. 10 per cent. 0.85 per cent. complement c.c. 0.1 0.1 0.1 0.1 0.1 0.1 saline, c.c. 0.1 0.1 0.2 0.2 0.0 0.1 ^ a •2.2 Sensitized erythrocyte suspensioni c.c. 0.2 0.2 , 0.2 0.2 0.2 0.2 n at R J-H ^ ' ANTIGEN 195 A test for diagnosis is set up as given above. Fixation of complement may be allowed to take place for one-half hour in a water-bath at 37° C, for one hour in the incubator at 37° C, at room temperature for four hours or more' or in the ice-box for from four to eighteen hours. - Whichever method is employed an antigen titration fixed by the same niethod should be followed, as the antigenic range varies with the time and temperatm-e allowed for fixation. Positive and negative control sera should always be included in the test. Citron's standard for the strength of a complement-fixation reaction is used by this laboratory in the reading- of all tests (Plate V): Complete absence of hemolysis in tubes 1 and 2 = + + -t- +, very strong positive. Complete absence of hemolysis in tube 1, faint hemolysis in tube 2 = -(- + +, strong positive. Complete absence of hemolysis in tube 1, complete or nearly complete hemolysis in tube 2 = + +, positive. Partial hemolysis in tube 1, complete or nearly complete hemolysis in tube 2 = +, weak positive or doubtful. Nearly complete hemolysis in tube 1, complete hemolysis in tube 2 = ±, doubtful. Complete hemolysis in all tubes = — , negative. The distinction between -| — \- and + is very important, as the latter reaction, we have heard, is frequently non-specific. We therefore make a positive diagnosis on only +H-,+ + +,or+-|- + + reactions." An antibody content titration is made for the purpose of measuring more accurately than in the diagnostic test just described the amboceptor con- tent of an immune serum. Our technic is given in table below. The serum is first titrated in a 10 per cent, solution; if fixation is complete through 0.01 c.c. a similar titration of a 1 per cent, dilution is made. The amount of antigen used is double the antigen unit; for example, if 0.05 c.c. of a 10 per cent, solution of antigen is the smallest amount giving complete fixation with a homologous immune serum, 0.1 c.c. of that dilution is used in the antibody content titration. The reading is made as in an antigen titration, after the erythrocytes have settled. The antigen and serum controls (tubes 11 and 12) should, of course, be hemo- lyzed. An antibody unit is the smallest amount of serum that with two units of an homologous antigen gives complete fixation of complement. The number of antibody units per cubic centimeter may be calculated by dividing 1 c.c. by the minimum fixing dose; that is, if 0.05 c.c. of a 1 in 10 dilu- tion is the antibody unit, the number of antibody units per cubic centi- meter equals tt.ottt = 200. An antibody content titration is used for the standardization of antimicrobic sera, such as the antigonococcus and antistreptococcus horse sera. The power of a serum to fix complement ' In our experience, room temperature is not desirable for fixation, as complement may deteriorate and the hemolysis of control tubes cannot be relied upon even though the hemolytic system when standardized is active. ^ A + Wassermann reaction following the treatment of a case that has given a + +, + + +, or + + + + reaction is considered positive and indicates the need of further treatment. 196 COMPLEMENT-FIXATION is not, however, always parallel with its protective or therapeutic power. Besredka found that antistreptococcus serum of high protective power had little complement-fixing power, and vice versa. Antibody Content Titration. Immune Antigen in serum. dilution Sensitized Number of Diluted 1 determined 10 per cent. 0.85 per cent. erythrocyte tube. to 10 (etc.). by titration. complement. saline. suspension. c.c. c.c. c.c. c.c. c.c. 1 0.1 0.1 0.1 0.0 •-H 0.2 2 0.09 0.1 0.1 0.01 P o 2 » 0.2 3 0.08 0.1 0.1 0.02 •0.2 i 0.07 0.1 0.1 0.03 j3 CO 0.2 5 0.06 0.1 0.1 0.04 ■3 « 11 0.2 6 0.05 0.1 0.1 0.05 0.2 7 0.04 0.1 0.1 0.06 I-? 0.2 • 8 0.03 0.1 0.1 0.07 ■^i 0.2 9 0.02 0.1 0.1 0.08 5| 0.2 IQ 0.01 0.1 0.1 0.09 ^t 0.2 11 0.0 0.2 0.1 0.0 §1 0.2 12 0.2 0.0 0.1 0.0 0.2 flrS- II a The Wassermann Reaction. — The antigen originally employed by Wassermann, Neisser, and Bruck, and still preferred by some workers, consists of a saline extract of liver from a syphilitic fetus. The finely divided tissues are mixed in the proportion of one to four with normal salt solution, to which 0.5 per cent, carbolic has been added, agitated at room temperature for twenty-four hours, centrifuged, and the super- natant fluid dra\vn off into sterile vessels and kept in the ice-box until needed. Since the cultivation of the Treponema pallidum has been made possible, antigens have been made from the organisms in pure culture instead of from luetic tissue; but the results have, on the whole, been less satisfactory, i. e., fewer cases of syphilis give a positive Wassermann reaction with a Treponema pallidum antigen than with a luetic tissue antigen, or even with a lipoid antigen from normal tissues. The exact nature of the antigen that produces the antibodies (called by Citron Luesreagin) taking part in the Wassermann reaction is unknown. It appears to be neither the pure spirochetes, nor a pure lipoid substance. Although a pure lipoid cannot stimulate the production of antibodies when inoculated into an experimental animal, it reacts in vitro with the Luesreagin in the blood of the syphilitic. The Wassermann reaction then is a lipotropic or lipoidophilic reaction and not due to the interaction of specific antigen and antibody. An extract of heart, liver, or kidney in 96 per cent, alcohol may be used. Some serologists recommend the addition of cholesterin to a crude alcoholic extract; such an antigen gives a higher percentage of positive reactions than most other antigens, but its reliability has not yet been established. In our experience a cholesterinized antigen is of special value in deternaining the effectiveness of antiluetic treatment, as old cases of syphilis that are or have recently been under treatment are more apt to react with this antigen than with a crude alcoholic antigen. In making a fresh diagnosis it is not wise to depend on a cholesterinized antigen alone, since cases of scarlet fever, leprosy, and other non-syphilitic conditions have been found to react strongly with this antigen. These false positive reactions do not occur with a crude alcoholic antigen. A positive reaction is, however, presumptive of syphilis and a negative THE WASSERMANN REACTION 197 reaction has more value in excluding a diagnosis of syphilis than has a negative reaction with the crude alcohohc antigen. A safe and stable antigen is Noguchi's acetone insoluble fraction of beef heart, liver, or kidney, prepared by extracting macerated tissue with ten times the amount of absolute alcohol at room temperature for several days, filtering, evaporating the filtrate to dryness, taking up the residue with ether, treating the ethereal solution with five times its volume of acetone, and making a satu- rated solution of the precipitate in absolute methyl alcohol. In our laboratory the most satisfactory results have been obtained with a crude alcoholic extract of guinea-pig hearts prepared as follows: hearts (from pigs bled to death for complement) are minced and washed in tap-water until free from blood and macerated in C. P. 96 per cent, alcohol in the proportion of 1 gram of heart to 5 c.c. of alcohol. Extraction is allowed to take place in the ice-box for at least three months and then in the incubator for a week. The extract is filtered through paper and kept in the ice-box, where it remains stable for at least a year. The titre of different antigpns varies from 1 to 20 to 1 to 100. In diluting the antigen the first 10 c.c. of saline are added drop by drop and the remainder slowly, the mixture being effected by gently rotating the receptacle, not by shaking. The table on page 198 gives the classical Wassermann technic. The readings are made according to Citron's standard. Our technic differs in three respects from Wassermann's : in the size of the test, in the amount of antigen used, and in the method of fixation. In our experience the results obtained through using all the reagents in one-tenth the amounts used by Wassermann are absolutely reliable, provided the test is accurately performed, and the saving in material is considerable. Another advantage in the small size is that a very small amount (about 0.2 c.c.) of the patient's serum is sufficient for a test. Concerning the amount of antigen to be used in the test, we have found the use of a constant amount to give the same results as the use of a varying amount. For fixation of complement a foiu'-hour period in the ice-box is allowed instead of incubation, as it has been found that about 10 per cent, more positive reactions in cases of syphilis are obtained by this method, and without any apparent danger of obtaining false positive reactions. Though biologically non-specific the Wassermann reaction is clinically specific except, perhaps, in cases of leprosy, yaws, sleeping sickness, and scarlet fever.^ With these exceptions a positive reaction indicates the presence of luetic .infection, either active or latent. A positive reaction may be given in any stage of syphilis but is most apt to occur in the secondary stage. A negative reaction at any stage of the disease does not exclude the possibility of syphilis. Antiluetic treatment, especially mercury, frequently results in the reaction becoming negative, though the disease may still be active. Treatment with salvarsan and even with mercury and the iodides, may at first or after a few treat- ments cause a negative Wassermann to become positive for a time. A cm-e can only be pronounced when the reaction has remained negative for at least a year after intensive treatment and when it then still remains negative after a provocative treatment with salvarsan. ■ The ingestion of alcohol or the administration of an anesthetic within twenty-four hoiu-s of collection of the blood specimen interferes with 1 With the use of the crude alcoholic antigens, in one-quarter the anticomplementary dose, we have never obtained false reactions in such cases ever by ice-box fixation. 198 COMPLEMENT-FIXA TION the accuracy of the test, alcohol weakening the reaction and an anesthetic giving rise to false positive reactions. Spinal fluids may give a positive reaction when the blood is negative in cases in which the brain or cord is involved, hence this fluid should always be tested when a negative report is received from a blood exami- nation in which disease in such locations is suspected. Classical Wassbrmann Test. Luetic liver 10 per 0.85 per Sensitized Number extract Patient's cent, com- cent. erythrocyte of tube. diluted. serum.' plement. saline. suspension. c.c. c.c. c.c. c.c. c.c. 1 1.0 0.2 1.0 0.8 2.0 Incubated until 2 0.5 0.1 1.0 1.4 Incubated 2.0 control tubes 3 2.0 0.2 1.0 1.8 one hour. 2.0 are completely 4 2.0 0.0 1.0 0.0 2.0 hemolyzed. To avoid error in the interpretation of the complement-fixation reaction, each diagnostic test should be accompanied by the following controls : II. Duplicate diagnostic test to detect error in technic. 2. Test for natural antisheep amboceptor to avoid false negative reactions. 3. Test for anticomplementary unit. 4. Test for lytic action upon cells alone. Antigen controls f Diagnostic test with known positive human ■I serum. Diagnostic test with known nega- \ tive human serum. Test tor anticomplementary unit. Test for fixation unit. Test for lytic action with complement when no amboceptor is used. 1 . For specificity of antigen. The Complement-fixation Test for Gonococcus Infection. — ^This method was first applied to the study of gonococcus infection by Miiller and Oppenheim (1906). The technic developed by Schwartz and McNeil has been widely followed and the value of the test in the diagnosis of secondary gonococcus infections, especially in conditions in which a bacteriological diagnosis is difficult, as in arthritis, has been thoroughly established. A polyvalent antigen, one made from as many different strains of the gonococcus as possible, is essential.^ A twenty- four-hour growth on salt-free veal^ agar neutral to phenolphthalein is washed off with sterile neutral distilled water, autolyzed in a 56° C. water-bath for one hour, and in an 80° C. water-bath for one hour (to destroy any ferment that might render the antigen unstable), filtered through paper pulp and a Berkefeld filter,' and heated on three successive ' In this laboratory ten strains of gonococci isolated by Torrey are used in the prepa- ration of the antigen. ' Medium from bob-veal is most desirable, made in the regular way but without salt. ' New Berkefeld filters are very alkaline, and before use for the filtration of bacterial antigens they should be boiled in distilled water at least three times for five minutes each time, and scrubbed thoroughly with a small brush in fresh water, after each boiling. After the filter is set up, hot, neutral, distilled water should stand in it for about five minutes. Then hot, neutral distilled water should run through under gentle pressure until the fluid is clear and neutral to phenolphthalein, when the filter is ready for use. Under high pressure a filter still alkaline might test neutral. After use, the filter should be boiled in distilled water, scrubbed, and dried in the air. A filter used for gonococcus antigen should never be used for any other bacterial antigen, unless it is first boiled in 1 per cent, sodium hydroxide solution and reneutralized. PERTUSSIS 199 days at 56° C. for^one-half hour for sterilization. This antigen is stable for at least six months and is highly specific.' Immediately before use the antigen must be made isotonic by the addition of one part of 9 per cent, saline solution to nine parts of the antigen. The test is performed in the usual way. The optimum period for fixation has been found to be six hours in the ice-box. Readings must be made with great care. A positive diagnosis should not be made unless 0.02 c.c. of serum gives complete inhibition of hemolysis, as serum containing heterologous amboceptors, for example, strepto- coccus, may give a -f- reaction with gonococcus antigen. Only gono- eoccus amboceptors give a +-h, + + +, or -f--|--|--l- reaction. Cases of anterior gonorrheal urethritis and acute vulvovaginitis rarely give a positive complement-fixation test. A positive reaction is indi- cative of the presence or recent activity in the body of a focus of living gonococci. A positive reaction may persist for from six to eight weeks after a cure has been effected. Persistently negative results obtained through a considerable period of time indicate the probability of a cure. The Complement-fixation Test for Glanders. — Complement-fixation has proved to be a valuable aid in the diagnosis of glanders. It is generally considered specific and more reliable than the agglutination test. In this laboratory the antigen is prepared from a twenty-four-hour growth of B. mallei on salt-free veal-agar 1.6 per cent. acid. The growth is washed off with sterile distilled water, autolyzed for six to eight hours at 80° C, filtered through paper pulp and a Berkefeld, and the filtrate sterilized at 56° C. on three successive days for a half-hour. This antigen, like the gonococcus and all other aqueous extracts, must be made isotonic before use. The test is performed like the others, but the optimum period for fixation has been found to be 6 to 18 hours in the ice-box. The New York Health Department condemns all horses that give a -\ — | — h + complement-fixation reaction when it is con- firmed by the eye mallein test, while those that give a strongly -|- + and H — h + are suspected of having a sUght glanders infection.^ Streptococcus Infections. — ^The value of complement-fixation in the diagnosis of streptococcus infections is still uncertain. In this laboratory fairly satisfactory results are obtained with a sahne antigen prepared as follows: A twenty-four-hour growth on salt veal-agar neutral to phenolphthalein is washed off with 0.85 per cent, saline solution. The emulsion is heated at 60° C, one hour, left in the ice-box twenty-four hours, and centrifugalized. The supernatant fluid is used for tests, either water-bath or incubator being used for fixation. The test seems to be specific, but is far from perfected. Pertussis. — Complement-fixation has been used for the determina- tion of the etiological cause of whooping-cough and for the diagnosis of the disease. The results of many investigators, including ourselves, 1 The specificity of a gonococcus antigen may be best determined by an antigen titra- tion against antimeniugococcus and antistreptococcus sera of high antibody content. ' At autopsy, macroscopic lesions are always shown by horses giving a + + + + comple- ment-fixation test, rarely by those giving a -|-+ora + + + reaction. 200 COMPLEMENT-FIXATION confirm the findings of Bordet, that the Bordet-Gengou bacillus is the etiological cause of whooping-cough. As to the diagnostic value of the test, reports vary. The figures given by the workers who use active serum are undoubtedly too high, as investigations in this laboratory- have proved that active serum may give non-specific fixation with even a highly specific antigen. We have found the most satisfactory antigen, that is, the strongest and most specific, to be obtained by shaking and then autolyzing at 56° C. A twenty-four-hour culture on the Bordet- Gengou potato-blood-agar medium is scraped off with a platinum spud and put in sterile distilled water. The emulsion is shaken for about two hours, left in a thermostat at 56° C. for about eighteen hours, and centrifugalized. The supernatant fluid is used for tests, being made isotonic immediately before using. Water-bath or incubator is used for fixation; room temperature, at which some workers allow fixation to take place, gives such variable results as to be unsuitable for routine work. We obtain a positive reaction in about 50 per cent, of whooping-cough cases in the paroxysmal stage. The administration of whooping-cough vaccine may increase the strength of the complement- fixation reaction, but experiments with normal individuals have proved that vaccine in itself, in the absence of whooping-cough^ does not bring about a positive reaction. Immune serum of high antibody content may be produced by the intraperitoneal inoculation of rabbits once a week with a live culture of the Bordet-Gengou bacillus, beginning with a dose depending on the condition of the rabbit. The height of immunity is usually reached through five or six inoculations; if other inoculations are given, the antibody content decreases. The rabbits are bled nine days after the last inoculation. Complement-fixation Test for Tuberculosis. — ^The application of the Bordet-Gengou phenomenon to the study of clinical and experimental tuberculous infections in man and various animals has been repeatedly and variously tried. No method has as yet been devised which yields results in any way comparable in value to those afforded by the Wasser- mann reaction in syphilis. Antigens made from the various single and combined constituents of the tubercle bacillus, representing both the protein as well as the lipoid fractions, emulsions of living or dead tubercle baciUi and also the substances elaborated by the tubercle bacillus in many kinds of nutrient media have all been employed. When it is remembered that in tuberculous infections the usual antibodies are scant or wholly wanting in the serum it is not surprising that investi- gators have encountered such great difficulties in devising a satisfactory method. The very nature of the disease, involving as it does, even after arrest and healing, the presence of an infective focus, requires the greatest refinement of the technic in order that a discrimination can be made between latent or inactive infection and active infection or disease. The majority of the methods already recommended yield little or no information more than that elicited by proper tuberculin tests. The method of Besredka, in which the antigen is prepared from a culture on an egg mediiun has given good results in his hands and also with Bron- PARASITIC SKIN DISEASES 201 fenbrenner. Miller and Zinsser, using the Wassermann technic with an antigen made by grinding living tubercle bacilli with dry salt and then emulsifying the ground bacilli, report encouraging results and go so far as to say that by this method they are able to detect active tuber- culosis. Until further confirmation is forthcoming, the complement- fixation method in tuberculosis should be considered as being of scientific, rather than of clinical value. It is in too early a stage of development to be relied upon in either diagnosis or prognosis and it would seem that its greatest value might lie in the light it may shed upon the mechanism of tuberculosis immunity. An excellent review of the literature is given by Stimson and by Miller. Meningitis. — ^The complement-fixation method has been successfully applied in epidemic meningitis by Briick, but the diagnosis can more readily be made by the bacteriological examination of cerebrospinal fluid. Complement-fixation is a valuable means of differentiating strains of meningococci. The antigen used here for this purpose is prepared in the same manner as gonococcus antigen. Water-bath fixation is used for all titrations. Typhoid. — Complement-fixation is a valuable method for corroborating the Widal. Its exact clinical value and specificity have not yet been established. The use of a highly polyvalent antigen is essential. Gar- bat's method of preparing the antigen is to wash off with sterile distilled water a twenty-four-hoiu" growth on agar. Heat the emulsion at 60° to 70° C. for twenty-four hours, shake with glass beads for twenty- four hours, and centrifugalize until the supernatant fiuid is absolutely clear. A positive reaction usually appears only after bacteria have disappeared from the circulation; it becomes stronger during con- valescence and persists for several months afterward. Parasitic Skin Diseases. — Kolmer and Strickler, corroborated by others, reported rather favorable results of fixation of complement in ringworm and favus. REFERENCES. Bzsredka: Ztschr. f. Immuntatsforsch., 1914, xxi, 77. BoRDET and Gengou: Ann. de I'lnst. Past., 1901, xv, 290. KoLMEH and Strickler: Complement Fixation in Parasitic Skin Diseases, Jour. Am. Med. Assn., 1915, Ixiv, 800. Miller, H. R.: Tlie Clinical Value of Complement Fixation in Tuberculosis, Jour. Am. Med. Assn., Ixvii, 1519. Miller, H. R.: A Review of the Complement-fixation Test in Tuberculosis, Jour. Lab. and CHn. Med., August, 1916, i. No. 11. Miller and Zinsser: Tr. New York Path. Soc, February 7, 1916, also Proc. Soc. Exp. Biol, and Med., 1916, xiii, 134. Neisser und Sachs: Berl. klin. Woch., 1905, xlii, 1388; 1906, xlii, 67. Olitskt and Bernstein: Non-specific Reactions in Antigen Made from Serum Media, Jour. Infect. Dis., 1916, xix, 253. Olmstead and Luttinger: Complement Fixation in Pertussis, Arch. Int. Med., 1915, xvi, 67. Stimson, A. M. : Complement Fixation in Tuberculosis, Hygienic Laboratory, Bulletin No. 101, 1915, United States Public Health Service. Wassermann und Bhuck: Deutsch. Med. Woch., 1906, xxxii, 449. CHAPTER XII. AGGLUTINATION AND PRECIPITATION OF MICEO- ORGANISMS AND THEIR PROTEINS. THE NATURE OF THE SEBUM SUBSTANCES CONCERNED. By the phenomenon of agglutination is meant the aggregation into clumps of uniformly disposed microorganisms in a fluid because of the action of an homologous immune serum. If the organisms are motile they become immobile. Many other substances other than those in serum cause the agglutination of cells. We are only interested in these because they may cause confusion. This phenomenon, while it had been noted by earlier observers (Charrin and Roger in 1889), was first extensively studied by Gruber and Durham in 1896, who determined that the serum of those passing through certain infections contained a specific substance (agglutinin) which caused the infecting organisms to clump. Several months later Widal reported that in typhoid fever the development of agglutinins could be used for diagnostic ptu"poses. It was thus demonstrated by these studies and those of Grunbaum, Bordet and others that through agglutinins a new means was available for the identification of bacteria and in many cases the nature of the infecting organism causing disease. As to the nature of these phenomena a number of theories have been advanced. There is a close analogy between agglutination and the flocculation of colloidal suspension by electrolytes. As in the case of the immune body, there is positive proof that the agglutinin combines directly with agglutinable substances in the bacterial body, the two bodies effecting a loose combination. But since a certain amount of sodium chloride or other inorganic salt is necessary (Bordet) it must be classed as a physicochemical reaction. Dead bacteria agglutinate as well but more slowly than living bacteria. The antigenic substance is designated as agglutinogen and the antibody as agglutinin. Ehrlich considers that the agglutinin consists of a haptophore or combining atom group which is stabile and of a ferment group which is labile (receptors of the second order, Fig. 68). The latter causes the phe- nomenon of agglutination. In some types of infection there is a great accumulation of agglu- tinins in the blood. Thus in typhoid patients and convalescents distinct agglutination has been observed in dilutions of 1 to 5000, and this re- action persisted for months, though not, of course, in the same degree. Even normal blood serum, when undiluted, often produces agglutina- tion through group agglutinins. But the specific agglutinins, which are formed only in consequence of an infection, are characterized by this. THE NATURE OF THE SERUM SUBSTANCES CONCERNED 203 that they produce agglutination even when the serum is highly diluted, and, furthermore, that after this dilution the action is generally specific — I. e., the high dilutions of cholera-immune serum agglutinate only cholera spirilla, of typhoid-immune serum only typhoid bacilli, etc. This specificity, however, for some bacteria such as the colon and dysentery group, is not always absolute, as the group agglutinins previously present may be in unusual amount. It was formerly assumed that agglutination was a prerequisite for bdcteriolysis. This, however, is not so, for both in cholera and in typhoid immunity bacteriolytic substances have been observed without agglutinins, and agglutinating substances without bacteriolysins. Characteristics of Agglutinins and Agglutinogen. — The union of agglutinin with agglutinogen in bacteria is a physicochemical reaction, and is quantitative. By chemical means it is possible again to separate a portion of the agglutinin from bacteria saturated with it and use it to agglutinate bacteria anew. The amount of bacteria in the emulsion used to test the amount of agglutinin must therefore be known if a quantitative determination is desired. An emulsion one hundred times as dense as another would require one hundred times as much agglutinin- to give an equally complete reaction. Heating the serimi above 60° C. injures the agglutinin but slightly, above 70° C. greatly, and above 75° C, destroys it. Agglutinins changed by heat, acids, and other influences may become "agglutinoids," which are comparable to toxoids, e. g., agglutinating sera heated to a certain temperature lose their power to agglutinate but act upon bacteria so that they are unable to be agglutinated by active serum. Heating the bacteria above 60° C. diminishes their agglutinability. Dreyer found that if a twenty-four-hour bouillon culture of Bacillus coli required 1 part of agglutinin to agglutinate it, then if heated to 60° C. it required 2.3 parts; if to 80° C, 18 parts; if to 100° C, 24.6 parts. He found the siu-prising fact that long heating of the culture restored to some extent its ability to be agglutinated by smaller amounts of agglutinins. Heated thirteen hours to 100° C, the culture was agglutinated by four parts. Agglutinin does not dialyze through animal membranes. In dilute solution agglutinin slowly deteriorates. Dried, it lasts longer. It is precipitated with the globulins by ammonium sulphate. When a solution containing agglutinin is passed through a stone filter the first few cubic centimeters contain no agglutinin. The next contain a moderate amount and the remainder the same as the solution. It is important to remember that in low dilutions of a serum agglu- tination may fail, while in higher dilutions agglutination may take place readily. Weak and strong acids agglutinate bacteria, while medium acidity does not. Alkalis inhibit agglutination. The Development of Agglutinin. — Experimental or natural infection of animals and men is followed in seven to ten days by an appreciable 204 AGGLUTINATION AND PRECIPITATION development of agglutinin. This development is much greater in certain bacterial infections than in others. Group Agglutination. — Many varieties of organisms have, among the different protein substances composing their bodies, some that are common to other microbes which are more or less allied to them (Fig. 73). If these substances are of the. type that excite agglutinins, we have from an animal immunized by any one of them a serum acting on other organisms somewhat in proportion to the amount of agglutinin- producing protoplasm which they have in common with the infecting organism. These agglutinins, acting on substances common to other microorganisms which are generally but not always allied varieties are called, therefore, group agglutinins. Thus, in a case, the infecting para- typhoid bacilli type B were agglutinated 1 to 5700; typhoid baciUi, however, only 1 to 120, while paratyphoid baciUi type A were agglutin- ated only 1 to 10. In a case of typhoid fever an agglutination of para- typhoid type B occurred with a dilution 1 to 40, while typhoid baciUi were agglutinated with 1 to 300. Typhoid Bacillus Dysentery Bacillus Fig. 73. — Specific and common agglutinins producing protoplasm. The bacteria which are agglutinated by one and the same serum need not at all be related in their morphological or other biological character- istics, as at first assumed. Conversely, microorganisms which, because of the characteristics mentioned, are regarded as entirely identical are some- times sharply differentiated by means of their agglutination. In other words, the "groups" arrived at by means of a common agglutination have no necessary relation to species as the term is usually employed, but only of chemical similarity. This is indicated by the diagrams in Fig. 73. The letters indicate chemical substances capable of stimulating the production of agglutinin and of combining with it when made. Thus both the typhoid and colon will stimulate B agglutinins and react to them, while type A agglutinins are produced only by the typhoid bacilli, and type D only by B, coli. Because of this lack of absolute specificity of the agglutination reaction the clinical diagnosis of the type of infectipn or the absolute identification of bacteria through the agglutination test can only be determined in those cases where the group agglutinins are not abundantly present. This suffices for some infections such as those caused by the typhoid bacillus and the cholera spirillum, but not for others such as those due to the colon group of bacilli. With the use of absorption methods given below the specific agglutinins can be separated in most cases from the group agglutinins. THE NATURE OF 'THE SERUM SUBSTANCES CONCERNED 205 The Relative Development of Specific and Group Agglutinins.— The study of a large number of series of agglutination tests obtained from young goats and rabbits injected chiefly with typhoid, dysentery, paradysentery, paracolon, colon, and hog-cholera cultures has shown that there is considerable uniformity in the development of the specific and group agglutinins. The specific agglutinins develop a larger amount of their total in the early days, being in the second week usually from five to one hundred times as abundant as the group agglutinins. Later the total amount of the group agglutinins tends to approach more nearly to that of the specific, and may reach as high as 20 to 50 per cent. In a number of tests carried out by us we found that many group agglutinins supplement spiecific ones in their action, causing by their addition an increased agglutinating strength. In our experience the variety of microorganism used for inoculation is, if equally sensitive, agglutinated in a higher dilution by the combined specific and group agglutinins produced through its stimulus, than any microorganisms affected merely by the group agglutinins. It is true that related bac- teria were at times agglutinated in higher dilutions than the variety injected; this, if not due to greater sensitiveness, was caused by normal group agglutinins present in the animal before immunization. For this reason untreated horse serum is a very dangerous substance to use in differentiating the intestinal bacteria unless the serum is tested for group agglutinins. The great height to which the group agglutinins may rise is seen in the following table: Agglutinin in the Seeum of a Hobse Injected with Paeadysentery Bacillus, Culture Type Manila. After 18 injections. After 21 injections. Culture. 1 : 3000 1 : 5000 1 : 10,000 1 : 3000 1 : 5000 1 : 10,000 Paradysentery type Manila + + — — ++ ++ + + Colon B.X ++ ++ — ++ ++ + + The great amount of agglutinins acting upon the colon bacillus X. is remarkable. A sertun is here seen to be acting in dilutions as high as 1 to 10,000 upon a culture possessing different characteristics from the one used in the injections. Although a considerable proportion of the group agglutinins acting on colon bacillus X. was undoubtedly due to the stimulus of the injec- tions of the paradysentery culture, still a portion of them was probably due to the agglutinins developed by the stimulus of the absorbed intestinal bacteria. In the table given below is seen the marked accumu- lation of agglutinins which may occur in a normal horse before injections are begun: Culture. A young horse before inoculation. 1 : 100 1 : 500 1 : 1000 1 : 5000 Dysentery B., Japan . . . + — — — Paradysentery, Mt. Desert . + — — — Paradysentery, Manila . . ++ ++ ++ — Colon B.X + + + — — 206 AGGLUTINATION AND PRECIPITATION The Relative Accumulation of the Group and Specific Agglutinins for the Organism Injected and for Allied Varieties. — ^A test was carried out with different types of dysentery bacilli. For the Manila culture of Flexner, which is nearest to the colon in its characteristics, the specific agglutinins were, in the serum of an animal which had received injec- tions of the Manila cultures, at the end of the fourth month five times as abundant as the group agglutinin acting on the Mt. Desert culture of Park, which represents a type lying between the Flexner and Shiga cultures. For the dysentery bacillus (Shiga) the development of agglu- tinins was the least (Fig. 74). 1st 2d 3d 4th 5th 6th* 7th 1: 500 — ^ I: 400 / > I: 300 / \ l: 2 00 y / .^v V |: 100 y • / \ V . i: 00 ^niiiv^^^ •r^:-'^ - ,"■ ^N Fig. 74. — The rise and fall of common and specific agglutinins during seven months in a rabbit injected with the Manila culture. Colon bacillus X. ■ Paradysentery type (Mt. iPesert). • — • — • ^ Paradysentery type (Manila). Dysentery type (Japan). . Test dates for all four sera. * Injections stopped. Another point of interest is that the proportional amounts of agglu- tinins from the different cultures varied at different times. If on tests made of a single bleeding we had attempted to draw conclusions as to the relative development of specific and group agglutinins between the cultures, we should have had an imperfect view. Many conflicting statements in literatiu-e are undoubtedly due to this lack of appreciation of the variability in the relative amount of these two types of agglutinins during a long process of immunization (Fig. 75). The Use of Absorption Methods for Differentiation between Specific and Group Agglutinins due to Mixed Infection and to a Single Infection. — It is now well established that if an infection is due to one micro- organism there will be specific agglutinins for that organism and group agglutinins for that and other more or less allied organisms. If infection is due to two or more varieties of bacteria, there will be specific agglu- tinins for each of the microorganisms and group agglutinins produced because of each of them. The following experiments will illustrate these points: A rabbit immunized to B. typhi agglutinated B. typhi 1 to 5000, B. coli 31 1 to 600. After saturation with B. typhi all agglutinins were removed for both microorganisms. A rabbit immunized to both B. typhi and B. coli 31- agglu- tinated B. typhi! to 4000, B. coli 31 1 to 1000, After saturation with 5. THE NATURE OF THE SEBUM SUBSTANCES CONCERNED 207 typhi the serum did not agglutinate B. typhi, but agglutinated B. coli 31 1 to 900. After saturation with B. coli 31 it failed to agglutinate B. coli 31, but still agglutinated B. typhi 1 to 3500. Some other strains of B. coli still agglu- tinated in 1 to 20 or more because many strains included in this group act as differently toward each other in respect to agglutinins as they do to the typhoid bacilli. ISOOO 1 :4500 / 1 :4000 1:3500 1 :3000 ^- lasoo x .^r 1 I.-2000 ^ ^/ |:i500 >^ ^y 1:1000 ^>^^ .x-D-y I: 500 ^^/ l: 00 . -^ FiG. 75. — Similar conditions to those noted in previous chart, except that a young goat has been used for the injections of the colon bacillus X. The great accumulation of common agglutinins for the paradysentery bacillus in the third month of the injections of the bacillus X is very striking. . Tests made. The foilowing tables give the outcome of several experiments: Absobption by the Typhoid Bacillus of Group Agglutinins Acting UPON A NuMBEiB OF VABIETIES OF B. CoLI WHICH WEEE PeODUCED BY Another Variety of B. Coli. Agglutination by Serum OF Rabbit Immunized to B. Coli X. Before addition of After attempt at absorption with typhoid bacilli at 22° C. typhoid bacilli. 6000 5000 500 20 500 30 250 30 250 10 10 less than 10 less than 10 less than 10 less than 10 less than 10 Colon bacillus X Colon bacillus 1 Colon bacillus 2 Colon bacillus 3 Colon bacillus 4 Colon bacillus 5 Colon bacillus 6-18 Tjrphoid baoUlus The absorption tests were carried out by adding the bacilli from recent agar cultures to a 10 per cent, solution of the serum in a twenty- four-hour bouillon culture. The mixture was allowed to stand for twenty-four hours at about 22° C. It was found that the agglutinin in a simple dilution of serum when left at 37° C. rapidly deteriorated. Thus, in an extreme instance a serum positive at 1 to 1500, when diluted with bouillon or salt solution 1 to 25 and left at 37° C. for twenty-four hours, lost 30 to 40 per cent, of its strength; at 22° C. it lost at times 15 to 20 per cent. Left for three hours, the loss only was 5 to 10 per cent. 208 AGGLUTINATION AND PRECIPITATION l=2000„„,, - I:i800 E S i *■ .H .5' l:i600 ^ II f 1 ^' ■ € i ; l:l400 .= si 1, ^ Mar i,a^ •? . , l:i200 o Manila t l:iOOO HtDesert. Coney | 0) 0) |: 800 Japan | No™il 1^ c £ 1: 600 1 1 1 ^ 1" s - ■D I: 400 1 1 ' 1 T -g 1 1 ' 1 ' -^ 1 ..1 I Si 1 *" I: 200 Ml' 1 ^1 sri ri l: 100 i i : 1 1 ^i o<3 fe 1 "1 l: 00 1 i 1 i Till! Yl: fi 1 i Fig. 76. — Showing the effect of saturating with bacilli of types of Shiga, Manila, and Mt. Desert, a serum from a horse which had received combined injections of dysentery bacilli of the three types. Note that the Manila type removed almost all the specific and group agglutinins acting upon its own type and the group agglutinins upon the Coney Island and normal types, leaving the specific agglutinins for types Shiga and Mt. Desert. The same is true for types Shiga and Mt. Desert when they were used. ■ — Manila paradysentery. Japan dysentery. Mt. Desert paradysentery. Atypical paradysentery. Normal. The great number/ of varieties of the colon group of bacUli that are in the normal intestine and which are absorbed slightly in health and more markedly in intestinal diseases make the use of absorption test for diagnostic purposes too compHcated except for peculiarly important cases. Loss of Capacity in Bacteria to be Agglutinated or to Absorb Agglu- tinins because of Growth in Inunune Sera. — ^The loss of these character- istics by growth in sera has been demonstrated by Marshall and Knox. The experiments of Collins and ourselves are recorded because they were undertaken in a slightly different way and also because a certain number of confirmatory observations are of value. The maltose-fermenting paradysentery bacillus of Flexner was grown on each of eleven consecutive days in fresh bouillon solutions of the serum from a horse immunized through oft-repeated injections of the bacillus. The solutions used were 1.5,' 4, and 15 per cent. The serum agglutinated the culture before its treatment in dilutions up to 1 to 800, and was strongly bactericidal in animals. After the eleven transfers the culture grown in the 15 per cent, solution ceased to be agglutinated by the serum and ceased to absorb its specific agglutinins. The cultures grown in the 1.5 and 4 per cent, solutions agglutinated well in dilutions up to 1 to 60 and 1 to 100 and continued to absorb agglutinins. The recovery of the capacity to be agglutinated was very slow, the culture being transplanted from time to time on nutrient agar. After growth for sixteen weeks, during which it was transplanted forty-three times, it agglutinated in dilutions of 1 to 200. The culture grown in 4 per cent, agglutinated 1 to 500, and the one in 1.5 per cent. 1 to 800. This diminution and final cessation of development of agglutinable substance in bacteria grown in a serum rich in agglutinin and immune THE NATURE OF THE SEBUM SUBSTANCES CONCERNED 209 bodies is interesting both as showing the variation of the bacteria and as one means of adapting themselves to resist destruction, since the bacteria which ceased to be agglutinated or absorb agglutinin probably also were less affected by other antibodies. These changes are probably due to alteration in the external layer of the cells, as can be noted in those forming visible capsules such as pneumococci. It is possible that there is also less development of agglutinogens. Non-agglutinable Strains.^-Sometimes strains of microorganisms are obtained which possess all the other characteristics of a type and yet do not agglutinate. When grown on artificial media for some time they may gain the property. Organisms freshly obtained from the blood and tissues frequently agglutinate to a much less degree than after several transfers on media. 4th mo. 6th 12th 14th 16th Fig. 77. — Relation of agglutinative to bactericidal power. Horse injected with culture of dysentery bacilli over a period of sixteen months. Agglutination index. Bactericidal index. Test dates. Relation between Agglutinating and Bactericidal Power. — ^In spite of proof to the contrary good observers hold to the belief that there is some relation between the agglutinating and the bactericidal strength of a serum. The tests we carried out on the serum of a number of horses showed no such relation. In Fig. 77 are recorded a number of comparative tests during a period of sixteen months. While such an experiment shows no definite relation between agglutinin and immune bodies it must be remembered that these antibodies are the products of processes which are governed by similar laws, and so they have many points in common. Variation in the Agglutinating Strength of a Serum. — There is usually a continued increase in the amoxmt of agglutinin in the blood of an infected person from its first appearance at the end of from four to twenty-one days until convalescence and then a decrease. At times, however, there is a marked variation from day to day, so that it may be abundantly present one day and very slight in amount the next. 14 210 AGGLUTINATION AND PRECIPITATION Mode of Obtaining Serum from Blood or Blisters for Examination.— Fluid blood serum can easily be obtained in two ways: First, the serum may be obtained directly from the blood, thus: The tip of the finger or ear is pricked with a lancet-shaped needle, and the blood as it issues is allowed to fill by gravity a capillary tube having a central bulb. The ends of the tube are then sealed by heat or with melted wax, or candle-grease, and as the blood clots a few drops of serum separate. To obtain larger amounts of serum for a microscopic examina- tion the blood is milked out from the puncture into a small homeopathic vial or test-tube. One cubic centimeter of blood can easily be collected in this way. The vial is then corked and placed on the ice to allow the serum to separate. As a rule one or two drops of serum are obtain- able at the end of three or four hours. Second, the serum may be obtained from blisters. This gives more serum, but causes more or less delay. The method is as follows: A section of cantharides plaster, the size of a 5-cent piece, is applied to the skin at some spot on the chest or abdomen. A blister forms in from six to eighteen hours. This should be protected from injury by a vaccine shield or bunion plaster. The serum from the blister is collected in a capillary tube, the ends of which are then sealed. Several drops of the serum can easily be obtained from a blister so small that it is practically painless and harmless. The serum obtained is clear and admirably suited for the test. A piece of blotting paper soaked in strong ammonia when placed on the skin and covered by a watch-glass or strips of adhesive plaster will quickly raise a blister. A little vaselin should be smeared on the skin surrounding the blotting paper. The whole blood may be used. It is dropped on glass or stiff paper, dried, and when it reaches the laboratory it is brought into solution. The Reaction. — The reaction can be observed either macroscopically by sedimentation of the agglutinated clumps in a test-tube or on a glass sUde, or microscopically in a hanging drop. The Macroscopic Tube Method of Estimating Amount of Agglutinins. — The tests are carried out with sets of test-tubes in racks as in the com- plement-fixation tests (see Chapter XI). Some prefer tapering tubes so that the sediment can be more easily estimated. As in complement- fixation tests, great care must be taken to have the test-tubes clean and free from chemicals. The bacterial suspension used, as in the micro- scopic test, must be standardized. The addition of the bacilli on a twenty-four-hour slanted agar tube to 150 c.c. of normal saline solution gives a good suspension. There must be a uniform cloudiness, not too heavy, with no signs of spontaneous agglutination. A series of dilutions are made in which the suspensions of bacteria are of similar strength. Salt Suspension of Final ube. Serum, solution, bacteria. solution, CO. c.c. C.G. c.c. 1 0.4 1.5 0.1 1 to 5 2 0.2 1.7 0.1 1 to 10 3 0.1 1.8 0.1 1 to 20 4 0.05 1.85 0.1 1 to 40 5 0.0 1.9 0.1 control THE NATURE OF THE SERUM SUBSTANCES CONCERNED 211 After thorough mixing, the tubes are allowed to stand in the incu- bator for one hour. They are then observed and the amount of floccu- lent precipitate noted. They are then placed in the ice-box at 10° or under, examined in twelve to twenty-four hours, and readings again made. Some varieties of bacteria agglutinate much more rapidly than others and for each, one must learn the proper time for reading the results. The tubes holding the greatest dilution of serum in which the fluid has cleared by the complete precipitation of all the bacteria shows the measure of the agglutinins. The control tube must be closely examined to note any sedimentation. Heating the mixed serum solution and bacteria from thirty to sixty minutes to 55° C. gives good results for some varieties of bacteria. The Macroscopic Slide Method. — This method allows a rapid diag- nosis of colonies from plates inoculated with suspected material, such as feces, and can be employed in examinations for cholera, typhoid, para- typhoid and dysentery bacilli. A highly potent serum, whose specific and group agglutinating strength is known, should be used, or false posi- tive results will be obtained; hence the method is not of much worth in clinical diagnosis of blood. The method is as follows: A loopful of saline solution (as control) and one of the highly potent specific serum in low dilution are placed on one slide, and a sufficient amount of the suspected colony to give a slight turbidity is added to each. Flocculation begins in the serum almost at once if the organism tested is specific. A negative reaction is not exclusive as relatively inagglutinable strains may be infrequently encountered, though with highly potent serum there are always some evidences of a reaction. Colonies apparently typical but not distinctly agglutinable should be fished for further identification. Colonies giving a positive reaction should also be fished for verification unless experience has shown that the serum used does not give false positive results with allied types. Microscopic Reaction. — If the reaction is observed through the micro- scope in a hanging drop, a formation of clumps is seen which, if it takes place rapidly, reveals the reaction almost completed at the first glance, that is, most of the bacilli are in loose clumps and nearly or altogether motionless (Figs. 78 and 81). Between the clumps are clear spaces containing few or no isolated bacilli. If the reaction is a little less com- plete a few bacilli may be found moving slowly between the clumps in an aimless way, while others attached to the clumps by one end are apparently trying to pull away, much as a fly caught in fly-paper struggles for freedom. If the agglutinating substances are present, but still less abundant, the reaction may be watched through the whole course of its development. Immediately after mixing the blood and the culture together it will be noticed that the bacilli move more slowly than before the addition of serum. Some of these soon cease all progressive movement, and it will be seen that they are gathering together in small groups of two or more, the individual bacilli being still somewhat separated from each other. Gradually they close up the spaces between 212 AGGLUTINATION AND PRECIPITATION them, and clumps are formed.- According to the completeness of the reaction, either all of the bacilli may finally become clumped and im- mobilized or only a small portion of them, the rest remaining freely motile, and those clumped may appear to be strugghng for freedom. With blood containing a large amount of agglutinating substances all the gradations in the intensity of the reaction may be observed, from those shown in a marked and immediate reaction to those appearing in a late and indefinite one, by simply varying the proportion of blood added to the culture fluid. The reaction takes place more quickly when put in the incubator at a higher temperature (36° C.) . Fig. 78. — Grilber-Widal reaction. Bacilli gathered into one large and two small clumps, the few isolated bacteria being motionless or almost so. Pseudoreactions. — If too concentrated a solution of dry blood from a healthy person is employed a picture is often obtained which may be mistaken for a reaction. Dissolved blood always shows a varying amount of detritus, partly in the form of fibrinous clumps, and prolonged microscopic examination of the mixture of dissolved blood with a culture fluid shows that the bacilli, inhibited by substances in the blood, often become more or less entangled in these clumps, and in the course of one-half to one hour very few isolated motile bacteria are seen. The fibrinous clumps alone, especially if examined with a poor light by a beginner, may be easily mistaken for clumps of bacilli. Again, the bacilli may be immobilized after remaining for one-half to two hours, by slight drying of the drop or by becoming attached to clumps of fibrin or other detritus. The reaction in disease is chiefly due to specific substances, but clumping and inhibition of movement similar in character may be caused by group agglutinins such as exist in normal horse and other serums. This is a very important fact to keep in mind. In order to help the student to thoroughly understand what com- prises a reaction Wilson prepared a set of drawings, which are here reproduced. The culture to be tested should be of about twenty hours' growth, either in bouillon or on agar. If on the latter a suspension THE NATURE OF THE SERUM SUBSTANCES CONCERNED 213 is made in broth or normal salt solution. A loopful of the fluid con- taining the bacteria is placed on the cover-glass, and to it an equal quantity of the desired serum dilution is added. '*''"" v^'4o' ^-"mm;^^ Fig 79. — Microscopic field, showing the top of a hanging drop in a normal typhoid culture. Fig. 80. — Microscopic field, showing a cross-section of the drop in Fig. 79. In making the hanging drop to be examined it is necessary to have it of such a depth that it will show at least three focal planes, other- wise the examination will be incomplete and unsatisfactory. The moist chamber must be well sealed by vaselin so as to prevent drying, and kept at a temperature of at least 20° and not over 35° C. Fig. 81. — Microscopic field, showing the top of a drop with the typhoid reaction. Fig. 82. — Microscopic field, showing a cross-section of the drop in Kg. 81. Fig. 79 shows a microscopic field of the top of a hanging drop of a normal bouillon culture of typhoid bacilli. The culture is twenty 214 AGGLUTINATION AND PRECIPlfAtlON hours old and the organisms are freely motile. This represents the eotltrol drop used for comparison with the drop of the same culture to which has been added a little of the blood of a person suspected to have typhoid. Note these points in Fig. 79; the organisms are eveilly distributed throughout the field, except at the edge of the drop, where they are gathered in great numbers; they show great activity here, seemingly trying to crowd to the very edge. This attraction is prob- ably due to the action exerted on the organisns by the oxygen in the air, which naturally exerts positive chemotaxis oti all aerobic organisms. Fig. 80 shows a cross-section of the drop represented in Fig. 79, and it will be noticed that the bacilli are evenly distributed throughout the drop, except at one place in the focal plane a, and again in the focal plane c. It sometimes happens that there is a substance adhering to a sup- posedly clean cover-glass which attracts the bacilli to that point, where they appear as fairly well-defined clumps, more or less like the true clumps due to the agglutinating substance in typhoid blood. The increase in organisms at the bottom of the drop in the focal plane c is easily accounted for by the fact that gravity naturally carries the dead and non-motile organisms to the bottom, these frequently assuming the character of clumps. If a field can be found in any focal plane of the hanging drop free from clumps, one can be quite sure that any clumping present is not due to any agglutinating substance which necessarily will affect organ- isms in every focal plane. Fig. 81 shows the microscopic appearance of the tof of a drop where the reaction is present. Notice first that the organisms have been drawn together in groups and that the individuals of each group appear to be loosely held together. Viewed under the microscope these clumps are practically quiescent, there being very little movement either of the individual organisms or of the clump as a whole. The edge of the drop is practically free from organisms, showing that the air no longer ' exerts any influence on them. Fig. 82 shows a cross-section of the hanging drop shown in Fig. 81. The clumps are evenly distributed throughout the drop, with perhaps some increase in the numbers and compactness of the clumps at the bottom. Fig. 83 shows the microscopic appearance of the tof of a hanging drop of a bouillon culture to which has been added some blood of a patient sufl^ering from a febrile condition not caused by typhoid infec- tion, but which exerts a slight non-specific influence on the typhoid organisms. It will be seen that the reaction is incomplete and that there are many organisms at the edge of the drop. The air exerts the same influence on the bacilli that it did before the addition of the blood. Note the character of the clumps, generally small and compact at the centre, 'with the bacilli at the edge of the clump, usually attached by one end only. THE NATURE OF THE SERUM SUBSTANCES CONCERNED 215 Very frequently these clumps have the appearance of being built up around a piece of detritus present in the clump. All the organisms comprising the clump seem to have retained part, at least, of their motility, those on the edges being particularly motile, so far as their free ends are concerned. When motility is very much inhibited these clumps have a peculiar trembling movement, which is like the molecular movement described by Brown. Fig. 84 shows a cross-section of the drop represented in Fig. 83. Note the same character of the clumps in every focal plane: the large number of motile bacilli and the number attracted to the edge of the drop by the air. Fig. 83. — Microscopic'field, showing the top of a drop of culture with reaction not due to typhoid. Fio. 84. — Microscopic field, showing a cross-section of Fig. 83. Comparison of Tube and Microscopic Slide Methods. — ^The reaction is the same in both and one is as reliable as the other. The ice-box readings are apt to be higher than those obtained by microscopic examination. For diagnostic examinations where haste is necessary and small amounts of serum are available, as in typhoid fever, the microscopic method is preferred. When a delay of twenty-four hours is no handicap and the serum is abundant as in tests for glanders in horses the macroscopic tube test is chosen. For the identification of bac- teria the macroscopic test is generally used. Dead cultures are more frequently used in the macroscopic method because the motility is of no importance. As noted above, the growth of bacteria in fresh blood containing agglutinins inhibits the development of agglutinable substance in bac- teria or causes them to produce substances which prevent the union of agglutinin with them. Bacteria should therefore not be grown on serum media when they are to be used in agglutination tests. Even the addi- tion of ascitic fluid to broth has some effect. 216 AGGLUTINATION AND PRECIPITATION THE PHENOMENON OF PRECIPITATION. Precipitin. — ^The injection into the animal body of the protein sub- stances of bacteria or of other cells was found to stimulate the develop- ment of agglutinin almost as much as the injection of the unaltered cell. Kraus, in 1897, found that when a little immune serum was added to the bacteria-free filtrate of a culture of the organism used to produce the immunization there occurred a precipitate. This he designated as the "precipitin reaction." This same reaction took place between the serum of an animal and various protein substances, such as white of egg, blood serum, or milk, with which the animal had been injected. That part of the protein which produces the precipitin is called precipi- tinogen (antigen). Precipitins in their development, their resistance to heat and chemicals, and in their specific and non-specific forms are similar to agglutinins. The specificity of precipitins is, like that of the agglutinins, not absolute. Group precipitins act upon similar chemical substances derived from cells having very different characteristics. The precipitin test is usually employed in blood identification and in testing sera and tissue extracts rather than bacterial filtrates. As the reaction depends on the formation of a precipitate it is iinportant that the solution of serum and antigen be absolutely clear before being placed together. With bacterial antigens the test is carried out by placing a constant amount of the precipitinogen in a row of test-tubes and adding decreasing amounts of immune serum (precipitin). A series of control tubes of each serum should be observed. The tubes are placed in the incubator. With potent serums precip- itation becomes visible after one-half to two hours. With serum or tissue extracts as antigens a constant amount of antiserum is placed in a series of tubes and then equal parts of dilutions of the antigen, the dilutions ranging from 1 to 10 to 1 to 5000 or beyond. A very highly potent serum will give a precipitate with even dilutions of 1 to 10,000 or beyond. Production of Precipitating Serums. — Animals, usually rabbits, are injected with bacterial suspensions heated to 60° C. or with the filtered bacterial extracts obtained by emulsifying bacteria grown on agar in salt solution and shaking in a shaking machine for forty-eight hours. The injections are made just as in the production of agglutinins, but a longer period of immunization is necessary to produce a highly potent precipitating serum. Sera or protein extracts are injected intravenously in amounts of from 1 to 3 c.c. at 5-6 day intervals. Three or four injections usually suffice. Some rabbits do not respond well so that several should be injected. CHAPTER XIII. OPSONINS. OPSONIC INDEX. LEUKOCYTE EXTRACT. OPSONINS. We find that phagocytosis is most mar Iced when the disease is on the decline or the infection mild, but is usually absent in rapidly increasing infection. This would seem to indicate that the course of the infection is often abeady determined before the leukocytes become massed at the point of its entrance. The first determining influence is given by the condition of the tissues and the amount of bactericidal substances contained in them, and then, later, in cases where the bacteria have been checked, comes the additional help of the leukocytes. If the tissues are wholly free from bactericidal and sensitizing substances, neither they nor the leukocytes, nor both combined, can prevent the increase. The simple engulfing by the cells of bacteria is not necessarily a destructive process. (See also p. 164). The interest in the subject of the opsonins is largely due to the investigations and influence of Wright. We should, however, recognize the important earlier work of others. Denys and Leclef had previously shown that in the case of rabbits immunized against streptococci, the increased phagocytosis was due to an alteration in the serum and not to changes induced in the leukocytes. They demonstrated that the leukocytes of the immunized animal when placed in normal serum showed no greater phagocytic activity than normal leukocytes did, and that therefore the substances in the serum favoring phagocytosis united with the bacteria. Wright and Douglass showed definitely that phagocytosis of bacteria in normal serum depends upon a special sub- stance in the serum which becomes fixed to the bacteria and prepares them for phagocytes. They called this substance opsonin. Neufeld and Rimpau discovered the same point independently. Wright dealt mostly with normal serum, while Neufeld used serum from immunized animals. Wright originated the idea of estimating the changes in the opsonic power of the blood for the purpose of guiding the use of vaccines in the treatment of bacterial infections. Thus he states: "I have found that there exists in the serum of the successfully inoculated patient an increase of opsonin. This is a substance which lends itself to very accurate measurements by a modification of Leishman's method. By the aid of this method the patient's progress or regress can be very accurately followed." Where vaccines are injected, Wright states, there "supervenes a negative phase where there is a diminished content in protective substances. This is succeeded by a positive phase. This inflowing wave of protective substances rapidly flows out again, but leaves behind in the blood a more or less perma- nently increased content of protective substances. When a small dose of vaccine is given the negative phase may hardly appear, but the positive phase may be correspondingly diminished. Where an unduly large dose of vaccine is 2l§ OPSONtNS— OPSONIC I NDeX— LEUKOCYTE EXTRACT inoculated the negative phase is prolonged and much attenuated. The posi- tive phase may in such a case make default. "It will be obvious that, if we, in the case of a patient who is already the subject of a bacterial invasion, produce by the injection of an excessive dose of vaccine a prolonged and well-marked negative phase, we may, instead of benefiting the patient, bring about conditions which will enable the bacteria to run riot in his system. "Now consideration will show that we may obtain, according as we choose our time and our dose wisely or unwisely, either a cumulative effect in the direction of a positive phase or a cumulative effect in the direction of a nega- tive phase. We may in other cases, by the agency of two or more successive inoculations, raise the patient by successive steps to a higher level of immun- ity, or, as the case may be, bring down by successive steps to a lower level. We can select the appropriate time and dose with certainty only by examining the blood and measuring its content in protective substances in each case before reinoculating." I.T 1.0 1.5 1.4 1.3 1.3 1.1 l.O .0 .8 .7 .6 .5 ' ' \ \ \ n \ / \ - j 7 \ \, A ,\ lA K. ^ — < • ' ^ / / <• / / V ^ V DATE 3| 1*1 \'o\ \b\- \l\ \s\ |a| |lo| |ll| |l2| |l:)| |lt| |l5| OofohKi-, 19 ){( Fig. S5. — Opsonic curve showing the slight immediate rise and the later negative and positive phases following inoculation. The changes here are more regular than generally occurs. ^ An immense amount of investigation has revealed the fact that an estimation of the opsonin cannot be obtained accurately enough in single tests to be a safe guide to be used in diagnosis or treatment unless the variation from the normal is exceptionally great, and that the opsonic content is not alone a safe guide for the measure of the total antibodies in the blood. THE OPSONIC INDEX. Technic. — Wright's technic of measuring the opsonic power is a slight modification of the Leishman method and is as follows: An emulsion of fresh human leukocytes is made by dropping twenty drops of blood from a finger prick into 20 c.c. normal salt solution containing 1 per cent, sodium citrate. The mixture is centrifuged, the super- natant clear fluid removed, and the upper layers of the sedimented blood cells transferred by means of a fine pipette to 10 c.c. normal salt solution. After centrifuging this second mixture the supernatant THE OPSONIC: INbEX 2i9 fluid is pipetted off and the remaining suspension used for the opsonic tests. Such a "leukocyte emulsion," of course, contains a mixture of leukocytes and of red blood cells; the proportion of leukocytes, how- ever, is touch greater than in the original bloods The bacterial emulsion Fig. 86. — Opsonic outfit. 220 OPSONINS— OPSONIC INDEX— LEUKOCYTE EXTRACT is prepared by gently rubbing a little of the culture to be tested in salt solution (0.85 to 1.2 per cent.). When thoroughly mixed the fluid is centrifuged for a few minutes so as to remove any clumps. The emulsion should be so thick that in a trial test the leukocytes take up about five apiece on the average. One volume of the leukocytes is mixed with one volume of the bac- terial emulsion and with one volume of the serum to be tested. This is best accomplished by means of a pipette, suggested. by Wright, whose end has been drawn out into a capillary tube several inches in length. With a mark made about an inch from the end it is easy to suck up an equal volume of each of the fluids, allowing a small air-bubble to inter- vene between each volimie. All three are now expelled on a slide and thoroughly mixed by drawing back and forth into the pipette. Then the mixture is sucked into the pipette, the end sealed, and the whole put into the incubator at 37° C. (Fig 86, d and e). The identical test is made using a normal serum in place of the serum to be tested. Both tubes are allowed to incubate fifteen minutes and then the end of the tube is broken off, a large drop mounted on a clean slide the surface of which was previously roughened by emery-paper, and a spread made with a second slide as in ordinary blood work, only a little thicker and using no force whatever. After drying in the air the smears are stained without previous fixation either with a 1 per cent, aqueous solution of methylene blue or some other suitable stain. The degree of phagocytosis is then determined in each by counting a consecutive series of fifty or one hundred leukocytes and finding the average number of bacteria ingested per leukocyte. This number for the serum to, be tested is divided by the niunber obtained with the normal serum (obtained by pooling the serum of healthy persons) and the result is regarded as the opsonic indegc of the serum in question. Thus, if the tubercle baciUi, sensitized by a patient's blood, are taken up by the leukocytes to the average number of three per leukocyte, and bacilli from the same emulsion sensitized by normal blood are taken up by leukocytes to the average of five, then the index will be three-fifths of one, or 0.6. Phago- cytic index (normal pooled blood), 5; phagocytic index (patient's serum), 3; opsonic index, | or 0.6. In this case the index would indicate a deficiency in opsonins. The presence of a high opsonic index Wright regards as indicative of increased resistance. He further states that the fluctuation of the opsonic index in normal healthy individuals is not more than from 0.8 to 1.2, and that an index below 0.8 is therefore almost diagnostic of the presence of an infection with the organism tested. A number of workers u3e the washed whole blood. This saves in original labor, but makes the search for a suitable niunber of leukocytes more difficult. The Dilution or Extinction Method recommended by Dean and by Klein. The degree of dilution of the serum necessary for the extinction of its opsonic index is determined; that is, the serum to be tested is diluted until a dilution is found which shows the same small amount THE OPSONIC INDEX 221 of phagocytosis shown in preparations in which no serum is used, namely, an index below 0.5. Klein claims that results by this method are more accurate than by the method of Wright. The method is too tedious for practical use in routine work. Combined Method. — Simon esthnates the percentage of phagocyting cells in the mixture containing the serum to be tested and compares this with the mixtures containing normal serum. He does this not only with the undiluted serum, but also in dilution of 1 to 10 or 1 to 100 in salt solution. Most workers now agree that the use of the opsonic index is limited to experimental investigations. The reasons for this opinion follow: The AcctTRAcy with which the Opsonic Power of the Blood Can be Determined by Wright's Method. — ^An examination of any slide will show that the different leukocytes vary in their size and in their ■ content of bacteria. This is due partly to variation in phagocytic activity, and partly to the interference of the red blood cells, which are present in great numbers in the emulsion and separate the bacteria in different degrees from the white cells. These and other reasons bring it about that the different leukocytes vary greatly in the number of bacteria they take up and in their distribution on the slide. Partly to overcome this, large numbers of leukocytes are counted. Beyond one hundred, or at most one himdred and fifty, the increase of accuracy hardly compensates for the extra labor. The following table shows the difference between counting larger or smaller numbers of cells in five opsonic tests as determined by counting different numbers of cells in one specimen: Opsonic Index Estimations in Five Blood Specimens. Cells counted. Average number of bacteria in each leukocyte. 50 ... . 1.18 1.88 1.34 1.42 1.90 100 ... . 1.22 1.78 1.24 1.42 1.59 150 ... . 1.18 1.62 1.22 1.44 1.50 200 ... . 1.18 1.51 1.22 1.46 1.37 600 ... . 1.28 1.62 1.23 1.36 1.36 1,200 .... 1.34 1.44 1.25 1.30 1.42 It is noticed that the variation between the average cell count obtained from fifty cells and larger numbers is much greater than between that obtained at from one hundred or one hundred and fifty. It is necessary to have the counts that are compared all counted by the same person, as each individual has a somewhat different method and may average higher or lower for all counts than any other person. When two specimens of blood are tested not only the inaccuracy of counting due to the different arrangement of the unequally filled cells on the slides to be counted is met, but the fact that in making the test the conditions are not similar, for in different mixtures slightly dif- ferent proportions of leukocytes, bacteria, and red cells will always be mixed together. If smears from a series of tubes of the same blood are compared with a series of smears from one of the tubes, the former will always show the greater variation. 222 OPSONINS— OPSONIC INDEX— LEUKOCYTE EXTRACT This variation is much greater than most examiners believe. North has collected a series of tests carried out in nearly all the important laboratories in the Eastern United States that are working upon opsonins. The results recorded prove absolutely that while an average counting error of only about 10 per cent, is present, there may be an exceptional error of at least 100 per cent., and one of at least 20 per cent, may be expected once in about every ten determinations. The following is a fair average of the correctness of routine tests by experienced workers. Absolute Count of Bactehia in One Hundebd Leukocytes. Blood specimen A. Blood specimen B. Blood specimen C. Tube 1 . . .156 Tube 1 . . .142 Tube 1 . . . 89 Tube 2 . . .168 Tube 2 ... 182 Tube 2 . . .102 Tubes ... 172 Tube 3 ... 188 Tube 3 ... 121 This error, which occurs because of the technic, applies not only to the examination of the specimen of blood, but also to the measure we employ to estimate the amount of opsonins. As these are not stable, we cannot have a standardized solution, as we do with anti- toxins. We must therefore determine our measure afresh in each test, taking for this purpose a supposedly normal blood. Wright, from a great many tests, has determined that the opsonic power of the blood in non-infected persons for tubercle bacilli does not vary, as a rule, more than 10 per cent, above or below the average power of healthy blood. For staphylococci there is more variation. It is found also that many things besides infection decrease the amount of opsonins in the blood. Hemorrhage, fatigue, starvation, and other influences which lower the resistance of the body have this effect. - Wright gets this measure as uniform as possible by determining the average opsonic strength of five supposedly healthy persons at the time of each test. If any one of these five is considerably below or above others it is omitted for the day. The measure so obtained will probably vary about 5 per cent, from day to day, though seldom getting far away from what we might call the absolute normal. The following results were obtained by us from examining at one test a number of supposedly normal persons against staphylococci. The variation is greater than in Wright's reports. Opsonic Counts in Test of Twenty-one Normal Seha with Stock Staphylococcus Culture. 1 . . . . . 4.13 8 . . . . 3.82 2 . . . . . 2.93 9 . . . . 3.95 3 . . . . . 2.78 10 . . . . 3.98 4 . . . . . 4.37 11 . . . . 4.27 5 . . . . . 3.58 12 . . . . 3.69 6 . . . . . 2.90 13 . . . . 3.80 7 . . . . . 3.56 14 . . . . 3.59 15 ... . 9.09 16 ... . 5.17 17 ... . 4.04 18 . . . .3.82 19 . . . .4.00 20 ... . 3.79 21 .... 3.44 The Influence upon the Opsonic Test of the Specific Differ- ences BETWEEN Strains of a Single Species. — The general practice THE OPSONIC INDEX 223 in laboratories is to use stock cultures of tubercle bacilli, staphylococci, and other bacteria for the opsonic tests. To obtain a culture from a case may be at first impossible and, if successful, causes a delay of at least one or two days. The culture when obtained may also, as is frequently the case with pneumococci and streptococci, fail to opsonize readily. These and other reasons tend to establish the use of laboratory stock cultures, and yet we must acknowledge that when we test the amount of opsonins by both the stock and fresh cultures a marked difference sometimes develops. The Leukocytes to be Employed. — ^To many it seems a matter of indifference whether one person's leukocytes or another's are used, but our experience agrees with that of others that the leukocytes from different persons not only vary in their activity, but also in their selective action, and that the index is not the same when obtained with one person's leukocytes as with another's. The Influence of the Strength of the Bacterial Emulsion. — The more abundant the bacteria the greater will be the number taken up by the leukocytes. It is very important therefore that the tests be made with the same strength of emulsion. 3.0 -2.5 3.0 1.5 1.0 ^ 8 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 20 27 28 29 30 31 Fig. 87. — x = injection of vaccine. Three types of lines = 3 cases. The Opsonic Variation during Treatment by Inoculations.— Wright lays stress on the considerable uniformity of the degree and persistence of development of opsonins after inoculation. We have found in a small percentage of cases typical increases and decreases, as seen in Fig. 85, but in the majority of those inoculated there has been great irregularity. Frequently the negative phase does not occur or at least it is not detected. The following chart for three staphylo- coccus cases illustrates this (Fig. 87): The Variation from Day to Day in the Amount of Opsonins in Supposedly Healthy Persons.— It has already been noted that in getting our measure we test a number of persons and exclude the blood 224 OPSONINS— OPSONIC INDEX— LEUKOCYTE EXTRACT of those which varies greatly from the average. We are so in the habit of seeing the index of normal blood placed at unity because it is each day the measure of comparison that even investigators are apt to think of the indices of normal persons as being unchanged from day to day. This is not the fact. A glance at the next chart (Fig. 88), in which four cases of tuberculosis are charted together with two normal persons, shows that the variation is only slightly greater in infected than in normal cases. If one normal person is charted against another for several weeks, marked differences will usually appear. The indices of the twenty-one normal cases tested against staphylococci (page 222) illustrate this variation in the amount of opsonins in normal blood. March 19UI | 3 3 4 5 J s a lU 11 12 Vi TBO. 1.7 1.0 1.5 1.4 \ \ ' \ \ \ TBC. \ \ — tN \ A ! TBC. 1.3 1.2 \ l\ \ / \, . T \\ / \ jf\ M \ ^? s ' / \ -^ \ J / 1 CALLtHAN- TBO. BOLDUAN- 1.1 -^, -H.^ ^1 s A \ Aj f 1.0 .9 A \ W' \ J \ K s N K pA f X ^N \/ \ / i" ^--^ \ ■^ K^ •I A \ if k .8 .7 \ ''**-1» ^ :j<' \ ') )^ \ V- 1 — ~^^ / >'---, ^\M ^, \ ~v i ^:d Fig. 97. — Section of lung. X 150 ; blastomycetes in large syncytial cell masses. (From Fontaine, Hasse, and Mitchell.) Shortly after Tokishige's pubUcation a similar disease occurring in horses in Italy and southern France was identified as being caused by saccharomyces. Cultures of this yeast, however, differ somewhat Fig. 98. — Blastomycosis in infant aged eight months, showing lesion on left cheek. (Kessler.) from that obtained in Japan, so that Busse is inclined to regard the two as two different species of blastomycetes. Kartulis, in Alexandria, described about a hundred cases of a skin 244 THE PATHOGENIC MOLDS affection occurring in the gluteal regions of men and characterized by an elongated finger-like swelling, which breaks and emits a purulent discharge, forming an unhealed sinus. In the discharge and surrounding tissues are numerous blastomycetes which Kartulis after cultivation and study considered a variety of the ordinary fermenting yeast (Saccha- romyces cerevisise Hansen). The cases were cured by excising the growth. Kessler reported a skin lesion in an infant (Fig. 98) probably due to a similar blastomycete, since the lesions healed after treatment with potassium iodide. The description of the yeast isolated is too incomplete to identify it. Some years ago the attempt was made to connect the development of cancerous growth with blastomycetes. This Was due in a measure to a certain similarity between the yeasts and the cell inclusions or so- called "parasites" of cancer, and further, to the fact that 'when yeasts are injected into the animal body tiunor-like nodules are often developed at the site of inoculation and in the internal organs. But these nodules are not tumors in the pathological sense of the term, but merely masses of blastomycetes mixed to a very variable degree with inflammatory tissue proliferations. Yeasts are Gram-positive, or Gram-amphophile. They stain readily with other stains in the young state. Older forms stain very irregularly. They are cultivated — sometimes with difficulty directly from the tissues — on ordinary media, but best on media made slightly acid (p. 110). Some pathogenic varieties grow best at blood heat. Most varieties grow easily at room temperature. The important part played by yeasts in certain industries is treated in Part III. REFERENCES. AsHFOHD, B. K.: Studies in Moniliasis of the Digestive Tract in Porto Rico, Am. Jour. Med. Sc, 1915, cl, 680. Bahr, p. H.: Researches on Sprue, Tr. Soo. Trop. Med. and Hyg., 1914, No. 5, vii. Brown, P. K., and Cummins, "W. T. : A Differential Study of Coccidiodal Granuloma and Blastomycosis, Arch. Int. Med., 1915, xv, 608 (with bibliography). Busse: In KoUe u. Wassermann's Handbuch d. path. Mikroorg., 1913, Jena. Ind. Ed. Cummins, W. T., and Sander, J. : The Pathology, Bacteriology and Serol. of Coccidiodal Granuloma, Jour. Med. Res., 1916, xxxv, 243. De Beurmann et Gougebot: Traits des Sporotrichoses, Paris, 1912. - Fontaine, Hasse and Mitchell: Arch. Int. Med., 1909, iv, 101. Foster, M. H.: Favus and Ringworm of the Nails, Jour. Am. Med. Assn., 1914, Ixiii, 640. Hektoen and Perkins: Jour. Exp. Med., 1900, v, 77. Jackson, E.: Blastomycosis of the Eyelids, Jour. Am. Med. Assn., 1915, Ixv, 23. Kessler: Jour. Am. Med. Assn., 1907, xUx, 550. Klocker: Tr. by Allan and Millar, 1903, New York and London. Meyer, K. F. : The Relation of Animal to Human Sporotrichoses, Jour. Am. Med. Assn., 1915, Ixv, 579. Ricketts: Jour. Med. Res., 1901, vi, 377. Ruediger: Jour. Infect. Dis., 1912, xi, 193. Sabouraud: Ann. de dermat. et de syph.,1892 and 1893. Schenk: Johns Hopkins Hosp. Bull., 1898, p. 286. Wood, E. J. : The Occurrence of Sprue in the United States, Am. Jour. Med. Sc, 1915, cl, 692. Zinsser: Proc. New York Path. Soc, 1907. CHAPTER XVI. THE PYOGENIC COCCI. THE STAPHYLOCOCCI (MICROCOCCI). Practically all microorganisms have been shown by experiment to induce, under certain conditions, the formation of pus by their products when inoculated into the animal body; but, while this has been demon- strated, the researches of bacteriologists show that only a few species are usually concerned in the production of acute abscesses in man. Of these the two most important, by reason of their frequent occurrence and pathogenic power, are Staphyheocmis aureus and Streptococcus pyogenes. These two organisms are often found in the same abscess; thus, Passet, in 33 cases of acute abscess, found Staphylococcus aureus&lone in 6, aureus and albus associated in 11, albus alone in 4, albus and citreus in 2, Strepto- coccus pyogenes alone in 8, albus and Streptococcus in 1, and alhus, citreus, and Streptococcus in 1. The staphylococcus is likely to enter as a mixed infection into most infections due to other bacteria, and is almost always met with in all inflammations of the skin and mucous mem- branes or in cavities connected with them. Staphylococci were first obtained from pus by Pasteur in 1880. In 1881 Ogston showed that they frequently occurred in abscesses, and in 1884 Eosenbach fully demonstrated their etiological importance in circumscribed abscesses, osteomyelitis, etc. Of the staphylococci those producing yellow and white pigments are by far the most important since they are the pathogenic varieties. The Staphylococcus (Pyogenes) Aureus. — ^The Staphylococcus aureus is one of the commonest pathogenic bacteria, being usually present in the skin and mucous membranes, and is the organism most frequently concerned in the production of acute, circumscribed, suppurative inflammations. Morphology. — Small, spherical cells, having an average diameter of 0.7;u to O.Qfi, occurring solitary, in pairs as diplococc"!, in short rows of three or four elements, or in groups of four, but most commonly in irregular masses, simulating clusters of grapes; hence the name staphylococcus. (See Fig. 99.) Staining. — It stains quickly in aqueous solutions of the basic aniline colors and with many other dyes. When previously stained with aniline gentian violet it is not decolorized by Gram's method. When slightly stained each sphere frequently is seen to be already dividing into two semispherical bodies. Biology. — ^The Staphylococcus (pyogenes) aureus is a non-motile, aerobic, facultative anaerobic micrococcus, growing at a temperature 246 THE PYOGENIC COCCI from 8° to 43° C, but best at 25° to 35° C, The staphylcoccci grow readily on all the common laboratory media, such as milk, bouillon, nutrient gelatin, or agar. A slightly alkaline reaction to litmus is best for the growth of the staphylococci, but they also grow in slightly acid media. Cultivation. — Growth in Nutrient Bouillon. — ^The growth of the staphy- lococcus is rapid, reaching about 500,000,000 per c.c. at the end of twenty- four hours at 30° C. The bouillon is cloudy and frequently has a thin pellicle. Later a slimy sediment forms. The odor is disagreeable. In peptone-water, growth occurs with indol production. Growth on Gelatin. — Grown on gelatin plates it develops, at room temperatm-e, within forty-eight hours, punctiform colonies, which when examined under a low-power lens, appear as circular disks of a pale brown color, somewhat darker in the centre, and surrounded by a smooth border. The colonies grow rapidly. The appearance of the growth is most characteristic. Immediately surrounding the colonies, which are of a pale golden yellow color, there is a pitting of the surface of the gelatin, due to its liquefaction. By suit- able light a number of these shallow de- pressions with sharply defined outlines may be seen on the gelatin plate, having a diameter of from 5 to 10 mm., in the centres of which lie the yellow colonies. Later the liquefaction becomes general, the colonies running together. In stab cultures in gelatin a white confluent growth at first appears along the line of puncture, followed by a funnel-shaped liquefaction of the medium, which rapidly extends to the sides of the test-tube. At the end of two days the yellow pigmentation begins to form, and this increases in intensity for eight days. Finally, the gelatin is completely liquefied, and the staphylococci form a golden-yellow or orange-colored deposit at the bottom of the tube. Under unfavorable conditions the Staphylococcus aureus gradu- ally loses its ability to make pigment and to liquefy gelatin. The liquefi cation is due to a ferment called gelatinase formed by the staphy- lococci. It may be separated from the cocci by filtration (Loeb). Growth on Agar. — In streak and stab cultures on agar a whitish growth is at first produced, and this at the end of a few days becomes a faint to a rich golden yellow on the surface. The yellow pigmentation is produced only in the presence of oxygen; colonies formed at the bottom of a stab culture or under a layer of oil remain white. Milk. — ^Milk is coagulated at the end of from one to eight days. Potato. — ^The staphylococci grow readily on potato and produce abundant deep colored pigment. Growth on Loffler's Solidified Blood Serum. — Growth vigorous, with fairly good pigment production. Some varieties slowly liquefy the serum. Fig. -Staphylococcus. X 1100 diameters. THE STAPHYLOCOCCI ' 247 Growth on Blood Agar. — ^If nutrient agar to which a Httle animal blood has been added is streaked with staphylococci there appears, at the end of twenty-four hours at 35° C, about the growth a clear zone, owing to the hemolytic effect of the staphylococcus products. Acids Produced. — In media containing carbohydrates there is, as a result of the growth of the Staphylococcus aureus, a production of acid in considerable quantities, consisting chiefly of lactic, butyric, and valerianic acids. These acids have been supposed to play a part in the production of pus, in which, according to some observers, they are often present. No gas is formed. Resistance. — ^The staphylococcus is distinguished from most other non-spore-bearing pathogenic bacteria by its greater power of resistance to outside influences, desiccation, etc., as well as to chemical disin- fectants. Cultures of the Staphylococcus pyogenes in gelatin or agar retain their vitality for a year or more. Suspended in water its thermal death-point varies, with different cultures and averages about two hovirs at 50° C, one-half hour at 60° C, ten minutes at 70° C, and five minutes at 80° C. Upon silk threads and in media rich in organic matter its resistance is greater, but subjected to 80° C, for thirty minutes or boihng for two minutes it is almost surely killed. Cold has but little effect. Thirty per cent, of the organisms remained alive after being subjected by us to freezing in liquid air for thirty minutes. These are average figures. Some cultures are more resistant than others. They are quite resistant to direct sunlight and drying. Dried pus contains living staphylococci for weeks and even months, and they can be found alive in the fine dust of the air in living rooms and operating rooms. Resistance to chemicals is given in Part III. Pigment Formation.— Pigment formation, already described, is con- sidered within a limited species characteristic. Thus different strains of Staphylococcus aureus produce a pigment varying from a pale brown to a deep golden yellow. It usually becomes less intense upon prolonged cultivation. The pigment is classed as a lipochrome (Schneider). It is soluble in alcohol, chloroform and ether. In order to test the amount of color produced, Winslow and Rogers recom- mend the following method: A portion of the growth is removed on a loop needle and spread out on white drawing paper with a rough surface. After drying at room temperature the color is compared with a standard color chart. Pathogenesis. — The pathogenic effect of the Staphylococcus {pyogenes) aureus on test animals varies considerably, according to the mode of apphcation, the virulence of the special culture employed and the species of animal used. In man a simple rubbing of the surface of the imbroken skin with pus from an acute abscess is, as a rule, sufficient to produce a purulent inflammation, and the introduction of a few germs from a septic case into a wound may lead to a fatal pyemia. These conditions can only be reproduced in lower animals with difficulty, and by the inoculation of large quantities of the culture. Small subcutaneous injections, or the inoculation of open wounds in mice, guinea-pigs, and 248 ^ THE PYOGENIC COCCI rabbits, are commonly without result; occasionally abscess formation may follow at the point of inoculation, which usually ends in recovery. The pus-producing property of the organism is exhibited in proportion to the virulence of the culture employed. Slightly virulent cultures, which constitute the majority of those obtained from pus taken from the human subject, when injected subcutaneously in large quantities (several cubic centimeters of a fresh bouillon culture) into rabbits or guinea-pigs, give rise to local pathological lesions — acute abscesses. When virulent cultures are used — usually those recently isolated from human infections — 0.5 c.c. of a fresh bouillon culture is sufficient to produce similar results. The abscesses generally heal without treatment; sometimes the animals die from marasmus in consequence of the suppurative process. In intra- peritoneal inoculations the degree of virulence of the culture employed is still more evident in the effects "produced. The animals usually die in from two to nine days. The most characteristic pathological lesions are found in the kidneys, which contain numerous small collections of pus, and under the microscope present the appearances resulting from embolic nephritis. Punctiform, whitish-yellow masses of the size of a pea are found permeating the pyramids. Many of the capillaries and some of the smaller arteries of the cortex are plugged up with thrombi, consisting of micrococci. Metastatic abscesses may also be observed in the joints and muscles. The micrococci may be recovered in piu-e cultures from the blood and the various organs; but they are not numer- ous in the blood and are often difficult to demonstrate microscopically. Intravenous inoculations of animals are followed by similar pathological changes. Orth and Wyssokowitsch first pointed out that injection of staphylococci into the circulation of rabbits, whose cardiac valves have previously been injured, produced ulcerative endocarditis. Subsequently Weichselbaum, Prudden, and Fraenkel and Sanger obtained con- firmatory results, thus establishing the fact that when the valves are first injured, mechanically or chemically, the injection into a vein of a pure culture of Staphylococcus aureus gives rise to a genuine ulcerative endocarditis. It has been further shown by Ribbert that the same result may be obtained without previous injury to the valves by injecting into a vein the staphylococcus from a potato culture suspended in water. In his experiments not only the micrococci from the surface, but the superficial layer of the potato were scraped off with a sterilized knife and mixed with distilled water and the successful result is ascribed to the fact that the little agglomerations of micrococci and infected fragments of potato attach themselves to the margins of the valves more readily than isolated cocci would do. Not infrequently, also, in intravenous inoculations of young animals there occurs a localization of the injected material in the marrow of the small bones. This may take place in full-grown animals when the bones have been injured or fractured. The experimental osteomyelitis thus produced has been demonstrated to be anatomically analogous to this disease in man. An increase in virulence of certain strains may be obtained by successive passage through susceptible animals. THE STAPHYLOCOCCI 249 Toxic Substances Produced. — Filtrates of cultures contain toxic sub- stances. Injected into the peritoneal cavity they excite peritonitis. Under the skin they produce infiltration or abscess formation. In the blood they injure both the red and white corpuscles. Cultures of the staphylococcus, when sterilized by boiling and injected subcutaneously, produce marked positive chemotaxis and often local abscesses. Leber found also that sterilized cultures introduced into the anterior chamber of the rabbit's eye would bring about a fibro- purulent inflammation, the cornea becoming infiltrated, and perforation alongside of the sclerotic ring finally taking place. This was followed by the formation of pus in the anterior chamber and recovery. These local changes follow the inoculation of small quantities only of the dead cultures; but when large amounts are injected into a vein or into the abdominal cavity, toxic effects are produced. The hemolytic effects of certain products of virulent staphylococci have recently been studied. In cultures they can be detected about the third or fourth day of incuba- tion and reach their maximum on the ninth to fourteenth day. Virulent staphylococci are more apt to produce this substance than the non- virulent, but there is no definite rule. The specific hemolysin, known as staphylolysin, is destroyed by heat- ing for twenty minutes at 56° C. An antibody for this is formed by inoculating animals with culture filtrates. A substance called leukocidin is produced which injures leukocytes, it also produces an antibody. The gelatin-liquefying enzyme — gelatinase — has already been men- tioned. Occurrence in Man. — ^The staphylococcus {Staphylococaus aureus) has been demonstrated not only in furuncles and carbuncles, but also in various pustular affections of the skin and mucous membranes — impetigo, sycosis, purulent conjunctivitis and inflammation of the lacrimal sac; in acute abscesses formed in the lymphatic glands, the parotid gland, the tonsils, the mammse, etc.; in metastatic abscesses and purulent collections in the joints; in empyema, infectious osteo- myelitis, ulceraitive endocarditis, pyelonephritis, abscess of the liver, phlebitis, meningitis, etc. It is one of the chief etiological factors in the production of pyemia in the various pathological forms of that condition of disease. It is remarkable how many staphylococci may be present in the blood without a fatal result, if the source of infection is removed. We met with one case in which over 800 staphylococci were present in 1 c.c. of blood. A week later only 5 were found. The patient finally died from pneumonia. Not all persons are equally susceptible to infection by the staphy- lococcus; those who are in a cachectic condition or suffering from con- stitutional diseases, hke diabetes, are especially predisposed to infection. In healthy individuals certain parts of the body, as the back of the neck and the buttocks, are more liable to be attacked than others, with the production of furuncles, carbuncles, etc. In persons in whom sores are readily caused, in consequence of disturbances of nutrition, as in exhausting diseases, the micrococci settle at the points of least resistance. 250 THE PYOGENIC COCCI Such conditions are present in the bones of debihtated young children, in fractures, and in injuries in general. Immunity. — ^Rabbits have been rendered immune by means of inoculations with either dead or living cultures. Unless the inoculations are carefully made the animals frequently succumb. The staphylococci injected into an immunized animal are more rapidly taken up by the leukocytes than when injected into an untreated animal. (See Opson- ins, p. 218.) A serum having some protective power has also been elaborated. Hiss claims good results from the use of leukocyte extracts in animals infected with Staphylococcus aureus. (See p. 225.) Therapeutic Use of Vaccine. — ^The treatment of abscesses, boils, and other localized staphylococcus infections as well as general infections by injections of repeated doses of vaccine is considered in Part III. Staphylococcus (Pyogenes) Albus. — ^It is morphologically identical with the Staphylococcus {pyogenes) aureus, and is probably the same organism which has lost the property of producing pigment. On the average it is somewhat less pathogenic and seldom produces pyemia or grave infections. The surface cultures upon nutrient agar and potato have a milk-white color. Its biological characters are not to be dis- tinguished from the Staphylococcus aureus. The majority of bacteriologists agree with Rosenbach, that the aureus is found at least twice as frequently in human pathological processes as the albus. Staphylococcus Epidermidis (Albus) (Welch).— Probably identical with the Staphylococcus {pyogenes) alhus. With reference to this micro- coccus, Welch says: "So far as our observations extend — and already they amount to a large number — this coccus may be regarded as nearly, if not quite, a constant inhabitant of the epidermis. It is now clear why I have proposed to call it the Staphylococcus epidermidis alhus. It possesses such feeble pyogenic capacity, as is shown by its behavior in wounds as well as by experiments on rabbits,- that the designation Staphylococcus pyogenes alhus does not seem appropriate. Still I am not inclined to insist too much upon this point, as very probably this coccus — ^which has hitherto been unquestionably identified by others with ordinary Staphylococcus pyogenes albus of Rosenbach — ^is an attenuated or modified form of the latter organism, although, as already mentioned, it presents some points of difference from the classical description of the white pyogenic coccus." According to Welch, this coccus differs from the Staphylococcus albus in the fact that it liquefies gelatin more slowly, does not so quickly cause coagulation in milk, and is far less virulent when injected into the circulation of rabbits. It has been shown by the experiments of Bossow- ski and of Welch that this microorganism is very frequently present in clean wounds, and that usually it does not materially interfere with the healing process, although sometimes it appears to cause suppuration along the drainage tube, and is the common cause of "stitch abscesses," THE MICROCOCCUS TEfRAGENUS . 261 Staphylococcus (Pyogenes) Citreus.— Isolated by Passet (1885) from the pus of acute abscesses, in which it is occasionally found in asso- ciation with other pyogenic cocci. It is distinguished from the other species only by the formation of a lemon-yellow pigment. Other StaphylococcL^Other varieties have been occasionally met with which differ in some respects from the typical varieties. This difference may be in the fact that they liquefy gelatin more slowly or not at all, or in pigment formation, or in agglutination, or in still other respects. None of these varieties seem to be of importance. THE MICROCOCCUS TETRAGENUS. This organism was discovered by Gaffky (1881). It is not infrequently present in the saliva of healthy individuals and in the sputum of con- sumptive patients. In sputum it is sometimes an evidence of mouth contamination rather than lung infection. It has been observed repeatedly in the walls of cavities in pulmonary tuberculosis associated with other pathogenic bacteria, which, though playing no part in the etiology of the original dis- ease, contribute, doubtless, to the progressive destruc- tion of the lung. Its pyo- /ti genie character is shown by Fig. 100. — Micrococcus tetra- genus. Stained with methylene blue. X 1000 diameters. Fig. 101. — Micrococcus tetragenus from peri- toneal fluid. Stained with fuchsin. (Frankel.) X 1000 diameters. its occasional occurrence in the pus of acute abscesses. Its presence has also been noted in the pus of empyema following pneumonia. Morphology. — Micrococci having a diameter of about \ix, which divide in two planes, forming tetrads, and bound together by a transparent, gelatinous substance, enclosing the cell like a capsule. In cultures the cocci are seen in various stages of division as large, round cells in pairs of oval elements, and in groups of three and four (Figs. 100 and 101). When the division is complete they remind one of sarcinse in appearance, except that they do not divide in three directions and are not built up like diminutive cotton bales (see also Plate II, Fig. 4, and Plate III, Fig. 11). 2S2 THE PYOGENIC COCCt Staining.— This micrococcus stains readily with the ordinary aniline dyes; the transparent gelatinous envelope is only feebly stained. It is not decolorized by Gram's method. Biology. — ^The growth of this micrococcus is slow under all conditions. It grows both in the presence and absence of oxygen; it grows best from 35° to 38° C, but may be cultivated also at the ordinary room temperature — about 20° C. Growth on Gelatin. — On gelatin plates small, white colonies are developed in from twenty-four to forty-eight hours, which, when ex- amined under a low-power lens, are seen to be spherical or lemon- shaped, grayish-yellow disks, with a finely granular or mulberry-like surface, and a uniform but somewhat roughly dentated border. When the deep colonies push forward to the surface of the gelatin they form white, elevated, drop-like masses, having a diameter of 1 to 2 mm. In gelatin stick cultures the gelatin is not liquefied. Growth on Agar and Blood Serum. — ^The colonies appear as small trans- parent, round points, which have a grayish-yellow color and are slightly elevated above the surface of the medium. Pathogenesis. — Subcutaneous injections of a culture of this micro- coccus in minute quantity is usually fatal to white mice. The micro- cocci are found in comparatively small numbers in the blood of the vessels and heart, but are more numerous in the spleen, lungs, liver, and kidneys. Intraperitoneal injections given to guinea-pigs and mice are followed by purulent peritonitis, beautifully formed cocci in groups of four being obtained in immense numbers from the exudate. Rabbits and dogs are not affected by large doses of a culture subcutaneously or intravenously administered. In man it is generally non-pathogenic, except in the conditions already cited. It has been obtained by us as the only organism in a case of chronic conjunctivitis. The serum from immunized cases has not been used therapeutically in human infection. Vaccines may be employed as with staphylococci. THE STREPTOCOCCI. The streptococci in their relation to human infection outweigh in importance most other disease-producing organisms. Under this name must be included not only the streptococci which excite inflam- mation in man, but all non-motile, non-spore-bearing spherical bacteria which divide, as a rule, in one plane only and remain attached in longer or shorter chains. Owing to the variations in morphology, cultural characteristics and virulence of this group of bacteria it has so far been impossible to give absolutely satisfactory differentiation of varieties in the group. The relation between pathogenicity and other char- acteristics is not clear-cut. The relation of the pnemnococcus to the group has not been decided. The recent work of Rosenow on the transmutation of streptococci and pneumococci calls for confirmation. Such a transformation has not been proved to occur in any other types THE STREPTOCOCCI 253 as distinct as these have appeared to be. The first classification was based upon pathogenicity and morphology. The first pathogenic strepto- cocci were discovered by Koch in stained sections of tissue attacked by septic processes, and by Ogston in the pus of acute abscesses (1881). Pure cultures were obtained by Fehleisen (1883) from a case of erysipelas. The cultural and pathological characters were studied by him and it was shown to be capable of producing erysipelas in man. Rosenbach (1884) and Krause and Passet (1885) isolated the streptococcus from the pus of acute abscesses and gave it the name of Streptococcus pyogenes. It was first thought that the streptococci of erysipelas, of acute abscesses, of septicemia, of puerperal fever, etc., belonged each to a different species, because they seemed to possess differences in their biological and patho- logical characteristics, according to the source from which they were obtained. But it is now thought that the slight differences among the majority of streptococci from these diseases are but acquired variations of organisms derived from the same species. The first grouping of the pathogenic forms on biochemical char- acteristics was made by Schottmuller in 1903. His three types were: (1) Streptococcus pyogenes or erysipelatus, which shows hemolysis on blood agar plates; (2) Streptococcus mitior or viridans, which produces a green halo about the colonies on blood agar plates, and (3) Strepto- coccus mucosus which produces a mucoid growth and a dark green zone. Park and Williams (1905) and others since then, showed that Strepto- coccus mucosus should be placed with the pneumococci under the common name pneumococcus mucosus. Gordon (1903) and Andrews and Border (1906), using nine test substances on over a thousand strains, divided streptococci from various sources into nine groups. This work has been for the most part corroborated, but there has been much unfavorable criticism of such a classification based only on biochemical reactions. Serological reactions while encouraging, have not yet given clear-cut results. They have special importance in helping to indicate the proper serum to use in infections. Lyall recently has repeated the work with the carbohydrates and has tried to show the relationship between the characteristic carbohydrate reaction and the reactions on erythrocytes, but evidence being gathered by Krumwiede and others tends to show that the production of methemoglobin is not clear cut enough to be used as a primary basis for classification. Holman's classification (page 254) may be giyen as a later type of classification based on erythrocyte and carbohydrate re- actions. Dextrose and saccharose are not used in differential tests, since prac- tically all streptococci ferment these sugars. Streptococcus Pyogenes. — Under this heading are included all those streptococci that produce hemolysis and ferment dextrose, lactose, saccharose, and salicin but that do not ferment mannit. Morphology. — The cocci, when fully developed are spherical or oval. They have no flagella or spores. They vary from 0.4/i to 1/t in diameter. They vary in dimensions in different cultures and even in different parts 254 THE PYOGENIC COCCI o o o o o H « U a J3 o a- o o 1 a w zo -J^ + a 1 G ea ;-. iXSHi. (•;■,. s 1 ■4 9idJ e l),r."- . + -1 ^ 'i + ^ +_ ^ Hi +_ +_ -Str. ignavus — Str. equinus -Str. non-hemolytious iii "r-Str. nou-hemolyticus ii -Str. salivarius 03 Str. mitis — Str. non-hemolytious i -Str. fecalis I - _a "3 +- —Str. subacidus -Str. equi I Str. hemolytifcus iii a "3 03 + Str. hemolyticus ii I Str. anginosus a +_ -|-_ +_ Str. pyogenes -Str. hemolyticus i + Str. infrequens THE STREPTOCOCCI 255 of a single colony. They multiply by binary division in one direction only, forming chains of eight, ten, twenty, and more elements, being, however, often associated distinctly in pairs. On solid media the cocci occur frequently as diplococci, but usually they grow in longer or shorter chains. Frequently certain cocci exceed their fellows greatly in size. Fig. 102. — Streptococci in peritoneal fluid, partly enclosed in leukocytes. X 1000 diameters. Fig. 103. — Streptococcus growing in long chains in bouillon culture. X 1000 diameters. especially in old cultures, when they may be considered the result of involution processes. Hueppe formerly called these arthrospores. Some varieties have distinct capsules when growing in the blood and in blood-serum media. 4. Fig. 104. — Streptococci from solidified serum culture appearing mostly in diplo- cocci. X 1000 diameters. Fig. 105. — Streptococci in throat exu- date smeared on cover-glass. X lOOU diameters. Staining. — ^They stain readily by aniUne colors and the pyogenic varieties give a positive reaction by Gram's method. Some species, mostly saprophytic, growing in short chains are negative to Gram's stain. Biology. — Streptococci of this type grow readily in various liquid and solid culture media. The most favorable temperature for their develop- ment is from 30° to 37° C., but they multiply rather freely at ordinary 256 THE PYOGENIC COCCI room temperature — 18° to 20° C. They are facultative anaerobes, growing both in the presence and absence of oxygen. Cultivation. — Growth on Gelatin.^ — ^Tubes of gelatin which have been inoculated with these streptococci by puncture with a platinum needle show on the surface no growth beyond the point of entrance. In the depth of the gelatin on the second or third day a distinct, tiny band appears with granular edges or made up of granules. These granules may be very fine or fairly coarse. They are nearly translucent, with a whitish, yellowish, or brownish tinge. With characteristic cultures the gelatin is not liquefied. Growth on Agar. — On agar plates the colonies are visible after twelve to thirty hours' growth at 37° C, and present a beautiful appearance when magnified sufficiently to see the individual cocci in the chain. The colonies are small, not averaging over 0.5 mm. in diameter (pin- head). From different sources they vary in size, thickness, mottling, color, and in the appearance of their borders. The streptococcus growing in short chains in bouillon shows but little tendency to form true loops, but rather projecting rows at the edges of the colonies, while those growing in long chains show beautiful loops, -which are characteristic of this organism. Growth in Bouillon. — Most streptococci of this type grow well in slightly, alkaline bouillon at 37° C, reaching their full development within thirty- six to forty-eight hours. Those which grow in long chains usually give an abundant flocculent deposit and leave the liquid clear. The deposit may be in grains, in tiny flocculi, in larger flakes, or in tough, almost membranous masses, the differences depending on the strength of union between the pairs of cocci in the chains. Some of the strepto- cocci growing in long chains, however, cause the broth to become cloudy. This cloudiness may be only temporary or it may be lasting. Those growing in short chains, as a rule, cloud the broth, this cloudiness remaining for days or weeks. A granular deposit appears at the bottom of the tube. An addition of 0.5 to 1 per cent, glucose aids the develop- ment of streptococci, but the acid produced tends later to hasten their death and make them lose virulence. A trace of calcium aids the growth. This is best added as a piece of marble, which has the additional advantage of neutralizing some of the acids produced. Growth in Ascitic or Serum Bouillon.— The development in this, which is the best medium for the growth of all streptococci, is more abundant than in plain bouillon. The liquid is usually clouded, and a precipitate occurs after some days, the fluid gradually clearing. The addition of blood serum frequently causes streptococci, growing in short chains in nutrient bouillon, to produce long chains. The reverse is also true, and in the blood all forms are usually found, some, at least, being diplococci or in short chains. Effect on Inulin. — This is not fermented by varieties of the pyogenes type. Growth on Solidified Blood Serum. — ^This is also an excellent medium for the streptococcus. Tiny, grayish colonies appear twelve to eighteen hours after inoculation. THE STREPTOCOCCI < 257 Growth in Milk. — All streptococci grow well in milk. As a rule when growth is luxuriant a marked production of lactic acid with coagulation of the casein occurs. Development of Hemolytic Substances. — Hemolysis is demonstrated within blood agar plates or in fluid containing blood in test-tubes. The plate method is as follows: If 1 c.c. of fresh or defibrinated blood is added to 6 c.c. of melted agar at 40° to 45° C, well shaken, inoculated with characteristic streptococci and poured into a Petri dish there will appear in twelve to twenty-four hours tiny colonies surrounded by clear zones of about i to ^ inch in diameter. The tube method is used for quantitative determination of hemolysis. The titration is made by adding decreasing amounts of a definite culture (eighteen-hour 2 per cent, peptone ascitic broth culture, accord- ing to Lyall), to a constant quantity of washed red blood cells (1 c.c. of a 5 per cent, suspension of sheep's red cells, Lyall). The tubes are incubated in water-bath at 37° C. for one hour and readings are then made. Pneumococci and many streptococci, grouped as Streptococcus viridans, which occxu" together with characteristic forms in the throat, lungs, and elsewhere, on the other hand, produce only narrow zones of a green pigment. Anthony in our laboratory has found that from a streptococcus producing abundant hemolytic substances strains may be obtained, by selecting certain colonies, which fail to make them. She has not been able to obtain from strains producing in first cultures the green pigment only any strains producing hemolytic substances. Duration of Life Outside of the Body. — ^This is not, as a rule, very great. When dried in blood or pus, however, they may live for several months at room temperature, and longer in an ice-chest, and in gelatin and agar cultures they live for from one week to three months. In order to keep streptococci alive and vigorous, it is best to transplant them frequently. They may be kept alive for a long time in semisolid agar stick cultures at room temperature or in serum or ascitic fluid bouillon in small sealed glass tubes in the ice-chest. Resistance to Heat and Chemicals is given under Disinfection, Part III. Non-hemolytic Streptococci. — Non-hemolytic streptococci, most of which produce a green zone (methemoglobin) on blood agar plates, are not of such active virulence as are the hemolytic type. They are slowly invasive and may produce chronic inflammation of a low grade. The methemoglobin produced by many of the strains in variable amounts is probably a combination of reduction and oxidation processes (Heubner, Cole and Blake). It has no relation to virulence and very little to carbohydrate reactions. Based on cultural and agglutinative reactions this group is apparently heterogeneous. Pathogenesis.— The majority of test animals are not very susceptible to infection by streptococci from tumors, and hence it is difiicult to obtain any definite pathological alterations in their tissues through the inocula- tion into them of cultures of this organism by any of the methods ordinarily practised. White mice and rabbits, under similar conditions, are the most susceptible, and these animals are therefore usually X7 258 THE PYOGENIC COCCI employed for experimentation. Streptococci, however, differ greatly in the effects which they produce in inoculated animals, according to their animal virulence, which is very different from hiunan virulence. The most virulent, when injected in the minutest quantity into the cir- culation or into the subcutaneous tissues of a mouse or rabbit, produce death by septicemia. Those of somewhat less virulence produce the same result when injected in considerable quantities. Those still less pathogenic produce septicemia, which is mild or severe, when injected into the circulation; but when injected subcutaneously, they produce abscess or erysipelas. The remaining streptococci, unless introduced in quantities of 20 c.c. or over, produce only a slight redness, or no reaction at all, when injected subcutaneously, and little or no effect when injected directly into the circulation. Many of the streptococci obtained from cases of cellulitis, abscess, empyema, and septicemia belong to this group. A number of varieties of streptococci have thus been discovered, differing in virulence and in their growth on artificial media; but all attempts to separate them into various classes, even with the use of specific serum, have largely failed, because the differences observed, though often marked, are not constant, many varieties having been found to lose their distinctive characteristics, and even to apparently change from one class to another. A further objection to any of the existing classifications of streptococci, which are based on the manner of growth on artificial culture media, is that it has been impossible to make any which would at the same time give even an approximate idea of their virulence. Experiments have proved that streptococci originally virulent may become non-virulent after long cultivation on artificial media, and, again, that they may return to their original properties after being passed through the bodies of susceptible animals. The peculiar type of virulence which they may acquire tends to perpet- uate itself, at least for a considerable time. One important fact that experience teaches us is that those strepto- cocci which are the most dangerous are those which have come imme- diately from septic conditions, and the more virulent the case the more virulent the streptococci are apt to be for animals of the same species. There seems also to be a strong tendency for a streptococcus to produce the same inflammation, when, inoculated, as the one from which it was obtained; for example, streptococci from erysipelas tend to produce erysipelas, from septicemia to produce septicemia, etc. Streptococci, however, obtained from different sources (abscesses, puerperal fever, erysipelas, etc.) are in many instances capable, under favorable conditions, of producing erysipelas when inoculated into the ear of a rabbit, as has been proved by experiment, provided they possess sufHcient virulence. Occurrence in Man. — Streptococci have been found to be the primary cause of infection in the following diseases: Erysipelas, circumscribed and extensive acute abscesses, impetigo, cellulitis (circumscribed as well as diffused), sepsis, puerperal infection, acute peritonitis, angina, THE STREPTOCOCCI 259 bronchopneumonia, periostitis, osteomyelitis, synovitis, otitis media, mastoiditis, enteritis, irregular cases of rheumatic fever, meningitis, pleurisy, empyema, and endocarditis. Associated with other bacteria in diseases of which they were the specific cause, they have also been found as the secondary infection in many diseases, such as in pulmonary tuberculosis, bronchopneumonia, septic diphtheria, and diphtheritic scarlatina. In cases of septic thrombus of the lateral sinus following mastoiditis there is almost certainly a streptococcus septicemia. Libman has shown that an examination of the blood may be useful in diagnosis. In diphtheritic false membranes this micrococcus is very commonly present, and is frequently the source of deeper infection, such as abscesses and septicemia; and in certain cases accompanied by a diphtheritic exudation, in which the Loffler bacillus has not been found by competent bacteriologists, it seems probable that the Streptococcus pyogenes, alone or with other pyogenic cocci, is responsible for the local inflammation and its results. These forms of so-called diphtheria, as first pointed out by Prudden, are most commonly associated with scarlatina and measles, erysipelas, and phlegmonous inflammation, or occur in indi- viduals exposed to these or other infectious diseases. So uniformly are long-chained streptococci present in the pseudomembranes of patients sick with scarlet fever, that many investigators have suspected a special variety of them to be the cause of this disease. The same is true for smallpox. Many varieties are regularly found, however, in the throat secretion of healthy individuals (in 100 examinations by us we found long-chained streptococci in 83, and probably could have found them in some of the others by longer search). Their abundance in scarlet fever and smallpox is most probably due to their increase in the injured mucous membrane and entrance into the circulation when the protective properties of the blood have been lowered. Recently, Rosenow and his co-workers have claimed that a variety of non-hemolyzing streptococci is the cause of poliomyelitis (see Filtrable Viruses). Septic Sore Throat — Streptococcus infection of the throat appears at times as a severe epidemic. Most of these epidemics have been traced to the milk supply. As Smith, Brown, Krumwiede and Valentine and others have shown, these streptococci probably always come originally from a septic human throat. Streptococci of human origin may invade the milk ducts and multiply in the udder without causing any physical signs of mastitis. The bovine streptococci normally producing mastitis (Mathers) have no relation to septic sore throat. The local infection is usually accompanied by severe general symptoms and by suppurative foci elsewhere. Occurrence in Animals. — ^Besides streptococci similar to those in man, animals are infected by strains that are negative to Gram and fluidify gelatin. Udder infections of the cow and glandular diseases of the horse are frequently due to these. The streptococcic inflammations in animals are almost as frequent and serious as they are in man. 260 THE PYOGENIC COCCI - Eifect on Tumors. — Fehleisen inoculated cultures, obtained in the first instance from the skin of patients with- erysipelas, into patients in the hospital suffering from inoperable malignant growths — lupus, carcinoma, and sarcoma — and he obtained positive results, a typical erysipelatous inflammation having developed around the point of inocu- lation after a period of incubation of from fifteen to sixty hours. This was attended with chilly sensations and an elevation of temperature. Persons who had recently recovered from an attack of erysipelas frequently proved to be immune. These experiments were undertaken on the ground that malignant tumors had previously been found to improve or entirely disappear in persons who had recovered from accidental erysipelas. This fact was therapeutically applied to the treatment of malignant tumors. Then the mixed toxins of the strepto- coccus and B. prodigiosiis were given, and it became apparent that the toxins of the latter organism were much the more important. In some cases of inoperable sarcoma this method met with considerable success (Coley). The injections cause severe reactions. Production of Toxic Substances.- — ^There is no doubt that the strepto- coccus causes fever, general symptoms of intoxication, and death by means of toxic substances which it forms in its growth; but we know very little about these substances or how they are produced. The cell substance of streptococci possesses only slight toxicity. Ruediger has shown that a specific streptolysin is formed which produces a true antibody. The poisons while partly extracellular are mostly contained in the cell substance. Heat destroys a portion of them. They appear to attack especially the red blood cells, and this hemolytic action seems to be to some degree in proportion to the virulence of the organism. Susceptibility to Streptococcus Infection. — ^As with the ever-present staphylococci, whose virulence, as we have seen, is usually slight, the streptococci are more likely to invade the tissues, forming abscesses or erysipelatous and phlegmonous inflammation in man when the standard of health is reduced from any cause, and especially when by absorption or retention various toxic organic products are present in the body in excess. It is thus that the liabihty to these local infec- tions, as complications of operations or sequelae of various specific infectious diseases, in the victims of chronic alcohoUsm, and consti- tutional affections, etc., are to be explained. It seems established that the absorption of toxic products formed in the alimentary canal as a result of the ingestion of improper food, or in consequence of abnormal fermentative changes in the contents of the intestine, or from constipation predispose to infection. Immunity. — In none of the streptococcus inflammations do we notice much apparent tendency to the production of immunizing and curative substances in the blood by a single infection. Several general infections usually progress to a fatal termination after a few days, weeks, or months. It is true, however, that cases of erysipelas, cellulitis, and abscess, after periods varying from a few days to months, tend to recover, and to a certain extent, therefore, we THE STREPTOCOCCI 261 may assume that protective agents have been produced. In these cases, however, we know from experience that faulty treatment, by lessening the local or general resistance, would, as a rule, cause the subsiding infection again to progress perhaps even to a more serious extent that the original attack. Koch and Petruschky tried a most interesting experiment. They inoculated cutaneously a man suffering from a mahgnant tumor with a streptococcus obtained from erysipelas. He developed a moderately severe attack, which lasted about ten days. On its subsidence they reinoculated him; a new attack developed which ran the same course and over the same area. This was repeated ten times with the same results. This experiment proved that in this case, at least, the immunizing substances produced by repeated attacks of erysipelas were insufficient to make the tissues and lymph sufficiently bactericidal to prevent infection. The severe forms of infection, such as septicemia following injuries, operations, and puerperal infections, show little tendency to be arrested after being well established. Having in mind the above facts, let us consider the results aheady obtained in the experimental immunization and treatment of animals and men suffering from or in danger of infec- tion with streptococci. Knorr succeeded in producing a moderate immunity in rabbits against an intensely virulent streptococcus by injections of very slightly virulent cultures. Marmorek was the first to attempt the production of a curative serum on a large scale. Influence of Serum from Immunized Animals upon Streptococcus Infections in Other Animals. — ^In the table are given the results fol- lowing the injection of small amounts of a serum which represents in immunizing value what about one-third of the horses are able to pro- duce when given in gradually increasing doses a living, virulent strepto- coccus. In the following experiments the serum and culture were injected subcutaneously into rabbits, into some inoculated after mixing and into others separately and on opposite sides of the body. Showing Strength or Avbbage Grade or Antistreptococcic Serum Given by Selected Horses after Six Months op Injection of Suitable Amounts op Living Streptococci. Weight of Amounts inoculated. rabbit. Serum culture. Results. Autopsy. 1. Inoculated together . . 1430 * 0.25 c.c. 0.01 c.c. Lived 2. Inoculated together . . 1350 0.125 c.c. 0.01 c.c. Lived 3. On opposite sides . 1770 0.1 c.c. 0.01 c.c. Lived 4. On opposite sides . 1630 0.1 c.c. 0.01 c.c. Lived Controls: 1. Rabbits injected with cul- 1750 0.001 c.c. Died in Streptococcic ture only. 4 days infection. 2. Rabbits injected with cul- 1870 0.001 c.c. Died in Streptococcic ture only. 24 hrs. infection. The above results have been repeatedly obtained, and are absolutely conclusive that the serum of properly selected animals, which have been repeatedly injected with living streptococci in suitable doses. 262 THE PYOGENIC COCCI possesses bactericidal properties upon the same streptococcus when it comes in contact with it within the bodies of animals. Polyvalent Serum. — ^Results of investigators show that the majority of hemolytic streptococci met with in cellulitis, erysipelas, and abscess will be influenced by the same serum. In order that the serum may have specific antibodies for the variety of streptococci causing each separate infection, each horse is now injected with a large number of different varieties of streptococci. This serum will not be a,s good as if made by the streptococcus infecting the treated case, but will be fairly efficient for all cases. The non-hemolyzing streptococci obtained from cases of pneumonia and endocarditis and other exceptional infections are apt to have individual characteristics. Here a polyvalent serum is of little value. Dosage. — ^It is found that the immune bodies to be effective must be in sufficient concentration. The intravenous injections are usually 100 c.c. for an adult. Preparation of the Serum. — The preparation of antiserums is given in Part III under. Applications of Serum Therapy. Stability of the Serum.— It is fairly stable but, after several months, the serimi loses some of its protective value. It should be kept in a cool and dark place. Standardization of the Value of the Serum.- — ^There is at present no satisfactory way. The value of the serum is sometimes measured by the amount required to protect against a multiple of a fatal dose of a very virulent streptococcus of the same type as the one used to inject the horses. The dose is usually a thousand times the average fatal amount of a very virulent streptococcus. Other methods of standardization, such as the estimation of the amount of opsonins or agglutinins present, are also used but are not as conclusive. Therapeutic Results.— To estimate the exact present and future value of antistreptococcus serum is a matter of the utmost difficulty. Many of the cases reported are of little or no help, because, no cultures having been made, we are in doubt as to the nature of the bacterial infection. In the cases of pyerperal fever, erysipelas, and wound infection that we have seen, thie apparent results under the treatment have not been uniform. We have frequently observed favorable results which appeared to be due to the serum when doses of 50 to 100 c.c. were given intravenously. In a number of cases of septicemia in which for days chills had occurred daily they ceased absolutely or lessened under daily doses of 20 to 50 c.c. The temperature, though ceasing to rise to such heights, did not average more than one or two degrees lower than before the injec- tions. In some cases the serum treatment was kept up for four weeks. Some cases convalesced; others after a week or more grew worse and died. In some cases the temperature fell immediately upon giving the first injection of serum, and after subsequent injections remained THE STREPTOCOCCI 263 normal, and the cases seemed greatly benefited. As a rule, in these cases no streptococci or any other organisms were obtained from the blood. In bronchopneumonia due to streptococci and in the mixed infections accompanying laryngeal diphtheria, tonsillitis, smallpox, and phthisis, we have seen little effect. The results obtained here in New York by both physicians and surgeons in streptococcus infections have not, on the whole, been very encouraging. In some of the cases where apparently favorable results were obtained other bacteria than streptococci were found to be the cause of the disease. We believe that the following conclusions will be found fairly' accurate : The serum will in animals limit an infection already started if it has not progressed too far. The apparent therapeutic results in cases of human streptococcus infection are variable. In some cases the disease has undoubtedly advanced in spite of large injections, and here it has not seemed to have had any effect. In other cases good observers rightly or wrongly believe they have noticed gjeat improvement from it. Except rashes, few have noticed deleterious results, although very large doses have been followed in several instances, for a short time, by albuminous lu^ine. In suitable cases we are warranted, we believe, in trying it, but we should not expect very striking results. For our own satisfaction, and to increase our knowledge, we should always have satisfactory cultures made when possible, and the strepto- cocci, if obtained, tested with the serum used in the treatment. In the cases where we want most to use the serum, such as puerperal fever, septicemia, ulcerative endocarditis, etc., we find that it is very difiicult to make a bacteriological diagnosis from the symptoms, and in over one-half of the cases even the bacteriological examination carried out in the most thorough way will fail to detect the special variety of bacterium causing the infection. This is often a great hindrance to the proper use of curative antistreptococcic serum, for it, of course, has no specific effect upon the course of any infection except that due to the strepto- coccus and the full effect only on its own type. Care should be taken to get the most reliable serum; much on the market is worthless, and as it is weak, and the testing for strength is difficult or impossible, full doses (50 to 100 c.c.) of serum should be given if the case is at all serious, for the dose is limited only by the amount of horse serum which we feel it safe to give, not because we have given sufficient protective substance. Intravenous injections give better results than those given subcutaneously. Studdiford has obtained good results by adding to the intravenous injection the packing of the septic uterus with gauze impregnated with the serum. Scarlet Fever. — In Vienna for some years the serum of horses treated at each injection with a number of strains of streptococci derived from scarlet-fever cases has been used in this disease. The serum given in large doses of 100 to 200 c.c. has apparently given good results in about 264 THE PYOGENIC COCCI half of those treated. It is only used in severe cases. Moser has.ehiefly advocated its use. One of us had the opportunity of looking over the histories of his cases. Although left in doubt as to its value, it appears to us as worth a trial. Our own results and NicoU's,. in cases occurring in the Health Department hospitals, have been rather favorable. Streptococcus Vaccine. — ^The preparation and use of streptococcus vaccines is given in Part III, under Practical Applications of Vaccines. Complement-fixation. — ^The method and results of this test have been considered in Part I, p. 199. Bacteriological Diagnosis. — Streptococci, using the name in a broad sense, can often be demonstrated microscopically by simply making a smear preparation of the suspected material and staining with methy- lene-blue solution or diluted Ziehl's stain. In order to demonstrate them microscopically in the tissues the sections are best stained by Kiihne's methylene-blue method. In all cases, even when the micro- scopic examination fails, the cocci may be found by the use of culture media, such as broth or plated agar at 37° C. To obtain them from a case of erysipelas it is best to excise a small piece of skin from the margin of the erysipelatous area in which the cocci are most numerous; this is crushed up and part of it transferred to ascitic or serum bouillon, and part is streaked across freshly solidified agar in a Petri dish on which a drop of sterile rabbit's blood has been placed. Both are kept in the incubator at 37° C. In septicemia the culture method is always required to demonstrate the presence of streptococci, as the microscopic examination of specimens of blood is not sufficient. For this purpose from 10 to 15 c.c. of the blood should be drawn from the vein of the arm aseptically by means of a hypodermic needle, and to each of three tubes containing 10 c.c. of melted nutrient agar kept at about 43° C, 1 c.c. of blood is added. After thorough mixing, the contents are poured into Petri dishes. The remainder is added to several flasks containing 100 c.c. of nutrient broth, in order to produce a development of the cocci, which are found in small nmnbers in the blood. Petruschky is of the opinion that the cocci can be best shown in blood by animal inoculation. Having with- drawn from the patient 10 c.c. of blood by means of a hypodermic , syringe, under aseptic precautions, he injects a portion of this into the abdominal cavity of a mouse, while the other portion is planted in bouillon. Mice thus inoculated die from septicemia when virulent streptococci are present in only very small numbers in the blood. If a successful inoculation takes place we can, through the absence -or presence of the development of capsules, often differentiate between the pneumococcus and the streptococcus. Cultures may fail to do this. The development of a wide, clear zone about the colonies (upon blood agar), without a development of green pigment, indicates that the streptococci belong to the pyogenes type. The absence of a definite zone and the development of a green color indicates that they are pneu- mococci, or streptococci of the viridans type. The growth in the Hiss inulin serum medium will generally differentiate between the two, as THE STREPTOCOCCI 265 the pneumococci usually coagulate the serum, while the great majority of streptococci do not. Agglutination may also be tried but this would only differentiate certain strains. The morphological and cultural characteristics of the streptococcus give us, unfortunately, no absolute knowledge as to the influence which the protecting serum will have. The actual test is here our only method. The detection of the strepto- coccus in the blood is in itself an unfavorable prognostic sign. The blood cultures in many cases of supposed septicemia give no results, for many of these cases develop their symptoms and even die from the absorption of toxins from the local infection, such as an amputation wound or an infected uterus or peritoneum, and the bacteria never invade the blood. When we get negative results we are, as a rule, utterly unable to test the case with curative serums with any accuracy, for the sepsis may be due to either the streptococcus, colon bacillus, staphylococcus, or a number of other pathogenic varieties of bacteria. In mixed infections when the streptococci present may not be isolated from original plates, plates made-from aneighteen-hour culture in suitable broth may contain many isolated colonies. REFERENCES. Blake, F. G.: The Formation of Methemoglobin by Streptococcus Viridans, Jour. Exp. Med., 1916, xxiv, 315. Cole, R.: Jour. Exp. Med., 1914, xx, 363. Garr6: Beit. z. klin. Chir., 1893, xi. Heubnek: Arch. Exp. Path. u. Pharm., 1913, Ixxii, 239. HoLMAN, W. L. : The Classification of Streptococci, Jour. Med. Res., 1916, xxxiv, 377 (with bibliography). Kktjmwiede and Valentine: A Bacteriological Study of an Epidemic of Septic Sore Throat, Jour. Med. Res,, 1915, xxxiii, 231. Krumwiede and Valentine: A Study of the Agglutination and Cultural Relation- ship of Members of the So-called Streptococcus Viridans Group, Jour. Infect. Dis., 1916, xix, 760. Loeb: Centralbl. f. Bakt., 1902, xxxii, 471. Mathers: Different Streptococci and their Relation to Bovine Mastitis, Jour. Infect. Dis., 1916, xix, 222. Ogston: Brit. Med. Jour., 1881, i, 369. Rosenbach: Mikroorganismen bei Wundinfektiou, Wiesbaden, 1884, Ruedigeb: Jour. Amj Med. Assn., 1903, xli, 962; Jour. Inf. Dis., 1906, iii, 663 and 755. Schneider: Arb. a. d. bakt. Inat. Karlsruhe, 1891, i. Smith and Brown: A Study of Streptococci Isolated from Certain Presumably Milk-borne Epidemics of Tonsillitis, Jour. Med. Res., 1915, xxxi, 455. - CHAPTER XVII. THE DIPLOCOCCUS OF PNEUMONIA (PNEUMOCOCCUS, STREPTOCOCCUS PNEUMONIA, MICROCOCCUS LANCEOLATUS). THE DIPLOCOCCUS OF PNEUMONIA. The diplococcus of pneumonia was observed in 1880 almost simul- taneously by Sternberg and Pasteur in the blood of rabbits inoculated with human saliva. In the next few years Talamon, Friedlander, A. Frankel, Weichselbaum, and others subjected this microorganism to an extended series of investigations and proved it to be the chief etiological factor in the production of lobar or croupous pneumonia in man. The relationship of this organism to the, streptococcus group is spoken of in the preceding chapter. The outcome of the various investigations proved that the acute lung inflammations, especially when not of the frank lobar pneumonia type, are not excited by a single variety of microorganism, and that the bacteria involved in the production of pneumonias are also met with in inflammations of other tissues. In any individual pneumonic inflammation it is also found that more than one variety of bacteria may be active, either from the start or as a later addition to the original primary infection. Among all the microorganisms active in exciting pneumonia, the diplococcus of pneumonia is by far the most common, being almost always present in primary lobar pneumonia and as frequently as any other germ in acute bronchopnemnonia and metastatic forms. Besides the different varieties of pnetimococci the following bacteria are capable of exciting pneumonia : Streptococcus pyogenes, Staphylococcus pyogenes, Bacillus pneumonia. Bacillus influenzce, Bacillus pestis, Bacillus diph- thericB, Bacillus typhi, Bacillus coli, and the Bacillus tuberculosis. Since the varieties of bacteria exciting acute pneumonia, with the exception of the pneumococcus, are met with more frequently in other inflamma- tions and have been described elsewhere, they will only be noticed in this chapter so far as their relation to pneumonia demands. Morphology. — ^Typically, the pneumococcus occurs as spherical or oval cocci, usually united in pairs, but sometimes in longer or shorter chains consisting of from three to six or more elements and resembling the streptococcus. The cells, as they commonly occur in pairs, are somewhat oval in shape, being usually pointed at one end — hence the name lanceolatus or lancet-shaped. When thus united the junc- tion, as a rule, is between the broad ends of the oval, with the pointed ends turned outward; but variation in form and arrangement of the THE DIPLOCOCCUS OF PNEUMONIA 267 cells is characteristic of this organism, there being great differences according to the source from which it is obtained. As observed in the sputum and blood it is usually in pairs of lancet-shaped elements, which are surrounded by a capsule. (See Fig. 106.) When grown in fluid culture media longer or shorter chains are frequently formed, which can scarcely be distinguished from chains of certain streptococci, except that, as a rule, the length of the chain is less and the pairs of diplococci are farther apart. In cultures the individual cells are almost spherical in shape, and except in certain varieties are rarely surrounded by a capsule. ■ (See Fig. 107.) The pneumococeus is by some classed as a streptococcus. Rosenow claims that a typical pneumococeus may be easily changed into a typical streptococcus. (See preceding chapter.) *. *- J^ ^'1 "■ - ^^ Fig. 106. — Diplococcus of pneumonia from blood, with surrounding capsule stained by method of Hiss. The capsule is best seen in stained preparations from the blood and exudates of fibrinous pneumonia or from the blood of an inoculated animal, especially the mouse, in which it is commonly, though not always, present. It is seldom seen in preparations from cultures unless special media are employed. Flagella are not present. Staining. — ^It stains readily with ordinary aniline colors; it is not decolorized after staining by Gram's method. The capsule may be demonstrated in blood or sputum by the methods given (p. 78). Biology. — ^It grows equally well with or without oxygen; its parasitic nature is exhibited by the short range of temperature at which it usually grows — ^viz., from 25° to 42° C. — best at 37° C. In the cultivation of this organism neutral or sUghtly alkaline media should be employed. The organism when freshly isolated grows feebly on the serum-free culture media ordinarily employed for the cultivation of bacteria — viz., on nutrient agar and gelatin, in bouillon. The best medium for its 268 DIPLOCOCCUS OF PNEUMONIA growth is a mixture of one-third human or animal blood serum or ascitic or pleuritic fluid and two-thirds bouillon, or nutrient agar streaked with human, horse, or rabbit blood. Growth on Agar. — Cultivated on plain nutrient agar, after twenty- four to forty-eight hours at 37° C, the deep colonies are hardly visible to the eye. Under the microscope they appear light yellow or brown in color and finely granular. The surface colonies are larger, equaling, in size those of streptococci, but are usually more transparent. If blood serum or ascitic fluid is added to the agar the individual colonies are larger and closer together, and the growth is more distinct in con- sequence and of a grayish color. The surface colonies are almost circular in shape under a magnification of 60 diameters, finely granular in structure, and may have a somewhat darker, more compact centre, surrounded by a paler marginal zone. With high magnification cocci in twos and short rows often distinctly separated are seen at the edges. Fig. 107. — Pneumocoocus from bouillon culture, resembling streptococcus. Fig. 108. — Pneumococci stained for cap- sule by Huntoon's method. (Huntoon.) Growth on Blood Agar. — ^The colonies on blood agar are greenish and no hemolysis is present. This is a useful characteristic in isolating pneumococci from sputum, although many streptococci give similar colonies. Methemoglobin is produced in broth cultures as shown on addition of freshly washed red corpuscle suspensions. Growth on Blood Serum. — ^The growth on Lofiler's blood-serum mixture is very similar to that on agar, but somewhat more vigorous and char- acteristic, appearing on the surface as a delicate layer of dew-like drops. Growth in Bouillon. — In bouillon, at the end of twelve to twenty-four hours in the incubator, a slight cloudiness of the liquid will be found to have been produced. On microscopic examination cocci can be seen to be arranged in pairs or longer or shorter chains. After one or two transplantations the pneu- mococci frequently fail to grow. Growth in Milk. — It grows readily in milk causing coagulation with the production of acid, though coagulation is not constant with some forms inter- mediate between the streptococcus and pneumococcus. THE DIPLOCOCCUS OF PNEUMONIA 269 Growth on Gelatin. — ^The growth on gelatin is slow, if there is any develop- ment at all, owing to the low temperature — viz., 24° to 27° C. — above which even the most heat-resistant gelatin will melt. The gelatin is not liquefied. Special Media. — ^When cultures are grown on serum-free media the vitality of some cultures may indeed be indefinitely prolonged; but after transplanta- tion through several generations it is found that the cultures begin to lose in virulence, and that they finally become non-virulent. lii order to restore this virulence, or to keep it from becoming attenuated, it is necessary to interrupt the transplantation and pass the organism through the bodies of susceptible animals. The vitaUty is prolonged and the virulence less rapidly lost if serum ascitic fluid or blood is present in the medium. Serum or ascitic broth, serum semi- solid (stab cultures) or blood-streaked agar are the most satisfactory media for the preservation of cultures. Action of Bile and Bile Salts. — If 0.1 c.c. of rabbits' bile be added to 1 or 2 c.c. of a broth culture of pneumococci the culture becomes clear due to the dissolving of the cocci. A 10 per cent, solution of sodium taurocholate has the same action. Streptococci are not dissolved by bile. Serum and glucose interfere with the reaction. Hiss Serum-water with and without Inulin. — ^These are very useful. The inulin is fermented by typical pneumococci with coagulation of the serum, while most streptococci fail to ferment the inulin. This medium is therefore of considerable diagnostic value. Calcium Broth with or without Dextrose. — ^The addition of a small piece of marble to each tube of broth is the most satisfactory way of preparing it. Marble broth for this purpose was suggested indepen- dently by Bolduan and Hiss as very satisfactory for growth. Resistance to Light and Drying (see Media). — On artificial culture media the pneumococci tend to die rapidly. This is partially due to the acid produced by their growth. In sputum they live much longer. Pneumonic sputum attached in masses to clothes, when dried in the air and exposed to diffuse daylight, retains its virulence, as shown by injection in rabbits, for a period of nineteen to fifty-five days. Exposed to direct sunlight the same material retains its virulence after but a few hours' exposure. This retention of virulence for so long a time under these circumstances is accounted for by the protective influence afl^orded by the dried mucoid material in which the micrococci were embedded. Guarnieri observed that the blood of inoculated animals, when rapidly dried in a desiccator, retained its virulence for months; and Foa found that fresh rabbit blood, after inoculation and cultivation in the incubator for twenty-four hoiu-s, when removed at once to a cool, dark place, retained its virulence for sixty days. There are many conditions, therefore, in which the virulence of the micrococcus is retained for a considerable length of time; the fine spray expelled in coughing and loud speaking that remains suspended in the air soon dries so completely that probably no pneumococci survive after two hours. The action of chemical disinfectants is given in Part III under Dis- infection. Attenuation of Virulence. — The loss of virulence which occurs when the micrococcus is transplanted through several generations in culture fluid containing no blood has already been referred to. An attenuation 270 DIPLOCOCCVS OF PNEUMONIA of virulence, it has been claimed, takes place also spontaneously in the course of pneumonia. This attenuation is probably only apparent. If a little sputum is taken at different periods in the disease and planted in ascitic bouillon the resultant cultures do not vary greatly in virulence. Restoration and Increase of Virulence. — ^The simplest and perhaps the most reliable method of restoring lost virulence for any susceptible animal is by passage through the bodies of highly susceptible animals of the same species. Growth in fresh blood also increases it for the homologous animal. Maintenance of Virulence. — ^This is best done by drying the spleen of an infected animal in a dessicator. The spleen should be removed just before or immediately after death. The virulence is preserved in this way for a month or longer. Toxin Froduction. — We have little exact knowledge upon the nature of the substances produced by or through the growth of the pneumo- cocci in animal tissues or artificial media. Rosenow showed that the autolysis of virulent pneiunococci in NaCl solution brings into the solution a group of substances which inhibits the action of the pneumo- cocco-opsonin. Such extracts are also toxic for animals. Pneumococci dissolved in bile uniformly yield a toxic product. A hemotoxin is present in the bile extracts and can also be extracted by other methods. Occurrence in Man during Health.— It is probable that in crowded communities the pneumococcus is present on the mucous membranes of most persons. We have found it generally present not only in the throats of persons living in New York City, but also in those of persons living on farms and in the Adirondack Mountains. It is commonly present only on the mucous membranes of the bronchi, trachea, pharynx, and nostrils. The healthy lung seems to be generally free from it. The type found is usually one of the less virulent members of Group IV (see below). Carriers of the more virulent types also occur, usually due to contact with a case of pneumonia. Pathogenicity in Man. — Pneumococci, characteristic or atypical, are present in fully 95 per cent, of characteristic cases of lobar pneumonia. Usually no other bacteria are obtained from the lungs. Atypical cases usually show the same conditions, but they may be due to strepto- cocci, influenza bacilli, etc. The more recent the infection the greater is the number of bacteria found in the diseased lung area. As the dis- ease progresses these decrease in number until finally at the crisis they disappear from the tissues, though at this time and long after convales- cence they may be present in the sputum. In atypical forms of pneu- monia they may remain longer in the tissues, and in walking pneumonia they may be absent in the original centres of infection or present only as attenuated varieties, while the surrounding, newly formed foci may contain fully virulent cocci. It has been shown by Netter that more than one-half of the cases of bronchopneumonia, whether primary or secondary to some other disease, as measles and diphtheria, both in children and adults, are due to the diplococcus of pneumonia. Others, such as Pearce, have found other microorganisms, especially the strep- THE DIPLOCOCCUS OF PNEUMONIA 271 tococci, in the majority of cases. These findings will be considered at the end of the chapter. The pneumococci are found partly in the alveoli and bronchioles of the inflamed lung and partly in the lymph channels and blood capil- laries. Most of the organisms are found free, but a few are found in the leukocytes. Through the lymph channels they find their way to the pleura and to adjacent lymph glands. From the capillaries they find their way to the general blood current, and thus to distant parts of the body. In about 20 per cent, of cases the pneumococci are so abundant that they can be found in cultures made from 5 to 10 c.e. of blood. In a number of instances the fetus has been found infected. The pneumococci are also responsible for: Inflammations Complicatin|r Pneumonia. — In every case of lobar pneu- monia and in most cases of bronchopneumonia, pleurisy is developed, which is excited by the same microorganism that was predominant in the pneumonia. With pneumococci the exudate is usually moderate and of a fibrinous character, but may be more abundant and of a sero- fibrinous or purulent character. When the plemsy is marked it is more apt to continue after the cessation of the pneumonia. Pleurisy due to pneumococci is more apt to go on to spontaneous recovery than that due to streptococci or staphylococci. The most frequent pneumococcic infections next to pleurisy, follow- ing a pneumonia, are those of the middle ear, pericardium, endocardium, and meninges, and these not infrequently arise together. Pneumo- coccic inflammations of the heart valves are apt to be followed by extensive necrosis and growth of vegetations. In these cases pneimio- cocci can sometimes be found in the blood for many weeks. Pericarditis due to pnemnococci is a frequent complication, but is usually very slightly developed. Meningitis due to pneumococci may be either fibrinous or purulent or both and is apt to be secondary to otitis, mas- toiditis, or pneumonia. Arthritis, periarthritis, and osteomyelitis are rarer complications of a pneimiococcic pneumonia. Besides moderate parenchymatous inflammation of the kidney, which occurs in most cases of pneumonia, well-marked inflammation may occur in which pneumo- cocci exist in the kidney tissues in large numbers. The presence of pneumococci in the blood after death has been amply proved by mmierous investigations. In many instances, they have been recovered from the blood during life. Lambert, as a rule, found them in all fatal cases twenty-four to forty-eight hoiu-s before death. The conveyance of the infective agent by means of the blood and the lymph to all parts of the body explains the multiplicity of the affections complicating a pneumonia, which are caused by this micrococcus; and not only the secondary, but also the primary diseases, as of the brain and meninges, may be explained in the same way. Presence in Inflammatory Process Not Secondary to Pneumonia.— It is now known that the pneumococcus may infect and excite diseases in many tissues of the body independent of any preliminary localiza- tion in the lung. As a rule these processes are acute and usually run a 272 ' DIPLOCOCCUS OF PNEUMONIA shorter and more favorable course than similar inflammations due to the streptococci. The most frequent primary lesions excited by the pneumococcus after lobar pneumonia, bronchopneumonia, and bronchitis are probably meningitis, otitis media with its compHcating mastoiditis, endocarditis, pericarditis, rhinitis, tonsillitis, conjunctivitis, and keratitis; septicemia, arthritis, and osteomyelitis; inflammations of the epididymis, testicles, and Fallopian tubes; peritonitis, etc. Pneumococcic peritonitis and appendicitis are not so very frequent. The exudate is usually seropurulent. Conjunctivitis due to pneumococci frequently occurs in epidemic form and is frequently associated with rhinitis. From statistics collected by Netter the following percentages of diseases were caused by the pneumococcus: Pneumonia 65 . 9 per cent, in adults. Bronchopneumonia 15.8 Meningitis 13.0 Empyema 8-5 Otitis media 2.4 Endocarditis 1-2 In 46 consecutive pneumococcus infections in children there were: otitis media 29 cases. Bronchopneumonia 12 " Meningitis 2 " Pneumonia 1 case. Pleurisy 1 " Pericarditis 1 " The pneumococcus and streptococcus are the two most frequent organisms found in otitis media. The cases due to the pneumococcus are apt to run the shorter course, but have a tendency to spread to the meninges and cause a meningitis. The pneumococci may also find their way into the blood current. This usually foUows after sinus thrombosis. In bronchitis the pneumococcus is frequently met alone or in combination with the streptococcus, the influenza bacillus, or other bacteria. In certain epidemics pneumococcic bronchitis and pneumonia simu- late influenza very closely and cannot be differentiated except by bacteriological examinations. Primary pneumococcic pleurisy is frequent in children: it is very often purulent, but may be serous or serofibrinous. Its prognosis is better than that in cases due to other organisms. Frequently we have streptococci and staphylococci associated with the pneumococci. Pathogenesis in Lower Animals. — Most strains of the Micrococcus lanceolatus are moderately pathogenic for numerous animals; mice and rabbits are the most susceptible, indeed some strains are intensely virulent for these animals, while guinea-pigs and rats are much less susceptible. Pigeons and chickens are refractory. In mice and rabbits the subcutaneous injection of small or moderate quantities of pneu- THE DIPLOCOCCUS OF PNEUMONIA 273 monic sputum in the early stages of the disease, or of a twenty-four- hour ascitic broth culture from such sputum, or of a pure, virulent ascitic broth culture of the micrococcus, usually results in the death of these animals in from twenty-four to forty-eight hours. The course of the disease produced and the postmortem appearances indicate that it is a form of septicemia — ^what is known as sputum septicemia. After injection there is loss of appetite and great debility, and the animal usually dies some time during the second day after inoculation. The postmortem examination shows a local reaction, which may be of a serous, fibrinous, hemorrhagic, necrotic, or purulent character; or there may be combinations of all of these conditions. The blood of inoculated animals immediately after death often contains the micrococci in very large nunibers. For microscopic examination they may be obtained from the blood, and usually from pleural and peritoneal exudates when these are present. True localized pneumonia does not usually result from subcutaneous injections into susceptible animals, but injections made through the thoracic walls into the substance of the lung may induce a typical fibrinous pneumonia. This was first demonstrated by Talamon, who injected the fibrinous exudate of croupous pneumonia, obtained after death or drawn during life from the hepatized portions of the lung, into the lungs of rabbits. Wadsworth showed that by injecting virulent pnemnococci into the lungs of rabbits which had been immunized, a typical lobar pneumonia was excited, the bactericidal property of the blood being suflicient to prevent the general invasion of the bacteria. Pneumonia may be produced in dogs, and in rabbits less easily, by intratracheal injections. Varieties of the Pneumococcus. — ^As among all other microorgan- isms minutely studied, different strains of pneumococci show quite a wide range of variation in morphology and virulence. Some of the variations are so marked and so constant that they make it necessary to recognize several distinct varieties of the pneumococcus, and to class as pneumococci certain varieties which have before this been classed as streptococci — e. g., the so-called Streptococcus mucosus ca'psulatns (Streptococcus mucosus Schottmiiller) — ^when first isolated from pneumonic exudate or elsewhere, and planted on artificial culture media containing seriun, grows as a rounded coccus with a small dense distinct capsule, principally in short or medimn chains; it produces a large amount of mucus-like zooglea, forming very large spreading colonies; it promptly coagulates fluid-serum media containing inulin. It is also very virulent for mice, but only moderately virulent for rabbits. After a number of culture generations on ordinary nutrient agar it apparently loses most of these characteristics. It then grows in small colonies principally as naked diplococci which may be elongated and pointed, produces no zooglea, and loses most of its virulence for mice and rabbits. It still coagulates inulin-serum media, and when transferred to serum media regains its former morphological characteristics. For these reasons we consider this organism a distinct variety of the pneumo- 18 274 DIPLOCOCCUSOF PNEUMONIA coccus. This variety of pneumococcus has been isolated by us from the lungs after death following lobar pneumonia, out of twenty con- secutive autopsies, as the only organism present twice, and with another variety of pneumococcus once. Together with other varieties it was isolated from four out of twenty specimens of pneumonic sputum, and from sixty specimens of normal throat secretion five times. In 1905 Park and WiUiams showed that this variety should be placed with the pneumococci under the common name pneumococcus mucosus. Agglutination Reactions. — That agglutinins are produced in animals by the injection of pneumococci was shown by Neufeld, Clairmont, and others. It has since been shown that this test may be used as a means of diagnosis. Neufeld, Collins, Cole and Dochez and others have shown that certain pneumococci may be grouped according to serum reactions. According to Cole and Dochez, the groups based upon theu- agglutina- tion reactions are as follows: Group I and II, typical pneumococci; Group III, pneumococcus mucosus; Group IV, heterogeneous strains. Each strain in the last group seems independent as far as serum reactions are concerned; it is really not a group but rather an assembling together of isolated strains. More recently Avery has shown that occasionally pneumococci occur which are related to Group II in that the serum of a type organism protects animals against infection by these subtypes, at least to a very considerable degree, but the reverse, that a serum produced by one of them protects against the others or the standard Type II, has not been demonstrated. More than one subgroup evidently exists. Variations in the agglutinative characteristics of pneumococcci can be induced. Passage through a susceptible animal causes a return of its type reaction. The following table by Cole gives the types he found in pneumonia in a series of cases occurring in New York City, and the relative mor- tality of the cases according to the infecting type: Infecting Strain, Mortality, type. Cases. per cent. Deaths. . per cent. I 34 47 8 24 II 13 18 8 61 III ...... 10 14 6 60 IV 15 21 1 7 Total ... 72 100 23 32 In different localities, and at different times in the same locality, the type percentage will probably vary greatly and the mortality also varies somewhat. WoUstein and Benson found that Group IV occurred more frequently in the pneumonias of children and that the mortality due to this group was high (40 per cent.). Only a limited number of observations are available on the groups occurring in meningitis. Valentine gives the following series: I, 2 cases; II, 7 cases; III, 2 cases; IV, 2 cases. Total, 13. Pneumococci taken from normal throats or from slight inflammation fall usually into Group IV among the heterogeneous strains. , The strains THE DIPLOCOCCUS OF PNEUMONIA 275 belonging to Group I are widely scattered, being found not only through- out the United States but also in Europe. Rapid Method for Determination of Infecting Type in Sputum.— The following method is employed at the Rockefeller Institute Hospital: The sputum coughed up from the lung is injected intraperitoneally into a mouse. After four or five hours the mouse is killed and the peri- toneal cavity washed out with salt solution. The washings are then centrifuged slowly to throw down the fibrin and cells and the super- natant fluid is then drawn off and centrifuged to throw down the cocci. The sedimented cocci are then suspended in saline and mixed in equal parts with known Types I and II serums respectively, and the presence or absence of agglutination determined after incubation for one or two hours. Type III can usually be diagnosed by the mucoid character of the peritoneal exudate. With cultures obtained from other sources, broth cultures free of serum are employed for agglutination. Immunity. — Following an attack of pneumonia some immunity is established, but this lasts only a short time. After successive injec- tions of gradually increasing doses of virulent pneumococci into certain animals (horse, sheep, goat, rabbit), a serum of protective and some curative power in experimental animals is obtained. The mode of action of this serum is still the subject of study. According to Wright, Neufeld, and others, its activity is due primarily to the opsonins. The mechanism of the crises in pneumonia is hot imderstood, but the presence of increased opsonins and other protective substances at this time suggests that it is a phenomenon dependent upon the increase of these substances and associated cellular activity. Serums and Vacines are given in Part III under Applied Therapy. Chemotherapy. — Morgenroth and Levy showed that ethylhydro- cuprein (optbchin), a derivative of hydroquinone, was of value in the treatment of experimental pneumococcus infections. This substance has an almost specific action on pneumococci and in vitro its greater action on the pneumococcus group serves to separate this group from the related cocci. It acts even in dilutions as great as 1 in 1,000,000. It has a protective and curative action in guinea-pigs and mice infected with pneumococci. Small doses of optochin increase considerably the protective power of the type homologous antipneumococcus serum in experimental animals (Moore). The pneumococcicidal action of the serum after administration of the drug was studied in rabbits by Moore and in patients suffering from acute lobar pneumonia by Moore and Chesney. These observers state that when patients receive by mouth 0.024 to 0.026 gram of optochin hydrochloride per twenty-four hours per kilogram of body weight, the serum acquires a pneumococcicidal action and that the drug is apparently helpful in the treatment of lobar pneimionia due to pneumococci. The use of the drug in humans may give rise to amblyopia or amaurosis, which is generally transitory if the drug be discontinued 276 DIPLOCOCCUS OF PNEUMONIA when these symptoms appear. Moore and Ghesney have collected from the literature 786 cases of lobar pneumonia treated with the drug, among which the mortality was 12.84 per cent.; the eye symptoms referred to above occurred in 4.4 per cent, of the cases and in one of these this disturbance was more or less permanent. Tagendreich and Russo have shown that pneumococci subjected to sublethal concentrations of optochin in the test-tube can be rendered "fast" or resistant to the drug within a few days, and Morgenroth has shown that the same phenomenon can be observed to take place in vivo (mice) inefficiently treated. Moore and Chesney recovered a "fast" strain of pneimiococcus from one of their fatal cases treated with optochin. REFERENCES. Aveby: Biologic Classification, Jour. Exp. Med., 1915, xxii, 804. Immunity Factors in Pneumococcus Infection in the Dog, Jour. Exp. Med., 1916, xxiv, 7 (see also ibid., xxiv, 25). Cole: Pneumococcus Infection and Lobar Pneumonia, Arch. Int. Med., 1914, xiv, 56 (general discussion with bibliography). DocHEz: Protective Substances in Human Serum during Lobar Pneumonia, Jour. Exp. Med., 1912, xvi, 663. DocHEZ and Aveby: Antiblastic Immunity, Jour. Exp. Med., 1916, xxiii, 61. Moobb: Jour. Exper. Med., 1915, xxii, 389. Moore: Ibid., 551. MooBE and Chesney: Archives Int. Med., 1917, xix, 611. MoBGENEOTH and Levy: Berl. klin. Wohnschr., 1911, xlviii, 1560, 1979. Neufeld and Handel : Pneumococcus, Handbuch der pathogenen Microorganismen, KoUe and Wasserman, 2 Aufl., Bd. iv. Report of Respiratory Commission, Department of Health, New York City, Studies on the Pneumococcus, Part I, Reprinted from Jour. Exp. Med., 1905, vii, 401. Studies on the Pneumococcus, Part II, Reprinted from Jour. Infect. Dis., 1906, iii, 774. Stillman: Contribution to Epidemiology, Jour. Exp. Med., 1916, xxiv, 649. Steykee: Variations in Pneumococci Induced by Growth in Immune Sera, Jour. Exp. Med., 1916, xxiv, 49. WoLLSTEiN and Benson: Types of Pneumococci, Infants and Children, Am. Jour. Dis. Child., 1916, xii, 254. CHAPTER XVIII. MENINGOCOCCUS OR MICROCOCCUS (INTRACELLULARIS) MENINGITIDIS, AND THE RELATION OF IT AND OF OTHER BACTERIA TO MENINGITIS. While, as has already been stated, the pneumoeoccus is a cause of isolated cases of meningitis the meningococcus is the most frequent cause of purulent meningitis either sporadic or endemic. In 1887 Weichselbaum discovered a micrococcus in the exudate of cerebro- spinal meningitis in 6 cases, 2 of which were not complicated by pneumonia. He obtained it in pure cultures, studied its characteristics, and showed that this organism was clearly distinguishable from the pneumoeoccus and especially by its usual presence in the interior of pus cells, on which account he called it Diplococcus intracellularis meningitidis. In 1895 Jaeger and Scheurer drew especial attention to the etiological relationship of the organism to the epidemic form of cerebrospinal meningitis. They also believed it to be very probable that in most cases of primary meningitis it is from the mucous mem- brane of the nasal cavities and the sinuses opening out from them that both the diplococcus of pneumonia and the micrococcus intracellularis find their way through the blood or perhaps directly through the lymph channels to the meninges. The former we know to be almost constantly present in the nasal cavities, and the latter we have reason to believe is not infrequently there. The nasal secretion of carriers is more dan- gerous to the community than that of infected persons because carriers are not recognized as such usually and mingle freely with the general public. The prevalence of epidemics in winter and spring, a time favor- able to influenza and pneumonia, also suggests the respiratory tract as the place of the infection and where an increase in virulence takes place. We do not as yet know why meningitis follows in some persons and not in others after infection of the mucous membranes. Morphology. — ^This organism occurs as biscuit-shaped micrococci, usually united in pairs, but also in groups of four and in small masses, sometimes solitary, and small degenerated forms are found. It has no well-defined capsule. Cultures resemble strongly those of gonococci. In the sediment from spinal fluids (see Fig. 109) meningococci are both intra- and extracellular. When they are intracellular the prognosis is better, and frequently in cases where they are extracellular at first they later become intracellular. Staining. — ^They stain with all the ordinary aniline dyes and are readily decolorized by Gram's solution. A smear from a culture shows charac- teristic irregularity in staining, some of the cocci taking the counter- 278 MENINGOCOCCUS stain poorly and some staining deeply. The positive cocci described by Jaeger and others were probably contaminating organisms. The cells show no definite capsule. Cultivation.— They grow between 25° and 38° C, best at about 35° C. They are most easily isolated on 2 per cent, glucose ascitic agar neutral to phenolphthalein. They are kept well on semisolid media. The liver medium recommended by Dopter (p. 105) is an excellent one for stock transplants. The reaction of the media is very important. A cultiu-e can rarely be isolated on plain nutrient agar. Different strains of meningo- cocci vary somewhat in the ease with which they may be culti- vated, their ability to ferment carbohydrates, virulence for ani- mals, agglutinability, degree of digestibility in leukocytes, and power of resistance to immune serum. Fig. 109. — Diplococous intracellularis men^ ingitidis in pus cells. X 1100 diameters; After having been isolated for some time, a tolerably good growth develops at the end of forty-eight hours in the incubator. On semitransparent media (glucose-ascitic agar) the colonies may be seen as a flat layer, each about i inch in diameter, grayish white in color, finely granular, rather viscid, and non-con- fluent unless very close together. On Loffler's blood serum the growth forms round, whitish, shining, viscid-looking colonies, with smooth and sharply defined outlines; these may attain diameters of ^V to | inch in twenty-four hours. The colonies tend to become confluent and do not liquefy the serum. From the spinal fluid in acute cases, where the organisms are apt to be more abundant, a great many minute colonies may develop instead of a few larger ones. On agar plates the deep-lying colonies are almost invisible to the naked eye; somewhat magnified they appear finely granular, with a dentated border. On the surface they are larger, appearing as pale disks, almost transparent at the edges, but more compact toward the centres, which are yellowish gray in color. On blood agar or serum agar the growth is much more luxuriant than on plain agar and larger than the gonococcus. Cultivated in artificial media, while it often lives for weeks, it may die within four days, and requires, therefore, to be transplanted to fresh material at short intervals. Resistance. — It is readily killed by heat, disinfectants, sunlight, and drying. A few cocci may remain alive for one to three days in the dried state. To maintain cultures it is necessary to make transfers frequently. Agglutination Characteristics. — ^A considerable percentage of cultures of meningococci are relatively inagglutinable. Strains that are agglu- tinable respond to the agglutinins developed in an animal immunized with a true strain. Careful absorption tests are capable in many instances of separating true meningococci from other Gram-negative organisms. This serum reaction is practically never used for diagnosis PATHOGENESIS 279 because it. is so variable and unreliable. It may, however, be used in conjunction with other tests in standardizing protective serum. Further work is necessary before we can know the limits of specific agglutinin-producing powers of the aberrant strains classed by Dopter (1909) under the term parameningococcus. WoUstein, testing many strains from different parts of the world, for agglutination, opsonization, complement -fixation and serum protection, comes to the conclusion that while certain strains exhibit differences enough to- warrant the employment of them in the production of a polyvalent serum, they show too many variations in serum reactions to be separated into a distinct class. They should rather be considered special strains among menin- gococci. This conclusion is in general accord with those obtained through the extensive researches of Elser and Huntoon (1909). The method used by us for making the agglutination test is as follows : A light emulsion is made by rubbing up the growth of an eighteen-hour neutral veal agar culture in salt solution. Equal amounts (0.5 c.c.) of an emulsion and the various serum dilutions are made, the tubes in- cubated at 55° C. overnight and the readings made in the morning. Pathogenesis. — This organism has a low grade and very variable pathogenicity for laboratory animals. Following a suitable intraperi- toneal injection the temperature of the guinea-pig falls, the hair stands out, the abdomen becomes distended and the muscles rigid. The animal hunches up in a corner and seems very sick. Not infrequently " there is prolapse of the rectum. If the dose is fata.1 death usually occurs in from ten to forty-eight hours. At autopsy there is fluid exudate into the abdomen and sometimes into the pleural cavity, congestion or hemorrhage of the adrenals, hemorrhages into the mesentery, central tendon of the diaphragm and into the whole peritoneum. Frequently it is possible to recover the meningococcus from the heart's blood when live cultm-e has been used. The organisms do not have to multiply to produce death and the autolysate or killed culture is just as fatal as live culture. Death is probably due to a bacterial poison freed by the disintegration of the meningococci. Very young cultures are apt to produce a septicemia. Rabbits injected either subcutaneously or intravenously lose weight rapidly. ^ Pathogenicity for Man. — ^The most marked lesions occur at the base of the brain. The cord is always affected. This is not true to the same extent in other bacterial infections. In some epidemics the course of the disease is very rapid. The mortality without serum treatment varies between 50 and 80 per cent. Presence of Meningococci in the Nasal Cavity of the Sick and Those in Contact with Them. — ^In 1 of his 6 cases Weichselbaum succeeded in obtaining diplococci from the nasal secretion. In 1901 Albrecht and Ghon demonstrated them in healthy individuals. Scheurer, in 18 cases, found the diplococci in the nasal secretions during life. In 50 healthy individuals examined they were found in the nasal secretions of only 2, 1 being a man suffering at the time from a severe cold. This man, it is interesting to note, had been employed in a room which 280 MENINGOCOCCUS had just previously been occupied by a patient with cerebrospinal meningitis. Lately, there has been a tendency to throw doubt on these findings, but from our experience in the 1906 epidemic in New York one can state that the meningococci are usually present in great numbers in the nose g,nd nasopharynx in most cases of meningitis during the first twelve days of illness. After the fourteenth day they cannot usually be found. In 1 case Goodwin, of our laboratory, obtained them on the sixty-seventh day. She also found them in five persons out of sixty tested who had been in close contact with the sick, and in two of fifty medical students. We have found them since then in an interne con- valescing from whooping-cough, who to his knowledge had not been in recent contact with any meningitis cases. Presence in Other Localities. — In addition to the situation already noted the meningococcus is frequently found in the blood in the early days of the disease. Elser found it in blood cultures in 10 out of 40 cases. The meningococci have also been found in the herpes and in the urine, a fact to be considered in ordering the hygiene of the sick room. Complicating Infections. — Occasionally we find secondary to the cere- brospinal meningitis, and due to the Micrococcus pneumoniae, cystitis, conjunctivitis, inflammation of the middle ear, arthritis, endocarditis, etc. The most frequent and serious complication of meningococcic meningitis is pneumonia probably due in many cases to the pneumo- coccus. Serum Treatment. — ^It is difiicult to apportion the credit for the pro- duction of the first protective serum. BonhofF and Lepriere produced in animals a serum which showed definite protection. The world-wide epidemic beginning in 1904 stimulated a number of laboratories to pro- duce sera in horses with the idea of treating human cases. Thus KoUe and Wassermann, Jochmann, Flexner, and ourselves immunized horses. The usual method was to begin with cultures recently obtained from human cases and grow them on ascitic agar or plain nutrient agar in tubes. The growth was scraped off, added to physiological salt solution, and heated to 55° to 60° for one hour. Living cultures were often substituted later. The original injections were quite small, being only one or two mod- erate-sized platinum loopfuls. Each succeeding injection was doubled in sizie each time, until the maximum dose of the growth on two Petri dishes was given, which size dose was continued to the end. The injections were given about every eight days. Horses give the best serum after eight months to one year. KoUe and Wassermann injected one horse with the watery extract of recent cultures. They used both the intravenous and subcutaneous methods. The Therapeutic Use of Serum.— In 1905 there was inaugurated in Hartford the use of subcutaneous injections of diphtheria anti- toxic serum in meningitis. This influenced us to prepare and try the subcutaneous injection of an antimeningococcus serum. The results reported by the physicians in some 20 cases did not seem to establish that any beneficial effects were obtained, so no further serum was issued. Later Kolle and Wassermann reported somewhat favorable results BACTERIOLOGICAL DIAGNOSIS 281 m a number of cases from the subcutaneous injection of a serum pre- pared by them. Meanwhile a serum prepared by Jochmann was employed by the intraspinal method in a series of cases. This method soon supplanted the subcutaneous injections. The first successful use of an immune serum in cases of human cerebro- spinal meningitis by the intraspinal method should therefore, so far as we know, be credited to Jochmann and the physicians who used his serum in the winter of 1905 and 1906. He reported a series of cases treated by the intraspinal method before the Congress for Internal Medicine held in Munich in April, 1906, and pubUshed his paper on May 17, 1906. He reported 40 cases, but gave details concerning only 17 patients, all treated by Kromer. Five of these patients died and twelve recovered, a mortality of 29 per cent. He directed that after lumbar puncture, 20 to 50 c.c. of fluid should be removed and then 20 c.c. of immune serum injected. These injections should be repeated once or twice if the fever did not abate or returned. He noticed in general a bettering of the headache, stiffness of neck, and mental condition. Jochmann showed that in animals colored fluids injected into the spinal canal in the lumbar region passed the full length of the canal. The serum was shown to possess both bacteri- cidal and opsonic power. Although the serum prepared in different laboratories in Europe was regularly used after Jochmann's report, it did not receive much attention in this country until Flexner, through his important experi- ments on infected monkeys, which demonstrated the value of the intra- spinal injections of the sermn, aroused medical interest and paved the way for him to try out the serum on a large scale. All cases treated by him were subjected to most careful bacterial tests and clinical observation. Eighteen months later, Flexner and Jobling published their report which fully corroborated the earlier results of Jochmann. The serum prepared at the Rockefeller Institute for Medical Research has been sent to many places, both in this country and in Europe. The results obtained have been of the utmost value in arriving at the value of the intraspinal treatment. Details in regard to the administration of serum and of vaccine are given in Part III under Applied Therapy. Bacteriological Diagnosis. — ^The fluid should be collected in a sterile container. It may be clear, cloudy, or bloody. If it is clear it may be a normal fluid or a fluid from a case of pohomyelitis or tuberculous meningitis. If it is cloudy it may be due to the ineningococcus, streptococcus pneumococcus, pneumococcus mucosus, influenza bacillus, or other rarer organisms. The blood in a bloody fluid may be due to a previous hemor- rhage or to the accidental piercing of a vein. The two conditions may be differentiated by centrifuging the fluid. If the supernatant fluid is yellow or reddish the hemorrhage is old. Clear fluid indicates a recent hemorrhage. It is unfortunate to have much blood in the fluid for it obscures the microscopic picture and then unless the culture is positive it is difficult to make even a tentative diagnosis. Clear fluids M MENINGOCOCCUS should always be centrifuged preferably for one hour at high speed. Cloudy fluids showing no organisms should also be centrifuged. The sediment should be used to make cultures and smears The smears are examined: (1) for pus cells and (2) for tubercle bacilli if the fluid is clear or for other organisms if it is cloudy. The sediment from clear fluids should always be stained with the tubercle stain and that from cloudy fluids with Gram's stain. Once in a long while a cloudy fluid will be found to be tuberculous or a clear fluid to be due to some pyogenic organism but these occurrences are so rare as to be practically negligible. Gram's stain differentiates influenza bacillus, and meningococcus from the streptococcus, pneumococcus, and pneumococcus mucosus. The finding of Gram-negative cocci either intra- or extracellular is presump- tive evidence of meningogoccic meningitis, but it is always well to follow it up with a culture. In order that cultures may grow, fluids should be examined at the earliest moment possible, for meningococci in fluids over twelve hours old are frequently autoly zed so that they will not grow. The following table gives the main differential points in making diagnosis from spinal fluids: Pressure. Amount, c.c. Appear- ance. Cytology. Baat. Albumin. Globulin. Animal inoc. Normal. Normal. 5-10 Clear. Very few cells. Sterile. =1= _ Negative. Meningismus Increased. 10-100 Clear. Very few cells. Sterile. ± _ Negative. Infantile Increased. 20-100 Clear; Early polynu- Sterile. +-++ +++ Negative paralysis some- times slight fibrin web. cleosis; later lymphocytosis up to 95 per cent; endothe- lial cells. or pneu- monia. Tuberoulous Increased. 30-120 Clear; Lymphocytosis Tubercle ++-+++ ++-+++ Tubercu- meningitis. fibrin web. up to 95 per cent. bacilli. losis in 4 weeks. Epidemic cer- Increased. 5-120 Cloudy Polynucleosi? Meningo- ++-++++ ++-++++ ebrospinal up to 98 per coccus. meningitis. cent. Meningitis Increased. 20-100 Cloudy Polynucleosis Infecting ++-++++ ++-++++ due to other up to 98 per organisms. organisms. cent. Differential Diagnosis Distinguishing Meningococci from Gono- cocci. — As a rule the portion of the body from which the organ- isms are obtained reveals their identity. When this is insufficient careful cultm-al and serological tests are required. McNeil has obtained a specific complement-fixation reaction (see p. 198). Other Gram-negative Cocci Resembling Meningococci. — ^Micrococcus Pharyngis (Siccus) (von Lingelsheim), Diploccccus Mucosus, Chromogenic Gram-negative Cocci, Micrococcus Catarrhalis. — ^These may be differen- tiated by cultiu-al characteristics. OtherOrganismsExcitingMeningitis.— 1. Theiz<6ercZe6aciZ/Ms. This is the most frequent cause of meningitis due to an organism other than the meningococcus. 2. The Pneumococcus. This diplocbccus is one of the most frequent exciters of meningitis, both as a primary and a secondary infection. 3. The Streptococcus pyogenes, Pneumococcus and Staphylococcus, Meiiingitis due to these organisms is almost always secondary to some other infection, such as otitis, tonsillitis, erysipelas, endocarditis, suppurating wound of scalp and skull, etc. MiCROCddCtJS cAtArrhAlis 28S 4. The Bacillus influenzce. Numerous reports have been published of the presence of influenza baciUi in the meningeal exudate. Those that are reliable state in almost every instance that the meningitis is secondary to infection of the lungs, bronchi, and the nasal cavities with their accessory sinuses. 5. The colon bacillus, the typhoid bacillus, that of bubonic plague and of glanders, all may cause a complicating purulent meningitis. 6. In isolated cases of meningitis complicating otitis media and other infections, other bacteria, such as the Micrococcus tetragenus, the Bacillus pyocyaneus, the gonococcus, etc., may be found. Meningitis due to other organisms than the meningococcus is almost invariably fatal. MICROCOCCUS CATARRHAUS (R. PFEIFFER). Micrococci somewhat resembling meningococci are found in the mucous membranes of the respiratory tract. At times they excite catarrhal inflamma- tion of the mucous membranes and pneumonia. These cocci are at present included under the designation of Micrococcus catarrhalis. Microscopic Appearance. — ^They usually occur in pairs, sometimes in fours; never in chains. The cocci are coffee-bean in shape, slightly larger than the gonococcus, and are negative to Gram's stain. The micrococci are not motile and produce no spores Cultivation.— They grow between 20° and 40° C, best at 37° C. and less rapidly at somewhat lower temperatures, developing on ordinary nutrient agar, as grayish white or yellowish white, circular colonies of the size of men- ingococci. The borders of the colonies are irregular and abrupt as though gouged out. They have a mortar-like consistency. On serum-agar media the growth is more luxuriant. Gelatin is not liquefied. BouiUon is clouded, often with the development of a pellicle. Milk is not coagulated, but dextrose serum media may be. Gas is not produced. Location of Organisms. — ^In the secretion of normal mucous membranes they are occasionally present. In certain diseased conditions of the mucous membranes they may be abundant. Pathogenic Effects in Animals. — ^For white mice, guinea-pigs, and rabbits, some cultures are as pathogenic as meningococci, while others are less so. Differential Points Distinguishing them from the Meningococci. — ^These organisms have undoubtedly been at times confused. Some assert that the meningococci grow only above 25° C. Many cord cultures of meningococci grow below this point. Some assert that the menningococci will not grow on 5 per cent, glycerin agar. Many undoubted cultures do. Careful agglutinin- absorption tests are of great differential value, but can only be carried out safely by one accustomed to them. The meningococci tested by us have removed all the agglutinins acting upon meningococci from a specific meningococcus serum while the allied organisms have removed only about 60 per cent, of them. The probability is that the organisms described by different writers as Micro- coccus catarrhalis were not all the same variety, and some of them were meningococci. Vaccine Therapy. — Good reports have been made of the results of injecting the dead organisms in cases of infections due to this micrococcus. REFERENCES. Ei.SEK and Huntoon: Jour. Med. Res., 1909, xx, 377. Flexner: Jour. Exp. Med., May 1, 1913, No. 5, xvii, S53. A. Sophian: Jour. Am. Med. Assn., March 23, 1912, Iviii, 843. Weichsblbaum: Fortschr. d. Med., 1887, p. 573. Wollstein: Jour. Exp. Med., 19l4, xx, 201. CHAPTER XIX. THE GONOCOCCUS OR MICROCOCCUS GONORRHEA MICROCOCCUS MELITENSIS. The period at which gonorrhea began to inflict man is unknown. The earliest records make mention of it. Wherever civilized man has penetrated, gonorrhea is prevalent among the people. Except for a period after the fifteenth century it was generally recognized as a communicable disease and laws were made to control its spread. The differentiation between the lighter forms of gonorrhea and some other inflammations of the mucous membranes was, however, almost impos- sible until the discovery of the specific microorganism by Neisser in 1879. The organism was first observed in gonorrheal discharges and described by him under the name of "gonococcus;" but though several attempted to discover a medium upon which it might be cultivated, it was reserved for Bumm, in 1885, to obtain it in pure culture upon coagulated human blood serum, and then after cultivating it for many generations to prove its infective virulence by inoculation into man. The researches of Neisser and Bumm established beyond doubt that this organism is the specific cause of gonorrhea in man. Gonorrhea is in almost all cases among adults transmitted through sexual inter- course. Gonorrheal ophthalmia is a frequent accidental infection at birth, and vaginitis in the young child is frequently produced by the carelessness of the nurse or mother carrying infection. Microscopic Appearance. — Micrococci, occurring mostly in the form of diplococci. The bodies of the diplococci are broadened and, as shown in stained preparations, have an unstained division or inter- space between two flattened surfaces facing one another, which give them their characteristic "coffee-bean" or "kidney" shape. The older cocci lengthen, then become constricted in their middle portion, and finally divide, making new pairs (Fig. 110). The diameter of an associated pair of cells varies according to their stage of development from 0.8/x to 1.6/* in the long diameter — average about 1.25ju — ^by Q.Qfi to Ifi in the cross diameter. Extracellular and Intracellular Position of Gonococci. — ^In gonorrhea, during the earliest stages before the discharge becomes piurulent, the gonococci are found mostly free in the serum or plastered upon the epithelial cells, but later almost entirely in small, irregular groups in or upon the pus cells and epithehal cells, and always extranuclear. With the disappearance of the pus formation more free gonococci appear. Discharge expressed from the urethra usually contains more STAINING 285 free organisms than the natural flow. Gonococci sometimes appear irregular or granular, the so-called involution forms. These are found particularly in older cultures and in chronic urethritis of long standing.. Single pus cells sometimes contain as many as one hundred gonococci and seem to be almost bursting and yet show but slight signs of injury. These diplococci are also found in or upon desquamated epithelial cells. There is still discussion as to whether the gonococci actively invade the pus cells or only are taken up by them. There is no evidence that the gonococci are destroyed by the pus cells (Fig. 111). In gonorrhea of the conjunctiva they are contained in the epithelial cells, sometimes in large numbers. They form dense groups which contain forms similar to those seen in older cultures, showing metachromatic granules in round, swollen, pale blue bodies. These groups finally present an appear- ance somewhat like the cell inclusions found in "trachoma" (p. 415). \ ''\ Aft-. Fig. 110. — Smear from pure culture of Fig. 111. — Gonococous, fuchsin stain. X gonococcus on agar. X 1100 diameters. 1000 diameters. (Frankel and Pfeiffer.) (Heiman.) Staining. — The gonococcus stains readily with the basic aniUne colors (p. 75). Loffler's solution of methylene blue is one of the best staining agents for demonstrating its presence in pus, for, while staining the gonococci deeply, it leaves the cell protoplasm but faintly stained. Fuchsin is apt to overstain the cell substance. Beautiful double- stained preparations may be made from gonorrheal pus by treating cover-glass smears with methylene blue aiid eosin. Numerous methods for double staining have been employed, with the object of making a few gonococci more conspicuous. None of them has any specific characteristics such as the Gram stain. It is now established that gonococci from fresh cultures and from recent gonorrheal infections are, when properly treated by Gram's method, quickly and surely robbed of their color and take on the contrast stains. The removal of the stain from gonococci in old flakes and threads from chronic cases is not so certain. This difference is mostly due to the fact that 286 GONOCOCCUS OR MICROCOCCUS GONORRHEA equally uniform specimens cannot be prepared. The decolorized gono- cocci are stained by dipping the films for a few seconds into a 1 to 10 dilution of carbol-fuchsin or a solution of Bismarck brown. This stain- ing should be for as short a time as suffices to stain the decolorized organisms. This method of staining cannot be depended upon alone absolutely to distinguish the gonococcus from all other diplococci found in the urethra and vulvovaginal tract, for, especially in the female, other diplococci are occasionally found which are also not stained by Gram's method. It serves, however, to distinguish this micrococcus from the common pyogenic cocci, which retain their color when treated in the same way, and in the male urethra it is practically certain, as few organisms have been found in that location which in morphology and staining are identical with the gonococcus. It is certainly the most distinctive characteristic of the staining properties of the gonococcus, and it is a test that should never be neglected in differentiating this organism from others which are morphologically similar. Biology. — Grows best at blood temperature; the limits being roughly 25° and 40° C. It is a facultative anaerobe. It is not motile and pro- duces no spores. Culture Media. — The gonococcus requires for its best growth the addition to nutrient agar of a small percentage of blood serum or some equivalent. The media which have proven of value may be found in the chapter on Media. After continued cultivation gonococcus cultures frequently grow on media containing no serum. Occasional strains grow on ordinary glycerin or glucose nutrient agar and even on plain nutrient agar from the start. Viability .-^Cultures usually die in forty-eight to seventy-two hours when kept at room temperature. In the ice-box they may live for several weeks. They frequently live for one week in the thermostat at 36° C. on and in semisolid media. Appearance of Colonies. — A delicate growth is characteristic. At the end of twenty-four hours there will have developed translucent, very finely granular colonies, with scalloped margin. The margin is sometimes scarcely to be differentiated from the culture medium. In color they are grayish white, with a tinge of yellow. The texture is finely granular at the periphery, presenting yellowish punctated spots of higher refraction in and around the centre (Fig. 112). Surface Streak Culture. — ^Translucent grayish-white growth, with rather thick edges. Resistance. — ^The gonococcus has but little resistant power toward outside influences. It is killed by weak disinfecting solutions and by desiccation in thin layers. In comparatively thick layers, however, as when gonorrheal pus is smeared on linen, it has lived for forty-nine days, and dried on glass for twenty-nine days (Heiman). It is killed at a temperature of 45° C. in six hours and of 60° in about thirty minutes. Occurrence of Gonococci in Nature.— Qutside of the human body or material carried from it gonococci have not been found. PATHOGENESIS 287 Pathogenesis. — Non-transmissible to all animals. Both the living and dead gonococci contain toxic substances. Injected in considerable amounts into rabbits, they cause infiltration and often necrosis. Applied to the urethral mucous membrane there is produced an inflam- mation of short duration. In gonorrhea the secretion is believed to be due to these intracellular toxins. Repeated injections give only slight immunity. The filtrate of recent gonococcus cultures contains little toxin. The etiological relation of the gonococcus to human gonorrhea has been demonstrated beyond question by the infection of a number of healthy men with the disease by the inoculation of pure cultures of the microorganism. Disease Conditions Excited by Gonococci. — ^Affections due to this organism are usually restricted to the mucous membranes of the urethra, prostate, neck of bladder, cervix uteri, vagina, and conjunctiva. The conjunctival, vaginal, and rectal mucous membranes are much more sensitive in early childhood than in later life. .The usual course of the inflammation is as follows: The gonococci first increase upon the mucous membranes which show congestion, infiltration with serous exudate and accumulation of leuko- cytes. The cocci then penetrate the epithelial layer down to the submucous connective tissue. Re- covery or a prolonged chronic in- flammation may then persist. The original infection of the urethra or vagina and cervix may remain localized or spread to adjacent parts or through blood and lymph be carried to all parts of the body. Gono- cocci thus cause many cases of endometritis, metritis, salpingitis, oopho- ritis, peritonitis, prostitis, cystitis, epididymitis, and arthritis. Abscesses of considerable size, periostitis, and otitis are occasionally due to the gonococcus. Gonorrheal Ophthalmia. — ^We have corroborated the statement of Stephenson and others that the gonococcus, though a frequent cause of ophthalmia neonatorum is .not the only cause, in fact, in only about two-thirds of these cases is the gonococcus the cause of the inflammation. Endocarditis and Septicemia. — Cases of gonococcus endocarditis and septicemia are not infrequent. Gonococcus septicemia may occur in connection with other localizations or alone. Nearly every year one or two of these cases are met with in every general hospital. In a considerable number of cases where gonococci are obtained from the blood the patients recover. The fever is sometimes typhoid-like in character. Fig. 112. — Colonies of gonococci on pleuritic fluid agar. (Heiman.) 288 GONOCOCCUS OB MICROCOCCUS GONORRHEM Complications. — General infections with gonococci are often followed or accompanied by neuralgic affections, muscle atrophies, and neuritis. Urticaria occasionally occurs. Immunity. — ^Immunity in man after recovery from infection seems to be only shght in amount and for a short period if present at all. It is known that the urethra in man or cervix uteri in woman may contain gonococci which lie dormant and may be innocuous in that person for years, but which may at any time excite an acute gonorrhea in another individual or, under stimulating conditions, in the one carrying the infection. Animals may, however, be immunized, and their blood is both bactericidal and slightly antitoxic. Therapeutic Use of Serum and Vaccine. — The use of sera in acute gonorrheal joiiit inflammation has given in a considerable percentage of cases good results and seems to be worth trying. It seems to be useless in acute gonorrhea of the mucous membranes. Vaccines (heated cultiu-es) have also been used with apparently real benefit in joint inflammations and even, in very localized chronic infections of the urethra, bladder, and elsewhere. They have also been used in acute vaginitis in young children. In our cases the symptoms abated sooner than we expected, but the gonococci persisted. The dose is from twenty to a thousand milhons given every three to seven days. The benefit of serum and vaccine in septicemia is doubtful. Sensitized serum seem to do good in some cases. The use of vaccine and other points on the use of serum are given in Part III under Practical Applications of Vaccines and Serums. Complement-fixation.^ — By the method of McNeil, (p. 198) very good results have been obtained in the deep-seated chronic infections, but the reverse is true in the superficial acute cases. The method is used regularly in the New York City Health Department. Agglutination. — ^Torrey has shown that gonococci resemble pneumo- cocci in that there are a number of different strains which have different specific and but few common agglutinins. The agglutination test is of no practical value in diagnosis. Duration of Infections and of Contagious Period. — There is no limit to the time during which a man or woman may remain infected with gonococci and infect others. We have had one man under observation in which twenty years had elapsed since exposme to infection, and yet the gonococci were still abundant. It is now well established that most of the inflammations of the female genital tract are due to gonococci, and that many of such infections are produced in innocent women by their husbands who are suffering from latent gonorrhea. Bacteriological Diagnosis of Gonorrhea.— In view of the fact that occasional non-gonorrheal forms of urethritis exist, and also that micrococci morphologically similar to Neisser's diplococcus are at times found in the normal vulvovaginal tract of adults, it becomes a matter of importance to be able to detect gonococci when present, and to differentiate these from the non-specific organisms. Besides this, the gonococci which occur in old cultures and in chronic urethritis BACTERIOLOGICAL DIAGNOSIS OF GONORRHEA 289 of long standing sometimes take on a very diversified appearance. From a medicolegal and social stand-point, therefore, the differential diagnosis of the gonococcus has in certain cases a very practical signifi- cance. There are three methods of differential diagnosis now available — the microscopic, the cultural, and complement-fixation. The method employed in the last test is given on p. 198. Animal inoculations are of no value, as animals are not susceptible, and, of course, human inoculations are usually impossible. In the microscopic diagnosis it should be borne in mind that after the acute serous stage has passed, the specific gonococci in carefully made preparations are always found largely within the pus cells. Diplococci morphologically similar to gonococci occurring in other portions of the field and outside of the pus cells should not be considered specific by this test only. It should also , be remembered that the gonococci are decolorized by Gram's method, while other similar micrococci which occiu- in the lu^ethra are, as a rule at least, not so decolorized. Organisms having these characteristics can for all practical purposes be considered as certainly gonococci if obtained from the urethra. From the vulvovaginal tract the certainty is not so great, since other diplococci are found in pus from this area more frequently than from the urethra which stain as gonococci; here cultures should also be made. In chronic urethritis Heiman allows the patient to void his urine either immediately into two sterilized centrifuge tubes or first into two sterile bottles. The first tube will contain threads of the anterior urethra; the second tube will be likely to contain secretion from the posterior urethra and from the prostate gland if, while urinating, the patient's prostate be pressed upon with the finger. Tubes containing such urine are placed in the centrifuge and whirled for three minutes at twelve hundred or more revolutions per minute, at which speed the threads are thrown down. The centrifuged sediment will be found to contain most of the bacteria present, epithelial cells, and, at times, spermatozoa. When the examinations are negative and it is important to be certain, either massage or injections of a solution of silver nitrate may be employed. The latter by causing a temporary irritation with increase of secretion will almost surely cause a discharge of gonococci if any infection is present. In acute cases where the pus is abundant the specimen for examina- tion may be collected, by passing a sterihzed platinum-wire loop as far up into the urethra as possible and withdrawing some of the secretion. In vulvovaginitis the procedure should be as follows: For obtaining the vaginal material the labia are held well apart by an attend- ant wearing sterile rubber gloves. A sterile slender cotton swab is used which passes easily into the vagina without touching any external part but entrance. (If pus from the cervix is desired a speculum should be used.) The swab is rubbed gently about vaginal wall, then withdrawn and rolled (not rubbed) quickly 19 290 GONOCOCCUS OR MICROCOCCUS GONORRHEA over a slide (slide sterilized and held face down if culture is to be made). In mak- ing this smear care is used not to pass the swab over the same surface twice. In this way a beautifully spread film is made. The swab is then returned to its holder (and covered and numbered if culture is to be made). The air-dried slide is rewrapped in its filter paper, numbered and sent to laboratory where it is stained by Gram's method (p. 78). The technic for making culture is as follows: A small amount of rich sterile ascitic fluid is added to the tubes containing the swab. (If patient is at a distance from laboratory the ascitic fluid is sent in a separate tube, and is inoculated from swab just after smear is made; then swab is withdrawn from ascitic fluid tube, placed in its own tube, and both tubes are sent immediately to the laboratory.) After being stirred up in this fluid the swab is withdrawn and smeared in strokes radiating from the centre over ascitic agar (1-4) plates containing 2 per cent, glucose. Other plates of the same medium are stroked with platinum loopfuls of the ascitic fluid emulsion. From two to four plates are made and placed in thermostat at 36° C. After twenty-four hours the plates are examined and if gonococcus-like colonies are seen they are fished. Then a smear is made from the whole of one of the most characteristic streaks, stained by Gram's method and examined for gonococcus-like organisms. Grouping Cases. — From the microscopic examination of wellrmade and well- stained smears (stained by Gram's method), and, when necessary, from cultures and from clinical appearance as well, the cases are divided into four groups, as follows: 1. Positive cases, i. e., those showing leukocytes fiUed with morphologically typical gonococci in smear or showing typical cultures, or showing both. 2. Suspicious cases, i. e., those showing in smears any suspicious intracellular diplococci and 50 per cent, or more of polymorphonuclear leukocytes. 3. Observation cases, i. e., those showing in smears 50 per cent, or over of polymorphonuclear leukocytes, but no suspicious intracellular diplococci; or those having the clinical symptoms of discharge and inflammation and showing less than 60 per cent, of polymorphonuclear leukocytes. 4. Negative cases, i. e., those showing in smears less than 50 per cent, poly- morphonuclears, and no suspicious intracellular diplococci, and no chnical evidence of the disease. Isolation of Groups. — Each group is kept isolated. Later Smears. — From the first three groups smears are made once a week until a negative smear is obtained. From negative groups smears are made if any suspicious sjrmptoms appear. Negative Diagnosis from Later Smears. — Three successive well-made and well- stained negative smears from first three groups, at intervals of not longer than three days, are considered a negative diagnosis. _ Later Cultures. — From cases where morphologically typical gonococci per- sist in smear, cultures may be made and gonococcus-like organisms isolated and ■ studied to find out if they are true gonococci. Bacteria Resembling Gonococci. — ^A few micrococci which resemble gonococci in form and staining have been described. These assume importance largely because they may be confused with the gonococcus. They occur occasionally on the conjunctival and vaginal mucous mem- branes. One of these microorganisms, the Micrococcus catarrhalw (see p. 283), has an importance of its own. Others are probably unimpor- tant. When absolute certainty is demanded cultural and serological tests must be applied. Differential diagnosis from meningococci is given on p. 282. MALTA FEVER. The Micrococcus Melitensis. — ^This microorganism was first dis- covered in the spleen in a case of Malta fever by Bruce in Malta in MICROCOCCUS ZYMOGENS 291 1887. The disease is chiefly confined to the shores of the Mediter- ranean, but cases o? it have been observed elsewhere. Infected goat- herds have been found in Texas. Clinical Symptoms. — Prodromal symptoms follow an incubation period of five to fourteen days. Headache, sleeplessness, loss of appetite, or vomiting accompany a high fever. The fever lasts for weeks, with intermissions and remissions. A fever period of one to three weeks may occur from time to time during a period of many months. The spleen and liver are enlarged. The mortality is slight. Autopsy. — ^The spleen is large and very soft. The liver is also large and congested. Both organs show parenchymatous degeneration. Distribution of Micrococci. — These are most abundant in the blood and at the height of the fever and are present in organs and in the urine from the second day to the end of the disease. Morphology and Biology. — ^Very small rounded or slightly oval organ- isms, about 0.3m in their greatest diameter. They are usually single or in pairs. In old cultures involution, almost bacillary, forms occur. They are not motile and are Gram-negative. Cultivation. — At 37° C. they grow rather feebly on nutrient agar and in broth. The colonies are not usually visible until the third day. They appear as small round disks, slightly raised, with a yellowish tint in the centre. The broth is slightly clouded after four to six days. The culture remains alive for several weeks or months. In gelatin the growth is very slow. Gelatin is not liquefied. Pathogenesis in Animals. — ^Monkeys are susceptible. They pass through the disease much like man. They can be infected by sub- cutaneous or oral inoculation. Guinea-pigs and rabbits are less easily infected. Infected goats pass the organisms in feces, urine, and milk. The milk is believed to be the chief source of infection. By safeguarding the milk the disease has been largely eliminated. Contact infection cannot, however, be completely excluded. Horses and cows are also susceptible. Methods of Diagnosis. — ^The diagnosis of Malta fever can frequently only be made with the help of bacteriological examination. Blood cultures during the febrile period or cultures of the urine are usually employed. An agglutination reaction with the patient's serum, in dilu- tions of 1 to 1000 or higher is diagnostic. Animals injected with the coccus produce a serum agglutinating in high dilutions. This can be used to identify suspected cultures. MICROCOCCUS ZYMOGENS. MacCallum and Hastings observed this micrococcus in a case of acute endocarditis. It has since been found in a few other pathological processes. It occurs in pairs and short chains. It grows well on agar, ferments lactose and glucose, and slowly liquefies gelatin. REFERENCES. MacCalujm and Hastings: Jour. Exp. Med., 1899, iv, 521. Williams and Rosenberg: Arch, of Ophthal,, 1916, xlv, 109, CHAPTER XX. THE BACILLUS AND THE BACTERIOLOGY OF DIPHTHERIA. The lesions of diphtheria are caused by toxemia. The concentrated poison at the seat of the exudate causes intense local inflammation, while in the more severe cases the absorbed poison diffused throughout the body causes widespread cellular injury, giving rise to definite lesions of the cells of muscle, nerve, and other tissues. Historical Notes. — This specific contagious disease can be traced back under various names to almost the Homeric period of Grecian history. From time to time during the following centuries we hear of epidemics both in Italy and in other portions of the civilized world which indicate that the disease never absolutely ceased. In 1771 Bard, an American, advanced strong reasons for believing that membranous croup and pharyngeal diphtheria were different manifestations of the same disease process. In 1821 Bretonneau published his first essay on diphtheria in Paris and gave to the disease its present name. His observations were so extensive and so correct that little advance in knowledge took place until the causal relations of the diphtheria bacilli and their associated microorganisms to the disease began to be recognized. Evidence of Causal Relationship. — As early as 1840 observers began to notice microorganisms in the pseudomembranes. Gradually the obser- vations became more exact. The most importance was attributed to micrococci. In the year 1883, however, bacilli which were very peculiar and striking in appearance were shown by Klebs to be of constant occurrence in the pseudomembranes from the throats of those dying of true epidemic diphtheria. He described the peculiar staining of the organisms. One year later Loffler separated these bacilli from the other bacteria and grew them in pure culture. When he inoculated the bacilli upon the abraded mucous membrane of susceptible animals more or less characteristic pseudomembranes were produced, and frequently death or paralysis followed with characteristic lesions. These animal experiments have been fortified by a number of accidental human inoculations in laboratories with pure cultures of bacilli with sub- sequent development of diphtheria. The Diphtheria Bacillus. — This bacillus is one of the most inter- esting of bacteria. Grown in the animal body or in suitable culture fluid, it produces a powerful toxin. Its morphology and staining are peculiar. Outside of the body it grows best on serum media. Morphology. — ^When cover-glass preparations made from the exudate or from the cultures grown on blood serum, are examined, the diphtheria STAINING 293 bacilli are found to possess the following morphological characteristics : The diameter of the bacilli varies from 0.3^ to 0.8m and the length from 1/i to 6iU. They occur singly and in pairs (see Figs. 113 to 120) and very infrequently in chains of three or four. The rods are straight or slightly curved, and usually are not uniformly cylindrical throughout their entire length, but are swollen at the end, or pointed at the ends and swollen in the middle portion. The average length of the bacilli in pure cultures from different sources frequently varies greatly, and even from the same culture individual bacilli differ much in their size and shape. This is especially true when the bacilli are grown in asso- ciation with other bacteria. The two bacilli of a pair may lie with their long diameter in the same axis or at an obtuse or an acute angle. The bacilli possess no spores, but have in them highly refractive bodies, some of which are the starting-point for new bacilli. There are no flagella. For mode of division see p. 37. Fig. 113. — One of the very characteristic forms of diphtheria bacilli from blood-serum cultures, showing clubbed ends and irregu- lar stain. X 1100 diameters. Stain, methy- lene blue. Fig. 114. — Extremely long form of diph- theria bacillus. This culture has grown on artificial media for over twenty years and produces great amounts of toxin. X 1100 diameters. Staining. — The Klebs-LofHer bacilli stain readily with ordinary aniline dyes, and retain fairly well their color after staining by Gram's method. With Loffler's alkaline solution of methylene blue, and to a less extent with Roux's and dilute Ziehl's solutions (p. 78), the bacilli from blood-serum cultures especially, and from other media less constantly, stain in an irregular and extremely characteristic way. (See Fig. 113.) The bacilli do not stain uniformly. In many cultures round or oval bodies, situated at the ends or in the central portions, stain much more intensely than the rest of the bacillus, usually showing metachromatism (metachromatic granules. See p. 34 and Plate III.) Sometimes these highly stained bodies are thicker than the rest of the bacillus; again, they are thinner and surrounded by a more slightly stained portion. Other bacilli have barred staining. The bacilli stain in this peculiar manner at a certain period of their growth, so that only a portion of the organisms taken from a culture at any one time will show the characteristic staining. The young cultures have the most 294 BACILLUS AND THE BACTERIOLOGY OF DIPHTHERIA regular forms, an eighteen-hour growth showing more clubbed forms than at twelve hours. After twenty-four hours the bacilli do not stain quite as well. In still older cultures it is often difficult to stain the bacilli, ■ ^ -^ ^ % v^ ^ " -%- A .\ Fig. 115. — Diphtheria bacilli character- istic in shapes, but showing even staining. X 1000 diameters. Stain, methylene blue. Fig. 116. — Non-virulent diphtheria ba- cilli, showing stain with Neisser's solutions. This appearance was formerly supposed to be characteristic of virulent bacilli. Bodies of bacilli in smear, yellowish brown ; granules, dark blue. Fig. 117. — B. diphtheria agar culture. Bacilli small and uniform in shape. X 1000 diameters. >^%^''^ m ^-«( Fig. 119. — B. diphtheriae. Forty-eight hours' agar culture. Many segments; long, Indian-clubbed ends. One year on artifi- cial media. X 1410 diameters. Fig. 118.— B. diphtherise. Forty-eight hours' agar culture. Thick, Indian- clubbed rods and moderate number of segments. One year on artificial culture media. X 1410 diameters. Fig. 120. — B. diphtherise. Twenty-four hours' agar culture. Coccus forms. Seg- mented granular forms on Loffler's serum. Only variety found; in cases of diphtheria at Children's Home. X 1410 diameters. GROWTH OP CVLTVRE MEDIA 295 and the staining, when it does occur, is frequently not at all character- istic. The same round or oval bodies which take the methylene blue more intensely than the remainder of the bacillus are brought out still more distinctly by the Neisser stain (p. 78). The Neisser stain has been advocated in order to separate the virulent from the non-virulent bacilli, without the delay of inoculating animals; but in our hands, with a very large experience, neither the Neisser stain nor other stains, such as the modifications of the Roux stain, have given much more information as to the virulence of the bacilli than the usual methylene-blue solution of Lofiler. A few strains of virulent bacilli fail to show a marked characteristic stain, and quite a few pseudodiphtheria bacilli show the dark bodies. There are also in many throats bacilli which seem to have all the staining and cultural characteristics of the virulent bacilli, and yet have no relation to the disease diphtheria, that is, they produce no diphtheria toxin. As will be stated more fully later, nothing but animal inoculations with the sus- pected bacilli together with control injections of diphtheria antitoxin will separate harmless bacQli from those capable of producing diphtheria. The Morphology of the Diphtheria Bacillus on Serum-free Media. — This varies considerably with different culture media employed. On glycerin agar or simple nutrient agar there are two distinct types. One grows as smaller and, as a rule, more regular forms than when grown an serum culture media (Fig. 117). The other type shows many thick, Indian-club forms with' a moderate number of segments (Figs. 118-120). Short, spindle-, lancet-, or club-shaped forms, staLoing uniformly, are all qbserved. The bacilli which have developed in the pseudomembranes or exudate in cases of diphtheria resemble in shape young baciUi grown on agar. Biology. — ^The Klebs-Lofiler bacillus is non-motile and non-liquefying. It is aerobic and facultative anaerobic. It grows most readily in the presence of oxygen. It does not form spores. It begins to develop but grows slowly at a temperature of 20° C. or even less. It attains its maximum development at 37° C. In old cultures in fluid media Williams has observed fusion of one bacillus with another. The fused forms live the longest. Growth on Culture Media. — ^Blood serum. — Blood serum, especially coagulated in the form of Lcjffler's mixture, is the most favorable medium for the growth of the diphtheria bacillus, and is used particu- larly for diagnostic purposes in examining cultures from the throats of persons suspected of having diphtheria. For its preparation see p. 101. If we examine the growth of diphtheria bacilli in pure culture on blood serum we shall find at the end of from eight to twelve hours small colonies of bacilli, which appear as pearl-gray, whitish-gray or, more rarely, yellowish-gray, shghtly raised points. The colonies when separated from each other may increase in forty-eight hours so that the diameter may be one-eighth of an inch. The colonies lying together become confluent and fuse into one mass when the serum is moist. The diphtheria colonies after a growth of twelve hours become larger than those of the streptococci but remain smaller than those of the staphylococci. 296 BACILLUS AND THE BACTERIOLOGY OF DIPHTHERIA Growth on Agar.— On 1 per cent, slightly alkaline, nutrient or glycerin- agar the growth of the diphtheria bacillus is less certain and luxuriant than upon blood serum; but the appearance of the colonies when examined under a low-power lens, though very variable, is often far more characteristic. (Fig. 121.) For this reason nutrient agar in Petri dishes is used to obtain diphtheria bacilli in pure culture. Certain strains of the diphtheria bacillus after having been transplanted for several generations on serum culture media, grow well, or fairly well, on suitable nutrient agar, but when fresh from pseudomembranes they grow on this medium with great difficulty, and the colonies develop so slowly as to be covered up by the more luxuriant growth of other bacteria when present; or they may fail to develop at all. If the colonies develop deep in the substance of the agar they are usually round or oval, and, as a rule, present no extensions; but if near the surface, commonly from one, but sometimes from both sides, they spread out an apron-like ex- tension, which exceeds in surface area the rest of the colony. When colonies develop entirely on the sur- face they are more or less coarsely granular, and usually have a dark centre and vary markedly in their thickness.' The colonies from some are almost translucent; from others are thick and almost as luxuriant as the staphylococcus. The edges are sometimes jagged, and frequently shade off into a delicate lace-like fringe; at other times the margins are more even and the colonies are nearly circular. Peculiarities in the growth of the diphtheria bacillus upon agar are of prac- tical importance. If a large number of the, bacilli from a recent culture are implanted upon a properly prepared agar plate a certain and fairly vigorous growth will always take place. If, however, the agar is inoculated with an exudate from the throat, which contains but a few bacilli, no growth whatever may occur, while the tubes of coagulated blood serum inoculated with the same exudate contain the bacilli abundantly. Because of the uncertainty, therefore, of obtaining a growth by the inoculation of agar with bacilli unaccustomed to . this medium, agar is not a reliable medium for use in primary cultures for diag- nostic purposes. A mixture composed of 2 parts Of a 1.5 per cent, nutrient agar and 1 part of sterile ascitic fluid makes a medium upon which the bacillus grows much more luxuriantly, but not so characteristically. Isolation of the Diphtheria Bacillus from Plate Cultures. — Nutrient or glycerin agar should be melted and poured in the Petri dish for this purpose. After it has hardened, the medium in a number of plates is streaked across with bacteria from colonies on the serum culture, which appear in size and color like the diphtheria bacilli. Other plates are made from a general mixture of aU bac- teria, selected, as a rule, from the drier portion of the serum. Still others are inoculated from the pellicles of ascitic broth cultures. The plates are left in Fig. 121. — Colonies of diphtheria bacilli. X 200 diameters. PATHOGENESIS 297 the incubator for about sixteen hours at 37° C. In the examination of the plates one should first seek for typical colonies, then, if these are not found, for any that look most nearly like the characteristic picture. Diphtheria colonies are very apt to be found at the edges of the streaks of bacterial growth. The fishings from the colonies are inoculated upon Loffler's blood serum, or into ascitic bouillon. Growth in Bouillon.— -The diphtheria baciUi from about one-half the cultures grow readily in broth slightly alkaline to litmus; the other strains grow feebly. .The characteristic growth in neutral bouillon is one showing fine grains. These deposit along the sides and bottom of the tube, leaving the broth nearly clear. A few cultures in neutral bouillon and many in alkaline bouillon produce for twenty-four or forty-eight hours a more or less diffuse cloudiness, and frequently a fUm forms over the surface of the broth. On shaking the tube this film breaks up and slowly sinks to the bottom. This film is apt to develop during the growth of cultures which have long been cultivated in bouillon, and, indeed, after a time the entire development may appear on the surface in the form of a friable peUicle. The diphtheria bacillus in its growth, causes a fermentation of meat sugars and glucose, and thus, if these are present, changes the reaction of the bouillon, rendering it distinctly less alkaline within forty-eight hours, and then, after a variable time, when all the fermentable sugars have been decomposed, more alkaline again through the progressing fermentation of other substances. Among the products formed by its growth is the diphtheria toxin. Grovrth in Ascitic or Serum Bouillon. — ^All varieties of diphtheria bacilli grow well in this medium, even when first removed from the throat. They almost always form a slight pellicle at the end of twenty- four to forty-eight hours. This culture medium is, as pointed out by Williams, of the greatest value in attempts to get pure cultures of the diphtheria bacillus from solidified serum cultures containing few bacilli among many other bacteria. Plate cultures are made from the pellicle. The fluid is prepared by adding to the nutrient bouillon 25 per cent, ascitic fluid or blood serum. Growth in Gelatin. — The growth on this medium is much slower, more scanty, and less characteristic than that on the other media mentioned. This largely is on account of the lower temperature at which it must be used. Growth in Milk. — The diphtheria bacillus grows readily in milk, beginning to develop at a comparatively low temperature (20° C). The milk remains unchanged in appearance, as lactose is not fermented by the diphtheria bacillus. Pathogenesis.— In Lower Animals. — The diphtheria bacillus through its toxins is, when injected into their bodies, pathogenic for guinea-pigs, rabbits, chickens, pigeons, small birds, and cats; also in a lesser degree for dogs, goats, cattle, and horses, but hardly at all for rats and mice. In spite of its pathogenic qualities for these animals true diphtheria occurs in them with extreme rarity. As a rule supposed diphtheritic inflammations in them are due to other bacteria which cannot produce disease in man. The cat is the only animal that we have known to contract true diphtheria from contact with the disease. Cobbett reports a case in a colt. At the autopsy of animals dying from the poisons produced by the bacilli, the characteristic lesions described by Loffler are found. At the seat of inoculation there is a grayish focus surrounded by an area of congestion; the subcutaneous tissues for 298 BACILLUS AND THE BACTERIOLOGY Of DIPHTHERIA some distance around are edematous; the adjacent lymph nodes are swollen; and the serous cavities, especially the pleura and the peri- cardium, frequently contain an excess of fluid usually clear, but at times tiu-bid; the lungs are usually congested, the suprarenals are markedly congested. In the organs are found numerous smaller and larger masses of necrotic cells, which are permeated by leukocytes. The heart and certain voluntary muscular fibers and nervous tissues usually show degenerative changes. Occasionally, there is fatty degeneration of the" liver and kidneys. The number of leukocytes in the blood is increased. From the area surrounding the point of inoculation virulent bacilli may be obtained, but in the internal organs they are only occasionally found, unless enormous numbers of bacilli have been injected. Par- alysis, commencing usually in the posterior extremities and then grad- ually extending to the whole body and causing death by paralysis of the heart or respiration, is also produced in many cases in which the inoculated animals do not succumb to a too rapid intoxication. In a number of animals we have seen recovery take pla'ce three to six weeks after the onset of the paralysis. Tissue Changes in Natural (Human) Infection. — The characteristic lesions are a pseudomembranous inflammation on some of the mucous membranes or occasionally on the surface of wounds and the general hyperplasias and parenchymatous inflammations produced by the absorbed toxic substances. Pneumonia is apt to occur as a compli- cation of laryngeal diphtheria. The membrane may be simply a thin pellicle, which is easily removed without causing bleeding or it may be thick and firmly attached and leaving when removed a ragged bleeding surface. The tissue beneath the pseudomembrane is always intensely conjested and often hemorrhagic. The cells show marked degenerative changes. Causes of Death. — These are chiefly toxemia, laryngeal obstruction and bronchopneumonia. Septicemia due to other bacteria is frequently an additional factor. Diphtheria Toxin.— This poison was assumed by LofEer (1884) to be produced by the bacilli, but it was first partially isolated by Roux and Yersin, who obtained it from cultures of the living bacilli by filtration through porous porcelain. It has not yet been successfully analyzed, so that its chemical composition is unknown, but it has many of the properties of protein substances, and can well be desig- nated by the term active protein. It resembles in many ways the ferments. After injection into the body there is a latent period before its poisonous action appears. The poison produced is probably com- posed of a mixture of several nearly related toxins. Diphtheria toxin is totally destroyed by boiling for five minutes, and loses some 95 per cent, of its strength when exposed to 75° C. for the same time; 60° C. destroys very little. Lower temperatures only alter it very gradually. Kept cool and from light and air it deteriorates very slowly. Desiccated at low temperatures and kept dry in a vacuum it keeps unaltered for long periods. (See also pp. 151-153.) COMPARATIVE VIRULENCE AS ESTIMATED BY TOXIN 299 The Production of Toxin in Culture Media.— The artificial production of toxin from' cultures of the diphtheria bacillus has been found to depend upon definite conditions, which are of practical importance in obtaining toxin for the inocu- lation of horses, and also of theoretic interest in explaining why cases of -appar- ently equal local severity have such different degrees of toxic absorption. The ' researches of Roux and Yersin laid the foundation of our knowledge. Their investigations have been continued by Theobald Smith, Spronck, ourselves, and others. After an extensive series of investigations we (Park and Williams) came to the following conclusions: Toxin is produced by fully virulent diph- theria bacilli at all times during their life when the conditions are favorable. Under less favorable conditions some baciUi are able to produce toxin while others are not. Diphtheria baciUi may find conditions suitable for luxuriant growth, but unsuitable for the production of toxin. The requisite conditions for good development of toxin, as judged by the behavior of a number of cul- tures, are a temperature from about 32° to 37° C, a suitable cultm-e medium, such as a 2 per cent, peptone nutrient bouillon made from veal, of an alkalinity which should be about 9 c.c. of normal soda solution per liter above the neutral point to litmus, and prepared from a suitable peptone (Witte) and meat. The cultm-e fluid should be in comparatively thin layers and in large-necked Erlen- meyer flasks, so as to allow of a free access of air. The greatest accumulation of toxin in bouillon is after a duration of growth of the culture of from five to nine days, according to the peculiarities of the culture employed. At a too early period toxin has not sufficiently accumulated; at a too late period it has begun to degenerate. In our experience the amount of muscle sugar present in the meat makes no appreciable difference in the toxin produced when a vigor- ously growing bacillus is used, so long as the bouillon has been made sufficiently alkaline to prevent the acid produced by the fermentation of the sugar from producing in the bouillon an acidity sufficient to inhibit the growth of the bacilli. With the meat as we obtain it in New York, we usually get better results with unfermented meat than with fermented. In Boston, with the same bacillus. Smith gets his best results from the bouillon in which the sugar has been fer- mented by the colon bacillus. Instead of colon bacilli, yeast may be added to the soaking meat, which is allowed to stand at about 25° C. The preliminary fermentation of the meat sometimes produces poisonous substances which are deleterious to the horses. We have obtained especially good results with veal broth made from calves two to four weeks old (bob veal). When strong toxin is desirable the muscle is separated from all fat, tendon and fibrous tissue before being chopped. (See p. 109.) Under the best conditions we can devise toxin begins to be produced by bacilli from some cultures when freshly sown in bouiUon some time during the first twenty-four hours; from other cultures, for reasons not well understood, not for from two to four days. In neutral bouillon the culture fluid frequently becomes slightly acid and toxin production may be delayed for from one to three weeks. The greatest accumulation of toxin is on the fourth day, on the average, after the rapid production of toxin has commenced. After that time the number of living baoUli rapidly diminishes in the culture, and the con- ditions for those remaining ahve are not suitable for the rapid production of toxin. As the toxin is not stable at 35° C, the deterioration taking place in the toxin already produced is greater than the amount of new toxin stiU forming. Bacilli, when repeatedly transplanted from bouillon to bouillon, gradually come to grow on the surface only. This characteristic keeps the bacilli in contact with the oxygen and seems to aid in the development of toxin. Comparative Virulence as Estimated by Toxin Production of Dif- ferent Cultures. — The virulence of diphtheria bacilli from different sources, as measured by their toxin production, varies considerably. Thus, as an extreme instance, 0.002 c.c. of a forty-hour bouillon culture of our most virulent strain will kill a guinea-pig, which would require 300 BACILLUS AND THE BACTERIOLOGY OF DIPHTHERIA 0.1 c.c. of culture of our least virulent strain to kill. This difference frequently depends on the unequal growth of the bacilli, one culture having fifty times as many bacilli as the other. When the different strains are grown on ascitic broth, upon which their growth is usually good, the majority of cultures are nearly equal in virulence, but some still show marked differences. Moreover, the diphtheria bacilli differ some- what in the tenacity with which they retain their power to produce toxin when grown outside the body. The bacillus that we have used to produce toxin in the laboratory of the Board of Health has retained its power unaltered for twenty-two years in bouillon cultures. Other bacilli have apparently lessened their capacity for toxin production after being kept six months. Brown and Smith report on a culture in which 5 c.c. was the minimum toxic dose of the filtrate. The passage of diphtheria bacilli through the bodies of susceptible animals does not increase their toxic production to any considerable extent. Comparative Toxicity of Bacilli and Severity of Case. — From the severity of an isolated case the toxicity of the bacilli cannot be deter- mined. The presence of slight amounts of antitoxin in the blood of the person attacked and the association of other bacteria are at least two of the disturbing factors. The most toxic bacillus we have ever found was obtained from a mild case of diphtheria simulating tonsillitis. Another case, however, infected by this bacillus proved to be very severe. In localized epidemics the average severity of the cases prob- ably indicates roughly the toxicity of the bacillus causing the infection, as here the individual susceptibility of the different persons infected would, in all likelihood, when taken together, be similar to that of other groups; but even in this instance special conditions of climate, food, or race may influence certain localities. It must be remembered that bacilli of like toxic power may differ in their liability to infect the mucous membrane; that is, the virulence of the organism may not go hand in hand with the toxicity. Toxic Bacilli in Healthy Throats. — Fully toxic bacilli have fre- quently been found in healthy throats of persons who have been brought in direct contact with diphtheria patients or diphtheria carriers without contracting the disease. It is therefore apparent that infec- tion in diphtheria, as in other infectious diseases, requires not only the presence of toxic bacilli, but also a susceptibility to the disease, which may be local or general. We now know that 70 per cent, of all persons are protected from infection because of having antitoxin present in their blood. Among the predisposing influences which con- tribute to the production of diphtheritic infection may be mentioned the breathing of foul air and living in overcrowded and ill-ventilated rooms, impure food, certain diseases, more particularly catarrhal inflammations of the mucous membranes, and depressing conditions generally. Under these conditions an infected mucous membrane may become susceptible to disease. In connection with Beebe (1894) we made an examination of the throats of 330 healthy persons who had not come in contact, so far as known, with diphtheria, and we found DIPHTHERIA-LIKE BACILLI 301. toxic bacilli in 8, only 2 of whom later developed the disease. In 24 of the 330 healthy throats non-toxic bacilli similar to the toxic diphtheria bacillus were found. Very similar observations have since been made in Boston and by others in many widely separated countries. In 1905 Von ShoUy in our laboratory examined 1000 throats of those who had not knowingly been in contact with diphtheria and found toxic diphtheria bacilli in 0.5 per cent, of the cases. We have found toxic bacilli in about 5 per cent, of cases of scarlet fever. This indicates they are more prevalent in throats than a single culture from normal cases indicate. Persistence of Diphtheria Bacilli in the Throat.— The continued presence of toxic diphtheria bacilli in the throats of patients who have recovered from the disease has been demonstrated by all inves- tigators. In the investigations of 1894 we found that in 304 of 605 consecutive cases the bacilli disappeared within three days after the disappearance of the pseudomembrane; in 176 cases they persisted for seven days, in 64 cases for twelve days, in 36 cases for fifteen days, in 12 cases for three weeks, in 4 cases for four weeks, and in 2 cases for nine weeks. Since then we have met with a case in which they persisted with full toxicity for eight months. It is safe to say that in over 10 per cent, of the cases a few bacilli persist two weeks after the disappear- ance of the exudate and in over 1 per cent, four weeks. It is extremely difficult to prove that a case is absolutely clear, as the bacilli may remain hidden in the epithelial cells of some tonsillar crypt and not be detected by cultures. Diphtheria-Uke Bacilli Not Producing Diphtheria Toxin. — In the tests of the bacilli obtained from hundreds of cases of suspected diph- theria which have been carried out during the past twenty years in the laboratories of the Health Department of New York City, in over 95 per cent, of cases the bacilli derived from exudates or pseudomem- branes and possessing the characteristics of the Loffler bacillus have been found to be toxic, that is, producers of diphtheria toxin. But there are, however, in inflamed throats as well as in healthy throats, either alone or associated with the toxic bacilli, occasionally bacilli which, though morphologically and in their behavior on culture media identical with the Klebs-Loffler bacillus, are yet producers, at least in artificial culture media and the usual test animals, of no diphtheria toxin. Between bacilli which produce a great deal of toxin and those which produce none we find a few minor grades of toxicity. . We believe, therefore, in accordance with Roux and Yersin these non-toxic baciUi should be considered as possible attenuated varieties of the diphtheria bacillus which have lost their power to produce diphtheria toxin. This supposition is, however, not proven and it may be that the ancestors of these bacilli were never toxin producers. These ob- servers, and others following them, have claimed that the toxic bacilli can be artificially attenuated; but the reverse has not been proven that bacilli which produce no specific toxin have later been found to develop it. In our experience some cultures hold their toxicity even 302 BACILLUS AND THE BACTERIOLOGY OF DIPHTHERIA when grown at 41° C. for a number of months, while others become slightly attenuated rather quickly. We have never yet been able to change a toxic culture into an absolutely non-toxic one, and we have never accomplished the reversed process. We believe, therefore, that for practical purposes bacilli which produce no toxin in animals can be disregarded as a possible source of human diphtheria. Among the non-toxic diphtheria-like bacilli which are obtained fre- quently from normal or slightly inflamed throats or from other mucous membranes are some that may be slightly virulent for guinea-pigs, since they may kill, as we found, in doses of 2 to 5 c.c. of broth culture subcutaneously or intraperitoneally injected. Animals are not protected by diphtheria antitoxin from the action of these bacilli, showing that their poisonous action is not due to diphtheria toxin. At autopsy the bacilli are usually found more or less abundantly in the blood and internal organs. The fact that large injections of antitoxin serum hastens the death of guinea-pigs injected with these bacilli has given rise to the notion that injections of antitoxin might be dangerous in persons in whose throats these bacilli were present, either as saprophytes, or possibly, as inciters of slight disease. It is not the antitoxin, but the serum, which in large doses injures the vitality of the guinea-pigs and so slightly hastens death. The amount of serum required to produce this effect is far in excess of that properly given in man. These bacilli were first described by Davis from our laboratory and later by Hamilton in 1904. The possibility of their being present affords no reason to avoid giving antitoxin in suspected cases. When pathogenic in man they are usiially only feebly so. Diphtheroids. — ^There are a large number of small more or less irregular bacilli bearing some morphological resemblance to the diphtheria bacillus which have been loosely termed diphtheroids. They have been found in the nose, throat, eye and other parts of the body, both in health and in disease (Hodgkin's disease, leukemia, etc.). Several of them, e. g.,B.hoffmanni, B. xerosis, are considered distinct species (see below). Since this type of organism is so widely distributed, any specific patho- genic properties attributed to them, unless positive proof is offered, must be received with reserve. Bacillus Hofmaimi (Pseudodiphtheria Ba- cilli). — These bacilli are rather short, plump, and more uniform in size and shape than Fig. i22.-Pseudodiphtheria ba- ^^^ t™e Lofflcr baciUus (Fig. 122). On ciiii. (B. hofmanni.) blood serum their colony growth is very similar to that of the diphtheria bacilli. The great majority of them in any young culture show no polar granules when stained by the Neisser method, and stain evenly throughout with the alkaline methylene blue solution. They do not produce acid by the' fermentation of glucose, as do all known virulent and many non- micro-aMrophilic diphtheroids 303 virulent diphtheria bacilli; therefore there is no increase in acidity in the bouillon in which they are grown during the first twenty-four hours from the fermentation of the meat sugar regularly present. They are found in varying abundance in different localities in New York City, in about 1 per cent, of the normal throat and nasal secretions, and seem to have now at least no connection with diph- theria; whether they were originally derived from diphtheria bacillus is doubtful. They have been called pseudodiphtheria bacilli, and more properly B. hofmanni. In bouillon they grow, as a rule, less luxuri- antly than the diphtheria bacilli. Some of the varieties of the pseudo- diphtheria bacilli are as long as the shorter forms of the virulent bacilli. When these are found in cultures from cases of suspected diphtheria they may lead to an incorrect diagnosis. These bacilli are found occasionally in all countries where search has been made for them. There are also some varieties of diphtheroids which resemble the short B. hofmanni in form and staining, but which produce acid in glucose bouillon. Bacillus Xerosis. — Diphtheria-like bacilli were found by Kutschert and Neisser (1884) in the condition known as xerosis conjunctivae. Since then they have been found frequently in normal conjunctivae. These bacilli went by the name of xerosis bacilli. Under this name, no doubt, different observers have placed several varieties of bacilli morphologically somewhat similar to the diphtheria bacilli. Knapp first limited the . name to bacilli that ferment saccharose and not dextrin, in contradistinction to diphtheria bacilli which ferment dextrin and not saccharose. The fermentation tests for three of the diphtheria group, according to Knapp, may be shown as follows : Species. Dextrose. Levulose. Galactoss. Maltose. Saccharose. Dextrin. B. diphthera . . + + + + — + B. xerosis . . . + .+ + + + — B. hofmanni . . — — — — ■ — — Note. — One per cent, sugar in Hiss serum-water media. These reactions do not separate a large number of the non-virulent bacilli morphologically like the diphtheria bacilli mentioned before. Micro-aerophilic "Diphtheroids." — Certain "diphtheroid" bacilli chose for their most abundant growth that part of the medium which contains only a small amount of oxygen. Thus in a deep shake agar tube culture, a band of growth appears a short distance from the surface of the medium and few colonies appear below this band. Among these microaerophilic bacilli (see p. 120) the acne bacillus is of interest to us because of its probable etiological importance in acne vulgaris. Bacillus Acne. — ^A short, rather plump, irregularly shaped, bacillus was first reported as occurring in acne pustules and comedones by Unna in 1894. A morphologically similar bacillus was grown by Sabouraud, in 1897, in pure cultures on an acid glycerin agar. Whether or not this was the acne bacillus remains doubtful, since pure aerobic cultures are so difficult to obtain. In 1907 Halle and Civatte found an anaerobe in the sebaceous follicles from all faces examined. This threw doubt 304 BACILLUS AND THE BACTERIOLOGY OF DIPHTHERIA on the etiological relationship of this bacillus to acne. But Gilchrist and others found a similar bacillus most abundantly in acne lesions, though it was frequently accompanied by a small coccus. Flemming, in 1909, found a similar anaerobic diphtheroid in large numbers in acne vulgaris and obtained abundant anaerobic growths on oleic acid glycerin agar. The bacillus now accepted as B. acne has the following character- istics: The cultures are made up of short, rather irregularly club- shaped, non-motile rods averaging 2^ by 0.5ju. These are Gram-positive and take ordinary stains well. They grow best with only a small amount of oxygen; that is, in shake tube cultures the most abundant growth appears in the form of a hazy band about half an inch from the surface of the medium and about a fourth of an inch thick. This develops in about three to five days at 36° C. In our laboratory a semisolid medium (p. 99) has been found to be best for the isolation of this bacillus. A shake culture is made in a tube of this medium. If the tube, after sterilization by carbolic acid, is cut just below the zone of growth and new shake culture tubes made from a portion taken from the periphery of this zone, repeating on subcultures if necessary, a pure culture will finally result. The question of the efficacy of vaccine treatment in acne vulgaris with B. acne and with Staphyhcoccus pyogenes is taken up in the chaptei* on the Therapeutic Application of Vaccines in Section III. Persistence of Varieties of the Bacillus Diphtheria and of Diphtheria-like Bacilli. — ^The fact that there are distinct differences between strains of bacilli producing diphtheria toxin which are as great as between these and some strains of bacilli producing no diphtheria toxin has, we think, been fuUy established. But that such varieties are true subspecies with constant characteristics, one variety not changing into another of the estabhshed forms, has not been accepted by aU. On the contrary, the opinion is held by some investigators that all of the various forms of diphtheria-like bacilli are the result of more or less transitory variations of the same species and hence that the toxic forms are the result of a rapid adaptation to environment and consequent pathogenesis of the non-toxic forms, both typical and atypical. This question is of great practical importance in methods of handling the persons harboring these bacilli. Wesbrook, Wilson, and McDaniel, make a provisional classification based upon the morphology of the individual bacilli, into three groups, called granu- lar, barred, and solid, two of the groups into seven types and the other into five, two of the types corresponding with those in the other groups not having been seen. They state that there is generally a sequence of types in the varia- tions which appear throughout the course of the disease, the granular types, as a rule, predominating at the outset of the disease, and these giving place wholly or in part to the barred and solid types shortly before the disappearance of diphtheria-Hke organisms. The inference drawn from this work is that the diphtheria bacillus may be rather easily, especially in the throat, converted into non-granular, solidly staining forms of the "pseudodiphtheria" type, and that the converse may occur, and that therefore all diphtheria-like bacilli must be considered a possible source of danger. In studying the subject WilUams (1902) came to the following conclusions: Though some cultures change on some of the media, each changes in its own way, and each culture still has its distinct individuality. After many culture generations, especially when transplanted at short intervals, the different varie- TRANSMISSION OF DIPHTHERIA 305 ties of toxic diphtheria bacilli tend to run in lines parallel with a common norm, which seems to be a medium-sized, non-segmented bacillus, producing granules in early cultures on serum and growing well on all of the ordinary cultiu"e media. The non-toxic morphologically typical- bacilli must be classed with the toxic varieties as one species, though there is little doubt that more minute study would show that the former constitute a distinct group. The atypical pseudo forms, however, which show no tendency to approach the norm of the typical forms, must be classed as distinct species. Ail of the pseudo and the non-toxic morphologically typical varieties when inoculated into the peritoneum of guinea-pigs in immense doses cause death. In studying successive direct smears from the throats of diphtheria patients no evidence of change from one type to the other was noted. Attempts have been made to give more virulence and some toxicity to a few of these varieties by successive peritoneal inoculations, and by growing the organisms in symbiosis with several other organisms, but in no instance has any increase of pathogenicity or decided change in morphological or cultural characteristics been noted. Since there are so many different forms or varieties of diphtheria- like bacilli, it is quite possible that some of them are derived from strains of the diphtheria bacillus and that under certain conditions they readily regain its characteristics. This seems to be the only way to explain the apparent discrepancies in the results obtained by different observers. Such closely related varieties, however, do not appear to be common and we have up to the present been unable to obtain them. So we may safely say that in this region, at least, non-toxic diphtheria-like organisms retain their characteristics under various artificial and natural conditions, and that they may be regarded from a public health stand-point as harmless. Resistance to Heat, Drying, and Chemicals. — The thermal death-point with ten minutes' exposure is about 60° C, with five minutes 70° C. Boiling kills almost instantly. The bacillus has about the average resis- tance of non-spore-bearing bacteria to disinfectants. In the dry state and exposed to diffuse light diphtheria bacilli usually die in from a few hours to a few days, but they may live for months; when in the dark, or pro- tected by a film of mucus or albumin, they may live for even longer periods. Thus we found scrapings from a dry bit of membrane to contain vigorous and toxic living bacilli for a period of four months after removal from the throat, and if the membrane had not been at that time completely used, living bacilli could probably have been obtained for a much longer period. On slate- and lead-pencils, toys, tumblers, as well as on paper money, they may live for several weeks, while on metal coins they die in twelve to thirty-six hours. In culture media, when kept at the blood heat, they usually die after a few weeks; but under certain conditions, as when sealed in tubes and protected from heat and light, they retain their life and toxicity for years. The bacillus is not sensitive to cold, for we found about 10 per cent, of the bacilli to retain their vitality and toxicity after exposure for two hours to several hundred degrees below zero. At temperatures just below, freezing they may remain alive for a number of weeks. Transmission of Diphtheria. — ^The possibility of the transmission of diphtheria from animals to man cannot be disputed; we have met 20 306 BACILLUS AND THE BACTERIOLOGY OF DIPHTHERIA two instances in which cats had malignant diphtheria, and many other animals can be infected, but as we have said there are few authentic cases of such transmission on record. So-called diphtheritic disease in animals and birds is usually, if not always, due to other microorganisms than the diphtheria bacilli. The toxic bacilli have been found on soiled bedding or clothing of a diphtheria patient, or drinking-cups, candy, shoes, hair, slate- pencils, etc. These sources of infection by which the disease may be indirectly transmitted are, however, the less important ones. The usual source of the bacilli are the discharges of diphtheria patients; the secre- tions from the nose and throat of convalescent cases of diphtheria in which the toxic bacilli persist, and from the healthy throats of individ- uals who acquired the bacilli from being in contact with others having virulent germs. When we consider the number of healthy carriers and that it is only the severe types of diphtheria that remain isolated during their actual illness, the wonder .is not that so many, but that so few, persons contract the disease. Susceptibility to and Immunity against Diphtheria. — An individual susceptibility, both general and local, to diphtheria, as in all infec- tious diseases, is necessary to contract the disease. Age has long been recognized to be an important factor in diphtheria. Children within the first six months of life are but little susceptible, exceptionally infants of a few weeks are attacked. The time of greatest susceptibility is between the second and tenth year. After the tenth year susceptibility decreases. Young animals born of mothers immune to diphtheria possess nearly the same degree of immunity as their mothers. They gradually lose this but retain traces up to six to twelve months. The human infant is now known to receive immunity from its mother. This immunity has been shown to depend on the presence of antitoxin. The Persistence in Man's Blood of Homologous and Alien Antitoxin.— Antitoxins and other antibodies produced in an animal disappear more rapidly when introduced into the blood of another species than into one of the same species. In man an alien serum must be used in all except exceptional cases. In our experiments in guinea-pigs we have found that the homologous antitoxin was retained in appreciable amounts for at least six months, while the heterologous antibodies were noticeable to the same extent for only four weeks. There is a rather rapid loss of horse-produced antitoxin during the first few days and then a slow loss, becoming more and more gradual until final elimination at the end of ten days to three weeks. The larger the amount of antibodies injected the longer will be the time before the elimination of effective amount. For a discussion of the nature of antitoxin see page 168. For a presentation of the therapeutic use of antitoxin see Part III. Mixed Infection in Diphtheria.— Toxic diphtheria bacilli are not the only bacteria present in human diphtheria. Various cocci and bacilli more particularly streptococci, staphylococci, pneumococci, and influ- PSEUDOMEMBRANOUS EXUDATIVE INFLAMMATIONS 307 enza bacilli, are always found actively associated with Loffler's bacillus in diphtheria, playing an important part in the disease and leading often to serious complications (sepsis and bronchopneumonia) . Investi- gations indicate that when other pathogenic bacteria are associated with the diphtheria bacilli they mutually assist one another in their attacks upon the mucous membrane, the streptococcus being par- ticularly active in this respect, often opening the way for the invasion of the LofHer bacillus into the deeper tissues or supplying needed con- ditions for the development of its toxin. In most fatal cases of broncho- pneumonia following laryngeal diphtheria we find not only abundant pneumococci or streptococci in the inflamed lung areas, but also in the blood and tissues of the organs. *As these septic infections due to the pyogenic cocci are in no way influenced by the diphtheria antitoxin, they frequently are the cause of the fatal termination. Other bacteria cause putrefactive changes in the exudate, producing alterations in color, e. g., B. pyocyaneus and offensive odors, e. g., B.fusiformis. Fig. 123. — Vincent's bacillus with accompanying spirochetes. Pseudomembranous Exudative Inflammations Due to Bacteria other than the Diphtheria Bacilli. — ^The diphtheria bacillus, though the most usual, is not the only microorganism that is capable of produc- ing pseudomembranous inflammations. There are numerous bacteria present almost constantly in the throat secretions, which, under certain conditions, can cause local lesions very similar to those in the less- marked cases of true diphtheria. •The streptococcus and pneumococcus are the two forms most frequently found in these cases, but there are also others, such as Vincent's bacillus, which, under suitable condi- tions, excite this form of inflammation, but without serious constitu- tional symptoms. The pseudomembranous . angina accompanying scarlet fever, and to a less extent other diseases, may not show the presence of diphtheria 308 BACILLUS AND THE BACTERIOLOGY OF DIPHTHERIA bacilli, but only the pyogenic cocci, especially streptococci, or, more rarely, some varieties of little-known bacilli. The deposit covermg the inflamed tissues in these non-specific cases is, it is true, usually but not always, rather an exudate than a true pseudomembrane. Eelation of Bacteriology to Diagnosis.— We believe that all expe- rienced clinicians will agree that, when left to judge solely by the appearance and symptoms of a case, there are certain mild exudative inflammations of the throat of similar appearance some of which trans- mit diphtheria while others do not. The doubtful cases that have the diphtheria bacilli in the exudate, are capable of giving true characteristic diphtheria to others, or later develop it characteristically themselves, while those in whose throats no- diphtheria bacilU exist can under no condition give true charac- teristic diphtheria to others or develop it themselves. It is, indeed, true, as a rule, that cases presenting the appearance of ordinary follic- ular tonsillitis in adults are not due to the diphtheria bacillus. On the other hand, in small children mild diphtheria very frequently occurs with the semblance of rather severe ordinary follicular tonsillitis, due to the pyogenic cocci; and in large cities where diphtheria is prevalent all such cases must be watched as being more or less suspicious. Most observers agree with us in thinking that if in any case exposure to diphtheria is known to have occurred, even a slightly suspicious sore throat should be regarded as probably due to the diphtheria bacilli. If, on the other hand, no cases of diphtheria have been known to exist in the neighborhood, even cases of a more suspicious nature would probably not be regarded as diphtheria. Now that we know about 50 per cent, of children have a fairly constant supply of antitoxin in their blood, we have reason to believe that many of these doubtful cases are simply diphtheria carriers in which the bacilli are taking little or no part in making the lesions. Like any carriers they are dangerous to others. The presence of irregular-shaped patches of adherent grayish or yellowish-gray pseudomembrane on some other portions than the tonsils is, as a rule, an indication of the activity of the diphtheria bacilli. Restricted to the tonsils alone, their presence is less certain. Occasionally, in scarlatinal angina or in severe phlegmonous sore throats, patches of exudate may appear on the uvula or borders of the faucial pillars, and still the case may not be due to the diphtheria bacilli; these are, however, exceptional. Thick, grayish pseudomembranes which cover large portions of the tonsils, soft palate, and nostrils are almost invariably the lesions produced by diphtheria bacilli. The Very great majority of cases (Si pseudomembranes of exudative laryngitis, in the coast cities at least, whether an exudate is present in the pharynx or not, are due to the diphtheria bacilli. Nearly all membranous affections of the nose are true diphtheria. When the membrane is limited to the nose the symptoms are, as a rule, very slight; but when the nasopharynx is involved the symptoms are usually grave. Most cases of pseudomembranes and exudates, entirely confined BACTERIOLOGICAL DIAGNOSIS 309 to portions of the tonsils in adults, are not due to the diphheria bacilli. Cases presenting the appearances found in scarlet fever, in which a thin grayish membrane lines the borders of the uvula and faucial pillars, are rarely diphtheritic. As a rule pseudomembranous inflam- mations complicating scarlet fever, syphilis, and other infectious dis- eases are due to the activity of the pathogenic cocci and other bacteria, induced by the inflamed conditions of the mucous membranes due to the scarlatinal or other poison. The possibility of these persons being carriers of diphtheria bacilli must always be kept in mind. Paralysis following a pseudomembranous inflammation is an almost positive indication that the case was one of diphtheria, although slight paralysis has followed in a very few cases in which careful cultures have revealed no diphtheria bacilli. These, if not true diphtheria, must be considered very exceptional cases. Bacteriological Diagnosis. — From the above it is apparent that fully developed characteristic cases of diphtheria are readily diagnosticated, but that many of the less marked, or at an early period undeveloped, cases are diflficult to differentiate. In these cases cultures are of the utmost value, since they enable us to isolate those in which diphtheria-like bacilli are found, and to give preventive injections of antitoxin to both the sick and those in contact with tbem, if this has not already been done. As a rule cultures do not give us as much information as to the gravity of the case as the clinical appearances, for before the lapse of the twelve hours required for the laboratory report, the extent of the disease usually allows a diagnosis. The reported absence of bacilli in a culture must be given weight in proportion to the skill with which the culture was made, the suitableness of the media, the location of the disease, and the knowledge and experience of the one who examined it. Diphtheria does not occur without the presence of the diphtheria bacilli; but there have been many cases of diphtheria in which, for one or another reason, no bacilli were found in the cultures by the examiner. In many of these cases later cultures revealed them. The reverse is also true, the presence of diphtheria bacilli in throats without clear clinical signs of diphtheria in no sense makes it a case of diphtheria. In a convalescent case the absence of bacilli in any one culture indicates that there are certainly not many bacilli left in the throat, but even repeated cultures cannot absolutely prove their total absence, for in some deep tonsillor crypt a few bacilli may remain in the epithelial cells. The physician must have intelligence to use advantageously laboratory findings. Teclmic of the Bacteriological Diagnosis. — Colkction of the Animal Blood Serum and its Preparation for Use in Cultures. (Seep. 101.) Swab for Inoculating Culture Tubes. — The swab we prefer to use to inoculate the serum is made as follows: A stiff, thin, iron rod, 6 inches in length, is roughened at one end by a few blows of a hammer, and about this end a little absorbent cotton is firmly wound. Each swab is then placed in a separate glass tube, and the mouths of the tubes are plugged with cotton. The tubes and rods are then sterilized by dry heat at about 150° C. for one hour, and stored for future use. These iron rods have proved more serviceable for making 310 BACILLUS AND THE BACTERIOLOGY OF DIPHTHERIA inoculations than platinum-wire needles or wooden sticks, especially in young children and in laryngeal cases. It is easier to use the cotton swab in such cases and it gathers up so much more material for the inoculation that it has seemed more reliable. The wood unless very carefully selected is apt to break and is too thick to use in the nose. For convenience and safety in transportation "culture outfits" have been devised, which consist usually of a small wooden box containing a tube of blood serum, a tube holding a swab, and a record blank. These "culture outfits" may be carried or sent by messenger or express to any place desired. Directions for Inoculating Culture Tubes loith the Exudate. — ^The patient is placed in a good light, and, if a child, properly held. The swab is removed from its tube, and, while the tongue is depressed with a spoon, is passed into the pharynx (if possible, without touching the tongue or other parts of the mouth), and is rubbed gently but firmly against any visible membrane on the tonsils or in the pharynx, and then, without being laid down, the swab is immediately inserted in the blood-serum tube, and the portion which has previously been in contact with the exudate is rubbed a number of times back and forth over the whole surface of the serum. This should be done thoroughly, but it is to be gently done, so as not to break the surface of the serum. The swab should then be placed in its tube, and both tubes, thin cotton plugs having been inserted, are reserved for examination or sent to the laboratory or collecting station (as in New York City). If sent to the Health Department laboratories for examination the blank forms of report which usually accompany each "outfit" should be filled out and forwarded with the tubes. Where there is no visible membrane (it may be present in the nose or larynx) one swab should be rubbed over the mucous membrane of the pharynx and tonsils, and another in the nasal cavities, and a culture made from these. In very young children it should be remem- bered that the throat often contains food or vomited matter. This should be cleared away before using the swab to make the bacteriological examination easier. Under no conditions should any attempt be made to collect the material shortly after the application of strong disinfectants (especially solutions of corrosive sublimate) to the throat. Cultures from the nostrils are often more successful if the nostrils are first cleansed with a spray of sterile normal salt solution. Examination^of Cultures. — ^The culture tubes which have been inocu- lated, as described above, are kept in an incubator at 37° C. for at least twelve hours, and are then ready for examination. When great haste is required, even five hours will often suffice for a sufficient growth of bacteria for a skilled examiner to decide as to the presence or absenec of the bacilli. The absence of bacilli at this period cannot be relied upon. In primary cultures it is wise to reincubate tubes taken out under sixteen hours in which no bacilli were found. A small percentage of these will yield positive results after a few hours' further incubation. On inspection it will be seen that the surface of the blood serum is dotted with numer- ous colonies, which are just visible. No diagnosis can be made from siinple inspection; if, however, the serum is found to be liquefied or BACTERIOLOGICAL DIAGNOSIS 311 shows other evidences of contamination the examination will probably be unsatisfactory. In order to make a microscopic preparation, a clean platinum needle is inserted into the tube and quite a large number of colonies are swept with it from the surface of the culture medium, a part being selected where the most suitable colonies are found. A sufficient amount of the bacteria adherent to the needle is washed off in a tiny droplet of water previously placed on the glass slide and smeared over its surface. The bacteria on the glass are then allowed to dry in the. air. The glass slide is then passed quickly through the flame of a Bunsen burner or alcohol lamp, three times in the usual way, covered with a few drops of Loffler's solution of alkaline methylene blue, and left without heating for five to ten minutes. It is then rinsed off in clear water, dried, and mounted in balsam. When other methods of staining are desired they are carried out in the proper way (see Methods of Staining). In the great majority of cases one of two pictures will be seen with the xV oil-immersion lens — either an enormous number of character- istic Lofiler bacilli, with a moderate number of cocci,, or a pure culture of cocci, mostly in pairs or short chains. (See Streptococcus.) In a few cases there will be an approximately even mixture of LofHer bacilli and of cocci, and in others a great excess of cocci. Besides these there will be occasionally met preparations in which, with the cocci, there are mingled bacilli more or less resembling the Loffler bacilli. These bacilli, which are usually of the pseudodiphtheria type of bacilli (see Fig. 122), are especially frequent in cultures from the nose. In not more than one case in twenty will there be any serious diffi- culty in making the diagnosis, if the serum in the tube is moist and has been properly inoculated. In the doubtful case another culture must be made or the bacilli plated out and tested in pure culture. Direct Microscopic Examination of the Exudate. — ^An immediate diag- nosis without the use of cultures is often possible from a microscopic examination of the exudate. This is made by smearing a slide or cover-glass with a little of the exudate from the swab, drying, heating, stainingj and examining it microscopically. This examination, however, is much more difficult, and the results are more uncertain than when the slides are prepared from cultures. The bacilli from the membrane are usually less typical in appearance than those found in cultures, and they are mixed with fibrin, pus, and epithelial cells. They may also be very few in number in the parts reached by the swab, or bacilli may be found which closely resemble the Loffler bacilli in appearance, but which differ greatly in growth and in other characteristics, and have absolutely no connection with them. When in a smear containing mostly cocci a few of these doubtful bacilli are present, it is impossible either to exclude or to make the diagnosis of diphtheria with certainty. Although in some cases this immediate examination may be of the greatest value, it is not a method suitable for general use, and should always be controlled by cultures. When carried out in the best manner an experienced bacteriologist may obtain markedly accurate results. Higley, 312 BACILLUS AND THE BACTERIOLOGY OF DIPHTHERIA in a series of consecutive throat cases, made the same diagnosis from the direct examination of smears as the Health Department laboratory made from the culture. To get the exudate he used a probe armed with a loop of heavy copper wire which had been so flattened as to act as a blunt curette. He then made thin smears from the exudate. After drying and fixing by heat the smears were stained for five seconds in a solution made by adding five drops of Kiihne's carbolic methylene blue to 7 c.c. of tap-water. After washing and drying they were stained for one minute ii^ a solution of 10 drops of carbol-f uchsin in 7 c.c. of water. The dilute solution should be freshly prepared. The diphtheria bacilli will appear as dark red or violet rods, and their contour, mode of division, and arrangement are manifest. Animal Inoculation as a Test of Toxicity. — If the determination of the toxicity of the bacilli found is of importance, animal inocu- lations must be made. Experiments on animals form the only method of determining with certainty the toxicity of the diphtheria bacillus. For this purpose alkaline broth cultures of forty-eight hours' growth should be used for the subcutaneous inoculation of guinea-pigs. The amount injected should not be more than one-fifth per cent, of the body Weight of the animal inoculated, unless controls with antitoxin are made. In the large majority of cases, when the bacilli are toxic, this amount causes death within seventy-two hours. If a good growth is not obtained in nutrient bouillon, ascitic broth should be used. At the autopsy the characteristic lesions already described are found. Bacilli which in cultures and in animal experiments have shown them- selves to be characteristic may be regarded as true diphtheria bacilli, and as capable of producing diphtheria in man under favorable conditions. For an absolute test of specific toxicity antitoxin must be used. A guinea-pig is injected with antitoxin, and then this and a control animal, with 2 c.c. of a broth culture of the bacilli to be tested; if the guinea-pig which received the antitoxin lives, while the control dies, it was surely a diphtheria bacillus which killed by means of diphtheria toxin — or, in other words, not simply a toxic bacillus, but a toxic diph- theria bacillus. Quite a number of bacilli have been found which kill 250-gram guinea-pigs in doses of 2 to 15 c.c, and yet are unaffected by antitoxin. These bacilli, though slightly virulent to guinea-pigs, pro- duce no diphtheria toxin, and so cannot, to the best of our belief, .produce diphtheria in man (see p. 301). The intradermal test may be used satis- factorily by the trained worker in testing toxin and antitoxin, resulting in the saving of the number of pigs used in the test. Rabbits can also be used in place of guinea-pigs. The Schick Beaction. — Recently Schick published a method by which the presence of antitoxin in the blood and tissue can be determined very easily. A minute quantity of toxin is injected intracutaneously, and a local reaction follows if there is less than ^V of a unit of antitoxin per cubic centimeter of blood. This amount is considered sufficient to pro- tect against diphtheria. The explanation of the test is that when no THE SCHICK REACTION 3l3 antitoxin is present the toxin acts on the skin; when antitoxin is present it neutralizes the toxin so that no poisoning occurs. A negative reaction therefore indicates the presence of antitoxin. A standard diphtheria toxin is diluted at first 1 to 10 in 0.5 per cent, phenol; this dilution will keep in the ice-box with little dieterioration for at least two weeks. For use further dilutions are made in normal saline, of such strength that 0.1 c.c. contains ^V of the minimum lethal -dose for the guinea-pig. This amount is injected intracutaneously on the flexor surface of the arm or forearm. If the injection has been made properly a definite wheal appears which lasts for several minutes. It is of the utmost importance that the injection be made in and not under the skin. A positive reaction appears in twenty-four to thirty-six hours, and is characterized by a circumscribed area of redness and slight infiltration which measures from 1 to 2 cm. in diameter. It persists for seven to ten days, and on fading shows superficial scaling and a persistent brownish pigmentation. The test represents a true irritant action of non-neutralized toxin. Pseudoreactions are seen occasionally in young children and rather frequently in adults, who may have a large amount of antitoxin. These are local sensitization phenomena of a general protein character, and can be distinguished from the true reaction. They appear earlier, are more infiltrated, less sharply circnmscribed and disappear in twenty-foiu" to forty-eight hours. On fading they leave a faintly pigmented area which may show superficial scaling. They can also be obtained with the neutralized toxin and at times with dilutions of plain broth. By the use of this test a large number of individuals have been shown to be naturally immune. Negative reactions were obtained in 93 per cent, of the newborn, in 57 per cent, during the first year of life, in 37 per cent, between two and five years, and 50 per cent, between five and fifteen years. In adults the negative reactions were as high as 90 per cent. By applying the test, therefore, passive immunization with antitoxin of a large number of individuals can be omitted and the disagreeable symptoms of sensitization avoided. Of 400 cases of scarlet fever, showing a negative Schick reaction and receiving no immuniz- ing dose of antitoxin, not one developed clinical diphtheria. Of these cases 25 per cent, were carriers of virulent diphtheria bacilli. An early series of cases in institutions gave the results shown in the table below. Summary op Schick Tests in 2700 Normal CmLDHEN. 2 to 4 years 4 to 6 years 6 to 8 years 8 to 10 years 10 to 12 years 12 to 14 years 14 to 16 years 2700 2120 580 21.4 Per cent. Total. -Schick +Schick. +Schick. 62 42 20 32.2 318 236 82 25.7 444 347 97 21.8 597 462 135 22.6 584 459 125 21.4 506 416 90 17.7 189 158 31 16.4 3l4 BACILLUS AND THE BACTERIOLOGY OP DIPHTHERIA From these figures, then, we may assume that the percentage of individuals susceptible to diphtheria is greatest between the ages of one to four years. It is less during the second six months of life and less in older children and in adults. Our results in scarlet fever cases closely approach those obtained by Schick, whose table shows positive reactions in 7 per cent, of the newborn, in 43 per cent, during the first year of life, in 63 per cent, between two and five years and 50 per cent, between five and fifteen years. In adults the positive reactions were not more than 10 per cent. Toxin-antitoxin Inoculations. — ^Behring was the first who emphasized the practical application of toxin-antitoxin mixtures in which the toxin had been so neutralized as to be no longer poisonous, but still retained some toxin in loose combination. The use of this immunizing agent is the result of a long series of investigations as to the possibility of producing antitoxin through injection of these mixtures. In 1895 Babes carried out successful experiments in guinea-pigs. Since 1897 the horses in the Health Department of New York City, which are used to produce diphtheria antitoxin, have been immunized at first with these neutral mixtures. In 1903 one of us published the records of a number of horses showing that three injections might cause produc- tion of several hundred units of antitoxin in each cubic centimeter of serum. Theobald Smith later made a careful study of the subject in guinea-pigs and suggested the toxin- antitoxin injections in children for practical inimunization but never tried it. Behring deserves, therefore, the credit not of the discovery of the method but of actually applying the toxin-antitoxin mixtures for the immunizing of persons against diphtheria. We have personally watched the result in a series of over 1000 cases, that had been actively immunized with diphtheria toxin-antitoxin. These susceptible individuals were selected by means of the Schick test out of a total of about 10,000 children and adults in 10 different institutions. The mixtm^es of toxin-antitoxin that were used for immunization contained about 2L+ doses of toxin to each cubic centimeter, and were either neutral (66-70 per cent. L+ to each unit of antitoxin) or slightly toxic (80-90 per cent. L+ to each unit of antitoxin) to the guinea-pig. The dose was varied from 0.5 c.c. to 1.0 c.c, and the number of injections from one to three. The injections were made subcutaneously at intervals of seven days. The local reactions at the site of injection were generally mild; in the older children and adults, the redness and swelling were more marked. General symptoms, like malaise, and temperature of 100° 102° F., were noted in 10 to 20 per cent, of the cases; in a few the temperature reached 104° F. The symptoms lasted twenty-four to forty-eight hours, and then rapidly subsided. Both local and general symptoms were especially evident in those who showed a susceptibility to the protein by giving a combined pseudo and true Schick reaction. No harmful after-effects were noted in several thousand injections. The retests with the Schick reaction showed that only 30 to 40 TOXIN-ANflTOXIN JNOCVlATIONS 315 per cent, became immune three weeks after the first injection; about 50 per cent, at four weeks, 70 to 80 per cent, at six weeks, and 90 to 95 per cent, at eight to twelve weeks. The best results were obtained with the full immunization, consisting of three injections of 1 c.c. each, given at weekly intervals. The duration of the active immunity was studied in a group of children that was followed up for over one and one-half years; these cases showed that the active immunity per- sisted for at least that length of time. It is possible that the immunity induced by the injections of toxin-antitoxin starts a continued cellular production of antitoxin which otherwise would have appeared much later in life. From their results Park and Zingher conclude that it is advisable to immunize children soon after the first year of life so as to afford them a protection against diphtheria at a time when the disease is most dangerous. In addition such young children, by not having any hyper- sensitiveness to the bacillus protein, show very mild local and con- stitutional symptoms after the injections. An immune child population could thus be developed with the result that fresh clinical cases would be prevented and the bacillus-carrier menace would probably soon dis- appear as a hygienic factor in our communities. Interesting and parallel results were noted in guinea-pigs and horses. Guineor-pigs are fairly resistant to active immunization with diphtheria toxin-antitoxin, and in that respect they show an almost complete parallelism to the positive Schick cases among human beings. After injections of toxin-antitoxin, an antitoxic immunity develops slowly from the sixth to eighth week. Horses, on the other hand, as a rule, correspond to those human beings who are naturally immune in their behavior toward small doses of toxin-antitoxin. They both give a ready response, even after a single injection of toxin-antitoxin, and show a distinct increase in the antitoxin content toward the end of the first week. Occasionally a horse is found that has no antitoxin in the control bleeding; such animals respond slowly to small doses of toxin-antitoxin. It is probable that the tissue cells of the naturally immune human beings and the majority of horses have acquired the property of giving a quick and easy response to the stimulation of diphtheria toxin. REFERENCES. Brown and Smith: Jour. Med. Research, vol. xxx, No. 3, p. 443. Clahk: Jour, of Inf. Dis., 1910, vii, 335. Davis: Medical News, April 29, 1899. Fleming: On the Etiology of Acne Vulgaris and its Treatment by Vaccines, Lancet, 1909, i, 1207. Halle and Anatte: Ann. de. dermat. at de Syph., 1907, p. 184. Pahk, Zingheb and Sbbota: Archives of Ped., 1914. Schick: Die Diphtheritoxin-Hautreaktion des Menschen, etc. Miincben. med. Wchnschr., 1913, ix, 2608-2610. Stanton, E. M.: The Isolation and Cultural Characteristics of B. Acne, Centralbl. f. Bakt., orig. I, 1912, Ixvi, 386. -XJnna: Histopathology of Diseases of the Skin, 1894. Wesbeook, Wilson, and McDaniel: Transactions of the Association of American Physicians, 1900. Williams: Journal of Med. Research, June, 1902, viii 83. CHAPTER XXI. INTESTINAL BACTERIA. SOUR MILKS. Conditions Influencing Bacterial Development. — ^The alimentary tract both as regards foodstuff, varying degrees of oxygen tension and reaction gives so many different conditions for growth that many varieties of bacteria find an optimum environment at some level. Thus, the mouth offers, in general, an aerobic arfea, although with its various crevices between the teeth, in the folds of the mucous membranes and the crypts of the tonsils, anaerobic types may flourish with aerobic types, which absorb the oxygen. The stomach when its acidity is normal, destroys bacteria unless protected by food particles. As soon as the duodenum is reached the oxygen tension becomes low and from there on an anaerobic condition exists. More aerobic conditions are reached in the lower colon and rectum. It is evident that the degree of digestion of the food and the proportion of unabsorbed carbohydrate and protein at different levels will also influence the flora, likewise the products of bacterial growth at one level will influence the growth of other types at this and at lower levels. As the intestinal contents pass downward, they carry with them the bacteria from a higher level, but if the conditions are not favorable for their growth they are quickly overshadowed by the more adapted types. In the lower part of the large intestine and in the rectum due to the gradual loss of water, there is marked tendency of the bacteria in fecal mass to die or to become so attenuated as to be incapable of further growth. Development of the bitestinal Flora. — ^At birth the meconium is sterile unless fetal infection has taken place due to general infection in the mother. Shortly after birth bacteria and yeasts are found entering either through the rectum or by way of the mouth from swallow- ing saliva or food. Very soon, through further injestion of bacteria and by multiplication of different types at their optimum levels, a more or less distinctive bacterial flora is established, distinctive in that the breast-fed infant, the artificially fed infant and the adolescent or adult each has, considering the dominant types, a somewhat characteristic flora. This flora is susceptible, however, to variation through the administration of cathartics, by changes in diet, by implantation through feeding of cultures or during alimentary infections. Importance of the Intestinal Flora. — ^Whether the development of such an intestinal flora is of physiological advantage to the host is open to question. Successful experiments in raising animals with a sterile intestinal tract show that such a flora is not a physiological DOMINANT INTESTINAL BACTERIAL TYPES 317 necessity. Experiments in which animals so raised did not do well, cannot be considered evidence against this view, as the conditions of the experiment may have deleteriously influenced development. Undoubt- edly the character and balance of the flora does act as a protection under some circumstances, being antagonistic to the implantation of an exogenous pathogenic type. Abnormal variations, however, in balance between the dominant types may be the basis of intestinal symptoms (see below). In one sense part of the intestinal flora is a potential menace to the host. The intestinal mucosa is an obstacle to the passage of the bacteria into the tissues, lymph, and blood. A few, however, evidently escape as the fairly frequent localization in the gall-bladder, kidney and elsewhere show. Probably this occurs constantly in small numbers, but because of the slight virulence of the majority of such types they are promptly disposed of. Fig. 124. — B. bifidus, representing the various forms described; the irregularly stained or vesicular forms being from old cultures. X about 1800 diameters. Dominant Intestinal Bacterial Types.— B. bifidus.— A strict anaerobe isolated by Tissier from the stools of breast-fed infants and from the superficial ducts of the mammary glands of mothers. It is Gram- positive or contains a Gram-positive granule, the remainder of the bacillus being Gram-negative. In stools it is a slender bacillus with one end tapering, the other club-shaped. In cultures it has the property of developing bifid ends. It produces acid freely but no gas from lactose or other sugars. B. Acidophilus.— A type of bacterium characterized by its acid toler- ance. B. bulgaricus (see below) is related to this type, both being included in the aciduric group. It is a Gram-positive, non-spore-bearing pleomorphic- bacillus, frequently forming chains in cultures and produc- ing acids freely from carbohydrates. A similar gas-producing type has been described. Enterococcus or Micrococcus Ovalis. — A Gram-positive, oval coccus, usually in pairs or chains. Capsules may be present. Due to delayed gl8 INTESTINAL BACTERIA cleavage pseudobacillary types may develop on certain media. It is aerobic, facultative anaerobic, non-liquefying, producing acid from glucose and lactose, and coagulating milk. B. Putrificus. — ^An anaerobic slender. Gram-positive, motile putre- factive bacillus resembling B. tetani in its morphology. B. Mesentericus. — Like B. subtilis, it is essentially a Gram-positive, spore-bearing, aerobic, proteolytic bacillus, that is, forming products of decomposition in protein media when glucose is absent. B. (aerogenes) capsulatus. — See p. 332. B. CoU and Allied Types.— See pp. 323-334. Intestinal Flora of Breast-fed Infants. — ^The Micrococcus ovalis types preponderate in the duodenum and are most numerous when digestion is in progress. The dominant type in the remainder of the small intes- tine is B. aerogenes, B. coli being most numerous about the cecal level. B. bifidus finds it optimum in the large intestine. Putrefactive bacteria are uncommon in normal breast-fed infants. Fecal smears show a preponderance of Gram-positive types. Intestinal Flora of Artificially Fed Infants. — ^The Gram-negative B. coli and B. aerogenes as well as the M. ovalis are relatively increased as compared with the breast-fed infant, while B. bifidus types are diminished, being replaced by colon types and B. acidophilus. Proteo- lytic action is evidenced in the alkaline reaction of the feces. In general the distribution of the types follows that in breast-fed infants. The character of the flora is susceptible to variation according to the balance of carbohydrate to protein and by changes in the carbohydrate employed. Intestinal Flora of Adults. — The duodenum except during digestion has a low bacterial content. Cocci predominate in the upper small intestine. B. coli types become numerous in the lower small intestine, and together with B. mesenteric types predominate about the cecal level. These and, to a limited extent, the proteolytic anaerobes constitute the flora of the large intestine. The flora as compared with that of infants is character- ized by the absence of essential carbohydrate-fermenting types. The carbohydrate of adult life is mostly starch and the products of its cleavage are probably quickly absorbed, leaving the more slowly cleaved and absorbed protein as available foodstuff for the bacteria. The colon types and the aerobic and anaerobic proteolytic types which can utilize this material therefore dominate the flora. Variations in the Bacterial Flora and Their Significance. — The flora as given for the three types is basically dependent on the diet. Thus, in the nursing infant, the carbohydrate is sufficient in amount to supply the conditions for growth of the essentially fermentative types. These because of their acid production limit to a marked degree the multi- plication of other types, especially the proteolytic types. Although in the case of the artificially fed infant the available carbohydrate is less in amount, it is still suflBcient to lead to relatively the same result. If, however, the protein proportion of the food be increased, a relative or complete suppression of acid-producing types with an increase of the proteolytic types occurs, approximating the flora of the adult. The TYPES OF SOUR MILK 319 activity of B. coli as regards end-products depends also on the avail- ability of carbohydrates. If present in sufficient amounts, lactic acid and other end-products of carbohydrate fermentation are produced. When absent, the end-products of protein cleavage, HjS, NH3, and indol, are produced. It is easy to see how carbohydrate fermentation acts as a check to excessive putrefactive activity, which condition is delete- rious to the host. The bacterial flora, as a rule, is little influenced by the injestion of bacteria in the foods. When infection occurs, such as with dysentery or cholera, these organisms and others similar to them may dominate the intestinal- flora and a specific disease develops. Apart from these well-known types of disease there are other morbid conditions which, although of unknown etiology, are intimately associated with changes of the intestinal flora, even if these changes are not of etiological importance. These can be subdivided as to whether (a) the products of bacterial proteolysis, (6) the products of carbohydrate fermentation or (c) both are at fault. The last is the least understood. In the first case a condition of auto-intoxication develops with the presence of ethereal sulphates in the urine. In the second the produc- tion of abnormal or the excessive production of the products of fer- mentation as simple hyperacidity or less-known irritative products due to the overactivity of B. aerogenes or B. capsulatus are the probable cause of the symptoms. Diarrheal conditions due to the latter, even though chronic, do not show evidences of toxemia. On this basis attempts have been made to influence these conditions. Cathartics, intestinal antiseptic and starvation have no appreciable beneficial value. Changes in diet so adapted as to limit or supply the optimum foodstuff for certain types or the addition to the diet of exog- enous lactic acid-producing bacilli and the products of their growth have been more successful. The latter method first recognized by Herter has been popularized by Metchnikoff's propaganada of sour-milk therapy. Types of Sour Milk (Lactic Acid Milks). — ^Milk will sour naturally due to lactic acid bacteria which are contaminations of the milk from its surroundings. Buttermilk is a naturally soured product. Metchnikoff advised the use of B. bulgaricus (see below) found in the sour milk of Eastern Europe and in Asia. Many lactic acid-milk preparations are available. To control the appearance of the end-product and to have a uniformly palatable preparation, it is necessary to heat the milk and add a starter. Thus "koumyss" is fermented by lactic acid bacteria and yeasts, "maadsoun" and "zoolak" by B. bulgaricus and lactic acid producing streptococci and diplococci. Many milk dealers now offer lactic acid milks under various trade names, the starters used being those mentioned in various combinations. B. bulgaricus alone gives too sour a milk. B. Bulgaricus. — ^The types included under this name belong to the aciduric group of bacilli. They are non-motile, Gram-positive (except involution forms which are negative) aerobic, facultative anaerobic, 320 INTESTINAL BACTERIA non-liquefying bacilli occurring singly or in chains. They require carbo- hydrates for their growth, growing best on or in milk or whey media, producing large amounts of acids, which, however, do not readily inhibit their growth. On whey agar, they produce characteristic colonies (see Fig. 125). Milk is quickly coagulated. White and Avery have differentiated two groups depending upon the amount of lactic acid produced. w •iSJ^ji I i V «■*■'■-.' - ■ '- Fig. 125.- -B. bulgaricus; seventh day (44°) colony. . Whey agar plate. (White and Avery.) X 50 diameters. Other B. Bulgaricus Preparations. — Other methods of using B. bul- garicus in the form of liquid cultures, dried tablets or oily suspensions are of less value than milk preparations. Their use is based on the assumption that the bacilli will be active in the intestinal tract, which is an assumption of doubtful validity. Furthermore, they do not supply the preformed acid, at least to any extent, as contained in milk prep- arations. Bendick has found that many of the tablets and probably the oily suspensions as well, contain very few viable organisms; and the bacilli in liquid preparations tend to die out rapidly. At best, this preparation does not compare with sour milk in its bacillary content. In conditions of toxemia the- limitation of proteins with liberal drinking of sour-milk preparations, flooding the intestinal tract with pre- formed acids is beneficial, but the implantation of B. bulgaricus as an intestinal type probably does not occur. For the latter purpose it would seem more logical to feed B. acidophilus which is naturally an intestinal TYPES OF SOUR MILK 321 parasite. In the special fermentative types of diarrhea due apparently to an overgrowth by B. capsulatus the liberal use of acid buttermilk Fig. 126. — B. bulgaricus. X 1000 diameters. (Piffard.) Fio; 127. "Lactic acid" milk contaiuing B. bulgaricus and a lactose-fermenting Btreptococcus. and a carbohydrate-free diet will lead to improvement. The same treat- ment should be applied if B. aerogenes is the apparent causative agent. 21 322 INTESTINAL BACTERIA The Influence of Available Foodstuffs in Specific Intestinal Infections. — In infections due to members of the typhoid-dysentery group or to the cholera vibrio the presence or absence of available carbohydrates may have an important bearing on the products developed by these bacteria and thus on the disease. Each will probably utilize carboi^ydrate if available, in preference to proteins (protein sparing). There is no evidence that the acids produced from carbohydrate cleavage are any more harmful than those produced by any non-pathogenic fermentative organism. Furthermore, the acid produced would limit their further growth. Not only may the products of protein cleavage be harmful in themselves, but the toxicity of the products of growth, whether a free toxin is produced or not, is enhanced in amount when growth is due to the utilization of protein material. A free carbohydrate diet, therefore, should be valuable in these diseases not only to spare protein but as an incentive to increased activity of the aciduric types which would aid by limiting multiplication. Although difhculties due to inflammatory conditions of the intestines and other causes may interfere with the efficiency of this mode of treat- ment, its results in general have been good. Coleman and Schaffer, with the high calory diet in typhoid fever, have not only demonstrated that the nitrogen and weight loss is prevented to a large extent, but that the toxemia is reduced. Liberal feeding of lactose in dysentery has yielded similar results.' 1 For fuller discussion and bibliography see : Herter: Bacterial Infections of the Intestinal Tract, New York, 1907. Kendall: Bacteriology — General, Pathological and Intestinal, Philadelphia, 1916. CHAPTER XXII. THE COLON-TYPHOID GROUP OF BACILLI. There are a number of varieties of bacilli normally occupying the intestines of man and animals which, because they have similar charac- teristics and live in the colon, are generally grouped together as colon bacilli. Many of the varieties occurring in animals are culturally like those found in man. These bacilli are only pathogenic under unusual conditions. The specific pathogens, typhoid, paratyphoid, including the types responsible for meat poisoning, dysentery, and paradysentery bacilli, also have among themselves and between them and the colon bacilli resemblances and are often classed together in the group of the colon-typhoid bacilli. The chief characteristics common to this whole group are: (1) a similar morphology, i. e., short, rather plump non-spore-bearing rods with a tendency to thread formation; (2) a Gram -negative staining reaction; (3) similar growths on agar and gelatin; (4) non-liquefaction of gelatin (a few closely related organisms, such as B. clcacce, liquefy gelatin very slowly). In order to see more clearly the main points of difference among the subdivisions of this great group the following tabulations may be studied. GeOTJP or COLON-TYPHOID Bacilli. Colon group . Paratyphoid group (intermediate group) B. (coli) communis B. (coli) communior 1 B. (acidi) lactici B. aerogenes types B. paratyphosus, A B. paratyphosus, B B. paratyphosus, C B. enteritidis (Similar types, intermediates between colon and typhoid have been isolated from water and other sources, prob- ably non-pathogenic.) Normal inhabitants of the intes- tines, under certain conditions become pathogenic. Usually pathogenic for man or animals in varying degree. B. typhosus Pathogenic for man. Dysentery group B. dysenteriae (mannite not fermented). B. paradysenterise (man- nite fermented) . Three varieties. Pathogenic for man, paradysen- teriae usually less so than dysen- tery type. Alkaligenes group . B. alkaligenes. Occasionally pathogenic. 324 COLON-TYPHOID GROUP OF BACILLI \ Essential Dipfeebntial Reactions — Colon-typhoid Group. Acid and gas Glucose — acid and gas — Colon group. (colonies, red (Russell medium — slant, acid on Endo and butt, acid and gas.) Lactose Conradi) not fermented, (colonies, color- less on Endo, blue on Conradi) Glucose — acid and gas — Paratyphoid-enteritidis group. (Russell medium — slant, unchanged butt, acid and gas.) Glucose — acid only — B. typhosus — dysentery group. (Russell medium — slant, unchanged butt, acid only.) Glucose — not fermented — B. alkaligenes. (Russell medium — slant, unchanged butt, unchanged.) Among the non-lactose fermenters motility or its absence is of differ- ential value. The fermentative characteristics of value in differentiating members of the subgroups are given with each subgroup. Russell's Medium. — ^The reactions as given are based on the fact that the bacilli growing aerobically, that is, on the slant, only utilize the carbohydrate when present in amounts over 0.1 per cent., whereas growing anaerobically stabbed in the butt, they must utilize the carbo- hydrates for their oxygen supply and therefore ferment the trace of glucose present. By adding 1 per cent, saccharose intermediate types are differentiated. THE COLON GROXIP. The first description of an organism of the colon type was by Emmerich (1885), who obtained it from the intestinal discharges of cholera patients. A similar organism was found by Escherich (1886) in the feces of healthy infants. He gave it the name of Bacterium coli commune. It has since been demonstrated that closely allied types of bacilli are normal inhabitants of the intestines of most of the lower animals. They are transferred through the feces as manure and sewage to cultivated land, surface waters, etc. During warm weather they may multiply' outside of the animal body. Those strains having the chief cultural characteristics of the original strains are classed as colon bacilli, while those differing considerably from it are, while considered in the general group, given different names, such as paracolon, "atypical coli," etc. These types are especially common in water and their atypical reactions are often due to the suppression of certain characteristics due to unfavorable environment. The colon group is divided into four main subgroups, B. communis, B. communior, B. aerogenes, and B. (acidi) lactici. These subgroups have numerous varieties. The colon group has interest not only because it excites disease at times in man and animals, but also because it is an index of fecal pollu- tion from man or animals. If from man it indicates the possibility of THE COLON GROUP 325 infection with tlie typhoid or dysentery bacilli. For the significance of the colon group in water see Bacteriology of Water. B. (Coli) Communis. — This organism is taken as the type of the group in the following description. Only differentiating points are noted in the other varieties. Morphology. — B. comviunis varies in morphology. The typical form (Fig. 128) is that of short rods with rounded ends, from OAfi to 0.7/* in' diameter by 1/x to 3/i in length; sometimes, espe- cially when the culture media are not suitable for their growth and in tissues, the rods are so short as to be almost spherical, resembling micro- cocci in appearance, and, again, they are somewhat oval in form or are seen as threads of 6/i or more in length. The various forms may often be associated in the same culture. The bacilli occur as single cells or as pairs joined end-to-end, rarely as short chains. There is nothing in the morphology of this bacillus sufficiently characteristic for its identification. Fig. 128. — Colon bacilli. Twenty-four-hour agar culture. X 1100 diameters. Flagella. — Upon some varieties seven or eight peritrichic flagella have been demonstrated, the non-motile types show none. The flagella are shorter and more delicate than those characteristic of the typhoid bacilli. Staining. — The B. communis stains readily with the ordinary aniline colors; it is always decolorized by Gram's method. Under certain conditions the stained bacilli exhibit bipolar granules. Biology. — ^It is ah aerobic, facultative anaerobic, non-liquefying non- spore-bearing bacillus. It develops best at 37° C, but grows well at 20° C, and slowly at 10° C. It is usually motile, but the movements in some of the cultures are so sluggish that a positive opinion is. often diflBcult. In fresh cultures, frequently only one or two individuals show motility. Cultivation. — ^The B. communis develops well on all the usual culture media. Its growth on them is usually more abundant than that of the 326 COLON-TYPHOID GROUP OF BACILLI typhoid bacillus or the dysentery bacillus, but the difference is not sufficient for a differential diagnosis. Gelatin. — In gelatin plates, colonies are developed in eighteen to thirty-six hours. They resemble greatly the colonies of the typhoid bacillus, except that many of them are somewhat larger and more opaque. (See Figs. 43 to 45, page 115.) When located in the depths of the gelatin and examined by a low-power lens they are at first seen to be finely granular, almost homogeneous, and of a pale yellowish to brownish color; later they become larger, denser, darker, and more coarsely granular. In shape they may be round, oval, or whetstone-like. When the gelatin is not firm the margins of many colonies are broken by outgrowths, which are rather characteristic of colon bacilli. In stab cultures in gelatin the growth usually takes the form of a nail with a flattened head, the surface extension generally reaching out rapidly to the sides of the tube. Nutrient Agar. — In plate cultures: Surface colonies mostly circular, finely granular, and rather opaque. The deep colonies are likely to have protuberances. In streak cultures an abundant soft, white layer is quickly developed, but the growth is not characteristic. Bouillon. — In bouillon the B. communis produces diffuse clouding with sedimentation; in some cultiu-es a tendency to pellicle formation on the surface is occasionally seen. Potato. — On potato the growth is rapid and abundant, appearing after twenty-four to thirty-six hours in the incubator as a yellowish brown to dark cream-colored deposit covering the greater part of the siu-face. But there are considerable variations from the typical growth; there may be no visible growth at all, or it may be scanty and of a white color. These variations are often due to variations in the potato. Milk. — ^Milk is usually coagulated in from one to four days at 37° C. Coagulation is due principally to the production of lactic and acetic acid. Some strains produce acid but coagulation does not occur. It is possible that a lab ferment is partly responsible for the coagulation. Chemical Activities. — ^Behavior Toward Carbohydrates. — In cultures of B. communis many carbohydrates, especially sugars, become fermented with production of acid and gas. The important fermentation products, both qualitatively and quan- titatively, are produced from grape-sugar, probably according to the following reaction: 2C6H,206 + mo = 2C8H603 + CH3COOH + CjHsOH + 2CO2 + 2H2 Grape-sugar. Water. Lactic acid. Acetic acid. Ethyl alcohol. Carbonic Hydrogen. acid. There are reasons to think that lactic acid is first produced and that from this other acids and products develop. Under aerobic conditions lactic acid is produced in excess of acetic, while in the absence of oxygen the reverse is likely to be true. Gas Production. — ^When B. communis is grown in a solution of glucose (dextrose), CO2 and H2 are produced, in the proportion of ICO2 to IH2 CHEMICAL ACTIVITIES 327 up to ICO2 to 3H2. Anaerobic conditions aid gas formation. Very slight traces of gases other than H2 and CO2 are produced. The amount of gas varies in different varieties; the closed arm of the tube half-filled, and the H2 and CO2 in the proportion 2 to 1, is the characteristic type. The fermentation is not a simple hydrolytic action, but one in which combinations between the C and atoms are sundered and formed. This is not an oxidation process, but a change through breaking down — ^that is, a true decomposition. What oxidation takes place is chiefly due to the oxygen liberated from splitting the sugar molecules. The fermentative reactions of the main subgroups based on the work of Smith and of MacConkey are as follows:' B. (coli) communis — saccharose, negative — duloit, positive. B. (coli) communior — saccharose, positive — dulcit, positive. B. (acidi) lactici — saccharose, negative — dulcit, negative. B. aerogenes — saccharose, positive — dulcit, negative. Kligler found that a subdivision according to the fermentative reac- tions in salicin correllated more fully with other characteristics and suggests the following classification: B. communis — saccharose, negative — salicin, positive (dulcit usually positive). B. communior — saccharose, positive — salicin, negative (dulcit usually positive). B. lactici — saccharose, negative — salicin, negative (dulcit usually negative). B. aerogenes — saccharose, positive — salicin, positive (dulcit usually negative). The last group usually gives a positive Voges-Proskauer reaction (see Bacteriology of Water). Effect of B. Communis in Nitrogenous Compounds. — Indol Formation. — B. communis does not liquefy gelatin nor peptonize any albumins. They do, however, break down some of the higher nitrogenous compounds into smaller atom groups. The first noted of these compounds was indol, C6H4\pTT/>CH. This is one of the most important products of colon activity, although some varieties lack the ability to produce it. The maximum amount of indol is present about the tenth day. In the intestinal canal in health very little indol appears to be produced by the bacilli. Sulphurated hydrogen is liberated from sugar-free protein substances. Reduction Processes. — Nitrates are reduced to nitrites and from them ammonia and free nitrogen. Litmus and other dyes are also reduced. Toxins. — ^The bodies of dead bacilli contain pyogenic and other substances, which, injected into the' circulation, produce paralysis of the striped muscle fibers, convulsions, coma, and death. Extracts from some cultures produce irritation of the mucous membranes of the large intestines with dysenteric symptoms. 328 COLON-TYPHOID GROUP OF BACILLI Growth with Other Bacteria. — The B. communis as well as other members of the colon group act antagonistically to many of the pro- teolytic bacteria in the intestinal tract, and so inhibit alkaline putre- faction otherwise caused by the latter. In milk the same antagonism exists, probably because of the acidity caused by the colon growth. Reaction to High and Low Temperatures. — B. communis is killed at 60° C. in from five to fifteen minutes. Frozen in ice a large propor- tion die, but some resist for six months. Frozen in liquid air 95 per cent, are killed in two hours. Resistance to Drying and Antiseptics. — Simply drying destroys the majority of organisms dried at any one time, but some bacilli of the number dried may remain alive, especially when held in the texture of threads, for five or six months, or all may die in forty-eight hours. To most antiseptics they are moderately resistant. They are killed in five to fifteen minutes in a 1 per cent, solution of carbolic acid. Effect of Acids. — B. communis grows in a wider range of acids and alkalis than most other bacteria. It develops in from 0.2 to 0.4 per cent, of mineral acids, in from 0.3 to 0.45 per cent., of vegetable acids, and in from 0.1 to 0.2 per cent, of alkalis. Effect of Intestinal Juices. — ^Gastric juice kills unprotected B. com- munis unless it is too greatly diluted by food. All the members of the typhoid-colon group are more resistant to the gastric juices than most non-spore-bearing bacteria. With the food they readily pass from the stomach into the intestines. They grow in bile and in the intestinal juices. Pathogenesis. — ^In Lower Animals. — Intraperitoneal and intravenous inoculation of guinea-pigs and rabbits may produce death, which, when it follows, usually takes place within the first forty-eight hours, accom- panied by a decided fall of temperature, the symptoms of enteritis, diarrhea, etc., and finally fibrinopurulent peritonitis. Subcutaneous inoculation into rabbits is followed usually by abscess formation at the point of inoculation. Dogs and cats are similarly affected. Cystitis and pyelonephritis may be produced by direct mjections into the bladder and ureters, if the urine is artificially suppressed. Angiocholitis and abscess are produced by direct injections into the liver. Osteomyelitis may follow the intravenous injections of cultures in young rabbits. . From experiments on animals it would appear that the explanation of the pathogenesis of the colon bacillus is undoubtedly to be found in the toxic effects of the chemical substance and products of the cells. In Man. — In normal intestines with intact mucous membranes the toxic products formed hy B. coli are absorbed but little or not at all, and the bacilli themselves are prevented from invading the tissues by the epithelial layer and the bactericidal properties of the body fluids. Possibly there is an acquired immunity to the colon varieties which have long inhabited the intestines. BACILLUS COLI IN DIARRHEA 329 B. coli was at first regarded purely as a saprophyte. Later, because of the postmortem invasion and the great ease of growth of the colon bacillus on ordinary media, the other extreme of attributing too much to it was taken. The bacilli -present in the intestines may, either by an increase in virulence or by a lowered resistance in the person, cause intestinal inflammation or more distant infections. Thus in the case of ulcera- tion in typhoid fever B. coli may produce peritonitis. In conditions approaching death they at times pass through the intact mucous lining. The spread of bacilli from the intestines may also cause disease in the gall-bladder or urinary tract. The specific serum reaction in the body is a sign of infection, but great care has to be observed in deciding that it is present, as group agglutinins also occur. Up to the present time it is very difficult to state in any colon infection whether the bacilli were previously present in the intestines or were derived from outside sources through water, food, or direct contact with other cases. Intestinal Lesions. — ^The lesions present in intestinal inflammation attributed to B. coli are those of enteritis; the duodenum and jejunum are found to contain fluid, the spleen is somewhat enlarged, and there are marked hyperemia and ecchymosis of the small intestines, together with swelling of Peyer's patches. Virulence of B. Coli from Normal and Diseased Intestines.— The virulence varies with the culture and the time since its recovery from the intestines. Other things being equal, it is usually more virulent from an intestinal inflammation. From severe diarrhea the colon bacilli in 0.25 c.c. bouillon culture may kill guinea-pigs if given intraperitoneally, while from the healthy bowel 2.5 c.c. are usually required. B. Coli in Sepsis. — When lesions of the intestinal mucous membranes exist, or in colon cystitis, pyelitis, or cholecystitis, there is frequently just before death a terminal dissemination of the bacilli and consequent septicemia. The colon septicemia is detected by blood cultures. At times very few bacilli are found, and then the blood infection may be less important than the local one. Cases occurring in typhoid and cholera have been observed, especially in relapses in typhoid. In very young infants a malignant septicemia with tendency to hemorrhages may be due to B. coli. In a few cases in which B. coli but no typhoid bacilli were present the course of the disease has been similar to typhoid fever. An epidemic probably due to colon infection of water has been noted. Infections through food and water are usually caused by other closely allied bacilli not belonging to the colon group. B. Coli in Diarrhea. — ^In diarrhea we find increased peristalsis, less absorption of foodstuff, increased and changed intestinal secretions. Tissier observed that under treatment with cathartics the colon varieties increased, while the anaerobic forms are inhibited. In diarrhea, there- fore, we should expect favorable conditions for multiplication with the inhibiting causes lessened. This makes one question the importance of the significance of numerous epidemics which have been reported of acute diarrhea in children from one to five years of age in which almost 330 COLON-TYPHOID GROUP OF BACILLI pure cultures of colon bacilli have been found. The symptoms in such epidemics begin with high fever which often rapidly falls, and frequent stools, only watery_or containing mucus and streaks of blood. These symptoms may quickly abate or go on to a toxic state characterized by heart weakness and drowsiness. This may lead to lung complica- tions or death. B. CoU in Peritonitis. — Here the lesions must be considered as being due to mixed infection. Not only perforation of the intestines in man, but injury to the intestinal walls, allows colon infection of the peri- toneum to take place. At first most of these cases were believed to be a pm-e colon infection, but now it is known that this idea came largely from the overgrowth of colon bacilli in the cultures. More careful investigations, through cultures and smears, have demonstrated the fact that streptococci, and less frequently staphylococci and pneumo- cocci, are also usually present in peritonitis arising from intestinal sources. The colon . bacilli found even in the same case commonly comprise many varieties. B. Coli in Inflammation of the Bile Tract. — The normal healthy gall-bladder is usually sterile. This is true in spite of the fact that bile is apparently a good culture medium for the colon group. Ligation of the neck of the gall-bladder usually causes a colon infection to take place within twenty-four hours. Obstruction of the bile-duct through various causes is fairly common in man. The gall-bladder then becomes infected, and following the inflammation of the mucous membranes there is. often the formation of gall-stones. Some cases of jaundice are believed to be due to colon inflammation of the gall-ducts. Atypical varieties of B. coli are frequently isolated from gall-bladder infections. Inflammation of the Pancreas.^Welch was the first to record a case of pancreatitis with multiple fat necroses due to B. coli infection. A few more cases have since been reported due to members of the colon group, either alone or in conjunction with the pyogenic cocci. Inflammation of the Urinary Tract.— As far back as 1879 Bouchard noted cystitis due to bacilli of the colon group. When cystitis is estab- lished the bacterial infection frequently spreads to the pelvis of the kidneys, causing a pyelitis or suppurative nephritis. In most cases of chronic cystitis the ureters and pelves of the kidneys become involved; any malformation of the ureters aids the process. P'rom the pelvis the bacteria push up into the urinary tubules and excite inflam- mation and multiple abscesses. Colon infection of the different parts of the urinary tract may occur at any age, from infancy upward. Instead of starting in the bladder it may begin in the kidney itself, the colon bacilli coming from the blood or peritoneum. In many of these cases the bacilli isolated from the urine are agglutinated in high dilutions of the blood from the patient. Although other bacteria — the pyogenic cocci, the proteus, the typhoid .. baciUus, etc. — may excite cystitis, still in 90 per cent, of all cases some of the colon group are found, and this percentage is even higher in young children. The clinical picture of colon infection is very variable. The IMMUNITY 331 lightest cases progress under the guise of a bacteriuria. The urine is passed a little more frequently and shows a fine granular cloudiness. The reaction is acid. The cell elements are but little increased. There is an excess of mucus. Albumin is absent or present in only a trace. The condition may last for weeks or months and then spontaneously disappear or grow worse. With a somewhat more severe infection there is painful urination, perhaps tenesmus, increase of pus cells and slight fever. In a conical glass a sediment of pus cells forms at the bottom, and clear urine remains above. If the infection passes to the kidney colicky pain and tenderness over the region of the kidneys is usually present. The most important symptom of pyelitis is an irregular inter- mittent fever resembling malaria. The albumin is increased in the urine and red blood cells may be seen. If a general nephritis arises the symptoms are all intensified and an anemic condition may develop. Septicemia may finally result. In most of these cases the microscopic examination is sufficient to make a probable diagnosis, since the bacteria are so abundant. The variety of colon bacillus present can, of course, only be told by cultures and other means. In the urine they appear as diplobacilli, or partly in short, almost coccus, forms, partly in long threads. As a rule motility is absent. Not infrequently the c'ultures appear to be identical with those of the Bacillus aerogenes. The characteristics of the urine itself have much to do with the probability of infection; the more acid urines being less likely to afford a proper soil for growth. Some urines are bactericidal even when they are neutral. The substances producing this condition are not known. The colon bacilli in the urine produce no appreciable effect on the reac- tion, but give up some of their toxins, which upon absorption cause the deleterious local and general effects. The serum of the patient usually agglutinates the cultures from the urine in 1 to 20 or 1 to 50 dilutions, but this property is sometimes absent, especially in light cases. In all cases in addition to the introduction of the colon bacillus a predisposing condition must be present, such as more or less marked retention of urine by an enlarged prostate or stricture, any unhealthy state of the mucous membrane or general depression of vitality. B. Coli as Pus Former. — ^Members of this group are frequently the cause of abscesses in the region of the rectum, urethra, and kidney. They rarely produce pus in other locations. B. Coli in Inflammation Not Previously Mentioned. — Broncho- pneumonia, lobar pneumonia, and pleurisy have occasionally been caused by colon bacilli, probably from blood sources. Not a few cases of meningitis and spinal meningitis in infants, follow localized B. coli infections. The symptoms are not well developed as a rule. Some cases of endocarditis and of conjunctivitis have also been noted. Immunity. — Natural infection in man, or inoculation into animals, is followed by the production of antibodies; agglutinins, precipitins, bactericidal substances, and opsonins being produced. The attempts to separate the members of this group by agglutination have shown the 332 COLON-TYPHOID GROUP OF BACILLI great dissimilarity of the different members in their immune reactions, although group reactions are marked. Natural agglutinins for this group are commonly present in the serum of man and animals. (See Agglutination.) Vaccine Therapy. — See Part III. Methods of Isolation. — They may be isolated from lesions on ordi- nary media, in bile media or on Endo or Conradi plates. The latter are used in isolation from feces or other mixed material. Blood cultures may be made in bile or broth. For the examination of water for members of the colon group, see Bacteriology of Water. Bacillus Aerogenes and Allied Encapsulated Bacilli. — ^The members of this group differ from the members of the colon group, already described, in producing a viscid or mucoid growth, and in smears a capsule is commonly demonstrable. For this reason the group is spoken of by some as the capsulatus group, and the aerogenes types described as allied varieties. There are wide variations in the group, not only in the degree of capsule production but in the ability to ferment different carbohydrates. In the absence of an accepted biological classi- fication of types, the varieties are best considered according to their source. For this reason the term B. aerogenes is collectively used for the types normally found in feces, milk, etc. B. Aerogenes. — ^Normally present in feces and also found in sewage and water. It is constantly present in milk and is one of the chief causes of the souring of milk and cream. Morphologically it is variable in length and capsule production. In cultures the growth is abundant and viscid or mucoid in variable degree. It usually ferments the various carbohydrates more vigorously than other colon types, the closed arm of the fermentation tube being usually full or nearly full of gas. Indol production is variable. B. aerogenes is much more resistant to acids and other deleterious substances than the other colon types. Pathogenesis. — ^Probably slight, although it has been isolated from infections of the urinary tract, peritoneum and liver and gall-bladder. In some of the cases reported there is a possibility that the infection was really due to other of the colon types. In the absence of reliable methods of differentiation there is always a question whether the more serious infections are not due to the more virulent bacillus of Friedlander. Bacillus Fneumonise, Friedlander (B. Mucosas Capsulatus). — ^This bacillus was first described by Friedlander in 1882. He confused this bacillus with the pneumococcus, then undescribed, and regarded it as the causa- tive agent in lobar pneimionia. Morfhology. — It varies from coccoid forms to longer bacilli; they may be single, in pairs, or in short chains. They are surrounded by a wide capsule which is easily demonstrable not only in the material directly from the lesions but also in smears from cultures. Biology. — It grows freely on ordinary media. On agar the colonies are characteristically mucoid, with a tendency to confluence. Indol is not produced. BACILLUS ASrOGENES AND ALLIED ENCAPSULATED BACILLI 333 Pathogenesis. — It is pathogenic for mice and guinea-pigs in variable degree. Injection intraperitoneally or into the lung is followed by peri- tonitis or local hepatization with septicemia. Rabbits are killed by some strains after intraperitoneal or intravenous injection ; subcutaneous injection, however, commonly causes only a local lesion. In man it causes both broncho- and lobar pneumonia. This type of pneumonia is relatively infrequent and is characterized by its very high mortality. The bacilli are present in the sputum, as a rule, in almost pure culture. " It is commonly present in the upper respiratory tract and the accessory sinuses, and may cause inflammation. By extension, infection of the ear occiu-s, which may be complicated by meningitis. The pleura or the pericardium may be infected as a complication of pneumonia. A septicemia may develop as a complication of pneumonia or infection elsewhere. Inflammations of the eye are occasionally due to this organism. B. Ozena. — This organism is culturally indistinguishable from the preceding. It receives its name from its constant presence in ozena or fetid rhinitis. Abel and others consider it the causative agent in this disease, basing their opinion on its constant presence. Bacillus of Rhinoscleroma. — ^This bacillus differs from the Bacillus pneumonioe only in its weaker fermentative -properties. Its name comes from its frequent presence in rhinoscleroma. Its etiological relationship to this disease is not established. (See Fig. 13.) Immunity and Serum Reactions of the Capsulatus Types. — Active immunity cannot, as a rule, be produced in animals by the injection of killed cultures or their products. Careful injections of animals, however, lead to the production of agglutinins, precipitins, and complement- fixing antibodies. The attempt to differentiate the various types by ■ means of serum reactions has as yet led to no uniform results. In common with other encapsulated bacteria they are not easily affected by sera. Forges found that the absence of agglutination could be attributed to the presence of the capsule and devised a method, of destroying the capsule by first treating the bacterial suspension with acids and heat and then neutralizing with alkali. Vaccine Therapy. — See Fart III. Other Bacilli Allied to the Colon Group. — ^There are various types, of which B. cloacce is an example, which resemble the colon bacillus in many respects but liquefy gelatin in varying degree, some very slightly and only after prolonged growth. REFERENCES. Kligler: Jour. Inf. Dis., 1914, xv, 187. MacConkey: Jour. Hyg., 1905, v, 333. Smith: Centralbl. f. Bakt., 1895, xviii, 494. CHAPTER XXIII. THE TYPHOID BACILLUS. This organism was first observed by Eberth, and independently by Koch, in 1880, in the spleen and diseased areas of the intestine in typhoid cadavers, but was not obtained in pure culture or its principal biological features described until the research,es of Gaffky in 1884. The absolute identification of the bacillus only became possible with the increase of our knowledge concerning the specific immune substances. Its etiological relationship to typhoid fever has been particularly difiicult to demonstrate, for although pathogenic for many animals when subcutaneously or intravenously inoculated, it was impossible to produce infection in the natural way. More recently infection of the higher apes has been produced. Nevertheless, the specific reactions of the blood serum of typhoid patients, the constant presence of the* Bacillus typhosus in the intestines and some of the organs of the typhoid cadavers, the very frequent isolation of this bacillus from the roseola, spleen, blood, and excretions of the sick during life, the absence of the bacilli in healthy persons, unless they have at some time been directly exposed to, or are convalescent from, typhoid infection, all these have demonstrated scientifically that this bacillus is the chief etiological factor in the production of the great majority of cases designated as typhoid fever. Morphology and Staining. — Typhoid bacilli are short, rather plump' rods of about \ii to Sju in length by 0.5/i to 0.8/i in diameter, having rounded ends, and often growing into long threads. They are longer and somewhat more slender in form than most of the members of the colon group of bacilli (Figs. 129 and 130). The typhoid bacilli stain with the ordinary aniline colors, but a little less intensely than do most other bacteria. Like the bacilli of the colon and paratyphoid groups, they are decolorized by Gram's method. Bipolar staining is sometimes marked. Biology. — ^The typhoid bacillus is a motile, aerobic, facultative anaerobic, non-liquefying bacillus. It develops best at 37° C; above 40° and below 30° growth is retarded; at 20° it is still moderate; below 10° it almost ceases. It grows slightly more abundantly in the presence of oxygen. It does not form spores. Resistance. — ^When a number of typhoid bacilli are dried most of them die within a few hours, the remainder gradually dying during the next few weeks. A few frequently remain alive for months. In their resistance to heat and cold they behave like the average non- spore-bearing bacilli. With rare exceptions they are killed by heating to 60° C. for one minute. CULTIVATION 335 ;,; Motility. — ^Typhoid bacilli, when living under favorable conditions, \are very actively motile, the smaller ones having often an undulating motion, while the larger rods move about rapidly. In different cultures, however, the degree of motility varies. Fig. 129. — Typhoid bacilli from nutrient agar. X 1100 diameters. Fig. l.SO. — Typhoid bacilli from nutrient gelatin. X 1 100 diameters. Flagella. — These are often numerous and spring from the sides as well as the ends of the bacilli, but many short rods have but a single terminal flagellum (Figs. 131 and 132; see also Plate III). Cultivation. — ^Its growth on most sugar-free culture media is quite similar to that of the Bacillus coli, but it is somewhat slower and not quite so luxuriant. Fig. 1.31. — Flagella, heavily stained, attached to bacilli. (Van Ermengen's method.) i' Fig. 132. — Typhoid bacillus with faintly stained flagella. (Loffler's method.) Growth on Gelatin Plates (Fig. 133). — The colonies growing deep down in this plate medium have nothing in their appearance to dis- tinguish them from submerged, colonies of the colon group; they are finely granular, with a sharp margin and a yellowish-brown color. The superficial colonies, however, particularly when young, are often quite characteristic; they are transparent, bluish white in color, with an irreg- ular outline, not unlike a grape leaf in shape. Slightly magnified they 336 TYPHOID BACILLUS appear homogeneous in structure, but marked by a delicate network of furrows. Surface colonies from some varieties" of colon bacilli give a similar picture. In stick cultures in gelatin the growth is mostly on the surface, appear- ing as a thin, scalloped extension, which gradually reaches out to the sides of the tube. In the track of the needle there is but a limited growth, which may be granular or uniform in structure, and of a yellowish-brown color. Growth in Bouillon. — ^This medium is uniformly clouded by the typhoid bacillus, but the clouding is not so intense as with the colon bacillus. When the bouillon is somewhat alkaline a delicate pellicle is sometimes formed on the surface after eighteen to twenty-four hours' growth. Growth on Agar. — ^The streak cultures on. agar are not distinc- tive; a transparent, filiform, gray- ish streak is formed. Growth on Potato.— The growth on this medium was formerly of great importance in identification, but now other media, giving more specific characteristics, have been discovered. When characteristic, the growth is almost invisible but luxuriant, usually covering the surface of the medium. Again, the growth may be quite heavy and colored yellowish brown, and with a greenish halo, when it is very similar to the growth of the colon bacillus. These differences of growth 'on potato appear to be chiefly due to variations in the substance of the potato, especially in its reaction. For the characteristic growth the potato should be slightly acid. A new lot of potato should always be tested with a typical typhoid bacillus as a control. Indol Reaction.— It does not produce even a trace of indol in peptone- water solution. This test was proposed by Kitasato for differentiating the typhoid bacillus from other similar bacilli such as those of the colon group, which, as a rule, give the indol reaction. The typhoid bacillus, like the colon bacillus, produces alkaline substances from peptone. Neutral Red. — In stick cultures in glucose agar or glucose broth cultures with neutral red as indicator the typhoid bacillus produces no change, while the colon bacillus decolorizes the medium and produces gas. Effect of Inhibiting Substances in Culture Fluids. — ^The typhoid bacillus is inhibited by weaker solutions of formaldehyde, carbolic acid, and Fig. 133. — A superficial colony and i deep cdlony of typhoid bacilli in gelatin X 20 diameters. DISTRIBUTION OF BACILLI IN HUMAN SUBJECT 337 other disinfectants than is the colon bacillus. Most typhoid-like bacilli resemble the typhoid bacillus in this respect. Some substances, such as brilliant green, inhibit the colon bacillus more. Action on Different Sugars. — The essential fermentative differences between the typhoid bacillus and paratyphoid colon group are given on p. 324. Acid is produced in mannit, maltose and xylose media. Dulcit is acted upon, as a rule, aft^r prolonged incubation. While the typhoid bacillus does not produce gas from dextrose, galactose, and levulose, it does produce acid from these substances. Milk. — ^The typhoid bacillus does not cause coagulation when grown in milk. A very slight acidity is evidenced in litmus milk from its action on traces of fermentable substances other than lactose. Production of Disease in Animals. — ^It is impossible, with the exception of the anthropoid apes, to produce a disease like typhoid fever in animals. Metchnikoff and Besredka fed apes with typhoid bacilli by adding fecal material to their food. After eight days' incubation, fever, diarrhea and invasion of the blood stream by the typhoid bacilli, developed. The general clinical picture resembled that of typhoid fever. Some of the animals died. Injection of other animals may produce sickness due to the toxemia produced by the substances in the bodies of the bacilli injected. Typhoid bacilli, freshly obtained from typhoid cases and introduced subcutaneously in animals, rapidly die. In the peritoneal cavity they may increase, causing a fatal peritonitis with toxic poison- ing. By accustoming bacilli to the animal body a certain degree of increased virulence for the animal can be obtained, so that smaller amounts of culture may prove fatal. In rabbits a septicemia can be produced by intravenous inoculation. Localization in the gall-bladder follows in some and a "carrier" state similar to that found in man results. Direct injection into the gall-bladder also results in the develop- ment of the carrier state. Distribution of Bacilli in the Human Subject. Toxic Effects. — Typhoid fever during its early stages, at least, is accompanied by a bacteriemia. The bacilli thus pass to all parts of the body and become localized in certain tissues, such as the bone marrow, lymphatic tissues and spleen, liver and kidneys. Wherever found in the tissues the typhoid bacilli are usually observed to be arranged in groups or foci; only occasionally are they found singly. These foci are formed during life, as is proved by the degenerative changes often seen about them; but it is possible that the bacilli may also multiply somewhat after death. Important Primary Characteristic Lesions in Man. — ^The lesions of the intestines which are most pronounced in the lower part of the ileum consist of an inflammatory enlargement of the solitary and agminated lymph nodules. Necrosis with ulceration frequently follows the hyperplasia in the more severe cases. In the severest cases the ulceration and sloughing may involve the muscular and peritoneal coats and perforation may occur. Peritonitis and death usually follow. In rare cases the perforation is closed by adhesions. The minute changes are a hyperplasia of normal elements .of the lymphatic tissue, namely, the lymph cells and the endothelium of the trabeculse and sinuses. In severer forms necrotic changes are apt to intervene. These 22 338 TYPHOID BACILLUS changes are attributed to the toxic substances formed by the typhoid baciUi, but may be directly brought about by the occlusion of the nutritive blood- vessels, as pointed out by Mallory. . The mesenteric lymph nodes undergo changes similar to those in the ileum. The spleen is enlarged because of congestion and hyperplasia. The liver and, to a less extent, the kidneys are apt to show foci of cell proliferation. In typhoid fever, as in other infectious diseases, toxic poisoning may be mani- fested by disturbances in the circulatory, respiratory, and heat-regulating mechanism as weU as by manifest lesions. In a few cases the intestinal lesions are absent. Some of the inflammatory compUcations which occur in typhoid fever are due to the growth of the bacillus in excessive numbers in unusual places in the body; but many of them are due to a secondary infection with other bacteria, especially the pyogenic cocci and baciUi of the colon group. Unusual Location of Typhoid Lesions Occurring as Complications of Typhoid Fever. — Cases of sacculated and general peritonitis, abscess of the liver and spleen, subphrenic abscess, osteomyelitis, periostitis, and inflammatory processes of other kinds have been reported as being due to the typhoid bacillus. In certain cases of typhoid pneumonia, serous pleurisy, empyema, and inflammations of the brain and spinal cord or their membranes, typhoid bacilli exclusively have occurred. The inflammation produced may or may not be accompanied by the formation of pus. There are indeed a number of cases now on record in which the typhoid bacillus has played the part of pm producer. The Importance of Mixed Infection. — Frequently when complications occur in typhoid fever they are due to secondary or mixed infection with the staphylococcus, pneumococcus, streptococcus, pyocyaneus, and colon bacillus. Frequently these bacteria are found side by side with typhoid bacilli; in such cases it is difiBcult to say which was the primary and which was the secondary infection. Primary Infection of Liver and Gall-bladder. — ^We find an appreciable number of chronic typhoid carriers with no history of having had typhoid. Although the symptoms may have been so slight as to have been over- looked, it is probable that bacilli may escape from the intestine into the portal circulation and thus infect the liver and gall-bladder (see Normal Carriers). Elimination of Typhoid Bacilli from Body. — ^Not infrequently typhoid bacilli are found in the secretions. They are present in the urine in about 25 to 50 per cent, of the cases during some part of the disease. Slight pathological lesions in the kidneys almost always occur in typhoid fever, but severe lesions also sometimes occur. In some cases the urine has very many typhoid bacilli. In cases of pneumonia due to the typhoid bacillus it is abundantly present in the sputum, and care should be taken to disinfect the expec- toration of typhoid patients. In typhoid fever the bacilli are almost always present in the gall-bladder. The bacilli are usually eliminated by the feces, being derived from the ulcerated portions of the intestinesf their growth within the intestinal contents is, with few exceptions, not extensive. They may be excreted, however, very early in the , disease, and have been found in the stools even during the incubation period. ELIMINATION OF TYPHOID BACILLI FROM BODY 339 Not only the great majority of cases examined bacteriologically and pathologically, but the epidemiological history of the disease, proves that the chief mode of invasion of the typhoid bacillus is by way of the mouth and stomach. The infective material is discharged principally by means of the excretions and secretions of the sick — namely, by the feces, the urine, and occasionally by the sputum. Typhoid Carriers. — ^The bacilli usually disappear from the body during the first week or two of convalescence. About 1 to 5 per cent, continue to excrete typhoid bacilli for years, perhaps for life. Petruschky, in 1898, reported that typhoid bacilli sometimes remained in the urine of typhoid convalescents for months. Gushing soon after observed a case who had had typhoid fever five years before. In 1902 Frosch, and a little later Conradi and Diigalski, reported persons who passed typhoid-infected feces months after recovery from typhoid fever. Some bacilli carriers did not know either that they had had typhoid fever or been in contact with it, and others knew only that they had been in contact with it. In Washington, during the typhoid season, a series of one thousand examinations showed that 0.3 per cent, of persons with no known exposiue, and who had not had typhoid were excreting typhoid bacilli. These cases may be called "healthy or normal carriers" in contradistinc- tion to "convalescent carriers." Lentz, in 1905, found out of a large number of examinations that about 4 per cent, of persons convalescent from typhoid fever were typhoid carriers. In our laboratory we have found six in one hundred and forty institution convalescents. The focus of infection is either the gall-bladder or bile ducts. The majority are women. A remarkable case of a cook ("Typhoid Mary"), discovered by Soper, was under our care for three years. A visitor of the family In which this woman was cook developed typhoid fever some ten days after entering the household. This was in 1901. The cook had been with the family three years and it is difficult to judge which infected the other. The cook went to another family. One month later the laundress in this family was taken ill. In 1902 the cook obtained a new place. Two weeks after arrival the laundress was taken ill with typhoid fever; in a week a second case developed and soon seven members of the household were sick. In 1904 the cook went to a home in Long Island. There were 4 in the family as well as 7 servants. Within three weeks after arrival 4 servants were attacked. In 1906 the cook went to another family. Between August 27 and September 3, 6 out of its 11 inmates were attacked with typhoid. At this time the cook was first suspected. She entered another family on September 21. On October 5 the laundress developed typhoid fever. In 1907 she , entered a family in New York City, and two months after her arrival 2 cases developed, 1 of which proved fatal. The cook was removed to the hospital March 19, 1907. Cultures taken every few days showed bacilli off and on for three years. Some- times the stools contained enormous numbers of typhoid bacilli and 340 TYPHOID BACILLUS again for days none would be found. She was released on parole in 1910, promising to report to the Health Department and not to engage in cooking. She broke her parole and disappeared. In 1915 in an epidemic of typhoid at a maternity hospital, a total of 25 cases developed. Investigation showed that food infection was the cause and the cook was identified as " Typhoid Mary." During the period of disappearance she infected a friend and was the cause of several cases in a small private sanatorium. She is known to have been the cause of at least 50 cases of typhoid fever. We recently traced some hundreds of cases of typhoid fever to a milk supply produced at a farm, looked after by a typhoid carrier who had typhoid fever forty-seven years ago. Treatment of Typhoid Carriers. — Medicinal treatment or surgery seems so far to have yielded only slight results. Urotropin in very large amounts is reported to have cured one case, in which operation alone had failed. Removal of the gall-bladder cannot be relied upon, as the bacilli may be present in the bile ducts. Meader reports some success with long-continued treatment by injections of killed cultures. Duration of Life in Water, Feces, Oysters, etc. — ^It is of importance to know for what length of time the typhoid bacillus is capable of living outside of the body; but, unfortunately, owing to the great difficulties in proving the presence of this organism in natm-al conditions, our knowledge on this point is still incomplete. In feces, in privies, on the ground, etc., the length of life of the typhoid bacilli is very variable, depending on the composition of the feces and the soil, and on the varieties of bacteria present; sometimes they live but a few hours, usually a day, exceptionally for very long periods. Thus, according to Levy and Kayser, in winter typhoid bacilli may remain alive in feces for five months. Foote says that they can be found in living oysters for a month at a time, but in numerous experiments we have not been able to find them after five days. Their life in privies and in water is usually short, often not over forty-eight hours and usually not over a week. The less the general contamination of the water, the longer the bacilli are apt to live. The life of the typhoid bacillus varies according to the abundance and varieties of the bacteria associated with it, and according to the presence or absence of such injurious influences as deleterious chemicals, high temperature, light, etc., to which it is known to be sensitive. In ice typhoid bacilli rapidly die, none probably ever live as long as six months (see p. 348). Methods of Communication. — ^The bacilU may reach the mouth by means of infected fingers or articles of various kinds, or by the inges- tion of infected food, milk, water, etc., or more obscure ways, such as the eating of raw oysters and clams or the contamination of food by files. Of great importance, though gradually lessening, however, is the production of infection by contaminated drinking water or milk. In a very large number of cases indirect proof of this mode of infection has been afforded by finding that the water had been contaminated with urine or feces from a case of typhoid. In a very few instances the proof has been direct — namely, by finding typhoid bacilli in the water. METhOM of communication S4l Examples of infection from water and milk have frequently come under our direct observation. The following instances may be cited: A large force of workmen obtained their drinking water from a well near where they were working. Typhoid fever broke out and continued to spread until the well was filled up. Investigation showed that some of the sick, in the early stages of their disease, repeatedly infected the soil with their urine and feces at a point which drained into the well. Another example occurred in which typhoid fever broke out along the course of a creek after a spring freshet. It was found that, far up near the source of the creek, typhoid feces had been thrown on one of its banks and had then been washed into the stream. The epidemic at Scranton, Pa., during the winter of 1907 was most interesting. A little over 1 per cent, of the inhabitants were attacked. No pollution of the water with typhoid feces or urine could be discovered, although typhoid bacilli were isolated from the water of a small inter- cepting reservoir by Dr. Fox. This was only accomplished by using large quantities of water. The bacillus isolated was identical by all known tests with the typhoid cultures from cases of typhoid fever. A stream entering the reservoir was exposed to pollution by men working on a nearby railroad. In water-borne outbreaks of typhoid fever it is not infrequent to have outbreaks of diarrhea preceding or accompanying the development of the typhoid fever. The number of cases of diarrhea may be many times greater than the number of typhoid cases. A sudden outbreak of diarrhea therefore may mean contamination of a water supply and should be considered as possible indication of an oncoming typhoid outbreak. An instance of milk infection secondary to water infection occurred in the case of a milk dealer whose son came home suffering from typhoid fever. The feces were thrown into a small stream which ran into a pond in which the milk cans were washed. A very alarming epidemic of typhoid developed, which was confined to the houses and asylums supplied with this milk. Milk-borne epidemics are most commonly due to infection of the milk by carriers. During the Spanish- American War not only water infection, but food infection was noticed, as in the case of a regiment in which certain companies were badly infected, while others nearly escaped. Each company had its separate kitchen and food supply, and much of the infection could be traced to the food, the contamination coming partly through the flies. A small number of cases have been traced to oysters. Contact infection is responsible for many cases. Even during outbreaks due to various causes, contact infection plays a large role in the development of secondary cases. These cases develop either by contact with known typhoid cases or through convalescent or healthy carriers in the households. The early undiag- nosed cases may have typhoid bacilli in their stools and act as a source of contagion; likewise short, mild cases develop which are never diag- nosed. For these reasons "typhoid precautions" should not await positive diagnosis, but should be instituted on the slightest suspicion. 342 TYPHOID BACILLUS Individual Susceptibility. — ^In this, as in all infectious diseases, individual susceptibility plays an important role in the production of infection. Where exposure is associated with the ingestion of few bacilli, the majority of persons escape infection. In many individuals there probably is some disturbance of the digestion, excesses in drinking, etc., or a general weakening of the power of resistance of the individual, caused by bad food, exposure to heat, overexertion, etc., as occurs with soldiers and prisoners, for example, to bring about the conditions suitable for the production of typhoid fever, when the dose of bacilli is small. Ingestion of larger numbers of bacilli and especially with repeated exposure, 50 per cent, or more of the exposed succumb to infection. Immunization. — ^After recovery from typhoid fever a considerable immunity is present which lasts for years. This is not absolute, as about 2 per cent, of those having typhoid fever have a second attack, which is usually a mild one. Specific immunization against experi- mental typhoid infection has been produced in animals by the usual method of injecting at first small quantities of the living or dead typhoid bacilli and gradually increasing the dose. The blood serum of animals thus immunized has been found to be highly bactericidal and to possess some protective power against toxic extracts of the bacilli. It is also rich in agglutinins, precipitins, and opsonins. These characteristics have also been observed in the blood serum of persons who are con- valescent from typhoid fever. The attempt has been made to employ the typhoid serum for the cure of typhoid fever in man, but, although a number of individual observers, have reported good results with one or another of the serums, most consider that little or no good is derived from serum. Vaccine and Serum. — See Part III. Diagnosis by Means of the Widal or Agglutination Reaction. — ^The chief practical application of our knowledge of the specific substances developed in the blood of persons sick with typhoid fever has been as an aid to diagnosis. Gruber-Widal Test. — ^The first application of the use of serum for the early diagnosis of typhoid fever on an extensive scale was made by Widal, and reported with great fulness and detail in a communication published in June, 1896. For history and theory of test see p. 179. Since then the serum test for the diagnosis of typhoid fever has come into general use in bacteriological laboratories in all parts of the world, and though the extravagant expectations raised at the time when Widal first announced his method of applying this test have not been entirely fulfilled, it has, nevertheless, proved to be of great assistance in the diagnosis of obscure cases of the disease, and is now one of the recognized tests for the differentiation of the typhoid bacillus. It should also be mentioned that to Wyatt Johnson, of Montreal, belongs the credit of introducing its use into municipal laboratories, suggesting that dried blood should be employed in place of blood serum. Use of Dried Blood. — Directions for Preparing Specimens of Blood. — ^The skin covering the tip of the finger or the ear is thoroughly cleansed, DIAGNOSIS BY WIDAL OR AGGLUTINATION REACTION 343 and is then pricked with a needle deeply enough to cause several drops of blood to exude. Two fair-sized drops are then placed on a glass slide, one near either end, and allowed to dry. Glazed paper may also be employed, but it is not as good, for the blood soaks more or less into it, and later, when it is dissolved, some of the paper fiber is apt to be rubbed off with it. The slide is placed in a box for protection. Preparation of Specimen of Blood for Examination. — In preparing the specimens for examination, the dried blood, if accuracy is desired, is first weighed and then brought into solution by adding to it the quantity of normal salt solution to make the desired dilution, remembering of course to allow for the loss of the water in the blood through drying. The loss of water is equivalent to about 80 per cent, of the weight of the blood. A drop of each dilution is placed on individual cover-slips and to each is added a drop of an eighteen-hour broth culture of typhoid. These are then mounted on hollow slides in the ordinary way. (See p. 71.) The drops, after being mixed, have in a 1 to 10 dilution a distinct reddish color and in 1 to 20 a faint pink tinge. The cover-glass with the mixture on the surface is inverted over a hollow slide (the edges about the concavity having been careftdly smeared with vaselin, so as to make a closed chamber). Ordinarily the dried blood is not weighed, but the measure of dilution is estimated by the color of the drop. To judge this the beginner must carefully make dilutions of fluid blood and notice the depth of color in 1 to 10 and 1 to 20 dilutions. Besides the faulty judgment of the dilution color by the examiner, the variation in depth of color of different specimens of blood makes the estimation of dilutions more or less inaccurate, but fortunately this does not greatly interfere with the value of the test. For use of serum or fluid blood see Agglutination for methods, p. 202. Advantages and Disadvantages of Serum, Dried Blood, and Fluid Blood for the Serum Test. — ^The dried blood is easily and quickly obtained, and does not deteriorate or become contaminated by bacterial growth. It is readily transported, and seems to be of nearly equal strength as the serum in its agglutinating properties. It must in use, however, be diluted with at least five times its original bulk with water, otherwise it is too viscid to be properly employed. The amount of dilution can be determined roughly by the color of the resulting mixture. Serum, on the other hand, can be used in any dilution desired, from a mixture which contains equal parts of serum and broth culture to any dilution desired; and this can be exactly measured by a graduated pipette, or, roughly, by a measured platinum loop. The disadvantages in the use of serum are entirely due to the slight difficulty in collecting and transporting it. If the serum is obtained from blood after clotting has occiu"red a greater quantity of blood must be drawn than is necessary when the dried-blood method is used. For scientific investigations and for accurate results, particularly in obscure cases, the use of serum is to be preferred to dried blood. Practically, however, the results are nearly as good for diagnostic purposes from the dried blood as from the serum; 344 TYPHOID BACILLUS The Culture to be Employed. — It is important that the culture employed for serum tests should be a suitable one, for although all cultures show the reaction, yet some respond much better and in higher dilutions than others. Cultures freshly obtained from typhoid cases are not as sensitive as those grown for some time on nutrient media. Those kept for a long time on artificial media sometimes show a decided tendency to spontaneous agglutination. Artificial cultivation is usually accom- panied by increase of capacity for agglutination. At present a strain known as Mt. Sinai is used at this laboratory. A broth culture of the typhoid bacillus developed at 25° to 35° C, not over twenty-four hours old, in which the bacilli are separated and actively motile, has been found to give us the most satisfactory results. If the broth culture is heavy it should be diluted. Cultures grown at temperatures over 38° C. are not apt to agglutinate so well as those grown at lower temperatures. Dilution of the Blood Serum to be Employed and Time Required for the Development of Reaction. — ^The serum test, as has been pointed out, is quantitative and not qualitative, that is, the diagnostic reaction must be in dilutions higher than occur because of the presence of normal or group agglutinins. (See p. 204.) It is most important to remember that it is purely a matter of experience to determine in any type of infection what agglutinating strength of a serum is of diagnostic value. The results obtained in the Health Department Laboratories, as well as elsewhere, have shown that in a certain proportion of cases not typhoid fever a slow reaction occurs in a 1 to 10 dilution of serum or blood; but very rarely does a complete reaction occur in this dilution within fifteen minutes. When dried blood is used the slight tendency of non-typhoid blood in 1 to 10 dilution to produce agglutination is increased by the presence of the fibrinous clumps, and perhaps by other substances derived from the disintegrated blood cells. From maiiy cases examined it has been found that in dilutions of 1 to 20' a quick reaction is almost never produced in any febrile disease other than that due to typhoid or paratyphoid bacillus infection. In typhoid fever such a distinct reaction often occurs with dilutions of 1 to 100 or more. It is possible that some cases of paratyphoid infection give a prompt reaction in 1 to 20 dilutions, but if this is so, it is not a serious drawback. Typhoid-bacillus carriers with the exception of transient contact or normal carriers, commonly have specific agglutinins in their blood whether they previously have had typhoid fever or not. The mode of procedure as now employed is as follows: Withserimi, one part of a 1 to 10 dilution is added to one of the bouillon cultures. With dried blood, as stated above, a solution of the blood is first made according to a known color, and the dilution is made from this. If there is no reaction — that is to say, if within five minutes no marked change is noted in the motility of the bacilli, and no clumping occurs — nothing more is needed; the result is negative. If marked clumping and immobilization of the bacilli immediately begin and become com- plete within five minutes, this is termed a marked immediate typhoid reaction, and no further test is considered necessary, though it is always advisable to confirm the reaction with higher dilutions up to 50 or DIAGNOSIS BY WIDAL OR AOGLUTINATION REACTION 345 more, so as to measure the exact strength of the reaction. If in the 1 to 20 dilution a complete reaction takes place within thirty minutes, the blood is considered to have come from a case of typhoid infection, while if a less complete reaction occurs it is considered that only a proba- bility of typhoid infection has been established. By many the time allowed for the development of the reaction with the high dilutions is from one to two hours, but to us twenty minutes with the comparatively low dilution of 1 to 20 seems safer and more convenient. Positive results obtained in this way may be considered conclusive, unless there be grounds for suspecting that the reaction may be due to a previous fairly recent attack or to recent vaccination. The failure of the reaction in one examination by no means excludes the presence of typhoid infection. If the case clinically remains doubtful, the examination should be repeated every few days. Use of Dead Cultures. — Properly killed typhoid bacilli respond well to the agglutination test. For the physician at his office the dead bacilli offer many advantages. The reaction is slower than with the living cultures and is observed either macroscopically or microscopically. A number of firms now supply outfits for the serum test. These outfits consist of a number of small tubes containing an emulsion of dead typhoid bacilli. Directions accompany the outfit. This method has been found of great value in comparative estimation of agglutinin by Dreyer and others. Proportion of Cases of Typhoid Fever in which a Definite Reaction Occurs, and the Time of its Appearance. — ^As the result of a large number of cases examined in the Health Department Laboratories, it was found that about 20 per cent, give positive results in the first week, 60 per cent, in the second week, about 80 per cent, in the third week, about 90 per cent, in the fourth week, and about 75 per cent, in the second month of the disease. In 98 per cent, of the cases in which repeated examinations were made (hospital cases) a definite typhoid reaction was present at some time during the illness. Persistence of the Reaction after Convalescence or after Immunization. — A definite typhoid reaction has been observed from three months to a year after convalescence, and a slight reaction, though much less than sufiicient to establish a diagnosis of typhoid infection, from one to fifteen years after the disease. In persons in whom the typhoid bacilli persist the serum reaction may last as long as the bacilli remain in the body. As a rule the reaction becomes progressively weaker during the first six months after vaccination, and has usually completely dis- appeared at the end of a year. Typhoid Bacilli in the Blood. — ^A blood culture is usually positive during the first week of typhoid fever, and is the best method for early diag- nosis. Bacilli appear in the blood even in the first few days of the disease. In the first week, nearly 100 per cent, positive results are obtained, in the second week 50 per cent, and then progressively less till the end of the disease. In relapses they reappear. Method. — ^The blood is drawn from the median basilic vein by syringe and is inoculated into either broth or bile medium. If broth is used, 346 TYPHOID BACILLUS several flasks are used with sufficient broth to dilute the blood 50 times or more. The essential thing is to dilute the blood sufficiently so that coagulation does not occur. The separated serum apparently is more strongly bactericidal and inhibits growth. Bile mediums are more convenient, as the bulk need not be so large. The bile inhibits coagulation. Three parts of bile medium are used to one part of blood. The bile is plated on Endo, or Conradi, after eight- een hours' incubation. It should also be examined after two or three days, as growth is sometimes delayed. Another method, not as satisfactory as the preceding, but valuable in emergencies, is to inoculate the blood into ten times its volume of distilled water. Hemolysis results and the proteins supply the nutrient substances for growth. Typhoid Bacilli in Feces. — ^The following table by Schroder gives the number and time of positive results: No. ejcamined. Positive. Per cent. First week 115 11 9.6 Second week 160 45 28.1 Third week 71 16 22.5 Fourth week 41 9 21 .9 Fifth week 26 4 15.3 Sixth week 12 2 16.6 Convalescent 190 10 7.9 These results were obtained using several media in combination. The more promptly the stools are plated after passage, the better is the chance of obtaining typhoid bacilli. Hiss, examining the stools very quickly after passage, obtained a very high percentage of positive results, in 43 consecutive cases, 37 of which were in the febrile stage and 6 convalescent. In a number of instances only one stool was examined, but even under these adverse conditions the average of positive results in the febrile stage was 66.6 per cent. Out of 26 cases of typhoid fever examined in hospitals, 21 were in the febrile stage and 5 convalescent. In 17 of the febrile cases the presence of typhoid bacilli, often in great numbers, was demonstrated. Thus in these care- fully followed cases the statistics show over 80 per cent, of the febrile cases positive. The bacilli were isolated from these cases as early as the sixth day, and as late as the thirtieth day, and in a case of relapse on the forty-seventh day of the disease. These observations, with regard to the appearance of the bacilli in the stools during the febrile stage and their usually quick disappearance, except in the permanent typhoid carriers, after the defervescence, have been confirmed by others. We have, however, a carrier under observation who was discharged from the hospital after three negative examinations. The bacilli are isolated in some cases before the appearance of the Widal reaction; and in some cases in which no Widal reaction was demonstrated. Between the seventh and twenty-first day of the disease, experience seems to indicate that the bacilli may be obtained from about 25 per cent, of all cases on the first exam- ination and from about 75 per cent, after repeated examinations. In some samples of feces typhoid bacilli die out or are overgrown within twenty- four hours; in others they remain alive for days or even weeks. This TYPHOID MCILLI tN THE tlRlNE W seems to depend on the bacteria present in the feces and upon its chemi- cal character. Probably the presence of typhoid bacilli in some stools and their absence in others must be explained largely by the character- istics of the intestinal contents. The short life of the typhoid bacillus in many specimens of feces suggests that stools be examined as quickly as possible. In fact, unless the physician wishes to take the trouble to have the sample of feces sent immediately to the laboratory, it is hardly worth while for the bacteriologist to take the trouble to make the test. (When specimens are delayed in transit the method suggested by Teague and Clurman can be employed. One part of feces should be added if fluid, or well rubbed up if solid, in two parts of 30 per cent, glycerin in 0.6 per cent, salt solution. This concentration of glycerin has no effect on the typhoid bacillus but prevents its overgrowth by the fecal bacteria.) Method. — ^The feces, if solid, are rubbed up with peptone-water, otherwise they can be used without further preparation. The fluidified feces should consist of a mixture of a generous amount of the whole mass, not merely a loopful. The density of the suspension to employ is a matter of experience or successive dilutions may be plated. For general use the Endo or Conradi medium are most frequently employed. We employ Endo and a brilliant green agar, using two concentrations of brilliant green (see p. 107) which by suppression of some or all of the fecal flora allows a heavy inoculation with coincident increase in positive results. None of the fluid enrichment media that we have tried have been satisfactory for general use, although they may be successful in isolated cases. We have had no experience with the recently published method of Teague and Clurman which may yield better results. Plates are streaked (see p. 108) and the colonies of typhoid can then be agglutinated directly either microscopically or by using the slide method. In the latter a drop of saline for control and a drop of high-titre serum are placed on a slide and. the colony rubbed up in the saline and then in the serum. With appropriate dilutions of the serum, agglutination will take placeal most immediately. For fishing, Russell's medium is used, which gives a tentative indica- tion of the nature of the organism, which can be confirmed by agglu- tination and further cultural tests. Typhoid Bacilli in the Urine. — Of great interest ' is the frequent occurrence of typhoid bacilli in large numbers in the urine. The results of the examinations of others as well as our own indicate that the typhoid bacilli are not apt to be found in the urine until the end of the second week of the fever, and may not appear until much later. From this on to convalescence they appear in about 25 to 50 per cent, of the cases, usually in piu-e culture and in enormous numbers, even as high as 100 million per cubic centimeter. They are found until several weeks or months after convalescence; in exceptional cases they persist for years. When we think of the, chances such cases have to spread infection as they pass from place to place, we begin to realize how epidemics can start without apparent cause. The more we investigate the per- 348 TYPHOID BACILLUS sistence of bacteria in convalescent cases of disease, the more difficult the prevention of their dissemination is seen to be. The disinfection of the urine should always be looked after in typhoid fever, and con- valescents should not be allowed to go to places where contamination of the water supply is possible, without at least warning them of the necessity of great care in disinfecting their urine and feces for some weeks. Method. — If the bacilli are numerous, as evidenced by examination in the hanging drop, they are easily obtained by direct plating, other- wise the urine should be centrifuged and the sediment plated, and larger amounts inoculated into bile medium for enrichment. Typhoid Bacilli in Rose Spots and Spleen. — ^Although the bacilU have been frequently isolated from rose spots, it is a less convenient method than blood cultures. Spleen puncture has been employed and although cultures are usually positive the operation is dangerous, and has been abandoned. Detection of Typhoid BaciUi in Water.^There is absolutely no doubt that the contamination of streams and reservoirs is a frequent cause of the outbreak of epidemics of typhoid fever, but the actual finding and isolation of the bacilli is a very rare occurrence. This is often due to the fact that the contamination has passed away before the bacteriological examination is undertaken, and also to the great difficulties encountered in detecting a few typhoid bacilli when they are associated with large numbers of other bacteria. The Importance of Ice in the Production of Typhoid Fever. — ^The total number of instances of typhoid fever which have been reasonably attributed to ice infection are remarkably few. One was in France, where a group of officers placed ice made from water polluted by a sewer in their wine and afterward a large percentage developed typhoid fever, while those of the same company not using ice escaped. A second instance was a small epidemic which occurred among those who used ice from a pond. It was found that water directly infected with typhoid feces had flowed over its frozen surface and been congealed there. The fact that freezing kills a large percentage of typhoid bacilli makes it indeed possible to conceive that ice from moderately infected water while still polluted contains so few living typhoid bacilU that only the exceptional person here and there becomes infected, and so the source of the infection remains undetected. If this be true and scattered cases occur, there should be at least some increase in some if not every year in March, April, and May in such a city as New York, where four-fifths of all the ice consumed is from the Hudson River, which is known to be contaminated with typhoid bacilli. Many persons place the ice directly in their drinking water. When we examine the records for the past ten years we find no increase of typhoid fever in Greater New York during those months, with the one exception of 1907, when we had in the borough of Man- hattan a sharp outbreak lasting four weeks. This outbreak did not occur at all in Brooklyn. As the people of Brooklyn drank different ICE IN THE PRODUCTION OF TYPHOID FEVER 349 water, but received ice from the same places of the Hudson River as those of Manhattan, this directed attention to the water or milk rather than the ice. Examination of the Croton watershed at the time showed that a small epidemic of typhoid existed there and that pollution of the water was probable. This suggested still more strongly that the water and not the ice was the cause of the typhoid infection. It happened that most of the cases occurred in those living in the section of the upper West Side, where only well-to-do people live. An investigation showed that the majority of the infected had used only artificial ice and several had used no ice in their water at all. Life of the Typhoid Bacillus in Ice in Laboratory Experiments. — ^The first important investigation was that of Prudden, who showed that typhoid bacilli might live for three months or longer in ice. This experi- ment is frequently wrongly interpreted, as when a recent writer states: "It has been amply demonstrated that the germs of typhoid fever are not killed by freezing and that they have been known to live in ice for long periods of time." It is true that in Prudden's experiment a few typhoid bacilli remained alive for three months, when the experiment was terminated, but those were but a small fraction of 1 per cent, of the original number. Following Prudden's experiment Sedgwick and Winslow, in Boston, and Park, in New York City, carried on independently a series of experiments. These led to the same conclusions. A table summarizing a final experiment of ours in which twenty-one different strains, mostly of recent isolation, were subjected to the test is given below: Life of Twenty-one Stbains of Typhoid Bacilli in Ice. Average number of bacilli Percentage typhoid in 1 gm. of ice. bacilli living. Before freezing 2,560,410 100.0 Frozen three days 1,089,470 -42.0 Frozen seven days 361,136 14.0 Frozen fourteen days ^ . 203,300 8.0 Frozen twenty-one days 10,280 0.4 Frozen five weelcs 2,950 ' . 1 Frozen nine weeks 127 . 005 Frozen sixteen weeks ....... 107 . 004 Frozen twenty-two weeks In these experiments twenty-one different flasks of Croton water were inoculated each with a different strain of typhoid bacilli. In one a little of the feces rich in typhoid was directly added. The infected water in each flask was then pipetted into thirty tubes. These tubes were placed in a cold-storage room in which the temperature varied from 20° to 28° F. At first tubes were removed and tested twice a week, later once a week. The object of using so many different strains was because it has become evident that some cultures live longer than others. At the end of five weeks the water infected with six cultures was sterile, at the end of sixteen weeks only four strains remained alive. Interesting investigations of Hudson River ice were carried out in 1907 by North. 350 . TYPHOID BACILLUS There was noticed a considerable difference between the niunber of bacteria in the top, middle, and bottom layers of ice. This is natural, since while water in freezing from above downward markedly purifies itself, 75 per cent, of the solids and a fair proportion of bacteria being eliminated, yet this cannot happen in the case of the snow blanket which becomes flooded by rain or by cutting holes through the ice. Here all impurities, such as dust and leaves which have fallen on the surface and dirt which may come from the water, remain with the bacteria which they carry, since all are retained in the porous snow. The bacteria in freshly cut bottom ice generally show the least destruc- tion by freezing. The river water in the forty specimens averaged 1800 bacteria per cubic centimeter, the top ice 306, the bottom ice 36, and the middle ice 14. Only four specimens of top ice had over 500 bacteria per cubic centimeter; none of the specimens of middle or bottom ice. The great destruction by freezing is noticeable in these figures. Even the top ice soiled by flooding and by the horses and men gathering it contained but 16 per cent, as many bacteria as the water from which it was obtained. The bottom ice, the last to be frozen, had but 2 per cent, of those in the water. Conclusions in Regard to Ice Pollution. — The danger from the use of ice produced from polluted water is always much less than the use of the water itself. Every week that the ice is stored the danger becomes less, so that at four weeks it has become as much purified from typhoid bacilli as if subjected to sand filtration. At the end of four months the danger becomes almost negligible, and at the end of six months quite so. Differential Diagnosis. — The typhoid bacillus is easily separated from the other members of the .group, and for practical purposes it is sufficient identification if the colony on the special plating media and the growth on Russell's medium is characteristic and is agglutinated in relatively high dilutions of a serum, so as to eliminate the action of group agglutinins. If a strain does not agglutinate freely, as commonly happens with strains isolated from the blood, it should be transferred daily on plain agar and again tested. The range of group agglutinins in the serum used should be known or error will result. In important examinations or strains from unusual sources as water, etc., a more extensive cultural study should be done. Likewise, absorp- tion of agglutinins and possibly the production of agglutinins for a known strain should be tested. B. (fecalis) Alkaligenes. — ^This group resembles the typhoid bacillus but produces acid from no sugars. It is frequently present in the intestines. It is usually only a saprophyte, but has been found in a few cases of disease in man. Other non-gas-producing typhoid-like bacilli have been isolated not only from the feces of man but also from feces of cholera-infected swine, cow's feces, and water. REFERENCES. Dbeyeb: Jour. Am. Med. Assn., 1916, Ixvi, 1297. Teague and Clubman: Jour. Inf. Dis., 1916, xviii, 663. Tbague an(l Clithman: Jqut. Med. Res., 1916, xxxv, 107. CHAPTER XXIV. PARATYPHOID GROUP. Gartner, in 1888, found the Bacillus enteritidis in association with a meat-poisoning epidemic. A cow sick for two days with profuse diarrhea had been slaughtered and the meat sold for food. Of the persons eating the meat fifty-seven became ill and one diedl Similar bacilli were isolated by Smith from swine suffering from hog cholera, from mice by Loffler, and the bacilli became known as the hog cholera group. Similar bacilli were isolated by others from cases resembling typhoid fever. Schottmiiller reported" in 1900 that he was able to differentiate the strains from 5 cases into two groups, one resembling typhoid very closely, the other corresponding to the enteritidis types. The work of Durham and Buxton gave the basis for classification into subgroups. Numerous types have been isolated from infections in man and animals, and most of the terms in use refer to their source; but many cannot be differentiated one from the other, so fall into distinct groups according to their bacteriological and immune reactions. Gas production from glutfose is usually given as the differential characteristic between this group and B. typhosus. This is true in general for the human pathogens. One type, B. sanguinarium from fowls, produces no gas. This characteristic is susceptible to variation and may be lost. From our own work it would appear that the ability to ferment rhamnose is the basic characteristic of the paratyphoid enteritidis group. The more the strain resembles the typhoid bacillus, however (especially the non-gas-producing type mentioned above), the lower is the avidity for this carbohydrate. The ability to ferment salicin or saccharose, serves to exclude many of the paratyphoid-like organisms found in feces. Indol also is not produced by the types known to be pathogenic for man. Parat3T)hoid A. — Disease Produced. — ^First isolated by Gwyn, later by Schottmiiller, Brown, and Kayser, and others. It produces a typhoid- like disease in man, but is of relatively infrequent occurrence. The bacillus may be present in the feces, urine, blood, and bile. The post- mortem findings have been variable. In a few the typical intestinal lesions of typhoid infections were present, in others they were absent. A diffuse inflammation of the intestines may be found. . Morphology and Biology. — Similar to typhoid in many respects. Krumwiede has recently shown that it differs from all other members of the paratyphoid-enteritidis group in that it does not produce acid from xylose. Not all the non-xylose-fermenting types from man are alike agglutinatively and he suggests that all be included in the para- typhoid "A" group and that the strains differing agglutinatively be 352 PARATYPHOID GROUP considered a subgroup. This fermentative reaction is more reliable than the differentiation by litmus milk. The reaction on litmus milk has, since Schottmiiller's work, been relied upon for differentiation. The reaction, however, is only quantitative and exceptions occur. Communicability. — Communicability is the same as for typhoid. It has been isolated in a few instances from animals, and perhaps in water and foodstuffs, but its distribution can in no way be compared with thiat of the other members of this group. Diagnosis. — See below. Paratyphoid B. — ^This group includes many types, isolated not only from disease of man but also from diseases of many animals. Disease Produced in Man. — This varies and may be subdivided into several clinical types, depending on whether the disease is primarily an infection or an intoxication. Typhoid-like Type. — A variable period of incubation is usually present. The onset is commonly more acute than in typhoid fever, and is usually accompanied by chills. As a rule it runs a shorter and milder course. In some epidemics abortive attacks occur with no symptoms except fever for a short time. The bacilli may be present in the feces, blood, urine, and gall-bladder; in the first two, early in the disease and com- monly during the whole course. Complications similar to those in typhoid may occur. Gastro-enteric Type. — ^The onset is acute and the symptoms come on promptly after eating the infected food or after twenty-four to forty- eight hours. The symptoms are those of an acute enteriti's ushered in by chills and rise of temperature. Nervous symptoms are common and may be severe. The bacillus can be isolated from the blood as early as the first day of the disease. In the mild cases cultures are nega- tive. They are present in the stools during the acute stage but quickly disappear in the acute cases. Cholera-like Types. — These differ from the preceding only in the degree of the toxic and intestinal symptoms. The bacilli may be present in the blood. Localized infections occur as complications, occasionally as primary infections without any other symptoms of paratyphoid infection elsewhere, such as pyelitis (sometimes associated with septicemia), cystitis, arthritis, periostitis, phlebitis, cholecystitis, etc. Infection with these types occasionally occurs as a complicating infection of other diseases. Morphology and Biology. — ^They closely resemble the other members of this group. For differential characteristics see pp. 324 and 351. Occurrence in Healthy Persons; Paratyphoid Bacillus Carriers ("A" and "B" Types). — ^When general invasion occurs the bacilli may be found in the urine after convalescence for shorter or longer periods of time. Its localization in the gall-bladder and its excretion in the feces by chronic carriers is fairly frequent. We recently had the opportunity of examining the feces of every man in A militia regiment which had been badly infected with paratyphoid "A", while at the Mexican border. We found nearly 4 per cent, of healthy carriers. This incidence of BACILLUS ENTERITIDIS 353 normal carriers is interesting in relation to normal carriers of B. typhosus. Is this greater incidence only apparent, due to the greater ease with which we can isolate B. paratyphosis as compared with B. typhosus or actual, because the former known to be less virulent, a greater number of persons escape infection but become carriers for a shorter or longer time? Careful bacteriological examination has revealed parathyphoid- like bacilli in the feces of a considerable percentage of healthy persons. This fact is unfortunately lost sight of, and etiological importance ascribed to such organisms when isolated during disease. Because of the range of group agglutination, such organisms may even be agglutin- ated by the patients' serum as well. Frequency of Paratyphoid Infections. — In this country the disease is relatively infrequent, though statistics are not available. It is frequent in Europe, especially in certain districts. In the present war the disease has been very common. In this country military camps have been favorable places for its development as seen in the outbreak of para- typhoid ("A") fever in the militia at the Mexican border. Communicability. — ^The individual case and the carrier either through contact or by the contamination of water and milk, etc., are the usual sources of infection for paratyphoid fever. Food infection, however contaminated, plays the important role in the development of the more acute types of disease. Infection has followed the consumption of the following foods: milk and milk products; fish; potato salad; variously prepared string beans. In water it is more capable of a saprophytic existence than the typhoid bacillus, and has been isolated more often. , Meats or edible organs from cattle, swine, horse, sheep, geese, as well as various products containing these, are most frequently associated with acute food poisoning or infection. Although contamination of meat may occur in other ways the important source in epidemics is meat from infected animals. Such meat may by contact infect meat from healthy animals. Pickled, salted, and smoked meat may be a source of infection. Cooking cannot be relied upon to kill the bacilU because of the slow penetration of the heat. It is occasionally found in the feces, meat, and organs of some of the domestic animals, although apparently healthy. It has also been isolated from meat products and market milk. B. Enteritidis. — ^In human pathology this bacillus plays the same role as the B. 'paratyphosus B. Meat is the usual source of infection, although other foodstuffs have been the cause. The disease is almost always of the acute type (see above). It is found in the feces, urine, and blood. Chronic carriers have not been found, although the bacilli have been foimd in gall-stones, and probably do occur. They may be present in the stools of healthy persons exposed to infection. Contact infection occurs, but infrequently. The pathological lesions are the same as those caused by the B. paratyphosus B. Infection of domestic animals occurs and the dissemination of the bacillus is the same as for paratyphoid B. It has not been found in water, probably not in milk, and infrequently in meat products. In cultures it is similar to B. paratyphosus B. For methods of diagnosis see below. 23 354 PARATYPHOID GROUP Virulence. — ^Although varying in degree, the various members of the group are highly virulent for white mice, guinea-pigs, rabbits, and rats. Death usually results from a septicemia. Toxic substances are freely .produced in fluid media, which are heat resistant. The paratyphoid A. is probably the least virulent of the group. Immunity .^-In experimental animals immunity is produced by feeding and by injection. Immunity is produced not only against the homol- ogous strain but, as a rule, against related strains. It has been sug- gested that because of the cross-protection of the members of these groups and the production of group antibodies active against typhoid or paratyphoid when one or the other is injected that a mixed vaccine of the typhoid-paratyphoid group be used. (Vaccines, see Part III.) DifEerential Diagnosis of Members of the Groups.— The cultural differentiation only separates the A type from the B and enteritidis types. The members of the latter two groups, whether from man or animals show differences, but definite cultural groups have not been established. Weiss and Rice found that inosite fermentation was essentially a characteristic of B. paratyphosus B. Krumwiede, Pratt and Kohn, studying the reduction of fuchsin, the fermentation of xylose, arabinose and dulcit, and also inosite, found that a grouping results which is highly suggestive in relation to some types and especially to host origin. The differences found, however, did not in many instances correlate with the agglutinative relationships. Much more work is necessary before the significance of such differences can be decided. Immunity reactions, such as agglutination, fail to separate the members within the different groups, although the different groups are easily separated. As with other allied bacteriaj however, group agglu- tinins are present. Other immune reactions give the same results. Variation in Types. — ^The related enteritidis-paratyphoid strains show a wide ability for variation, not only culturally but in their patho- genicity and virulence. The above-described differences hold good for strains isolated from natural infections. There is reason to believe that a strain from man can adapt itself and develop a full degree of virulence for an animal host, and vice versa. Even its original agglutination reactions may vary in degree, so that there is an apparent shifting from the paratyphoid to the enteritidis type; the antigenic properties do not change, however, in a parallel manner. In the study of the group, therefore, this should not be lost sight of and differential value thereby given to what may be variant characters. Even such characters as gas production may be lost or be extremely variable. Much work is needed before we shall be able to differentiate between dominant constant characters on which a classi- fication can be based and variable characters, and also learn the range of variability. The extraordinary degree of cross-agglutination which may occur even with totally dissimilar types has been shown by Smith and Ten Broek. Because of adaptability of the members of the group the presence of any organism belonging to this group should be looked upon as a potential source of infection for man. PARATYPHOID-ENTERITIDIS GROUP 355 Diagnostic Methods. — The diagnostic methods are the Widal reaction, blood culture, isolation from the stool or urine or other sites of infection. The use of the Widal reaction for a differential diagnosis, from typhoid or between infection with different members of the group, may occasion- ally lead to errors because of the presence of group agglutinins. The titre of the serum should be determined against each of the types, that type which is agglutinated in the highest dilution, being the probable infecting organism. For surety of diagnosis the other methods should be employed. Their application is the same as in typhoid fever. For final diagnosis of the cultures isolated, immune sera against each of the types is necessary. The range of group agglutination of the sera used must be known before they are used for identification. Paratyphoid-like Bacilli. — B. Paratyphosus C. — Bacilli having all characteristics of the paratyphoid group, but not agglutinated by para- typhoid or enteritidis serum, have been isolated from swine suffering from hog cholera, from man, meat, etc. The paratyphoid-like organisms found in feces, etc., but without pathological significance are frequently spoken of as belonging to this group. It seems to us better to reserve the term for the known pathogenic types. Types of the Paratyphoid-enteritidis Group Found in Animal Dis- eases. — B. Paratyphosus B. Types. — B. suipestifer, hog-cholera bacillus commonly present as a secondary invader. The disease is caused by a filtrable virus. It is commonly encountered in food poisoning in man. B. typhi murium, mouse typhoid bacillus, is used for the destruction of mice, successfully only when the technical details for the maintenance of virulence are carefully carried out. Instances of infection in man have been reported, and in Prussia its use is restricted by law. B. psittacosis, the cause of enteritis in parrots, infectious for man, the disease being spoken of as psittacosis. Pseudotuberculosis of Ouinea-pigs. — ^A paratyphoid bacillus is found in the pseudotubercular lesions in the spleen and liver and lymph nodes. Sporadic cases or epidemics are common. Recently the condition developed in a series of guinea-pigs which we had injected with milk to test for the presence of tubercle bacilli. Unless careful examination is made confusion with tuberculous lesions might occur. Enteritidis Types. — Rat Viruses. — Danyz isolated an organism from an epidemic of mice which had a high degree of virulence for rats and' mice. It has been widely used for the extermination of rodents. Similar strains have been isolated by others. They are obtainable in the market as Ratin, Danyz-Virus, Liverpool Virus, etc. Recently infection in man has been reported. Domestic animals, meat and meat products should be protected from contact with the virus. Other Types. — Infectious Abortion in Mares. — ^Various investigators have isolated paratyphoid types from infected mares which apparently constitute a distinct group. REFERENCES. Kbumwiede: Jour. Med. Res., 1916, xxiv, 335. Kbumwibdji!, Peatt and Kohn: Jour. Med. Res., 1917, xxxv, 357. ScHOTTMULLEE : JouT. Med. Res., 1916, xxv, 55. Ten BeoSk: Ibid., 1915, xxi, 503. Weiss and Rice: Jour. Med. Res.. 1917. xxxv. 403. CHAPTER XXV. DYSENTERY GROUP. Dysenteky may be divided into acute and chronic. Amebse appear to be the chief exciting factor in most cases of chronic dysentery, though bacilli of the colon group also play a part. In temperate climates acute dysentery is but very rarely due to amebse, but usually to the bacilli identified by Shiga or to allied strains identified by Kruse, Flexner, Park, Hiss, Strong and others. The usual summer diarrheas are not excited by the dysentery bacilli. Historical Note. — ^In 1897 Shiga found in the stools of cases of dysen- tery a bacillus which had not been before identified. In 1900 Flexner and Strong isolated bacilli which they at that time considere^d the same as those isolated by Shiga., In the same year Kruse, in Germany, isolated bacilli from cases of asylum dysentery which differed, however, in their agglutinative properties. Park and Dunham (1902) isolated a bacillus from a severe case of dysentery during an epidemic at Seal Harbor, Mt. Desert, Maine, which they showed differed from the Shiga dysentery bacillus in that it produced indol and differed in agglu- tinating characteristics. At first they considered it the same as the Flexner strain, but it was shown later by Park to be a distinct variety, and later found by him in widely separated epidemics. Martini and Lentz, in December, 1902, found that the Shiga type was present in separate epidemics in Europe, but also in some cases, that other types similar to those isolated by Flexner, Park, Kruse, and others were found. These types differed from the Shiga type in fermenting mannite and in agglutination. In January, 1903, Hiss and Russel showed that a strain isolated by them differed from the Shiga type in the same characteristics. German observers at first were inclined to consider the Shiga type as the only one producing dysentery, while the American observers considered both types of equal importance. Park investigated several epidemics and isolated only the Shiga type from some, from others either the Park-Hiss or the Flexner types, thus definitely proving the importance of the mannite fermenting types. The results obtained by others later were the same, so that no doubt exists that all of the types produce true dysentery. Morphological and Cultiiral Characteristics of Dysentery Bacilli. — Microscopic. — Similar to bacilli of the colon groups Staining. — Similar to bacilli of the colon group. Motility. — ^No definite motility has been observed. The molecular movement is very active. Flagella are absent. PATHOGENESIS 857 Appearance of Ci4liures. — On gelatin the colonies appear more like the typhoid than the colon bacilli. Gelatin is not liquefied. On agar growth is somewhat more delicate than that of the average colon cultures. On Potato. — ^A delicate growth just visible or distinctly brownish. In Bouillon. — Diffuse cloudiness with slight deposit and sometimes a pellicle. See p. 324 for comparison with other members of colon-typhoid group. The fermentation reactions vary, and on this basis the dysentery group is divided into subgroups and types. The first three groups are encountered in the United States. The last type is of infrequent occur- rence but is accepted as a separate type by Shiga and by Lentz. Glucose. Mannite. Maltose. Saccharose. Indol. B. dysenterise + — — — — B. paradysenterise : Type 1 (Park-Hiss) . . + + - - + Type 2 (Flexner) . . . + + + - + Type 3 (Strong) . . . + + - + + In differential tests 2 per cent, mannite and 2.5 per cent, maltose give surer and more prompt reactions than the usual 1 per cent. The fermentations as given are only of differential value with freshly isolated cultures. After artificial cultivation as noted by Hiss and by Lentz the Park-Hiss strains may ferment maltose, and the Flexner strains, saccharose. /' .^sv '-> / * * ^ -^'■» • * I. * '.■' K- ■ •! • . t 1 '«" • <* • # 1 . «^ * ,;j/ ■ '%■ • * * f • * '■ "■* 1 ,.^ Fig. 134.- -Dysentery bacilli. X 1000 diameters. Fig. 135. — Colony of dysentery bacilli in gelatin. X 40 diameters. Pathogenesis. — Animal Tests. — No characteristic lesions with one exception have followed the feeding of large quantities of bacilli. Dogs at times have diarrhea with slimy stools, but autopsy shows merely a hyperemia of the small intestine. The disease can be produced and occurs spontaneously in monkeys. Many animals are very sensitive to dead or living bacilli injected into the subcutaneous tissues, vein or peritoneal cavity. 358 DYSENTERY GROUP The autopsy of animals dying quickly from injection into the peri- toneum of living or dead bacilli shows the peritoneum to be hyperemic, the cavity more or less filled with serous or bloody serous exudate. The spleen is sometimes moderately swollen. The small intestine is filled with fluid, the large intestine is usually empty. The mucous membrane of both is hyperemic aiid sometimes contains hemorrhages. Conradi found ulcer formation in one case. Subcutaneous injections of dead or living cultures are followed by infiltration of tissues and frequently by abscess formation. The dysen- tery bacilli are not found in the blood or organs of animals. Toxins. — ^A highly poisonous endotoxin is found in autolysates and in filtered broth cultures. When injected into animals death results and on autopsy the same lesions are found as following the injection of live or dead bacilli. This endotoxin can be neutralized by large doses of immune serum. Kraus and others claim to have demonstrated the presence of an extracellular toxin against which an antitoxin can be produced. The Shiga type is the more toxic, the others less so or irregularly so. In Man. — The etiological significance of the dysentery bacillus in man is not only shown by its constant presence and by the presence of immune, bodies in the serum of infected persons, but also by experi- mental infection of man. Thus Strong infected two condemned criminals w^th pure cultures. Jehle infected himself. Kruse reported two acci- dental laboratory infections, and others have occurred. As a general rule infection with the Shiga type is more severe and the mortality higher than infection from other types. But the severity of individual cases varies widely during an epidemic. Prevalence of the Disease. — ^The disease is distributed over the whole world. The Shiga type and Type I of the paradysentery varieties are most commonly found in the United States. • Character of Disease in Man. — In the onset, acute dysentery is sudden ■ and ushered in by cramps, diarrhea, and tenesmus. Bacillary dysentery is a disease especially of the mucous membrane of the large intestines. The epithelium is chiefly involved. In the lightest cases a catarrhal inflammation alone is present, in the more severe the lymph follicles are swollen and some necrosis of epithelium takes place. In severe cases in adults the lesions are of a diphtheritic character and may be very marked. The entire lumen of the intestines may be filled with a fibrinous mass of pseudomembrane. In young children, even in fatal cases, the lesions may be more superficial. Distribution of the Bacilli. — ^The bacilli are only foimd in the intestines. They do not invade the rest of the body. The feces, therefore, are the only excretions containing them. Duration of life Outside of the Body.— They usually dje in stools in one to two days. In water they die out in several days to a week, exceptionally after a longer time. Communicability. — ^The infected person is mainly responsible for the spread of dysentery. Especially dangerous are the mild or subacute cases PATHOGENE&IS 350 or carriers. Such infection may be direct or it may be indirect through contamination of food, dishes, linen, and clothes; water may also be infected and be the cause of epidemics. The bacillus has been isolated from water in such epidemics. Bacillus Carriers. — Both healthy and convalescent carriers are found. The former may be equal to one-fourth or even one-half of the number of cases. Convalescents commonly excrete the bacilli for weeks and some become chronic carriers. The excretion is irregular, and slight relapses occur, when the bacilli are more numerous. The importance of finding and isolating the carriers diu-ing an epidemic is obvious. This is not always possible and general precautions against infection should be taken. Susceptibility. — ^The frequent occurrence of healthy carriers and mild cases along with severe and fatal cases shows the varying resistance to infection. Disturbances of digestion and other conditions lowering the general resistance, such as heat and fatigue, are factors in infection. Immunity.— Immune bodies appear in the blood some time after infection. In animals, immune bodies are also produced by injection. Agglutinins, bactericidal substances, precipitins, and opsonins are found. Neutralization of the toxic products is also possible with immune sera. True antitoxin is also probably produced. Active immunity in animals is produced with difficulty against the Shiga type because of its high toxicity. Against the other types, how- ever, immunity is more easily produced. . Vaccines and Serums. — See Part III. Diagnosis. — ^The use of the Widal reaction is limited, because the symptoms are sufficiently diagnostic in well-marked cases and, fm-ther, the reaction does not appear during the early acute stage. Later it may be used. Group or normal agglutinins are commonly present and inter- fere greatly with the value of the reaction in the paradysentery types. The action on the infecting subgroup is usually sharp, but infection by the types of the paradysentery group can frequently not be differentiated. An agglutination reaction is commonly present in chronic carriers. For even moderate diagnostic value the serum should agglutinate in 1 to 50 in infections with B. dysenteries and in 1 to 100 in other types. • . . . . Isolation of the bacillus is the only method of diagnosis for identi- fication of the type causing the infection. The mucous flakes in the stool should be selected for plating. The methods of isolation and identi- fication are the same as for typhoid or paratyphoid, except that the crystal violet should be omitted from the Conradi medium as the growth of many strains is inhibited more or less by aniline dyes. The bacilli are so abundant in the mucus of most cases that they can be readily isolated from nutrient agar plates. If Endo is used the plates should be fresh. When the color returns the growth of the Shiga types is usually inhibited. Differential Diagnosis of Type. — ^The cultural differences have been given. The actual nature of the bacillus should be verified by agglu- 360 DYSENTERY GROUP tination, although the cultural characters will suffice for a tentative diagnosis from a case clinically typical. Agglutination. — Sera from rabbits or goats are used, as horses develop normal and group agglutinins to. a high degree. For practical purposes two sera, one for B. dysenterioe and a polyv- alent serum for the paradysentery types may be used, further differ- entiation being determined culturally. In the use of sera for differen- tiation of the latter groups univalent sera for each of the types must be on hand and all the precautions against error from group agglutinins should be observed. Variations in agglutinability of freshly isolated strains adds to the difficulty. Variability. — Cultures, after prolonged artificial cultivation, may show variations in their sugar fermentations. By passage through animals (monkey, etc.), the original characteristics are usually restored. CHAPTER XXVI. BACILLUS PYOCYANEUS (BACILLUS OF GREEN AND OF BLUE PUS). BACILLUS PROTEUS (VULGARIS). BACILLUS PYOCYANEUS. The blue and green coloration which is occasionally found to accom- pany the purulent discharges from open wounds is usually due to the action of the Bacillus pyocyaneus. It was first obtained in pure culture and its significance noted by Gessard, 1882. Morphology. — Slender rods from O.S/i to Ifi broad and from 2fi to 6m long; frequently united in pairs or in chains of four to six elements; occasionally growing out into long filaments and twisted spirals. The bacillus is actively motile, a single flagellum being attached to one end- Does not form spores. Stains with the ordinary aniline colors; does ., - not stain with Gram's stain. y/^ Biology. — Aerobic, facultative anaerobic, liquefying, motile bacil- lus. Growing anaerobically it pro- duces no pigment. Grows readily on all artificial culture media at room temperature, though best at 37° C, and gives to some of them a bright green color in the presence of oxygen. In gelatin-plate cultures the colonies are rapidly developed, **.«'' i* *ftj * imparting to the mediimi a fluores- "^ **.*'*' cent green color; liquefaction begins ^^^ ise.-Baciiius pyocyaneus. (From at the end of two or three days, and KoUe and Wassermann.) by the fifth day the gelatin is usually entirely liquefied. The deep colonies, before liquefaction sets in, appear as round, granular masses with scalloped margins, having a yellowish-green color ; the surface colonies have a darker green centre, surrounded by a deli- cate, radiating zone. In stick cultures in gelatin, liquefaction occurs at first near the surface, in the form of a small funnel and gradually extends downward; later the liquefied gelatin is separated from the solid part of the medium by a horizontal plane, a greenish-yellow color being imparted to that portion which is in contact with the air. On agar a wrinkled, moist, greenish-white layer is developed, while the surrounding medium is bright green; this subsequently becomes darker in color, changing to blue green or almost black. In bouillon the green color is produced, and the growth appears as a delicate, flocculent sediment. Milk is coagulated and assumes a yellowish-green color. 362 BACILLUS PYOCYANEVS Figment. — ^Two pigments are produced by this bacillus^one of a fluorescent green which is common to many bacteria. This is soluble in water but not in chloroform. The other (pyocyanin) of a blue color is soluble in chloroform, and may be obtained from pure solution in long, blue needles. This pigment distinguishes the Bacillus pyocyanevs from other fluorescing bacteria. Pigment production is usually more marked with incubation at 22° C. The ability to produce pigment may be lessened or lost by artificial cultivation. Ferment. — Besides the ferment causing liquefaction of gelatin there is one which acts on albumin. It resists heat. This ferment called pyocyanase is able to dissolve- bacteria, and it has been stated to have some protective power when injected into animals. It has been used locally in diphtheria in a number of cases. We think it has no advantage over the cleansing preparations. Distribution. — ^This bacillus is very widely distributed in nature; it is frequently found on the healthy skin of man, in the feces of many animals, in water contaminated by animal or human material, in purulent discharges, and in serous wound secretions. Pathogenesis. — Its pathogenic effects on animals have been care- fully studied. It is pathogenic for guinea-pigs and rabbits. Sub- cutaneous or intraperitoneal injections of 1 c.c. or more of a bouillon culture usually cause the death of -the animal in from twenty-four to thirty-six hours. Subcutaneous inoculations produce an extensive inflammatory edema and purulent infiltration of the tissues; a sero- fibrinous or purulent peritonitis is induced by the introduction of the bacillus into the peritoneal cavity. The bacilli multiply in the body and may be found in the serous or purulent fluid in the subcutaneous tissues or abdominal cavity, as well as in the blood and various organs. When smaUer quantities are injected subcutaneously the animal usually recovers, only a local inflammatory reaction being set up (abscess), and the animal is subsequently iinmune against a second inoculation with doses which would prove fatal to an unprotected animal. Loew and Emmerich have shown that the enzymes produced in the pyocyaneus cultures are capable of destroying many forms of bacteria in the test- tube, and have a slight protecting value in the body. Its presence in wounds in man greatly delays the process of repair, and may give rise to a general depression of the vital powers from the absorption of its toxic products. This bacillus has been obtained in pure culture from pus derived from the tympanic cavity in disease of the middle ear, from cases of ophthalmia, and bronchopneumonia. Kruse and Pasquale have found the organism in three cases of idiopathic abscess of the liver, in two of them in immense numbers ^and in pure culture. Ernst and Schiirmayer report the presence of the bacillus pyocyaneous in serous inflammations of the pericardial sac and of the knee-joint. Ehlers gives the history of a disease in two sisters who were attacked simultaneously with fever, albuminuria, and paralysis. It was thought that they would prove to have typhoid fever or meningitis, but on the twelfth day there was an eruption of blisters, from the contents of which the Bacillus pyocyaneus was isolated. Krambals refers to BACILLUS PROTEUS 363 seven cases in which a general pyocyaneus infection occurred, and adds an eighth from his own experience. In this the Bacillus pyocyaneus was obtained postmortem from green pus in the pleural cavity, from serum in the pericardial sac, and from the spleen in pure culture. Schim- melbusch states that a physician injected 0.5 c.c. of sterilized (by heat) culture into his forearm. As a result of this injection, after a few hours he had a slight chill, followed by feverj which at the end of twelve hours reached 38.8° C; an erysipelatous-like swelling of the forearm occurred, and the glands in the axilla were swollen and painful. Wasser- mann reports an epidemic of septic infection of the newborn, starting in the umbilicus. In all there were eleven deaths. Lartigau found it in well water, and in great abundance in the intestinal discharges of a number of cases made ill by drinking the water. It has also been found in a certain number of cases of gastro-enteritis in which no special cause of infection could be noted. We may therefore conclude from these facts that the Bacillus pyo- cyaneus, although ordinarily but slightly pathogenic for man, may under certain conditions, as in general debility, become a dangerous source of infection. Children would seem to be particularly susceptible. Differential Diagnosis of the Pyocyaneus from other Fluorescing Bacteria. — ^This is easy enough as long as it retains its pigment-producing property. When an agar culture is agitated with chloroform a blue coloration demonstrates the presence of this bacillus. When the pyocyanin is no longer formed, however, the diagnosis is by no means easy, particularly when the pathogenic properties are also gone. Immunity. — ^Animal infection is followed by the production of anti- toxic and bactericidal substances. BACILLUS PROTEUS (VULGARIS). The term B. proteus is used for a group of bacilli classed as putre- factive bacteria, because they decompose protein substances with the production of a disagreeable odor. They were discovered by Hauser in 1885. The limits of this group are not well defined nor has it been determined how many varieties there are in this group. The following is a description of a typical liquefying variety. Morphology. — Bacilli varying greatly in size; most commonly occur- ring 0.6;u broad and 1.2|i long, but shorter and longer forms may also be seen, even growing out into flexible filaments which are sometimes more or less wavy or twisted like braids of hair. The bacillus does not form spores, and stains readily with fuchsin or gentian violet. It is Gram-negative. Biology. — An aerobic, facultative anaerobic, liquefying, motile bacillus. Grows rapidly in the usual culture media at room temperature. Growth on Gelatin. — ^The growth upon gelatin plates containing 5 per cent, of gelatin is very characteristic. At the end of ten to twelve hours at room temperature small round depressions in the gelatin are observed which contain liquefied gelatin and a whitish mass con- sisting of bacilli in the centre. Under a low-power lens these depres- 364 BACILLUS PYOCYANEUS sions are seen to be surrounded by a radiating zone composed of two or more layers, outside of which is a zone of a single layer, from which ameba-like processes extend upon the surface of the gelatin. These processes are constantly undergoing changes in their form and position. The young colonies deep down in the gelatin are somewhat more com- pact, and rounded or hump-backed; later they are covered with soft down; then they form irregular, radiating masses, and simulate the superficial colonies. When the consistency of the medium is more solid, as in 10 per cent, gelatin the liquefaction and migration of surface colonies are more or less retarded. In gelatin-stick cultures lique- faction take place rapidly along the line of puncture, and soon the entire contents of the tube are liquefied. Upon nutrient agar a rapidly spreading, moist, thin, grayish-white layer appears, and migration of the colonies also occurs. Milk is coagulated, with the production of acid. Cultures in media containing albumin or gelatin have a disagree- able, putrefactive odor, and become alkaline in reaction. Growth is most luxuriant at a temperature of 24° C, but is plentiful also at 37° C. It is an aerobic bacillus but it grows also in the absence of oxygen. In the latter condition it loses its power of liquefying gelatin. It produces indol and phenol from peptone solutions. The proteus develops fairly well in urine, and decomposes urea into carbonate of ammonia. Pathogenesis. — This bacillus is pathogenic for rabbits and guinea-pigs when injected in large quantities into the circulation, the abdominal cavity, or subcutaneously, producing death with symptoms of poison- ing. Hauser has obtained the Bacillus proteus {vulgaris) from a case of purulent peritonitis, from purulent puerperal endometritis, and from a phlegmonous inflammation of the hand. It is probable that in some instances food poisoning has been due to the contamination of foods by B. proteus. Because of the proteolytic power, toxic products " ptomaines" may develop as a result of its growth. Under these conditions decomposition has started and the food is disa- greeable both in taste and odor and for this reason food poisoning of this type is probably much more uncommon than that due to members of the paratyphoid-enteritidis group, or to B. botulinus where there is no change or only a slight change in odor and taste. Proteus vulgaris has been found to be the predominating organism in the alvine discharge in cases of cholera infantum. The prominent symptoms in these cases were drowsiness, stupor, and great reduction in flesh, more or less collapse, frequent vomiting and purging, with watery and generally offensive stools. The Proteus vulgaris appears to be next in importance to the Bacillus communis in the etiology of cystitis and pyelonephritis. The Proteus vulgaris is usually a harmless parasite when located in the mucous membrane of the nasal cavities. Here it only decom- poses the secretions, with the production of a putrefactive odor. It is found occasionally in the discharge from cases of otitis media in association with other bacteria. CHAPTER XXVII. THE BACILLUS AND THE BACTERIOLOGY OF TUBER- CULOSIS. Historical Note. — ^A knowledge of phthisis was certainly present among men at the time from which our earliest medical descriptions come. For over two thousand years many of the clearest thinking physicians have considered it a communicable disease; but it is only within comparatively recent times that the infectiousness of tulaerculosis has become an established fact in scientific medicine. ViUemin, in 1865, by infecting a series of animals through inocu- lations with tuberculous tissue, showed that tuberculosis might be induced, and that such tissue carried the exciting agent of the disease. He also noticed the difference in virulence between tuberculous material of human and bovine sources, and says that not one of the rabbits inoculated with human material showed such a rapidly progressive and widespread generalization as those receiving inaterial from the cow. Baumgarten demonstrated, early in 1882, bacHU in tissue sections which are now known to have been tubercle bacilli. But these investigations and those of others at the same time, though paving the way to a better knowledge of the disease, proved to be unsatisfactory and incomplete. The announcement of the discovery of the tubercle bacillus was made by Koch in March, 1882. Along with the announcement satisfactory experimental evidence was presented as to its etiological relation to tuberculosis in man and in susceptible animals, and its principal biological characters were given. He submitted his. full report in 1884. Innumerable investigators now followed Koch into this field, but their observations served only to confirm his discovery. Distribution of Bacilli. — They are found in the sputum of persons and animals suffering from pulmonary or laryngeal tuberculosis, either free or in the interior of pus cells; in miliary tubercles and fresh caseous masses in the lungs and elsewhere; in recent tuberculous cavities in the lungs; in tuberculous glands, joints, bones, serous effusions, mucous membranes, and skin affections. They are also found in the feces of those suffering from tuberculosis of the intestines or of those who have swallowed tuberculous sputum. They are frequently present in the blood in very small numbers, in large number only in acute miliary tuberculosis. Morphology. — ^The tubercle bacilli are slender, non-motile rods of about 0.3 fi in diameter by 1.5ai to 4/^ in length. (Plate VI, Figs. 1 and 2.) The morphology is extremely variable, especially on culture media, and varies with the type of medium used. Commonly they occur singly or in pairs, and are then usually slightly curved; frequently they are observed in smaller or larger bunches. Under exceptional conditions branching and club-shaped forms are observed. The tubercle bacillus is probably closely allied to nocardia. In stained preparations there •are often seen unstained portions. In old cultures irregular forms may develop, the rods being occasionally swollen at one end or presenting 366 BACILLUS AND BACTERIOLOGY OF TUBERCULOSIS lateral projections. Here also spherical granules appear which stain with more difficulty than the rest of the bacillus and also retain the stain with greater tenacity. The bacilli, however, containing these bodies are not appreciably more resistant than those not having them; therefore they cannot be considered true spores. The bacilli have a thin capsule, shown in one way by the fact that they appear thicker when stained with fuchsin than with methylene blue. Chemical Constituents of the Tubercle Bacilli. — ^Water 86 per cent.; dry substance 14 per cent., 25 per cent, of which is soluble in alcohol and ether, consisting of free fatty acids and fatty acids combined with the higher alcohol "mykol" to form a wax; lecithin proteins, other nucleo-albumins, and inorganic bases constitute the remainder. Staining Peculiarities.— These are very important, for by them its recognition in microscopic preparations of sputum, etc., is rendered possible. Owing to content of waxy substance it does not readily take up the ordinary aniline colors, but when once stained it is very difficult to decolorize, even by the use of strong acids. The more recently formed bacilli are much more easily stained and decolorized than the older forms. For methods of staining see pp. 77 and 80. Biology. — The bacillus of tuberculosis is a parasitic, aerobic, norv- mdtile bacillus, and grows best at a temperature of about 37° C„ limits 30° to 42° C. It does not form true spores. Resistance. — The bacilli, because of the protection given by their waxy substances, it has been assumed, have a somewhat greater resisting power than most other pathogenic bacteria. Frequently a few out of a great number of bacilli resist desiccation at ordinary temperatures for months; most bacilli die, however, soon after drying. There is a greater resistance shown by the tubercle bacillus than by most other non- ' spore-bearing bacilli to the action of the products of bacterial growth as in souring milk, in water, and in sewage. In water and sewage they may remain viable for weeks. They frequently retain their vitality for several weeks, or even months, in putrefying material, such as sputum. In cultures the bacilli do not live longer than three months. EXPLANATION OF PLATE VI. Fig. I. ^Tuberculous lymph node "giant cell" containing tubercle bacilli "human type." Bacilli red, rest of specimen blue. Ziehl-Neelsen stain. X 1000 diam. Fig. 2. — Tuberculous sputum from human case. Stain same as above. X 1000 diam. Fig. 3. — Tuberculous sputum, human case. Stained by Hermann's method. Tubercle bacilli violet, rest of specimen brown. Fig. 4. — Pus from tuberculous abscess in cow. "Bovine type" of bacillus. Stained same as Figs. 1 and 2. X 1000 diam. Fig. 5. — Section through leprous skin showing bacilli in clumps in and out of cells and large "leprous cell" containing a ball of bacilli. Stained with Ziehl-Neelseu. Fig. 6. — Photograph of human type of tubercle bacilli from sputum. Bacilli in red, ' rest of specimen blue. X 1000 diam. (Frankel and Pfeiffer.) PLATE VI ^ e> '^. ®^^.; 6 • 'f'l < •S"**- l\ f • I ^ . I Y^^ :5 BIOLOGY 367 unless the media be favorable, such as egg or serum; transplants after this time may fail to grow. A few bacilli, sufficient to infect guinea-pigs, may persist much longer. Cold has little effect upon them. When dry, some of the organisms stand dry heat at 100° C. for twenty minutes but are dead in forty-five minutes; but when in fluids and separated as in milk, they are quickly killed — viz., at 60° C. in twenty minutes, at 65° C. in fifteen minutes, at 70° C. the great majority in one minute, all in five minutes, at 80° C. the great majority in one-half minute, all in one minute, and at 95° C. in one-half minute. In some experiments they appear to withstand a higher temperature. As pointed out by Theobald Smith, when milk is heated in a test-tube in the usual way. the cream which rises on heating is exposed on its surface to a lower temperatiure than the rest of the milk, and as this contains a large percentage of the bacteria some of them are exposed to less heat than those in the rest of the fluid. Rosenau points out another source of error: If a moderate number of killed bacilli are injected, limited lesions will arise and caseation may follow. On killing and autopsying the animals, tubercle bacilli can then be demonstrated in smears from the lesions, and the inocula- tion is considered positive. If, however, this material is reinjected into a second pig, the latter will show nothing on autopsy. This capacity of dead bacilli to cause macroscopic lesions has long been shown by Prudden and Hodenpyl. Its importance, however, is not sufliciently considered. The resisting power of this bacillus to chemical disinfectants, drying, and light is considerable, but not as great as it is apt to appear, for, as in sputum, the baciljus is usually protected by mucus or cell protoplasm from penetration by the germicidal agent. It is not always destroyed by the' gastric juice in the stomach, as is shown by successful infection experiments in susceptible animals by feeding them with tubercle bacilli. They are destroyed in sputum in six hours or less by the addition of an equal quantity of a 5 per cent, solution of carbolic acid. Bichloride of mercury is less suitable for the disinfection of sputum as it combines with the mucus and forms a more or less protecting envelope. Iodoform has no effect upon cultures until 5 per cent, is added. The fumes from four pounds of burning sulphur to each 1000 cubic feet of air space will kill tubercle bacilli in eight hours when fully exposed to the action of the gas, providing they are moist, or abundant moisture is present in the air. Formaldehyde gas is quicker in its action, but not much more efficient. Ten ounces of formalin should be employed for each 1000 cubic feet of air space. The tubercle bacillus resists the action of alkaline hypo- "FiG. 137. — Tubercle bacilli. Impression preparation from small colony on coagu- lated blood serum. X 1000 diameters. v- 368 BACILLUS AND BACTERIOLOGY OF TUBERCULOSIS chloride solution (" antiformin") in dilutions which quickly dissolve non-acid fast bacteria. The tubercle bacillus in sputum when exposed to direct sunlight is killed in from a few minutes to several hours, according to the thick- ness of the layer and the season of the year; it is also usually destroyed by diffuse dayhght in from five to ten days when placed near a window in fine powder. Protected in cloth the bacilli survive exposure to light for longer periods. Tuberculous sputum expectorated upon sidewalks, etc., when left undisturbed in the shade may be infectious for weeks, but when exposed to the action of direct sunlight will, in many cases, especially in summer, be disinfected by the time it is in condition to be carried into the air as dust, but not before children and flies have an opportunity of getting into it. The action of sunlight and other more important hygienic reasons suggest that the consump- tive patients should occupy light, sunny rooms. Fig. 138. — Tubercle bacilli, bovine. X 1000 diameters. Fig. 139. — Tubercle bacilli, human. X 1000 diameters. Dried sputum in rooms protected from abundant light has occasion- ally been found to contain virulent tubercle bacilli for as long as ten months. For a year at least it should be considered dangerous. The Rontgen rays have a deleterious effect on tubercle bacilli in cultures, but practically none upon those in tissues. Multiplication of Tubercle Bacilli in Nature Takes Place Only in the Living Animal. — The tubercle bacillus is a strict parasite — that is to say, its biological characters are such that it could scarcely find natural conditions outside of the bodies of living animals favorable for its multiplication. Under exceptional conditions, such as in freshly expectorated sputum, tubercle bacilli may increase for a limited time. Cultivation of the Tubercle Bacillus. — On account of their slow growth and the special conditions which they require, tubercle bacilli cannot be grown in pure culture by the usual plate method on ordinary culture media. Koch first succeeded in cultivating and isolating this bacillus on coagulated beef serum, which he inoculated by carefully rubbing the surface with sections of tuberculous tissue and then leaving CULTIVATION OF THE TUBERCLE BACILLUS 369 the culture, protected from evaporation, for several weeks in the incu- bator. Cultures are more readily obtained of human or avian than of bovine bacilli. Growth on Coagiilated Dog or Bovine Serum or on Egg. — On these, one of which is generally used to obtain the first culture, the growth is usually visible at the end of ten days at 37° C, and at the end of three or four weeks a dis- tinct and characteristic development has occurped. On serum small, grayish- white points and scales first appear on the surface of the medium. As develop- ment progresses there is formed an irregular, membranous-looking layer. On egg the growth is in the form of more or less elevated colonies which may become confluent. Growth on Nutrient 3 to 5 Per Cent. Glycerin Agar. — Owing to the greater faciUty of preparing and steriUzing glycerin agar, it is now usually employed in preference, to blood serum for continuing to produce later cultures. When numerous bacOli have been distributed over the surface of the culture medium, a rather uniform, thick, white layer, which subsequently acquires a slight yellowish tint, is developed; when the bacilli sown are few in number, or are associated in scattered groups, separate colonies are developed, which acquire considerable thickness and have more or less irregular outlines. The growth appears similar to that shown upon bouillon as seen in Fig. 140. Fig. 140. — Growth of tubercle bacilli upon glycerin bouillon. (Kolle and Wasserman.) Growth on Nutrient Veal or Beef Broth Containing 5 Per Cent, of Glycerin. — Glycerin broth is used for the development of tuberculin and must be neutral to litmus, viz., between 1.5 and 2 per cent, acid to phenol- phthalein. On these media the tubercle bacillus grows readily if a very fresh thin film of growth from the glycerin agar or a small piece of pellicle removed from a previous broth culture is floated on the surface. This continues to enlarge as long as it floats on the surface of the liquid, and in the course of three to six weeks covers it wholly as a single film, which on agitation is easily broken up and then settles to the bottom of the flask, where it ceases to develop further. The liquid remains clear. A practical point of importance, if a quick growth is desired, is to use for the new cultures a portion of the pellicle of a growing bouillon culture, which is very thin and actively increasing. Growth on Potato. — A good growth from cultures and sometimes even from tissue takes place on potato, and this forms the most uniform medium for stock cultures. 24 370 BACILLUS AND BACTERIOLOGY OF TUBERCULOSIS Obtaining of Pure Cultures of the Tubercle Bacillus from Sputum, Infected Tissue, and Other Materials. — On account of the time required and the difficulties to be overcome, this is never desirable except when careful investigations of importance are to be undertaken. Pure cultures can be obtained directly from tuberculous material, if the tubercle bacilli are present in sufficient number and mixed infection is not present, by using the proper blood serum or egg culture medium (p. 106); but it is difficult to get material free from other bacteria which grow much more rapidly and take possession of the medium before the tubercle bacillus has had time to form visible colonies. It is usually necessary first to inoculate guinea-pigs, subcutaneously or intramuscularly, preferably in the thigh, and then obtain cultures from the animals as soon as the tuberculous infection has developed. In this way, due to the susceptibility of the guinea-pig to tuberculosis, cultures may be obtained from material containing very few tubercle bacilli, although contaminating bacteria may be very numerous. Animals inoculated usually die at the end of three weeks to four months. It is better, however, not to wait until the death of the animals, but at the end of four to six weeks to kill a guinea-pig without violence, using illuminating gas, chloroform, or ether in a closed tin or jar. (Animals which develop tuberculosis acutely are apt to have abundant tubercle bacilli and give successful cultures, while the chronic cases usually have few bacilli and may give unsuccessful cultures.) The animal after being killed is tied out in trays, and after washing with a 5 per cent, solution of carbolic acid, immediately autopsied. The skin over the anterior portion of the body having been carefully turned back, the inguinal nodes are removed with fresh instruments. The nodes on the side of injection are especially favorable for cultures. The abdomen is then opened and the spleen and retroperitoneal nodes removed. As the organs are removed they should be placed in Petri dishes and thoroughly minced with knife and forceps. Fresh instru- ments should be used for each operation. The sternal nodes may be used for cultures, but the lungs are almost useless, as the majority of cultures will be contaminated. The minced tissue is then placed on the surface of the culture media, both egg and glycerin egg being used, and evenly and thoroughly smeared over its surface, then the cotton plug is dipped in hot paraffin to aid in keeping the media from drying. The tubes are incubated in an inclined position. On egg, growth is visible in from seven to ten days, and well marked at the end of three weeks. Many tubes should be inoculated, as it is only with the dexterity acquired by practice that contaminations are avoided. As will be noted further on, the growth of the bovine type will be very sparse and on glycerin egg probably negative. Cultures may also be obtained with a fair proportion of successful results by the antiformin method or the method of Petroff. In the former (see p. 397) the washed sediment is inoculated on egg media. Petroff (see p. 106) digests sputum or feces with equal amounts of 3 per PATHOGENESIS ^371 cent. NaOH for one-half hour at 37° C, neutralizes with dilute HCl and inoculates the sediment obtained by centrifugalizing, on his special gen- tian-violet egg medium. A successful culture of an acid-fast bacillus does not necessarily mean a tubercle bacillus. Van Winkle in our laboratory has isolated two non-pathogenic acid-fast strains in this way from sputa. Pathogenesis. — The tubercle bacillus is pathogenic not only for man, but for a large number of animals, such as the cow, monkey, pig, cat, etc. Young guinea-pigs are very susceptible, and are used for the detection of tubercle bacilli in suspected material. When inoculated with the minutest dose of the living bacilli they usually succumb to the disease. Infection is most rapidly produced by intraperitoneal injection. If a large dose is given, death follows in from ten to twenty days. The omentum is found to be clumped together in sausage-like masses which contain many bacilli. There is no serous fluid in the peritoneal cavity, but generally in both pleural sacs. The spleen is enlarged, and it, as well as the liver and peritoneum, contains large numbers of tubercle bacilli. If smaller doses are given, the disease is prolonged. The peritoneum and internal organs — spleen, liver, etc., and often the lungs — are then filled with tubercles. On subcutaneous injection, for instance, into the thigh, there is a thickening of the tissues about the point of inoculation, which may break down in one to three weeks and leave a sluggish ulcer covered with cheesy material. The neighboring lymph nodes are swollen, and at the end of two or three weeks may attain the size of hazel-nuts. Soon an irregular fever is set up, and the animal becomes emaciated, usually dying within four to eight weeks. If the injected material contained only a small number of bacilli the wound at the point of inoculation may heal up and death be postponed for a long time. At autopsy the lymphatic nodes are found to have undergone cheesy degeneration; the spleen may be very much enlarged, and throughout its substance, which is colored dark red, are distributed masses of nodules. The liver is also commonly increased in size, streaked brown and yellow, and the lungs are filled with grayish- white tubercles; but, as a rule, the kidneys contain no nodules. Oc- casionally the lesions are limited to the inguinal and retroperitoneal nodes. Tubercle bacilli are found in the affected tissues, but the more chronic the process the fewer the bacilli present. Injection into the thigh is to be preferred for diagnostic purposes, the swelling of the local lymph nodes being then palpable. As soon as this is appreciable the node may be removed with or without killing the pig, the presence or absence of tuberculous lesions noted, and smears made for the detection of tubercle bacilli, thus saving time. It must be remembered that the pig may not show the usual picture of gener- alized tuberculosis, but only a swelling of the local lymph nodes. For- tunately tubercle bacilli are usually easily demonstrable in smears made from the crushed nodes. If there is any doubt the remaining tissue should be emulsified and reinjected into a second set of pigs. Another point to be considered is that other organisms may, rarely, give a picture difficult to distinguish macroscopically from tuberculosis, 372 BACILLUS AND BACTERIOLOGY OF TUBERCULOSIS as, for instance, streptothrix. To safeguard against error smears should be stained and tubercle bacilli demonstrated. Chronic guinea- pig septicemia may be accompanied by lesions in the spleen which' might be taken for tuberculous lesions. Rabbits are very susceptible to tuberculosis of the bovine type, less so to that of the human type. This will be given more in detail under the differences between human and bovine tuberculosis. Monkeys are very susceptible to infection with both types of bacilli. Cats, dogs, rats, and mice are susceptible, the last two usually show no tuberculous lesions, but there is great multiplication of the bacilli in the tissues. Tubercle Toxins. — ^The tubercle bacillus produces no true toxins. The bodies of the bacteria contain substances which cause necrosis of tissue with subsequent caseation or abscess. In broth cultures, after filtration, are present substances which produce fever and inflam- matory reactions of tissues. These substances as well as extracts from tubercle bacilli are highly toxic for tuberculous animals, little if at all for normal animals, and cause fever only in the former and the tissue inflammation spoken of as marked about tuberculous lesions. These poisons will be considered in detail later in connection with tuberculins. Action upon the Tissues of the Poisons Produced by the Tubercle Bacillus. — Soon after the introduction into the tissues of tubercle bacilli, either Uving or dead, the cells surrounding them begin to show that some irritant is acting upon them. The connective-tissue cells become swollen and undergo mitotic division, the resultant cells being dis- tinguished by their large size and pale nuclei. A small focus of pro- liferated epithelioid cells is thus formed about the bacilli, and according to the intensity of the inflammation these cells are surrounded by a larger or smaller number of lymphoid cells. When living bacilli are present and multiplying the lesions progress, the central cells degenerate and die, and a cheesy mass results, which later may lead to the formation of cavities. Dead bacilli, on the other hand, unless bunched together give off sufficient poison to cause the less marked changes only (Prudden and Hodenpyl) . Of the gross pathological lesions produced in man by the tubercle bacilli the most characteristic are small nodules, called miliary tubercles. When young, and before they have undergone degeneration, these tubercles are gray and translucent in color, somewhat smaller than a millet seed in size, and hard in consistence. But miliary tubercles are not the sole tuberculous products. The tubercle bacilli may cause diffuse growth of tissue identical in structure with that of miliary tubercles, that is, composed of a basement substance, containing epithe- lioid, giant, and lymphoid cells. This diffuse tuberculous tissue also tends to undergo cheesy degeneration. Point of Entrance of Infection. — Infection by the tubercle bacillus takes place usually through the respiratory tract or the digestive tract, including the pharynx and tonsils, more rarely through wounds of the skin. PATHOGENESIS 373 Tuberculosis may be considered to be caused chiefly by the direct transmission of tubercle bacilli to the mouth through soiled hands, lips, handkerchiefs, milk, etc., or- by the inhalation of fine particles of mucus thrown off by coughing or loud speaking, or of dust contaminated by tuberculous sputum or feces. Tuberculosis of Skin and Mucous Membranes. — When the skin or mucous membranes are superficially infected through wounds there may develop lupus, ulceration, or a nodular growth. The latter two forms of infection are apt after an interval to cause the involvement of the nearest lymphatic nodes. Tuberculosis of Respiratory Tract. — ^The lungs are the most frequent location of clinically recognizable tuberculous inflammation. On account of their location they are greatly protected from external infection. Most of the bacilli are caught upon the nasal or pharyngeal mucous membranes. Only a small percentage can find their way to the larynx and trachea, and still less to the smaller bronchioles. From the examination of the lungs of miners as well as from experimental tests there is no doubt but that some of the bacilli may find their way into the deeper bronchi. The deeper the bacilli penetrate the more - unlikely that they can be cast out. On the other hand, the lungs are the most likely point of localization of tubercle bacilli which find their way into the blood stream. It is in this way that the lungs frequently become infected. It is now well established that infection taking place through the intestine may find its way by the blood to the lungs and excite there the most extensive lesions with or without leaving any trace of its point of entrance. Even if infection of the lung is slight or entirely absent the tubercle bacilli then find their way to the bronchial nodes where lesions may develop. Lesions either active or latent, any- where in the body may be the source of a subsequent pulmonary infec- tion. Should bacilli find their way into the blood, infection of the lung tissue will result if the resistance is lowered for any reason. The nasal cavities are rarely affected with tuberculosis, but more often the retropharyngeal tissue. Tuberculosis of this tissue as well as that of the tonsils is apt to give rise to infection of the lymph nodes of the neck. It is believed that just as bacilli may pass through the intestinal walls to infect the mesenteric nodes, so bacilli may, without leaving any trace, pass through the tonsils to the nodes of the neck. Primary infection of the larynx is rare. Secondary infection is fairly common. The region of the vocal cords and the interarytenoid space are the special sites attacked. Infection by Inhalation of Dried and Moist Bacilli. — ^A common mode of infection is by means of tuberculous sputum, which, being coughed up by consumptives, is either disseminated as a fine spray aiid so inhaled, or, carelessly expectorated, dried and, broken up by tramping over it, sweeping, etc., distributes numerous virulent bacilli in the dust. As long as the sputum remains moist there is no danger of dust infection, but only of direct contact; it is when it becomes dry, as on handkerchiefs, bedclothes, and the floor, etc., that the dust is a source of danger. 374 BACILLUS AND SACTERIOLOGX OF TUSERCVLOSiS A great number of the expectorated and dried tubercle bacilli undoubtedly die, especially, as we have said, when acted upon by direct sunlight; but when it is considered that as many as five billion virulent tubercle bacilli may be expectorated by a single tuberculous individual in twenty-four hours, it is evident that even a much smaller proportion than are known to stay alive will suffice in the immediate vicinity of consumptives to produce infection unless precautions are taken to prevent it. The danger of infection is greatest, of course, in the close neighborhood of tuberculous patients who expectorate profusely and indiscriminately, that is, without taking the necessary means for pre- venting infection. We found that of 100 tuberculous men admitted to one of the consumption hospitals, only 20 claimed to have taken any care to prevent the contamination of their surroundings by their sputum. There is much less danger of infection at a distance, as in the streets for instance, where the tubercle bacilli have become so diluted that they are less to be feared. In rooms the sputum is not only protected from the direct sunhght, but it is constantly broken up and blown about by the walking, closing of doors, etc. In crowded streets on windy days infected dust must sometimes be in the air unless the expectoration of consumptives is controlled. Exhaustive experiments made by many observers have shown that particles of dust collected from the immediate neighborhood of con- sumptives, when inoculated into guinea-pigs, produce tuberculosis in a considerable percentage of them; whereas, the dust from rooms inhabited by healthy persons or dust of the streets does so only in an extremely small percentage. Fliigge is probably right in thinking that the dust which is fine enough to remain for a long time in sus- pension in the air is usually free from living bacilli. It is in the coarser though still minute particles, those in which the bacilli are protected by an envelope of mucus, that the germs resist drying for considerable periods. These are carried only short distances by air currents. Such reports are those given by Straus, who, on examining the nasal secretions of twenty-nine healthy persons living in a hospital with con- sumptive patients, found tubercle bacilli in nine of them, must be accepted with some reserve, since we know that in the air there are bacilli which look and stain like tubercle bacilli and yet are totally different. It may be said that the danger of infection from slight contact with the tuberculous is not so great as it is considered by many, but that on this account it is all the more to be guarded against in the immediate neighborhood of consumptives. Those who are most liable to infection from this source are young children; the adult members of the family, the nurses, the fellow-workmen, and fellow-prisoners of persons suffering from the disease are relatively very less likely to become infected. In this connection, also, attention may be drawn to the fact that rooms which have been recently occupied by consumptives are not infrequently the means of producing infection (as has been clinically and experimentally demonstrated) from the deposition of tuberculous dust on furniture, walls, floors, etc. The danger is not apt to last beyond PATHOGENESIS 375 three months. Fliigge has drawn attention to the fact that in coughing, sneezing, etc., very fine particles of throat secretion containing bacilli are thrown out and carried by air currents many feet from the patient and remain suspended in the air for a considerable time. To encourage us, however, we now have a mass of facts which go to show that when the sputum is carefully looked after there is very little danger of infecting others except by close personal contact. Tuberculosis of Digestive Tract. — ^Tuberculosis of the gums, cheeks, or tongue is rare. The tonsils and pharynx are somewhat more often involved. The stomach and esophagus are almost never attacked. The small intestines are rather frequently the seat of infection from bacilli swallowed with the food or dust-infected mucus. In a striking case four previously healthy children died within a short period of one another. Their nurse was found to have tuberculosis of the antrum of Highmore, with a fistulous opening into the mouth. She had the habit of putting the spoon with which she fed the children into her mouth so as to taste the food before it was given to them. As already noted, the bacilli frequently pass through the mucous membrane of the gastro-intestinal canal to the lymph glands with- out leaving any lesions. Infection by Ingestion of Milk and Milk Products. — Milk serves as a conveyor of infection, whether it be the milk of tuberculous mothers or the milk of tuberculous cows. In this case evidence of infection is usually shown in the mesenteric and cervical lymph nodes or general- ized tuberculosis may be caused, while the intestinal walls are fre- quently not affected. Bacilli accompanied by fat pass much more readily through the intestinal mucous membrane or that of the tonsils and pharynx. The transmission of tubercle bacilli by the milk of tuberculous cows has been abundantly proved. Formerly it was thought that in order to produce infection by milk there must be a local tuberculous affection of the udder; but it is now known that tubercle bacilli may be found in the milk in small numbers, when adjacent tissue is infected and when careful search fails to detect any udder disease. Schroder has shown that the feces are a very dangerous factor in the dissemination of tubercle bacilli. He compares feces in cattle to sputum in man, since the tubercle bacilli are swal- lowed by cattle and are to a great extent passed through the intestinal tract without destruction. He found that when milk from phthisical cows having healthy udders was obtained so as not to become infected by feces it was free from bacilli, but when obtained without special precautions it was frequently infected. The milk of every cow which has any well-developed internal tuberculous infection must therefore be considered as possibly containing tubercle bacilli. Rabinowitsch, Kempner, and Mohler also proved beyond question that not only the milk of tuberculous cattle, which showed no appreciable udder disease, but also those in which tuberculosis was only detected through tuber- culin, frequently contained tubercle bacilli. Different observers have found tubercle bacilli in 10 to 30 per cent, of the samples of unheated 376 BACILLUS AND BACTERIOLOGY OF TUBERCULOSIS city milk. Butter may contain tubercle bacilli in higher percentages of samples examined. When we consider the prevalence of tuberculosis among cattle we can readily realize that, even if the bovine bacillus infects human beings with difficulty, there is danger to children when they are exposed to this source of infection. The milk from cattle suffering from udder tuberculosis usually contains a few hundred bacilli per cubic centimeter, but may contain many millions. It is also impor- tant to mention the fact that mixed milk from a herd, though tending to dilute the milk of cows excreting tubercle bacilli, may be badly infected from one cow, especially if this cow has udder disease. Taking the abattoir statistics of various countries, we find that about 10 per cent, of the cattle slaughtered were tuberculous. A less probable source of infection by way of the intestines is the flesh of tuberculous cattle. Here the danger is considerably less, from the fact that meat is usually cooked, and also because the muscular tissues are seldom attacked. In view of the finding of the bovine type of bacilli in a considerable percentage of the cases of tuberculous children tested, the legislative control and inspection of cattle and milk is an absolute necessity. As a practical and simple method of preventing infection from suspected milk, sufficient heating of the milk used as food must commend itself to all. Human tubercle bacilli may be found in milk as instanced by feeding them one sample out of a series of city milks examined in the Research Laboratory by Hess. Method of Examining Milk for Tubercle Bacilli. — Thirty cubic centi- meters of milk are centrifuged at high speed and 10 c.c. of the lower milk and sediment collected. Four cubic centimeters of the cream is thinned with a little sterile water and injected into two guinea-pigs. The sedi- ment is injected in amounts of 3 to 5 c.c. into other pigs. Larger amounts than this are apt to kill too many pigs from the associated bacteria. Subcutaneous injection is to be preferred. There are certain precau- tions that must be taken in drawing conclusions, as the different types of acid-fast "butter bacilli" may cause lesions, and their presence will be noted in smears made from these lesions. To avoid this source of error, two methods are resorted to. If cultures are made from the sus- pected lesions on glycerin agar, these bacilli usually develop in a few days, whereas tubercle bacilli do not. When one is ready to kill the pigs, 2 c.c. of old tuberculin should be injected into each pig late in the day. The following morning the tuberculous pigs will be dead or dying. Autopsies should be done on all to confirm the test. The milk should be as fresh as possible to prevent the growth of other bacteria. Bovine Infection in Man. — Numerous investigations have been made on this point. To Ravenel probably belongs the credit of isolating the first bovine bacillus from a child. The following tables summarizing the results of a large series of cases give a fair idea of incidence of such infection. As will be seen, children are especially the ones infected, and usually the point of entry is clearly alimentary, as shown by the lesions. Cervical adenitis and abdominal tuberculosis are the most frequent types of infection. Generalized tuberculosis due to bovine PATHOGENESIS 377 infection is less frequent. Bone and joint tuberculosis is usually of the human type. The meninges are less commonly affected by the bovine type than by the human type. Infection of adults is very uncommon; and, though cases of pulmonary tuberculosis due to the bovine type of bacillus have been reported, such cases are rare. Table I. — Tabulation op Cases Reported.^ Diagnosis. Adults sixteen years and over. Human Bovine. Children five to sixteen years. Human. Bovine. Children under five years. Human. Bovine, Pulmonary tuberculosis . . '. Tuberculo.us adenitis, axillary Tuberculous adenitis, cervical Abdominal tuberculosis Generalized tuberculosis, alimentary origin Generalized tuberculosis Generalized tuberculosis, including meninges, alimentary origin Generalized tuberculosis, including meninges Tubercxilar meningitis .... Tuberculosis of bones and joints . Genito-urinary tuberculosis Tuberculosis of skin .... Miscellaneous cases: Tuberculosis of tonsils Tuberculosis of mouth and cervical nodes Tuberculous sinus or abscess Sepsis, latent bacilli .... 497 2 27 15 6 27 31 16 9 17 7 3 4 10 1 31 1 4 Totals 635 14 85 14 37 28 2 9 9 16 56 51 2 20 11 11 13 3 10 201 51 Mixed or double infections : 10 cases. Pulmonary tuberculosis, 20 years. 27 years. Abdominal tuberculosis, 70 years. Generalized tuberculosis (alimentary origin). 5i years. 18 years. 30 years. Generalized tuberculosis, 9 months. Generalized tuberculosis, including meninges (alimentary origin). Generalized tuberculosis, including meninges. years. years. i^Y years. Sputum, human and bovine types. Sputum, human and bovine types. Mesenteric nodes, human type. Retroperitoneal, human and bovine types. Spleen, human type. Mesenteric node, bovine type. Lung, culture not obtained. Mesenteric nodes, human and bovine types. Bronchial node, human type. Mesenteric nodes, human and bovine types. Bronchial nodes, bovine type. Mesenteric nodes, human type. Meninges, human type. Bronchial nodes, human type. Mesenteric nodes, bovine type. Bronchial nodes \ injected together. Spleen / bovine and human types. Meninges, human type. Lung, human type. Mesenteric nodes, human type. , Bronchial nodes, human and bovine types. Total cases, 1033. 1 Summary of cases reported up to July, 1912, exclusive of cases examined at the Research Laboratory (see Table II). In contrast to the next table, the above contains a large percentage of selected cases of alimentary types of tuberculosis, many of which showed only slight lesions. 378 BACILLUS AND BACTERIOLOGY OF TUBERCULOSIS Table II. — The Relative Peoportion op Human and Bovine Tubercle Bacilli Infections in a Large Series of Unselected Cases' Examined at the Research Laboratory. Diagnosis of Adults sixteen years and over. Children five to sixteen years. Children under five years. examined. Human. Bovine. Human. Bovine. Human. Bovine. Notes. Pulmonary tuberculosis 281 - 8 - 7 - Clinical diagnosis only known and therefore no positive details as to the extent of lesions elsewhere. Tuberculous adenitis, in- guinal and axillary 1 " 4 " " " See next. Tuberculous adenitis, cer- vical ■ 9 19 8 6 13 In two oases cultures were from axillary nodes but the primary focus was cervical. Another case died shortly afterward with pulmonary tubercu- losis. Abdominal tuberculosis 1 - 1 1 1 3 Milk supply of one child subsequently examined. Tubercle bacilli isolated. Generalized tuberculosis, alimentary origin 1 2 Only three cases given under this heading. Many of the cases in the following sub- divisions showed marked intestinal lesions and some possibly were of alimentary origin. Generalized tuberculosis 2 — 1 - 18 4 One bovine case had tuber- culous osteomyelitis of the metatarsal bone. Generalized tuberculosis including meninges 1 ~ 25 1 Tuberculous meningitis 1 - 2 - 26 2 No autopsy. Extent of lesions elsewhere unknown. Tuberculosis of bones and joints Genito-urinary tuberculosis 1 6 1 10 1 - 7 - The adult bovine case was tuberculosis of kidney. . Removal of kidney. Complete recovery. Tuberculosis of skin 1 - - - - - Tuberculous abscess 1 - - *• - - Possibly primary in bone. Totals 305 1 46 9 91 25 Double infection in one case. Both types isolated. Generalized tuberculosis including meninges, thirteen months. Mesenteric nodes gave human type. Meningeal fluid gave bovine type. Total oases, 478. A careful study of all the factors leads us to estimate that with the average raw milk supply about 10 per cent, of all deaths caused by tuberculosis in children under five is due to bovine infection. I Unselected cases from the hospitals of New York City. For full resum^ and discussion of results see Park and Krumwiede, Jour. Med. Res., vols, xxiii, xxv and xxvii. MIXED INFECTION 379 The tables on pages 377 and 378 give a summary of the results obtained in the larger investigations so far carried out. Hypothesis of Transmissibility of Tubercle Bacilli to the Fetus. — The transmission of tubercle bacilli from the mother to the fetus in animals occurs occasionally. With regard to tuberculosis in the human fetus the evidence is not so clear, though some 20 cases have been recorded of tuberculosis in newly born infants, and about a dozen cases of pla- cental tuberculosis. As to the infection of the fetus from the paternal side, where the father has tuberculosis of the scrotum or seminal vessels, we have no reason to suppose that such can occur. There are, how- ever, grounds for belief that infection in this way may take place from husband to wife. Attenuation. — Tubercle bacilli when subjected to deleterious in- fluences slowly decrease in virulence. Some strains lose their virulence when artificially cultivated for some time; some quickly, some slowly. Others retain their virulence indefinitely. Mixed Infection. — ^In regions where tuberculous processes are on the surface, such as skin infections, and also when the infection itself is multiple, as in diseases of the glands of the neck from tonsillar absorp- tion, there are frequently associated with the tubercle bacilli one or more other varieties of organisms. Those of most importance are the streptococcus, pneumococcus and influenza bacilli. While the influence of this secondary or mixed infection is not exactly known, ye;t both the local and systemic effects are undoubtedly unfavorable. In regard to pulmonary tuberculosis, it should be remembered that Baldwin has shown that caseation, ulceration and cavity formation may be produced experimentally in the lungs of animals by the tubercle bacillus alone. Further, it has been found that fever, emaciation and other characteristic symptoms of tuberculosis may be caused by the tubercle bacillus independently of any other associated microorganisms. The preponderance of the evidence supports the view that the lesions and symptoms in this type of infection may be caused by the tubercle bacillus alone, but not infrequently secondary organisms may contribute to the more severe symptoms or may be largely responsible for the unfavorable progress of the disease. Individual Susceptibility. — It was believed by many that in demonstrat- ing that tuberculosis was due to a specific bacillus that its occurrence was sufficiently explained; but they left out another important factor in the production of disease — individual susceptibility. That this susceptibility, or "predisposition," as it is improperly called, may be either inherited or acquired is now an accepted fact in medicine. It has even been thought that the physical signs and characters — the phthisical habit — which indicate this susceptibility can be externally recognized. At first the inherited susceptibility was considered more important than the acquired, but now much that was attributed to the former is known to be explained by the fact of living in an infected area. The acquired susceptibility may arise from faulty physical development or from depression, sickness, overwork, excessive use of alcohol, etc. 380 BACILLUS AND BACTERIOLOGY OP TUBERCULOSIS Unquestionably, vast differences exist in different individuals in the intensity of the tuberculous process in the lung. That this does not depend chiefly upon a difference in virulence of the infection is evident from the fact that individuals contracting tuberculosis from the same source are attacked with different severity, and that there is, as a rule, no great difference in degrees of virulence for animals in the tubercle bacilli obtained from different sources. The possibility of favorably in- fluencing many cases of an existing tuberculosis by treatment also proves that, under natural conditions, there is a varying susceptibility to the disease. Clinical experience teaches, likewise, that good hygienic con- ditions, pure air, good food, freedom from care, etc., increase resistance and are aids to recovery. Animal experiments have shown that not only are there differences of susceptibility in various animal species, but also an individual susceptibility in the same species. The doctrine of indi- vidual susceptibility, therefore, is seen to be founded on fact, although the reasons for it are only partially understood. Certain infectious diseases reduce the resistance to tuberculous infection, the most note- worthy example being measles. Tuberculosis Immunity. — ^As in other infectious diseases, various attempts have been made to produce an artificial immunity against tuberculosis, but the results so far have been disappointing. The expectation that the immunological mechanism operating in other bac- terial diseases would also be found in tuberculosis has not been realized. As a consequence of recent researches, we are coming to the opinion that the usual manifestations of humoral immunity have a secondary impor- tance in this disease and that to the tuberculous focus we must ascribe a greater role than has been given it in the past. Tuberculous infection presents certain phases which find no close analogy in other infectious processes. Several facts of fundamental importance have developed from the newer pathological and bacteriological investigations. An infection by the tubercle bacillus may take place, and foci of considerable size may develop without the production of sufficient disturbance to the bodily welfare to attract attention. Further, while. a frank tuberculous infection may come to a clinical cure, yet the focus remains,, and, even though completely walled off, may harbor virulent tubercle bacilli during the whole of the individual's normal life. It has now become established that a healed tuberculous focus present in the body, con- taining as it does bacillary protein, exerts an appreciable influence on the behavior of the body to subsequent infection. In the considera- tion of tuberculosis immunity, therefore, we must carefully bear in mind the distinction between tuberculous infection and tuberculous disease. Koch was the first to discover that animals already infected with living bacilli reacted differently to an injection of tubercle bacilli than did normal animals. When virulent tubercle bacilli are injected into a healthy animal, tubercles develop at or near the point of inoculation and the infection then progresses and the tubercle bacilli are carried to the spleen, liver, lungs and the intermediate glands with the formation of foci TUBERCULOSIS IMMUNITY 381 in these various organs. As a consequence, the animal finally succumbs to generalized tuberculosis. In the case, however, of an animal aheady tuberculous, the inoculation with virulent tubercle bacilli is followed by a quite different sequence of events. Shortly after such an inoculation there is a marked inflammatory reaction at the point of injection, followed by necrosis and possibly sloughing but with no advance of the infection beyond the point of inoculation. In other words, the animal suffers from a local toxic process but not from a true infection. Romer has given further details of this phenomenon. He found that if a small dose of tubercle bacilli be given at the second inoculation the local reaction soon subsides and healing results. But, on the other hand, if a large quantity of tubercle bacilli be injected the local process goes on to necrotic sloughing and the animal soon dies of cachexia. The tuber- culous animal, therefore, when subjected to an injection of tubercle bacilli is not truly infected but suffers from an intoxication, the degree of which depends upon the quantity of tubercle bacilli introduced. Recent studies in tuberculosis emphasize the fundamental importance of Koch's and Romer's observations, and have forced a radical revision in our theories of tuberculous infection and immunity. By means of the various tuberculin reactions, particularly the intracutaneous of Mantoux and cutaneous of von Pirquet, and by the most careful postmortem studies on a great number of individuals, we now know that a large majority of human beings are infected with the tubercle bacillus and develop demonstrable tuberculous lesions before they reach the age of eighteen. A few children thus infected may succumb to the disease but by far the greater number show no conspicuous signs of the infection, their foci heal, and without the tuberculin reaction or the opportunity of examination after death, the presence of tuberculous infection would pass unsuspected and undetected. These recent additions to our knowledge have given us a new and apparently a truer conception of the factors underlying tuberculous infection. It is quite probable that the majority of cases of pulmonary tuberculosis developing after the eighteenth year of age are caused not by infection from without but by the breaking down of an encapsulated focus acquired and healed during childhood. The greater morbidity during the age group of about twenty to thirty may be accounted for by the fact that this is usually the period of greatest physical and mental strain and stress and that it is this stress which breaks down the body's resistance to an infective focus, and which transforms a latent infection into active disease. Although it is evident that the presence of a healed focus protects . the individual to a certain degree against a subsequent infection by the tubercle bacillus, it is equally apparent that the exacerbations suffered by healed cases show that the protection acquired is at best only a relative one. We must therefore conclude tha there exists in the human race no absolute immunity to tuberculosis, and that such increased resistance as we may possess is gained only at the hazard of an early infection. 382 BACILLUS AND BACTERIOLOGY OF TUBERCULOSIS Immunization. — Koch, reasoning from his observations on the ability of the tuberculous animal to resist subsequent infection, attempted to increase this resistance by the injection of certain modified products of the tubercle bacillus. To this end he prepared his original tuberculin, . or "0. T." which, however, failed to fulfil his expectations. With the development of our modern theories of bacterial immunity the hope arose that protection might be actively acquired through the injection of the tubercle bacillus either' attenuated or dead or of its products by way of stimulating the body to the production of specific antibodies; or, again, that protection might be passively conferred by the adminis- tration of the serum of animals made immune to the tubercle bacillus. From observations on experimental animals and from serological tests on individuals treated with the various tuberculins, we now know that the injection of antigens derived from the tubercle bacillus fails to call forth in the treated individual any marked response in the way of the produc- tion of demonstrable bodies and that serum therapy is of little or no avail. It is true that in some cases resistance to the disease can be raised within certain limits, and that, in man, clinical cures may sometimes be effected by the aid of such therapeutic agents as tuberculin. Yet in these cases we find little or no evidence of the production of agglutinins, precipitins, lysins or complement-fixing antibodies. We are, conse- quently, forced to the conclusion that such immunity as has been established must have been the result of a physiological mechanism differ- ing from the processes operating in other infectious diseases, and recent investigations, particularly those of Krause, point to the tuberculous focus as the prime factor in the production of such immunity as develops in tuberculous disease. Experimental analyses of the physiological action of tuberculin have shed new light on the problem. That it is the tuber- culous focus which, in a large measure, determines the body's reactivity to tuberculin is shown by the fact that normal individuals, that is, individuals free from tuberculous infection, can tolerate the adminis- tration of a relatively large amount of tuberculin without exhibiting any appreciable symptoms. Tuberculin, therefore, is in itself non-toxic and the characteristic reaction following its application to the tuber- culous body must be looked upon as an allergic "phenomenon. Its manifestations are threefold: there is a local reaction at the point of application, a focal reaction at the site of infection and a general con- stitutional reaction. The local reaction appears as a more or less non- infective inflammation in the skin or on the mucous membranes, depend- ing upon the point of application, and it is this phase of the reaction which has been so profitably utilized for the diagnosis of tuberculous infection (see "Tuberculin as a Diagnostic Aid," below). The focal reaction consists in vascular changes at the tuberculous lesion. There is a hyperemia with a consequent softening of the focus, and a liberation of antigenic or toxic focal products. Krause has shown that these products are inherently toxic and their liberation into the blood stream may occasion the fever, the malaise and the other symptoms character- TUBERCULIN IN DIAGNOSIS AND THERAPY 383 istic of tuberculin intoxication. It is likely that in addition to t^is the antigen of the tuberculin may react with such immune bodies as may be present in the infected, and therefore hypersensitive, body and thus contribute to the general reaction. If the dose of tuberculin is not too great the hyperemia at the focus is transient and the body responds with an increased cellular activity at the focus, resulting, under favorable conditions, in a further prolifera- tion of connective tissue and a more complete encapsulation of the tuberculous lesion. Should the dose of tuberculin, on the other hand, be excessive or the bodily resistance be deficient the focal reaction may lead to a softening and breaking down of the lesion with consequent bleeding, dissemination of the liberated tubercle bacilli and extension of the lesion. Inasmuch as the internal location of the lesion frequently precludes an opportunity for observing the focal reaction we must look to the general constitutional manifestations as an index of the degree of reaction produced. The goal of tuberculin treatment, therefore, is the stimula- tion of the focus by doses of tuberculin so graded and adjusted to the sensitiveness of the individual that the body's ability to respond to and control the focal hyperemia is not overtaxed. In this way cell pro- liferation, encapsulation and ultimate healing of the focus may be promoted. Tuberculin in Diagnosis and Therapy. — ^From the foregoing it can be readily seen that by taking advantage of the allergic or hypersensi- tive state in tuberculosis we are able to use tuberculin both as a diagnostic aid and as a therapeutic agent. As a Diagnostic Aid. — ^The presence of a tuberculous focus in the body, by some mechanism not yet wholly understood, sensitizes the skin and mucous membranes in such a way that the application of tuberculin to these tissues causes a more or less severe inflammatory reaction at the point of application. The preparation used is the Old Tuberculin ("O. T.") of Koch and the following represents the methods of applica- tion in their order of sensitiveness as described by Hamman. 1. Intracutaneous test of Mantoux. 2. Subcutaneous-local test ("Stich Reaktion"). 3. Cutaneous test of von Pirquet. 4. Subcutaneous test. 5. Percutaneous test of Moro. 6. Conjunctival test of Calmette. The Intracutaneous Test of Mantoux. — ^While more difficult to carry out than the cutaneous test, the intracutaneous is the more delicate, and is the most widely used. The test is carried out as follows: "the inner surface of the forearm is cleansed with alcohol, then with ether, the skin is drawn taut, and the diluted tuberculin is injected from a tuberculin syringe (1 c.c. graduated in 50ths or lOOths) through a fine needle (a No. 26 preferably) which has been carefully inserted into, but not under, the skin. The total volume injected should be 0.1 c.c, and it is advisable to employ several dilutions of tuberculin in order to determine the degree of hypersensitiveness of the patient's skin. Usually four 384 BACILLUS AND BACTERIOLOGY OF TUBERCULOSIS simultaneous injections are given, using dilutions of old tuberculin oi 1 to 10,000,000, 1 to 1,000,000, 1 to 100,000, and 1 to 10,000, repre- senting respectively, 0.0000001, 0.000001, 0.00001 and 0.0001 gram of tuberculin. Separate sterile syringes should be used for each dilution and 0.1 c.c. of sterile salt solution should be similarly injected as a control. (See Methods of Diluting Tuberculins.) The reaction when positive appears in six to eight hours, reaches its maximum in twenty-four to forty-eight hours and generally subsides in six to ten days, and consists of infiltration, hyperemia, and, in severe reactions, vesiculation. The width of the area of infiltration and the degree of inflammation are noted. It not infrequently happens that a person fails to give a positive reaction at the first test yet shows the typical local manifestations when, the test is repeated. This has given rise to the impression that an injection of tuberculin sensitizes the individual. Without further discussion it may be stated that, as far as we know, there is no skin sensitiveness without infection. The appearance of a positive reaction at the second injection may be looked upon as a true reaction, and it is likely that the first injection while eliciting no response in the skin has served to stimulate the latent hypersensitiveness of the cells. The Subcutaneous-local Test or "Stich Reaktion."— The test is carried out as in the intracutaneous test with the difference that the needle is inserted into the subcutaneous tissue with the point of the needle directed toward the surface. The intracutaneous test is preferable. The Cutaneous Test of von Pirquet. — ^This is performed as follows : The inner side of the forearm is cleansed with alcohol and ether and two small similar scarifications or scratches are made about three inches apart. Oozing of blood is to be avoided. On one spot or scratch a drop of tuberculin is placed and allowed to dry on the scarification. The tuber- culin may be diluted to 25 per cent, if desired. The other spot is kept as a control. Both spots should be examined at the end of twelve, twenty-four, and thirty-six hours. A positive reaction appears after three to twenty-four hours and is usually at its height at thirty-six to forty-eight hours, and consists then of a slightly raised reddening of the skin somewhat circular in outline and usually about 10 mm. in diameter. Reactions under 5 mm. in diameter should be regarded as doubtful. The Subcutaneous Test.— The object of this test was to elicit a con- stitutional reaction by the injection of old tuberculin under the skin. Owing to the severe reactions frequently obtained which resulted in harm to the patient it is now considered the better practice to abandon this test in favor of the Mantoux or von Pirquet methods. The Percutaneous Test of Moro.— An ointment is made of equal parts of lanolin and tuberculin ("O. T."), and a small amount of it is rubbed into the skin on the chest. A positive reaction is shown by the develop- ment of reddening and papules. Ophthalmic Test of Calmette.— Owing to the occurrence of serious accidents this test is little used. A drop of a 2 per cent, solution of tuberculin is applied to the lower conjunctival sac. The reaction is indicated by secretion and reddening of the inner canthus caruncle or lower lid, which may include the entire conjunctiva, with edema of the lids. KINDS AND PREPARATION OF TUBERCULIN 385 Deductions and Limitations of the Tuberculin Test for Diagnosis. — A positive reaction indicates the presence of a tuberculous focus but not necessarily of tuberculous disease. It tells nothing of the location, extent or activity of the lesion. Krause gives the following basis for interpreting the reaction: There is no cutaneous hypersensitiveness without a focus (tubercle); this hypersensitiveness appears coincident with the establishment of the focus; it diminishes with the healing of the focus; it varies directly with the intensity of the disease. It should be mentioned, further, that many advanced cases, particularly those in cachexia, fail to show any appreciable response to tuberculin tests, and also that measles diminishes hypersensitiveness. Tuberculin: Kinds and Preparation. — There exist a large number of tuberculin preparations. The following, however, hold the highest favor and suffice for the needs of the diagnostician or practitioner: Tuberculin, Koch's "Old" (O. T.). — Cultures of tubercle bacilli after six weeks' growth on 5 per cent, glycerin broth^ are heated in the Arnold sterilizer to kill the bacilli, and filtered. This bacillus-free filtrate is evaporated to one-tenth its original bulk and after filtering through paper to remove the sediment it is ready for use. The tuberculin is therefore a heated 50 per cent, glycerin solution of the products of the bacilli in the culture fluid and such portions of the bacilli as go into solution. It is used chiefly for diagnostic purposes. Tuberculin B. F. (Bouillon Filtrate of Denys) is made from a broth culture as above. The culture is not heated, but filtered first through paper and then through a Berkefeld filter to insure sterility. It differs from the original tuberculin in that it is neither heated nor concentrated, and contains only such constituents of the bacillus as are soluble or are developed in the culture medium during cultivation. Tuberculin B. E. (Bacillus Emulsion) is produced by grinding up dried tubercle bacilli until no intact bacilli can be found on microscopic examination. This powder is then suspended in glycerin-water and heated to 60° C. for an hour or more to kill any viable tubercle bacilli. The proportion of bacillus to water is such that 1 c c. of fluid contains 5 mg. of bacillus substance. This preparation contains all the con- stituents of the tubercle bacillus in an unchanged state, and therefore corresponds to a bacterial vaccine. The undiluted products keep for a long time if kept cool and pro- tected from the light. The low dilutions keep for at least one month, but the high dilutions should not be used after two weeks. Method of Diluting Tuberculin. — The following describes a method of dilu- tion in terms of volume of finished tuberculin, and gives the weight equivalent. Although in the case of bouillon filtrate many use dilutions, taking into consider- ation the fact that the bouillon filtrate is not concentrated, or in the case of bacillus emulsion dilute according to the weight of solid substance contained, a uniform method for each seems advisable for reasons of simplicity. If we consider the finished product in terms of cubic centimeters or grams regardless of the contents, the following is the method of dilution. Dilutions should be made with sterile saline to which 0.25 per cent, carbolic has been added:^ 1 Broth should be made 1.5 per cent, acid to phenolphthalein. 25 386 BACILLUS AND BACTERIOLOGY OF TUBERCULOSIS Dilu- tions. A B C D E F Amount of tuber- culin. 1 c.c. 1 c.c. of dilution A 1 c.c. of dilution B 1 c.c. of dilution C 1 c.c. of dilution D 1 c.c. of dilution E Amount of diluent. 9 c.c. 9 c.c. 9 c.c. 9 c.c. 9 CO. 9 c.c. Content of tuberculin terms of finished product. Using bacillus emulsion content of solid substance in eat^h dilution will be: 1 c.c. = 0.1 c.c. or gm., or 100 c.mm. or mg. 1 c.c. = 0.01 c.c. or gm., or 10 c.mm. or mg. 1 c.c. = 0.001 c.c. or gm., or 1 c.mm. or mg. 1 c.c. = 0.0001 c.c. or gm., or 0.1 c.mm. or mgm. 1 c.c. = 0.00001 c.c. or gm., or 0.01 c.mm. or mgm. 1 c.c. = 0.000001 c.c. or gm., or 0.001 c.mm. or mgm. 1 c.c. = 0.5 mgm., or 0.5 mgm. 1 c.c. = 0.05 mgm., or 0.05 mgm. 1 c.c. = 0.005 mgm., or 0.005 mgm. 1 c.c. = 0.0005 mgm., or 0.0005 mgm. 1 c.c. = 0.00005 mgm., or 0.00005 mgm. 1 c.c. = 0.000005 mgm., or 0.000005 mgm. Tuberculin Treatment. — Tuberculin is not a cure for tuberculosis. It promotes healing and relapses are less frequent after its use. It should be used as an addition not as a substitute for the recognized methods of treatment. It is a two-edged weapon and should be employed only by those who have a thorough understanding of its possibilities for good, and unfortunately, for harm.^ United States Government Directions for Inspecting Herds for Tuberculosis. — "Inspection should be carried on while the herd is stabled. If it is necessary to stable animals under unusual conditions or among surroundings that make them uneasy and excited, the tuberculin test should be postponed until the cattle have become accustomed to the conditions they are subjected to, and then begin with a careful physical examination of each animal. This is essen- tial, because in some severe cases of tuberculosis, on account of saturation with toxins, no reaction foUows the injection of tuberculin, but experience has shown that these cases can be discovered by physical exaroination. This should include a careful examination of the udder and of the superficial lymphatic glands, and auscultation of the lungs. "Each animal should be numbered or described in such a way that it can be recognized without difiiculty. It is well to number the stalls with chalk and transfer these numbers to the temperature sheet, so that the temperature of each animal can be recorded in its appropriate place without danger of con- fusion. The following procedure has been used extensively and has given excellent results: "(a) Take the temperature of each animal to be tested at least twice, at intervals of three hours, before tuberculin is injected. "(&) Inject in the evening, preferably between the hours of six and nine, 0.4 c.c. of Koch's tuberculin previously diluted to 5 c.c. with sterile water. The injection should be made with a carefully sterilized hypodermic syringe. The most convenient point for injection is back of the left scapula. Prior to the injection the skin should be washed carefully with 5 per cent, solution of car- bolic acid or other antiseptic. "(c) The temperature should be taken nine hours after the injection, and temperature measurements repeated at regular intervals of two or three hours until the sixteenth (eighteenth)^ hour after the injection. "(d) When there is no elevation of temperature at this time the examination may be discontinued; but if the temperature shows an upward tendency, meas- Tor details of tuberculin treatment see Haman and WoUman, "Tuberculin in Diagno- sis and Treatment," Appleton, New York, 1912. 2 The directions allow temperatures to be stopped the sixteenth hour, but even when there is no reaction at all it is much safer to always take temperatures for eighteen hours. We have found now and then a tuberculous cow that reacted on the eighteenth hour for the first time. : TUBERCLE BACILLUS OF DOMESTIC ANIMALS 387 urements must be continued until a distinct reaction is recognized or until the temperature begins to fall. " (e) If a cow is in a febrile condition tuberculin should not be used, because it would be impossible to determine whether, if a rise of temperature occurred, it was due to the tuberculin or to some transitory illness. " (/) Cows should not be tested within a few days before or after calving, for experience has shown that the result at these times may be misleading. "(g) In old, emaciated animals and in retests, use twice the usual dose of tubercuUn, for these animals are less sensitive. " (h) Condemned cattle must be removed from the herd and kept away from those that are healthy. " (t) In making postmortems the carcasses should be thoroughly inspected, and all the organs should be examined." Antituberculous Serum. — Every conceivable way of obtaining the true products of the tubercle bacilli has been tried, so as to cause the injected animals to produce antibodies both antitoxic and bactericidal. In spite of much conflicting testimony, it is probably safe to assert that no serums now obtainable have any great value. Prophylaxis. — All energies should be directed to the prevention of tuberculosis, not only by the enforcement of pi-oper sanitary regulation as regards the care of sputum, milk, meat, disinfection, etc., but also by continued experimental work and by the establishment of free consump- tive hospitals, and by efforts to improve the character of the food, dwellings, and conditions of the people in general, we should endeavor to build up the individual resistance to the disease. It may be years before the public are sufficiently educated to cooperate with the sanitary authorities in adopting the necessary hygienic measures to stamp out tuberculosis entirely; but, judging from the results which have already been obtained in reducing the mortality from this dread disease, we have reason to believe that in time it can be completely controlled. Among the numerous medical agents that have been tried without avail to protect animals against the action of the tubercle bacillus may be mentioned tannin, menthol, sulphuretted hydrogen, mercuric chloride, creosote, creolin, phenol, arsenic, eucalyptol, etc. Agglutination and Complement-fixation Reaction for Diagnosis. — The results obtained with the agglutination reaction by various observers have been very conflicting. At present the test cannot be advised as useful in diagnosis, as the sera of cases suffering from tuberculosis fre- quently fail to give a reaction, while the sera from those having no detectable tuberculosis frequently cause a good reaction. A reaction in dilutions of 1 to 10 or 1 to 15 is considered a positive test (For Comple- ment-fixation see p. 200.) The Tubercle Bacillus of Domestic Animals and its Relation to Human Tuberculosis. — ^Among the domestic animals tuberculosis is most common in cattle. On account of the milk which they provide for our use, and which is likely to contain bacilli, the relation of these to human tuberculosis is a matter of extreme importance. The chief seat of the lesions is apt to be the lymphatic nodes or lungs, and with them the pleura; less often the abdominal organs and the udder are affected. In pigs the abdominal organs are more often 388 BACILLUS AND BACTERIOLOGY OF TUBERCULOSIS involved, then the lungs and lymphatic glands. In sheep, horses and goats tuberculosis is rare. Differences between Tubercle Bacilli of Human and Bovine Type- As has been already noted in the tables given of the incidence of bovine and human infection, it is possible to tell in any case the type of infec- tion. The essential differences are in cultural characteristics and in virulence for rabbits and calves. Cultural Differences. — ^The bovine bacillus grows very poorly when isolated, the human bacillus very freely. This is noted on plain egg, but to a less extent than on glycerin egg. The glycerin restrains or adds little to the growth of bovine bacilli, but increases markedly the amount of growth of the human bacillus. In fact, primary cultures on glycerin egg of bovine material commonly fail to grow. This difference is very noticeable in the first few generations and is sufficient in the great majority of instances for differentiation to one who has had some experience with such cultures. Further, the majority of human strains can be transplanted to glycerin potato or glycerin broth and give vigorous growth in- the first few generations, whereas the bovine bacillus fails to grow or growth is very slight. After further cultivation the bovine bacillus gradually increases its amount of growth until it is indistinguishable from the human type. This increase in luxuriance may be rapid or very slow. Rabbit Virulence. — ^The bovine bacillus is exceedingly virulent for rabbits by any method of inoculation; the human bacillus only slightly so. The best method of differentiation is by intravenous inoculation. A small amount of culture is weighed after the moisture has been extracted with filter paper, and a suspension made in normal saline and diluted so that 1 c.c. = 0.01 mg. of culture; this amount is then injected into the ear vein of a rabbit. If the rabbit survives for from forty to fifty days, and on autopsy shows only lesions in the lungs or kidneys or both, the strain is of the human type. With the bovine tj^e of bacillus the rabbit will die in the majority of instances before or about this time, if not it may be killed. On autopsy a progressive generalized tuberculosis will be found. The lesions in the lungs will be very marked, the tubercles having become confluent with caseous centres. The liver or spleen or both will be peppered with tubercles. Tubercles will be present in the great majority of cases in the superficial lymph nodes and also in those of abdomen and thorax. There may be tubercles on the heart, in the rib marrow, or over the peritoneum. These two differences alone are sufficient to differentiate in every case the type of bacillus. It must be insisted upon again that the cul- tural characteristics be observed in the early generation and, further, that the virulence be tested in early generations. In case the bovine culture does not afford sufficient material for weighing, a suspension can be made and compared with a weighed suspension. Virulence for Calves. — In proving the non-identity of the two bacilli, calf experiments were resorted to. This was necessary as the supposed bovine cultures from children would have to be virulent for calves to STABILITY OF THE DIFFERENT TYPES OF BACILLI 389 the same extent as cultures from bovine material. The commonly used method was the subcutaneous inoculation in the side of the neck with 50 mg. of culture. The human type of bacillus caused only a local lesion or at most a spreading to the nearest lymph node. The bovine bacillus, on the other hand, caused a generalized tuberculosis which was or was not fatal. Sufficient data has been accumulated to make this test practically unnecessary for the determination of type. Differences in Morphology. — ^The bovine bacillus tends to be shorter, thicker, and solidly stained; the human type tends to be longer, slimmer, usually bent, and shows beading and irregularities in staining. We have found this difference most marked on glycerin egg, slight or impercep- tible on other media. Besides the above differences Theobald Smith made the interesting discovery that the production of acid differed with the two types when grown on glycerin broth. The bovine type renders the bouillon less and less acid; this may even progress until the medimn becomes slightly alkaline to phenolphthalein. The human type causes a preliminary fall in the acidity; as growth progresses the acidity is then gradually increased, and may exceed the original acidity of the broth used. This difference is evident in tuberculin made from the two types of bacilli. The bovine tuberculin is alkaline or very slightly acid while human tuberculin is markedly acid. The change is only noticed when glycerin is used in the media. The work of more recent investigators would seem to show that this difference, like all differences between the types, is purely quantitative, and that different strains vary in their reactions and give intermediate reactions between these two extremes. Bird (Avian) Tuberculosis. — ^Tuberculosis is very common among fowl. The bacillus grows easily and freely on glycerin media. It tends to form a moist or even slimy growth, and commonly produces an orange pigment. It is able to grow at a higher temperature than mam- malian tubercle bacilli, the latter failing to grow above 41° C, the former growing at even higher temperatures. Guinea-pigs are less susceptible to inoculation with avian tubercle bacilli, and the virulence for these animals is usually quickly lost. Rabbits are much more susceptible. Rats and mice are spontaneously infected with avian tubercle bacilli and are supposed to be an important factor in spreading the disease. Birds are refractory, with few exceptions, to infection with the mam- malian tubercle bacillus. Parrots, however, are susceptible to infection with all three types and commonly have spontaneous tuberculosis caused by the human type of bacillus. Stability of the Different Types of Bacilli. — The fact that the agglu- tination reactions and the tuberculin reactions of the different types is similar shows their close relationship. This has led to the endeavor to change one type into the other. This is usually attempted by passage through animals. The results have been peculiar. Some cultures have been passed through a series of calves without any change except for a moderate increase in virulence. Other cultures seem to have completely changed their type. We believe that this is not a change of 390 BACILLUS AND BACTERIOLOGY OF TUBERCULOSIS type, but an additional bovine infection. Strong negative evidence found is the fact that the bovine bacillus when infecting man loses none of its characteristics, though present in the human body for years. | Tuberculosis of Cold-blooded Animals. — The bacilli of this group are of interest mainly because of the claims of Friedmann that immunity against the tubercle bacillus is produced by their injection, j and the publicity given to his claims that vaccines of these organisms have a curative influence in tuberculosis. The results of the use of the so-called Friedmann vaccine have, however, been unsatisfactory. A similar atteriipt to employ the cold-blooded types for immunization is the use in cattle as advised by Klimmer. These types have been isolated from spontaneous, tuberculous-like lesions of frogs, lizards, turtles, fishes, etc. They have little resemblance to the mammalian types of tubercle bacilli other than their acid-fastness. They grow rapidly and luxuriantly on ordinary media and their optimum ■ of growth is at 20° to 30° C, higher temperatures inhibiting their growth. They are not pathogenic for warm-blooded animals, although toxic symptoms or limited lesions may be produced by the injection of large doses as with most other non-pathogenic acid-fasts. Tuberculins prepared from them are not toxic for tuberculous mammalia except in large doses, so that a specific reaction can be excluded. It is possible but very improbable that they produce some antibodies which would protect 'against mammalian bacilli to a limited extent. Methods of Examination for Tubercle Bacilli. — One of the most important results of the discovery of the tubercle bacillus relates to the practical diagnosis of tuberculosis. The staining peculiarities of this bacillus render it possible by the bacteriological examination of micro- scopic preparations to make an almost positive diagnosis in the majority of cases. A still more certain test in doubtful cases is the subcutaneous or intraperitoneal injection of guinea-pigs, which permits of the deter- mination of the presence of numbers of bacilli, so small as to escape detection by microscopic examination. For the animal test, however, time is required — at least three weeks, and, if the bacilli present are very few in number, at least six weeks — before any positive conclusion can be reached, for when only a few bacilli are present tuberculosis develops slowly in animals. In disinfection experiments where many dead bacilli i are injected, care must be taken to exclude the local effect of dead bacilli. In doubtful cases a second guinea-pig should be injected with material from the first. Microscopic Examination of Sputum for the Presence of Tubercle Bacilli. — 1. Collection of Material. — ^The sputum should be collected in a clean bottle (two-ounce) with a wide mouth and a water-tight stopper, and the bottle labelled with the name of the patient or with some other distinguishing mark. The expectoration discharged in the morning is to be preferred, especially in recent cases, and the material should be coughed up from the lungs. Care should be taken that the contents of the stomach, nasopharyngeal mucus, etc., are not discharged during the act of expectoration and collected instead of pulmonary J Microscopic examination of sputum 39l sputum. If the expectoration be scanty the entire amount discharged in twenty-four hours should be collected. In pulmonary tuberculosis the purulent, cheesy, and mucopurulent sputum usually contains bacilli; while pure mucus, blood, and saliva, as a rule, do not. When hemorrhage has occurred, if possible some purulent, cheesy, or muco- purulent sputum should be collected for examination. The sputum should not be kept any longer than necessary before examination, for, though a slight delay or even until putrefaction begins, does not vitiate the results so far as the examination for tubercle bacilli is concerned, it almost destroys any proper investigation of the mixed infection present; it is best, therefore, to examine it in as fresh a condition as possible, and it should be kept on ice until examined if cultures are to be made. 2. Methods of Examination. — Examination for Tubercle Bacilli. — Pour the specimen into a clean, shallow vessel, having a blackened bottom — a Petri dish placed upon a sheet of dull black paper answers the purpose — and select from the sputum some of the true expectora- tion, containing, if possible, one of the small white or yellowish-white cheesy-looking masses or "balls." From this make rather thick cover- glass or slide "smears" in the usual way. In doubtful cases a number of these coarse or fine particles should be placed on the slide. The material being thick should be evenly spread and very thoroughly dried in the air before heating. (For methods of staining see pp. 77 and 80.) Occasionally one is able to demonstrate the presence of tubercle bacilli with the Hermann stain where Ziehl's carbol-fuchsin gives nega- tive results. At least two smears should be made and examined, if possible, before a negative report is given. Many of the incipient cases will require several examinations before bacilli are found. Some will remain consistently negative. It must be remembered that lesions may exist and that without ulceration the bacilli do not find their way into the sputum. Methods for Concentrating the Bacilli. — Uhlenhuth advises the use of antiformin. This is a patented preparation consisting of a mixture of sodiimi hydroxide and sodium hypochlorite solution. If this is mixed with sputum so that the total strength is about 15 per cent, of antiformin, the sputum quickly becomes fluid. This should be thinned with water or alcohol to help reduce the specific gravity of the mixture and centrifuged. The sediment is then mixed with water and recentrifuged, and the washed sediment used for smears. Besides the dissolving action, antiformin kills most of the bacteria in the sputum, but not the tubercle bacilli, though they are slowly affected, so that sediment may be used for cultural purposes or injection into guinea-pigs. A comparison of the above methods made by us gave the following results : Of twenty-eight sputa negative with carbol-fuchsin, two showed bacilli after a few minutes, search with the crystal violet stain. On restaining with carbol-fuchsin and giving only a light counter-stain with methylene blue the negative slides were also positive. Of the remaining twenty-six, four (15 per cent.) were quickly positive in the antiformin 392 BACILLUS AND BACTERIOLOGY OF TUBERCULOSIS sediment when stained with crystal violet, whereas only three were positive with carbol-fuchsin and only after restaining as above. It is advisable, therefore, in using carbol-fuchsin to have only a light counter- stain to make the method most efficient, and control the results with crystal violet if negative. In place of sedimenting the bacilli, the dissolved sputum may be shaken up with a hydrocarbon. When the hydrocarbon separates out from the sputum the waxy tubercle bacilli adhere to it and are collected in a layer between the dissolved sputum and the hydrocarbon. If the hydrocarbon is heavy, chloroform, they are carried down, if light, ligroin, they are carried up. Kinyoun has modified the original ligroin method as given below. We have had very satisfactory results with its use. As a routine method it saves time and gives a high percentage of positive results. Bottles of about 15 c.c. capacity containing about 2 c.c. of a 1 per cent, solution of cresol are used for collection. The cresol is added to limit decom- position of the sputum if its transit to the laboratory is delayed. When received 1 c.c. of ligroin (specific gravity not less than 0.715 or more than 0.72) is added and the bottle filled with an alkaline solution of hypochlorite of lime. If the bottle is full, about one-third must be poured out to allow for the addition of the solution. The hypochlorite solution is prepared as follows: Three packages of chlorinated lime are weighed, and for each 90 grams, 65 grams of sodium carbonate are taken. The lime is mixed with 500 to 600 c.c. of water and the carbonate is dissolved in 1500 c.c. of water by boiling. The carbonate is then added to the lime and thoroughly mixed. After standing twelve to twenty-four hours the solution is filtered off. The amount of available chlorine is estimated and the solution diluted so that the chlorine is 0.56 per cent. Then 7.5 grams of catistic soda are added to each 100 c.c. of the filtrate. The solution shoiild be kept cool and in the dark. Fresh lots should be prepared about every three months. After the addition of the ligroin and the solution of chlorine the bottles of sputum are placed in a shaking machine and thoroughly shaken for five to ten minutes. The bottles can then be allowed to stand until the ligroin rises, which takes several hours, or this can be hastened, placing the bottles in a centrifuge, with special cups to accommodate the bottles, and run for about ten minutes at moderate speed. When the ligroin rises to the top a soapy layer develops at the point of con- tact with the fluidified sputum and the tubercle bacilli are collected in this layer. The soapy layer is taken up with a platinum loop and smears made on glass sUdes fixed by heat and stained. Individual slides must be used and the slides must be stained separately or error will result, as the bacilli are not firmly fixed to the slides. All the antiformin methods must be used with caution, as it is easy to see how error can creep in from contamination with other acid-fast bacilli. (See also the Petroff method given above.) Detection of Tubercle Bacilli in Urine and Feces, etc. — ^The cath- eterized urine is centrifuged. If little sediment appears, the upper portion of the fluid is removed and more urine added and again centri- fuged. If the urine is rich in salts of uric acid, the same may be dim- inished by carefully warming the urine before treating it. If too alkaline add a little acetic acid. A possible source of error is the presence of smegma bacilh. DETECTION OF TUBERCLE BACILLI 393 The feces are examined for any purulent or mucous particles. If none are found, larger masses of feces are removed and then the rest diluted and centrifugalized. The antiformin methods are a great aid in the examination of feces. The examiner must remember that bacilli swallowed with the sputum may appear in the feces. In examining cerebrospinal fluid for tubercle bacilli it must be remembered that the majority of the bacilli are entangled in the delicate clot that forms. Whenever possible the fluid after withdrawal should be allowed to stand until this filmy clot develops, which is then fished out and examined. If this is impossible the fluid should be centrifuged and the sediment stained. This is also the case in other serous fluids, but in ascitic or pleuritic fluid they are usually very few in number. (For sputum washing see p. 136.) Inoculation of Animals. — ^The inoculation of suspected material into guinea-pigs produces tuberculosis; even if the number of bacilli is very small. When no bacilli can be detected by microscopic examination this can be done for diagnostic purposes. The material should be injected subcutaneously as already described. Cultivation. — This requires so much time that it is not generally used except in important investigations upon the nature of the tubercle bacilli. The special methods have already been given. CHAPTER XXVIII. OTHER ACID-FAST BACILLI: BACILLUS OF LEPROSY, BACILLUS OF RAT LEPROSY, BACILLUS OF JOHNE'S DISEASE IN CATTLE, AND THE GROUP OF NON- PATHOGENIC ACID-FAST BACILLI. LEPROSY BACILLUS— B. LEPR.ffi. The bacillus of leprosy was discovered by Hansen and Neisser (1879) in the leprous tubercles of persons afflicted with the disease. This discovery was confirmed by many subsequent observers. Morphology (in Tissues). — Small, slender rods resembling the tubercle bacilli in form, but somewhat shorter and not so frequently curved. The rods have pointed ends, and in stained preparations unstained spaces, similar to those observed in the tubercle bacillus, are seen. They stain readily with the aniline colors and also by Gram's method. Although differing slightly from the tubercle bacillus in the ease with which they take up the ordinary aniline dyes, they behave like tubercle bacilli in retaining their color when subsequently treated with strong solutions of the mineral acids' and alcohol. The difference in staining characteristics is too slight to be relied upon for diagnostic purposes (see Plate VI). Bacilli Isolated from Leprous Lesions. — No acid-fast organism was grown from leprous lesions until Clegg reported, in 1908, that he had been able to cultivate an acid-fast bacillus by growing it in symbiosis with ameba and cholera. Since then Duval, Kedrowski, Twort, and many others have reported the finding of various more or less acid- fast organisms in leprous lesions. These organisms may be grouped as follows ;! Bacilli of the Diphtheroid Type.^ — In serum media the colonies are yellowish white and develop best at 37° C, although a slight growth occurs at room temperature. When growth occurs on broth, the medium remains clear, and a pellicle is produced. The morphology is variable. They are either solidly stained or irregularly stained like other types of diphtheroids. They are Gram-positive and may show som'e resist- ance to decolorization after staining with carbol-fuchsin, especially the metachromatic granules. Pathogenicity, none or questionable. Acid-fast Chromogenic Bacilli. — This type of bacillus is difficult to isolate but after isolation grows freely at both 37° and 20° C. on 1 The results of all the investigations cannot be given. For a fuller discussion and bibliography see the excellent resumi of Wolbach and Honeij: Jour. Med. Research, 1914, xxix, 367. LEPROSY BACILLUS 395 most of the ordinary media. The growth is luxuriant, moist, and a yellow to deep orange color develops. The individual bacilli vary in morphology from coccoid to filamentous bacilli, some showing meta- chromatic granules, others showing clear areas. They are acid-fast but less so than the tubercle bacillus. Pathogenicity. — Lesions similar to leprosy produced in Japanese dancing mice and in monkeys. Anaerobic Bacilli. — In this place it is sufficient to state that such organisms have been isolated. Acid-fast Non-chromogenic Bacilli. — These types are characterized by their feeble, slow growth on artificial media, and growth only takes place at 37° C, and then only on special media. Morphologically they vary from plump to long, slender bacilli, often beaded or bipolar in appearance. Pathogenicity, none. Fig. 141. — Leprosy bacilli in nodule. (Kolle and Wassermann.) What conclusion is to be drawn from such variable results is difficult to say. Has the diphtheroid bacillus any relationship to the more acid- fast types ? This is a possibility when one considers that under certain circumstances it shows some resistance to decolorization. Then, too, the leprosy bacilli in tissue may decolorize easily, although they are abundant, as shown by staining with polychrome methylene blue. On the other hand, the repeated isolation of diphtheroids from the lymph nodes In other conditions raises a strong element of doubt as to the etiological significance of this organism. A careful comparison of the diphtheroids from various conditions, especially their resistance to decolorization, might give us some help. As to the chromogenic acid-fasts, the character of their growth, viz., the luxuriant growth upon ordinary media at low temperatures is not that of a highly specialized parasite. As to the apparently successful animal inoculations it must be remembered that lesions have been produced by acid-fast bacilli, known to be non-pathogenic. Why such 396 OTHER ACID-FAST BACILLI bacilli should be frequently isolated from leprous lesions is still to be explained. The non-pigmented types are more consistent with our idea of what the bacillus should be but whether they are actually the etiological organism remains to be seen. The serum reactions, such as agglutination and complement-fixation, have added no evidence as to the etiological significance of any one of the bacilli isolated. Each of the bacilli mentioned has. been agglu- tinated by sera of lepers. The complement-fixation reactions await a successful specific test for the individual acid-fast organisms. Pathogenesis. — Numerous inoculation experiments have been made on animals with portions of leprous tubercles, but there is no con- clusive evidence that leprosy can be transmitted to the lower animals by inoculation. The inference that this bacillus bears an etiological relation to the disease with which it is associated is based chiefly upon the demonstration of its constant presence in leprous tissues. The bacilli are found in all the diseased parts, and usually in large numbers, especially in tubercles on the skin, in the conjunctiva and cornea, the mucous membranes of^ the mouth, gums, and larynx, and in the interstitial processes of the nerves, teisticles, spleen, liver, and kidneys. The rods lie almost exclusively within the peculiar round or oval cells of the granulation tissues which compose the leprous tubercles, either irregularly scattered or ajffanged parallel to one another. In old centres of infection the leprosy cells containing the bacilli are larger and often polynuclear. Giant cells, such as are found in tuber- culosis, are claimed to have been observed by a few investigators (Boinet and Borrel). In the interior of the skin tubercles, the hair follicles, sebaceous and sweat glands are often attacked, and bacilli have sometimes been found in these (Unna, etc.) . Quite young eruptions often show a few bacilli. A true caseation of the tubercles does not occur, but ulceration results. During acute exacerbations with develop- ment of new lesions bacilli have been observed in the blood. In the anesthetic forms of leprosy the bacilli are found most commonly in the nerves and less frequently in the skin. They have been demon- strated in the sympathetic nervous system, in the spinal cord, and in the brain. The Bacillus leprae occurs also in the blood, partly free and partly within the leukocytes, especially during the febrile stage which precedes the breaking out of fresh tubercles (Walters and Doutrelepont). The bacilli have also been found in the intestines, in the lungs, and in the sputum, but not in the urine. With regard to the question of the direct inheritance of the disease from the mother to the unborn child there is considerable difference of opinion. Some cases have been reported, however, in which a direct transmission of the bacillus during intra-uterine life seems to be the only or most plausible explanation of the infection. At the same time, we have no positive experimental evidence to prove that such an infec- tion does take place. Although many attempts have been made to infect healthy individuals with material containing the bacilli of leprosy. LEPROSY BACILLUS 397 the results are not conclusive. Even the experiments made by Arning, who successfully infected a condemned criminal in the Sandwich Islands with fresh leprous tubercles, and which have been regarded as positive evidence of the transmissibility of the disease in this way, are by no means conclusive; for, according to Swift, the man had other oppor- tunities for becoming infected. The negative results, together with the fact that infection does not more frequently occur in persons exposed to the disease, may possibly be explained by the assumption that the bacilli contained in the tuberculous tissues are mostly dead, or much more probably that an individual susceptibility to the disease is requisite for its production. The widespread idea, before the discovery of the leprosy bacillus, that the disease was associated with the constant eating of dried fish or a certain kind of food, has now been entirely abandoned. The relation of leprosy to tuberculosis is sufficiently evident from their great similarity in. many respects. This is rendered still more remarkable by the fact that leprosy reacts, both locally and generally, to an injection of tuberculin in the same manner as tuberculosis, but to a somewhat less extent. Rat Leprosy. — ^The interest in this disease lies in the fact that diphtheroid and chromogenic acid-fast bacilH similar to those described above have been isolated from leprous rats. Bacillus of Johne's Disease, Cbronic Enteritis or Faratubercular Dysentery of Cattle. — This disease is comparatively common in this country and is char- acterized by chronic diarrhea and emaciation, commonly leachng to death. The intestinal mucosa is thickened, and the lesions are not limited. Tubercle formation and necrosis are absent. The baciUi are present in the lesions in enormous numbers. Twort succeeded in cultivating the organism and his work was verified by Holth and Meyer. A tubercuhn made from this organism win probably be of diagnostic value. Animals having this disease do not react to the ordinary tuberculin test but do react to large doses of tuberculin made from the avian type of tubercle bacillus. The bacillus is not pathogenic for guinea-pigs or rabbits, although local abscesses may be produced. Non-pathogenic Acid-fast Bacilli. — ^These have no importance further than historical interest and the fact that they may be present in materials suspected of containing tubercle bacilli and thus lead to error. They vary widely in their acid-fastness, especially when artificially cultivated. Differential staining methods have been devised to separate them from the tubercle baciUus and although in a general way the decolorization by prolonged action of acid and alcohol is presmnptive evidence against suspected bacilli being tubercle bacilli, it is an luisafe procedure. Tubercle baciUi vary in their acid-fastness but the non-pathogenic types vary even more widely, some being extremely resistant to decolorization. Many of the non -pathogenic types grow rapidly at low temperatines and in cultures can thus be quickly differentiated from tubercle bacilli. They can be separated from tubercle baciUi by inoculating animals in which no progressive lesions will develop, although limited lesions may be produced if they are injected in large numbers. The guinea-pig may be injected with 2 c.c. of tuberculin and if infected with tuberculosis will die, but if by other acid- fast bacilli, will show little of no reaction. If a second group of guinea-pigs are inoculated with a small amount of the infected tissue from the inoculated pigs there wiU develop progressive tuberculosis if the doubtful bacilli were tubercle bacilli, and practically no lesions if they were grass bacilli. Cultures from the lesions may also be an aid in differentiation. 398 OTHER ACID-FAST BACILLI Bacillus of Lustgarten. — ^This bacillus was found by Lustgaiten in 1884 in syphilitic lesions or ulcers. It is undoubtedly a saprophyte. It is very similar morphologically to the smegma bacillus and may be identical with it. It is of historical interest only. Smegma Bacillus. — This bacillus is present in smegma from the prepuce or vulva. Its only interest is the danger of mistaking it for tubercle bacilli in the examination of urine, especially if the latter be carelessly collected. Timothy and Other Grass Bacilli. — On various grasses, in cow manure, in butter and in milk there have been found bacilli with varying degrees of acid- fastness. Similar bacilli have also been demonstrated in water. They make the direct microscopic examination of such material for tubercle bacilli of Uttle value and the nature of any acid-fast organisms so found must be determined by animal inoculations. CHAPTER XXIX. GLANDERS BACILLUS (BACILLUS MALLEI). B. ABORTUS (BANG). GLANDERS BACILLUS. The Bacillus mallei was discovered and proved to be the cause of glanders by several bacteriologists at almost the same time (1882). Bouchard, Capitan and Charin obtained it in mixed cultures, while it was first accurately studied in pure culture by Loffler and Schiitz. It is present in recently formed nodules in animals affected with glanders, in the nasal discharge, in pus from the specific ulcers, etc., and oc- casionally in the blood. Morphology. — Small bacilli with rounded or pointed ends, from nutrient agar cultures, 0.25^ to 0.5/i broad and from 1.5/i to bfi long; usually single, but sometimes united in pairs, or growing out to long filaments, especially in potato cultm-es. The bacilli frequently break up into short almost coccus-like elements (Fig. 142). Staining. — The bacillus mallei stains with difficulty with the ani- line colors, best when the aqueous solutions of these dyes are made feebly alkaline; it is decolorized by Gram's method. This bacillus presents the peculiarity of losing very quickly in decolorizing solu- tions the color imparted to it by the aniline staining solution. For this reason it is diflScult to stain in sections. Loffler recommends his alkaline methylene-blue solution for staining sections, and for decolorizing, a mixture containing 10 c.c. of distilled water, 2 drops of strong sulphuric acid, and 1 drop of a 5 per cent, solution of oxalic acid; thin sections to be left in this acid solution for five seconds. Biology. — A non-motile bacillus, whose molecular movements are so active that they have often been taken for motility. It is aerobic, but moderate multiplication occurs in the depths of culture media. Grows well on culture media at 37° C. Development takes place Fig. 142. — Glanders bacilli. Agar culture. X 1100 diameters. 400 GLANDERS BACILLUS slowly at 22° C. and ceases at 43° C. The bacillus does not form spores. Exposure for ten minutes to a temperature of 55° C, or for five minutes to a 3 to 5 per cent, solution of carbolic acid, or for two minutes to a 1 to 5000 solution of mercuric chloride destroys its vitality. As a rule the bacilli do not grow after having been preserved in a desic- cated condition for a week or two; in distilled water- they may live twenty-five days. It is doubtful whether the glanders bacillus finds conditions in nature favorable to a saprophytic existence. A solution of chlorinated lime, containing 1 part of free chlorine per 1000, is useful as a disinfectant of stables and utensils; it kills the bacillus in from one to two minutes. Strong sodium carbonate solution (washing soda) is also useful. Cultivation. — (For obtaining pure cultures see page 401.) — ^It grows well at 37° C. on glycerin-veal agar, an acidity of 1.5 to 2.5 (phenolphthalein) being the most favorable. Upon this medium, at the end of twenty-four to forty-eight hours, whitish, transparent colonies are developed, which in six or seven days may attain a diameter of 7 or 8 mm. On blood serum a moist, opaque, slimy layer develops, which is of a yellowish-brown tinge. The growth on cooked potato, that is, sterilized, is especially characteristic. At the end of twenty-four to thirty-six hours at 37° C. a moist, yellow, transparent layer develops; this later becomes deeper in color, and finally takes on a reddish-brown color, while the potato about it acquires a greenish-yellow tint. In bouillon the bacillus causes diffuse clouding sometimes with a pellicle, ultimately with the formation of a more or less ropy, tenacious sediment. The broth should also be as acid in reaction as the glycerin-veal agar above. The addition of potato juice to either of these media is most favorable to the growth of the organism. Milk is coagulated with the production of acid. Pathogenicity. — ^The bacillus of glanders is pathogenic for a number of animals. Among those which are most susceptible are horses, asses, guinea-pigs, cats, dogs, ferrets, moles, and field mice; sheep, goats, swine, . rabbits, white mice, and house mice are much less susceptible; cattle are immune. Man is susceptible, developing both the acute and chronic forms. Infection not infrequently terminates fatally, usually in about 60 per cent, of the cases. Doubtless many cases are not recog- nized as glanders, but are mistaken for other diseases such as pyemia, rheumatism, typhoid and syphilis. (Fitch.) When pure cultures of Bacillus mallei are injected into horses or other susceptible animals true glanders is produced. The disease is characterized in the horse by the formation of ulcers upon the nasal mucous membrane, which have irregular, thickened margins, and secrete a thin, virulent mucus; the submaxillary lymphatic glands become enlarged and form tumors which are often lobulated; other lymphatic glands become inflamed, and some of them suppurate and open externally, leaving deep, open ulcers; the lungs are also involved, and the breathing becomes rapid and irregular. Acute generalized glanders may cause death in one to six weeks, young horses being MODE OF SPREAD 401 especially susceptible. At postmortem usually the lungs show either tubercle-like nodules or pneumonic areas. Nodules may occur also in the liver, spleen and lymph glands. Of the bones, the ribs are the most often involved and contain caverns filled with a tenacious yellowish substance. In farcy, which is a more chronic form of the disease, cir- cumscribed swellings, varying in size from a pea to a hazel-nut, appear on different parts of the body, especially where the skin is thinnest; these suppurate and leave angry-looking ulcers with ragged edges, from which there is an abundant purulent discharge. The lymphatics leading from these ulcers become inflamed, stand out as tense hot cords under the skin and from them new "farcy buds" may develop. Chronic cases may run on for years until an acute exacerbation due to overwork or adverse conditions brings death. The bacillus of glanders can be obtained in pure cultures from the interior of suppurating nodules and glands which have not yet opened to the surface, and the same material may give successful results when inoculated into susceptible animals. The discharge from the nostrils or from an open ulcer may contain comparatively few bacilli, and these being associated with other bacteria which grow more readily on the culture media than the Bacillits mallei, make it difficult to obtain pure cultures from such material by the plate method. In that case, however, guinea-pig inoculations are useful. Of test animals guinea-pigs and field mice are the most susceptible. In guinea-pigs subcutaneous injections are followed in four or five days by swelling at the point of inoculation, and a tumor with caseous contents soon develops; then ulceration of the skin takes place and a chronic purulent ulcer is formed. The essential lesion is the granulo- matous tumor, characterized by the presence of numerous lymphoid and epithelioid cells, among and in which are seen the glanders bacilli. The lymphatic glands become inflamed and general symptoms of infec- tion are developed in from two to four weeks; the glands suppiu-ate, and in males the testicles are involved. This fact is used as a means of diagnosis. (See Straus Reaction, p. 406.) Finally purulent inflam- mation of the joints occurs and death ensues from exhaustion. The formation of the specific ulcers upon the nasal mucous membrane, which characterizes the disease in the horse, is rarely seen when guinea-pigs are inoculated. In these the process of the disease is often prolonged or remains localized on the skin. Guinea-pigs succumb more rapidly to intraperitoneal injection, usually in from eight to ten days. Attenuation of virulence occurs in cultures which have been kept for gome time on artificial media and inoculation with such cultures may give a -negative result, or, when considerable quantities are injected, may produce a fatal result at a later date than is usual when small amounts of a recently isolated culture are injected. Mode of Spread. — Glanders occurs as a natural infection only in horses and asses. The disease is occasionally communicated to man by contact with affected animals, usually by inoculation on an abraded sur- face of the skin. The contagion may also be received on the mucous 26 402 GLANDERS BACILLUS membrane. Infection. has sometimes been produced in bacteriological laboratories. It is transmissible also from man to man. Washerwoman have been infected from the clothes of a patient. The infective material exists in the secretions of the nose, in the pus of glanders nodules, and frequently in the blood; it may occasionally be found in the secretions of glands not yet affected, as in the urine, milk, and saliva, and also in the fetus of diseased animals (Bonome). From recent observations it is found that glanders is by no means an uncommon disease among apparently sound horses, sometimes taking a mild course and remaining latent for a considerable time. Therefore, horses appearing healthy, may spread the disease through the public drinking troughs and black- smith shops. Immunity. — Attempts have been made to produce artificial immunity against glanders but so far with unsatisfactory results. Various workers, as Straus, Fenger and also Ladowski, have reported the production of immunity in the smaller animals, such as dogs, cats and rabbits, by the injection of either living or killed cultures. In the horse, the most important animal from the economic standpoint, other observers have reported not only immunity but cures by the use of vaccine and also subcutaneous mallein; however, corroboration is still lacking. Such substances as mallein and vaccine when injected do produce immune bodies which can be demonstrated by the serodiagnostic tests. It is unfortunate, however, that the presence of these immune bodies in the blood does not indicate, necessarily, a practical immunity of the animal against infection. (Mohler and Eichhorn.) Since the indiscriminate use of vaccine causes confusion in the blood tests of horses, it is not now being distributed by the Health Department of the City of New York. In man the therapeutic value of vaccine is not yet fully determined. But few cases so far have been reported — one by Bristow and White and one by Cramp recovered after the use of an autogenous glanders vaccine. The use of mallein in man subcutaneously, has been reported in six cases. (Robins.) In one it gave a reaction and in three it was of supposed benefit therapeutically. Diagnosis of Glanders.— The chief methods (Mohler and "Eichhorn) are: physical examination; serodiagnostic tests — complement-fixation and agglutination reaction; mallein reactions — eye and subcutaneous; "Straus reaction" (inoculation of guinea-pigs with either suspected material or cultures) ; postmortem examination. Physical Examination. — In horses those cases with clear-cut clinical • symptoms (p. 400) offer little difficulty to the veterinarian. However, the easily applied eye mallein reaction should be used for confirmation. These two methods are sufficient under such circumstances. It is the latent or occult cases, showing only a little fever, or none, that require additional tests for diagnosis. Since these cases are frequently the distributors of the diseases their early recognition and extermination is imperative. To detect these cases the application of the serodiagnostic reactions supplemented by the use of eye mallein is necessary (p, 404). DIAGNOSIS OF GLANDERS 403 Collection of Blood for Serodiagnostic Tests. — In obtaining blood from horses a large-sized hypodeimic needle, which has been sterilized, is inserted into the jugular vein which has been brought into view by pressing the thumb upon it from below; the blood is allowed to flow through the needle into a sterile neutral- ized tube or flask, 8 to 10 c.c. being sufficient. In the case of human beings, the median basiUc vein at the bend of the elbow is used. Under aseptic conditions 5 to 10 c.c. of blood are drawn either by means of a sterile hypodermic syringe, or allowed to flow through a large hypo- dermic needle, as above. Complement-fixation Test — In 1909 Schiitz and Schubert applied this method to the diagnosis of glanders. It gives excellent results, for it picks up 97 per cent, of positive cases according to Miessner and Trapp. Its failures lie chiefly in the early stages of the disease (see p. 406), and for this reason it should be paralleled by the agglutination reaction. The combination of these two tests gives, according to Huytera and Marek, a percentage of 99 successful tests. The use of a polyvalent antigen for the detection of glanders, as of gonorrhea, is important. Agglutination Reaction. — It was first applied byMacFadyean (1896) who used the microscopic method as in the Widal reaction. Later Schiitz and Miessner (1896) found the macroscopic method more practicable. In the early stages of glanders this method is most valuable (see p. 406). It picks up about 84 per cent, of positive cases (Anthony and Grund), the failures occurring chiefly in old chronic cases. Schiitz and Miessner claim that a culture recently passed through a guinea-pig (once in three weeks) is essential for a good test fluid. In our hands cultures kept on artificial media do well if passed through a pig once in two inonths. Not every strain of B. mallei agglutinates well, consequently a suitable one must be chosen. A univalent test fluid is the most logical, for in a polyvalent fluid a strain not agglutinated by a particular serum would cloud the supernatant fluid at the same time that a reaction occurred in the bottom of the tube. The macroscopic agglutination test may be carried out in several ways : Iriciibator Method. — ^The procedure of Schiitz and Miessner, with slight modifications, is as follows: A forty-eight-hour acid-glycerin-agar culture of B. mallei is washed off with normal saline solution containing 0.5 per cent, pure carbolic acid. This suspension is heated at 60° C. for two hours. It is then filtered through cotton, and enough of the carbolic salt solution added to reduce it to a faintly cloudy suspension. This should be standardized by comparing it with a known test fluid, if possible, and testing it with known negative and positive sera. This test fluid wiU keep in the ice-box for several weeks. The active serum to be tested is then made up with normal salt solution to a 1 to 40 dilution. From this the final dilutions of 1 to 500, 1 to 800, 1 to 1000, etc., ^re made by adding 0.24 c.c, 0.15 c.c, and 0.12 c.c. respectively to 3 c.c. of the standardized test fluid in each test-tube. The tubes are well shaken and incubated twenty-four to seventy-two hours with positive and negative control sera in the same dilutions. If a reaction occurs the upper part of the fluid is clear, while a veil-like sediment is found at the bottom. A strong positive reaction (1 to 1000) may occiu- in twenty-four hoiu-s. A nega- tive reaction shows the sediment in a definite "button" at the bottom of the tube and the fluid above is cloudy. 404 GLANDERS BACILLUS With the centrifuge method of Miessner and others!, cited by Mohler and Eichhorn the time factor is greatly reduced. The tests are incubated for a half- hour at 37° C, then centrifugalized at 1600 revolutions for ten minutes, and kept at room temperature for two hours before reading. This is done by looking down on the tubes from above toward a dark background. Indefinite reac- tions may be read the next day after standing at room temperature. Special tubes with perfectly rounded bottoms are essential for this method, and the dilutions are made up in only 2 c.c. of the test fluid. Rapid Method. — Povitzky, of this laboratory, is applying the following method of macroscopic agglutination with time-saving results. Technic: A fresh culture grown for forty-eight hours on glycerin-potato-veal agar (2.5 acid to phenolphthalein) is washed off with a small amoimt of sterile normal salt solution. This thick milky suspension is heated for one hour at 70° C. After filtering through cotton, normal salt solution is added until the suspension is only faintly cloudy. This test fluid must be standardized by comparison with a known standard, etc., as described under the incubator method of agglu- tination. The active serum to be tested is then diluted with normal salt solu- tion and added in suitable amounts to 3 c.c. of the test fluid (see procedure under incubator method). The various dilutions and controls are then placed in a water-bath at 37° to 40° C. for two hours. Very active reactions can be read as early as the end of the first hour; others at the end of the second hour. The tubes are then set in the ice-box overnight and any delayed reactions may be read next morning. Although this method is only a little more rapid in^time than the centrifuge method, its easier technic and lack of compUcated apparatus recommend it. The limit of agglutination in the normal horse is 1 to 500, most of the reactions occurring at 1 to 200 or 400. Since, however, some cases of chronic glanders do not react above 1 to 500, this reaction should be regarded with suspicion and checked by the complepient-fixation test and eye mallein; so also reactions below 1 to 1000. Reactions of 1 to 1000 are positive, some horses running up to 1 to 2000 or 3000. In practical work dilutions higher than 1 to 1000 are unnecessary and this test should always be checked by the use of eye mallein and the comple- ment-fixation test. In human cases a reaction by the "rapid method" (Povitzky) of 1 to 500 and above, is considered positive. Normal human blood reacts seldom above 1 to 100, but it may reach 1 to 200 or even 400 in exceptional cases. Mallein Reaction.— Malleiji is like tuberculin in that it consists of glycerinated bouillon which contains the products of the growth and activity of B. mallei cultivated in it. It was discovered by Kelnig, a Russian veterinarian,, in 1890. (For the preparation of the two kinds of mallein — eye and subcutaneous, see p. 406.) The eye mallein reaction is the most recently developed test for glanders in animals, yet it has taken a preeminent place in diagnosis for it is already the Federal test for the interstate shipment of horses. The simplicity of the application of eye mallein, the short time — twenty-four hours- required, and the comparatively easy reading of results, after a suitable experience, make it possible for the veterinarian to apply prompt tests on suspected horses. In healthy horses the error of this test has been shown (Schnarer) to be only 0.39 per cent., while in glanderous horses the test gives 88.8 per cent, positive, 3.5 per cent, negative, and 7.5 per cent, doubtful. DIAGNOSIS OP GLANDERS 405 As with the complement-fixation test the doubtful and negative reac- tions occur chiefly in the early stages of the disease ; consequently its use alone, without the agglutination reaction, to check the complement- fixation test is subject to error unless retests at suitable intervals are planned and carried out. Technic of Application. — -When 2 or 3 drops of concentrated mallein aie instilled into the conjunctival sac, no reaction save a .slight lacrimation and congestion results in healthy horses. In glanderous animals this goes on, at the end of from five to seven hours, to profuse lacrimation, redness, edema, and the formation of pus. There may be only a drop of pus at the inner canthus of the eye, or all degrees to profuse purulent discharge. Unless pus is present, the reaction is not considered positive. There is a slight rise in temperature in those cases showing a marked reaction, but as the local reaction is very distinct, the tedious task of taking temperatures as in the subcutaneous method is superfluous. Another advantage of this method over the subcutaneous inoculation, aside from its simplicity, is the fact that it can be repeated after twenty- four hours in doubtful cases; also it does not interfere with subsequent serodiagnostic tests. With very few exceptions, a second test in a glanderous horse gives a prompt reaction. Subcutaneous Infection of Mallein. — Although the injection of mallein subcutaneously is one of the oldest and most reliable methods for the diagnosis of glanders it should be applied only after the complement-fixation, agglutination and eye mallein tests have been used, since the subcutaneous injection of mallein, as also any glanders antigen including vaccines, interferes with the sero- diagnostic tests. In spite of the fact that it picks up 89 per cent. (Huytera and Marek) of the positive cases, its cumbersome technic of prolonged temperature taking, the detention of the horses from work, etc., all contribute to render it unsuitable as an early test. Before applying the subcutaneous test the temperature of the horse should be taken at least three times at intervals of three hours. If there is fever the mallein should not be given. The injection of maUein (usually about 2 c.c.) should be made about 10 p.m. In a glanderous horse there will be a local reaction and a general reaction with fever. The temperature begins to rise usually three or four hours after the injection and reaches its maximum between the tenth and twelfth hours. Sometimes the highest point is not reached until fifteen to eighteen hours after the injection. This rise in temperature is from 1.5° to 2° C. (2° to 3.5° F.). The temperature taking should be con- tinued every two hours , beginning not later than eight to ten hovus after the mallein was given. The general condition of the animal is more or less pro- foundly modified and the local reaction is usually very marked around the point of injection. Here, in a few hours, there appears a warm, tense and very painful swelling. Running from this will be found hot, sensitive lines of sinuous lymphatics directed toward the neighboring lymphatic nodes. This edema increases for twenty-four to thirty-six hours and persists for several days, not disappearing entirely for eight or ten days. In healthy animals the rise of temperature is usually only a few tenths of a degree but it may reach 1 ° C. This rise should always be considered, however, in connection with the general and local reactions. At the point of injection the mallein produces only a small edematous tumor which, instead of increasing diminishes rapidly and disappears in about twenty-four hours. 406 GLANDERS BACILLUS Occurrence of the Reactions of the Various Tests after Infection. — Agglii- tinins increase above normal in four or five days and continue to rise in the early stages of the disease, diminishing as the disease becomes chronic. Specific amboceptors for the complement-fixation test may be demonstrated in from seven to ten days and remain during the entire course of the disease. The subcutaneous mallein test may, as a rule, be relied upon for diagnosis fifteen days after infection, while the eye mallein test is reliable twenty-one days after infection. (See Report on Detection on Glanders, in References at end of this chapter. Effect of One Test on the Others. — The serodiagnostic tests are influenced in three to six days after a subcutaneous injection of mallein or any glanders antigen including vaccines. The period of influence varies from six to eight weeks after the injection of mallein, and lasts three months or longer -after the injection of glanders antigen or vaccines. "Straus Reaction." — This test consists of introducing into the peri- toneal cavity of a male guinea-pig some material or a culture from the suspected products. If the B . mallei is present, the diagnosis may usually be made within two to five days from the tumefaction of the testicles showing evidences of pus formation from which pure cultures can be obtained. An objection to this method, however, is that occasionally from the injection of impure material, as in the nasal secretion, the animal may die of septicemia; but if an uncontaminated specimen can be obtained, as from the lymphatic glands, this method is satisfactory accord- ing to most authors. Nevertheless, while a positive result is conclusive evidence of glanders, failure of the pig to develop lesions is not proof of its absence, for other workers find such inoculations fail in about one- half the animals injected, even after a prolonged period. Postmortem Lesions. — Postmortem lesions are given on page 400. The confirmation of the findings of all positive (or doubtful) tests by careful examinations at autopsy is most desirable in order to extend our present data along these lines. Preparation of Mallein.^ — ^Mallein for subcutaneous injection is pro- duced by growing B. mallei (preferably a variety of strains) for six to eight weeks in a 5 per cent, glycerin-nutrient veal bouillon, about 2.5 acid to phenolphthalein. (See chapter on Media, p. 110.) Each flaSk or bottle of the culture is then tested for purity by the examination of smears and cultures made on neutral veal agar — a medium unfavorable to the growth of B. mallei. If pure, the broth culture is killed by steam- ing in the Arnold sterilizer for one hour. After sedimentation in the ice- box for a few days the supernatant fluid is filtered first through paper pulp and then through the Berkefeld filter. Carbolic acid is added to give 0.5 per cent. For eye mallein the same procedure as above is followed except that after filtering the liquid through paper pulp, it is measured and then evaporated over the hot- water bath to one-tenth its volume. The viscid liquid is then sterilized by heating in the Arnold sterilizer for three- 1 The methods given are those in use in the Health Department of the City of New York. B. ABORTUS 407 quarters of an hour. The precipitate which has formed is thrown down either by centrifugahzing, or by sedimentation in the ice-box. This latter method takes about two weeks. Before use both the subcutaneous and the eye mallein should be sub- jected to potency tests on both glandered and normal horses. B. ABORTUS (BANG). This organism, first described by Bang, is the cause of contagious abortion in cattle. It is a small, pleomorphic Gram-negative bacillus which when first isolated is micro-aerophilic, becoming aerobic on culti- vation. Its main interest medically is its common presence in milk and the possibility of human infection arising from this source. The presence in the blood of agglutinins and complement-fixing substances has been demonstrated in an appreciable number of children and in some instances of aborting women (Larson and Sedgwick, Nicoll and Pratt). Only once, however, has the organism been isolated from human tissues, viz., a tonsil, which cannot be considered as an infec- tion. Whether such antibody reactions are dUe to intestinal absorp- tion of the products of the bacilli contained in the milk or to the passive absorption of antibodies (Cooledge) in the milk of infected cows cannot be answered. At least it does not seem that they are due to infection. The other interest is that B. abortus when injected into guinea-pigs gives rise to lesions very similar to those of tuberculous origin (Smith and Fabyan). It is evident that this may be a source of error in ex- amining milk for the presence of tubercle bacilli by inoculation. Spontaneous infection of guinea-pigs has also occurred in a labora- tory, the source of infection being inoculated pigs. REFERENCES. Anthony and Gbund: Collected Studies, Bureau of Lab., City of New York, 1913, vii. 291. Bbistow and White: New York State Jour. Med., 1910, p. 236. Cramp: Jour. Am. Med. Assn., 1911, Ivi, 1379. Cooledge: Jour. Med. Research, 1916, xxxiv, 459. Fitch: Cornell Veterinarian, July, 1914. HuYTERA and Marek: Centralbl. f. Bact., 1909, Band lii. Larson and Sedgwick: Amer. Jour. Dis. Child., 1913, vi, 326. Miessner and Trapp: Centralbl. f. Bact., 1909, Band lii. MoHLER and Eichhorn: Bxill. XJ. S. Dept. Agriculture, 1914, No. 70. MoHLER and Eichhorn: U. S. Dept. Agriculture, Bureau of Animal Industry, 1912, Circ. 191. NicoLL and Pratt: Am. Jour. Dis. Child., 1915, x. Report on Detection of Glanders, Proc. Fiftieth Meeting Am. Vet. Assn., 1913, p. 291. Robins: Studies from Royal Victoria Hospital, Montreal, 1906, ii, 1. Schnurer: Proc. Tenth International Veterinary Congress, London, 1914. Smith and Fabyan : Centralbl. f. Bakt., 1912, Ixi, 549. Smith and Fabyan: Jour, of Inf. Dis., 1912, xi, 464. CHAPTER XXX. THE GROUP OF HEMOGLOBINOPHILIC BACILLI. BORDET-GENGOU BACILLUS. THE INFLUENZA BACILLUS. A DISEASE called influenza can be traced back to the fifteenth century and probably existed at a much earlier date. At times but few endemic cases occur, and then a great epidemic spreads over the civilized world. The last great epidemic reached Russia from the East in the fall of 1889 and . gradually spread over Europe and to America, reaching the latter country in December of that year. Since then we have had more or less of it, especially during the winter months. Many acute inflammations of the respiratory mucous membranes due to pneumococci and streptococci, give symp- toms similar to those due to the influenza bacillus. The rapidity of the spread of epidemics of influenza sug- gested that persons were the carriers of the infection, while the location of the disease pointed to the respiratory tract as the location of, and to the expectoration as the chief source of infection by, the microorganisms. After numerous unsuccessful attempts, during the epidemic of 1889 and succeeding years, to discover the specific cause of influenza, Pfeiffer (1892) succeeded in isolating and growing upon blood agar a bacillus which abounded in the purulent bronchial secretion of patients suffering from epidemic influenza, which he showed was the probable cause of the disease. Canon, working at the same time, found a similar bacillus in the blood of several cases of the disease. Though B. influenzcB has been shown to have definite pathogenic powers, its specificity in epidemic influenza has not been fully proved. Morphology.— Very small, moderately thick bacilli (0.2/x to 0.3m) in thickness to 0.5/i to 2/i in length), usually occurring singly or united in pairs, and occasionally showing threads, are found in spreads from the Fig. 143. — Influenza bacilli. X 1100 diameters. THE INFLUENZA BACILLUS 409 sputum and young cultures. In later cultures threads may be produced in great abundance. No capsule has been demonstrated. Staining. — The bacillus stains rather faintly with the ordinary aniline colors — best with dilute Ziehl's solution (water 9 parts to Ziehl's solu- tion 1 part), or Loffler's methylene-blue solution, with heat. Giemsa's method stains them well and brings out the polar granules which sometimes develop in these bacilli. They are not stained by Cram's method. Biology. — ^An aerobic, facultative anaerobic (contrary to the accepted opinion), non-motile bacillus; does not form spores; no growth occurs with most cultures below 22° C, or above 41° C. Cultivation. — This bacillus is best cultivated at 37° C, and on ordinary nutrient culture media containing hemoglobin (p. 103). At the end of eighteen hours in the incubator very small circular colonies are developed, which, under a low magnification (100 diameters), appear as shining, transparent, homogeneous masses, and even at 600 diameters scarcely show indications of the individual organisms. Older colonies are sometimes colored yellowish brown and in the centre are characteristic heapings. Another characteristic feature of the influenza bacillus is that the colonies tend to remain separate from each other, although when they are thickly sown in a film of moist blood upon nutrient agar they may become confluent. Transplantation of the original cul- ture to ordinary agar or serum cannot be successfully performed, owing to the want of sufficient hemoglobin; but if sterile rabbit, pigeon, horse or human blood be added to these media, even in very small amounts (1 to 500), especially if the agar is very hot (90° C.) at the time the blood is added, transplantation may be indefinitely performed, provided it is done every three or four days. Cultures if kept moist, may remain alive for a variable time. By a series of beautifully carried out experi- ments Pfeiffer showed that not only were the red blood cells the necessary part of the blood needed for the growth of the influenza bacillus, but that it was the hemoglobin in the cells that was the essential element. Though in pure cultures this organism will not grow without hemo- globin, in the presence of certain other organisms it will grow abun- dantly for successive culture generations without blood. Meissner and others found that it would grow well with members of the diphtheria group, and with the pyogenic cocci. We have found, contrary to others, that staphylococci tend to inhibit growth. Resistance and Length of Life. — The influenza bacillus is very sensitive to desiccation; a pure culture diluted with water and dried is destroyed with certainty in twenty-four hours; in dried sputum the vitality, according to the completeness of drying, is retained from twelve to forty-eight hours. It does not grow, and soon dies in water. In blood- bouillon cultures at 20° C. it retains its vitality for several weeks. In moist sputum it is difiicult to determine the duration of its life, since the other bacteria overgrow and make it impossible to find it. It probably can remain alive for at least two weeks. The bacilli are 410 GROUP OF HEMOGLOBINOPHILIC BACILLI very readily killed by chemicals, disinfectants, and succumb to boiling within one minute and to 60° C. within five minutes. Eifect on Animals. — The bacillus of influenza is only slightly virulent for experimental animals. It is, however, definitely though moderately pathogenic for some animals, especially the rabbit, and several observers have found that such pathogenic power (from one blood-agar slant to yxr of a slant) may be decidedly increased by successive passage through the susceptible animal. Guinea-pigs are also quite susceptible to some strains. Pure, abundant cultures are always obtained from the heart after intraperitoneal inoculation. WoUstein has been able to produce cerebrospinal meningitis in the monkey by subdural inoculation. Toxic products cause toxic symptoms in rabbits only when inoculated in large quantities. Immunity. — Short immunity may be established after an attack, though in some cases animals seem to be more sensitive to a second attack. That rabbits may be hyperimmunized is shown in the process of obtaining antibodies. At least in three experiments made by Pfeiffer on monkeys, these animals, after recovering from an inoculation with bacilli, seemed to be much less susceptible to a second injection. Complement-fixation. — Specific antibodies may be obtained in most rabbits, which definitely fix complement. The greatest difficulty in demonstrating the phenomenon of complement-fixation with this group of organisms, is the preparation of suitable antigens. We have found that with the use of only small quantities of heated blood in the medium and by employing the method of shaking the cultures and then incubating overnight to help the autolysis, very satisfactory antigens have been obtained. Definite groups among these bacilli have been demonstrated by this reaction. Pathogenicity for Man. — The invasion of the body by the influenza bacillus is more widespread than was earlier supposed. Very frequently the influenzal process invades portions of the lung tissue. In severe cases a form of pneumonia is the result, which is lobular and purulent in character and accompanied by symptoms which may be somewhat characteristic for influenza, or, again, almost identical with broncho- pneumonia due to the pneumococcus. The walls of the bronchioles and alveolar septa become densely infiltrated with leukocytes, and the spaces of the bronchial tubes and alveoli become filled. The influenza bacilli are found crowded in between the epithelial and pus cells and also penetrate the latter. There may be partial softening of the tissues or even abscess formation. Bacilli are found in fatal cases to have penetrated from the bronchial tubes not only into the peribronchitic tissue, but even to the surface of the pleura, and in a few cases they have been obtained in pure cultures in the pleuritic exudation. The pleurisy which follows influenza, however, is usually a secondary infection, due to the streptococcus or pneumococcus. Presence in Other Parts of the Body.— Influenza bacilli are found at times in the blood during the early days of an acute infection, while there is marked fever (Ghedini found them in 50 per cent, of his cases); THE INFLUENZA BACILLUS 411 and sometimes in bad cases in young children a septicemia develops before death. It is found at times in otitis media accompanying influenza, and has been found in cases of meningitis (Wollstein), in many cases of acute and subacute conjunctivitis (Williams) and in several cases of peritonitis, appendicitis, and cystitis. The general, cerebral, gastric, and other symptoms produced are usually due to the absorption of the toxic products of the specific organism, these poisons being particularly active in their effects on the central nervous system. Presence of Influenza Bacilli in Chronic I^uenza and in Tuber- culosis. — Ordinarily influenza runs an acute or subacute course, and not infrequently it is accompanied by mixed infections with the pneumo- coccus and streptococcus. Pfeiffer was the first to draw attention to certain chronic conditions depending upon the influenza bacillus. Bacilli may be retained in the lung tissue for months at a time, remaining latent a while, and then becoming active again, with a resulting exacer- bation of the disease. Consumptives frequently carry influenza bacilli for years and are particularly susceptible to attacks of influenza. Williams, in the examination of sputa in cases of pulmonary tuberculosis, found abundant influenza bacilli to be present in a large proportion of the samples of sputum from consumptives, and this not only in winter but also in the summer, when nO influenza was known to be present in New York. Taken together with results elsewhere, this indicates that at all times of the year many consumptives carry about with them influenza bacilli, and that very likely many healthy persons as well as persons suffering from bronchitis also harbor a few. Given favorable conditions, we have at all times the seed to start an epidemic. Epidemiology. — ^The discovery of this bacillus enables us to explain many things, previously unaccountable, in the cause of epidemic in- fluenza. We now know, from the inability of the influenza bacillus to exist for long periods in dust, that the disease is not transmissible for great distances through the air. We also know that the infective material is contained only in the secretions. Sporadic cases or the sudden eruption of epidemics in any localities from which the disease has been absient for a long time, or where there has been no new importation of infection, may possibly be explained by assuming that the bacilli, as already mentioned, often remain latent in the lungs or bronchial secretions or secretions of the conjunctiva for many months, and per- haps years, and then become active again, when under favorable circumstances they may be communicated to others. Bacteriological Diagnosis. — This is of importance for the identification of clinically doubtful cases, which, from their symptoms, may be mistaken for bronchitis, pneumonia, or tuberculosis. In acute uncomplicated cases the probable diagnosis can be frequently made by microscopic examinations of stained preparations of the sputum. In chronic cases or those of mixed infection few or many bacilli may be found and the culture method may be necessary to give even a probable diagnosis. The bacillus of influenza is not readily separated by its morphological, staining, and cultural peculiarities from 412 GROUP OF HEMOOLOBINOPHILIC BACILLI other bacteria belonging to the influenza group, and at present by the spread method it is almost impossible to identify it certainly. Examination of Sputum for Influenza Bacilli.— 1. Sputum coughed from the deeper air passages and not from throat scraping should be used. 2. The sputum should be expectorated into a sterile bottle, which should then be placed immediately in cracked ice to transport to the laboratory. 3. Blood-agar plates should be made by placing a drop of fresh rabbit or horse blood, obtained aseptically, on the centre of a hardened agar plate. 4. One of the more solid masses of the sputum should be taken from the bottle with sterile forceps and placed on a plain agar plate. A small portion of this mass should be separated with a sterile platinum needle and drawn through the blood on the blood-agar plate from the centre out in different directions. The larger part of what is left of this small portion is then placed in a similar manner over a second blood agar, and from this to a third, sterilizing the needle between the transfers. The plates should be placed in the incubator at about 36° C. for twenty-four hours. 5. After the plates are planted two smears should be made, from the sputum, one stained by Gram and the other by weak carbol-fuchsin. v 6. After twenty-four hours the plates are examined under low power. The influenza colonies use up the hemoglobin, and in parts of the blood-agar plate where the blood is of right thickness such colonies show as almost clear white areas surrounded by the red blood. With a higher power (No. 6 or 7 objective), if such areas seem to be made up of fine indefinite granulations, they are prac- tically sure to be influenza colonies. Most influenza colonies are more highly refractive than other light colonies, and they show this characteristic best when they grow on the edge of a blood mass. Many influenza colonies also show heap- ings in the centre. Influenza colonies growing away from the blood cells are less characteristic in appearance and less easily differentiated from other similar bacteria. 7. Fishings from the influenza-like colonies should be planted on blood- agar tubes, and if, after twenty-four hours in the thermostat, the resulting growth should consist of influenza-like organisms, plantings should be made on plain agar. The first generation on plain agar may show slight growth because of the blood carried over from the original tube, but the second gen- eration should show no growth if the organism is the influenza bacillus. 8. The agglutination characteristics of the cultures should be tested in the serum from a rabbit injected with a single typical culture, and in the serum from one injected with a number of cultures. The agglutination tests should be carried out in order to gain knowledge in regard to their worth. The cultures tested in the Research Laboratory have shown considerable variation. Other Bacilli Resembling the Influenza Bacillus. — There are a number of bacilli which differ slightly in morphology and growth in culture from the characteristics of the typical influenza bacillus. These were grouped under the name "pseudo-influenza bacilli." But so far there have been shown no characteristics distinct enough to separate these hemoglobinophilic organisms into other than "strains" or "varie- ties" of the one specie, B. influenzce. For example, the influenza-like bacilli found first in whooping-cough by Jochmann and others, Miiller's "trachoma bacillus," Koch- Weeks' bacilli, the bacilli foimd by Cohen in meningitis, and those reported occasionally in other parts of the body — all of them seem to be so closely related that they should be considered one species or, at the most, varieties of one species until more specific characteristics can be demonstrated. INFLUENZA-LIKE BACILLI IN CONJUNCTIVITIS 413 Relation of the Clinical Symptoms to the Bacterial Excitant.— There is no doubt that other infections are also included under the clinical forms of influenza, and during an epidemic of bronchopneumonia, irregular types of lobar pneumonia, and cases of bronchitis frequently have symptoms so closely alike that the nature of the bacteria active in the case is very frequently different from that supposed by the clinician. Thus in four consecutive autopsies examined by the writers the influenza bacillus was found almost in pure culture in one case believed, from the symptoms, to be due to the pneumococcus, and entirely absent in two of the three believed to be due to it. Except for these examinations the clinician would be of the opinion that he had clearly diagnosed bacteriologically the cases, while in fact he had been wrong in three of the four. The striking symptoms in acute respiratory diseases are frequently due more to the location of the lesions than to the special variety of organisms producing them. In epidemics of influenza there are, of course, many cases which, on account of their characteristic symptoms, can be fairly certainly attributed to the influenza bacillus. Even under these circumstances error may be made, as, for instance, two cases of apparently typical influenza were reported in a household and both showed a total absence of influenza bacilli. The pneumococcus was present in almost pure culture. For Testing the Agglutination of Influenza Bacilli in the Hanging Drop. — Grow the cultures on coagulated blood-agar slants (see p. 409) . When twenty to twenty-four hours old, make a suspension of the bacilli in normal salt solution, controlling the number of bacilli by ex- amining a hanging-drop preparation. The influenza bacilli agglutinate rather slowly, so it usually takes four or five hours to get a good reaction. Serum Therapeutics. — No protective serum has been produced which has been used successfully in human cases, but Wollstein obtained one which was of value in experimentally produced infection in monkeys. Vaccine Treatment. — So far this has not been proven to be of marked value. Certain cases may be slightly helped, but too little is still known about the best conditions for use to make it of universal value. Sensi- tized vaccines have been said to give good results. INFLUENZA-UKE BACILU IN CONJUNCTIVITIS (INCLUDING TRACHOMA). The Koch-Weeks' Bacillus. — ^This bacillus was first observed by R. Koch in 1883 while making certain investigations into inflammation of the eye occurring during an epidemic of cholera in Alexandria. It was later, in 1887, more specifically described by Weeks in New York. Weeks obtained it in cultures growing with the xerosis bacillus from cases of "pink-eye," or acute contagious conjunctivitis. Morax stated that he was able to obtain pure cultures only until the third culture 414 GROUP OF HEMOGLOBINOPHILIC BACILLI generation. Others state that on human serum or hydrocele fluid they have obtained growths for many generations. Kamen concludes that it is a strict hemoglobinophile. Our studies led us to agree with this last conclusion. The successive cultures obtained with some sera are probably due to the presence of hemoglobin in amounts too small to be easily detected, but large enough to allow growths of hemoglobinophilic bacilli. The few differential points claimed between this bacillus and influenza bacilli do not hold (Williams) and so the question of their relationship is not settled. Fig. 144. — Koch-Weeks' bacillus from ("pink-eye") — third generation. X 1000 diameters. (Weeks.) Fig. 145. — Secretion of mucopua from conjunctiva in "pink-eye." X 1000 diam- eters. (Weeks.) Other Microorganisms in Conjunctivitis. — Many organisms are found in diseases of the eye, but few of these present evidence of specific etiology. Morax-Axenfeld Bacillus. — In certain subacute inflammations of the conjunctiva, especially noticeable about the angle of the eyes (angular conjunctivitis), Morax (1896) and later Axenfeld found a bacillus which they consider the cause of the disease. Morphology. — Short (about 2fi long), thick, non-motile bacilli, gener- ally in twos, but sometimes single or in short chains. They take the ordinary stains easily, but are decolorized by Gram's stain. Cultures. — ^At 37° C. the bacilli produce a delicate growth on media containing blood or serum. Later cultures grow slightly on nutrient veal agar. They grow slightly, if at all, at room temperature. Upon serum agar, they form delicate grayish colonies. Upon Loffler's blood serum after twenty-four to thirty-six hours the growth appears as an indentation of the medium due to liquefaction. This liquefaction continues slowly for a variable time. In ascitic broth cloudiness is produced within twenty-four hours. Pathogenicity. — Lower animals so far have shown themselves refrac- tory. In human beings inoculations of pure cultures have produced subacute conjunctivitis. TRACHOMA 415 Bacillus of Zur Nedden.— This bacillus has been found by Zur Nedden in certain ulcers of the cornea and is supposed to be the etiological factor in this disease. Morphology.— Small (usually less than 1/* long) rather slender single, sometimes slightly curved, non-motile bacilli. They may occur as diplo- bacilli, but they do not form chains. They stain easily, sonjetimes faintly at ends. They are decolorized by Gram. Cultures. — ^They are easily grown on all laboratory media. Upon agar, within twenty-four hours they produce rounded, raised, translucent, slightly fluorescent colonies, which are more or less confluent and, under the low power, are rather coarsely granular. Upon potato they form a thick, yellowish growth. Milk is coagulated; gelatin is not liquefied; dextrose is fermented without gas. No indol is produced. Pathogenicity.- — Pure cultures inoculated in the cornea of guinea-pigs produced ulcers. Trachoma. — Many studies have been made on the etiology of trachoma (progressive follicular inflammation of the conjunctiva fol- lowed by cicatrization) and allied conjunctival affections. Halber- stadter and Prowazek (1907) state that the cause of trachoma is a small germ which grows in a characteristic way in the conjunctival epithelial cells. The organism itself, they say, is so small that at first it cannot be seen, only the mantle which it produces is demonstrable. This stains blue with Giemsa, and as the organisms grow in bunches, one sees at first in the neighborhood of the nucleus only a bunch of small, blue, coccus-like bodies. The organism finally appears as a minute red granule within the blue body. As it continues to increase in numbers and size, the blue mantles finally disappear, leaving a mass of small rounded or slightly elongated red bodies. The bodies are only found in the early acute cases. Prowazek named them Chlamydozoa on account of their mantle, and thinks they should occupy a place between bacteria and protozoa. Our studies given below show that there is similarity of these inclu- sions to iiests of growing hemoglobinophilic bacilli. The Koch-Weeks' bacillus has been frequently reported as occurring in trachomatous eyes, from the time of Koch (Collins, Morax, Miiller, etc.). Markus states that the Koch- Weeks' bacillus is the cause of "Schwellungskatarrh" (our papillary conjunctivitis). Miitermilch goes further in declaring that "often repeated infection with the same microorganism, e. g., bacillus of Koch-Weeks, produces a series of exacerbations on an already inflamed conjunctiva and finally produces the picture of trachoma,." Miiller, who isolated hemoglobinophilic bacilli from the largest series of trachoma cases reported, had positive results chiefly in his "acute trachoma" cases. Miiller thought at first that his bacillus was the cause of trachoma, but others thought that it was an influenza bacillus and had nothing specifically to do with trachoma (Zur Nedden, Morax and others). 416 GROUP OF HEMOGLOBINOPHILIC BACILLI We, however, agree partly with both workers. We have demonstrated the continued presence of hemoglobinophilie bacilli in cases showing successively acute, subacute, and chronic inflammation of the con- junctiva and their increase in numbers during acute exacerbations of chronic cases. We have also shown that the bacilli found in cases of "pink-eye" are indistinguishable, thus far, from hemoglobinophilie bacilli found in the chronic cases. Furthermore, we have pointed out a relation- ship between hemoglobinophilie bacilli and trachoma inclusions. In studying closely the morphology in cultures of these baciUi, we were struck by the fact that they frequently grew in more or less dense clumps of extremely minute and irregular coccoid forms. This led us to the conclusion that possibly they form the Prowazek inclusions found in trachoma, and when we found that these bacilli and the inclusions were found coincidently and repeatedly in so many cases diagnosed as acute papillary trachoma (called by us papillary conjunctivitis), the possibility became a probability and we proceeded to study the morphology of the cultures more minutely. The morphology varies somewhat with the age of the inoculated culture, the number of culture generations, the kind of medium, and the strain. After forty- eight hours' incubation the forms become somewhat more irregular. Then in three days most of the bacilli have become extremely minute, many showing only as red- dish granules (the ' ' elementary bodies of Prowazek " ) , while scattered through the culture are swollen spheroidal bodies taking a fainter clear blue stain (the larger ' ' initial bodies of Prowazek ") , in some of which are minute reddish granules (more of the "elementary bodies of Prowazek"). A number of irregular light blue bodies are also scattered through the culture. Where the bacteria are densely grouped more red granules may appear in the centre of the group than at the periphery and more blue bodies at the periphery than in the centre. In short, aU of the changes described by Prowazek and others as characteristic of trachoma inclusions are seen in the growing cultures of these hemoglobinophilie bacilli. Similar day-to-day studies were undertaken with the other types of bacteria found most frequently in the eyes diagnosed as trachoma, e. g., streptococci, staphylococci, gonococci, a minute Gram-negative, non-hemoglobinophilic bacillus not before described, found in a few cases of papillary conjunctivitis, xerosis bacillus, and Micrococcus tetragenus, but in none of these varieties except the gonococcus were similar changes found in the same marked degree. The fact that the gonococcus cultures showed such definite appearances similar to the trachoma inclusions led us to make a special study of a series of ophthalmia neonatorum cases. In aU of the inclusion cases where gonococci are found, apparent transition forms between gonococcus and inclusion are very evident, and we find from a further microscopic study of these slides that the inclusions on the whole present certain characteristics different from those found in our series produced, accord- ing to our hypothesis, by nests of growing hemoglobinophilie bacilli. From this comparative study of "inclusions" and cultures we have reached the following conclusion: In many cases of "papillary conjunctivitis" and a certain number of cases of ophthalmia neonatorum, as well as in a certain number of cases of inflammation of the mucous membranes of other parts of the body (e. g., vagina, urethra), the trachoma inclusions found are due to one or more varieties of hemoglobinophilie bacilli; in a certain number of cases of gonorrheal ophthalmia as well as in gonorrheal inflammation of the mucous membranes of other parts of the body, the trachoma inclusions are due to the gonococcus. According to one of the later THE BORDET-GENGOU BACILLUS 417 reports of Leber and Prowazek and the reports of Noguchi and Cohen, certain inchision conjunctivitis cases may be caused by microorganisms other than the two mentioned above. THE BORDET-GENGOU BACILLUS (B. PERTUSSIS). In 1906 Bordet and Gengou announced that they had discovered the etiological factor of whooping-cough to be a small bacillus found in pre- dominating numbers in whooping-cough sputum. To this organism they gave the name Bacillus pertussis. Their claim that this bacillus is the real cause of whooping-cough they based upon their results with the complement-fixation test which they had been first to describe. Morphology. — The pertussis bacillus is a short, oval rod, varying in size from about 0.2/u to 0.3^ in diameter and from O.Sju to 2ai in length. It occurs singly, sometimes in twos joined at the ends, and very excep- tionally in short chains. Motility. — ^It is non-motile. Staining. — ^The pertussis bacillus is decolorized by Gram's method. It is stained faintly by the ordinary aniline dyes. Bipolar staining is demonstrated very well by Gram's method and by toluidin blue (p. 78). Cultivation. — ^The pertussis bacillus grows best at 35° C, to 37° C. It grows slowly at room temperature. It is aerobic, facultative anaerobic. When first isolated in pure culture it grows only upon the glycerin-potato-blood-agar, recom- mended by Bordet and Gengou (p. 105). In later generations it grows more or less capriciously, upon the ordinary culture media. Isolation. — The pertussis bacillus is isolated from sputum with difficulty, owing to the fact that in cultures it is frequently overgrown by other organisms found in the respiratory tract. The sputum should be collected from the patient during the early sta,ge of the disease, best in the first week. The thick grayish portion of the sputum should be selected for the culture. This material is streaked over the surface of a plate of Bordet-Gengou medium. The plates are incubated at 35° C. to 37° C. In forty-eight hours to three days very minute, dis- crete, elevated colonies appear. When these colonies occur in abun- dance, the blood at their periphery, is lighter red. In pure cultures this lightning of the blood is marked and may appear as hemolysis. Identification. — (a) Differential Diagnosis by Cultmre. — There are other bacilli occurring in whooping-cough sputum so closely resembling the pertussis bacillus in morphological and staining characteristics that they cannot be distinguished in smears. These bacilli, however, can be differentiated by their growth upon various culture media. The most important of these organisms is the influenza bacillus, which never grows alone without the presence of hemoglobin in the culture medium. There is frequently found a Gram-negative bacillus which makes a pro- 27 418 GROUP OF HEMOOLOBINOPHILIC BACILLI fuse growth upon all media from the first generation in pure culture. The following table gives the chief points of differentiation: Growth on Bordet- Coagulated Glycerin- Plain Bordet-Gengou Gengou horse-blood ascites-agar agar slants. plates. slants. slants. slants. B. pertussis .... Lightens the First generation After several After several After several medium. in pure culture generations generations generations abundant non- abundant abundant tenacious spreading tenacious tenacious growth moist growth growth in growth in occurs in twenty-four forty-eight forty-eight slowly. hours. hours. hours. B. influenzae Darkens the First generation. First genera- Never grows. Never grows. medium. delicate growth. tion: abundant moist spread in twenty- four hours. Intermediate group of Markedly ^ First generation : First genera- First genera- First genera- Gram-negative bacilli. lightens the abundant tion: pro- tion: pro- tion: .pro- mediimi. moist spread fuse moist fuse moist fuse moist in twenty-four spread spread spread hours. in twenty- in twenty- in twenty- four hours. four hours. four hours. (6) By Agglutination. — Agglutinins are easily obtained in rabbits for the B. pertussis and by absorption tests we have shown that all strains tested by us belong to the same species. It is easy therefore to identify any strain isolated. It is not so easy, however, to diagnose a human case of whooping-cough by this method since agglutinins are not produced to any extent in the natural affection. (c) Complement-fixation. — Bordet and Gengou regarded the com- plement-fixation as the main support for their claim of the specificity of their organism. They report positive complement-fixation in the majority of human cases tested by them. Other investigators, how- ever, have reported negative tests of complement-fixation. The irregular reports given of the complement-fixation test in human beings may be accounted for by different methods used without making comparative studies and sufficient controls. The question of the best method of making the complement-fixation test is still under experi- ment (p. 199). Different observers have used different methods of preparing both antigen and serum and have employed different hemo- lytic systems. In experimental animals, specific antibodies are produced which give a positive complement-fixation test. Pathogenicity. — The specific pathogenicity of the pertussis bacillus for man still lacks proof, though the evidence is strongly in favor of this organism being the cause of pertussis. Several investigators have reported its pathogenicity for monkeys, dogs, rabbits, and guinea-pigs. Mallory claims that a certain number of bacilli between the cilia of the epithelial cells in the trachea and bronchi constitute the specific lesion. He claims that he has fulfilled Koch's laws by the finding of these bacilli in experimental animals in the same situation and by the recovery of the culture from these animals. He acknowledges, however, that his results in animals are complicated by the fact that these animals THE BACILLUS OF SOFT CHANCRE 419 are frequently infected by the Bacillus hronchisepticus (accepted as the cause of distemper in dogs) which is morphologically similar to the pertussis bacillus and that it apparently has the same power (as Theor bold Smith and others have pointed out) to cling to the cilia of the epithelial cells in the respiratory tract. In human beings Mallory states, judging from the examination of hundreds of controls, only pertussis cases showed this lesion. The results of treatment by vaccine are given in Part III. THE BACILLUS OF SOFT CHANCRE. This bacillus was fii'st specifically described and obtained in pure culture by Ducrey in 1889. An experimental inoculation is followed in one or two days by a small pustule. This soon ruptures and a small round depressed ulcer is left. About this other pustules and ulcers develop which tend to become confluent. The base of the ulcer is covered with a gray exudate and its edges are undermined. There is no induration such as in the syphilitic chancre. The secretion is seropurulent and very infectious. The process usually extends to the neighboring lymphatics, which become swollen and may result in abscesses. These are known as "buboes." Morphology. — About 1.5/* long and 0.4/* thick, growing often in chains and in cultures, sometimes twisted together in dense masses. It stains best with carbol-fuchsin, and shows polar staining. It is Gram-negative. Cultural Characteristics. — The following method of cultivation has given the best results: Two parts of liquefied agar at 50° C. are mixed with one part human, dog, or rabbit blood. The blood from the cut carotid of a rabbit may be allowed to run directly into the agar tube, to which the pus from the ulcerated bubo is then added in proper proportion, and the whole placed in the incubator at 35° C. The pus may be obtained by puncture and aspiration from the unbroken ulcer, or if the ulcer is already open it is first painted with tincture of iodine and covered with collodion or sterile gauze. After twenty-four to forty- eight hours, some pus having collected under the bandage, inoculations are made from it. The bacillus grows well also in uncoagulated rabbit- blood serum or in condensation water of blood agar. In twenty-four to forty-eight hours, on the surface of the media, well-developed, shining, grayish colonies, about 1 mm. in diameter may be observed. The colonies remain separate, but they increase in numbers after further transplantation. The best results are obtained when the pus is taken close to the walls of the abscess. Smears show isolated bacilli or short parallel chains with distinct polar staining. The characteristic soft chancre is produced in man after inoculation of cultures. Animals in general cannot be infected, but positive results have been obtained with monkeys and cats. The organisms are especially characteristic in the water of con- densation from blood agar, the bacilli being thinner and shorter, with 420 GROUP OF HEMOGLOBINOPHILIC BACILLI rounded ends; sometimes long, wavy chains are found. In rabbit- blood serum at 37° C. a slight clouding of the medium is produced and small flakes are formed, consisting of short bacilli or moderately long, curved chains, showing polar staining. The bacillus lives several weeks on blood agar at 37° C, but it soon dies in cultures on coagulated serum. All other ordinary culture media so far tried have given negative results, and even with the media described, development is difficult and often fails entirely. The chancre bacillus possesses but little resistance to deleterious outside influences. Hence the various antiseptic bandages, etc., used in treatment of the affection soon bring about recovery by preventing the spread of inoculation chancre. REFERENCES. BoEDET et Gengou: Add. de 1' lost. Past. 1906, xx, 573. Malloby and Hobner: Jour. Med. Res. 1912, xxvii, 115; 1913, p. 391. OliMSTEAD AND PoviTZKY: JouT., Med. Res. 1916, xxxiii, 379. Povitzky: Arch, of Int. Med., 1916, xvii, 279. Weeks: New York Med. Ree., 1887, xxvi, 571. Williams and Co-workebs: Jour. Inf. Dis., 1914, xiv, 261. CHAPTER XXXI. MICROORGANISMS BELONGING TO THE HEMORRHAGIC SEPTICEMIA GROUP. A NUMBER of bacilli of similar characteristics have been described as causing certain infectious diseases of lower animals, marked by the appearance of hemorrhagic areas throughout the body (hemorrhagic septicemia of Hueppe). The bacilli are short, non-motile, non-spore- bearing organisms. They exhibit bipolar staining and are Gram-nega- tive. They do not liquefy gelatin. They are found in rabbit septicemia, fowl cholera, swine plague, and a similar disease in cattle. The bacillus of bubonic plague seems to be closely related to the bacteria of this group. BACILLUS OF CHICKEN CHOLERA. (Bacillus Amsepticus.) In 1880 Pasteur carried on some fundamental studies on the bacillus of chicken cholera. The bacillus was isolated from a widely disseminated acute disease of fowls and smaller birds. Characteristics. — It is a short (0.5-1/j long), non-motile bacillus with marked polar staining. In general, its characteristics are similar to those of other members of this group. Pathogenicity. — Pure cultures are very pathogenic for chickens and rabbits, less so for sheep, pigs, and horses. Chickens are infected even by feeding minute amounts. A septicemia is produced which is rapidly fatal. BACILLUS OF SWINE PLAGUE. (Bacilliis Suisepticus.) This organism is morphologically and culturally similar to the B. ■avisepticus. It differs in pathogenesis in that it is naturally a disease of swine, characterized by a more or less chronic bronchopneumonia followed by septicemia. The gastro-intestinal tract is not markedly affected. The disease is generally fatal in young pigs. The "bacillus of hog cholera" (see p. 355) may often be found as a mixed infection with the B. suisepticits. BACILLUS OF BUBONIC PLAGUE (BACILLUS PESTIS). Historically we can trace the bubonic plague back to the third century. In Justinian's reign a great epidemic spread over the Roman empire and before it terminated destroyed in many portions of the country nearly 50 per cent, of the people. The fourteenth century saw the whole 422 MICROORGANISMS BELONGING TO SEPTICEMIA GROUP of Europe stricken with this "black death." Europe and America have of late been practically free, but in India the disease still breaks out in all its horrors so that at the present time over 500,000 persons die annu- fl* Fig. 146. — Bacillus pestis from agar culture. X 1100 diameters. Fig. 147. — Bacillus pestis from bouillon culture. X 1100 diameters. ally from it. Among the most fatal forms of infection is that of the lungs. Pneumonic cases are not only very serious, but they readily spread the infection. The bacillus exciting the disease was discovered simulta- neously by Kitasato and Yersin (1894) during an epidemic of the bubonic plague in China. It is found in large numbers in the seropurulent fluid from the recent buboes charac- teristic of this disease and in the lymphatic glands; more rarely in the internal organs except in pneumonic cases when the lungs and sputum contain immense numbers. It occurs in the blood in acute hemorrhagic cases and shortly before death. It also occurs in malignant cases in the feces of men and animals. The bacillus, as we have stated, is closely allied to the hemorrhagic septi- cemia group. Morphology. — The bacilli in smears from acute abscesses or infected tissues are, as a rule, short, thick rods with rounded ends. The central portion of the bacillus is slightly convex. The bacilli are mostly single or in pairs. Bacilli in short chains occur at times. The length of the bacilli varies, but on the average is about Fig. 148. — Involution forms on salt agar. (KoUe and Wassermann.) BACILLUS OF BUBONIC PLAGUS 423 1.6;tt (1.5/i to 1.7/i), breadth 0.5/* to 0.7//. Besides the usual oval form, the plague bacillus has many exceptional variations which are charac- teristic of it. In smears, especially from old buboes, one looks for long bacilli with clubbed ends (similar to involution forms (Fig. 148), yeast- like forms, and bladder shapes. Some of these stain with difficulty. Staining. — ^They stain readily with the ordinary aniline dyes, and especially well with methylene blue, the ends being usually more deeply colored than the central portion; they do not stain by Gram's method. Biology. — ^An aerobic, non-motile bacillus. Grows best at 30° to 35° C. Does not form spores. Grows on the usual culture media, which should have a slightly alkaline reaction. Does not liquefy gelatin. Grows well on blood-serum media. It grows rapidly on glycerin agar, forming a grayish-white surface growth. The bacilli appear, as a rule, as short, plump, oval bacilli, but a few present elongated thread forms which are very characteristic. In bouillon which is kept undisturbed a characteristic appearance is produced, the culture medium remaining clear while a pellicle forms on the surface from which projections sprout downwai'd (stalactite formation) toward a granular or grumous deposit which forms on the walls and on the bottom of the tube. In bouillon and most fluid media the growth is in the form of short or medium chains of very short, oval bacilli, which look almost like streptococci. It does not coagulate milk, but produces a slight acidity. It produces no indol in peptone media. Pathogenicity. — Plague is a rodent disease transmissible to man. The most important rodent is the rat. Mice may become infected Other rodents may be infected and be the means of maintaining the disease in endemic areas, as the marmot in Thibet or the ground squirrel in California. The tarbagan was primary source of infection in the epidemic of pneumonic plague in Manchuria (1910-11). Other species of rodents may later be found to be reservoirs of the disease. These animals as well as guinea-pigs are easily infected artificially by feeding or application of the bacilli to the mucous membranes or to the skin. In the last, infection is sure to follow if a slight puncture or scratch is present. Monkeys and rabbits can also be infected. The bacilli may lose their virulence after prolonged artificial cultivation. In rodents the disease may be either acute or chronic. A septicemia develops as a terminal event and fleas feeding at this time become infected. In man, the disease is bubonic pneumonic or septicemic in type. Septicemia is the usual mode of termination of the bubonic and pneumonic types of the disease. The diagnosis of natural infection in rats is made macroscopically, although occasionally the disease does occur without evident lesions. In acute plague the engorgement of the subcutaneous bloodvessels and the diffuse pink color of the subcutaneous tissues and muscles is strongly diagnostic. . The superficial lymph nodes are very much enlarged and frequently surrounded by edema or hemorrhagic areas. The spleen is very much enlarged and soft. The liver is mottled with small hemor- rhages and yellowish punctate areas of necrosis. Generally, there is an 424 MICROORGANISMS BELONGING TO SEPTICEMIA GROUP excess of fluid in the pleural cavity. In the more chronic disease, abscesses of the peripheral lymph nodes or more commonly mesenteric or splenic purulent or caseous foci are found. Frequently the rats submitted for examination are badly decomposed but pure cultures of the plague bacillus may be readily obtained by applying the material from the lesions to the freshly shaven abdomen of a guinea-pig. The plague bacilli penetrate the skin through the slight scarification due to shaving, whereas the other bacteria do not and a general infection results. Epidemiology. — The disease is maintained in the rodents. Although direct contagion may occur, the most important mode of transfer is by rodent fleas. The bacilli taken up with infected blood multiply in the digestive tract of the flea but the mouth parts seem not to remain infected. Fleas may remain infected for weeks or even one or two months if the Fig. 149. — Bacilli in smear from acutely inflamed gland. temperature is low. The mechanism of infection is probably the deposi- tion of infected feces during the act of feeding or regurgitation of infected blood from a previous meal. Plague bacilli can invade the tissues through a flea bite, thus if a non-infected flea is allowed to bite a rat and a drop of plague cultures be placed on the bite infection results. Bubonic plague in man is due to transfer of infection by means of fleas. The evidence for this is only indirect based on the facts that rodent fleas will bite man, the parallelism between outbreaks of plague in man with the prevalence of the disease in rats and the seasonal rise in the number of fleas, the lymphatic involvement suggesting primary skin infection, etc. The pneumonic type may develop after infection by fleas due to blood invasion and pulmonary localization. Should this occur an outbreak of pneumonic plague may develop due to man-to- man contagion. This is most likely to occur during cold weather and under conditions of close contact. This form of disease is extremely contagious, with a mortality of 90 per cent, or more. The mortality of bubonic plague varies from 30 to 90 per cent. BACILLUS OF BUBONIC PLAGUE 425 Other insects than fleas may possibly be a factor in infection, for example bed-bugs. It is possible that human vermin were a factor in the widespread outbreaks of "black death" during the middle ages. Immunity. — Like typhoid infection, a single attack of the plague bacillus protects, with rare exception, from a second infection. Yersin, Calmette, and Borrel have succeeded in immunizing animals against the bacillus of bubonic plague by the intravenous or intraperitoneal injection of dead cultures, or by repeated subcutaneous inoculation. They also succeeded in immunizing rabbits and horses, so that the serum afforded protection to small animals, after subcutaneous injection, of virulent cultures, and even cured those which had been inoculated, if administered within twelve hours after injection. The serum has considerable antitoxic as well as bactericidal properties. It also contains specific agglutinins which may be made use of in diagnosis. For use of vaccine and serum see Part III. Duration of Life Outside of the Body. — In cultures protected from the air and light the plague bacilli may live ten years or more (Wilson) . In the bodies of dead rats they may live for two tnonths. In sputum from pneumonic cases the bacilli lived ten days. Upon sugar sacks, food, etc., they may live six to fifteen days. Resistance to Deleterious Influences. — ^The bacilli resemble the colon bacilli in their reaction to heat and disinfectants. Bacteriological Diagnosis. — ^Material is obtained in bubonic plague by puncture or incision of a lymph node, in pneumonic plague the sputum is employed. Direct smears if plague-like bacilli are present are valuable in making a rapid presumptive diagnosis, cultures and cutaneous inocu- lation of guinea-pigs should also be made. Antemortem blood cultures are frequently positive especially late in the disease. In postmortem examinations the heart blood and spleen should be examined. Wher- ever the material is badly contaminated or even decomposed the cuta- neous method of inoculation of guinea-pigs should be resorted to. For rodents see p. 423. Plague-like Disease in Rodents. — ^McCoy and Chapin (1912) found an organism {Bacillus tularense) in a disease of Calif ornian ground squirrels, which show lesions similar to those of plague. The bacilli have been cultivated by McCoy and Chapin on an egg-yolk medium. Wherry and Lamb' report two cases of conjunctivitis and lymph- adenitis in man due to this bacillus, as well as an epidemic among wild rabbits. 1 Jour. Am. Med. Assn., 1914, Ixiii, 2041. CHAPTER XXXII. THE ANTHRAX BACILLUS. Anthrax is an acute infectious disease which is very prevalent among animals, particularly sheep and cattle. Geographically and zoologically it is the most widespread of all infectious disorders. It is much more common in Europe and in Asia than in America. The ravages among herds of cattle in Russia and Siberia and among sheep in certain parts of France, Hungary, Germany, Persia, and India are not equalled by any other animal plague. Local epidemics have occasionally occurred in England, where it is known as splenic fever. In this country the disease is rare. In infected districts the greatest losses are incurred during the hot months of summer. The disease also occurs in man as the result of infection either through the skin, the intestines, or in rare instances through the lungs. It is found in persons whose occupations bring them into contact with animals or animal products, as stablemen, shepherds, tanners, butchers, and those who work in wool and hair. Two forms of the disease have been described — the external anthrax, or maglignant pustule, and the internal anthrax, of which there are intestinal and pulmonary forms, the latter being known as "wool-sorters' disease." Owing to the fact that anthrax was the first infectious disease which was shown to be caused by a specific microorganism, and to the close study which it received in consequence, this disease has probably contributed more to our general knowledge of bacteriology than any other infectious malady. Davaine, in 1863, announced to the French Academy of Sciences the results of his inoculation experiments, and asserted the etiological relations of the microorganism to the disease, ^ith which his investigation showed it to be constantly associated. Pollender corroborated this statement. In 1877 Koch, Pasteur, and others established its truth by obtaining the bacillus in pure cultures, and showing that the inoculation of these cultures produced anthrax in susceptible animals as certainly as did the blood of an animal recently dead from the disease. Morphology. — Slender, cylindrical, non-motile rods, having a breadth of l/i to 1.25 II,, and ranging from 2/i or 3^ to 20/* or 2i)ix in length. Some- times short, isolated rods are seen, and, again, shorter or longer chains or threads made up of several rods joined end-to-end. In suitable culture media very long, flexible filaments may be observed, which are frequently united in twisted or plaited cord-like bundles. (See Figs. 151 and 152.) These filaments in hanging-drop cultures, before the development of spores, appear to be homogeneous or nearly so; MORPHOLOGY 427 but in stained preparations they are seen to be composed of a series of rectangular, deeply stained segments. When obtained directly from the blood of an infected animal the free ends of the rods are slightly rounded, but those coming in contact with one another are quite square. In cultures the ends are seen to be a trifle. thicker than the body of the cell and somewhat concave, giving the appearance of joints of bamboo. At one time much stress was laid upon these peculiarities as distinguishing marks of the anthrax bacillus; but it has been found that they are the effects of artificial cultivation and not necessarily characteristic of the organism under all conditions. The bacillus is inclosed in a transparent- envelope or capsule, which in stained preparations (from albuminous material) may be distinguished by its taking on a lighter stain than the deeply stained rods which it surrounds. Fig. 150. — Anthrax bacillus. X 900 diameters. Agar culture. Fig. 151. — Spores heavily stained (in specimen red). Bodies of disintegrating bacilli faintly stained (in specimen blue). X 1000 diameters. Spore Formation. — Under favorable conditions in cultures spores are developed in the bacilli. These spores are elliptic in shape and about one and a half times longer than broad. They first appear as small, refractive granules distributed at regular intervals, one in each rod. As the spore develops the mother-cell becomes less and less distinct, until it disappears altogether, the complete oval spore being set free by its dissolution. (See Fig. 151 and Plate III, Fig. 22). Irregular sporulation sometimes takes place, and occasionally there is no spore formation, as in varieties of non-spore-bearing anthrax. Sporeless varieties have been produced artificially by cultivating the typical anthrax bacillus under certain conditions, among which may be mentioned the addition of antiseptics, as carbolic acid, and of continued high temperature (43° C). Varieties differing in their pathogenic power may also be produced artificially. Pasteur produced an " attenu- 428 ANTHRAX BACILLUS ated virus" by keeping his cultures for a considerable time before replanting them upon fresh soil. Anthrax cultures containing spores retain their vitality for years; in the absence of spores the vitality is much more rapidly lost. When grown in liquids rich in albumin the bacilli attain a considerable degree of resistance; thus dried anthrax blood has been found to retain its virulence for sixty days, while dried bouillon cultures only did so for twenty-one days. Dried anthrax spores may be preserved for many years without losing their vitality or virulence. They also resist a comparatively high temperature. Exposed in dry air they require a temperature of 140° C. maintained for three hours to destroy them; but suspended in a liquid they are destroyed in four minutes by a temperature of 100° C. Staining .^The anthrax bacillus stains readily with all the aniline colors, and also by Gram's method, when not left too long in the decolor- izing solution. In sections good results may be obtained by the employ- ment of Gram's solution in combination with carmine, but when only a few bacilli are present this method is not always reliable, as some of the bacilli are generally decolorized. McFadyean-Heine Methylene-blue Reaction. — In imperfectly fixed film preparations (pass through flame three times in a second with film side up) the capsule disintegrates. When a solid film is stained for a few seconds in an old solution of methylene blue. Washed in water and dried with filter paper, the bacteria are surrounded by a varying amount of a reddish-purple amorphous or fixed granular deposit. McFadyean says this does not occur with other morphologically similar bacteria. Biology. — The anthrax bacillus grows easily in a variety of nutrient media at a temperature from 18° to 43° C, 37° C. being the most favorable temperature. Under 12° C. no development takes place as a rule, though by gradually accustoming the bacillus to a lower tempera- ture it may be induced to grow under these conditions. Under 14° C. and above 43° C. spore formation ceases. The lower limit of growth and i)i sporulation is of practical significance in determining the question whether development can occur in the bodies of animals dead from anthrax when buried at certain depths in the earth. Kitasato has shown that at a depth 1.5 meters the earth in July has a temperatue of 15° C. at most, and that under these conditions a scanty sporulation of anthrax bacilli is possible, but that at a depth of 2 meters sporulation no longer occurs. The anthrax bacillus is aerobic — that is, its growth is considerably enhanced by the presence of oxygen — but it grows also under anaerobic conditions, as is shown by its growth at the bottom of the line of puncture in stick cultures in solid media; but under these conditions it no longer produces the peptonizing ferment which it does with free access of air. Furthermore, the presence of oxygen is absolutely necessary for the formation of spores, while carbonic acid gas retards sporulation. This explains, perhaps, why sporulation does not take place within the animal body either before or after death. It is also capable of leading a saprophytic existence. The bacillus is non-motile. BIOLOGY 429 Growth in Gelatin. — In gelatin-plate cultures, at the end of twenty- four to thirty-six hours at 24° C., small, white, opaque colonies are developed, which, under a low-power lens, are seen to be dark gray in the centre and surrounded by a greenish, irregular border, made up of wavy filaments. As the colony develops on the surface of the gelatin these wavy filaments spread out, until finally the entire colony consists of a light gray, tangled mass, which has been likened to a Medusa head (Fig. 152). At the same time the gelatin begins to liquefy, and the colony is soon surrounded by the liquefied medium, upon the surface of which it floats as an irregular, white pellicle. In gelatin-stick cultures at first, development occurs along the line of puncture as a delicate white thread, from which irregular, hair-like projections soon extend perpen- dicularly into the culture medium, the growth being most luxuriant Fig. 152. — Colonies of Bacillus anthracis upon gelatin plates: a, at the end of twenty- four hours; b, at the end of forty-eight hours. X 80. (F. FlUggej) near the surface, but continuing also below. At the eiid of two or three days' liquefaction of the medium commences at the surface and gradually progresses downward. Growth on Agar. — ^The growth on agar-plate cultures in the incubator at 37° C. is similar to that on gelatin, and is still more characteristic and beautiful in appearance. A grayish-white layer is formed on the surface within twenty-four hours, which spreads rapidly and is seen to be made up of interlaced threads. Growth in Bouillon. — ^The growth is characterized by the formation of flaky masses, which sink as a sediment to the bottom of the tube, leaving the supernatant liquid clear. Pathogenesis. — ^The anthrax bacillus is pathogenic for cattle, sheep (except the Algerian race), horses, swine, mice, guinea-pigs, and rabbits. Rats, cats, dogs, chickens, owls, pigeons, and frogs are but little sus- ceptible to infection. Small birds — the sparrow particularly — are 430 ANTHRAX BACILLUS somewhat susceptible. Man, though subject to local infection and occasionally to internal forms of the disease, is not as susceptible as some of the lower animals. In susceptible animals the anthrax bacillus produces a true septicemia. Among test animals mice are the most susceptible, succumbing to very minute injections of a slightly virulent virus; next guinea-pigs, and then rabbits, both of these animals dying after inoculation with virulent bacilli. Infection is most promptly produced by introduction of the bacilli into the circulation or the tissues, but inoculation by contact with wounds on the skin also causes infection. It is difficult to produce infection by the ingestion even of spores; but it may readily be caused by inhalation, particu- larly of spores. Subcutaneous injections of these susceptible animals results in death in from one to three days. Comparatively little local reaction occurs immediately at the point of inoculation, but beyond this there is an extensive edema of the tissues. Very few bacilli are found in the blood in the larger vessels, but in the internal organs, and especially in the capillaries of the liver, the kidneys, and the lungs, they are present in great numbers. In some places, as in the glomeruh of the kidneys, the capillaries will be seen to be stuffed full of baciUi, and hemorrhages, probably due to rupture of capillaries by the mechanical pressure of the bacilli which are developing within them, may occur. The pathological lesions in animals infected by anthrax are not marked except in the spleen, which, as in other forms of septicemia, is greatly enlarged. Occurrence in Cattle and Sheep.^Cattle and sheep are affected chiefly with the intestinal form of anthrax, infection in these animals commonly resulting from the ingestion of food containing spores. The bacillus itself, in the absence of spores, is quickly destroyed by the gastric -juice. The disease usually takes a rapid course, and the mortality is high — 70 to 80 per cent. The pathological lesions consist of numerous ecchymoses, enlargement of the lymphatic glands, serous, fatty, and hemorrhagic infiltration of the mediastinum and mesentery, of the mucous membranes of the pharynx and larynx, and particularly of the duodenum, great enlargement of the spleen, and parenchymatous changes in the lymphatic organs. The blood is very dark and tar-hke. Bacilli are present, especially in the lymph spaces, in enormous masses. Sheep are also subject to external anthrax, infection taking place by way of the skin; cattle are seldom infected in this way; At the Fig. 153. — Section of liver from mouse dead of anthrax Septicemia. X 1000 diameters. (From Itzerott and Niemann.) OCCURRENCE IN MAN 431 point of inoculation there develops a hard, circumscribed boil — the so-called anthrax carbuncle; or there may be diffuse edema with great swelling of the parts. When death occurs the appearances are similar to those in intestinal anthrax, except that the duodenum is usually less affected; but in all cases metastasis occurs in various parts of the body, brought about, no doubt, by previous hemorrhages. Occurrence in Man. — ^The disease does not occur spontaneously in man, but always results from infection, either through the skin, the intestines, or occasionally by inhalation through the lungs. It is usually produced by cutaneous infection through inoculation of exposed sur- faces — the hands, arms, or face. Infection of the face or neck would seem to be the most dangerous, the mortality in such cases being 26 per cent., while infection of the extremities is rarely fatal. External anthrax in man is similar to this form of the disease in animals. There are two forms: malignant pustule or carbuncle, and, less commonly, malignant anthrax edema. In malignant pustule, at the site of inoculations, a small papule develops, which becomes vesicular. Inflammatory induration extends around this, and within thirty-six hours there is a dark, brownish eschar in the centre, at a little distance from which there may be a series of small vesicles. The brawny induration may be extreme. There may also be considerable edema of the parts. In most cases there is no fever; or the temperature at first rises rapidly and the febrile phenomena are marked. Death may take place in from three to five days. In cases which recover the symptoms are slighter. In the mildest form there may be only slight swelling. Malignant anthrax edema occurs in the eyelids, and also in the head and neck, sometimes the hand and arm. It is cha,racterized by the absence of the papule and vesicle forms, and by the most extensive edema. The edema may become so intense that gangrene results; such cases usually prove fatal. The bacilli are found on microscopic examination of the fluid from the pustule shortly after infection; later the typical anthrax bacilli are often replaced by involution forms. In this case resort may be had to cultures, animal inoculation, or examination of sections of the extirpated tumor. The bacilli are not present in the blood until just before death. Along with the anthrax bacilli pyogenic cocci are often found in the pustule penetrating into the dead tissue. Internal anthrax is much less common in man; it does, however, occur now and then. There are two forms of this: the intestinal form, or mycosis- intestinalis, and the pulmonic form, or wool-sorters' disease. Intestinal anthrax is caused by infection through the stomach and intestines, and results probably from the eating of raw flesh or unboiled milk of diseased animals. That the eating of flesh from infected animals is comparatively harmless is shown by Gerlief, who states that of 400 persons who were known to have eaten such meat not one was affected with anthrax. On the other hand, an epidemic of anthrax was produced among wild animals, according to Jansen, by feeding them on infected 432 ANTHRAX BACILLUS horse flesh. It is evident, therefore, that there is a possibility of infection being caused in this way. The recorded cases of intestinal anthrax in man have occurred in persons who were in the habit of handling hides, hair, etc., which were contaminated with spores; in those who were conducting laboratory experiments, and rarely it has been produced by the ingestion of food, such as raw ham and milk. The symptoms produced in this disease are those of intense poisoning, chill, followed by vomiting, diarrhea, moderate fever, and pains in the legs and back. The pathological lesions are similar to those described in animals. Wool-sorters' disease, or pulmonic anthrax, is found in large establish- ments in which wool and hair are sorted and cleansed, and caused by the inhalation of dust contaminated with anthrax spores. The attack comes on with chills, prostration, then fever. The breathing is rapid, and the patient complains of pain in the chest. There may be a cough arid signs of bronchitis. The bronchial symptoms in some instances are pronounced. Death may occur in from two to seven days, The patho- logical changes produced are swelling of the glands of the neck, the formation of foci of necrosis in the air passages, edema of the lungs, pleurisy, bronchitis, enlargement of the spleen, and parenchymatous degenerations. Prophylaxis against Anthrax Infection. — Numerous investigations have been undertaken with the object of preventing infection from anthrax. The efforts of Pasteur to effect immunity in animals by preventive inoculations of "attenuated virus" of the anthrax bacillus opened a new field of productive original research. Following in his wake many others have devised methods of immunization against anthrax infection; but the one adopted by Pasteur, Chamberland, and Roux has alone been practically employed on a large scale. According to these authors, two anthrax cultures of different degrees of virulence attenuated by cultivation at 42° to 43° C, are used for inoculation. Vaccine No. 1 kills mice, but not guinea-pigs; vaccine No. 2 kills guinea-pigs, but not rabbits. The animals to be inoculated — viz., sheep and cattle — are first given a subcutaneous injection of one to several tenths of a cubic centimeter of a four-day-old bouillon culture of vaccine No. 1 ; after ten to twelve days they receive a similar dose of vaccine No. 2. Prophylactic inoculations given in this way have been widely employed with apparently good results. Serum Treatment. — ^The serum of immunized animals has been used in Italy with apparently some good results. In this country Eichhorn has reported good results. Bacterial Cultures for Diagnosis. — ^The detection of the anthrax bacillus is ordinarily not difficult, as this organism presents morpho- logical, biological, and pathogenic characteristics which distinguish it from all other bacteria. In the later stages of the disease, however, the bacilli may be absent or difficult to find, and cultivation on artificial media and experimental inoculation in animals are not always followed by positive results. Even in sections taken from the extirpated pustule it is sometimes difficult to detect the bacilli. In such cases only a DIFFERENTIAL DIAGNOSIS 433 probable diagnosis of anthrax can be made. It should be remembered that the bacilli are not found in the blood until shortly before death, and then only in varying quantity; thus blood examinations often give negative results, though the bacilli may be present in large numbers in the spleen, kidneys, and other organs of the body. The suspected material should be streaked over nutrient agar in Petri plates and inoculated in mice. Differential Diagnosis. — Among other bacteria which may possibly be mistaken for anthrax bacilli are Bacillus subtilis and the bacillus of malignant edema. The former is distinguished by its motility, by various cultural peculiarities, and by being non-pathogenic. The latter differs from the anthrax bacillus in form and motility, in being decolorized by Gram's solution, in being a strict anaerobe, and in various pathogenic properties. (See also description in chapter on Water Analysis.) The diagnosis of internal anthrax in man is by no means easy, unless the history points definitely to infection in the occupation of the individual. In cases of doubt cultures should be made and inoculations performed in animals. REFERENCES. EiOHHORN, A.: Experiments in Vaccination against Antlirax, U. S. Dept. Agiic, Bull. No. 340, 1915. 28 CHAPTER XXXIII. ANAEROBIC BACILLI. In this chapter we are grouping several species of bacteria which have in common only the characteristic that they cannot grow in pure cultures in the presence of oxygen. Those of most interest to us are B. tetani, B. (anthracis) symptomatici, B. (edematis) maligni, B. wehhii, B. botulinus and B. fusiformis. THE BACILLUS AND THE BACTERIOLOGY OF TETANUS. Tetanus is a disease which is characterized by a gradual onset of general spasms of the voluntary muscles, commencing in both man and the horse most often in the muscles of the jaw and neck, and extending in severe cases to the muscles of the body. The disease is usually associated with a wound received from four to fourteen days previously. Tetanus has been reported for many centuries. The writings of Hippocrates clearly describe the symptoms. In 1884 Nicolaier, under Fliigge's direction, produced tetanus in mice and rabbits by the subcu- taneous inoculation of particles of garden earth. The Italians, Carle and Rattone, had just before demonstrated that the pus of an infected wound from a person attacked with tetanus could produce the same disease in rabbits. Finally, Kitasato, in 1889, obtained the bacillus of tetanus in pure culture and described his method of obtaining it and its biological characters. Occurrence in Soil, etc. — The tetanus bacillus occurs widely through- out the world as a common inhabitant of the soil, especially in places where manm-e has been thrown, being abundant in many localities not only in the superficial layers, but also at the depth of several feet. It has been found in many different substances and places — in hay dust, in horse and cow manure (its normal habitat is the intestine of the herbivora), in the mortar of old masonry, in the dust from horses' hair, and in the dust in rooms of houses, barracks, and hospitals. The tetanus bacilli are more numerous in certain localities than in others — for example, some parts of Long Island and New Jersey have become notorious for the number of cases of tetanus caused, by small wounds — and they are fairly common in New York City. As a rule they are more abundant in regions where the temperature is high. In some islands and countries in the tropics cases of puerperal tetanus and tetanus in the new;born are very frequent. Tetanus bacilli are found in the intestines of about 15 per cent, of horses and calves living in the vicinity of New York City. They are also present to a somewhat less extent in the intestines of other animals and of man. THE BACILLUS AND THE BACTERIOLAGY OF TETANUS 435 Morphology. — From young gelatin cultures the bacilli appear as motile, slender rods, with rounded ends, 0.3/i to 0.5At in diameter by 2n to 4(1 in length, usually occurring singly, but, especially in old cultures, often growing in long threads. They form round or nearly round spores, thicker than the cell (from In to 1.5fi in diameter), occupy- ,----_ ing one of its extremities and giving to /^. A the rods the appearance of small pins / V ""^ (Fig. 154). _ _ _ / . \ Staining. — It is stained with the ordinary, aniline dyes, and is not de N> ordmary. aniime dyes, and is not de- ,' .4 lation over the entire surface of the _, , ,1 , i .J ] abdomen, and the muscles present a ' ' ■ '■ .S ■. I dark red or black appearance, even *• \ V ' ' more intense in color than in malig- ■ ' j ' , nant edema, and there is a consider- " ' / / " ^ ' able development of gas. The ^ ' , , lymphatic glands are markedly ..1 ^-■- , nyperemiC. ^ ^ ^ Fie, 155, — Bacilli of symptomatic anthrax, The disease occurs chiefly in showing spores. (After Zettnow.) cattle, more rarely in sheep and goats; horses are not attacked spontaneously — i.'e., by accidental infec- tion. In man infection has never occurred. The usual mode of natural infection by symptomatic anthrax is through wounds which penetrate not only the skin, but the deep, intercellular tissues; some cases of infection by ingestion have been observed. The pathological findings present the Conditions above described as occurring in the experimental animals. Distribution Outside of the Body. — Symptomatic anthrax, like anthrax and malignant edema, is a disease due to soil infection, being confined especially to places over which infected herds of cattle have been pastured. By contamination of deep wounds acquired by animals in infected pastures, the disease is spread. The spores are extremely resistant. Toxins. — Under favorable conditions extracellular toxins are formed. Injections of the toxin into animals excite the production of antitoxins. 446 AN AMOBIC BACILLI Preventive Inoculations. — Recovery from one attack of symptomatic anthrax protects an animal against a second infection. Active immunity can be produced by vaccines of attenuated organisms. A dried powder of the muscles of animals which have succumbed to the disease is used as a vaccine and subjected to a suitable temperature to insure attenua- tion of the virulence of the spores contained therein. Two vaccines are prepared — a stronger vaccine by exposing a portion of the powder to a temperature of 85° to 90° C. for six hours, and a weaker vaccine by exposing it for the same time to a temperature of 100° to 104° C. In- oculations are made with this attenuated virus into the end of the tail — ^first the weaker and later the stronger. The results obtained from this method of preventive inoculation seem to have been very satisfactory. THE MAUGNANT EDEMA BACILLUS. Bacilli of this group are widely distributed, being found in the super- ficial layers of the soil, in putrefying substances, in foul water. This bacillus was discovered (1877) by Pasteur in animals after infection with putrid flesh, and named by him "vibrion septique." He did not obtain it in pure culture. Koch and Gaflky (1881) carefully studied this microorganism, described it in detail, and gave it the name "Bacillus edematis maligni" (Fig. 156). Fig. 156. — Bacilli of malignant edema. 1, bacilli; 2, with .spores; 3 and i, deep colonies in dextrose nutrient agar. (KoUe and Wassermann.) Morphology.— The edema bacillus is a rod of from 0.8m to 1^ in width, and of very varying length, from 2ai to lO/ti or more, according to the conditions of its cultivation and growth. It is usually found in pairs, joined end-to-end, but may occur in chains or long filaments. It is motile, and does not produce a capsule. It forms spores which are situated in or near the middle of the body of the rods, exceptionally near the ends. The spores vary in length and are oval in form, being often of greater diameter than the bacilli, to which they give a more or less oval shape. The bacilli stain readily by the usual aniline colors employed and are Gram-amphophile. BACILLUS WELCH 1 1 GROUP 447 Biology.^An obligate anaerobe growing best on or in albuminous media but also growing well on ordinary media especially if available carbohydrates are present. Growth occurs at 20° C, but is more rapid and abundant at 37° C. Growth on Agar. — On dextrose agar plates the colonies appear as dull, whitish points, irregular in outline, and when examined under a low- power lens are seen to be composed of a dense network of interlacing threads, radiating irregularly from the centre toward the periphery. Growth in Gelatin. — ^The colonies are similar to those on agar, a liquefied zone developing after several days. Resistance. — ^The spores are very resistant and because of this the soil remains infected. Other Media. — ^Milk is coagulated and then digested. Blood serum is liquefied. An offensive odor develops due to the proteolysis. Glucose is fermented with the production of acid and gas. Pathogenicity. — ^Malignant edema is mostly confined to domestic animals, horse, sheep, cattle and swine. It follows the contamination of wounds with infected soil or other infectious material; and also occurs as a complication of surgical operations. It is therefore a frequent contaminant after war wounds. The depth of the wound as well as the introduction of foreign particles and other bacteria are factors in infection, as the inoculation of washed spores frequently fail to produce infection. An extensive hemorrhagic edema of the sub- cutaneous tissues develops from the site of the wound. The serous effusion is frothy from gas production and has a foul odor. As a rule in the larger animals the bacilli do not invade the blood until after death. The bacillus is pathogenic for the smaller laboratory animals, rabbit, guinea-pig, mouse; a septicemia developing as well as the local edema. Immunity. — Recovery from infection is followed by immunity. It is claimed that an extracellular toxin is produced, also a leukocidin. BACILLUS WELCHH GROUP (BACn.LUS AEROGENES CAPSULATUS). The first bacillus of this group to be described minutely was found by Welch in the bloodvessels of a patient suffering with aortic aneurysm; at autopsy, made in cool weather, eight hours after death, the vessels were observed to be full of gas bubbles. Since then it has been found in many cases in which gas has developed from within sixty hours of death until some hours after death. It occurs most frequently after external cutting operations and wounds. These cases are, as a rule, marked by delirium, rapid pulse, high temperature, and the development of emphy- sema and discoloration of the diseased area or of marked abdominal distention when the peritoneal cavity is involved. Members of this group are present, as a rule, in the intestinal canal of man and animals and are apt to be found in the dust of hospitals and elsewhere. Herter has shown that they are present in excessive numbers in certain diseases of the digestive tract. These cases are apt to develop anemia. 448 ANAEROBIC BACILLI Different strains of bacilli belonging to this group have appeared under different names and their exact relationship is still uncertain. Thus, B. phlegmonis emphysematosse of Frankel is probably the same as B. welchii. B. perfringens (Veillon and Zuber, 1898) and B. enteritidis sporogenes (Klein, 1895) are closely related if not the same organism. Morphology. — Straight or slightly curved rods, with rounded or sometimes square-cut ends; somewhat thicker than the anthrax bacilli and varying in length; occasionally long threads and chains are seen. The bacilli in the animal body, and sometimes in cultures, are enclosed in a transparent capsule. Spores are usually absent in the tissues and often in cultures. Dunham showed that the culture isolated by Welch formed spores when grown on blood serum. Some strains since isolated make spores readily. Fig. 157.- -Baoillus aerogenes capsulatus. 1, bacilli; S, spores; 3, culture in dextrose- nutrient agar. » Biology. — Anaerobic, non-motile, non-liquefying bacilli. They are positive to Gram, but are more easily decolorized than many bacteria. Growth is rapid at 37° C, in the usual culture media in the absence of oxygen, and is accompanied by the production of gas. Nutrient gelatin is not liquefied, but it is gradually peptonized. On agar colonies are developed which are from 1 to 2 mm. or more in diameter, grayish-white in color, and in the form of flattened spheres, ovals, or irregular masses, covered with hair-like projections. Bouillon is diffusely clouded, and a white sediment is formed. Milk becomes acidified and coagulated, then partially digested, giving a worm-eaten appearance to the clot. A large amount of butyric acid is produced. Isolation. — When quantities up to 2.5 c.c. of fresh bouillon cultures are injected into the circulation of rabbits and the animals killed shortly after the injection, the bacilli develop rapidly, with an abundant forma- tion of gas in the bloodvessels and organs, especially the liver. This procedure is one of the best methods of obtaining the bacilli. The material suspected of containing the bacillus alone or associated with other bacteria is injected intravenously into rabbits, which are killed five minutes later and kept at 37° C. for sixteen hours, and cultures made from the liver and heart blood. BACILLUS BOTU LINUS 449 Pathogenicity. — Its pathogenicity is usually not marked in healthy animals, although Dunham found that the bacillus taken freshly from human infection is sometimes very virulent. As we have said bacilli of this type are one of the frequent infections after irregular unclean wounds such as those received in war. In these infections there is marked destruction of tissue, especially in the muscles. This is due possibly to the large quantity of butyric acid produced from the glycogen of the muscles. Many attempts have been made to demonstrate a specific toxin production by these bacilli (Harde, Weinberg) but the results have not been clear-cut. Robertson has made a study of the isolation and types of anaerobes found in wounds. It is suggested by Welch that in some of the cases in which death has been attributed to the entrance of air into the veins the gas found at the autopsy may not have been atmospheric air, but may have been produced by this or some similar microorganism entering the circulation and developing shortly before and after death. The same may be true for gas in the uterine cavity. BACILLUS BOTULINUS. This bacillus, while not pathogenic for man, is as first shown by Van Ermengen the causative agent in a characteristic type of food poisoning, the symptoms being due to the toxin produced by the growth of this bacillus in foods. The clinical symptoms of this type of food poisoning are neuroparalytic in character. There are secretory disturb- ances as increase or suppression of the salivary secretions or a thick secretion of mucus in the mouth and pharynx. Disturbances of accom- modation, ptosis, double vision, and dysphagia are the common motor symptoms. Obstinate constipation and retention of urine as well as disturbances in heart action and respiration occur. Fever is absent. Death is not uncommon and is due to respiratory paralysis. Morphology. — ^The bacilli are large, 4 to 6m long and 0.9 to 1.2^ wide, with somewhat rounded ends. They are slightly motile, Gram- positive, and have oval spores, usually terminal. Short chains may be produced. Biology. — ^It is anaerobic, growing best at 22° to 25° C. Gelatin is liquefied and glucose fermented with the production of acid and gas. A butyric acid odor develops in cultures. The colonies on gelatin are yellowish, translucent, coarsely granular, the granules moving slowly when liquefaction begins. The older colonies are brownish, more opaque and show fine thorn-like extensions. Resistance. — ^The spores are not very resistant to heat, being killed in fifteen minutes at 85° C. or in one-half hour at 80° C. Toxin Production. — ^A thermolabile extracellular toxin is produced in glucose-broth cultures. Guinea-pigs, rabbits, mice, .cats and monkeys are susceptible to the toxin and succumb not only to injection but also when the toxin is given by mouth. As little as 0.0001 c.c. may produce symptoms. After an incubation period, dyspnea, convulsions and 29 450 ANAEROBIC BACILLI paralysis occur and death is due to respiratory paralysis. The paralyses are due to degeneration of the ganglion cells of the anterior horn and of the bulbar centres. According to Dickson this is secondary to dis- turbances in blood supply due to thrombosis associated with meningeal hemorrhage. An antitoxin has been produced. Foods Involved. — Meat and meat preparations, such as canned or pickled meats and sausages are the usual offending foods. Until recently it was thought that preserved or, canned fruits or vegetable products were not a factor. Cases due to such foods have been reported and Dickson has shown that the toxin develops in a medium of peas or beans. All these foods have one thing in common, they are prepared for weeks or months before they are consumed. The source of contamination is not definitely known, though the organism has been isolated from pigs' feces. Diagnosis of Botulism. — The diagnosis usually rests on the clinical manifestations. Verification may be attempted by examination of the suspected food. A thick emulsion of the food is made in saline and from it glucose gelatin plates and fermentation tubes (with tissue) are inoculated. A portion of the suspension can be heated (60° C, one-half hour) and similarly inoculated. From the growth in the fermentation tube, heated and unheated anaerobic plates are made. The pure cultures obtained are then tested for toxin production by animal inoculation. Saline extracts of the suspected food, if they produce characteristic symptoms in experimental animals, confirm the clinical diagnosis in man. Prophylaxis. — Although a rancid butter odor may be present in contaminated foods, it may be slight and not recognized. Cooking will destroy the toxin. Cleanliness in preparation is an aid in preventing contamination and also in limiting the numbers of associated bacteria which, by their growth, would aid in establishing anaerobic conditions. Brine for pickling should contain at least 10 per cent, of salt as this concentration prevents the growth of the bacilli. Incomplete steriliza- tion is the source of danger with canned goods. BACILLUS FUSirORMIS (BACILLUS OF VINCENT'S ANGINA). This organism together with spirochetes was found by Plaut and Vincent in pseudomembranous inflammation of the throat (Vincent's Angina). The constancy of its presence and the preponderance of their numbers in smears from this condition suggests strongly their etiological connection with the disease. Fusiform bacilli and spirochetes are also encountered in gangrene, noma, ulcerative stomatitis, gingivitis, dental caries and even around the gum margins of dirty teeth, especially if there are deposits of tartar. Although B. fusiformis is present in many con- ditions, either because of the greater virulence of certain strains or because of reduced resistance of the tissues they, together with the spirochetes are probably the essential agent in the production of Vincent's angina and noma and act as a contributing factor in other lesions. The fusiform baciUi are anaerobic and have been cultivated. The most BACILLUS TYPHI-EX ANTHEM ATICI 451 successful methods of isolation and cultivation are those of Xrumwiede and Pratt who studied fifteen strains from various conditions. The typical bacillus is double-pointed, containing one or more granules. In culture the morphology is \'ariable. In fluid media there is a tendency to produce filamentous types, which may form tangled, thread-like masses. The colonies are characterized by thread-hke outgrowths. The fifteen cultures studied fell into saccharose fermenting and saccharose non-fermenting groups, but this difl^^erence had no correlation to the source of the culture. The fusiform bacillus and the spirochetes accompanying it in the lesion were never encountered in cultures made from single colonies although Tunnicliff claimed that the fusiform bacillus and the spirochete were only different forms of the same organism. Fig. 158. — Vincent's bacillus with accompanying spirochetes. BACILLUS TYPHI-EXANTHEMATICL Historical. — The bacillus typhi-exanthematici was first isolated by Plotz, in 1914, fjom the blood of individuals suffering from typhus fever. A year later, Plotz, Olitsky and Baehr published the report of an exten- sive study of bacteriological, serological and animal investigations which contained much evidence that this organism is the causative agent in typhus fever. Continued investigations by these workers and by others working in cooperation with them have made the evidence stronger. The following description is from their reports and from a private com- munication from Dr. Baehr, giving information of work to be reported in the Journal of Infectious Diseases early in 1917. Morphology. — ^The organism is a small, slender bacillus, the average length being about one micron. In young cultures the organisms are small and uniform in size; in very old subcultures there is often an ad- mixture of various sized bacilli from coccoid forms up to some measuring two microns in length. The bacilli are usually straight, though slightly curved forms occur occasionally. The ends are slightly pointed, less often rounded. (Fig. 159.) With special stains an occasional organism will show a fine polar body at one end, more rarely at both. 452 ANAEROBIC BACILLI Reaction to the Gram Stain. — The organism in subcultures is Gram- positive. According to Olitsky, Denzer and Husk and Baehr and Plotz, the colonies in occasional blood cultures may consist only of exceedingly minute baciUi which are completely decolorized by Gram's method. In subsequent subcultures these bacilli always become Gram- positive. A similar experience was the rule in organisms isolated from typhus infected lice. Fig. 159.— Typhus bacillus. (Plotz.) Cultural Characteristics. — The bacillus typhi-exanthematici is an obligate anaerobe. After artificial cultivation for many months, it has- been possible to obtain slight aerobic growth with occasional strains. The organism, as far as tested, grows only upon a medium containing_ ascitic or hydrocele fluid and glucose. Although growth occurs in fluid media con- taining these ingredients, the optimum medium contains 1 part of ascitic fluid and 2 parts of 2 per cent, glucose agar. Even on slants of this medium a large amount of culture material must be subinooulated in order to obtain growth and it is advisable to smear it in thick streaks upon the surface. The sub- cultures upon such slants incubated in Buchner tubes containing equal parts of pyrogallic acid and 30 per cent, sodium hydrate solution usually show evi- dences of growth after about three days. The growth is profuse by the sijrth or seventh day and at that time is soft and creamy in character, raised above the surface and of a glistening white color. In the medium itself a diffuse clouding gradually develops as the growth increases (the precipitation phenom- enon of Libman) . Strains of the Bacillus typhi-exanthematici isolated in the United States, Mexico, Serbia and Russia are identical in their action upon sugars. They invariably ferment glucose, maltose, galactose and inulin with marked acid production and precipitation of protein, but have no action upon saccharose, lactose, raffinose, arabinose, dextrin and mannit. Method of Isolating the Organism from the Blood. — The method used in ma,king the anaerobic blood culture is that of Liborius-Veillon, using the ascitic fluid glucose agar recommended by Libman as the optimum medium for bac- terial cultivation. The essential ingredient of the medium is ascitic or hydrocele fluid. It must be clear, free of bile or blood pigment and should possess a specific gravity of more than 1015. Filtered ascitic fluid or one which contains a preservative or has been sterilized by heating should not be used. Pleural fluids are prob- ably useless. The technic of the culture consists in withdrawing with a syringe and BACILLUS TYPHI-EXANTHEMATICI 453 needle 15 c.c. of blood from a vein and dividing this among 8 large test-tubes (20 X 2 cm.) half-filled with melted 2 per cent, glucose agar. To each tube in turn, ascitic or hydrocele fluid equivalent to at least one-third of the volume of the agar {i. e., 6 to 10 c.c.) is added. The contents are then poured once or twice into another sterile test-tube in order to secure thorough mixing. After thorough solidification of the mixture, each tube is covered with a layer of plain agar, 2 or 3 cm. deep, and then the cotton stopper paraffined. The tubes are then incubated at 37° C. Appearance of Colonies. — Colonies usually appear in the culture tubes in about nine or ten days. Occasionally they may occur as early as the fifth or as late as the nineteenth day. They first appear as small opaque spots which, by direct light, are white. During the subsequent two or three days they grow rapidly larger, usually assuming a " Y" shape, and develop a brownish zone of pre- cipitation in the medium about them. In consistency the growth is always exceedingly soft, and the color, especially of the older colonies, is pale brown. The organism may be isolated from the blood during the entire febrile course of the disease, from the first to the last day. In two cases Plotz was still able to grow the bacillus from the bipod twelve and thirty- six horn's respectively after the crisis. In the mild cases only one to three colonies may develop in the eight tubes; in severe cases as many as ten to twenty may be present. During chills which occur frequently at the onset of the illness and occasionally during its first week numerous colonies develop from the blood indicating that enormous numbers of the bacteria may be present in the blood. The percentage of positive blood cultures also varies with the severity of the disease, as can be seen from the following table taken from an article by Baehr and Plotz : Tvoe of disease Mortality ~ Niimber of Percentage with positive "^ percent. cases studied, blood cultures percent. New York typhus (Brill's disease) 0.2 34 53 Russian ts^phus ....5.5 24 79 Mexican typhus .... 20.0 8 100 Balkan typhus, 1914 . 18 to 60 7 100 The organisms can also be pretty regularly isolated from monkeys and guinea-pigs in which the disease has been produced by the inoculation of typhus blood. Pathogenicity. — On artificial media the organism loses its patho- genicity very rapidly. Organisms which have been grown on artificial media for less than a week are sometimes still pathogenic for guinea-pigs but those cultivated for longer periods are always non-pathogenic. After an incubation period more, irre'gular than of typhus fever, the animals may develop a febrile illness during which in several cases the bacillus has been recovered from the blood. Development of Antibodies after Typhus Fever. — Specific agglutinins, precipitins, opsonins and complement-fixing antibodies are regularly present in the blood of typhus convalescents. They usually appear during the second week of the disease, increase as the crisis is approached, reach their maximal titer during the first or second week of convalescence and usually persist in the blood for months, in one case for two and one- half years. The curve of the course of development of the antibodies in this disease is typically an immunity curve. 454 ANAEROBIC BACILLI Baehr has made the observation that individuals who have been intimately exposed to typhus infection, may develop antibodies in their blood without having fever or other clinical evidences of the disease. None of these people developed the disease, although subsequently they were repeatedly exposed to infection. Some adults who are not known to have had typhus fever have been shown to have immune bodies in their blood. Prophylactic Immunization. — The repeated subcutaneous inoculation in human beings of vaccine made from these organisms is followed by the development in their blood of specific antibodies. In guinea-pigs, how- ever, although they develop immunity after an infection due to the blood, such an antibody production does not occur and thus far it has also been found impossible to confer immunity upon these animals by vaccination. Plotz, Olitsky and Baehr have recently carried out investigations in the Balkans and in Volhynia, Russia, where typhus fever was epidemic, upon the value of prophylactic immunization in human beings, and have reported that vaccination with Bacillus typhi-exanthematici markedly reduces the incidence of the disease. Transmission by the Body Louse. — ^The stomach of typhus infected lice contains immense numbers of bacilli which are morphologically identical with the Bacillus typhi-exanthematici. Such bacilli have been seen in typhus lice in Mexico by Ricketts and Wilder and Olitsky, Denzer and Husk. In smear preparations made from typhus lice, these bacillary organisms are decolorizable by Gram's method. Olitsky, Denzer and Husk succeeded in cultivating them from the lice by the anaerobic method described above, and found that after several genera- tions on artificial media the bacilli gradually became Gram-positive. The bacilli were then morphologically, culturally and serologically iden- tical with the Bacillus typhi-exanthematici and in the early subcultures were pathogenic for guinea-pigs. REFERENCES. Andebson and Leake: A Method of Producing Tetanus Toxin, Jour. Med. Research, 1915, xxxiii, 239. Baehr and Plotz: Blood Culture Studies on Typhus Exanthematicus in Serbia, Bulgaria and Russia, Jour. Infect. Dis., 1917, xx, 201. Denzeb and Olitsky: Jour. Inf. Dis., 1917, xx. 99. Hehde: Compt. rend. Soc. de biol., 1915, Ixxviii, 134, Hekteb: .Jour. Biol. Chem., 1906, ii, 1. Kbumwiede and Pbatt: Jour. Inf. Dis., 1917 xii, 199, and xiii, 438. Olitsky, Denzer and Husk: The Etiology of Mexican Typhus Fever (Tabardillo) , Jour. Am. Med. Assn., 1916, Ixvi, 1692; Jour. Infect. Dis., 1916. Olitksy: Jour. Inf. Dis., 1917, xx, 349. Plotz, Olitsky and Baehb: The Etiology of Typhus Exanthematicus, Jour. Infect. Dis., 1915, xvii, 1. Plotz: The Etiology of Typhus Exanthematicus, Jour. Am. Med. Assn., 1914, Ixii, 1556; La Presse Med., 1914, xliii, 411. RoBEBTBON, MuEiEL: Certain Anaerobes Isolated from Wounds, Jour, of Path, and Bacteriol., 1916, xx, 327. SiMONDS, J. P.: Classification of B. Welchii Group of Bacteria, Jour. Infect. Dis., 1915, i, 31. Weinberg: Proceed. Roy. Soc. Med., 1916, ix, 119. Wilcox, Harriet L.: Jour. Bact., 1916, i, 333. Wildeb: The Problem of Transmission in Typhus Fever, Jour. Am. Med. Assn., 1910, liv, 1373. CHAPTER XXXIV. THE CHOLERA SPIRILLUM (CHOLERA VIBRIO) AND SIMILAR VARIETIES. In 1883 Koch separated a characteristically curved organism from the dejecta and intestines of cholera patients — the so-called "comma bacillus." It was absent from the blood and viscera, and was found only in the intestines; and the greater the number, it was said, the more acute the attack. Koch also demonstrated an invasion of the mucosa and its glands. The organisms were found in the stools on staining the mucous flakes or the fluid with methylene blue or fuchsin, and sometimes alone; by means of cultivation on gelatin they were readily separated from the stools. Numerous control observations made upon other diarrheic dejecta and upon normal stools were negative; the comma bacillus was found in choleraic material only, or occasionally in small numbers in the stools of healthy persons who came in contact with cholera. Soon, however, other observers described comma-shaped organisms of non-choleraic origin. Finkler and Prior, for instance, found them in the diarrheal stools of cholera nostras, Deneke in cheese, Lewis and Miller in saliva. All of these organisms, however, differed in some respects from Koch's comma bacillus, and it has since been proved that none of them is affected by the specific serum of animals immunized to cholera. After a time, therefore, the exclusive association of Koch's vibrio with cholera or those in contact with it became almost generally acknowledged, until now it is regarded by bacteriologists everywhere to be the specific cause of Asiatic cholera. Certain sporadic cases of cholera-like disease, however, are undoubtedly due to other organisms. Morphology. — Curved rods with rounded ends which do not lie in the same plane, of an average of l^n in length and about OAjj. in breadth. The curvature of the rods may be very slight, like that of a comma, or distinctly marked, particularly in fresh unstained prepara- tions where the adhesion of two individuals presents the appearance of a half-circle. By the inverse junction of two vibrios S-shaped forms are produced. Longer forms are rarely seen in the intestinal discharges or from the cultures grown on solid media, but in fluids, especially when grown under unfavorable conditions, long, spiral filaments may develop. The spiral forms are best studied in the hanging drop, for in the dried and stained preparations the spiral character of the long filaments is often obliterated. In film preparations from the intestinal contents in typical cases it will be found that the organisms are present in enormous numbers, and often in almost pure culture. In old cultures 456 CHOLERA SPIRILLUM irregularly clubbed and thickened involution forms are frequent, and the presence in the organisms of small, rounded, highly refractile bodies is often noted. Staining. — The cholera spirillum stains with the aniline colors usually employed, but not as readily as many other bacteria; a diluted aqueous solution of carbol-fuchsin (1 to 10) is the most rehable staining agent. It is decolorized by Gram's method. The organisms exhibit one long, fine flagellum attached to one end (other spirilla often have two or more end flagella). Biology. — The cholera spirillum is aerobic, liquefying, and extremely motile. It grows readily on ordinary culture media, best at 37° C, but also at room temperature, 22° C. Fig. 160. — Contact smear of colony of Fig. 161. — Cholera spirilla preparation cholera spirilla from agar. X 700 diam- from gelatin-plate culture of cholera. X eters. (Dunham.) 800 diameters. Gelatin Plate Cultures. — ^In gelatin plate cultures, characteristic colonies are produced. After twenty-four hours' growth, there is a uniform granular appearance of the surface, which because of the high degree of refraction gives the appearance of being powdered with broken glass. Color is practically, absent or there may be a slight yellowish tint thus differing from B. coli. If growth is continued, liquefaction appears about the colony and its appearance gradually changes. The character- istic colony in gelatin used to be of the greatest practical importance. With the introduction of special selective media, however, the use of gelatin has been practically discontinued. Gelatin Stab Culture.— In gelatin stab culture, a small funnel of lique- faction appears after twenty-four hours. This deepens and broadens, until at the end of a week liquefaction may be complete. Agar.— On agar a moist, shiny, grayish-yellow layer develops. On the surface of alkaline-agar plates the individual colonies are characteristic. They are round, transparent and have a rather distinctive opalescent sheen. This characteristic appearance is made use of for isolation. On Dieudonn4 alkaline-blood agar the colonies are large and circular. On transmitted light there is a glassy transparency, on reflected light they are grayish. On alkaline-egg agar the colonies, when examined BIOLOGY 457 by transmitted light, appear to be deep in the agar and have a dis- tinctive hazy appearance due to the development of a halo about the colony. A zone of clearing may develop where growth is continued. As some fecal bacteria other than vibrios may develop on these media, the development of a typical colony is a great aid in isolation. LofQer's blood serum is rapidly Hquefied at 37° C. Milk is not coagu- lated. Glucose and saccharose are fermented with great rapidity, acid only being formed. Growth in fluid media is abundant and usually characteristic, most so in peptone-water. Peptone-water is diffusely cloudy to a moderate degree, but at the surface the cloud is much denser. Due to the greater supply of oxygen, the vibrios seek the surface and multiply there more freely. Reaction of Media. — Cholera grows best on media that are strongly alkaline to litmus. They can grow in an excess of alkali sufficient to inhibit the growth of many of the bacteria found in feces. This is most evident where an alkali-albumin mixture is used to increase the alkalinity. (See Special Media.) Cholera-red Reaction. — All cholera strains give this reaction. This is important as many non-cholera spirilla do not give this reaction. Changes due to Artificial Cultivation. — Chol- era strains which have been in cultivation for some time show more spiral forms. They grow less typically or not at all on selective media and fluid cultures may develop a wrinkled pellicle. Their digestive powers are also lessened, as gelatin-liquefaction or the Fig. 1 62 .—Choi era colonies in liquefaction of Loffier's serum medium, gelatin; twenty-four to thirty- 17 ... 1 V, • J J u • 1 six hours' growth. X about 20 Variations can also be induced by special diameters. conditions. Hemolysin Production. — Six strains were isolated by Gottshlich at El Tor from Pilgrims to Mecca, who died with diarrheal symptoms but had shown no clinical evidences of cholera. These strains are called "El Tor" strains. They give the serum reactions of cholera vibrios but produce a strong hemolysin. Kraus and Ruffer believed them not to be true cholera, as hemolysin production was considered by the former as an attribute possessed only by non-cholera types. Further investigations have shown that cholera vibrios may develop, lose, or show wide variations in their ability to produce hemolysins. It is apparently a potential power of all vibrios and of no value in differen- tiation. For this reason the El Tor strains must be considered as cholera vibrios. Resistance and Vitality. — Development Outside of Body. — If a culture is spread on a cover-glass and exposed to the action of the air at room temperature the spirilla will be dead at the end of two or three hours, unless the layer of culture is very thick, in which case it may take twenty-four hours or more to kill all the spirilla. This indicates that 458 CHOLERA SPIRILLUM infection is rarely if ever produced by means of dust or other dried objects contaminated with cholera spirilla. The transmission of these organisms through the air, therefore, can only take place for short distances, as by the spray from infectious liquids. The cholera bacillus is also injuriously affected by the abundant growth of saprophytic bacteria. It is true that when associated with other bacteria, if present in large numbers, and if the conditions for th;eir development are particularly favorable, the comma bacillus may at first gain the upper hand, as in the moist linen of cholera patients, or in soil impregnated with cholera dejecta; but later, after two or three days, even in such cases, the bacilli die off and other bacteria gradually take their place. Thus, Koch found that the fluid contents of privies twenty-four hours after the introduction of comma bacilli no longer contained the living organisms; in impure river water they were not demonstrable for more than six or seven days, as a rule. In the dejecta of cholera patients they were found usually only for a few days (one to three days), though rarely they have been observed for twenty to thirty days, and on one occasion for one hundred and twenty days. In unsterilized water they may also retain their vitality for a relatively long time; thus, in stagnant well water they have been found for eighteen days, and in an aquarium containing plants and fishes, the water of which was inoculated with cholera germs, they were isolated several months later from the mud at the bottom. Koch found them in the foul water of a tank in India, used by the natives for drinking purposes. In running river water, however, they have not been observed for over six to eight days. In milk they are finally destroyed by acidity due to the growth of the milk bacteria, in sterilized milk they may sm-vive eight to ten days. For the cholera organisms the conditions favorable to growth are a warm temperature, moisture, a good supply of oxygen, and a considerable proportion of organic material. These conditions are fully met, outside the body, in very few localities. The comma bacillus has the average resistance of spore-free bacteria, and is killed by exposure to moist heat at 56° C. in half an hour, at 80° in five minutes, at 95° to 100° C. in one minute. The bacilli have been found alive kept after a few days in ice, but ice which has been preserved for several weeks does not contain living bacilli. Chemical disinfectants readily destroy the vitality of cholera vibrios. For disinfection on a small scale, as for washing the hands when con- taminated with cholera infection, a 0.1 per cent, solution of bichloride of mercury, or a 2 to 3 per cent, solution of carbolic acid or, better, lysol may be used. For disinfection on a large scale, as for the disin- fection of cholera stools, strongly alkaline milk of lime is an excellent agent. The wash of cholera patients, contaminated furniture, floors, etc., may be disinfected by a solution of 5 per cent, carbolic acid and soap water. For the disinfection of drinking water, chlorinated lime, 1 or 2 parts per million of free chlorine may be used for fairly pure water. Five parts per million would probably be effective even in polluted water. DISTRIBUTION IN THE BODY 459 Pathogenesis. — Not one of the lower animals is naturally subject to cholera. Koch succeeded in producing symptoms and intestinal lesions in guinea-pigs similar to those in man by introducing cultures by catheter after neutralizing the contents of the stomach with a solution of car- bonate of soda and inhibiting peristalsis by the use of opium. Thomas injected a dilute suspension of cholera spirilla into the ear vein of young rabbits, and thus caused symptoms and lesions of the intestinal mucous membrane resembling those of cholera in man. The spirillum was recovered from the feces. Metchnikoff was successful with suckling rabbits by rubbing a small amount of a culture on the teats of a mother rabbit. Intraperitoneal injections with cholera spirilla kill guinea-pigs acutely, but intestinal lesions are rarely found. Accidental Human Infection. — There are several cases on record which furnish the most satisfactory evidence that the cholera spirillum is able to produce the disease in man. In 1884 a student in Koch's laboratory in Berlin, who was taking a course on cholera, became ill with a severe attack of cholera. At that time there was no cholera in Germany, and the infection could not have been produced in any other way than through the cholera cultures which were being used for the instruction of students. In 1892 Pettenkofer and Emmerich experimented on them- selves by swallowing small quantities of fresh cholera cultures obtained from Hamburg. Pettenkofer was affected with a mild attack of choler- ine or severe diarrhea, from which he recovered in a ftew days without any serious effects, but Emmerich became very ill. On the night following the infection he was attacked by frequent evacuations of the characteristic rice-water type, cramps, tympanites, and great prostration. His voice became hoarse, and the secretion of urine was somewhat diminished; this condition lasting for several days. In both cases the cholera spirillum was obtained in pure culture from the dejecta. Finally, there is the case of Dr. Oergel, of Hamburg, who accidentally, while experi- menting on a guinea-pig, allowed some of the infected peritoneal fluid to squirt into his mouth. He was taken ill and died a few days after- ward of typical cholera, though at the time of his death there was no cholera in the city. Lesions in Man. — Cholera in man is an infective process of the epithelium of the intestine, in which the spirilla clinging to and between the epithelial cells produce a partial or entire necrosis and final destruction of the epithelial covering, which thus renders possible the absorption of the cholera toxin formed by the growth of the spirilla. The larger the surface of the mucous membrane infected and the more luxuriant the development of baciUi and the production of toxin, the more pronounced will be the poisoning, ending fatally in a toxic paralysis of the circulatory and thermic centres. On the other hand, however, there may be cases where, in spite of the large number of cholera bacilli present in the dejecta, severe symptoms of intoxication may be absent. In such cases the destruction of epithelium is not produced or is so slight that the toxic substance absorbed is not in sufficient concentration to give rise to the algid stage of the disease, or for some reason the spirilla do not produce toxin to any extent. Distribution in the Body. — ^The cholera spirilla are found only in the intestines and are believed never to be present in the blood or 460 CHOLERA SPIRILLUM internal organs. The lower half of the small intestine is most affected, a large part of its surface epithelium becoming shed. The flakes floating in the rice-water discharges consist mostly of masses of epithelial cells and mucus, among which are numerous spirilla. The spirilla also penetrate the follicles of Lieberkiihji, and may be seen lying between the basement membrane and the epitheUal lining, which become loosened by their action. They are rarely found in the connective tissue beneath, and never penetrate deeply. In more chronic cases other microorganisms play a greater part and deeper lesions of the intestines may occur. Cholera Toxins. — Koch assumed that the severe symptoms of cholera were due to the absorption of a toxin produced by the growth of the vibrio in the intestines. The toxic effects are apparently due to sub- stances which are an integral part of the organism and are only liberated by the breaking down of the vibrio. Suspensions of killed vibrios, when injected into animals, give the same symptoms as living cultures, although quantitatively less toxic. The endotoxin is labile, and is best shown in cultures which are killed by chloroform or by heating to 56° C. for one hour. More active chemicals or a higher degree of heat changes it from a more specific toxin to a niore general protein poison. The bacteria-free filtrates of fresh fluid cultures are only slightly toxic; old cultures, however, due to breaking down of the vibrios may be very toxic. This toxicity is due mostly to substances similar in action to the general class of ptomaines. Kraus has, however, been able to demonstrate, what he considers an extracellular toxin, in young broth cultures. Metchnikoff and Roux have also attempted to prove the existence of an extracellular toxin by growing cholera vibrios in collodion sacs implanted in the peritoneum of guinea-pigs. The production of an antitoxin against such toxins has not been accomplished. Communicability.— Origin of Epidemics.— The two fundamental epidemiological facts are, that the vibrio leaves the body only in the feces, and the mode of infection is by way of the mouth. The feces of the cholera patient during the acute stage of the disease is extremely rich in vibrios, which are at times present in almost pure culture. As the case recovers they decrease in number, but persist after recovery for seven to fourteen days, in exceptional instances longer. In oijly one case have they persisted for three months, that is, chronic carriers do not exist. These persons constitute the "convalescent carriers." In this connection, the mild cases which are undiagnosed or overlooked are important. Another group of persons may act as sources of infec- tion, viz., excrete cholera spirilla in their stools. These are the " healthy or contact carriers." Not all persons who ingest cholera develop the disease. In a number, the vibrio will multiply in the intestine to a limited extent and be excreted in the stools, although no clinical evidences of disease are present. These healthy carriers are important not only as insidious spreaders of infection, but they may be potential cases of cholera. Should their resistance be lowered they may develop the disease. The transfer of the infectious agent to the mouth may occur in SPECIFIC SERUM REACTIONS 461 several ways: by personal contact, by fomites, and by contamination of food and water. Where a moderate number of cases are developing in a district having fair sanitary conditions, contact, especially with mild cases and carriers, or indirectly, fomites or infection of food are the sources of infection. As other factors enter, such as the contamina- tion of the soil and privy-vaults with subsequent infection of well and river water and of green vegetables, the cases increase in number. Where a general water supply becomes contaminated an explosive, widespread epidemic follows. It is easy to understand how localized epidemics are kept active, by the development of healthy carriers, who help maintain the contamination of their surroundings; and, given also the climatic conditions, such as heat and moisture, to favor the continued vitality or even multiplication outside of the body, how endemic foci persist. The transfer of infection by fomites, such as body and bed linen or dishes, etc., is only dangerous when direct. Drying quickly lessens the danger. Flies may. be a factor by mechanical transfer of the virus. The susceptibility to infection of different individuals varies and conditions may lower or raise the resistance of the individual. The occurrence of healthy carriers illustrates this. Such carriers may be very numerous. Abel and Clausen, for example, found that 14 of 17 persons belonging to families of 7 cholera patients, had cholera vibrios in their stools. In a group of immigrants who were exposed, we found 10 per cent, were healthy carriers. The resistance of the individual depends upon his general good health. Gastric and intestinal disorders due to indiscreet eating or drinking, or other causes undoubtedly favor infection or in the case of healthy carriers may cause the development of cholera. At the New York Quarantine Station two such cases promptly developed: one after the administration of a dose of salts, the other after a drinking bout. Cholera Iminuiiity. — Eight to ten days after recovery the serum of cholera patients contains protective substances. If a guinea-pig is injected intraperitoneally with living cholera vibrios, and serum from the patient be given, the pig recovers. Similar protective substances are found in the serum of animals injected with sublethal doses of live vibrios or with killed organisms. The serum is not antitoxic, for although it will protect an animal from a lethal dose of living vibrios, by preventing their multiplication, it has little effect, when a fatal dose of killed organisms or toxic extracts is given. Similarly an animal may be highly immune to the injection of living cultures but on intestinal infection will show no resistance to the poisonous products absorbed. The antibodies present in the seriun are precipitins, agglutinins, opsonins, bacteriolytic and bactericidal substances. Prophylactic Vaccination and Serum Therapy. — See Part III. Specific Serum Reactions. — Agglutinins. — Because of the acuteness of the disease, the agglutination reaction is valueless for diagnosis. It may be of diagnostic help in recovered cases, where no bacteriological 462 CHOLERA SPIRILLUM diagnosis has been made. Agglutination, is used, however, for identifica- tion of the cholera vibrio. In this way cholera and non-cholera vibrios can be separated with certainty, because a cholera-immune serum usually contains few group agglutinins for non-cholera types. Freshly isolated strains agglutinate freely; in fact, seem to give better reactions than stock cultures, although strains vary in their agglutinability; relatively inagglutinable strains such as are encountered among freshly isolated strains of typhoid are not found. PfeifEer Phenomenon. — This phenomenon (see page 178), which is a bactericidal test in vivo, can be employed to differentiate the cholera vibrio from other vibrios. The bactericidal serum should have a high titer, viz., 1 c.c. of a 1 to 1000 dilution should be able to dissolve a loop of cholera vibrios when injected intraperitoneally together. Each factor in the test must be controlled. ISOLATION OF CHOLERA VIBRIO FROM STOOLS. OUTLINE OF PROCEDURE. Feces. > Peptone-water. > Peptone-water. \ /% / : 4* •I' ^ 4- ( Direct } Selective Media f Smear, if positive, hanging drop 1 microscopic > —> J with and without serum. Plate I examination J /Agglutination of\ | for pure cultures for verifica- | \suspicious colony./ [ tion, etc. J Direct Microscopic Examination. — This is of great value in suspected cases but of no use in the examination of carriers. The presence of many typical, extremely motile vibrios warrants a tentative diagnosis of cholera. In exceptional cases, not cholera, a great abundance of vibrios may be found. This led us in one instance unnecessarily to isolate a nurse, who while caring for a cholera patient, developed nervous diarrhea. Peptone-water. — ^Inoculate with feces and incubate six to twelve hours. Examine smears from surface growth. If the vibrios are numerous, pre- pare hanging drops with and without immune serum. If vibrios are not found in the smear, or if too few in number for hanging drop observa- tion, subinoculate into peptone-water, or selective media, or both. Subculture Peptone-water. — This second enrichment is probably never required except in the examination of suspected carriers. In four instances we have found cholera vibrios in subculture where they were not evident in smears from the first peptone culture. This second enrichment helps also to exclude some of the vibrios other than cholera; some will have died out, some will not have enriched beyond the amount present in the first peptone culture; these are not cholera. Smears from the surface growth are examined and if positive, examination ISOLATION OF CHOLERA VIBRIO FROM STOOLS 463 of hanging drops carried on. Where haste is not a factor as in carrier examinations the examination of the first peptone culture may be omitted. In making smears and hanging drops from peptone-water especial care should be taken not to shake the tubes and that the loop is just touched to the surface. In the examination of, smears little time need be spent; if very few vibrios are present, further enrichment is necessary. If they are present in sufficient numbers for agglutination, they are found at once. The use of the surface growth for agglutination is open to certain criticisms, but in practical work it has given us accurate results. Where cholera or cholera-like vibrios, having the same ability to enrich as true cholera, are present, the surface growth is nearly a pure culture. The reliability of the peptone method, as outlined, was best shown in a series of examinations where we found two carriers. In this series 50 per cent, of the stools contained vibrios. The results were checked by the isolation of pure cultures. With some experience a great number of examinations can be carried through in. this way, using peptone- water only, with a minimum of preparation and equipment. Selective Media.— Inoculation may be done directly from the feces or after enrichment in peptone-water. The advantage of such media is that they may be heavily inoculated. The colonies which develop are used for agglutination either in hanging drop, or more convenient, the macroscopic slide method (see under Typhoid). Various modifica- tions of Dieudonne's have been suggested. The alkaline-egg medium has the advantage that a distinctive vibrio colony develops allowing quick selection for agglutination should non-vibrios develop. A second- ary plating on plain agar is necessary, if pure cultures are to be obtained from selective media. The other bacteria, which are only restrained, may be transferred in fishing a colony and thus yield mixed culttues. Alkaline agar may be employed for plating either directly from the feces or from peptone enrichments. As stated, the colony is distinctive and no difficulty Will be encountered if the vibrios are fairly numerous.' Saccharose Peptone-water. — ^This medium has been suggested by Bendick to avoid the time-coiisuming microscopic examination of peptone tubes. The stool is first inoculated in peptone-water and the surface growth then inoculated into this medium. Because of the ability of the cholera vibrio to rapidly split saccharose, decolorization occurs in five to eight hours.' The tubes which do not decolorize in this time can be discarded, those decolorized are examined for vibrios, which, if present, are isolated by plating. Because of the presence of a fermentable sugar, the growth is diffused and the surface is not satisfactory for agglutination. To avoid this difficulty duplicate peptone tubes could be planted, however, and used 1 The direct plating of stools on agar or gelatin is sometimes of practical importance. In no disease but cholera do vibrio colonies approach or exceed in number those develop- ing from the ordinary fecal bacteria. Even without identification by agglutination, such a condition gives us a practically certain diagnosis of cholera. 464 CHOLERA SPIRILLUM for agglutination when the saccharose peptone tubes were decolorized. This method promises to be of the greatest value where many specimens must be examined in the search for carriers. Examination of Suspected Carriers. — ^The simplest procedures possible must be employed when the daily examinations may run into many hundreds or even thousands. The peptone method outlined reduces the bacteriological work very much, especially if the first peptone tubes are not examined. The collection of stools under these circumstances is impracticable. Ordinary swabs moistened in peptone-water may be inserted into the rectum to obtain fecal material. Individual glass specula may be employed to aid in the introduction of the swabs. Where feasible much time is saved by giving an identification card a number and dropping the swab at the time of taking into a similarly numbered tube of peptone-water. The sterilization of the peptone tubes, supported in blocks of wood or racks protected by a cover instead of by cotton plugs, and the use of individual wire loops (sterilized in bundles) for transfers or smears saves a great deal of time. The removal and replacing of cotton plugs and the burning of the platinum loop usually employed is thus avoided. In transferring, the wire is dropped into the second tube. The exposure of the tubes when the cover is removed for inoculation or transfer does not lead to interference by contamination. Isolation from Water. — ^As the cholera vibrios are few in number in water a large volume should be used. About 1 liter of water is taken and 100 c.c. amounts placed in Ehrlenmeyer flasks and to each is added 10 c.c. of a tenfold strength peptone-water. These are shaken and incubated for eighteen hours. The surface growth is then subinoculated in peptone-water for further enrich- ment if necessary and plates made for isolation. Especial care must be taken in fishing as vibrios other than cholera may be present in the water, and there is some evidence that cholera vibrios lose their agglutinability to some extent after living in water. These difiiculties may be surmounted by many fishings and the use of a high titre serum in low dilutions in order to select cultures for final identification. Spirilla (Vibrios) More or Less Allied to the Cholera Spirillum. — ^Various types of spirilla may be isolated from stools, water, and other sources. Some are practically identical, morphologically and culturally, with true cholera. It would be well to limit the term " cholera-like vibrios" to this group. Another group of vibrios, similar to cholera in some respects but differing in others might be termed "non-cholera vibrios." Much of the practical interest in these, types was lost when the sero- logical methods for identification were introduced. In emergency work, where immune serum is not at hand, they are of extreme practical importance. A few of the types are of interest because of their patho- genicity for animals. Some of these vibrios enrich in mixed cultures in peptone-water like true cholera, others enrich to a limited extent or die out. The majority of the vibrios found in stools during routine examinations for cholera spirilla can be excluded culturally if serum be not available. Of 50 vibrios thus isolated we found that 43 did not give the cholera-red reaction. Of the 7 that did, 2 produced no acid from glucose and 4 ALLIED VIBRIOS 465 produced acid and gas. One produced acid only, but could be excluded, as it formed a tenacious pellicle on peptone-water. The following is a short list of vibrios of interest historically or because of their pathogenicity for animals: Vibrio metschnikovii ; source, epizootic gastro-enteritis of fowls, present in intestinal contents and in blood. Typical vibrio with one flagellum, liquefies , gelatin and gives cholera-red reaction. A minute amount of culture inoculated into a cutaneous wound causes a fatal vibriosepticemia in pigeons and guinea- pigs. Vihio massavah; source, stools, considered cholera vibrio when first isolated, fom fla^ella, pathogenicity like Vibrio metschnikovii. Vibrio septicus; source, stools, case of cholerine, cholera-like cultiually and morphologically, minute amounts cause a rapidly fatal septicemia in guinea-pigs. Spirillum finkler-prior ; source, feces in cholera nostras, does not give cholera- red reaction. Vibrio ivanoff and Vibrio berlionensis ; source, former artificially inoculated stools for disiiSection tests, latter, water artificially inoculated to determine viability of cholera vibrio in water. Both are probably variants produced by artificial conditions; they give the immune reactions of cholera spirilla. REFERENCES. Creel: Am. Jour. Public Health, December, 1911, p. 889. Kbumwibde, Pratt and Grund: Jour, of Infect. Dis., 1912, x, 134. CHAPTER XXXV. PATHOGENIC MICROORGANISMS BELONGING TO THE HIGHER BACTERIA (TRICHOMYCETES). Observers are still of different opinions in regard to the classifi- cation of this group of organisms (see Chapter II, p. 41). Foulerton and his associatjes have made an extensive study of this group, both saprophytic and parasitic varieties, and they agree with some others in caUing attention to the acid-fast character of some of the varieties and to the apparent relationship of the group to B. tubercvlosis, B. mallei, and B. diph- therice. To us, however, the relationsliip does not seem to be close enough to place all of these organisms in one group. We have shown (Chapter II) that the apparent branching in B. diphtheriwis not a true branching. If these are not classed with the true bacteria, they should either be put in a group by themselves or be classed with the cladothrix group since their apparent branching takes place in a manner similar to that described as occurring in the latter group. Foulerton considers all organisms in the group classed as higher bacteria as belonging to a single genus, streptothrix, which he places with the hjrphomy- cetes, or mold fungi, because of their growth in branching threads from spore-like bodies. He says that streptothrix and actinomyces are absolutely synonymous tenns, and that the majority of pathologists consider them so. Wright and others do not agree with this view (see below) . More minute work, both clinical and experimental, should be done on this group of infections before a classification can be accepted. Leptothrix Infections. — Leptothrix forms are frequently found in the human mouth {Leptothrix buccalis), and a few writers have claimed that under certain conditions these may become pathogenic, but since no corroborative work has been done, and very little is known about the 'group, no opinion can be formed of the worth of these observations. Cladothrix Infections. — ^The organisms found in the comparatively few cases which have been considered by their observers to be due to cladothrix have not been minutely enough studied to decide definitely as to their true or false branching, the characteristic chosen to separate them from the nocardia; hence it is difficult to separate the two groups, but an attempt should be made, since the difference said to exist between them is an important one from a morphological stand-point. Clinic- ally, however, according to the reports, the cases cited are very similar to those said to be due to nocardia (streptothrix) and to actinomyces. Gasten found in a case of clinically typical actinomycosis, in which abscess cavities were found along the spinal column, not the usu&l actinomyces in the yellow, granular pus, but a fine mass of filaments. Cultures grew on all the ordinary media, best at incubator tempera- ture, but also at lower temperature on gelatin. The gelatin-stick culture, which was especially characteristic, formed on the surface THE ACTINOMYCES , 467 a whitish button; deUcate threads stretched out in all directions from the point of inoculation. On agar and potato rumpled, folded films formed on the surface, with white deposit which contained spores. Animal inoculation gave positive results only in a few cases of intra- peritoneal injection of rabbits and guinea-pigs. Purulent nodules were found in the peritoneum. Gasten called the organism Cladothrix liquefaciens. Eppinger found on postmortem examination of a case of chronic cerebral abscess, which was the result of purulent meningitis, in the pus and abscess walls, etc., a delicate fungoid growth which he suc- ceeded in cultivating on various media. On sugar agar it formed yellow, rumpled colonies which finally developed into a skin. On potato it grew rapidly, but the colonies remained small, at first a white granular deposit, which afterward turned red, and on the twentieth day resembled a crystallized almond. It did not grow well on gelatin. In bouillon it formed on the surface a small white granule, which became deeper in the centre as it grew and sank to the bottom as a white deposit. The bouillon remained clear. Migroscopically the fungus consisted of fine threads without branches which exhibited distinct motility. No flagella were observed. It was judged to be a cladothrix, to which the name "asteroides" was given by the author. It proved to be quite pathogenic for rabbits and guinea-pigs, and produced an infection called pseudotuberculosis. Mice were not affected by inoculation. THE ACTINOMYCES. The little clumps produced by this group of parasites were first seen by von Langenbeck in 1845, and the organisms were later discovered by Bollinger (1877) in the ox. They were given the name of actinomyces, or ray fungus, by the botanist Harz. They were reported in human beings by Israel in 1878. The characteristics of the microorganisms, described first by Bostrom (1890) and then by Wolf and Israel (1891), differed greatly and -have led to confusion. Bostrom's organism grew best aerobically and developed well at room temperature. He noted the intimate rela- tion of the organism to those on fragments of grain, and this led to the finding of similar microorganisms in the outer world on grains, grasses, etc. There is no doubt that some suppurative processes have been due to organisms having these characteristics, but they do not seem to excite true actinomycosis. Wolf and Israel described a microorganism from two human cases, which differs from that described by Bostrom, but agrees with the microorganisms obtained by most of the more recent investigators from typical actinomycosis. It grew best under anaerobic conditions and did not grow at room temperature. Its growth was much less luxuriant than Bostrom's microorganism. On the surface of anaerobic agar slant cultures on the third, fourth, and fifth day numerous minute 468 MICROORGANISMS BELONGING TO HIGHER BACTERIA isolated dew-drop-like colonies appeared, the largest pin-head in size. These gradually became larger and formed balUike, urregularly rounded, elevated nodules varying in size up to that of a millet seed, exceptionally attaining the size of a lentil or larger. As a rule the colonies did not become confluent, and an apparently homogeneous layer of growth was seen to be made up of separate nodules if examined with a lens. In some instances the colonies presented a prominent centre with a lobulated margin and appeared as rosettes. A characteristic of the colonies was that they sent into the agar root-like projections. In aerobic agar slant cultures no growth or a slow and very feeble growth was obtained. In stab cultures the growth was sometimes limited to the lower portion of the line of inoculation or was more vigorous there. In bouillon, after three to five days, growth appeared as small white flakes, partly floating and partly collected at the bottom of the Fig. 163. — Smear from bouillon culture of actinomyces. X 1500, diameters. (From Wright.) tube. Growth occurred in bouillon under aerobic conditions, but was better under anaerobic conditions. The organisms grew in branching and interlacing filaments, which later tend to break into segments (see Fig. 162). The microorganism in smear preparations from agar cultures appeared chiefly as short homogeneous, usually straight, but also comma-like or bowed rods, whose length and breadth varied. In many cultures short, plump rods predominated, and in others longer, thicker, or thinner individuals were more numerous. The ends of the rods often showed oval or ball-like swellings. Swollen clubs were formed irregularly in the presence of blood or serous fluids. On intraperitoneal inoculation guinea-pigs and rabbits after four to seventeen weeks showed, after being killed, tumor growths mostly in the peritoneal cavity and in one instance in the spleen. Microscopic examination of the tumors showed in all cases but one the presence of typical actinomyces colonies, in most cases with typical "clubs." The The ACTINOMYCES 469 general histological appearance of the tumors was like that of actino- mycotic tissue. Wolf in a later paper reports that an animal inoculated in the peri- toneal cavity with a culture of the same organism had lived a year and a half. At the autopsy several tumors were found in the peritoneal cavity, and in the liver a large typical tumor in which were many colonies which by microscopic examination were shown to be typical club-bearing actinomyces colonies. Wright, in 1905, made an extensive study of actinomycosis and added greatly to our knowledge of it. Naked-eye Appearance of Colonies of Parasite in Tissues. — ^In both man and animals they can be readily seen in the pus from the affected regions as small, white, yellowish or greenish granules of pin- head size (from 0.5 to 2 mm. in diameter). When pus has not formed they lie embedded in the granulation tissue. Fig. 164. — A typical "club' -bearing colony of actinomyces. (From Wright.) X 325 diameters. Microscopic Appearance. — ^Microscopically these bodies are seen to be made up of threads which radiate from a centre and present bulbous, club-like terminations (Fig. 164). These club-like termi- nations are characteristic of the actinomyces. They are generally arranged in pairs, closely crowded together, and are very glistening in appearance. They are more common in bovine than in human lesions. They have been thought to be reproductive elements, but they are probably simply a reaction of the filament end to the host tissue. The threads which compose the central mass of the granules are from 0.3m to 0.5m in diameter. The threads show true branching and in the older colonies show a segmentation which gives them the appearance of chains of cocci. Sometimes the whole centre of the colonies seems to be a mass of coccus-like bodies most of which are considered spores or conidia; the clubs are from 6m to 8m in diameter. The threads and spores are stained with the ordinary aniline colors, also by Gram's solution; m hen stained with gentian violet and by Gram's 470 MICROdRGANISMS BELONGING TO HIGHER BACTERIA method the threads appear more distinct than when stained with methylene blue. The clubs usually lose their stain by Gram's method and take the contrast strain. Isolation of Actinomyces. — Certain strains of actinomyces grow aero- bically and others anaerobically. The anaerobic strains are grown with difSculty. A large number of solidified blood serum or serum agar tubes are inoculated with the hope that one or two will develop a growth. The cultures appear much like those of tubercle bacilli. They grow, however, into the medium, and take on a yellowish hue. Wright recommends that granules, preferably obtained from closed lesions, are first thoroughly washed in sterile water or bouillon and then crushed between two sterile glass slides. (In bovine cases make sure the granule has filamentous masses, for, if not, no culture will grow.) The crushed granule is transferred to a tube ofmelted Iper cent, glucose agar at 40° C. The material is thoroughly distributed by shaking and the tube placed in the incubator. A number of granules after washing should be placed on the inside of a sterile test-tube and allowed to dry. In this way, should the material be contaminated, the drying of the granules for several weeks may kill off the other organisms. The tube should be examined daily. If a number of living filaments were added to the agar a large number of colonies will develop. These will be most numerous in a zone five to twelve millimeters below the surface (microaerophiles). The cultures are quite resistant to outside influences; dried, they may be kept for a year or more; they are killed by an exposure of five minutes to a temperature of 75° C. Experimental Inoculation in Animals. — True progressive infection is rarely or never obtained by the injection of pure cultm-es into rabbits, guinea-pigs, or other small animals. In cattle, however, the disease has been produced from cultures. The cultiu-es form the characteristic " club"-bearing colonies in the tissues of experimental animals. These, colonies are either enclosed in small nodules of connective tissue or are contained in suppurative foci within nodular tumors made up of connec- tive tissues in varying stages of development. Wright does not accept the prevalent belief, based on the work of Bostrom, Gasperini, and others, that the specific infectious agent of actinomycosis is to be found among certain branching microorganisms, widely disseminated in the outer world. He thinks that these forms belong to a separate genus, Nocardia, and that those cases of undoubted infection by them should be called nocardiosis and not actinomycosis. The term actinomycosis should be used only for those inflammatory processes the lesions of which contain the characteristic granules or nodules. That Nocardia ever forms these characteristic structures in lesions produced by them has not been convincingly shown. Occurrence. — Actinomycosis is quite prevalent among cattle, in which it occurs endemically; it is more rare among swine and horses. Many cases have in recent years been reported in man. The disease is rarely communicated from one animal to another and no case is known where a direct history of human contagion has been obtained. The cereal grains, which from their nature are capable of penetrating NOCARDIA INFECTIONS 471 the tissues, have been found in centres of actinomycotic infection in the lower animals. The microorganism may also be introduced by means of carious teeth. Cutaneous infection has been produced by wood splinters, and infection of the lungs by aspiration of fragments of teeth containing the fungus. The presence of the microorganism in cereal grains, which was formerly accepted, is denied by Wright and therefore certainly placed in doubt. The further distribution of the fungus after it is introduced into the tissues is effected partly by its growth and partly by conveyance by means of the lymphatics and leukocytes. Not infrequently a mixed infection with the pyogenic cocci occurs in actinomycosis. Characteristics of Disease in Man and Animals. — In the earliest stages of its growth the parasite gives rise to a small granulation tumor, not unhke that produced by the tubercle bacillus, which contains, in addition to small round cells, epithelial elements and giant cells. After it reaches a certain size there is great proliferation of the svirrounding connective tissue, and the growth may, particularly in the jaw, look like, and was long mistaken for, osteosarcoma. Finally, suppuration occurs, which,' according to Israel, may be produced directly by the fungus itself. The course of the disease is very chronic. Usually the first sign is a point of infiltration about the lower jaw or lower on the neck. This almost painless swelling increases and finally softens in its centre. The necrotic tissue finally forces a passage externally or, passing downward, infects the pleura, lungs, mediastinum, or ribs. As a rule the disease is not accompanied by fever. In cattle the disease is usually situated in some portion of the head, especially in the jaw, tongue, or tonsils, hence called lumpy jaw, wooden tongue, etc. Primary lung, intestinal, and skin lesions are not infrequent. These local lesions sometimes scatter and produce a general infection and the udder may be involved. Treatment. — ^In 1892 Nocard showed that cases in animals might be cured by iodide of potassium, calling attention to the fact that Thomassen had recommended this treatment in 1885. It is given in doses of 1| to 2| drams once a day. Salmon and Smith (U. S. Bureau of Animal Industry, Circular No. 96) give directions as to its use. Mycetoma (Madura Foot). — This is a purulent inflammation of the foot occurring primarily in warm climates. The inflammation is accompanied by much irregular enlargement of the foot. Three varieties of this condition have been described based upon the color of the granules found in the diseased area: (1) white, (2) black, and (3) red. The white variety has been studied by Musgrave and Clegg (1907) , who have isolated an organism resembling somewhat actinomyces and somewhat the organism isolated by Wright (1898) from a black variety of the disease which is probably a true mold. (See "Trench Foot.") NOCARDIA (STREPTOTHRK) INFECTIONS. The most familiar name of this group of microorganisms is strepto- thrix, but this name had already been used for another genus; there- 472 MICROORGANISMS BELONGING TO HIGHER BACTERIA fore, according to the rules of nomenclature, nocardia, which name was proposed by Trevisan in 1889 for the organism discovered by Nocard in fardn des boeufs, should be employed. Wright calls attention to the misuse of the term streptothrix, and gives the reasons for the employment of the term nocardia in its place. From widely scattered localities and at long intervals of time reports have been published describing unique cases of disease produced by varieties of microorganisms belonging to the genus nocardia. In some of these cases, points of similarity can be recognized in the clinical symptoms and the gross pathological lesions, while others differ widely in both respects. Nocardia have been found in brain abscess, cerebro- spinal meningitis, pneumonic areas, and in other pathological conditions. Eppinger injected cultures into guinea-pigs and rabbits, and observed that they caused lesions similar to tuberculosis. ConsoHdation of por- tions of both lungs, thickening of the peritoneum, and scattered nodules resembling tubercles were noted by Flexner in a case of human infection due to nocardia in which the pathological picture of the disease re- sembled so nearly that of tuberculosis in human beings that the two diseases could be separated only by finding the causative microorganism in each case. But in no two cases reported up to the present time have the descriptions of the microorganisms found agreed in all par- ticulars. In some cases no attempt at cultivation was made. In other cases numerous and careful plants on various culture media failed to develop the specific organism. As late as the year 1904 Tuttle was able to find the reports of only twelve cases in which nocardia was found in suflScient abundance to have been an important, if not the principal, factor in producing disease. These cases were all fatal, and only once was the character of the disease recognized during life. As the clinical symptoms and the lesions in the human subject as well as in the animals experimentally inoculated with nocardia often resemble those of miliary tuberculosis, the question is naturally suggested whether cases of nocardia tuberculosis are not more numerous than the few reported cases would indicate. The almost universal prevalence of genuine tuberculosis and the extreme gravity of the disease have so long occupied the attention and study of the medical profession that much is taken for granted, and in cases in which the symptoms and lesions resemble with some closeness those characteristic of the well- known disease they may easily be set down without question to the account of the tubercle bacillus. The cases of nocardiosis reported which simulated tuberculosis have been fatal, and the lesions for the most part have been widely distributed, but in a number of cases old lesions have been found which suggest that the disease may have been locahzed for a longer or shorter time, and then, by some accident, may have become rapidly general. In this respect also these cases may resemble tuberculosis. Whether all cases of nocardiosis in the human subject are general and fatal or, as in tuberculosis and actinomycosis, whether there may be cases of localized disease which recover, are questions which have not yet been decided. The methods employed NOCARDIA INFECTIONS 473 to demonstrate the presence of tubercle bacilli render nocardia more or less invisible. Again, unless the observer keeps in mind the possi- bility of nocardia infection, he may not appreciate the importance of finding slender threads with or without branches, and may consider them accidental baciUi, or varieties of leptothrix or non-pathogenic fungi. As the lungs have appeared to be the seat of the primary infection in most of the cases of human nocardiosis it is very desirable that all cases presenting the physical signs of tuberculosis, in which repeated examinations fail to discover the tubercle bacillus, should be system- atically examined for threads. In this way alone can the frequency of the disease be determined. Gram's method of staining is one of the most reliable agents .for demonstrating these organisms. Varieties of nocardia are widely distributed and are not very infrequently met with, but as yet, with the exceptions mentioned above, very little is known about them. Tuttle's report of a case of general nocardia infection at the Pres- byterian Hospital gives such a good clinical, bacteriological, and patho- logical picture of an acute case of this infection that a considerable portion of it is repeated here : Six days before her admission to the hospital her illness began with a severe chill and fever and pain in her left side and back. The following day the pain in the side was worse and breathing was difficult. She began to cough and had some expectoration, but no blood was noticed in the sputa. At irregular inter- vals she had alternating hot and chilly sensations. On admission the patient complained of pain in the left side of the chest, cough, fever, weakness, and prostration. Her temperature was 103° and her pulse and respirations were rapid. The history of the disease and the physical signs indicated an attack of acute lobar pneumonia, the area of consolidation being small and situated in the lower part of the left upper lobe in front. Frequent and violent coughing, with almost no expectoration, pain in the affected side and in the lumbar region, restlessness and sleeplessness, and involuntary urination were the symptoms noted during the first four days in the hospital. The pneumonic area increased somewhat and extended backward to the posterior axillary line, and the temperature was continuous at 103° to 103.5°. On the fifth day the temperature fell 2° and signs of resolution appeared in the consolidated area. The apparent improve- ment, however, was of short duration. On the sixth day the temperature rose to 104.5°, and continued to rise each day, reaching 107.5° shortly before death, which occurred on the ninth day in the hospital and the fifteenth day of the disease. There were repeated attacks of profuse sweating. On the day before her death three indurated swellings beneath the skin were noticed. One, on the left forearm, about the size of a walnut, apparently contained pus. Two, of smaller size, were situated in the right groin. Blood cultures from a vein in the arm, taken on the sixth day, remained sterile. The leukocyte count on the seventh day was 36,000. Autopsy.-^— On the right arm, the left forearm, the abdominal wall, and on both thighs there are eight to ten slightly projecting, rounded, fluctuating, subcutaneous swellings from 5 to 1 inch in diameter. The skin over most of these nodules is unaltered, but over the larger ones there is a shght bluish dis- coloration. The nodules are composed of bluish gray, thick, mucilaginous matter, which is very tenacious and can be drawn out into long threads. The lower lobe is thickly studded with miliary tubercles, and scattered through the 474 MICROORGANISMS MLOnGING fO HtCfHER BACTERIA entii-e lung are suppurating foci. Liver and spleen normal. Kidneys: The surface is evenly dotted with minute white spots, which suggest septic emboli rather than tubercles. A few prominent white nodules from i to 5 inch in diameter, contain thick, tenacious matter (Fig. 166). Section shows that the en- tire substance of the kidney is densely studded with these minute white granules. The gross pathological conditions were interpreted before nocardia was found as follows: An old tuberculous nodule in the right lung; acute miliary tuberculosis in the right lung and peritoneum; acute lobar pneumonia, affecting the left lung; septic infarctions and pyemic abscesses of both lungs, heart muscle, both kidneys, pancreas, mesenteric lymph nodes, and subcutaneous connective tissue. The miliary tubercles of the right lung and peritoneum presented the characteristic appearance of genuine tuberculosis. They were minute, hard, gray, almost translucent nodules, while the granules in the kidneys were of an opaque white or yeUowish-white color. . _ Fig. 165. — Portion of kidney showing minute and large areas of infection. Microscopic Examination.— Smears from the abscess beneath the skin and on the surface of the kidneys were stained with methyl blue, carbol-fuchsin, and by Gram's method. The smears resemble those made of tenacious sputum. There is a large amount of mucoid material containing a considerable number of leukocytes. Occasionally irregu- larly curved, thread-shaped microorganisms are found. They vary considerably in length and thickness, and broken and apparently degenerating fragments are seen. The more slender threads are evenly stained, but some fragmentation or beading of the protoplasm can generally be observed. The thicker threads and broken fragments show deeply stained globules and irregular bodies in a faintly visible rod or thread-shaped covering. Some branching threads are observed, but more commonly they are not branching. No other microorganisms are found m the smears. Sections from the lower lobe of the right lung, stained with hematoxylin and eosin, show in certain places the identical microscopic appearances which are considered characteristic of tuberculosis. Stained by Gram's method, with care not to decolorize too completely, threads like those described in the abscesses are found NOCARDIA INFECTIONS 475 in great abundance, but rather faintly stained. No threads can be Found within the typical tubercles with giant cells, but in the zones of small cells around them they are seen in great numbers, winding about among the cells and forming a sort of network. In the minute foci of small cells one or two fragments of threads are generally seen, and a moderate number in the small abscesses. In the areas of more diffuse infiltration these threads are abundant. No other microorgan- isms can be found except in the pneumonic area of the left lung, where some groups of cocci are seen. Fig. 166. — Streptothrix from bouillon Fig. 167. — Young streptothrix threads culture. (From Tuttle.) showing terminal buds. (From Tuttle.) Culture Experiments. — Six tubes of Loffler blood serum were inocu- lated from the kidneys and kept at 37° C. On the third day minute white colonies appeared in some of the tubes, and on the fifth day all the tubes showed from three to ten or twelve similar colonies in each. The colonies increased in size until some of them reached a diameter of one-eighth of an inch. The color, at first white, changed to yellowish- white and then to a decided pale yellow. The well-developed colonies cling firmly to the surface of the medium and were not easily detached or broken up. The growths in all of the tubes were absolutely pure, and consisted of branching threads like those found in the sections. Loffler's blood serum seems to be the most suitable medium for cultures. The growth on this medium is more^ rapid and abundant than on any of the other media tried. On plain agar and glycerin agar the growth is the same as on blood serum, but is less rapidly developed. In bouillon the growth is slow. If the tube is not disturbed or jarred, minute white tufts are seen clinging to the surface of the glass. But if the tube is shaken even slightly they sink slowly to the bottom, forming a white, fluffy layer. These growths when undisturbed resemble minute balls of thistle-down. The yellow color is not apparent even in the mass at the bottom of the tube. It is strictly aerobic. Morphology (Figs. 166 and 167). — On blood serum the threads are comparatively thick and coarse, but those growing in bouillon are very slender and delicate. The main trunk also is often thicker than 476 MICROdRGANISMS BELONGING TO HIGHER SACfERlA the branches. When unstained they are homogeneous gray threads, without any appearance of a central canal or double-contoured wall. There is never any segmentation of the threads. When properly stained there is always a distinct beading or fragmentation of the protoplasm, but overstaining with fuchsin produces rather coarse, evenly stained rods. The branching is irregular and without symmetry, and the branches are placed at a wide angle, very nearly, and some- times quite,- at right angles. This is best seen in specimens taken from liquid media. The irregularly stellate arrangement of the branches, which was observed by Eppinger in his original specimen, is often seen in young organisms floated out from a liquid medium. Spore Formation.— On examining the deep orange or red-colored growth upon potato, one is siu-prised to find that the threads have entirely disappeared and that the specimen -consists of moderately large coccoid forms. These represent the spore form of the organism, and when planted upon blood serum the branching threads again appear. The spores stain readily with carbol-fuchsin and are not easily decolorized. They are spherical, or nearly so, but often appear somewhat elongated, apparently from beginning germination. They are killed by exposure to moist heat of 65° to 70° C. for an hour, but are more resistant to dry heat. Drying destroys the threads after a comparatively short time, but the spores retain their vitality for an indefinite period. A dried-up potato culture retains its vitality at the end of almost four years. The identity of Tuttle's microorganism is not fully established. It is undoubtedly a nocardia, but it does not agree in all particulars with any of the varieties described. Animal Inoculations. — ^A number of rabbits and guinea-pigs were inoculated subcutaneously upon the abdomen and in the neighborhood of the cervical, axillary, and inguinal lymph nodes with colonies broken up in salt solution. Indurated swellings were produced at the point of inoculation and a number of abscesses resulted. The abscesses developed rapidly and some of them opened spontaneously, while others were incised. The material evacuated did not resemble ordinary pus, but was thick and mucilaginous and exceedingly tenacious, like that from the subcutaneous abscesses of the patient described above. The microscopic appearance was the same, and the nocardia threads were found in considerable numbers. Several rabbits and guinea-pigs and two cats received peritoneal inoculations, but none of them showed any sign of infection. When rabbits were inoculated intravenously, a rapidly fatal general infection was produced, and the lesions were similar in kind and distribution to those described in the human subject. Other Cases Reported. — Ferre and Faguet found in Bordeaux, in a cerebral abscess in the centrum ovale, a branching fungus, colored by Gram, which corresponded to nocardia. It grew on agar in round, ochre-colored colonies; on potato there was little growth visible; slimy, tough colonies, which became gray and remained free from white dusting on the surface. Inoculations in rabbits and guinea-pigs were negative. NOCARDIA INFECTIONS 477 Varieties of nocardia have been found in the human vagina. We have found a variety of nocardia in several cases of stillbirth with invasion of the placenta with the same organism. Numerous cases have been observed in which nocardia proved to be the cause of chronic lung diseases, clinically suspected to be tuberculosis. Treatment. — Recently autogenous vaccines have been tried in certain cases, but it is yet too soon to determine with what result. REFERENCES. Bostrom: Beitr. path. Anat., etc., 1890, vol. ix. Foulerton: The Strepotrichoses and Tuberculoses, Lancet, 1910, clxxvili, 551, 626, and 769. MusGRAVE and Cleqg: Phila. Jour. Sc, 1907, iii, 2, 477. Nocard: Ann. de rinst. Past., 1888. TnTTLE: Med. and Surg. Rep., Presbyterian Hosp., New York, 1904, vi, 147. Wright: Jour. Exp. Med., 1898, Iii, 421, and Jour. Med. Res., 1905, viii, 349, and Osier's Modern Medicine, 1907, i, 327. CHAPTER XXXVI. FILTRABLE VIRUSES. DISEASES OF UNKNOWN ETIOLOGY. FILTRABLE VIRUSES. There exists a class of infectious diseases from which it has been quite impossible up to the present time to demonstrate visibly any individual microorganisms, although infective material from such diseases may, with certain precautions, be passed through stone filters of varying degrees of porosity (for types of filters, see p. 93) ; the fil- trates will contain the virus and be capable of reproducing the disease with all its characteristics when inoculated into a susceptible animal. Examined microscopically, even with the highest powers, the filtrate is limpid, and, except in a few diseases which will be described in detajl later on, not a sign of characteristic particulate matter can be seen. Such a filtrate therefore contains either ultramicroscopic organisms or small organisrus with refraction and staining powers so faint that they cannot be demonstrated by our present methods. Certain precautions must be observed in such filtrations. In the first place the filter must be shown by actual test to be free from infec- tion. Any and all of the test organisms must be absolutely retained and none pass into the filtrate. A few minute species of germs have been found to pass certain grades of filters, especially under pressure. Thus the bacillus of guinea-pig pneumonia, which is O.Sai x O.lix, passes Berkefeld V (Wherry) ; a spirillum isolated^ by von Esmarch passes, according to him, the Berkefelds, Chamberland F, and other filters; a minute water flagellate was found by Borrel to pass through the coarser filters. Recently, Wolbach and Binger and others have shown that several spirochetes, pathogens as well as saprophytes, pass through Berkefelds V, N, and W, by pressure and by suction. It is seen, therefore, that certain known organisms within the limits of visibility may pass even some of the finer filters. The filtration must be completed within a moderate time, because even bacteria as large as the typhoid bacillus may, in media containing a certain amount of albuminous material, grow, in time, through the filter. The material to be filtered should, furthermore, be greatly diluted and first filtered through filter paper or other coarse filter in order to avoid the clogging action of extraneous material. If after all the proper precautions have been taken the filtrate is pathogenic, it must be shown that the symptoms are due to a micro- organism and not to a toxin. This may be decided by inoculating a series of animals successively with the filtrate obtained from a pre- viously so inoculated animal. DISEASES PRODUCED BY FILTRABLE AGENTS 479 From our present knowledge, filtrable agents may be divided into two groups: (1) those which have not yet been morphologically demon- strated (ultramicroscopic?); (2) those which are shown to be within the limits of visibility. A third groiip may be made of the diseases produced by viruses of questionable filtrability. GROUP I— DISEASES PRODUCED BY FILTRABLE AGENTS OF UNKNOWN MORPHOLOGY. Foot-and-mouth Disease.— This is probably the first disease shown to be produced by a filtrable agent. It is a highly infectious disease of cattle. Other domestic animals may also be attacked. It occurs seldom in man and then among people handling infected cattle or drinking the milk of infected animals. The disease in cattle is characterized by the appearance of vesicles in the mouth and around the coronet of the foot as well as between the toes. Lofiler and Frosch, in 1898, discovered that after diluting the contents of an unbroken vesicle with twenty to forty times its volume of water and passing the resulting fluid through a Berkefeld filter, the filtrate contains the virus. This virus remains infectious for some time. One attack of the disease usually produces a certain amount of immunity. " LofHer claimed to have produced a serum which has immunizing properties but his work has not been corroborated. Mosaic Disease of Tobacco.— This is the second disease in which a filtrable virus was demonstrated. The disease causes the young tobacco leaves to become devoid of chlorophyll in spots which enlarge, turn brown, and the imderlying tissue becomes necrotic. Beijerinck, in 1899, showed that the filtrate from a porcelain filter promptly repro- duced the disease on tobacco leaves. Cattle Plague (Rinderpest). — ^This fatal European and African disease of cattle is characterized by inflammation of the intestinal mucous membrane. The blood is infectious and filtrates of it through Berkefelds and Chamberland F (Nicolle and Adil-Bey, 1902) produce the disease. No organism can be seen. Immunity follows one attack. It can also be produced by a subcutaneous inoculation of bile from an infected animal. Yellow Fever. — ^The undiluted serum from cases of this disease has been shown by the American commission (1901) and others (see p. 578) to pass the Berkefelds and the Chamberland B filters as clear filtrates, and in this form to be infectious; therefore some forms at least of the specific organism are probably ultramicroscopic. Rabies. — No bacteria have been discovered that are considered as factors. The probability of the Negri bodies being protozoa and the exciting factor is considered in Chapter XLV. The virus of rabies has been shown to be filtrable (Remlinger, 1903). Poor and Steinhardt (1913) showed that gland virus is more filtrable than that from the brain. It may pass through the coarser Chamberland filters. Hog Cholera. — de Schweinitz, Dorset, Bolton, and McBryde demon- strated in 1905 that the blood of hogs suffering from hog cholera con- 480 FILTBABLE VIRUSES tains a filtrable virus which is capable of producing the disease on inoculation into healthy hogs. This virus passes Chamberland B and F filters. It leaves the body in the urine and probably enters another animal through contaminated food. King, Baeslack, and Hoffman found in 1913 a short motile spirochete {Spirochceta suis) in the blood in a series of cases of hog cholera. They think this organism may be a stage of the specific organism which may produce filtrable granules. This disease is a very fatal and contagious disease of swine char- acterized by fever and ulcerative enteritis. Immunity follows one attack. An immune serum is produced. Animals are also immunized by a mixture of infected blood and immune serum (sensitized virus). The hog cholera bacillus was earlier supposed to be the cause of this disease. South African Horse Sickness. — This is a warm-weather disease, more common in animals that do not pass the night under cover. The horses are uneasy, have difficulty in breathing, and a reddish froth exudes from their mouths. The temperature rises in the daytime, but has a tendency to drop at night. In severe cases an edematous swelling of the head and neck may appear. MacFadyen succeeded in passing blood serum of a diseased horse (diluted) through the Berkefelds and a Chamberland F, not through a Chamberland B filter. Chicken Sarcoma. — In 1910 Rous discovered a spindle-celled tumor in chickens, which he has reproduced in other chickens by transplanta- tion, by inoculation of dried tumor tissue, and by the inoculation of Berkefeld filtrates from emulsions of tumor cells. Novy's Rat Virus. — Novy has found an extremely virulent filtrable virus from a disease of rats, which passes all filters. He has not yet published full reports. GROUP II.— DISEASES PRODUCED BY FILTRABLE AGENTS SHOWN TO BE VISIBLE. Contagious Pleuropneumonia of Cattle.— This malady affects cattle, but not other species. Typically, there is an inflammation of the lungs and the pleura which is invasive and causes necrosis of the diseased parts. Nocard and Roux succeeded in cultivating the organism in collodion sacs placed in the peritoneal cavity of rabbits, using a mixture of serum and bouillon. After two weeks a very faint turbidity appeared in the sacs; coincidently the fluid became infective. Nocard and Roux described the organisms as minute spheres and spirals just within the limits of visibility. They showed (1899) that the organisms passed the Berkefelds and a Chamberland F, but not a Chamberland B filter. Immunity is produced by a single attack. It has also been produced artificially by the inoculation of cultures or virulent exudates. Epidemic Poliomyelitis.— This is a disease which affects chiefly the central nervous system. It occurs usually in children, and appears sporadically and in epidemics in many countries. In New York, in 1907, there was an epidemic of over 2000 cases and in Texas, in 1912, DISEASES PRODUCED BY FILTRABLE AGENTS 481 there was a large outbreak. During the summer of 1916 one of the largest epidemics known occurred in New York City and the surrounding country. In nature, the virus enters probably through the upper air passages. The chief symptoms of the disease are fever, with or without sore throat, hypersensitiveness followed by paresis and paralysis. The mortality is low. There is usually permanent injury to parts of the motor areas of the nervous system, with resulting deformity. The principal microscopic changes are a marked exudation of polynuclear leukocytes into the lymph spaces and the cerebrospinal fluid. The changes are usually specially marked in the anterior commissure and the cornea of the cervical and lumbar regions of the cord, but the whole nervous system is more or less affected. The bloodvessels may become thrombosed and ruptured. The disease seems to be closely related to, if not identical, with certain diseases manifested by acute encephalitis and bulbar paraly- sis, e. g., Landry's paralysis. That the disease also bears a marked resemblance to rabies is quite apparent. The finding of the virus in the salivary glands and the demonstration of its filtrability give added evidence of its similarity to rabies. Until recently nothing was definitely known of the etiology. Now we have a series of studies which seem to have practically cleared up the subject. Landsteiner and Popper (1909) reported the transmission of acute poliomyelitis to apes. They inoculated the spinal cord intraperi- toneally and produced typical symptoms and lesions, but did not succeed in translnitting from ape to ape, probably because they used a mild virus. Flexner (1909) and Lewis transmitted the disease from monkey to monkey by means of intracerebral inoculations. Landsteiner and Levaditi (1909) also transmitted it from monkey to monkey. They found that the virus lives some time outside of the body; that the degenerated nerve cells are taken up by phagocytes, and there is an analogy between the lesions of poliomyelitis and those produced by rabies. They demonstrated that the virus is filtrable. Leineri and Wiesner transmitted the disease from monkey to monkey, found young animals more susceptible than older ones, and spinal fluids, blood, and spleen negative. Flexner transmitted the disease by inoculating into the blood or peritoneal cavity, also by the subcutaneous method, and independently found the virus to be filtrable. Landsteiner and Leva- diti found the virus in the salivary glands and suggested the saliva, moist or dry, as a source of contagion. Soon after this (1911) Noguchi and Flexner announced that they had obtained cultures in media similar to that used by Noguchi in cultivating spirochetes (see p. 509). In such media in about five days the pieces of tissue employed become surrounded by an opalescent haze which increases for about five days more, and a sediment gradually forms. Giemsa's stain shows the presence of minute globoid bodies (0.15 to 0.3 diam.) in pairs, short chains, and masses. Cultures were also obtained from the filtered virus. Monkeys inoculated with certain strains of this organism grown for a variable number of culture genera- tions may die with typical lesions of the disease. The authors consider 3X 482 FILTRABLE VIRUSES these bodies the cause of the disease. They further report that the cultures are filtrable. Levaditi states that he cannot obtain the results of Noguchi, but that he obtains evidence of growth by the living tissue method. Rosenow as a result of studies during the recent epidemic states that he has found an aerobic streptococcus which shows in early culture generations specific pathogenicity for the central nervous system, producing paralysis in animals. When grown under anaerobic con- ditions he says it shows only minute forms. He therefore concludes that it is the cause of the disease and that it is probably the same organism as that described by Flexner and Noguchi. During this epidemic we also have been studying the etiology of this disease, but our studies are not yet far enough advanced to draw conclusions. We have isolated the Flexner-Noguchi organism in 3 cases out of 50, and definite streptococci in more than half the cases. We can say that the Flexner-Noguchi organism is distinctly different from any aerobic streptococcus studied but we have not yet had enough monkeys to test its ability to produce poliomyelitis in them. Eabbits and possibly guinea-pigs have also been found to be somewhat susceptible to the virus (Krause and Meinicke, Marks, Rosenau) . These authors state that the disease in rabbits does not resemble that in man. Immunity. — One attack seems to give certain immunity. Flexner and Amoss state that, experimentally, the virus inoculated into the blood is capable of being neutralized by intraspinous injections of immune serum. Relapsing Fever. — Todd and Wolbaeh have found (1914) that SpirochcBta duttoni may be forced through even the finer grades of Berkefeld filters by a pressure of from fifty to ninety pounds to the square inch. For filtrability of other spirochetes see page 515. GROUP m.— DISEASES PRODUCED BY VIRUSES OF QUESTIONABLE FILTRABILITY. Smallpox and Related Diseases. — The details of the disease smallpox are considered in a later chapter. In 1908 Casagrandi reported that the virus was filtrable under pressure through coarser Berkefeld filters. Vaccine virus had already been reported as filtrable by some investi- gators, and as non-filtrable by others. Our results agree with those of the latter observers. The opinion as to the filtrability of the other "pox" diseases is not unanimous. Dengue. — Ashburn and Craig claim to have reproduced dengue in susceptible individuals by a procedure similar to that employed in yellow fever. The virus passes a Berkefeld filter. The intermediary host in natural infection is claimed by them to be Cv,lexfatigans. Measles. — This very definite infectious exanthematous disease still remains among those of unknown etiology. Both cell inclusions (Field, Ewing, and others), bacilli and cocci have been reported as having an etiological relationship. The reports have not been corroborated. In 1905 Hektoen produced measles in two hiunan cases by the inoculation OTHER DISEASES OF UNDETERMINED ETIOLOGY 483 of blood drawn from an infected case at an early stage of the disease. Anderson and Goldberger, in 1911, inoculated monkeys with measles blood and demonstrated the virus in the blood of the inoculated animals and in the secretion of the upper air passages. They also demonstrated the period of infectivity of the blood. They reported positive results with Berkefeld filtrates. Tunnicliff has just reported finding a small anaerobic coccus in the blood of 42 out of 50 cases of measles. Scarlet Fever. — Scarlet fever is an acute febrile, highly infectious dis- ease, characterized by a diffuse, punctate, erythematous skin eruption, accompanied by catarrhal, croupous, or gangrenous inflammation of the upper respiratory tract and by manifestations of systemic infection. The disease was probably known long before the Christian era, but the present name does not appear until the time of Sydenham (1685), who differentiated the disease from measles. It is very generally disseminated, but is much more common in temperate climates than in the tropics. The specific exciting factor is thought by some to be a streptococcus, of the Streptococcus pyogenes type, but the evidence in favor of this view is very slight (see Streptococcus pyogenes). Mallory (1914) found certain bodies occurring in a series of forms in and between the epithelial cells of the epidermis and free in the super- ficial lymph vessels and spaces of the corium. He. gave the name Cy'clasterion scarlatinale to these bodies in consequence of the frequent wheel and star shapes of the rosette. Field and others think that these bodies are not specific germs. Recently Mallory reports the finding of a small pleomorphic bacillus which he thinks may be the specific cause. The reports that a filtrable virus is found lack corroboration. Trachoma. — This condition has already been considered in a previous chapter. Bertarelli reported that he was able to produce a specific filtrate. Nicolle also reported positive results with filtrates. The* descriptions of the disease, however, are too vague to allow an opinion as to the truth of these reports. Other Diseases said to be due to Filtrable Viruses. — Several other diseases of less importance have been listed as belonging to those produced by a filtrable agent. When a well-known disease such as epidemic cerebrospinal meningitis is reported as being due to a filtrable virus, much corroboration is needed before we can accept the statement as a fact. Recently Kruse corroborated by Foster reported that a filtrable virus is obtained from common colds which would reproduce the disease. OTHER DISEASES OF UNDETERMINED ETIOLOGY. Rocky Mountain Spotted Fever.— This is an acute infectious disease characterized by fever and a more or less hemorrhagic eruption. Some years ago Wilson and Chowning thought they found a protozoan in the blood similar to babesia in Texas fever. Their findings have not been corroborated. Their investigations proved, however, that rabbits are susceptible to the disease and that a tick of the genus Dermacentor probably carries the infection. Then Ricketts and Gomez made some very interesting studies on the disease. They found that guinea-pigs 484 FILTRABLE VIRUSES and monkeys are susceptible as well as rabbits, and they further founc that in guinea-pigs and monkefys an attack of spotted fever produces a strong active inherited immunity characterized by a serum wit! high protective but low curative power, and that the production of th( serum in the horse with the use of serovaccination in man may givt practical results. They found a moderate number of diplococcoic bodies in the blood of infected guinea-pigs and monkeys, and fewer ir man. They found that the virus is transmitted by the infected female tick to her young through the eggs. If the larvae from these eggs an allowed to feed upon normal guinea-pigs, these animals come down with the disease. Immense numbers of these apparent organisms are found in afPected eggs and none were found at first in normal eggs. Afterward Ricketts found a few, but he thought these might be an avirulent species of the same organism. The salivary glands, alimentary sac and ovaries of infected female ticks are swarming with these bodies, while normal ticks seem to have none. Lastly, Ricketts found that these bodies agglutinate with specific serum, 1 to 300 dilution. Recently Wolbach has corroborated some of these findings and demonstrated a similar organism in characteristic lesions of experimental animals bitten by infected ticks. Friek recently reported the finding of small bodies in centrifuged blood cells, which he did not classify. Typhus Fever. — The occurrence of this infectious disease in epidemic form had disappeared from civilized lands until the recent war broke out. The fact that the body louse may transmit the disease helps explain why it has always been classed as a filth disease. It still occurs endemically in parts of Europe and North and South America,. In Mexico it occurs in epidemics where it is known as Tabardillo. In New York it occurs occasionally in mild form under the name of Brill's disease. Brill thought it was a new disease, but Anderson and Goldberger have shown that typhus fever and Brill's disease are the same. Nicoll and Krum- wiede observed four cases which clinically verified these findings. The disease is characterized by high temperature and a petechial rash. NicoUe (1909) showed that the old world typhus can be transmitted to the chimpanzee and from this to the macacus with typical eruptior in each case. He also showed that the disease is transmitted by th< ordinary body louse (Pediculus vestimenti) . Anderson and Goldberge: (1909) were the first to transmit typhus fever of Mexico (tabardillo] to monkeys. They were able to transmit directly from human being! to the macacus and capuchin and from monkey to monkey. Rickett; and Walker (1910) also found that the macacus was directly susceptible to the disease. They based their diagnosis chiefly upon a rathei indefinite fever and in most cases somewhat distinct symptoms o illness. They also found that the monkey may pass through an attacl of typhus so mild that it cannot be recognized clinically, but it result: in immunity. The immunity test is a reliable proof of the previou occurrence or non-occurrence of typhus at least within a period of om month. They found that typhus was transmitted to the monkey b; the bite of the louse. They further state that in stained preparation OTHER DISEASES OF UNDETERMINED ETIOLOGY 485 of blood of patients taken from the seventh to the twelfth days of the disease they invariably found a few short bacilli similar to those which belonged to the hemorrhagic septicemia group. In moist preparations they saw similar forms in all cases. No motility was observed. No cultures could be obtained. They examined the dejecta of many lice and found similar bodies in large numbers in infected lice and occasion- ally in non-infected, lice. Pediculus capitis may also transmit the disease. Recently, Plotz reported the isolation in pure cultures of a Gram- positive pleomorphic anaerobic bacillus from cases of typhus and of Brill's disease. His report is given on page 451. Attempts at filtration by Ricketts and Wilder, Anderson and Goldberger and Nicolle, Connor and Wilder showed that filtered blood inoculated into monkeys is probable non-filtrable. Olitzky has just gone over this work and added experiments of his own which seem to prove that the virus is non-filtrable. Chicken-pox. — Kling claims to have been able to vaccinate against this contagious disease with the clear contents of a fresh vesicle in as early a stage as possible. He inserted the point of a sterile lancet into such a visicle and then into the patient's skin, repeating six times. This work was corroborated by Rabinoff. Mumps. — Comparatively few studies have been made of this infec- tious disease, probably because of its low mortality. By puncturing the parotid gland, Laveran and Catrin obtained a diplococcus which stains easily, is Gram-negative, and grows on ordinary culture media. No satisfactory specific studies have been made of this organism. Hess, in 1915, used serum from recovered cases of mumps in a children's home as a protective measure, with the result that none of the 17 so inoculated came down with the disease on exposure. Pellagra.-^-This disease has been much studied recently. The theory that it is due to the ingestion of damaged corn received a check from the work of Siler, Garrison and MacNeal (1913) who stated that they were unable to get evidence, to support this theory. They consider the disease a specific infection caused by unknown means. Since then the work of Goldberger and others added much to the evidence that the disease is due to a deficiency of certain substances in the diet which may be corrected by including in the diet suitable proportions of fresh animal and leguminous protein food. Verruga peruviana, a South American disease has been shown by Strong and his co-workers to be a disease distinct from oroya fever (see p. 552) and to be produced by an unknown virus transmitted, according to Townsend, by the bite of a gnat {Phlebotomus ■verrucanamun) . REFERENCES. Anderson and Goldbekger: Public Health Report, 1910, 1912 and 1913; Jour. Am. Med. Assn., 1911, Ivii, 113. AsHBURN and Ceaig: Jour. Inf. Dis., 1907, iv, 440. Ewing: Jour. Inf. Dis., 1909, vi, 1. Field: Jour. Exp. Med., 1903, vii, 343. 486 FILTRABLE VIRUSES Flexneb and Lewis: Jour. Am. Med. Assn., 1909, liii, 2095, and 1913, xli, 1639; Jour Exp. Med., 1910, xii, 227. Flexner: Huxley Lecture, Lancet, 1912, ii, 1271. Flexnek and Noguchi: Jour. Exp. Med., 1913, xyiii, 461. Flexner and Co-workers: Jour. Exp. Med., 1913 and 1914. Foster^ Geo B.: The Etiology of Common Colds, Jour. Am Med. Assn., 1916, Ixvi 1180. Frick, L. D.: Rocky Mountain Spotted Fever, Pub. Health Rep., March 3, 1916, p, 516. Goldberger: U. S. Pub. Health Rep., 1914, October 22, 1915; Jouv. Am. Med. Assn., 1916, Ixvi, 471. Hektoen: Jour. Inf. Dis., 1905, ii, 238. Hess, A. F.: A Protective Therapy for Mumps, Am. Jour. Dis. Child., 1915, x, 99. Kino, Baeslack and Hoffman: Jour. Inf. Dis., 1913, xii, 39 and 206. Kling: Hygiea, 1913, Ixxv, 1032; Berl. klin. Woch., 1915, lii, 13. Khadsb u. Meinioke: Deut. med. Wchnschr., 1909, xxxv, 1825. Krause, W.: Die Erreger von Husten, Mlinch. med. Wchnschr., 1914, Ixi, 1547. Landsteiner and Levaditi: Compt. rend. Soc. de biol., 1909, xlvii, 592 and 787. LOFFLER and Frosoh: Centralbl. f. Bakt., 1898, xxiii, 371. Mallory and Medlar: The Etiology of Scarlet Fever, Jour. Med. Res., 1916, xxxv, 209. Marks: Jour. Exp. Med., 1911, xiv, 116. MoHLER, J. B.: Foot and Mouth Disease, Farmers' Bull. 666, U. S. Dept. Agric, 1915. NiGOLL, Krumwiede, Pratt and Bullowa: Jour. Am. Med. Assn., 1912, Ux, 521. Nicolle: Compt. rend. Acad. Sc, 1909 and 1911; Ann. del'Inst. Past., 1910, 1911, and 1912. Olitsky: Jour., Inf. Dis., 1917, xx, 349. Plotz: Jour. Am. Med. Assn., 1914, Ixii, 1556. Poor and Steinhardt: Jour. Inf. Dis., 1913, xii, 202. Rabinofp: Arch, of Fed., 1915, xxxii, 651. Read, Carrol, Agramontb and Lazear: Philadelphia Med. Jour., 1900. Remlinger: Ann. de I'Inst. Pasteuir, 1903, xvii, 834. RiCKETTS and Wilder: Jour. Am. Med. Assn., 1910, Ix, 309. Rous and Morphy: Jour. Exp. Med., 1912, xv, 270. Strong and Co-wokkers: Jour. Am, Med. Assn., 1913, 1x1, 1713. Townsand: Jour. Am. Med. Assn., 1913, Ixi, 1717. Tunnicliff: Jour. Am. Med., Assn., 1917, Ixviii, 1029. Wilder: Jour. Inf. Dis., 1911, ix, 9. Wolbach: Jour. Med. Res., 1912, xxviii, 1 (with bibliography); also 1916, xxxiv, 121, and xxxv, 147. CHAPTER XXXVII. FLAGELLATA.I The flagellates that are pathogenic to man belong chiefly to the genera Trypanosoma and Leishvfiania, but certain other genera are found occasionally in man which may be mentioned because of the possibility of their becoming pathogenic. According to the classifica- tion of Calkins (1908) the flagellates parasitic in man are from three orders, the Monadida, the Heteromastigida, and the Polymastigida; according to that of Doflein and Koehler (1913), they are from two, Protomonadina (one to three flagella), and Polymastigina (four to eight flagella). Six families, according to the morphology of the flagella, are distinguished, three of which contain forms parasitic for man, namely, Cercomonadidw (one flagellum and, no blepharoplast), TrypanosomedcB (one flagellimi, a blepharoplast, and an undulating membrane), and Bodonidw (two flagella). Under the Trypanosomedce, which include most of the pathogenic forms, are placed the genera Leptomonas, Herpetomonas, Leishmania, Trypanosoma, and Schizotrypanum and Endotrypanum. Under the Bodonidw are placed the genera Bodo and Prowazekia, and in the order Polymastigina the genera Trichomonas and Lamblia have been found in human beings. Hartmann puts the Trypanosomata, with other blood parasites, in an order, the Binucleata, and makes the Spirocheta an appendix of this order. According to this arrangement the Hemospcnidia are taken from the Sporozoa and placed in the Binucleata with the Trypanosomata. The malarial organisms supposedly lose through their parasitism many of the characteristics ascribed to this order. Material and Methods for Study. — ^A number of flagellates (Bodo, for instance, see p. 518) are found in the large intestine of the lower animals. The toad, the grass lizard, and the guinearpig- may, contain some interesting forms. As these forms are easily obtained and remain ahve a long time outside of the body, they are well fitted for class study. The feces are obtained by pressing lightly over the anus of the animal, or if the whole intestinal tract is to be examined, by sacrificing the animal and dissecting out the parts wanted. The material is placed in a clean watch-glass and thinned if necessary with physiological salt solution. Hanging drops may be made in physiological salt solution or in such a solution made a httle thick by the addition of gelatin in order to retard the motion of the flagellates some- what so they may be better studied. Permanent preparations may be made according to directions given in Part I. As most of the pathogenic members of this group may be difficult to obtain in the living condition at any stated time, they must be studied by students principally in stained smears and sections. If one can obtain rats infected with Tr. lewisi, others with one or more patho- genic forms still others with Spirocheta obermderi, the infecting organisms ' See Part I for general description of Protozoa. 488 FLAGELLATA can be kept alive by frequent reinoculation of the heart's blood, subcutaneous or intraperitoneally into the fresh animal, or cultures may be carried on (si below). But this is an expensive and tiresome work in those laboratories whe such work is not being carried on, and generally one must rely on the pe manent preparation. In the development in the second host one must alf study the stained specimens in the great majority of instaaces. The fresh specimens of blood are obtained from the tail tip of the rat, ( the ear of the dog; they may be examined, after dilution with physiologica.1 sa solution, in the hanging drop, or in a drop spread under a cover-glass and ringe with vasehn. For permanent preparations films of the blood are spread, fixei and stained in the usual way; Giemsa's method of staining (p. 82) is very sati factory. For section work of the various organs the fixatives and methods given c page 84 may be used. Special methods are given under each organism. Artificial Cultures of Blood Flagellates (see p. 499). — ^These, accordin to Novy and MacNeal, may be made on a culture medium consistin of a mixture of ordinary nutrient agar with variable amounts of fres defibrinated rabbit or rat blood. The best all-round results ai obtained with equal parts of blood and agar. The agar is melted an cooled to 50° C, then the blood is added and thoroughly mixed. Th tubes are inclined imtil the medium stiffens, when they should be inoci lated at once with blood or other infected material containing livin trypanosomes. The surface of the medium should be very moist, s water of condensation may form. Generally evidence of growth may b observed in three or four days. CERCOMONAS. The members of this genus are round or oval flagellates with a long anteric flagellum and a more or less pointed posterior extremity which is sometimt ameboid. The vesicular nucleus is situated anteriorly, and passing through th organisms from flagellum to posterior extremity is an axial elastic fibril (Fij 169, a). Division into two daughter forms has been observed. A number of cercomonada, none of them well studied, have been observed i different animals as well as in man. They are very numerous in stagnant wate Cercovwnas hominis (Davaine, 1854) was observed in the dejections of cholera patient by Davaine. The body is 10^ to 12m long and pear-shape( .pointed posteriorly. The flagellum is twice as long as the body. Davain also reported a smaller form in the stools of a typhoid patient. Other observe: have noticed similar forms in human stools, some associated with "Ammba coli. Similar forms have been seen also in an echinococcus cyst of the liver, in tb sputum from a case of lung gangrene, in the exudate of a hydropneumothora; and a few times in the urine. They, are all probably harmless invaders. LEPTOMONAS, HERPETOMONAS, AND CRITHIDIA. Certain flagellates found in the digestive tract of mosquitoes, flies, an other insects are very similar to trypanosomes. Among them several speci( have been recognized, but they need to be more fully studied in order to dete mine their definite relationship to each other and to the genus trypanosomi Leptomonas is described as having a single flagellum directed forward an arising near a blepharoplast situated in the anterior part of the cell. Herpet( monas is distinguished from leptomonas by a flageUum containing two filamen and by a delicate filament extending from the blepharoplast toward the posteri( end. Crithidia has a rudimentary undulating membrane. The distinctioi LEI SH MAN I A 489 between these three genera and ,the genus Trypanosoma which have been recog- nized are: (1) the former contain no undulating membrane or only a rudi- mentary one, and (2) their centrosome or blepharoplast usually lies at the side of, or anterior to, the nucleus instead of posterior to it, as in Trypanosoma (Fig. 168). These distinctions, Novy claims, may disappear in the cultural forms of the three genera, when trypanosomes may show a rudimentary undulating mem- brane and an anterior blepharoplast. His caution in regard to confusing these insect flagellates with developmental stages of vertebrate blood parasites should be remembered. The Leishman-Donovan bodies found in kala-azar are closely related to the genus Leptomonas. They are considered a separate genus, Leishmania. Fig. 168. — Schematic drawings of flagellates belonging to the trypanosomedse, showing differential points: a, Cercomonas; b, Leptomonas and Leishmania; c, Herpetomonas ; d, Crithidia; e, Trypanosoma. LEISHMANIA (LEISHMAN-DONOVAN BODIES AND ALLIES). Certain fevers of severe malarial-like types known in different sec- tions of the tropics by different names (dum-dum fever, cachexia! malaria, kala-azar) have been shown to have a casual relationship by the finding of similar protozoon-like bodies in the lesions. These bodies were first minutely described by Leishman, in 1903, as being present in certain cells in the spleen of cases occurring in India, called by him dum-dum fever. He considered them as possibly trypano- somes, but did not name them. Later in the same year Donovan described similar bodies in cases of what he called malarial cachexia. The bodies were first called the Leishman-Donovan bodies; then Laveran and Mesnil who examined Donovan's preparations and considered the organisms similar to those causing Texas fever in cattle, called them Piroplasma donovani. Ross, however, thought they constituted a dis- tinct genus which he called Leishmania. This genus is now accepted, hence they are known as Leishmania donovani. Rogers and Patton place them with the genus Herpetomonas, but until we know more of the limits of variation of all these forms it seems best to make them a separate genus. They have since been found in different parts of India, in China, Tunis, Algiers, Arabia, Egypt, South Africa, Italy, Greece, 490 FLAGELLATA and Portugal. Wright, in this country, has reported in a case of tropical ulcer, or Delhi boil, from an Armenian immigrant, bodies which, accord- ing to his excellent photographs (Fig. 169) and description, must be identical with, or very closely related to, Leishman's bodies. On account of the different pathological conditions in which they are found, however, they are classed as a different species, Leishmania tropica Wright. The form found in infantile splenomegaly is considered another species, with the name Leishmania infantum Nicolle. Darling described an organism resembling that of kala-azar found in a fatal disease of tropical America. Though the organism, he says, resembles Donovani, he thinks it has enough points of difference to be placed in a different . . Fia. 169.— Leishmania tropica in a case of tropical ulcer. Smear preparation from the lesion stained with Wright's Romanowsky blood-staining fluid. The ring-like bodies with white central portions and containing a larger and a smaller dark mass are the microorganisms. The dark masses in the bodies are stained a magenta, while the per- ipheral portions of the bpdies in typical instances are stained a pale robin's egg blue The very large dark masses are nuclei of cells of the lesion. X 1500 approximately. (After Wright.) genus; therefore he gives it the name Histoplasma capsulatum, and calls the disease histoplasmosis. He says it differs from Donovani in the form and arrangement of its chromatin nucleus and in not possessing a chromatin rod. It has a refractile achromatic capsule. Morphology.— The bodies as seen in the cells of the host are circular to elliptical in shape, from 2fi to 4m in diameter, and contain two nuclei, a large oval one at one part of the periphery and a small circular or rod-shaped one (blepharoplast) near or at the opposite part of the per- iphery This smaller body stains more intensely than the larger one, while the cytoplasm of the parasite stains very dimly, sometimes show- ing only a faint peripheral rim. Near the rod-shaped nucleus may sometimes be seen a minute granule or rod which is the rudiment of LEISHMANTA 491 the flagellum. Any nuclear and cytoplasmic staining methods will bring out these points in Zenker-fixed material. Smears stain well by Wright or the Nocht-Romanowsky methods. It was not until these organisms were cultivated outside of the body that their relationship to the flagellates was established (Fig. 170 and Plate IV, i, Fig. 1, b). Fig. 170. — Cultural forma of L. infantum, showing the flagellate type. (After Nioolle.) Site in Body. — The bodies have been found in large endothelioid cells in the spleen, liver, bone marrow, lymph nodes, kidney, lungs, testes, skin, muscles, intestinal ulcers and in the leukocytes in the peripheral blood. In this last situation they are only found in appreciable numbers in advanced cases. 492 FLAGELLATA The large cells containing the parasites are supposed by Christophers to be the endothelial cells from the finest capillaries. Donovan states that he found small forms in the polynuclear leukocytes of the peripheral circulation when the temperature was above 103°. Marshall found them chiefly in the mononuclears. Rogers was able to demonstrate them more easily in the peripheral circulation by centrifuging the blood and examining the leukocytic layer. NicoUe and Conte and others obtained them in the mononuclears found in the fluid of blisters pro- duced experimentally on the skin. Cultivation. — Rogers has grown abundant pure cultures of the bodies in a slightly acid citrated blood medium at 20° to 22° C. NicoUe and later Novy have shown that L. infantum is pathogenic for dogs and that cultures may be obtained with comparative ease from the infected animals (Fig. 170). Nicolle has also cultivated L. tropicum. In these cultures the rounded organism elongates, the flagellum develops, and the Leishman-Donovan body becomes a flagellate, like the Leptomonas (Fig. 168) . Insect Carriers. — ^Rogers and Patton have shown that the bed-bug may transmit the disease, and Patton has demonstrated the develop- ment of the organism up to the fully flagellated stage in the gut of this insect. The Sargents have shown that L. infantum may be transmitted by the dog flea. Effect on Human Host. — The pathological changes are those following the degenerations, subsequent to the growth of the organisms in the large mono- nuclear cells. The symptoms in the cages of general infection are: (1) Very much enlarged spleen and leSs enlarged liver; (2) progressive anemia with pecuhar dark, earthy paUor of skin (kala-azar), progressive emaciation, and muscular atrophy; (3) long-continued, irregularly remittent, and intermittent fever (97° to 104°); (4) hemorrhages, such as epistaxis, bleeding from gums into subcutaneous tissue, producing purpuric eruption; (5) transitory edemas of various regions. There are often complications, such as congestion of lungs, dysentery, and cancrum oris. The blood shows practically no loss of red blood cells, but a diminution of hemoglobin; there is a decrease in the leukocytes, principally polynuclears, giving a relative increase of mononuclears. Negative points which help in the diagnosis are: absence of malaria, no typhoid or Malta fever reaction, resistance to medication, 'quinine, as a rule, having no effect though in early cases and with large continued doses a few good results have been reported. Puncture of spleen or liver with the finding of Leishman-Donovan bodies makes the diagnosis certain. Sometimes the bodies may be found in the peripheral blood. The duration of the disease is from a few months to several years. The percentage of deaths in systemic infections is great; in some forms of the disease at the height of an epidemic it may reach 98 per cent. Strangers from non- tropical countries are specially susceptible. The infection in children known as splenomegaly is similar to that in the adult. The local disease known as Delhi or Aleppo boil or tropical ulcer is a com- paratively non-dangerous circumscribed chronic ulcer in which the endothelial cells contain the organisms in large numbers (Fig. 169). Recovery is followed by marked immunity. Complement-fixation. — Makkas and Pappassoterion state that a specific antigen gives positive results with both kala-azar and syphilis, but a syphilitic or a non-specific antigen gives positive results only with syphilis, therefore both should be used. LEISHMANIA IN' UTA 493 Prophylaxis. — Segregation and perfect cleanliness, especially in regard to bed-bugs and fleas, are recommended as the best means of eradicating the disease. Uta. — A South American disease has been shown by Strong and his co-workers (1913) to be due to a species of Leishmania. The flagellate stage of the organism was obtained and animals were successfully inoculated from a human case. REFERENCES. Darling, S. T.: The Morphology of the Parasite (Histoplasma capsulatum) , etc., Jour. Exper. Med., 1909, xi, 515. DoFLEiN u. Koehlek: M. KoUe and Wasserraann, 1913. Sec. Ed., Jena. Makkas and Pappassotebion: Arch. d. Med., 1911. Maybe, M.: Leishmania. M. KoUe and Wasserraann, 1913. Sec. Ed., Jena. Nicolle: Le Kala-azar infantile, Ann. Inst. Pasteur, 1909, xxiii, 361 and 441. NovT, MacNeal and Tokry: Jour. Inf. Dis., 1907, iv, 223. Patton, W. S. : Scientific Memoirs by Officers of Medical and Sanitary Departments of Government of India, new series. No. 31. Sebgent: Ed. and Et. Bull. Soc. Path. Exp., 1912, v. 595. Strong and Co-workers: Jour. Am. Med. Assn., 1913, Ixi, 1713. Wenyon: Jour. Trop. Med., London, 1912, iii, 13. Wright: Jour. Med. Research, 1903, x, 472. CHAPTER XXXVIII. TRYPANOSOMA. Pathogenic Forms.— Very many species of trypanosoma have been described, and the number reported as distinctly pathogenic is increas- ing. Two of the latter are known to be pathogenic for man; a closely related form that was described by Chagas in 1909 is made a new genus. The table on page 495 gives a list of the better-known forms pathogenic for mammals with their chief differential characteristics. They are divided by Laveran into three groups according to the dif- ferent characters of the flagellum. Their general characteristics and broader classifications are given in Part I. (See also Plate IV, i, Fig. 2.) Historical Note.— The first species of trypanosome studied with any degree of fulness is the comparatively non-virulent T. lewisi. It was probably first seen in the blood of the rat in 1845, but was not well described until 1879, when Lewis studied it more fully. Since then it has been studied by many observers. It is found in the blood of from 2 to 3 per cent, of wild rats through- out the world. , , -r, The first of the more pathogenic trypanosomes was discovered by Evans in the blood of East Indian horses suffering from surra, but it was not well studied until 1893, when Lingard's important work on surra led, in a way, to all the subsequent work on diseases caused by trypanosomes. The next year a trypanosome was discovered by Bruce in the blood of horses and cattle suffering from nagana in Zululand and other parts of Africa. Bruce further demonstrated the important fact that the disease was transmitted by the bites of flies, the tsetse flies (Glossina). Aunouncenients of other pathogenic trypanosomes in different parts of the tropics quickly followed. In 1896 Rouget found that dourine, a disease of equines in Algiers and South Africa, was caused by a trypanosome {T. equiperdum) . Then the South African disease of horses, caUed mal de Caderas, was shown by Voges to be due to a ginular flagellate, while in 1902, Theiler found a variety of trypanosome in the blood of cattle in the Transvaal suffering from the disease called galziekte, or gaU sickness. The number of trjrpanosomes found in the tropics is constantly increasing — ^both pathogenic and non-pathogenic forms. Man was thought to be comparatively immune to trypanosomes until the important discovery was made that trypanosomes are the specific cause of a definite disease known as sleeping sickness, which occurs chiefly in the African negro. In 1898 Nepvieu reported having found trypanosomes in the blood of 6 out of more than 200 cases of human beings examined for malarial organisms and in a seventh case which was apparently in good health. The eighth case is reported by Button in 1901. The tenth and eleventh casefe were published by Manson in 1902. Broden published 2 more cases, and Baker 3. In 1904 Castellani stated that the cause of sleeping sickness of the negro is a trypanosome. He found trypanosomes in the centrifugalized cerebrospinal fluid of 20 out of 34 cases of this disease. His work has been fully corroborated. The trypanosomes found in these cases resemble each other; they are therefore included under the same name, Trypanosoma gambiense Button. A similar form found by Stephens and Fantham in 1911 in cases of sleeping sickness in PATHOGENIC FORMS 495 1 S 1 5 s 1 1 Intestines. Unknown. Gut and sali- vary glands. Unknown. ■i =^1 m d 0^ Proboscis. Proboscis 0" '0 1 ■ii|l 1 1 i-1 1 ax P 1 II 1 K m O H is |1 1 -3 ^- ^ 5 5 ^ ^ ^^il ■"*i 3. S 3 3 ■!i 1 « >o >>o ■i !z;t3 bi n t-i m h CO 1 H !» ^ H (-1 H 02 H H H m xh * JS 5 ll §J 1 Ir^ .1 » T3 ■e^ ■2 , -p" 3 fa d ■B-S i' iM d 03 3 a trS c o a CP aa cS > 05 H H H K W W W IS OS , 1 I1 d 2 a - 1% 1 > (l4 > > Q Ph 5' .s lO L-- . in »o IC 10 •p i-H ^ .-1 ^ ^ ^ d ^ s s o c o o *a • 1 d m s CO (N M CO CO CO 10 >s • a o oc 5 U) i-H w cd S 00 ■* Tti w »o ?H OS CO Oi (O S CO cr CO M CO CO r* 1-1 (N CO II 1"" la ■3 S<) •a H S -0 d a 1 3 t4 f N K CQ z iz; 4 .ss s 3 00 oc OS t-^ 2 * 00 05 01 Oi i-H i-i 01 Cft 05 2 ■o i ■3 0) p< :s 1 1 ^ i ■34i ^1 a i > s 1 pi: §1 ' p,a3 ■■ii .1 Si S § -a i a 3 h ■g J. •3>H ^1 H H H B H H H H H H H H ^1 ^ (M CO ■^ iO to t^ 00 01 «>a 1 o '-' ft - :^ s^ ..-"'A ■4- S A «*«. Z DAYS QUARTAN TYPE ... o 2 Pj \ /___ I DAY 2 DAYS /ESTIVO-AUTUMNAL TYPE J~J .r^ .Si. "3 ' o 2 DAYS SEXUAL FORMS (8 c ,- - E i ■a- J Kj ':~-i^ t <^ A B A. W- WILLIAMS, DEL. .0t Q ■•I'r.'p C D ^ 4 ^ -.V * e H ASEXUAL CYCLE iN THE BLOOD OP MAN 539 Each of these species undergoes the two phases of development already alluded to, one withm the red blood cells of human beings (the asexual phase) ; the other within the digestive tract of the mosquito (the sexual phase). The form changes which each parasite undergoes in humans and which the benign tertian undergoes (which may be considered a type for all) in both hosts are shown on Plate VII. Briefly, they may be described as follows : The Asexual Cycle (Schizogony) Occurring in the Blood of Man.— The young form is often difficult to find in fresh blood. A pale area is seen on an otherwise unaltered red corpuscle, situated usually eccen- trically, about one-tenth the size of the red corpuscle or about one-fourth its diameter, when at rest presenting a rounded appearance, but usually actively ameboid, throwing out distinct pseudopodia never remaining long in the same focal plane, frequently dipping, so to speak, into the substance of the corpuscle. It is often called the hyaline form because it is free from pigment, but it is not hyaline in the proper sense of the term. It is also called the ring form, because of its resemblance to a ring in stained preparations; but it is never a true ring. The ring appearance is produced by the formation of a large food vacuole. The organism continues to grow either on or in the red blood cell. Rowley claims that the organisms never enter the red blood cells. They only feed from the surface. Her beautiful demonstrations show that possibly the majority of them do this. The forms intermediate between this and the segmentation stage appear in the fresh blood simply as larger parasites, which are readily found ■ on account of the reddish-brown pigment granules that they contain. These granules begin to appear several^ hours after the organism has infected the red blood cell. At this time the organism is usually actively ameboid and the granules have a lively dancing motion, . due to protoplasmic currents in the parasite. The infected corpuscle is swollen and paler. In the estivo-autumnal form the infected red blood cells are smaller than normal. When the parasite has approached nearly to its full growth, it occupies the greater portion of the corpuscle, which is now more difficult to make out\ The pigment is still more evident, so that this form is therefore most readily found. At this stage ameboid movements are riot so active. When full growth is reached, segmentation occurs. The forms up to the period of segmentation are called schizonts. The morphological changes which have been going on in the parasite preparatory to segmentafion are best studied in properly stained smear preparations. In the living organism they become presently suffi- ciently distinct to be followed; the pigment gathers more or less centrally into a compact mass, and a peripheral notching indicates that the parasite is preparing to divide into a number of segments called mero- zoites; the number of these segments varies in the different species. (See table.) Suddenly the segments separate as small spheroidal 1 See table for number of hours in each species. 540 THE MALARIAL ORGANISMS— BABESIA < CI p o n m S < o a fl 3 a a " s ^S-o- » >i O ^r -J? 0) E-i 23 63-' I -isle . ■3'3.Sxi> O c3 3 a a S 3 >> o a t>. d43c« m "« o o « C o"^ o a-§.£-a:g.s-= o o o 5 O m ^9-^-9 5 2 O 03 o e;.3 o IS 02 ^»^ (u 9 fl ^■B & III S © d !3 S^ is r5 S S o o O B o ■^ oJ o a,53 tj P B OJ' g o ^Hw ft CD oS oj.S d O ki . c w a-g-B ^ S a:> a !§■* '3 2 o fa -^ j3 r o 03 ASEXUAL CYCLE IN THE BLOOD OF MAN 541 bodies, the young parasites. A corpuscular remnant and the pigment float away and are ultimately ingested by phagocytic cells. The young parasites attach themselves to red corpuscles as before and the human cycle is repeated. In a suitably stained preparation the young parasite (see Plate VII), appears to be a disk consisting of a central pale, unstained area, known as the achromatic zone, and of a basic (blue) periphery, the body, including a metachromatically stained, rounded, compact (red) chroma- tin mass, the nucleus, which tends to give the parasite the form of a signet ring. Later stages up to a certain number of hours show simply changes in size and outline of the body. The nucleus then divides by simple mitosis. Later it breaks up by amitotic division into an increasing number of small masses. By the time the chromatin division is com- pleted the chromatin masses will have assumed a rounded form, and will be seen to exhibit ultimately the same strong affinity for certain dyes which is seen in the compact chromatin body of the young ring- like form. At this stage the heretofore scattered pigment appears in one clump. Good technic will always show a corpuscular remnant even at this time. The achromatic zone mentioned will be seen to develop with the chromatin, and when the next step, namely, the division of the body of the parasite, is seen to be completed, there will be as many achromatic bodies as there are chromatin bodies, each division having a share of the basic mother-body, each representing the young parasite (merozoite). A certain number of the full-grown parasites do not segment and these are the forms which commence the life cycle in the mosquito. These forms. grow to produce the sexual forms, the macrogametocyte, or female organism, and the microgametocyte, or male organism. When mature these forms are generally larger than the mature schizont of the same species, the female organism being usually larger than the male and containing more food granules and a smaller nucleus. In the estivo-autumnal forms they are crescentic in shape, while in the other species they are spherical. In the circulating blood of human beings they show no further changes except to become freed from the corpuscle; but when the blood containing them is withdrawn and exposed for a short time to the air, an interesting series of changes in the microgametocyte is observed. The crescentic bodies are trans- formed into spherical bodies; the pigment of the microgametocytes becomes actively motile, due to internal agitation of the chromatin fibrils, which presently emerge as flagella-like appendages. Their movements are very rapid, causing corpuscles to be knocked about, and finally they become detached as the microgametes, or male elements, and go in search of the female element. In the withdrawn blood of birds, one may actually observe the process of conjugation in slide preparations even without the aid of a moist chamber and heat. This transformation of male bodies never occurs in the human blood. It will be seen that it belongs to the sexual cycle which occurs in the stomach of the mosquito. 542 THE MALARIAL ORGANISMS^BABESIA The Sexual Cycle (Sporogony) Developing in the Mosquito.— The common mosquito, often day-flying, belongs to the genus Culex; it cannot carry human malaria. It is easily distinguished from its night- flying or dusk-flying relatives. Anopheles (the malaria-carrying mos- quitoes comprise about eight genera of the subfamily anophelinae), by its assuming a difl'erent posture on the perpendicular wall. While the Culex holds the body more or less parallel with the surface, the body of the Anopheles stands off at a marked angle. Other differential points are the following (see Figs. 184-193) : Wings of Culex are unspotted; those of Anopheles are spotted (except in one rare species). The proboscis of Anopheles points toward the resting surface, while that of Culex does not do so. Anopheles species bite usually in the early evening, while those of Culex bite almost at any hour of the day. The male mosquito is readily told from the female by its plumed antennse, those of the female being inconspicuous. The eggs and the larvae of the two genera are quite distinct, as may be readily seen by glancing at Figs. 184 and 185. The anopheles mos- quitoes breed in practically any kind of a collection of water, though some species prefer slow -running water to quiet pools. The best known domestic carriers are usually found in barrels and cisterns. If an ordinary mosquito (Culex) is allowed to imbibe the blood of a malarial patient whose blood shows gametocytes there will be simply a digestion of such blood in the mosquito, and no development of the malarial organisms results. If, however, certain species of Anopheles ingest such blood, immediate changes follow. It should be remem- bered that only female mosquitoes are blood sucking; hence they alone can be responsible for the spreading of the disease. It should also be remembered that if the blood imbibed by the anopheles does not contain gametocytes, though it may contain earlier stages of the malarial organisms, no amount of such blood can cause general infection of the mosquito. The sexual cycle is similar in all species of the parasite. The flagellation of the male parasite described above will promptly take place in the stomach of the anopheles, 4 to 8 microgametes being formed; these conjugate with the female element (Plate VII) in a manner comparable to the impregnation of the ovum of higher animals by spermatozoids. The macrogametocyte becomes a macrogamete by the formation of a reduction nucleus which is thrown out of the organism. The product of conjugation, the ookinet (zygote), remains for a number of hours in the juices of the chyme stomach, changing gradually from a spherical, immobile body into an elongated motile wormlet (Plate VII, Sexual Forms, 1 h). This penetrates the epithelial lining of the stomach and rests in the tunica elasticomuscularis (Plate VII, Sexual Forms, Fig. 2) ; here it changes into an oval then into a round body, which grows in the course of the next few days enormously, forming a cyst which projects into the body cavity. Meanwhile the Fig. 184 Fig. 185 ^ \l\^ Fig. 186 Fig. 187 >' Fig. 190 Fig. 191 Fig. 192 Fig. 193 Chief comparative characteristics of Culex and Anopheles. (From Kolle and Hetsch.) Egg of Culex, Fig. 184, laid together in "small boat," those of Anopheles, Fig. 185, separate and rounded. Larva of C, Fig. 186, hangs nearly "at right angles to water surfaoe,those of A., Fig. 187, are parallel to surface. Body of C, Fig. 188, when resting is held parallel to wall in a curved position, that of A., Fig. 189, stands at an angle of about 45° and is straight; wings of C, Fig. 190, are generally not spotted, those of A., Fig. 191, are spotted. In C. the palpte. Fig. 192, of the female are very short, of the_^ male are longe- than the proboscis; in A., Fig. 193, the proboscis of both sexes are about of equal length. 544 THE MALARIAL ORGANISMS— BABESIA chromatin will have become very active. It will have divided into numerous nuclei, which become arranged around inactive portions, and filamentous sporozoites develop from this chromatin and surrounding pfotoplasm (Plate VII, Sexual Forms, Fig. 2). These sporozoites ulti- mately fill the cysts, which rupture, setting them free into the cavity of the mosquito's body; they then are .carried by the lymph to all parts of the body of the mosquito and thus reach a glandular structure in the thoracic cavity of the insect, the so-called salivary gland (poison gland), m which they accumulate in large numbers. This gland is in immediate connection with the biting and sucking apparatus. If, now, such an infected mosquito " bites" a human being, the lubricating fluid of the puncturing apparatus will carry sporozoites into the latter's blood and the human cycle begins. The stages of development in the mosquito require from seven to ten days, but only when the temperature is favorable. Cultivation. — Bass and Johns announced in 1911 that they had succeeded, by a rather complicated method, in obtaining a certain amount of development of H. Vivax in the test-tube (see p. 106 for details of method). Essentially, defibrinated malarial blood is used, to which is added | per cent, dextrose. Bass and Johns state that the layer of serum about the sedimented red blood cells must not be too deep (I inch) and that the leukocytes must be removed if more than one generation of the parasite is wished. The Thompsons (1913) state that they have gotten several generations by less attention to these details. They draw the malarial blood into the sterile test-tube (10 c.c.) in which there is a thick wire, and 0.1 c.c. of a 50 per cent, aqueous solution of glucose. The blood is defibrinated by gently stirring the wire for five minutes. The wire and clot are removed, and the blood is poured into smaller tubes, 1 inch in each. Rubber caps are placed over cotton plugs. These tubes are kept at 37° to 44° C. The corpuscles settle slowly, leaving about i inch clear serum. They have found it unnecessary to remove the leukocytes by centrifuging. They state that the malarial organisms are not destroyed by the leukocytes in the tube but by the changes in the serum. Their organisms grow through the whole depth of the layer of red blood cells. EfCect on Man (Pathogenesis). — ^As the organism grows at the expense of the red blood cells the principal change is in the blood. Melanemia, or the formation of pigment granules from the destroyed red blood cells, is one of the most characteristic features of malaria. As the disease progresses the red corpuscles show varying changes in form and hemoglobin content, not only the infected corpuscles, but others as well, thus showing that the organism produces either primarily or secondarily some toxic substances. The pig- ment occurs in two forms, melanin and hemosiderin. The second only gives the reaction for iron and is found in the internal organs, while the first is found everywhere in the circulating blood. The pigment is taken up by the leuko- cytes. There is usually a definite reduction of both red and white blood corpuscles, which is more marked in tertian and quartan malaria than in estivo- autumnal. There is a relative increase in the number of mononuclear leuko- cytes. The spleen shows marked hyperplastic inflammation and pigmentation. In intense estivo-autumnal cases the capillaries of brain and other organs may be fiUed with the parasites. We have observed parasites also in the large nerve cells of the brain. POINTS OF DIAGNOSIS 545 Toxic Production. — ^The relationship between segmentation and paroxysm is always noted in tertian cases, and it is reasonable to sup- pose that the occurrence of the paroxysm is referable entirely to the liberation of toxic substances resulting from metabolic activity of the parasite within the corpuscle. That there should be a toxic product seems highly probable, and its amount must be considered in heavy infections. Cases showing an infection of 1 to 5 per cent, of all cor- puscles are not infrequent; the destruction of from 50,000 to 200,000 or more corpuscles per cubic millimeter of blood leads to the rapid deglobularization of the blood; hence the deficiency in numbers; add to this the effects of the metabolic products, and little is left to the imagination to explain the pronounced anemia. Immunity from malaria appears to exist as natural and acquired immunity. Prophylaxis. — The fact that, with the extermination of the malaria- carrying mosquitoes, malarial fevers in man would be made impossible, remains established; the parasite must have its chance of rejuven- escence in the mosquito's stomach. The various methods of extermination are fully described in books which go minutely into the subject. The method of giving small doses of quinine to human beings exposed to Anopheles, and of thus getting rid of the organism itself within man, should be considered. In hot climates especially, where it is practically impossible totally to destroy the breeding places of the mosquitoes by drainage or oiling, this method is especially serviceable. In these countries, too, the use of adequate screening is of marked value. Points of Diagnosis. — ^By a study of the parasite taken from the circu- lating blood the examiner should be able to tell not only the species present, but also the progress the disease is making. Malarial parasites can always readily be found in recent primary infections, and it is usually only in old cases that the search becomes difficult; one is, however, generally rewarded by finding them if one looks long enough for them. A helpful sign is the finding of pigment in mononuclear leukocytes, which are seen about the time of a chill or of the period symptomatically corresponding to it. Free pigment cannot be used as a means of diagnosis, as it may be impossible to tell it from dirt or dust. In a primary infection of long standing the gametocytes may be found, and in relapses and in those cases treated by quinine, many atypical forms appear. A small dose of quinine may drive all parasites except the sexual forms out of the peripheral circulation; at all events the finding of them becomes, in the absence of gametocytes, a matter of time and experience, especially also as they may be much altered in appearance. The part most and first affected is the blue-staining body; later follow eccentricities of the chromatin, such as multiple bodies, and dwarfing, just such changes as might have occurred in time, if the body had been allowed to combat the parasite without the aid of drugs. In both cases the fever curve becomes atypical. It 35 546 THE MALARIAL ORGANISMS— BABESIA should be remembered that there is no quotidian form originating in this country. Quotidian paroxysms occurring here are either a double tertian, or a triple quartan infection. The notion that the parasites can be found only at the time of the paroxysm is still in the minds of many; it is erroneous. The gametocytes are quite resistant to qui- nine and other drugs, and it appears as if cases in which these forms are seen are much more prone to relapse than promptly treated recent primary infections. The macrogametocytes may remain quiescent for years in the blood, and then under certain conditions, probably through parthenogenesis, may again begin to develop and multiply, thus bringing about relapses. In the estivo-autumnal forms the crescentic gametocytes are gener- ally few, but at times large numbers of them develop. Of course they are absolutely characteristic. The young parasites are more or less characteristic in stained preparations (Plate VII). There may be as many as seven parasites in one corpuscle. Later the few heavy pigment granules are characteristic. In fatal cases the formation of crescents may not take place; the blood infection with young parasites is then enormous, every field of the microscope showing numbers of them. DESCRIPTION OF PLATE VIII. (After Goldhorn.) 1. Typical youDg tertian form; the corpuscle shows incipient degeneration; corpuscle to left above shows a blood platelet. 2. Abnormal young form, showing small accessory chromatin body. 3. Two parasites; one normal young form; the second a large form in crenated corpuscle ia an unusual abnormal form with very large achromatic area. 4. 5, 6. Estivo-autumnal parasites; single, double, and triple infection; central elongated chromatin bodies. These forms are about the largest usually seen in the peripheral blood; no degeneration of corpuscle. 7. Tertian parasite, about ten hours old; marked degeneration of corpuscle. 8. Double infection of a corpuscle in tertian fever; marked degeneration of corpuscle. 9. 10, 11. Large tertian parasites showing division of chromatin previous to segmenta- tion. 12 and 14. Complete segmentation of tertian parasite. 13. Double infection of corpuscle, one parasite reaching maturity, but showing unusually small segments; the second one atrophied. 15. Tertian parasite, old case; while the parasite is only half-grown, the chromatin has split into several compact masses. Degeneration of infected corpuscle. 16. Dwarfed tertian parasite, smaller than a red corpuscle, but showing five compact chromatin bodies; resemblance to quartan rosette. 17. Microgametocyte of tertian malaria; prominence of blackish pigment surrounding a large achromatic zone in which the microgametes lie coiled up. 18. Tertian macrogametocyte. 19 to 23. Crescentic bodies of estivo-autumnal malaria. 19. Typical gametocyte; pigment surrounding achromatic area; no chromatin shown; the "bib" is present. (Male?) 20. Semiovoid gametocyte. (Female?) 21. Pigment removed. Elliptical achromatic area in which the microgametes are seen. 22 and 23. Pigment removed; chromatin more compact; possibly female elements. 24. Froqi a case of pernicious malaria with rich infection ; only hyaline forms in per- ipheral blood. Below, a large blood platelet. Note. — As the amplification is not uniform, a comparison of the parasites with the blood corpuscles shown should be made in order to have a correct conception of their PLATE VI i; Photographs of Tertian and Estivo-autumnal Malarial Parasites in Different Stages of Development. (Goldhorn.) MALARIAL-LIKE PARASITES IN OTHER ANIMALS 547 In the study of estivo-autumnal fever, as well as in that of the other forms, it is to be remembered that crescents when found indicate that the disease is of some standing, for such sexual forms are not formed until the asexual propagation is waning. The recognition of these ovoidal and crescentic bodies is easy. But as there are no readily discoverable pigmented forms in the peripheral blood in the early stages, it is necessary to be thoroughly familiar with the young estivo- autumnal forms. Polychrome staining for them cannot be too much recommended, as there is little that is characteristic about them when they have been stained with methylene blue alone. Many a serious error has been made by adhering to the antiquated idea that parasites should be looked for in the fresh blood, as these young, non-pigmented, so-called hyaline forms cannot be readily recognized by the inexperi- enced, while it is an easy matter to know and classify them when properly stained. The recognition of the quartan parasite in its early stages in the fresh blood is not as difficult as that of the tertian form, because the outline is more distinct; but in stained preparations it is often indis- tinguishable from the latter. The living ameboid young form or schizont is more refractive than the young living tertian schizont, more like the estivo-autumnal form, and it is just as sluggish in its movements. Then, too, the corpuscle is often shrunken and looks as if it contained more hemoglobin than in the case of infection with the tertian parasite. The growing parasite rapidly becomes pigmented, but it shows fewer, larger, less motile pigment granules than the corresponding tertian one; moreover, the pigment is arranged around the periphery of the organism, while in the tertian form it is distributed through- out the protoplasm. The quartan parasite is apt to form a band across the infected corpuscle. Segments are few in number, as a rule, and the parasite remains dwarfed while the infected red blood cells are normal in size. The segments are generally arranged symmetrically around the central pigment, giving the so-called daisy or marguerite appear- ance to the parasite at this stage (Plate VII). In tertian fever the granular degeneration which the infected cor- puscles early undergo is diagnostic. In the first few hours it resembles the ordinary granular stroma degeneration with basic affinity, while it is later seen that the affinity of the then more numerous granules is more acid, or, at least, the staining is no longer orthochromatic, the blue being superimposed by a red; in other words, these granules stain later metachromatically. The greater the loss or transformation of the hemoglobin the greater the number of granules. This holds good only for tertian parasites, the estivo-autumnal variety causing practically no appreciable change, though the same technic is used. Malarial-like Parasites in Other Animals. — Two genera of protozoa closely related to the malarial organisms have been found in birds: (1) the proteosoma or hemoproteus; (2) the halteridium; both found 548 THE MALARIAL ORGANISMS— BABESIA in owls {HcBmoproteus noctucB Celli and Sanfelice). Points in their life history have been brought out by various observers, especially by Ross and by MacCallum. The complete life cycle of both forms, as worked out by Schaudinn, is considered by him and his followers to be of fundamental importance to the understanding of the rela- tionship of blood parasites. Schaudinn states that these organisms pass through a flagellate stage in the intestinal tract of the common mosquito (Culex pipiens) which has previously fed on owls infected with the intracellular organisms (halteridium and hemoproteus). Novy considers that this mosquito flagellate stage of Schaudinn is simply a growth of trypanosomes in the mosquito's intestinal tract which are normally found there, and that Schaudinn did not sufficiently control his work to warrant his conclusions. Malarial-like organisms have been found also in monkeys, cattle, dogs, and frogs, but they have been little studied. An interesting article by Bernberg-Gossler on malarial organisms in monkeys was published in 1909. In it the author describes a binucleate phase of these plasmodia and agrees with Hartmann in his recent classification of these organisms. Blackwater Fever. — This is a condition which occurs frequently, especially in Europeans in tropical countries. Its etiology has been the subject of much discussion. The chief symptoms are fever, hemoglo- binuria, delirium, and collapse. It frequently ends in coma and death. A few consider it a disease entity, but the majority of observers are inclined to believe it the terminal stage of a severe malarial infection. It is frequently associated with the demonstration of the malarial organism, but they are not always found. It may be that the invasion of the ganglion cells of the brain by the malarial organisms are the chief cause of the symptoms, aided in certain cases by the lysemia noted by Christophers and Bentley. GENUS BABESIA (PIROPLASMA). It was not until 1888 that there was a hint as to the real nature of the actual cause of "Texas fever" (bovine malaria, tick fever, hemo- globinuria) and allied diseases which attack field cattle in many parts of the world. Then Babes described inclusions in red blood cells in Roumanian cattle sick with the disease, though he did not decide upon the nature of the organism. No new studies were reported until 1893, when Theobald Smith and Kilborne gave such a complete description of this disease and its cause as occurring in Texas cattle that little concerning it has since been discovered. These authors describe as the cause of Texas fever, pigment-free ameboid parasites appearing in various forms within the red blood cells of infected animals. The organisms may be irregularly round and lie singly or they may be in pear-shaped twos, united by a fine line of protoplasm. BABESIA 549 Because of these double pear-shaped forms Smith and Kilborne named the organism Pyrosoma bigeminum} and they placed it provisionally among the heinosporidia. These authors also showed that the contagion was carried by a tick (see p. 550). Their work has been corroborated by many investigators in different parts of the world. Hartmann places this genus in his new order Binucleata, and he considers it an important form for showing the relationship of the endocellular blood parasites to the flagellates. Schaudinn, in 1904, was the first to call attention to the occurrence of nuclear dimorphism in B. cards and boms, and Luhe, Nuttall and Graham-Smith, Breinl and Hindle, and others have confirmed this observation. The second nuclear mass is generally in the form of a small granule similar to the blepharoplast pf undoubted flagellates. Morphology of the Parasite. — ^In the examination under 1000 diameters of fresh blood of sick cattle, according to Smith and Kilborne, are seen, in the red blood cells, double pear-shaped forms and single rounded or more or less irregular forms. The size varies, though generally it is the same among the bodies in the same red blood cell. The average size is 2/i to 4/i long and l^ju to 2m wide. The pointed ends of the double form are in apposition and generally touch, though in unstained specimens a connection between them cannot be seen. The axis forms either a straight line or an angle. The protoplasm has a pale, non-granular appearance, and is sharply separated from the protoplasm of the including red blood cell. The small forms are generally fully homogeneous, whereas the larger ones often contain in the rounded ends a large rounded body, 0.1/* to 0.2/i in size, which is very glistening and takes a darker stain. Within the largest forms in the centre of the thick end is a large round or oval body, 0.5/* to l/i, which sometimes shows ameboid motions. Piana and Galli-Valerio (1895 and 1896) and other observers have since described definite ameboid motion of the whole parasite. The motion of the whole parasite on the warm stage is not produced by the formation of distinct pseudopodia, but by a constant change of the boundary. The changes can succeed each other so quickly that it is scarcely possible to follow them with the eye. The motion may persist for hours. The single ones show motion, while the double ones remain unchanged. The parasites take most basic aniline stains well. The Romanowsky method or its modifications gives the best results (Plate IV, Fig. iii, b). Stained by this method the smallest forms appear as tiny rings, about one-sixth the diameter of the red blood cell. A part of the rim takes the red nuclear stain, the rest is blue. In the large mature pear-shaped organisms a loose mass of chromatin is at the rounded end and a dense, compact mass is situated nearer the pointed end. These mature, pear-shaped forms, Nuttall states, are the mark of distinction between Piroplasma (Babesia) and other intracorpuscular blood pftra- sites. These pyriform bodies are generally present in pairs, and 1 The generic name Pyrosoma, already in use for a well-known Ascidian genus, was altered to Piroplasma by Patton in 1895. In the meantime Starcovici (1893) had given the name Babesia boms to the form described by Babes; and as this form seems to be identical with that described by Smith and Kilborne the correct name of the genus should be Babesia while the species parasitic in cattle should be called Babesia bigemina. 550 THE MALARIAL ORGANISMS— BABESIA occasionally, in the acute form of the disease, sixteen pairs may be seen in a single blood cell. The number of red cells infected is about 1 per cent, of the whole. If the number increases to 5 per cent, or 10 per cent., it generally means the death of the animal. The parasites quickly disappear from the blood after the disap- pearance of the fever. In fatal cases many parasites are found in the red blood cells of the internal organs. They vary in number according to the stage at which death occurs, are most abundant in the liidneys (50 to 80 per cent, of all red corpuscles infected), and are found in fewer numbers in the liver, spleen, and other internal organs. R. Koch has described a baciUar form which he found in large numbers in red blood cells of acute fatal cases in East Africa. Between these and the pear- shaped forms he found all grades. This variety is probably a distinct species. MageUa-like appendages in Babesia have been described by several observers as occurring in the blood in mammals. More frequently they have been seen in the tick and in attempted cultures. They have been interpreted by some (Hartmann, CaUdns) as possible microgametes, by others (Breinl and Hindle) as true flagella, and by others (most observers) as fine pseudopodia. Smith and Kilborne showed that the infection is caused by a species of tick, Margarojms annulatus Say {Boophihis bovis) (Fig. 194), and Kossel gives Ixodes redivius as the tick causing transmission 9f the germ in the hemoglobinuria of Finland cattle. Fig. 194. — No. 1, Texas fever tick, Margaropus annulatus (Boophilus bovis). No. S, natural size. (Mohler.) X 15.4 The ticks feeding upon the blood of cattle and other mammals become sexually mature at their last moult. They then pair, and the fertilized females, after gorging themselves with the blood of their host, drop to the ground. Each female then lays about 2000 eggs, and within the shell of each egg a large quantity of blood is deposited to serve as food for the developing embryo. The female then shrivels up, becoming a lifeless skin. The newly hatched larvae containing in their abdomens some of the mother-blood, crawl about until they either die from starvation or have the opportunity of passing to the skin of a fresh host. If the mother-tick has drawn its supply of blood from cattle infected with piroplasma, her larvse are born infected with BABESIA 55l the parasite and become the means of disseminating the disease further. This mode of dissemination explains the long incubation period of the disease (forty-five to sixty days — ^thirty days for the development of the larvae and the remainder for the development of the parasite within the host). It is possible that the tick embryo acquires the infection secondarily from the blood it absorbs in the egg, and that the parasites do not pass through the ovum itself as in Nosema bombycis. This species of tick M. annulatus has been found also on sheep and ponies. So far it has not been possible experimentally to inoculate animals other than cattle with these parasites. Calves withstand the infec- tion better than older animals and a certain degree of immunity is reached in some of the older cattle in infected districts. The piro- plasmata taken in by such animals may remain as harmless parasites for some time. If, however, such cattle are weakened from any cause, their resistance to the organism may be lowered and they may there- fore pass through a more or less severe attack of the disease. Symptoms of the Disease. — Fever (40° to 42° C), anorexia, weakness, increased pulse and respiration, decreased secretion of milk, hemoglobinuria at the height of the fever, causing the urine to appear dark red like port wine or darker. The urine may contain albumin even if the hemoglobinuria is absent, but there are no red blood cells present, the color being due to the coloring matter of the blood only. There is icterus of the mucous membranes if much blood is destroyed. The prognosis varies in different epidemics from 20 to 60 per cent. Death may occur in three to five days after first symptoms appear. Recovery is indicated by a gradual fall of the fever. Treatment. — Quinine in large doses seems to have helped in some epidemics. Nuttall, Graham-Smith, and Hadwen have reported curative effects from trypanblau in both canine and bovine babesiosis (Piroplasmosis) . Prophylaxis. — Stalled cattle are not infected, but it is impracticable to keep large herds of cattle stalled. If the cattle are kept from infected fields for one or two years and other animals (horses and mules) are allowed to feed there the ticks may disappear. The burning of the field for one season may have a good effect. If animals cannot be taken from infected fields such fields should be enclosed. Ticks on animals may be killed by allowing the cattle to pass through an oil bath (paraffin, cottonseed oil, etc.), whereupon the ticks die from suffocation. The bath should be repeated after a week in order to kill any larvae which may have developed. All animals sent from infected regions should receive this treatment. Animals apparently healthy before the treatment, after the dis- turbing influence of the bath often develop the disease in an acute form and die. Certain birds in AustraUa seem to feed on the ticks, therefore such birds might be propagated. Various attempts have been made to give protection by the inoculation of fresh (not older than two or three days) blood from sUghtly infected animals. Some partial results have been reported, especially when the inoculations were made during the cold months. In Australia the inoculation of defibrinated blood from animals which have just recovered from the infection, but "whose blood still contains some parasites, has been tried. So far no absolute protection has been produced, neither does the parasite-free serum of animals which have entirely recovered from the disease seem to contain protective qualities. 552 THE MALARIAL ORGANISMS— BABESIA Cultivation. — Thompson and Fantham have reported successful development in the test-tube after the method of Bass and Johns for malaria. Nuttall and Graham-Smith report a study of canine piroplasmosis, and have drawn a cycle showing the usual mode of miitiplication in the circulating blood. They consider B. cams a species distinct from B. bovis and B. pithed (found by Ross, in 1905, in blood of a species of cercopithecus) though no morphological differences are given. Christophers has described probable sexual stages of development in the tick R. sanguineus, so that he has drawn a complete life cycle of the organism. Other Blood Organisms. — Blood organisms similar to those described in the hemoglobinuria of cattle have been found in cases of red water fever of cattle in England. They also occur in monkeys, dogs, sheep, horses, and pigeons. Nocard and Motas, who have made an extensive study of these parasites in the maUgnant jaundice (hemoglobinuria, malaria, or biliary fever) of dogs, state that though the parasites are morphologically similar to those infecting cattle, yet it is impossible to infect cattle or any other animal tried with them. They must therefore be considered a physiological variety. Strong and his collaborators reported that in an extensive study of Oroya fever, a tropical disease, they had determined the cause to be a Babesia-like organism which they called Bartonia bacilliformis. REFERENCES. Bass and Johns: Jour. Exp. Med., 1912, xvi, 567; Am. Jour. Trop. Dis., 1915, iii. 298. Behenbreg-Gossler: Beitrag zur Naturgeschichte der Malariaplasmodien, Arch, fiir Protistenkunde, 1909, xvi, 245. Christophers; Jout. Trop. Med., 1907, x, 323. Craig: The Malarial Fevers. Osier's Modern Medicine, Philadelphia, 1907, i; The Malarial Fevers, etc., 1909, Wm. Wood & Co., New York, Erst edition. Howard: Mosquitoes. Osier's Modern Medicine, Philadelphia, 1907, i. Kinoshita: Arch. f. Protisterk., 1907, viii, 294. Koch: Ztschr. f. Hygiene, 1901, xlv, 1. Marchiafava and Bignami: Malaria, Twentieth Century Practice, New York, 1900. Miyajami: Philip. Jour. Science, 1907 ii, 83. Nuttall and Graham-Smith: Jour. Hygiene, 1905, v, 485; 1906, vi, 586; 1907, vii, 232; also Parasitology, 1909, ii, 215, 229, 236. Rowley: Jour. Exp. Med., 1914, xix, 450. . Ruge: KoUe and Wassermann's Handbuch der Pathogenen Mikroorganismen, 1913, 2d ed., Jena. Smith, T. H., and Kilborne: United States Dept. of Agri., 1893, Bull. No. 1. Strong, Tyzzer, Brtjes, Sellards and Gastlaburn: Jour. Am. Med. Assn., 1913, Ixi, 1713; 1915, Ixiv, 805 and 965. Thayer and Hbrickson: The Malarial Fevers of Baltimore, Johns Hopkins Hospital Rep., 1895, v. Thompson and Thompson: Ann. Par. and Trop. Med., 1913, vii, 509. Thompson and Fantham: Ann. Par. and Trop. Med., 1913, vii, 621. CHAPTER XLIV. SMALLPOX (VARIOLA) AND ALLIED DISEASES. Introduction. — The diseases smallpox, cow-pox, vaccinia, horse-pox, sheep-pox, if not identical, are closely allied. Indeed, the following facts seem to prove that at least cow-pox and variola are very closely related, if not essentially the same disease : First, smallpox virus inocu- lated into calves produces, after passage through several animals, an affection exactly similar to cow-pox. The successful inoculation of the first series of cattle from smallpox is a matter of great difficulty, but so many experimenters have asserted that this has been done that there seems to be no doubt as to its truth. In our laboratory not one of many attempts to accomplish it has been successful. Second, both when occurring in nature and when produced by experiment the lesions of the two diseases are similar. Third, monkeys have been successfully protected against either disease by previous inoculation of the other; also, observations go to show that human beings inoculated with cow-pox vaccine are not susceptible to inoculation with smallpox virus, and that those who have within a varied time passed through an attack of smallpox cannot be inoculated successfully with cow-pox vaccine. These facts seem positively to prove that the two diseases are produced by organisms originally identical, one being modified by its transmission through cattle, the other through human beings. Variola is perhaps the most regularly characteristic of the diseases of man. It is highly infectious and is controlled only by vaccination. Notwithstanding the fact that we know definitely the exact site of the infective agent in this disease arid that certain experimental animals are susceptible to inoculation of the material containing the infective agent, most investigators are still undecided in regard to the nature of the chief exciting factor. A few, however, claim that certain bodies found chiefly in the epithelial cells of the skin and mucous membranes in the specific lesions are protozoa causing the disease. Definition. — Smallpox (Synonyms: Variola, la variola, Blattern, Pocken, Vajiiola) is an acute infectious disease characterized by an epidermic eruption of vesicles and pustules which, upon healing, produce cicatrices of varying extent and depth. Historical Note. — ^The first undoubted description of the disease was given by Rhazes in the tenth century, but it is evident that he did not consider it a new disease. To trace its original home seems to be impossible. It may have developed first in certain regions in Asia and Central Africa where it is at present endemic and is said to be uncontrolled by vaccination. Many outbreaks of the disease in the United States can be traced directly to the importation of African negroes. 554 SMALLPOX AND ALLIED DISEASES The disease, carried by the intercommunication, principally of war and com- merce, was widespread when Edward Jenner showed conclusively in 1798 that vaccination with cow-pox afforded protection. Now the few cases of variola that occur are seen in those who, through neglect or ignorance (sometimes wilful), have not been vaccinated. Etiology of Variola and Cow-pox. — It has been repeatedly shown that no bacteria similar to any of the known forms have a causal rela- tion to these diseases. In om- own laboratory we are able, by the inoculating of rabbits' skins, to produce extremely active vaccine virus in large quantities, absolutely free from microorganisms which grow under the conditions of our present methods of bacterial culti- vation. Such pure active vaccine, when emulsified in equal parts of ' glycerin and water and filtered through two or three thicknesses of the finest filter paper, gives a slightly opalescent filtrate, which in the hanging drop under high magnification shows many very tiny granules with an occasional larger one, and in smears shows no formed elements giving characteristic stains. This filtrate, from which no growth was obtained on artificial culture media, when rubbed over a freshly shaved rabbit's skin after the method of Calmette and Guerin, or when used to vaccinate hmnan beings, gives an abundant typical reaction. These facts show that some, at least, of the infective forms cannot as yet be made to grow outside of the body, that such forms are very minute, and that they do not stain characteristically with our usual methods of staining. In a few experiments we were unable to filter the virus through a Berkefeld filter under forty pounds' pressure, but this may have been due to the fact that we did not dilute our virus sufficiently. Since then Bertarelli and a few others have reported that the virus is slightly filtrable under pressure. Park and Wilson, however, on later experiments (1913) still obtained only negative results. Steinhardt, Israeli, and Lambert (1913) report that evidence of multiplication of germs may be obtained by growing the virus on living tissue in vitro. Since Guarnieri, in 1892, claimed that certain inclusions present in the epithelial cells of the lesions of smallpox in a rabbit's cornea (Fig. 195) were parasites, much attention has been given to the study of these bodies, commonly known as "vaccine bodies," yet opinions still differ as to their nature. Among the more important studies of these bodies are those, on the one hand, by Councilman and his asso- ciates, who believe them to be protozoa, and, on the other, by Ewing, who believes that all of the forms so far described are degeneration products, some specific, others not. Calkins, working with Councilman, thinks that his original tentative cycle is too elaborate. He still firmly believes that the bodies are protozoa, but that they belong among the rhizopoda and not among the microsporidia where he first placed them. Prowazek and others believe that the organisms of this group of diseases, as well as of rabies, scarlet fever, trachoma, and a few others, are all minute coccus- ETIOLOGY OF VARIOLA AND COW-POX 555 like forms which have the power of producing an envelope from the host cell substance, such envelope with its contained organism constituting the specific body which others have called a protozoon. Prowazek calls the group Chla- mydozoa and says they probably stand between the bacteria and the protozoa in systematic classification. From our studies on this whole group of diseases we have come to the conclusion that there is no close relationship between the trachoma bodies and the intracellular bodies of rabies, smallpox and scarlet fever (see pp. 415 and 561). Fig. 195. — Epithelial cells of a rabbit's cornea, containing many "vaccine bodies." Tissue fixed three days after inoculation with smallpox virus, a and 3, vaccine bodies; h and c. nuclei. X 1500 diameters. In our own work on sections, which has extended irregularly over a period of several years, we have gotten results which are somewha,t confusing, princi- pally because of the non-uniformity of the appearances of these bodies, both by different methods of demonstration and by the same methods at different times. There is no doubt that, whatever the nature of the bodies, they are easily affected by methods used for fixing, hardening, and staining them. This accounts in part for the varied results reported. However, in the most perfectly prepared specimens, judged according to the appearance of the red blood cells, leuko- cytes, and tissue cells at a distance from the lesions, we have found that the vaccine bodies, especially in corneal infection, show a more or less constant series of changes, somewhat similar to those described by Calkins in his "gemmule formation" and by Tyzzer in his development of the vaccine bodies. Our best results on corneas have. been obtained with the following technic: Fix in Zenker's fluid for from four to eight hours; wash in running water over- night; place in 95 per cent, alcohol (changing in two hours to fresh) for twenty- four hours, then in absolute alcohol for twenty-four hours. Imbed in paraffin. The cuts should be from 3m to 5/i thick. Stain with (1) eosin and methylene blue (Mallory) — eosin half an hour, methylene blue two minutes; (2) Heiden- hain's iron hematoxylin; (3) Borrel modified by Calkins. The vaccine bodies may be studied for a short time in the living cornea by rapidly excising an inoculated cornea, spreading it on a shallow agar plate and dropping a thin cover-glass over it. The struc- tured bodies are very clearly differentiated from the rest of the cell contents, and interesting changes have been observed in them. Too little work has been done, however, by this method, to draw any further conclusions in regard to their nature. Councilman and Tyzzer photo- 556 SMALLPOX AND ALLIED DISEASES graphed these living cornea bodies with the ultraviolet light, and the structure came out as the chromatin structures of known living cells. Pathogenesis. — ^For Lower Animals. — ^Various animals seem to contract the disease, or a modification of it, in nature. Horse-pox, sheep-pox, and cow-pox, all show similar pathological changes. Experimentally, probably aU mammals are susceptible, though in varying degrees. Most of them are more sensitive to vaccinia than to variola. ' The epidermis of rabbits, for instance, shows a beautifully typical eruption after inoculation with vaccine virus, while material from smallpox eruptions produces only diffused redness. The corneal "take," however, in both instances, is similar in intensity. Monkeys are equally susceptible to both forms of the disease. For Man. — ^Without vaccination hiunan beings seem to be equally suscep- tible to infection with variola, whatever their race or their condition in life or in whatever part of the world they live. Immunity. — The immunity caused by successful vaccination is not permanent, and varies in its duration in different individuals. Although it usually gives protection for several years and may give it for ten or fifteen years, it is not well to count on immunity for more than one year, and whenever one is liable to exposure it is well to be vaccinated. If this vaccination were unnecessary it will not be successful, while if it is successful we have reason to believe the individual was open at least to a mild smallpox infection. Protective Substances Present in the Serum of Animals after Suc- cessful Vaccination. — ^It has been frequently shown that the blood serum of a calf some days after an extensive vaccination possesses feeble protective properties, so that the injection of one or two liters of it into a susceptible calf would prevent a successful vaccination. A further and more convincing fact has been demonstrated by Huddle- ston and others, namely, that when active vaccine is mixed in certain proportions with serum from an animal which had just recovered from a successful vaccination, and the mixture is inoculated into a susceptible animal, there is no reaction. The Preparation of Vaccine.- The following is the method employed at the New York City Health Department. Seed Virus. — ^This may be prepared by one of the following methods: 1. Glycerinized bovine virus which has been ripened for two or three months. Such virus becomes attenuated after a varying number of calf-passages. 2. Rabbit Virus. — Rabbits vaccinated with stock bovine virus. This may also result in deterioration. 3. Human Virus. — ^This consists of the serum collected on sterilized bone slips from the vaccine vesicles of previously unvaccinated children. This produces good vaccine on calves; it is manifestly difficult to secure. 4. Glycerinized vaccine from calves which have been vaccinated with an emulsion of human vaccine crust. This also gives good virus, but it necessitates the constant collection of human crusts. 5. Human-Calf-Rabbit Seed. — This has been found to be the most economic, efficient, and reliable seed yet found by us. It is produced THE PREPARATION OF VACCINE 557 as follows : Crusts are collected from healthy children about nineteen days after successful vaccination. These crusts are cut up and emulsified with boiled water to a mucilaginous paste. This humanized seed is inoculated into an area about 6 inches square upon the abdomen of a calf, the remainder of the calf being vaccinated in the ordinary way. The pulp from this special area is separately collected and glycerin- ized in the usual way. It is then tested bacteriologically and clinically. This bovine virus from human seed is now used in a dilution of 1 part to 12 J parts of normal salt solution to vaccinate rabbits. The seed is rubbed sufficiently upon the freshly shaven skin of the back. Five days after vaccination the pulp is removed with a curette, weighed and emulsified in a mortar with the following solution: glycerin 50 per cent., sterile water 49.5 per cent., and carbolic acid 0.5 per cent., in the proportion of 1 part of pulp to 8 parts of the solution. Four rabbits should yield from 15 to 20 c.c. of this emulsion, an amount sufficient to vaccinate one calf. The regular supply of vaccine is pro- duced by vaccinating calves with this rabbit seed in the manner to be described. Animals. — ^The preferable animals are female calves, from two to four months of age, in good condition and free from any skin disease. These can easily be vaccinated on the posterior abdomen and insides of the thighs by placing them on an appropriate table. It is possible that, on account of the character of the available supply, older animals may be desirable, but the calves take more typically and are more easily handled. When an animal is too old to be thrown and held with- out difficulty it may be vaccinated on the rump each side of the spine; but the skin there is tougher than on the posterior abdomen and insides of the thighs, and the resulting virus, though efficient, is not so easily emulsified. Vaccination. — The hair should be clipped from the entire body when the animal is first brought into the stable and the calf should be cleaned thoroughly, including the feet and the tail. Just before vaccination the posterior abdomen and insides of thT thighs are shaved and the skin beneath washed in succession with soap and water, sterilized water and alcohol, and then dried with a sterile towel. On this area there are now made superficial linear incisions with a sharp knife, about a fourth of an inch apart. After they have been made they should be dried with a sterile towel or with sterile cotton and rubbed with the charged slips. One or two slips, depending on the amount of virus each slip holds, should be sufficient for vaccinating each incision. Collection. — On the fifth or sixth day, depending upon the rate of development of the vaccine vesicles, they should be ready for collec- tion. The entire shaved area is washed with sterile water and sterile cotton, and the crusts are picked off. The soft, pulpy mass remaining is then curetted off with an ordinary steel curette and the pulp placed in a sterilized vessel. The pulp should be mixed with four times its weight of glycerin and water (50 per cent, glycerin, 49 per cent, water, 1 per cent, carbolic acid). This is done by placing the pulp in 558 SMALLPOX AND ALLIED DISEASES a morter and gradually adding the fluid. The more watery the pulp, especially if it is not to be used immediately, the smaller should be the proportion of glycerin. The emulsion so produced can then be put up for issue in vials. Capillary tubes require special means of filling, and small vials filled and corked answer the purpose admirably. Care of the Calves. — The calves are inspected by a Department veterinarian at the time of purchase and during the period of detention previous to vaccination. After the vaccine has been collected the calves are immediately killed and their organs examined by the veterinarian. If, at autopsy, an animal be found tuberculous or otherwise diseased, the vaccine is discarded. The vaccine stable at Otisville, N.'Y., is a new building with screened windows and concrete floors and stalls, which are kept flushed with water to wash away the dejecta of the animals. The calves stand upon raised racks of galvanized iron. They are fed on milk, no hay or straw being used for any purpose. The calves are vaccinated and the vaccine pulp collected under careful aseptic precautions in a separate operating room, which has tiled walls and a concrete floor. The vaccine pulp when collected is placed in sterile glass containers, sealed, packed in ice and shipped at once by express to the vaccine laboratory. Preparation and Testing of Finished Product. — ^The mixture of vaccine pulp and diluent is pressed through a 40-mesh sieve with the pestle several times until there is no residue on the sieve. Then it is passed about twice through a 100-mesh sieve. Laboratory Tests for Purity. — ^The following tests for purity are now made: 1. Plating upon agar and counting colonies of organism. This is done weekly for five weeks. After one to three weeks the plates usually show no growth, the carbolic acid and glycerin having killed off the extraneous organisms. 2. Inoculation of glucose broth in fermentation tubes for evidence of gas-forming organisms. 3. Tests for Tetanus. — ^Two tests are carried out at the same time, using two different anaerobic methods. After the cultures have incu- bated six days, they are filtered and guinea-pigs are inoculated with the filtrate, and are watched for ten days for evidences of tetanus. 4. Test for Streptococcu's. — A guinea-pig is inoculated subcutaneously with the freshly prepared vaccine and watched for ten days for evidences of streptococcus infection. Clinical Tests for Potency.- — ^After all the laboratory tests for purity have been made and found satisfactory, and not until then, the vaccine is ready for use upon human beings, provided it be found to be efficient. To determine this point, fifteen inoculations of the vaccine are made upon previously unvaccinated children. These must all show a perfect take in order to pass the vaccine as up to standard. A clinical test of such vaccine is made every two weeks thereafter so long as the KEEPING OF VACCINE 559 vaccine is on the market. If one of these tests fails before the end of the period of guarantee, the vaccine is called in. Keeping of Vaccine. — Bulk vaccine is kept in cold storage at a tempera- ture of 4° to 8° F. below zero. Vaccine which has been put up in packages ready for issue is kept in an ice-box at a temperature of 33° to 40° F. REFERENCES. Councilman and Co-wokkbbs: Jour. Med. Research, 1904, xii, 1; Osier's Modern Medicine, Philadelphia, 1907, ii. Ewjng: Jour. Med. Research, 1904, xii, 509. Halbekstaedter and Prowazek: Zur Aetiologie des Trachoms, Deutsch. med. Wchnschr., 1907, xxxiii, 1285. Steinhardt, Israeli, and Lambert: Jour. Inf. Dis., 1913, xiii, 294; 1914, xiv, 87. Williams and Flournoy : Studies from the Rockefeller Institute for Medical Research, 1905, iii. CHAPTER XLV. RABIES. YELLOW FEVER. RABIES. Introduction. — Rabies (synonyms: Hydrophobia, Lyssa, Hundswuth, Rage) is an acute infectious disease of mammals, caused by a specific virus, and communicated to susceptible animals by the saliva of an infected animal coming in contact with a broken surface, usually through a bite. The name rabies (Latin) is given to the disease because of its most frequent and characteristic symptom — furor or madness. Hydrophobia (Greek, fear of water) is another name commonly used, which is also given because of a frequent symptom of the disease, the apparent fear of water. Lyssa is a Greek word meaning hydrophobia. Within the gray nervous tissue of rabid animals are peculiar protozoon- like structures known as "Negri bodies" which are diagnostic of rabies. The nature of these bodies is still a question of dispute (see below). Historical Note. — Rabies is probably one of the oldest diseases in exist- ence, but because of the occurrence of so few human cases, and because the disease develops so long after the bite, its source was for a long time not known nor was it recognized as a separate disease. Hippocrates does not mention it in his writing, but Aristotle about fifty years later (about 300 b. c.) speaks of its being purely an animal disease and being carried by the bite of one animal to another. Celsus, in the first century, was the first to give in writing a detailed description of human rabies. He speaks of it being produced by the bite of rabid animals and states that the wound must be thoroughly bathed and then burned with a hot iron in order to prevent the development of the disease, for after sjrmptoms appear death always follows. As Celsus was not a physician he probably obtained his knowledge from writings which have since been lost. Other writers soon after gave very true descriptions of the symptoms and handling of the disease. Many hundred years passed after this without adding anything to our knowl- edge of the disease, though authors on the subject were numerous. Van Sweiten in 1770 observed the paralytic form of rabies in human beings. At this time several authors, among them Morgagni and Zwinger, believed that the bite of a dog which was not suffering from rabies might produce the disease in man. In 1802 Bosquillon brought forth the original idea that belief in the existence of infectious material in rabies was a chimera and that hydrophobia was simply due to fright. This false idea. had adherents for a long time; even now, by a few people, it is thought to be a true one. Among the host of good observers who studied the disease during the latter part of the nineteenth century, Pasteur stands out as the discoverer, in 1880, of the fact that the disease may be prevented by inoculating gradually increas- ing doses of the virus into the person or animal bitten. This treatment with some modifications, the details of which wiU be given later, is stiU used, though many efforts have been made to develop an efficient serum treatment. Pasteur, as well as munerous other investigators, tried to discover the specific cause of rabies, but all of the results were negative. The importance of making a quick RABIES 561 diagnosis had become so evident that the efforts of many workers were directed toward this end alone. Pasteur and his immediate followers relied for their diagnosis entirely upon rabbit inoculations, and this meant a fifteen to twenty days' wait before the patient knew whether or not the treatment he was receiving wa§ necessary. In 1898 this time was shortened to about nine days in our laboratory by Wilson, who found that guinea-pigs came down with the disease much more quickly than rabbits. From time to time it has been thought that certain histological findings were diagnostic; for instance, the "rabic tubercles" of Babes, and the areas of "round- and oval-celled accumulation in the cerebrospinal and sympathetic ganglia" of Van Gehuchten and Nelis were said to be specific, but further study has shown that they are not absolutely specific for rabies. In many cases the whole picture of the grosser histological changes is sufficiently characteristic to warrant the diagnosis of rabies, but often it is not so. It was not imtil Negri, in 1903, described certain bodies (Negri bodies) seen by him in large nerve cells in sections of the central nervous system, that anything was found which seemed absolutely specific for hydrophobia. Negri claims that these bodies are not only specific for rabies, but that they are probably animal parasites and the cause of the disease. We independently found the same bodies. This work, especially so far as the diagnostic value of these bodies is con- cerned, has been corroborated by investigators in almost all parts of the scientific world, among them workers in our own laboratory who not only determined their worth in diagnosis, but investigated their nature. In our work emphasis was placed upon the fact that the demonstration of the "Negri bodies" by our "smear method" (see p. 562) wonderfully simplified the process of diagnosis. As a result of our studies we concluded that the Negri bodies are not only specific for rabies, but that they are living organisms, belonging to the protozoa, and are the cause of the disease; giving as our reasons the following facts: (a) They have a definite characteristic morphology; (6) this morphology is constantly cyclic, that is, a definite series of forms indicating growth and multiplication can be demonstrated; (c) structure and staining qualities, as shown especially by the smear method of examination, resemble those of certain protozoa. Since this report was published many more cases of rabies have been more or less studied by us and our former conclusions have been more firmly established. Indeed, the evidence as to animal nature of these cell inclusions seemed so convincing that Williams, in 1906, gave them the name Neuroryctes hydrophobiae} Calkins has since studied these bodies and agrees with Williams as to their nature. He called attention to the similarity between their structure and that of the rhizopoda. A number of observers, however, still believe that the Negri body as a whole is principally the result of cell degeneration and that .the • specific organism may be contained within it. Prowazek includes rabies with his " chlamydozoan diseases" (see p. 555). To anyone who has made a long and minute study of the two diseases, however, there can be no question in regard to the essential difference between the "trachoma bodies" and the "Negri bodies." Material and Methods for Study. — ^In New York one may stiU frequently obtain fresh brains of rabid animals, from veterinary hospitals or from the laboratories handling this material. Two methods have been used in helping to study the principal site of infection. (1) ijiimal inocidations. (2) Sections and smears. > Proceedings of the New York Pathological Society, 1906, vi, 77. 36 562 RABIES— YELLOW FEVER The first method is used as a decisive test in diagnosis when results from the second method are doubtful. The technic of the smear method used at present in the Research Laboratory of the New York City Health Department is as follows: 1. Glass slides and cover-glasses are washed thoroughly with soap and water, then heated in the flame to get rid of oily substances. . . . ■ 2. A small bit of the gray substance of brain chosen for examination is cut out with a small sharp pair of scissors and placed about 1 inch from the end of the shde, so as to leave enough room for a label. The cut in the brain should be made at right angles to its surface and a thin slice taken, avoiding the white matter as much as possible. 3. A cover-sUp placed over the piece of tissue is pressed upon it imtil it is spread out in a moderately thin layer; then the cover-slip is moved slowly and evenly over the shde to the end opposite the label. Only sUght pressure should be used in making the smear, but slightly more should be exerted on the cover- glass toward the label side of the sMe, thus allowing more of the nerve tissue to be carried farther down the smear and producing more well-spread nerve cells. If any thick places are left at the edge of the smear, one or two of them may be spread out toward the side of the shde with the edge of the cover-glass. 4. For diagnosis work such a smear should be made from at least three differ- ent parts of gray matter of the central nervous system: (1) From the cortex in the region of the fissure of Rolando or in the region corresponding to it in lower animals (in the dog, the convolution around the crucial sulcus) ; (2) from Ammon's horn, and (3) from the gray matter of the cerebellum. 5. The smears are partially dried in air and the method of fixation and staining given on p. 84 is applied to them. With this method the Negri bodies stain magenta, their contained granules blue, the nerve cells blue, and the red blood cells yellow (Plate IX, Fig. 1). Other methods we have found useful for staining smears are: (1) Giemsa's (p. 82), by which generally the "bodies" are a blue and the contained granules are azur. The cytoplasm of the nerve cells stains blue also, but with a success- fully made smear the cytoplasm is so spread out that the outUne and structure of most of the "bodies" are seen distinctly within it. The nuclei of the nerve cells are stained red with the azur, the nucleoh a dull blue, the red blood cells a pink yellow, more pink if the decolorization is used (Plate IX, Fig. 2). The "bodies" have an appearance of depth, due to their refractive qualities. (2) eosin-methylene-blue method of MaUory (p. 84). With this method of staining, the cytoplasm of the Negri bodies is a magenta, light in the small bodies, darker in the larger; the central bodies and chromatoid granules are a very dark blue, the nerve-cell cytoplasm a light blue, the nucleus a darker blue, and the blood cells a brilliant eosin pink. With more decolorization in the alcohol the "bodies" are not such a deep magenta, and the difference in color between them and the red blood cells is not so marked. In the technic of the section work (p. 84) the most important point is the time the material is allowed to remain in, Zenker. According to our experience two hours' fixation is not enough, three to eight hours is very good, and with every hour after eight hours the results become less satisfactory. Left in Zenker overnight the tissue is granular and takes the eosin stain more or less deeply, both of which results interfere with the appearance of the tiniest "bodies," especially of the very dehcate, minute forms found by us in sections from fixed virus. The sections may be stained by the eosin-methylene-blue method of MaUory (p. 84). In the sections made in this way we have been able to demonstrate clearly very minute forms, as well as good structures in the larger forms Giemsa's method for sections may also be used. Mann's method, recom- mended by others, has not given us such good results. Harris has published a staining method for both sections and smears, which brings the larger bodies out clearly, but which does not seem to give enough differentiation betweeo the smaller bodies mi the nucleoli of the nerve cells, PLATE IX t»oc .-B9 ^m^ 'SR ^ A. W. WILLIAMS, DEL. Nerve Cells in Spreads from Ammon's Horn. Magnification 1200 diameters. Figs. 1 and. 2 from dog "street rabies" show "Negri bodies" (NB); Fig. 3 from non-rabie eat, and Fig. 4 from dog distemper show indefinite inclusion (I) that might be mistaken for Negri bodies by the inexperienced, Negri bodies are structured, more intensely staining, and more refractive. Figs. 1, 3, and -4 are stained by fuehsin-methylene blue mixture, which stains Negri bodies (NB) red -with blue granules, nucleus (NDC) of nerve cells blue, and red-blood cells yello-w (RB). Fig. 2 is stained by Giemsa's mixture, which stains Negri bodies a robin's egg blue -with red granules, nucleus of nerve cells red, and red-blood cells salmon pink. RABIES 563 Morphology of the Negri Bodies.^ — ^The largest forms measured are about 18/i and the smallest about O.Sjit- They are round, oval, oblong, triangular, or ameboid. The latter are more numerous in the fixed virus of rabbits and guinea-pigs. Their structure is shown especially well in smears. "Whatever the variety or species of animal infected, the bodies present the same general characteristic structure; i. e., a hyaline-like cytoplasm with an entire margin, containing one or more chromatin bodies having a more or less complicated and regular arrangement. Fig. 196. — Negri body showing central chromatin with ring of small granules. X 2000. Their structure varies to a certain extent with their size. In fixed virus, with an occasional exception, only tiny forms are found. These are rounded or sometimes wavy in outline, as if possessing slight ameboid motion, sometimes elongated, extending along the rim of the host-cell nucleus, or along one of the nerve fibrils, as if moving there; with eosin and methylene blue they take a delicate light magenta stain, very similar to that taken by the small serum globules in the bloodvessels. Many of the organisms, however, show a small chromatin granule, situated more or less eccentrically, sometimes on the very rim of the body. In the larger forms the granule is large, in the smaller it cannot always be seen; some of the larger forms show from two to several granules and occasionally there is a body with the definite central body and the small granules about it. Detailed Characteristics of Structure in the Large Forms (Fig. 196).— In smears, as well as in sections, the cytoplasm appears quite homogeneous; there is no evidence of a reticulum or of a granular structure outside of the definite chromatoid granules. The smears, however, have brought out one important point in regard to the cytoplasm more clearly than the sections, and that is that it is more basophilic than acidophiUc in staining qualities. With the Giemsa stain, as we have aheady seen, it takes the methylene-blue stain more than the eosin red, and even with the simple eosin-methylene-blue stain the protoplasm appears as a deep magenta unless much decolorized. In studying the central bodies of these organisms, as they appear in the smears, one of the first things noticeable is that they are not surrounded by a clear space — that there is no sign of a vacuolar appearance in the body as there usually 564 RABIES— YELLOW FEVER is in the sections. We notice next that in the great majority of the organisms the central body stands out clearly, as decidedly different in structure, and slightly so in staining qualities, from the chromatoid granules which surround it. The general type of the structure of the central body is that of many well-known protozoan nuclei; that is, the chromatin is arranged in a more or less granule ring around the periphery of the central body or nucleus, leaving an achromatic or more acid-staining centre in which is situated, generally eccentrically a varying-sized karyosome. There are a number of variations from this principal type, according to stage of development (Plate IX, Figs. 1 and 2). Fragmented particles seem to be leaving the nucleus in certain forms, and in this way presumably the chromatoid granules are produced, thus forming chromidia. The chromatoid granules are most frequently arranged in a more or less complete circle about the nucleus. They are somewhat irregular in outline and size, being occasionally ring-shaped, sometimes elongated, often in twos, due probably to active changes of growth and division. They take generally a more mixed chromatin stain than the chromatin of the nucleus. Evidences of Division. — All stages in transverse division are seen. Many evidences of budding are also seen. The chromatoid granules seem to divide and pass out with part of the cytoplasm as a bud. This budding or unequal division appears to take place very early in the growth of the organism and to continue throughout growth until the parent body forms a mass of small organisms which may then break apart at the same time. The budding accounts for the number of small and large forms in a single cell (Plate IX, Fig. 1). Number. — They vary in number according to the stage of the disease and to the infectivity of the part. Site. — ^They are situated chiefly in the cytoplasm and along the fibers in the branches of the large nerve cells of the central nervous system. In parts of smears which are more broken up the bodies may appear as if lying free, and it is these bodies, if the pressure is not too great in smearing, that show the structure best. In some cases the bodies are distinctly localized in small scattered areas of the central nervous system. We have always found bodies in the spinal cord in abundance, but here they are especially prone to be localized in discrete groups of cells. Manouelian has found them in the ganglion cells of the salivary gland. That the organisms are present in various glands of the body (salivary thyroid, suprarenal capsule, etc.) is shown by the virulence of emul- sions from these organs. Cows' milk (Westbrook, McDaniel) and blood (Marie) have also been shown to be occasionally virulent. Cultivation of the Babies Organism. — ^Many attempts have been made to produce artificial cultures, but no reports have been so far corroborated. Noguchi, the latest claimant, states that he has obtained virulent cultures on the twenty-first transplant. Williams, Gurley, Krauss, Barbara, and Volpino have not been able to get the same results. Diagnosis of Rabies. — In our laboratory, for the past thirteen years, or since we have used the smear method in routine diagnosis, there have been many thousand cases in all examined, including suspected rabies and controls. RABIES 565 In all of our work controlled by careful animal inoculations we have never yet failed to have typical rabies develop in animals inoculated with material showing definitely structured Negri bodies. Negative results after inoculation with such material must be interpreted at present as due to some error in technic, such as regurgitation, or hemor- rhage at the time of inoculation, emulsion improperly made, not enough of the virulent material taken because of localization of the organisms, etc. Possibly individual resistance of the animal inoculated might play a part. We have used principally guinea-pigs, and some of them have shown enough irregularity in regard to the time in which they have come down with the disease to suggest a varied individual susceptibility,' if other factors can be ruled out. On the other hand, material in which we have failed to demon- strate typically structured bodies has produced rabies. All of this material, however, since we have improved our technic, has shown suspicious small forms similar to those found in rabbit-fixed virus. But any decomposing brain may also show in smears, bodies very similar to these tiny forms, therefore it is difficult to rule out rabies in such cases. Of course the animal test will probably always have to be used with brains that are too decomposed to show any formed elements except bacteria, unless a reliable chemical test can be dis- covered. Brains from animals dying of distemper may show small non-structured forms, somewhat like "fixed virus" forms (Plate IX, Figs. 3 and 4). So far we have not had rabies produced by fresh brains showing no Nagri bodies and no suspicious forms, but a few observers have claimed that such material has produced the disease. Therefore, until we can standardize our technic, we must in all such cases use animal inocu- lations. We may, however, be reasonably certain that a case showing such negative material was not a case of rabies. We may summarize our knowledge in regard to the worth of the smear method in diagnosis as follows: 1. Negri bodies demonstrated, diagnosis rabies. 2. Negri bodies and suspicious bodies not demonstrated in fresh brains, not rabies. 3. Negri bodies not demonstrated in decomposing brains, uncertain. 4. Suspicious bodies in fresh brains, probably rabies. The localization of the Negri bodies is an important point in mak- ing diagnoses. We have found well-developed bodies distinctly local- ized in different parts of the brain in several instances. In horses there may be small, widely scattered areas of well-structured forms throughout the cerebellum, while tiny, indefinite forms are scattered through the rest of the brain examined. In human brains well-developed forms are found in the corpus striatum and not in the rest of the brain. In several dogs the localization has also been marked. The Complement-binding Test in Rabies.— This test has been tried by Heller (1907), Friedberger (1907), and Baroni (1908), with negative 566 kABiES— YELLOW PEVER results. Berry (1910), and Olmstead and Wilson, in 1916, in oUf Research Laboratory, went over this work thoroughly and obtained similar negative results. Effect of Chemical and Physical Agents on Babic Virus. — Rabic virus appears to become attenuated under certain conditions of tempera- ture; indeed, if it be subjected for about an hour to 50° C. or for half an hour to 60° C, its activity is completely destroyed. A 5 per cent, solution of carbolic acid, acting for the same period, exerts a similar effect, as do likewise 1 to 1000 solutions of bichloride of mercury, acetic acid, or potassium permanganate. Gumming has shown that formalin is especially deleterious to the virus. The virus also rapidly loses its strength by exposure to air, especially in sunlight; when, however, protected from heat, light, and air it retains its virulence for a long period. The virus is readily filtered through all grades of Berkefeld filters, and from the glands through the coarser Chamberland. Poor and Steinhardt have shown that the filtered gland and brain virus seem to have similar characteristics. Pathogenesis. — Natural Infection. — The disease occurs in nature among the following animals given in order of their frequency: dogs^ cats, wolves, horses, cows, pigs, skunks, deer, and man; in fact, as all warm-blooded animals are more or less susceptible to inoculations, all may presumably contract the disease when an open wound is brought in contact with infectious material of a rabid animal. Rabies occurs in almost all parts of the world. It is most common in Russia, France, Belgium, and Italy; it is not infrequent in Austria and in those parts of Germany bordering on Russia. In this hemisphere it is infrequent in Canada, but in the United States the cases are increas- ing in numbers. In Mexico and South America it occurs occasionally; while in England, North Germany, Switzerland, Holland and Denmark, because of the enforced quarantine laws, and of the wise provision that all dogs shall be muzzled, it is extremely rare. In Australia it is unknown, probably because the law that every dog imported into the island must first undergo a six months' quarantine has always been enforced. Since 1915 it has decreased markedly in New York City due to the fact that the muzzling ordinance has been enforced. In this connection the question as to how long the sputum of a rabid dog may remain virulent after it drops from the animal is an interesting one. A case came under our observation in 1906 which illustrates this point. A child , of six years came down with typical rabies in a neighborhood where there had recently been several cases of canine rabies, but no history of a bite could be obtained. The parents were sure she had not been bitten. Six week's before, however, the child had fallen in the street and cut her cheek severely on a jagged stone. The wound was cauterized and healed without fiuther trouble. A mad dog had been on that street just before this occurred. It is reasonable to sup- pose that the stone had on it some of the sputum from that dog, and so the child was infected. Such a case would not occur very often, but the possibility should be considered. In regard to the question as to whether the bite of apparently healthy animals may give the disease, it may be said that, judging from laboratory experiments, RABIES 567 some animals may have a light attack of the disease and recover spontaneously; though such cases, if they occur, are probably extremely rare. That the bite of an infected animal may give the disease before that animal shows symptoms has been proved. Fifteen days is the longest time reported between a bite and the appearance of symptoms in the dog. Therefore, if an animal is kept under observation three weeks after biting another, without developing symptoms, he may be pronounced free from suspicion. Neither age, sex, nor occupation has any specific effect. The time of the year seems to have little effect, though most cases are said to occur during the summer months. The numbers vary with different years. The certainty with which the disease may be produced after a bite and the rapidity of its development have been found to be governed by three factors: (1) the quantity of the rabic virus introduced; (2) the point of inoculation; (3) the strength of the virus as determined by the kind of animal which affords the cultivation ground for the growth of the organism. It is a matter of common observation that in man slight wounds of the skin of the limbs and of the back or wherever the skin is thick and the nerves few either produce no results, especially when bites are made through clothes, or are followed by the disease after an extremely long period of incubation; while in lacerated wounds of the tip of the fingers where small nerves are numerous or where the muscles and nerve trunks are reached, or in lacerated woutids of the face where there is also an abundance of nerves the period of incubation is usually much shorter and the disease generally more virulent. These facts explain why only about 16 per cent, of human beings bitten by rabid animals and untreated appear to contract hydrophobia. Since the establishment of the Pasteur treatment for the disease, the percentage of developed cases after bites is very much less — a fraction of 1 per cent. Symptoms. — There is always a decided incubation period after the bite which varies within quite wide limits, but in the majority of cases it is from twenty to sixty days. Any period after six months is an exception; the shortest we have on record is fourteen days and the longest authentic period is seven months. A very few apparently authentic cases have been reported as developing in about one year, but reports of any time beyond this must be received with doubt. The symptoms may be divided into three stages: (1) The prodromal or melancholic stage; '(2). the excited or convulsive stage; and (3) the paralytic stage. When the second stage is the most pronoimced the disease is called furious or convulsive rabies; when this stage is very short or practically lacking and paralysis begins early, the disease is called dumb or paralytic rabies. In the dog the principal symptoms of each form may be summarized as fol- lows: (a) Furious rabies: change of behavior, biting (especially at those to whom the animal has been affectionate before), increased aggressiveness, characteristic restlessness, loss of appetite for ordinary food, with desire to eat unusual things, intermittent disturbance of consciousness, paroxysms of fury, peculiar howling bark, rapid emaciation, paralysis, beginning in the hindlimbs, death in great majority of cases in three to six days (exceptionally sUghtly longer) after the beginning of symptoms. (6) Paralytic rabies: short period 568 RABIES— YELLOW FEVER of excitation, paralysis of the lower jaw, hoarse bark, appetite and consciousness disturbed, weakness, with paralysis spreading in great majority of cases, and death four or five days after first symptoms. There may be a number of cases showing transition types between these two forms. In Human Beings. — Furious Rabies. — The first definite' symptoms are difficult and gasping breath with a feehng of oppression and difficulty in swallow- ing, the latter the most characteristic symptom. It is caused by convulsive contraction of the throat muscles. The attacks are brought out when attempt- ing to drink or swallow. The very thought of drinking may bring one on; and though there is no fear of water itself, there is fear of taking it because of the effect it produces. The convulsive attacks finally become more or less general over the whole body; in certain cases some parts are more affected by reflex excitation than others; for instance, there may be slight or no photophobia, while in exceptional cases, more frequently in dogs, the hydrophobia is also absent. Most of the special reflexes are increased. Pupils become irregularly con- tracted and widened until they finally remain fixed. Human beings are seldom dangerous to the people about them. In their convulsions they may bite things placed between their teeth, but not otherwise. At this time there is an increased flow of saUva, and one should avoid the contact with this in opened wounds. It may be so increased- that the patient may try to get rid of it by taking it from the mouth with the hand and throwing it about. As a general thing, however, the patient has full possession of his senses between the convulsive attacks until very late in the disease. The temperature is increased from 38° to 40° C, at first with morning remis- sions. Just before death it may rise as high as 42.8° C. (In lower animals the temperature sinks below normal just before death.) The pulse is generally over 100 and is irregular. This stage lasts from one to four days. Death may occur during a convulsion, but more often there is a paralytic sta^e, which lasts from two to eighteen hours. The convulsions become less frequent and the patient becomes weaker until finally there is a complete paralysis. At the beginning of this stage the patient may be able to drink water better than formerly. Death may occur at any time through paralysis of the heart or respiratory centre. Paralytic Rabies. — This form occurs quite seldom in human beings, more frequently in dogs, but not so often as the convulsive form. It is supposed to occur in humans and dogs after a more severe infection. Instead of periods of convulsions, the various muscles simply tremble and become gradually weaker until complete general paralysis supervenes. Sometimes paralysis develops very quickly and may be general before death from syncope or asphyxia occurs. This form generally lasts longer than ordinary rabies. Between these two typical forms of rabies there are many different types, giving quite varied pictures of the disease. Length of the Disease. — The majority of the cases of furious rabies die on the third or fourth day after the symptoms show themselves. The Umits of the reported cases are one to fifteen days, though there are reports of only one or two cases dying on any day after the ninth to the fifteenth. As the time when the symptoms really begin is difiicult to notice, these statistics are probably only approximately correct. In paralytic rabies the average time in which death occurs is five days. Treatment. — The old treatment of rabies consisted simply in encourag- ing bleeding from the wound, or in first excising the wound and then encouraging bleeding by means of ligatures, warm bathing, cupping- glasses, etc.; the raw surface was then freely cauterized with caustic potash, nitric acid, or the actual cautery. It is doubtful whether the disease ever manifested itself after such heroic treatment if the wounds RABIES 569 were small and the treatment was begun soon after the bite; but when the wounds were numerous or extensive the mortality was still high. As it was often impossible to apply cauterization to the wound rapidly or deeply enough to insure complete destruction of the virus, Pasteur and others were led to study the disease experimentally in animals with the hope of finding some means of immunization or even cure. These investigations finally resulted in the discovery of methods of preventive inoculation applicable to man. Pasteur's Method of Preventive Inoculation. — Pasteur's treatment is based upon the fact that rabic virus may be attenuated or intensified under certain conditions. He first observed that the tissues and fluids taken from rabid animals varied considerably in their virulence. Then he showed that the virus may be intensified by successive passage through certain animals (rabbits, guinea-pigs, cats) and weakened in passing through others (monkeys). If successive inoculations be made into rabbits with virus, either from the dog or the monkey, the virulence may be so exalted beyond that of the virus taken from a street dog, in which the incubation period is from twelve to fourteen days, that at the end of the fiftieth passage the incubation period may be reduced to about six or seven days when it remains fixed. This "fixed mms" was used by Pasteur and those after him in his preventive treatment because the dose could be more definitely regulated by subsequent attenuation or dilution. Original Method. — ^A series of spinal cords from rabbits dead from "fixed virus" infection are cut into segments and suspended in sterile glass flasks plugged with cotton stoppers and containing a quantity of some hygroscopic material, such as caustic potash; these are kept at a temperature of about 22° C. The cord when taken out at the end of the first twenty-four hours is found to be almost as active as the fresh untreated cord; that removed at the end of forty-eight hours is slightly less active than that removed twenty-four hours previoxisly; and the diminution in virulence, though gradual, progresses regu- larly and surely until, at the end of the eighth day the virus is inactive. Pasteur began his treatment with an emulsion of the cord kept until the four- teenth day. A certain quantity of this was injected into the animal that had been bitten ; this was followed by an injection of an emulsion of a twelve-day cord ; and so on until the animal had been injected with a perfectly fresh and there- fore extremely active cord, corresponding to the fixed virus. Animals treated in this way were found by Pasteur to be absolutely protected, even against subdiffal inoculation with considerable quantities of the most virulent virus, and thus Pastem-'s protective inoculation against rabies became an accom- plished fact. As it would be undesirable to inject any but persons who had actually been bitten by a rabid, or presumably rabid animal, PasteiK continued his experiments in order to see whether it would not be possible to cure a patient already bitten. He carried on, therefore, a series of experiments which led to the discovery that if the process of inoculation be begun within five days of the bite in animals in which the incubation period was at least fourteen days, almost every animal bitten can be saved; and that even if the treatment be commenced at a longer interval after the bite a certain proportion of recoveries can be obtained. Thus the application of this method of treatment to the human subject was not tried imtil it had been proved in animals that such protection could be obtained and that such protection would last for at least one year and probably longer. The chance of success in the human subject appears to be even greater than 570 RABIES— YELLOW FEVER in the dog or rabbit. Man's period of incubation is comparatively prolonged. Thus there is an opportunity of obtaining immunity by beginning the process of vaccination soon after the bite has been inflicted, the protection being com- plete before the incubation period has passed. Present Administration of Pasteur's Treatment in Human Beings. — The original method of Pasteur in its entirety was soon adopted in many lands, and his results were corroborated. Before long, however, a niunber of modifications were suggested by different observers, some slight, others more fundamental. Some have been widely used, such as Hogyes' dilution method; others have had a limited application in lower animals and are probably only of theoretic interest as regards man. Such are the intravenous inoculation of brain emulsions from street rabies into herbivora (Nocard and Roux, Protopopoff), and the intraperitoneal inoculations of large doses of fully virulent fixed virus into dogs, cats, or rabbits (Hellman, Heim, Remlinger). Immunity has been produced also in rats by allowing them to feed on rabid brains (Fermi, Repetto, Remlinger). Hogyes in Budapest was one of the first to use a different procedure. He claimed that the virus by Pasteur's method was attenuated only through the death of some of the specific organisms, that is, that there were simply fewer living organisms in the early doses given than in the later, and that therefore the same result might be obtained perhaps with even more accurate dosage by giving gradually decreasing dilutions of a fresh virulent cord. By diluting sufficiently he obtained a mixture which when inoculated did not produce rabies in the test animals, a result simi- lar to that following an 8- to 10-day dried cord. This dilution he used for the first inoculation and gradually stronger dilutions for the succeed- ing ones. Other methods of attenuating or diluting fixed virus have been used, such as exposure to the action of heat, cold, gastric juice, glycerin, or carbolic acid. The mixed treatment with specific serum and vaccine has also been employed, chiefly by Marie, by Remlinger, and by Babes. Methods of Attenuation by Gradual Drying. — Pasteur's classic method has imdergone modifications in three general directions: (1) lengthening or shortening the period of treatment; (2) starting the inoculations with a less attenuated cord; (3) increasing or decreasing the amount given at each injection. The method of drying the cords, however, has remained essentially the same as that used by Pasteur. The cord is removed by a modification of the method of Oshida in the following manner: Strict asepsis is preserved. The rabbit when com- pletely paralyzed (seventh day) is killed by gas or chloroform and is dropped into a 5 per cent, solution of carbolic acid for 5 minutes. It is then removed, the excess of carbolic solution is drained off, and an incision through the skin at the upper and inner part of the. thigh is. made. The skin is loosened by cutting around the lower portion of the trunk. It is then pulled by the hands toward the upper extremity of the animal and over the head to the ears, leaving the back exposed RABIES 571 and sterile throughout the entire length of the spine. The spine is then divided transversely near each extremity by bone-cutting forceps. The muscles are cut through about these areas so the spine may be more easily reached. With a long wire probe swabbed with cotton at one end the cord is pushed upward from its canal, freed from its nerves and membranes. The spine is steadied by lion-jawed forceps. The cord curls in a spiral as it emerges and rests on the sterile muscles of the neck. It is lifted with forceps, placed in a Petri dish and cut in two. A small piece is cut from one end and is dropped into a tube of broth to test its purity. A ligature with one long end is placed about each piece, both of which are then hung in a drying bottle (Fig. 197) . Fig. 197.- -One corner of constant temperature room showing drying bottle containing fixed virus cords being prepared for vaccine. Drying the Cord. — ^The drying bottles are sterile aspiration bottles with both openings plugged with cotton. A layer 1 inch high of sticks of caustic potash covers the bottom, and the pieces of cord are suspended from the top cotton plug by their attached ligatures. The bottles are then labeled and placed in the constant temperature room (Fig. 197) ot incubator, which is kept at a temperature of about 22° C. (70° F.). After twenty-four hours' drying the cord is known as one-day cord; after two days, two-day cord, etc. Pieces of cord cut off at any time and put into glycerin will retain about the same strength for several weeks. This procedure is followed in regions where there are few cases of rabies, and the daily killing of rabbits to keep up the vaccine would be a large expense. It may also be followed where treatment is sent by mail. The New York City Health Department used Pasteur's first schemata, with modifications, up to January, 1906, when they began treatment with a ten- and nine-day cord and finished with a one-day. They con- tinued with this until August, 1913. Since then they have been using the 572 RABIES— YELLOW FEVER more intensive method of the Hygienic Laboratory at Washington. From 1906 to 1913 inclusive they "treated 4282 cases infected by rabid animals, with a total mortality of 0.54 per cent, and a corrected mor- tality of 0.19 per cent. They have had 7 cases of definite paralysis with 2 deaths; 6850 cases in all, including those not bitten by rabid animals, were treated. Since it had been found that fresh rabbit-fixed virus inoculated sub- cutaneously into man is apparently harmless, the Berlin Institute, with the hope of obtaining an earlier immunization and a shorter treatment, began to give still earlier cords. In 1901 it began with the eight-day cord on the first inoculation, and was inoculating a two-day cord on the eighth day of treatment. Its* treatment lasted twenty-one days. This method was adopted at the Hygienic Laboratory in Washington in 1908, with slight variations for the different degrees of bites. Now only the intensive schema is used for all cases as follows: Days. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 Age of cord. ■ 8 7 6 4 3 5 4 3 3 2 2 1 5 4 4 3 3 2 2 4 3 2 3 2 1 Amount injected Adults. 2.5 2.5 2.6 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.6 2.5 2.5 2.5 2.6 2.5 2.5 2.5 2.5 5 to 10 years 2.5 2.5 2.5 2.5 2.5 2 2.5 1.5 2.5 2.6 2.5 2.5 2.6 2,6 2.6 2.5 2.5 2.5 2.5 2.5 2.5 1 to 5 years 2.5 2 2.5 2 2 1.5 2 1 2.5 2.5 2.5 2 2 2 2 2.5 2.5 2 2 2.6 2 Each dose contains 1 cm. of the indicated cord. The New York City Health Department has been using this schema since August, 1913, for severe head and face bites, but it reduces all doses by one-fifth, i. e., | cm. of the indicated cord is emulsified in 3 c.c. of normal salt solution, and 2| c.c. of this emulsion is inoculated. It also substitutes a two-day cord for the one-day cord of the eighth and twenty-first day in other than severe cases. In cases with very slight wounds which have begun treatment immediately the inoculations are carried only as far as the fifteenth day. The inoculations are made subcutaneously usually over the abdomen. Treatment by Mail.— The New York City Health Department was the first to send out treatment by mail to physicians for their own patients. Full directions are sent in the mailing case. One-fourth per cent, of carbolic acid is added to the emulsions prepared as above for the first three days' treatment; 20 per cent, glycerin is added to all other emulsions. The carbolic acid and the glycerin are added as pre- servatives, and are therefore omitted when the vaccine is administered to patients at the laboratory. The results from the treatment sent in this way seem to be equally as good as those from the treatment administered at the laboratory. More Intensive Treatment.— In Berlin, where intensive treatment has been longest used, they began to employ even fresher cords for begin- ning doses because they continued to have late deaths, though not quite BABIES 573 so often, after the more intensive methods they were using. Since 1910 the Pasteur Institute in Berlin has been using the following schema: Days 1 2 3 4 5 6 7 8 9 10 1112 13 14 15 16 17 18 19 2021 Age of cords 3 2 1 1 3 2 1 1 3 2 1 1 3 2 1 1 3 2 1 1 1 The dose is 2 c.c. of cord emulsion (1 part of cord in 5 parts of sterile physiological salt solution) inoculated once a day into the subcutaneous tissue of the abdomen. Children and adults receive the same dose. Simon gives the following statistics of the results of Berlin's increas- ingly intensive methods: Berlin Statistics. Period. Age of cord used for beginning inoculation. Cases. Paralysis. Mortality. Per cent. I. II. III. 1898-1906 1906-1909 1909-1910 Chiefly 8-day cord . 4-day cord. Sometimes 3-day cord 3-day cord for all cases . 2896 1490 819 b 2 3 21 7 5 0.7 0.47 0.6 Several others institutes are employing very intensive treatments, but their cases are still too few for consideration. Other directors still use the older methods. Rapid Drying of Babies Virus. — ^Recently Harris, of St. Louis, has published a new method of drying rabies virus and of regulating the dosage. Technic. — ^The brain and cord are removed aseptically and ground up in a sterile mortar with a sufficient quantity of CO2 snow thoroughly to freeze the tissue. The frozen nerve tissue and snow are then placed in a Scheibler jar over H2SO4, the jar being kept in a Frigo apparatus. A vacuum of from 5 to 2 mm. is produced in the jar, which is then kept at the temperature of 18° C. by an ice and salt mixture for a sufficient length of time to dry thoroughly the nerve substance, which then appears as a dry powder. About two days are required for one brain and cord, which lose about one-half of their virulence in the process. The powder is then sealed in tubes in vacuo and kept at a tem- perature below 0° C. until required for use. It has been found that by keeping the powder thoroughly dry and cold practically no further loss of virulence occurs for at least six months. Before storing the virus for use its strength in units is computed, the unit being the minimal infecting dose (M. I. D.) for a rabbit when injected intracerebrally. The advantages claimed for this method are: (1) the ease and econ- omy with which a large amount of virus can be prepared, it being neces- sary to prepare the virus for use even in large laboratories only at inter- vals of several months; (2) the possibility of more accurate dosage for the patients; (3) a shortened period of treatment; and (4) the inoculation of more virus units. The required amount of powdered virus is weighed out each morning, and the necessary dilutions in salt solution for the various patients are made from this. 574 RABIES— YELLOW FEVER Up to October 13, 1913, Harris had treated 240 cases exposed t( infection from dogs, in which the diagnosis of rabies was either provec by laboratory methods or strongly probable from veterinary diagnosis Of this number one patient, who had started treatment six days aftei the bite, died of rabies during the period of observation. No cases oi paralysis have been reported. We must wait for further statistics before being able to judge of the efficiency of this method. Fixed Virus Modified by Dialysis. — Gumming, of Ann Harbor, has devised a method of antirabic vaccination, by which he uses fixed virus which has been rendered avirulent by dialysis. The emulsion of fixed virus is placed in collodion sacs (prepared by the Novy method and sterilized in the autoclave at 105° C. for twenty minutes) and dialyzed in distilled water for from twelve to twenty-four hours. The resulting vaccine does not produce rabies on intracranial inoculations, but does produce immunity on subcutaneous inoculations. Experiments by Gumming on rabbits show that whereas the original Pasteur method protects against only twice the minimum lethal dose (minute directions for obtaining the M. L. D. are given) injected intracerebraUy, and the Hogyes method against one and one-half times the fatal dose, the dialysis method protects against at least three times the fatal dose. He also claims that immunity is produced at an earlier date than by the other methods. Treatment (2 c.c. of the vaccine) is given daily for from fifteen to twenty-five days. Gumming reports over 800 cases (62 per cent, bitten by animals proved to have been rabid) treated without a death and without complications. We must wait for further practical testing of this interesting method before deciding as to its comparative worth. Marie's Method. — For several years past the use of virus serum mixture has been in vogue at the Pasteur Institute in Paris, the technic of which is as follows: 1 gm. of the medulla of a rabbit dead of fixed virus is finely emulsified with 9 c.c. of 0.8 per cent, salt solution and filtered through linen. Two c.c. of this emulsion and 4 c.c. of antirabic serum (obtained from sheep, and inactivated at 56° C. for thirty minutes) are carefully mixed after standing for a time. Six c.c. of this mixture, which contains an excess of virus is injected into the patient. These injections are repeated on the next three days, after which the treatment proceeds according to the regular Pasteur schema, begin- ning with the use of a six-day cord on the fifth day. The antirabic serum is obtained from sheep which have been subjected to a long and strong course of treatment with fixed virus. It is claimed that a quicker immunity is produced by the serum-virus mixture than by the original Pasteur scheme, an advantage of especial value in the treatment of cases liable to become infected with a short incubation, such as bites on the head. Antirabic Serum.— The possibility that the serum of animals im- munized against rabies contains protective substances was suggested by Pasteur as early as 1889. The following year Babes recommended the use of the serum of vaccinated animals in combination with the Pasteur treatment. Since then the study of the amount and character of the antibody content of animals immunized against rabies has been carried on more or less extensively both from th? theoretic and the practical sides. RABIES 575 It was hoped that a serum could be obtained that would effect a cure for developed rabies just as diphtheria antitoxin does for developed diph- theria. But such a definite applicability of the serum has not developed. It was soon found that, while serum of certain vaccinated animals pos- sessed the property of neutralizing rabies virus in vitro, it had only a slight inhibiting power when inoculated into the living animal, and apparently no action at all by any method of inoculation after the disease had become manifest. Babes still claims, however, that the serum has enough effect in vivo to be used in treatment, and his serum treatment is based upon this claim. He gives as his reason for employ- ing serum at the end of treatment that he wishes to introduce into the patient at the time he most needs it the largest amount of anti- bodies. He also claims that the serum so given will prevent or cure the occasional paralyses which occur during treatment. Those who did not agree with Babes were led to test the practical use of the serum combined with the beginning vaccine inoculations. Remlinger, Marie, and others showed that a serum-virus mixture with a slight excess of virus will protect an animal against infection into the anterior chamber of the eye when inoculated during the three days following the vaccination. Thus Marie showed that immunity is pro- duced more quickly by these unsaturated mixtures of virus and serum than by the virus alone. If a surplus of serum is present the animals are not protected from a later infection. Marie, who has used the serum in his treatment of humans since 1904, prepares it as follows : The brains of two rabbits dying from fixed virus infection are finely rubbed up with physiologic salt solution in the proportion of 20 gm. in 180 c.c. This emulsion is filtered through fine cloth and heated for one-half hour at 37° C. Sheep are used for the inoculation. Each sheep receives intravenously 30 c.c. (3 gm. fixed virus) a week for six to eight weeks. Thirteen days after the last inoculation the first blood is drawn. Then in a period of two weeks, at 4 bleed- ings, 200 c.c. of blood are drawn. After a fourteen-day pause another series of inoculations are given and the animal is ready for another series of bleedings. From each animal yearly about 3 liters of antirabic serum are obtained. A strong serum is one that neutralizes 40 virus units in 1 c.c. A vims unit is 1 c.c. of five times the dilution of fixed virus that will surely kill a rabbit inoculated intracerebrally, e. g., the unit of a fixed virus that will surely kill a rabbit in 1 to 500 dilution is 1 c.c. of a 1 to 100 dilution. The nature of the antibodies in rabies serum has been the subject of many studies. Fermi and a few others claim that the antibodies are not specific. They say that they can obtain a similar serum after the inoculation of normal brain emulsions. Some even use normal brain emulsions in the treatment of their lighter cases. Certain investigators (Kraus, Marie, and others), while not able to corroborate all of these claims, have found that the sera of certain animals which are more or less refractory to rabies possess a small amount of rabicidal strength; e. g., 0,5 c,c, of normal chigken serum 576 - RABIES— YELLOW FEVER mixed with one unit of fixed virus (1 c.c. of 1 to 100 dilution) causes th( latter to become neutral in eighteen hours. All species of animals tried produce the specific antibodies, but no1 to an equal degree. Human beings and monkeys are said to have more antibodies after vaccination than rabbits. Centanni showed that immediately after vaccination the animal h not fully protected, though its serum may contain antirabic qualities, while later the animal is immune, though its serum may not be- able to neutralize the rabies virus. These facts point to a, cellular immunity. Results of Antirabic Treatment. — On the whole the results of pro- tective inoculations against rabies are marked. One has only to compare the statistics of mortality after bites from animals suffering from hydro- phobia with those given after any of the methods of treatment employed to see the benefit. As regards the best method to use, the case is dif- ferent. With many methods tried in many lands on a large number of cases, it would seem that we should be able by this time to determine their comparative worth. But the trouble is that the improvement on the whole is not great and the statistics are not kept uniformly or minutely enough to draw trustworthy comparisons. A slight decrease in mortality has been shown in the -statistics from most of the antirabic institutes of the world. But these figures tell us little about the actual value of the different methods. In order to be able better to judge, the statistics should uni- formly give many more details. Some institutes give such details, others do not. Until some such scheme as the following is carried out by all, we must change cautiously a treatment that has already given good results. 1. Diagnosis of biting animal : (a) Rabies, (b) probably rabies, (c) questionable, (d) not rabies, (e) nothing known. 2. Manner of making diagnosis : (a) By animal inoculation, (b) by microscopic examination, (c) by clinical diagnosis. 3. Site and character of bites (e. g., number, depth, laceration, pro- tected by clothing, etc.) : (a) Head, (b) hands, (c) other parts of body. 4. Time elapsing between bite and beginning of treatment. 5. Method of treatment used. 6. Complications during or after treatment, particularly paralysis. 7. Character and time of death. That the time after the bite makes a great difference is shown by the following table: Time intervening between bite and Number of beginning treatment. cases treated. Death. Percentages. Babes . . 1 to 2 days 3406 3 0.088 3 to 5 days 2541 2 0.077 5 to 6 days 809 1 0.124 Diatroptoff . 1 week 4602 26 0.560 2 weeks 961 16 1.660 3 weeks 313 10 3.190 RABIES 577 Immunity. — ^The immunity in human beings produced by the anti- rabic treatment apparently lasts a variable time. That it may not last more than fourteen months is shown by the history of one of our cases. The patient was an assistant in a hospital for dogs. He was given eighteen days' treatment after a light wound in the hand from a rabid dog. Four- teen months later he came down with typical hydrophobia. Since his treatment he had become very careless with cases of rabies, exposing wounded hands to saliva because he considered himself immune. He was warned that there might be danger. Six weeks before his death he put a wounded hand into the mouth of a rabid animal. There is little doubt but that this is a case of reinfection after loss of protection from the treatment rather than one of delayed hydrophobia. Marie has found complete immunity in dogs eighteen months after treatment. HI Effects of Treatment. — ^Local. — ^There is only slight local dis- comfort, increased a little if the emulsion contains glycerin. During the second week an erythema often appears about the point of inocula- tion, which Stimson regards as a manifestation of hypersusceptibility to foreign nerve tissue. It disappears in a few days. Constitutional. — Ever since the beginning of treatment occasional non-fatal affections of the nervous system have been reported, which occurred during or shortly after the course of treatment. These have varied in degree all the way from a slight neuritis, through paraplegias to paralyses of various parts of the body. Very occasionally the paralyses are marked and the patient dies. Cases of true paralytic rabies which may occur within the period required for the establishment of immunity by the treatment must be differentiated from cases occurring as a result of treatment. Simon published an extensive report of 84 cases occurring during the years 1888 to 1911 inclusive, and Fielder gives a corroborative report in 1916. Simon classifies the cases collected by him according to the diagnosis of the biting animal, with the rnortality in each group as follows: Positive group. *^umber Per cases. cent. Probable group. Number Per cases. cent. Questionable group. Number Per cases. cent. Negative group. Number Per cases. cent. Not known group. Number Per cases. cent 25 29.76 (2) 11 13.0 (4) 21 25 (5) . 17 20.23 (3) 10 11.9 (5) Nineteen deaths occurred, as seen from the figures in parentheses, or 22 per cent, of the 84 cases. In analyzing the effect of different methods of treatment oa paralyses, Simon gives the following summary: Number of Cases of cases treated. paralysis. Proportion. Classic Pasteur method ... . . . 32,676 6 1 in 5446 Modified Pasteur method 8,657 16 1 in 541 Hogyes method 51,417 3 1 in 17139 37 578 RABIES— YELLOW FEVER It is seen that the number of paralyses following the Hogyes methoc are markedly less than those following the other methods. From the studies so far made of these paralyses the possibility o there being different causes for different cases cannot yet be ruled out The chief theories advanced as to factors in producing the conditioi are six: 1. Due to "laboratory rabies" from the fixed virus vaccine inocu lated. 2. Due to "modified rabies" resulting from the treatment on th( street virus infection. 3. Due to a toxin produced by the rabies organisms. 4. Due to infection with extraneous organisms introduced wit! the virus during treatment. 5. Due to psychological disorders. 6. Due to the inoculation of a foreign protein followed by an anaphy^ lactic reaction. The Cauterization of Infected Wounds. — ^We believe that in cases in which the Pasteur treatment cannot be applied great benefit may be derived from the correct use of cauterization with fuming nitric add even twenty-four hours after infection, and that even in cases in which the Pasteur treatment can be given, an early cauterization will be ol great assistance as a routine practice and should be very valuable, as the Pasteur treatment is frequently delayed several days for obvious reasons, and then does not always protect. In the case of small wounds all the treatment probably indicated will be thorough cauterizatior with nitric acid within twelve hours from the time of infection. Oui experience in dealing with those bitten by rabid animals goes to shovi that physicians do not appreciate the value of thorough cauterizatior of the infected wounds. Preventive Measures in Animals. — Far more important than any treatment, curative or preventive, for hydrophobia in man is the preven- tion of rabies in dogs, through which this disease is usually conveyed. Were all dogs under legislative control and the compulsory wearing of muzzles rigidly enforced for two years where rabies prevails, hydro- phobia would practically be stamped out. This fact has been amply demonstrated by the statistics of rabies in countries (e. g., England) where such laws are now in force. New York City has recently had some measure of success in enforcing such laws. YELLOW FEVER. Yellow fever is an acute infectious disease of tropical countries, with no characteristic lesions except jaundice and hemorrhage. Othei lesions that exist are those common to toxemia. One attack usually produces complete immunity. Historical Note. — There have been many extensive studies on the etiologj of this disease with numerous announcements of the discovery of its specifit cause. Not one of the latter, however, has been corroborated. The BaciUui YELLOW FEVER 579 icteroides of Sanarelli (1897), found in the circulating blood and in the tissues of most yellow-fever patients, was thought by many to be the real organism, and for some time it was the subject of most jninute studies with the result that it, tod, has been placed with the rejected organisms. The epoch-making investigations of the United States Army Commission composed of Walter Reed, James Carroll, Aristides Agramonte, and Jesse W. Lazear (1901), established the truth, that this disease, like malaria, is carried from one infected person to another through the agency of a mosquito. Finley, in 1881, was the first positively to assert that the mosquito was the transmitter of the disease. He was, however, unable to prove his theory, and it remained for the commission conclusively to show that a distinct species of mosquito carried the infection. - The work of the American commission was fully corroborated by the French commission and by other workers. The principal facts established by the commission have been summed up by Goldberger as follows : 1. Yellow fever is transmitted, under natural conditions, only by the bite of a mosquito [Aedes calopus) that at least twelve days before has fed on the blood of a person sick with this disease during the first three days of his illness. 2. Yellow fever can be produced in man xmder artificial conditions by the subcutaneous injection of blood taken from the general circulation of a person sick with this disease during the first three days of his illness. 3. Yellow fever is not conveyed by fomites. 4. Bacillv^ icteroides Sanarelli stands in no causative relation to yellow fever. Though the specific parasite remains yet undiscovered, facts have been brought out by these studies which give some idea of its character. 1. It seems to require two hosts (a mammal and an arthropod) for the completion of its life cycle (analogies, Plasmodium malariae, Babesia bigeminum). 2. There is a definite time after the bite of the mosquito before the blood of the person bitten becomes infective (average five days), and a definite time that the blood remains infective (three days). 3. The blood during these three days is still infective after passing through the finest-grained porcelain filters (Chamberland B and F) . 4. The blood loses its virulence quickly (forty-eight hours) when exposed to the air at a temperature of 24° to 30° C. When protected from the air by oil and kept at the same temperature it remained virulent longer (five to eight days). Heated for five minutes at 55° C. it becomes non-virulent. 5. The bite of an infected mosquito does not become infectious until twelve days (at a temperature of 31° C.) after it has bitten the first patient. The cause of the disease still remains undiscovered, notwithstanding much study of human blood and other tissues and of infected mos- quitoes. The infective blood filtrates show nothing with the dark- field illumination except small dancing granules similar to those found in healthy persons. 580 RABIES— YELLOW FEVER Certain facts relating to the disease seem to point to protozoa as the cause; for instance, the necessity for a second host and the long incubation time required before that host becomes infective after biting a yellow-fever patient. The bodies described by Seidelm under the name paraplasma have , been shown by Agramonte to be tissue changes. The higher monkeys seem to be susceptible, though no complete experiments have been made with them. Fig. 198. — The yellow- fever mosquito (Aedes calopus). Adult female. (Howard.) Much enlarged. The Yellow-fever Mosquito (Figs. 198-203). — The name Stegomyia for this small tropical mosquito was suggested by the English entomol- ogist Theobald, who separated this genus from the genus Culex, with which it was formerly classed. It was first given the specific name Fasciata, but Blanchard proved that this had already been used and the name Calopus (Meigen, 1818) was found to be the proper one. Later the genus Stegomyia was shown to be invalid and the organism now goes by the name Aedes calopus Meigen. Salient characteristics of Aedes are: (1) The palpi in the male are as long or nearly as long, as the proboscis; in the female the palpi are uniformly less than one-half YELLOW FEVER 581 Fio. 199. — Tbe yellow-fever mosquito. Adult male. Much enlarged. (Howard.) Fig. 200. — The yellow-fever mosquito. Adult female, side view. Much enlarged. (Howard.) Fig. 201. — The yellow-fever mosquito. Egg. Greatly enlarged. (Howard.) 582 RABIES— YELLOW FEVER as long; (2) the legs are destitute of erect scales; and are alternately banded white and black; (3) the thorax is marked with lines of silvery scales. Aedes calojius is spread over a wide range of territory, embrac- ing many varieties of climate and natural conditions. It has been found as far north as Charleston, S. C, and as far south as Rio de la Plata. There is no reason to believe that it may not be present at some time or other in any of the intermediate countries. In the United States specimens of Aedes calopus have been captured in Georgia, Louisiana, South Carolina, and eastern Texas. The island o£ Cuba is Fig. 202. — The yellow-fever mosquito. Larva. Much enlarged. (Howard.) Fig. 203. — The yellow-fever mosquito. Pupa. Much enlarged. (Howard.) overrun with this insect. The fact that Aedes calopus has been known to exist at various times in Spain and other European countries may account for the spread of yellow fever which has occurred there' once or twice in former times; the same may be said of the country farther north in the United States, where Aedes calopus has not yet been reported, but which have suffered from invasions of yellow fever. The adult mosquito is so small that mosquito netting of 19 or 20 strands or meshes to the inch is required to keep the insect from entering a place. YELLOW FEVER 583 Brackish water is unsuited for the development of Aedes larvae. The species Aedes calopus seems to select any deposit of water which is comparatively clean. The defective drains along the eaves of tile roofs are a favorite breeding place in Havana and its suburbs; indoors they find an excellent medium in the water of cups of tin or china into which the legs of tables are usually thrust to protect the contents from the invasion of ants, a veritable pest in tropical countries. The same may be said of shallow traps, where the water is not frequently disturbed. Like other Culiddce, it prefers to lay at night. It is eminently a town insect, seldom breeding far outside of the city limits. Agramonte never found Aedes calopus resting under bushes, in open fields, or in the woods; this fact explains the well-founded opinion that- yellow fever is a domiciliary infection. The question of hibernation in the larval stage is important. Agra- monte failed to get larvae that could resist freezing temperature, and found that in the case of Aedes calopus this degree of cold was invariably fatal. The possibility of their being capable of life outside their natural element must also be considered from an epidemiological point of view. The dry season in the coimtries where this species seems to abound is never so prolonged as completely to dry up the usual breeding places. Experimentally, adult larvae removed from the water and placed over- night upon moist filter paper could not be revived the following morning. The question of the life period of the female insect is of the greatest importance when we come to consider the apparently long interval which at times has occurred between the stamping out of an epidemic of yellow fever and its new outbreak without introduction of new cases. The fact is that Aedes calopus is a long-lived insect; one individual was kept by Agramonte in a jar through March and April into May, in all for seventy-six days after hatching in the laboratory. These mosquitoes bite principally in the late afternoon, though they may be incited to take blood at any hour of the day. They are abundant from March to September, and even in November Agra- monte was able to capture them at will in his office and laboratory. The mosquito is generally beUeved to be incapable of long flight unless very materially assisted by the wind. At any rate, the close study of the spread of infection of yellow fever shows that the tendency is for it to remain restricted within very limited areas, and that when- ever it has travelled far beyond this, the means afforded (railway- cars, vessels, etc.) have been other than the natural flight of the insect. Experiments have demonstrated that not all mosquitoes which bite a yellow-fever patient become infected, but that of several which bite at the same time some may fail either to get the parasite or to allow its later development in their body. This condition is similar to that seen in Anopheles, with regard to malaria. The question as to the length of time infected mosquitoes remain dangerous to the non-immune community cannot be definitely answered 584 BABIES— YELLOW FEVER. at present; there is good presumptive evidence that the mosquito may- harbor the parasite through the winter and be enabled to transmit in the spring an infection acquired in the fall. There is reason to believe that the mosquito, once infected, can transmit the disease at any time during the remainder of its life. Freezing temperature, however, quickly kills the insect. Carrying out preventive measures based on the knowledge gained by the splendid work of the American Army Commission, yellow fever has been practically wiped out of Cuba, the Isthmian Canal Zone, and other infected areas. REFERENCES. Beekt: The Complement Binding Test in Rabies, Jour. Exp. Med., 1910, xii 338. Fielder: Jour. Am. Med. Assn., 1916, Ixvi. Harris: A Method for the Staining of Negri Bodies, Jour, of Infect. Dis., 1908, v, 566. HOGYES, Lyssa: Nothnagel's Specielle Pathofogie u. Therapie, Wien, 1897. Howard : The Yellow Fever Mosquito, Farmers' Bulletin, 547, United States Depart- ment Agriculture, Washington, D. C. Kerr and Stimson: The Prevalence of Rabies in the United States, Jour. Am. Med. Assn., 1909, liii, 989. Keaus and Baebaka: Deut. med. Wchnschr., 1914, xl, 1507. Manooj^lian: Ann. Inst. Past., 1914, xxviii, 233. Marie: L'Etude expferimentale de la Rage, Paris, 1909. NoGUCHi: Jour. Exp. Med., 1913, xviii, 314. Otto: Gelbfieber, In Kolle and Wassermann's Handbuoh d. path. Mikroorg., 1913 2d ed., Jena. PooK and Steinhakdt: Jour. Inf. Dis., 1913, xiii, 203. Reed and Carroll: Jour. Exp. Med., 1900, v, 215. Reed and Carroll and Agramonte: Jour. Am. Med. Assn., 1901, xxxvi, 413. The Yellow Fever Institute Bvilletin, No. 16, Yellow Fever, Etiology, Symptoms, and Diagnosis, by Goldberger, gives a good review with full literature to 1907. Volpino: Presse M6d., 1914, p. 79. Williams: Rabies, in Forscheimer's Therapeusis, 1916, v, Sec. Ed., New York, with references. WiLLLAMS and Gurley: Coll. Studies, N. Y. City Health Dept., 1914, p. 15. Williams and Louden: Jour, of Infect. Dis., 1906, iii, 460, with fuU list of references to date on Negri bodies. PART III. APPLIED MICROBIOLOGY. CHAPTER XLVI. THE PRACTICAL APPLICATION OF BACTERIAL VACCINES. The practical application of bacterial vaccines for prophylactic purposes developed as a result of the success of vaccination against smallpox and the Pasteur prophylaxis in rabies. The first successful application was that of Pasteur who immunized sheep against anthrax by the injection of attenuated cultures. The protection conferred by the injection of bacterial vaccines is associated with a rise in the anti- body content of the blood, and this increase has been assumed to be the basis of the protection. This increase may be only a part of the mechan- ism of protection, as the demonstrable antibodies disappear long before the immunity is lost. The view, however, that the protection was primarily due to the enhanced antibody content of the blood led to the therapeutic use of vaccines, with the hope that even during an infection they would stimulate further antibody production and thus be an aid in recovery. The rational application of bacterial vaccines as specific prophylactic or therapeutic agents presupposes a correct bacteriological diagnosis, or at least a clinical diagnosis which warrants the deduction that the infec- tion is due to a specific bacterium. It must be kept in mind that many of our bacterial names refer not to a single organism but rather to species in which the members of one subgroup frequently have little, if any, immunological relationship with menibers of other subgroups, or the members of subgroups may be completely heterogeneous as regards such relationship. On the diagnosis and a knowledge of immunological characteristics of the causative organism depends the selection of the vaccine and whether an autogenous or a stock vaccine is applicable. Polyvalent Vaccines. — Many of the so-called polyvalent vaccines are not polyvalent in the sense that they contain representatives based on immunological knowledge. In most instances they are mixtures of many strains under a group name. Vaccines of Mixtures of Bacteria, so-called Mixed Vaccines.— One type is a shotgun preparation advocated where a bacteriological diag- nosis is absent. Another type is a mixture of the organisms usually 586 APPLICATION OF BACTERIAL VACCINES found associated in certain types of inflammation. Unfortunately the types found in such inflammation are usually members of heterogeneous groups so that the applicability of such stock vaccines as specific agents is more apparent than real. Mixed vaccines of specific types of bacilli for prophylactic purposes are discussed later. Types of Vaccines. — Killed Bacteria. — ^This type of vaccine is the most commonly used. Heat is usually employed to kill the bacteria. The minimum temperature and time of exposure necessary to kill, should be used, as overheating may lower or destroy the antigenic value of the vaccine. Disinfectants such as carbolic, tricresol, etc., have been advised, but there is no definite evidence that they are superior to minimum heating. They are, however, often added as preservatives. Bacterial Extracts, Autolysates, Digested Bacteria. — ^The object of such preparations is to hasten absorption or to bring into solution the toxic elements of the bacteria. In approximating these conditions, however, the reactions become severe or even serious and such preparations should be very cautiously employed. Live Bacteria.^Their use is based on the fact that an unnatural portal of entry is associated with a local reaction only. The advantage would be that the unchanged bacteria should stimulate a higher immunity. There is always the possibility, however, that should they accidentally reach the normal portal of entry that disease might follow. The use of vaccines of bacteria attenuated in virulence by various methods is common in veterinary medicine. Sensitized Vaccines. — Living or Killed. — ^The bacteria are treated with immune sera to lessen or avoid the local and general reaction, Besredka and others claiming that this will, not interfere with their immunizing value. Larger and more frequent doses are thus possible and the sensi- tized bacteria are probably more quickly disintegrated and absorbed. The sensitized dead vaccines have been most commonly employed, and those left alive often die before use. The superiority of sensitized over the ordinary killed vaccine has not been demonstrated. A similar procedure, viz., the simultaneous injection of vaccines and immune serum is employed in certain prophylactic procedures, where the bacteriiun employed gives rise to excessive reactions. Preparation of Bacterial Vaccines. — Cultures are preferably grown on agar, although broth may be used. The growth after twenty-four to forty-eight hours, depending on the rapidity of growth of the organism, is washed off in a small amount of saline solution. This suspension should be well shaken to give an even distribution of bacteria and is then standardized before heating or the addition of preservatives. The Wright method is most commonly used for standardization. A capillary pipette (see p. 220) is marked about one inch from the tip. The finger is pricked and blood drawn up to this mark; a bubble of air is then allowed to enter the tube and the bacterial suspension drawn up to the mark. The contents are then mixed on a slide and thin smears (as for blood) made and stained. On the under lens of the eye-piece a one-quarter-inch square is marked PREPARATION OF BACTERIAL VACCINES 587 out with a pencil, and using an oil-immersion objective the number of bacteria and red cells appearing in this square are counted separately. About fifty squares are counted and the average per square obtained. The number of red cells per cubic centimeter being known, the number of bacteria are obtained according to the following proportion: Number of red cells per square: number of bacteria per square : : 5000 (millions): X (millions). Other methods are employed as direct counting either by the Prescott method, in a counting chamber as for blood cells or platelets, centrifuging and deter- mining the volume of bacteria or by determining the weight of the bacteria after evaporation and drying. The suspension is heated to not over 55° C. for one hour unless the organism is not killed, and a preservative tricresol 0.3 per cent., car- bolic acid 0.25 per cent, or lysol 0.25 per cent, added after making the necessary tests for sterility. For the American and English armies typhoid vaccine is only heated to 53° C. and the antiseptic relied on to finish sterilization. The vaccine is diluted for use with saline containing a preservative. The French add no antiseptic. Sensitized vaccines are prepared by adding immune serum to the suspension and after several hours sedimenting the bacteria by means of the centrifuge and washing them free of serum with saline solution and suspending them finally in saline solution. Gay and Claypole employ alcohol-killed sensitized vaccines. Length of Potency of Vaccines. — No definite knowledge is available. Until this is available it would seem advisable to limit the period to four or possibly six months. Therapeutic Application. — Theoretical Considerations. — The use of bac- terial vaccines in the course of an infection was primarily based on the conception that the infection was not a sufficient stimulus to the pro- duction of adequate curative antibodies. The additional stimulus of vaccines therefore would increase the antibody production and hasten cure. This conception is based on the idea of specific action. On this basis, localized lesions of a subacute or chronic character should be the type of case most benefited. Although this is, in general, the fact, the more acute and general types of infection are also benefited by vaccine therapy. It would seem that when the blood stream was invaded by bacteria, that all the tissues of the body would be stimulated to their maximum capacity of response. We must, however, distinguish three types of generalized invasion : first, where there is an initial blood invasion with secondary localization and the disappearance of the infective agent from the blood; second, acute local infections, with limited invasion, without multiplication in the blood (simple bacteremia); third, a septi- cemia, invasion and multiplication in the blood stream. The first two show the ability of the body to limit the blood infection by a response of the protective forces of the body. This, however, may not be the maximum possible response and in these conditions vaccines do increase the antibody response, an example being typhoid fever. Furthermore, the injection of vaccines leads to a cellular response in the form of a polynuclear leukocytosis or an increase if already existent. It is evident, however, that injudicious dosage might overwhelm the 588 APPLICATION OF BACTERIAL VACCINES body and break down the ability to respond, with serious consequences to the patient. In the septicemia types, intravascular multiplication with no tendency to localization usually indicates the inability of the body adequately to respond, and little is to be expected from the further stimulus of vaccines. , This is usually found to be the case. Considering specific response to vaccines, one would not expect much benefit in acute self -limiting types of infection, as the antibody response requires several days to become marked. Theoretically, if the antibodies cannot reach the focus of infection in sufficient concentration, vaccines would be of little help. This is prob- ably a factor in the generally poor results obtained with infections of bone sinuses or cavities. Vaccines are an aid, not a substitute for indicated surgical procedures, and the application of the latter should never be delayed. Incision, drainage and relief of tension, even if pus has not formed, are of curative value due partly to better circulation, also to the fact that drainage allows the exudation of fresh serum and cells, both of the utmost importance. Non-specific Response to Specific and Non-specific Vaccines. — ^The occurrence of prompt curative response following the injection of not only specific but also of non-specific substances cannot be explained on the basis of specific antibody response. Such a response, especially following intravenous injection of non-specific substances would at first glance overturn all our ideas of the specific action of therapeutic vaccines. The recent methods of treating typhoid fever by intravenous inoculation is a good example. The injection of an appropriate amount of typhoid vaccine or even paratyphoid or colon vaccine or albumose, is followed by a chill and a rapid rise of temperature followed by a pro- gressive fall, sweating, and marked subjective and general improvement. At first there is a decrease, then a sharp increase of the polynuclear leukocytes. In some cases the reaction develops into a recovery by crisis, in others the disease resumes its course and a second injection, giving a similar reaction, may or may not be followed by critical recovery. Similar curative reactions have been obtained in other infections. (See Gonococcus.) As far as we know at present, the following factors enter into such curative reactions. The sharp leukocytic response is undoubtedly of value. Increased antibody production cannot occur even with specific vaccines; furthermore, vaccination diwing the incubation period of typhoid fever has little, if any, effect on the subsequent disease, which should be the case if the mere increase of antibodies were the important factor. On the other hand, the best results are obtained after the tenth day of the disease, that is, after response to infection has been fully established and the injection at this time is followed by a rapid increase of the antibodies in the blood. What apparently takes place is a stimu- lation of the hemopoietic organs (the probable source of antibodies) with the release and dispersion of antibodies already formed and the throwing out of polynuclear leukocytes. A further factor as pointed out by Job- NON-SPECIFIC RESPONSE TO VACCINES 589 ling is the mobilization of non-specific ferments, the serum-protease, for instance, would act on the toxic products of the bacilli and reduce them to non-toxic products. The antiferment content of the serum is also influenced and may be a factor. Desensitization or a refractory condition of the cells to typhoid bacillus products has been suggested in explanation. The chill and rise of temperature are important, as without these curative effects are not obtained. How much of the reaction is due to the injected protein alone, or to the dissolution products resulting from the sudden response of the body is not clear. Normal persons react simi- larly but larger doses are required and the reaction is less intense in the nlilder infections. Where there is a focal infection, reactions at the focus are undoubtedly of curative value. This is most evident in gonorrheal arthritis. The ability to respond to non-specific or to specific substances run parallel, so that the data given as to selection of case apply equally. Although non-specific substances are of curative value, the relative value of these as contrasted with specific vaccines, or to what degree the curative response to specific vaccines is of non-specific character, is still to be determined. Theoretically, the added specific response to vaccine of the autogenous type should be of value. Non-specific substances may raise the resistance to specific infection, but for prophylactic pur- poses specific vaccines must be employed. The knowledge we have is no excuse for ignoring specificity nor for the general use of mixed and pseudospecific stock vaccines. It is unfortu- nate that many reports of beneficial results of vaccine therapy are valueless in this connection as the results are based on the assumption of specific action whereas the data given are no guarantee that such was the case and deductions are therefore impossible. Reaction to Vaccines. — ^The reaction may be local, focal or general. As a rule, with appropriate dosage reactions should be slight or, at most, moderate. With intravenous therapy a reaction (see above) is essential. Dosage. — ^The dosage given for prophylactic purposes is based on experience and should be closely adhered to. Children stand vaccines well, and full correction according to weight is unnecessary. For thera- peutic purposes it is usually better to start with small doses and increase these until the maximum is reached, which is usually the reacting dose in the individual. The period between injections may be from twenty-four to seventy-two hours or longer. Continued injections increasing in amount frequently lead to cure even when no beneficial results are seen at first. Control of Dosage. — ^The use of the opsonic index (see p. 217) has been abandoned, the dosage being based on reaction and clinical results. Negative Phase.— Wright thought that there was an initial period of depression following vaccination. Unless the dose is large or recklessly administered, this need not be considered. Prophylactic vaccination can be carried out even though the person is exposed to infection and is 590 APPLICATION OF BACTERIAL VACCINES not injurious even during the period of incubation. At most it may accentuate the onset of the disease. Mode of Injection. — This is usually subcutaneous. The intravenous method should not be undertaken unless one is conversant with the method and has a thorough understanding of its possible dangers. Intravascular agglutination with cerebral embolism, shock due to rapid dissolution of bacterial products and hemorrhage as in typhoid fever are the possible dangers. Several deaths can be directly attributed to this mode of injection. Intramuscular injections may also be used. PRACTICAL APPLICATION OF INDIVIDUAL VACCINES. The following is a brief summary of the modes of administration and the results in various types of infections. Staphylococcus Infections. — ^The best results have been obtained in chronic or recurrent types of infection, such as acne and furunculosis. Stock vaccines may be employed, but if failure is encountered autogenous vaccines should be tried. In the deep indurated types of acne other bacteria are probably important, as B. acne, and vaccines of the bacillus in doses of 2 to 20 million should be tried with the staphylococcus. In furunculosis the vaccine seems to be of more value in preventing new lesions than in the cure of the existing foci. Sycosis and other skin lesions with associated pustular lesions may be benefited by vaccines. Acute local lesions are probably little influenced, and in this and in the other conditions mentioned the usual modes of treatment should be employed. The greatest care should be taken to protect the healthy skin from discharges from the infected focus. Dosage. — ^The more extensive the lesions the smaller should be the initial dose. From 100 to 1000 million is the average increase in dosage, though larger doses may be given. Streptococcus Infections. — ^Local Infection. — ^Acute infections due to Streptococcus pyogenes are usually surgical conditions. In the subacute stage vaccines may be of help. Streptococci are found in many con- ditions, such as common colds, bronchitis, sinus involvement and mouth infection, but vaccines are of doubtful value. Puerperal infections are probably little affected by vaccines. The reported results in erysipelas are as variable as the disease itself. Erdman in an analysis of 800 cases could see no result. General Infections. — ^The data are insufiicient. There is a slight indica- tion that immune serum followed by vaccines is of some value. Dosage. — In severe or general infections an initial dose of 5 to 10 million, in more local lesions larger doses, may be given. The maximum is usually about 500 million. Pneumococcus Infections. — Pneumonia. — Only a moderate number of cases have been treated under conditions of specific relationship of vac- cine to the infecting pneumococcus. (See types under Pneumococcus.) Other favorable reports have been made without regard to this relation- ship. It may be that this disease can be favorably influenced by non- PRACTICAL APPLICATION OF INDIVIDUAL VACCINES 591 specific therapy. At best the data at hand do not indicate any very marked results. As a prophylactic the vaccine has been used. Other Infections. — See common colds, bronchitis, otitis, etc., p. 595. Dosage. — ^Ten to 500 million. Gonococcus Infections.^ — Urethritis is uninfluenced and the number of complications are not appreciably reduced. Vaginitis shows very little improvement, though this may be due, as Pearce points out, to immuno- logical differences in the gonococci from those found in adult infections. Epididymitis and Prostatitis. — Some cases have been benefited. Pelvic infection or general infections are little influenced. Periarthritis and similar conditions have been very successfully treated. Dosage. — From 25 to 500 million is the usual dosage, though up to 1 billion have been given. The dose should be rather rapidly in- creased until some degree of reaction is elicited. Although a reaction of any extent is to be avoided, a mild focal or febrile reaction is not only of value in indicating the limits of dosage but is also of curative value. Intravenous Administration. — Bruck and Sommer have advocated this method, and unusually good results are claimed not only in gonorrheal arthritis but also in epididymitis, prostatitis and even in urethritis. They used a preparation "Arthigon," 1 c.c. of which contains about 80 million gonococci. Miiller and Weiss, and Miller and Lusk have had good results with non-specific substances. Meningococcus Infections. — Prophylaxis. — ^Three injections at weekly intervals. of 250, 500 million and 1 billion respectively give rise to con- siderable antibody production. The data as to the protective value are too hmited to draw any conclusions. Therapeutic Application. — In some cases where lumbar puncture and serum administration, although repeated frequently, has had little effect an autogenous vaccine may be of value. DuBois and Neal .recommend an initial dose of 100 to 250 million; increased to 1 billion, giving the injections every two or three days. Micrococcus Catarrhalis. — ^See under Common Colds, etc. Dose 10 to 500 million. Typhus Fever. — ^The prophylactic use of vaccines is recommended by Plotz. As, however, he has not succeeded in producing immunity in experimental animals, it is well to reserve judgment on the favorable results reported in man. (See p. 454.) Typhoid Fever. — ^Prophylaxis. — ^This method had its inception in the demonstration by Pfeiffer and KoUe and by Wright in 1896, that the injection of killed bacilH caused the production of the same antibodies as found in the blood of convalescent typhoid cases. In 1898 Wright inoculated 4000 men in India and Leishman supervised the inoculation of the British troops in the Boer War. In 1909 vaccination was started in the United States Army under the direction of Russell. The results in the present war have been astonishingly good and equally good results have been obtained in civil life. Selection of Vaccine. — ^Various strains are used by the English, French and American Army medical men. The vaccine as advo- 592 APPLICATION OF BACTERIAL VACCINES cated by the United States Army oiEcers is prepared from a strain (Rawlings) of known antigenic value and the vaccine is only heated to 53° C, relying on the added tricresol to kill any bacilli not killed by this degree of heating. Unfortunately, some of the vaccines mar- keted are relatively inactive, possibly due to preparation, heating, pre- servation, or too long a labelled period of potency or a combination of these factors. Sensitized vaccines have been strongly advocated by Besredka, Gay and others. Administration. — A strong degree of immunity is only conferred by two large or three moderate doses. The army men receive 500 million, 1 billion and 10 billion bacteria. Successive Saturdays are most con- venient and it is preferable to give the vaccine in the afternoon so that the reaction, if it occurs, will occur while the subject is abed. Injection should be subcutaneous at the insertion of the deltoid. Reaction. — Usually only a local tender reddened area develops. In some cases it is more extensive and there may be some tenderness of the axillary nodes. Slight constitutional symptoms may develop but a severe general reaction is exceptional. The reaction is of no impor- tance except for the discomfort and has no relation to the subsequent immunity. There is no reason why vaccination should not be done during exposure to infection. Results of Immunization. — Among the many millions of men vaccinated during the present war there has been almost no typhoid fever. An excessive dose of infectious material may break down the protection which is only relative, but any extensive failure should raise a strong presumption that the vaccine employed was not satisfactory. Duration of Immunity. — The degree of immunity decreases after two and a half years but even after four or five years the rate among the vaccinated may be only one-fourth that of the unvaccinated. Under conditions of constant exposure to infection associated with strain and privation as in the present war, the immunization should be repeated each year. Therapeutic Use. — Subcutaneous Injection. — ^Watters has collected and analyzed 1120 cases. Seventy-one deaths occurred, that is, a mortality of 6.3 per cent. Probably 17 cases with 15 deaths could be omitted on the ground that they were moribund when treated or for other reasons. This would give a mortality of 5 per cent. The incidence of relapse was 6 per cent, in the cases where stated. Various-sized doses were employed by the different observers, and there is no correla- tion between dosage and mortality or relapse incidence. Many factors enter into the death-rate of this disease and the lowered rate cannot be directly attributed to the vaccines. In the different series also, the rate varied very widely. In general, the patients treated seemed brighter and the temperature averaged lower, and the febrile period was appreciably shortened. The more moderate dosage, 250 to 500 million would seem advisable. Intravenous Use. — ^This method was introduced by Ickikawa in 1912. Several hundred cases have been treated up to date by various observers. PRACTICAL APPLICATION OF INDIVIDUAL VACCINES 593 The reaction has been described (p. 588). About 50 per cent, quickly convalesce under this_ treatment. There is insufficient data to warrant deductions as to its influence on the death-rate. A number of cases have developed fatal hemorrhage from the bowel or elsewhere after inoculation and several cases have died shortly after injection; in 2 there was an associated pneumonia. Evidence of hemorrhage, pneu- monia or cardiac disturbance are therefore contra-indications. The dosage and vaccines employed have differed widely. From 50 to 250 million is the average. Although Gay advocates the use of sensi- tized vaccines, no one vaccine seems better than another. (See also , under Non-specific Vaccines.) Paratyphoid Infections. — Paratyphoid fever is comparatively un- common in this country, though under camp conditions it may became epidemic, as among the militia encamped on the Mexican border. It prevailed among the European troops before the use of vaccine. Prophylactic Vaccination. — ^There seems no reason why this should not be as effective as with typhoid, although the epidemiological data does not allow of conclusions as yet. Vaccination seems absolutely necessary among troops, etc., as evidenced by our experiences as well as the conditions in the European war. It would not seem, however, that vaccination would give as complete protection against the food-poison- ing types of infection where the dose of preformed toxin is large, although the immunity present may limit invasion by the bacilli. Vaccines of the individual types may be employed in the doses recom- mended for typhoid. Mixed vaccines have been more commonly employed, thus first dose 500 million typhoid, 250 million each of Paratyphoid "A" and "B," second and third doses twice this amount. When time is not a factor it would be well to adhere to our present methods of typhoid prophylaxis and give the paratyphoid vaccine (mixed types) separately.^ The European practice is to give them together. Bacillary Dysentery. — Dysentery vaccines are highly toxic and Shiga has employed the simultaneous injection of vaccine and serum. The results are not wholly satisfactory, although the mortality among the vaccinated was lowered. The prevalence of different types of bacilli adds to the difficulties. It has been claimed that the vaccines are of value in the treatment of carriers. Plague. — ^Prophylactic vaccination gives a relatively short period of immunity and is best undertaken during epidemics. The protection is only relative against bubonic plague but the mortality is also lowered. There is less protection against pneumonic infection. HafFkine advises 3 to 3.5 c.c. of his specially grown broth cultures, giving a second dose after eight to ten days. KoUe advises 2 mg. of the growth on agar. These amounts are fairly equivalent to 500 million bacilli. Cholera. — Prophylactic vaccination affords considerable protection, but the mortality rate of those who become infected is only slightly ^ The U. S. Army paratyphoid vaccine contains 1 billion B. paratyphosus A and 600 B. paratyphosus B to each cubic centimeter. The dosage is 0.5 c. c, 1 c. c. and 1 c. c. at 7- to lO-day intervals. 38 594 APPLICATION OF BACTERIAL VACCINES influenced. The reactions are frequently moderately severe. Live vaccines have also been employed. Combined Prophylatic Vaccines. — Castellani is the main advocate. Combined vaccines against the endemic types of disease would be of advantage when time was a factor.^ The dosage is as with the typhoid bacilli. The addition of the cholefa vaccine does appreciably increase the reaction. B. Pyocyaneus and B. Proteus. — Usually encountered as secondary invaders, some benefit has been reported from the use of vaccines. The dosage is from 25 million to 1 billion. Glanders. — There is some indication that vaccines are of value in subacute or chronic infections in man. The temperature should be carefully watched as the vaccine acts similarly to mallein. Dosage 10 to 100 million. B. Coli and Belated Types. — Genito-urinary infections seem, in some instances to be benefited, especially cystitis and possibly pyelitis after the acute symptoms have subsided. Vaccines may have some influence in diminishing the fever and discharge from sinuses after pelvic abscess, appendicitis, or cholecystitis. The mucoid B. aerogenes as well as inter- mediates resembling B. paratyphosus are frequently found in these types of infection. The dosage ranges from 25 to 500 million or more. Different strains vary in the degree of reaction produced. Atrophic Rhinitis and Rhinoscleroma. — ^The etiology of the former is not settled, Perez claiming that the " cocco-bacillus ozena" is the cause. Vaccines of this organism with or without B. ozense have given suggestive results. Rhinoscleroma is possibly influenced by vaccines. Other Infections due to Encapsulated Bacilli. — ^These types are encoun- tered in infections of the respiratory tract or by extension in sinus, middle ear and mastoid. It is doubtful if vaccines are of any value, at least they cannot be applied during the acute stage. Dosage. — ^The same as for B. coh. The initial dose of bacillus of Perez is 50 million to be increased until a focal reaction occurs. Pertussis. — Prophylaxis. — Some protection is conferred but it is difficult to judge of the degree with the data available. Hess vaccinated 244 children and 20 developed the disease, whereas of 80 equally exposed <'hil- dren 59 developed the disease. These results as well as those of Still and Luttinger warrant its application, especially as the procedure is innocuous. ; Therapeutic ,^^Hess could see no influence on the disease even where botK. prophylactic and therapeutic vaccines were given. It is difficult to correlate these results with the reports of others that the number of paroxysms and the duration of the disease is lessened. In some instances a prompt amelioration has been reported, so prompt as not to be explained on the basis of specific antibody production. We have been treating two series of cases, one with pertussis vaccine, the other with a vaccine of B. influenza which, though similar culturally, differs com- pletely immunologically. One vaccine shows results about as good as the other. It would almost seem as though we were dealing with a non- specific action on the mucous membrane condition. PRACTICAL APPLICATION OF INDIVIDUAL VACCINES 595 Dosage. — For children over one year 500 million, 1 billion and 2 billion at two-day intervals is recommended. Children under one year receive half these doses. If, after several days, improvement is not marked, further injections may be given. Prophylactic injections are given every third day, the doses being 500 million, 2 billion and 3 billion respectively. bifluenza Infections. — No data are available concerning epidemic influenza. The presence of influenza bacilli in inflammations of the mucous membranes, accessory sinuses and conjunctiva is not necessarily an indication of their etiological importance. The value of mixed vaccines containing influenza bacilli in this condition is problematical. The use of autogenous vaccines where the evidence points strongly to their etiological importance has given at best only suggestive results. Dosage. — Initial dose 10 million to 20 million, which can be increased to 200 million to 500 million. Tuberculosis (see Tuberculin Therapy in chapter on Tuberculosis). — Vaccines of the secondary invaders, in pulmonary phthisis have been tried with only meagre results. Focal Infection and Systemic Disease. — In general the conclusion seems warranted that vaccines are insufficient if the focus is not eradicated, and if eradicated, vaccines are not necessary. Each case is a problem in itself and a careful study of typical case reports is necessary to a comprehension of the subject. Streptococci and gonococci are the most frequent causative organisms. Miscellaneous Conditions. — Common Cold. — ^The beneficial results both prophylactic and curative, should be viewed with scepticism. Little is known concerning the etiology of common colds. Some outbreaks are due to a filtrable virus, according to Kruse and Foster. Micrococcus catarrhalis, influenza bacilli, pneumococci, streptococci, B. segmentosus, etc., have been encountered as the predominating organism. The types encountered are usually members of heterogeneous groups and it is difficult to see how mixed stock vaccines can have any influence at least from the specific stand-point. Possibly inoculation of such vaccines may have some obscure non-specific protective or ciu-ative influence on the mucous membranes. Bronchitis and Chronic Respiratory Conditions. — ^Autogenous vaccines of the predominating flora cause at most a slight amelioration in a few cases. Sinus and Middle-ear Infections.— The treatment of subacute or chronic infections has given little result, possibly because of the anatorriical conditions. Mouth Infections. — ^The use of vaccines in pyorrhea is advocated by some observers but they agree that local treatment is necessary as well, whereas others find that local treatment alone is all that is necessary. The vaccines employed only represent a small part of the aerobic flora, and the dominant anaerobic fusiform bacilli and spirochetes are ignored. Whether vaccines influence the general symptoms which may be associ- ated is another problem. (See Focal Infections.) 596 APPLICATION OF BACTERIAL VACCINES Selection of Vaccine, Dosage, etc., in above Conditions. — ^An autogenous vaccine based on a careful bacteriological examination is alone applicable. The stock vaccines widely advertised are, so far as oiu- present knowledge goes, non-specific vaccines, even though the contained organisms bear the same names as those encountered in the inflammation. The dosage for the individual types has been given. Pro rata reductions should be made according to the number of types in the vaccines. REFERENCES. Non-specific Factoks. JoBLiNG and Peterson: Jour. Am. Med. Assn., 1916, Ixvi, 1753. Hektoen: Jour. Am. Med. Assn., 1916, Ixvi, 1591. Ektsipelab. Erdman: Jour. Am. Med. Assn., 1913, Ixi, 2048. GoNOOoccTTS Infections. Inteavenotjs Therapy. Bkijck and Sommeks: Milnohen med. Wchnsehr., 1913, Ix, 1185. Menzeb: Med. Klin., 1913, ix, 1332. Fruhwald: Med. KKnik, 1913, ix, 1799. Kyle and Mtjcha: Wien. klin. Wchnsehr., 1913, xxvi, 1755. Bordack: Milnchen. med. Wchnsehr., 1913, Ix, 2622. Kreibick: Wien. klin. Wchnsehr., 1913, xxvi, 2024. GoNococcus Infection. Non-specific Therapy. MtJLLER and Weiss: Wien. klin. Wchnsehr., 1916, xxix, 249. Miller and Lusk: Jour. Am. Med. Assn., 1916, Ixvi, 1756. Sensitized Vaccines, Typhoid. Gay and Claypole: Arch. Int. Med., 1914, p. 671. Sawyer: Jour. Am. Med. Assn., 1915, Ixv, 1413. Nichols: Jour. Exp. Med., 1915, xxii, 780. Vaccines in Typhoid Fever. Sbbcutaneous. Watters: Med. Record, 1913, Ixxxiv, 518. Vaccines in Typhoid Fever. Intravenous. McWilliams: Med. Record, October 16, 1915; Jour. Immunol., 1916, i, 759. Gay and Chickering: Arch. Int. Med., 1916, xvii, 303. Combined Prophylactic Vaccines. Castellani: Cent. f. Bakt., 1915, Ixxvii, 63. Jour. Trop. Med., 1914, xvii, 326. Ozena. Horn: Jour. Am. Med. Assn., 1915, Ixv, 788. Pertussis Vaccine. Hess: Jour. Am. Med. Assn., 1914, Ixiii, 1007. CHAPTER XLVII. THE PRACTICAL APPLICATIONS OF SERUM THERAPY. The advisability of using sera in any particular disease is influenced by a number of considerations. The primary one is whether a serum has been obtained and is available which contains antibodies of suitable kind and amount to neutralize the toxins or aid the body cells and ferments in destroying the microorganisms. Others almost equally important are whether the infections can be identified from clinical signs alone, or only after additional laboratory examinations, and whether the sera can be brought into contact with the toxin or organisms at the essen- tial points, in the necessary concentration and within the required time. Practical serum therapeutics in the more important infections in which some results have been obtained will be considered in the light of the above considerations. SERUM TREATMENT OF LOBAR PNEUMONIA. This disease is so uniformly due to one organism that from the" clinical signs alone the physician can almost assume a pneumococcus infection. This knowledge is not as valuable as it seems since it has become evident that the characteristics which define a pathogenic microorganism are frequently so broad as to include a number of strains, which from the view-point of immune sera are as distinct as microorganisms with wide cultural difference. The term "pneumococcus" as pointed out by Neufeld is one of these broad group names which cover a number of strains each one of which is unaffected by the antibodies produced through immunizing injections with the others. A further extremely important differentiation between the types is that with our present methods some excite by their injection into suit- able animals abundant antibodies, while others do not. The reasons for this we do not understand. The most recent and most thorough investi- gation concerning the value of the specific serum for pneumonias due to each type of organisms has been carried on at the Rockefeller Institute for Infectious Diseases by Cole and his associates and the present favor- able outlook for treating a certain proportion of pneumonia cases is largely due to their work. (See chapter on Pneumococci.) Cole's, Longcope's and Richardson's statistics of the frequency and the mortality of pneumonias- due to the types are of great practical interest. Rockefeller Instit\ite (P. H. Longcope) . Univ. Penn. Hospital, Incidence, Mortality, (Richardson) . Number. CoE^' ■ cent. per cent. Incidence, Mortality, Cole. Longcope. Longcope. Cole. Longcope. No. per cent. per cent. I. . . 78 13 33 23 25.0 12.5 60 31 30 II. . . 75 11 32 21 29.0 72,7 39 20 25 III. . . 22 7 9 14 45.0 85.7 13 6 50 IV. . . 48 21 20 40 12.5 23.8 83 43 12 other bacteria 14 6 598 PRACTICAL APPLICATIONS OF SERUM THERAPY It is noticed that about 30 per cent, of the eases of pneumonia and about one-third of the total deaths are caused by infections with type I. These figures are very important because at present the results of serum injections follow their antibody content and seem favorable in type I infections, doubtful in type II and negative in type III and in type IV. Method of Administration. — ^The experimental work in animals and the observation of cases has led to the general use of larger injections and the substitution of the intravenous for the subcutaneous method. The size of the patient and the strength of the serum in antibodies have not been considered in controlling the size of the dose, although all accept the fact that theoretically they should. The serum should be stand- ardized, as in the case of antitoxins, 0.2 c.c. of a recent serum should protect against 100,000 f. d. of a very virulent type I pneumococcus. Serum should seldom be put out or used that contains less than half this strength. Cole advises that 80 c.c. of serum diluted with 80 c.c. of salt solu- tion be injected intravenously and repeated about every twelve hours until permanent improvement is noted. Usually he gives three to five doses. The repetition of the dose every twelve hours is founded on the desire to give it sufficiently often. Cole states that the transferred antibodies tend to disappear after a dose of serum. This is due, in his opinion, to combination with the antigen in the body. In our own experimental work we have found that a single dose of 0.1 c.c. protects a mouse for the next four days from an injection of 1000 fatal doses of living pneumococci, and as late as the fifteenth day, from 100 fatal doses. The mouse differs from the sick person in not having an infected area. In order that comparative results should be obtained, it is necessary to have all antipneumococcic sera labelled as to the strength and nature of antibodies and the date of the last potency test. The serum used should have been tested within three months, for it gradually deteriorates. Cole recommends that the" serum be only given in a case after the bacteriological test has shown the type. I believe, in severe cases a first dose of type I serum should be given as soon as possible and the later injections only after the bacteriological report. Except in hospitals Cole's plan means in most cases a delay of at least twenty-four hours and generally prevents the use of serum altogether. It is understood that the severe cases which receive an injection of type I serum and from which no bacteriological report is received cannot be considered as having any value in forming an opinion of the effect of the treatment; Thus if Dr. Longcope had treated all his cases with type I serum and had not identified the infecting type in each of the 22 deaths among his 52 patients he could have had no conception of the value of the serum, for only one of the deaths and 13 of the cases were due to the type I organism. SERUM TREATMENT OF EPIDEMIC MENINGITIS 599 The Results of Large Intravenous Injections of Specific Serum in Type I Infections.— There is in about 30 per cent, of the cases an almost immediate more or less severe chill with a considerable rise of temperature which lasts for a short period. This occurs usually after the first but sometimes after the later serum injections. If the blood contained pneumococci, they disappear within twelve hours after the injection. The temperature usually rapidly falls after the initial rise to a point lower than before the rise and the symptoms, as a rule, improve sooner than the average untreated case of equal severity. The mortality up to the present time has been much less than in the untreated cases. Serum sickness with rashes, painful joints, swelling of lymph glands and other symptoms occur to a greater or less extent in about 50 per cent, of the cases during convalescence. The serum sickness, while it lasts, is very annoying, but not dangerous. The fact that such really promising results have been obtained in cases due to type I leads us to hope that Wadsworth, Cole and others will succeed in their endeavors to get a therapeutically valuable serum for types II and III and possibly so concentrate the antibodies as to use a combined polyvalent serum. SERUM TREATMENT OF EPIDEMIC MENINGITIS. The intraspinal method of injecting serum from horses immunized to meningococci in. meningitis introduced by Jochmann has been approved by all. The collective investigation carried out under Flexner's super- vision practically settled its value, and if it were not for the somewhat dis- couraging reports received from England during 1915 it would not be necessary to more than allude to the value of the serum. Several reports from England mention a mortality under serum treatment of different collections of cases of from 52 to 63 per cent. and several experienced men have advised against the use of the serum. Should these results in any way cause us to change our favorable opinion? There are undoubtedly certain epidemics in which an unusual propor- tion of the cases develop a thick tenacious exudate which greatly hinders the successful use of the serum. A number of these have been found at the New York State Quarantine Station. Again the sudden great foreign demand for the serum caught some of the manufacturers with a small supply and the attempt to replenish their stock caused them to bleed their horses too frequently and thus obtain a serum which was found at a later period to be poor in antibody content. Unless the strains used are properly selected the antibodies may not be suitable to combine with the strain producing the local ep'demic. That one or more of these explanations for most of the poor results are true is rendered certain by the fact that when potent serum was used in several thousand cases among the English troops in 1916 the 600 PRACTICAL APPLICATIONS OF SERUM THERAPY results were very good. Since 1910 the New York City Health Department bacteriological laboratory has treated all cases of epidemic meningitis applying to it. Our mortality in different years has varied from 21 to 34 per cent. Drs. Sophian, DuBois and Neal, who have had charge of the serum treatment for. the laboratory, are absolutely convinced of the value of the treatment, not only from the clinical coiu-se but from the changes which take place in the spinal fluid as shown by the disappearance of the meningococci. Administration of the Serum.— Dr. Neal finds that too often the private physician fails to repeat the injections sufiiciently often. It is very rare, except in a case already convalescent, that it is correct to give daily injections for less than four days. If the organisms or symptoms do not disappear the injections of 10 to 25 c.c. of seriun should be con- tinued for many days. Fortunately lumbar puncture with removal of fluid is of value in the treatment. Unlike pneumonia, practically all cases of epidemic meningitis can be treated by a single polyvalent serum. The different strains do, indeed, differ, but they have group relation- ships and intravenous injections of suitably selected strains cause the horse to produce a seriun capable of influencing almost all strains. This has been recently carefully investigated by Dr. WoUstein at the Rockefeller Institute. Every lot of serum should be tested for its com- plement-fixing, opsonic and agglutinating power until some better method be devised to estimate its value. Its agglutinating value should be tested especially in relation to its activity against various strains. The results of the tests should be stated on the package. An agreement should be reached to make a standard unit. The fluid removed from every case of suspected epidemic meningitis should be examined microscopically and culturally, for only in this way can a correct diagnosis be made. Directions for Use of Serum. — ^The following directions are issued by the Research Laboratory of the New York City Board of Health. Perform a liunbar puncture under aseptic precautions in the third or fourth lumbar space. A general anesthetic should never be used. In hypersensitive patients a local anesthetic may be advisable. Have the patient lying on the side with the back arched so that there will be the greatest possible distance between the spines of the vertebrae. Find the notch nearest a line connecting the crests of the ilia. Intro- duce the needle, preferably a Qumke needle, in the midline and push forward and a little upward. The distance the needle goes in depends upon the age of the patient and the muscular development. It varies from I to 3 inches. Allow the cerebrospinal fluid to flow out until the pressure is so reduced that only 3 or 4 drops come per minute. If the fluid is cloudy, inject the serum immediately. The serum is warmed to body temperature and injected very slowly under the least possible pressiu-e. A funnel and the tube arrangement allowing it to run in by gravity should be used. The barrel of an ordinary syringe may be used as a funnel. The rubber tubing should be I to i of an inch in diameter and long enough so that the funnel ANTISTREPTOCOCCIC SERA 601 can be raised 12 to 15 inches. In general, the average dose for an adult is 20 to 40 c.c. and for infants and children 2 to 20 c.c. The amount depends as much upon the quantity of cerebrospinal fluid withdrawn as upon the age. An infant will frequently stand 10 to 15 c.c. without difficulty. The dose should usually be at least 5 to 10 c.c. less than the amount of cerebrospinal fluid withdrawn. When serum apparently runs in freely after a dry tap, it is advisable to proceed very slowly and to watch the patient carefully for the slightest change in pulse and respiration. In cases with very thick exudate which will not flow through the needle, gentle suction with a syringe may be tried. If that fails, a little serum injected will sometimes start the flow. When possible, further injections are made only after bacteriological examination has determined the cause to be the meningococcus. The antimeningitis serum does no harm in meningitis due to other organisms. In severe cases it is best to inject the serum every twelve hours until there is improvement. In moderate and mild cases it should be repeated each day for the first four days. Further administration depends upon the patient's general condition and the bacteriological examination of the fluid. Usually four to six injections are necessary, but as many as fifteen or more may have to be employed. During or immediately after the injection of serum the respiration may entirely cease or the pulse may become very rapid and thready. Such an occurrence, while alarming, is not necessarily serious and is best treated by immediate withdrawal of some of the serum if the needle is still in place. If the i^eedle has been withdrawn, or, if after some of the serum is removed the symptoms do not ameliorate, artificial respira- tion should be resorted to for the respiratory condition and adrenalin or other stimulants hypodermically for the heart. The successful treatment of cases of meningitis must always depend upon experience and judgment. It cannot be reduced to a rule of thumb. In all cases with meningeal symptoms a lumbar puncture should be done. No ill effects follow, on the contrary the relief of the pressure frequently produces beneficial results. ANTISTREPTOCOCCIC SERA. The same reasoning applies as with the antipneumococcic serum and the same dosage and method of administration. Some of the strains recur frequently, others infrequently in infec- tions. If we suspect that the infection is due to the hemolytic type of organism we can rightly give this type of serum with the hope of some good resulting. In the viridans type of infection we have no evidence that any good is done by serum. This is probably due to the fact that each organism produces its own antibodies. The organism should be identified as soon as possible and, if virulent in mice, tested against the serum in a mouse. Much further combined 602 PRACTICAL APPLICATIONS OF SERUM THERAPY clinical and laboratory investigation is required before a decision can be reached as to the value of the serum and the best dosage. If it were not for serum sickness there would be no question that injections should be made early before the infection has advanced and the streptococci acquired a somewhat greater resistance to the specific effect of the serum antibodies and ferments. The repeated local bathing of infected tissues with the serum seems to have a more beneficial action than that exercised by a non-specific serum. THE SERUM TREATMENT OF BACHiARY DYSENTERY. The earlier opinion has been confirmed that bacillary dysentery alone occurs in cold and temperate climates while in hot climates both bacillary and amebic cases occur in about equal numbers. The idea that summer diarrheas were frequently due to the dysentery bacilli has been discarded. The bacilli may be present at times in small numbers in these cases, but they have too little part in the disease process to require specific treat- ment. The dysentery bacilli may be divided into dysentery bacilli and paradysentery bacilh or into different strains of a group. Like so many other bacteria causing inflammations of mucous membranes the organisms exciting dysentery belong to a variety of strains. For sporadic cases and those occurring in the beginning of an out- break, a polyvalent serum must be used. In later cases in an epidemic where the strain has been identified the monovalent serum is to be used if possible. The serum in mild cases is given in doses of from 10 to 30 c.c. twice daily according to the size of the person. In severe cases as high as 100 c.c. two or three times in the twenty-four hours can be given and in desperate cases the serum has been given intravenously. If given intravenously, the serum must be warmed and given slowly. The doses are to be continued from day to day until permanent improve- ment is established. Injections are usually continued for two or three days. The majority of those who have used the serum con- firm Shiga's original belief that the results are good. Our own experi- ence in a number of severe cases is on the whole favorable. We believe it is unnecessary to use the serum in slight cases. In favorable cases within six to twelve hours the constitutional symptoms frequently improve. The abdominal pains are less, the mental condition is better and the pulse slower and stronger. There is frequently at this time some reduction in the number of stools, though they may be more copious and have increased sloughs. The serum on the market is not standardized and some of it is valueless. The necessity of using a polyv- alent serum in most cases also lessens its value. In spite of these objections the use of the serum in severe cases is strongly indicated. It is very desirable that records be kept of the cases treated and the results reported. Shiga believes the mortality to be reduced by the serum treatment THERAPEUTIC USE OF HUMAN BLOOD 603 from around 35 per cent, to about 9 per cent. Some report even more striking results, while still others have met with disappoint- ment. Undoubtedly the extent of mixed infections with other organ- isms, the possibility of some cases having been due to amebas and the use of inactive serum must be considered as being possibly the cause of the failure of the serum in these cases. Ruffer and Willmore showed that the Shiga type of serum had no effect on their cases which were due to P or Y type of paradysentery bacillus. Either a polyvalent serum ■ or a suitable monovalent serum was found effective. The usual serum after-effects may develop. THE THERAPEUTIC USE OF HUMAN CONVALESCENT BLOOD HAVING SPECIFIC ANTIBODIES. Convalescent serum or whole convalescent blood has been used in the treatment of early toxic cases of scarlet fever both here and abroad and has given encouraging results in the limited number of cases observed thus far. Reiss and Jungman recommended the intra- venous injection of 50 to 100 c.c. of pooled convalescent serum, while Zingher makes use of the intramuscular injection of whole convalescent blood, citrated or non-citrated, which he injects in quantities of 120 to 240 c.c. The blood causes no local inflammatory reactions in the muscles, and is rapidly absorbed. The convalescent serum or fresh whole blood is obtained from patients who are two or three weeks convalescent from scarlet fever. These donors should be free from syphilis and tuberculosis. The effect of convalescent serum or whole blood in the uncomplicated early toxic cases of scarlet fever is seen in a critical drop in temperature, beginning about six hours after the injection and ending in from twenty- four to thirty-six hours; an early fading of the rash; improvement in the circulation and character of the pulse; and especially in the general condition and mental symptoms of the patient. Zingher reported the results obtained at the Willard Parker Hospital with intra- muscular injections of whole convalescent blood in 15 very toxic cases of scarlet fever, selected out of a total of 900 admissions. A striking improvement was noted in 5 cases, improvement and final recovery in 6 more, while 4 patients died from various septic compli- cations. In the later septic cases, seen from the fifth to tenth day of disease, complicated by an extensive streptococcus exudate over fauces and tonsils, enlarged and tender cervical glands, a poor circulation and showing general septic temperature, fresh normal blood, injected in quantities of 120 to 240 c.c. and repeated if necessary in three or four days has shown very beneficial effects in some desperately ill cases. Fresh normal blood has no specific action in septic cases of scarlet fever, but it supplies definite nutritive, stimulating, and normal bactericidal substances. 604 PRACTICAL APPLICATIONS OF SERUM ThERAPY USE OF DIPHTERIA ANTITOXIN IN TREATMENT AND IMMUNIZATION. The antitoxin in the higher grades of globulin solution or serum is identical with that in the lower grades; there is simply more of it in each drop. In treatment, however, for the same amount of antitoxin we have to inject less foreign proteins with the higher grades, and therefore have somewhat less danger of rashes and other deleterious results. The amount of antitoxin required for immunization is 300 to 500 units for an infant, 500 to 1000 for an adult, and proportionately for those between these extremes. The larger doses are advised when the danger of infection is very great. After the observation of the use of antitoxin in the immunization of many thousand cases, we have absolute belief in its power to prevent an outbreak of diphtheria, when given in the amounts advised, for at least twelve days, and also of its almost complete harmlessness in the small doses required. When double the above quantities are given the immunity is prolonged on the average for about one week. If it is desired to prolong the immunity the anti- toxin injection is repeated every ten days. Treatment. — ^Although more than twenty years have elapsed since the introduction of diphtheria antitoxin in the treatment of diphtheria, good observers, although nearer together than at first, still differ in the amount which they believe should be injected and in the method of its administration. Before giving our own conclusions on the proper dosage, it is well to consider several important points upon which this dosage is founded. The amount of toxin in any case of diphtheria is comparatively small. One hundred units of antitoxin which would neutralize fifty times the amount of toxin sufiicient to kill a six-year-old child, would surely make harmless all the toxin present in the most malignant cases if it could gain access to it in time. If we gave antitoxin, there- fore, as many suppose, simply in sufficient amount to neutralize the poison in the body of an infected person, comparatively small amounts would be injected, but we have to give very much more than this because of the time it requires for much of the antitoxin to reach the toxin. This can be brought into direct contact with the toxin only by being absorbed into the blood and then passing through the capillary walls to the tissue fluids and cells. The greater the quantity of anti- toxin that is in the blood, the greater will be the speed that an appre- ciable amount will pass to the tissues. The combined endeavor of the clinical observer and the laboratory worker is to find the suitable dose which will give a sufficient concentration in the blood to neutraUze, as quickly as necessary, the toxin in the tissues. In the laboratory we can test the amount of antitoxin which is absorbed into the blood from any given dose and the amount which passes out to the tissues, while the clinical observer can note the changes which take place as he watches the case after antitoxin treatment. It is naturally a matter of great importance as to how the antitoxin USE OF DIPHTHERIA ANTITOXIN 605 is administered. When given subcutaneously, the swelling caused by its injection rapidly disappears by the absorption of the water, but the globulins and antitoxin remain behind in the tissues because Fig. 204. — ^Amount of antitoxin in 1 c.c. of serum from persons, at different intervals of time, after a single subcutaneous injection of 10,000 units. of the slow absorption of proteins. By testing many patients, it has been found that it takes twenty-four hours for the major part of the antitoxin to be absorbed bv the blood from the subcutaneous tissues UN TS 2 3 4 5 " 20 ^ 19 \ 18 N. 17 ^^^ 16 >, 15 14 It-POi ND8 ' 13 11 10 9 -=:!!ioi7- |0« a ^"— .."^ 7 R " r ^ 5 4 . _^ - — _/* JT V^^^ .2 3.0 .8 / / ^ N ^ / /^ N. ^ ^\ / 1 ^ *^^ 'N.s^ .^ '^-^ ^ ' / y^.- ■^ -. "^ / / / ^ ,-'*«- --. ^^ ^° »o„^ ■ 2.0 .8 .6 .4 1 / J^ ■ '^•^ f / ^7 '''l 1 /^.^ / --m ' """If I If ■f 1.0 If ■'' / / '' 1 " '/ / ^ --— - .2 /Xx^. ~ r — [5s> Fig. 206. — Amount of antitoxin in 1 c.o. of serum from persons, at different intervals of time (days), after a single subcutaneous injection of 10,000 units. of delay in the neutraUzation of the toxin in. a severe case is of impor- tance, but in a mild case, where dangerous poisoning is still remote, slight delay makes little difference. Infants and children are especially liable to laryngeal diphtheria, so that every case in a child presents a certain gravity which the adult does not present. The last point to be considered is whether a single or a multiple dose should be given. It must be realized that antitoxin has no effect whatever on injury which has already taken place. It is as useless as water on the ashes of a burned-out building. If the toxin is per^ manently united with the cell substance, antitoxin is no longer of any service. It is the early and sufficient dose which Is important. When we give a divided dose, we simply get the effect of the first portion during the interval before the giving of the second dose. If the second dose had been given with the first, we would have had its effect added. USE OF DIPHTHERIA ANTITOXIN 607 and so an insufficient dose made adequate. When the first dose has been of a sufficient size, the second and third injections, though harmless, are absolutely useless. The holding back of a part of the first dose so as to give it later, simply delays its action to a time when it cannot have much, if any, effect. For the last three years, we have used in the hospitals for contagious diseases only a single dose of antitoxin, which in mild cases, has been given subcutaneously; in moderate cases subcutaneously or intra- muscularly; and in severe cases, intravenously or intravenously and intramuscularly. After twenty years of experience in treatment and animal experimentation and consultation with physicians in New York and elsewhere, the following dosage, which is that adopted by the Health Departments of the City and State of New York, is advised. Dosage of Units op Antitoxin in Diphtheeia. single dose only. Infant, 10 to 30 pounds (under two years) Mild. Moderate. Severe. Malignant. 2000 3,000 5,000 3000 5,000 10,000 10,000 Child, 30 to 90 pounds (under 15 years) 3000 4,000 10,000 10,000 4000 10,000 15,000 20,000 Adults, 90 pounds and over. 3000 5,000 10,000 15,000 5000 10,000 20,000 40,000 METHOD OP ADMINISTRATION. Intramuscular Subcutaneous Intramuscular or ^ Intravenous i Intravenous or or and and intramuscular subcutaneous i intramuscular J intramuscular or or subcutaneous subcutaneous The above amounts are sufficiently large, and I think no appre- ciable advantage would be obtained by increasing them. Very much smaller doses are still able to do great good, as the general blood current soon becomes antitoxic and blocks any further passage of toxin from the diseased tissues to other portions of the body. The feebly antitoxic plasma gradually permeates the body. There will be, however, more delay in the improvement of the local process, for the neutralization of the concentrated toxin in the diseased tissues will be slower. The exudate or pseudomembrane will continue to increase for some hours after the complete neutralization of the toxin because the injury to the tissue takes time to manifest itself. Determination of the Presence of Antitoxin in the Livings Body and Results of Toxin-antitoxin Immunization. — See pp. 312-315. Danger in Giving Injections of Antitoxin. — About 1 in 10,000 persons develop, within a few minutes after an injection of serum, alarming symptoms. About 1 in 50,000 of the injected die. About 30 deaths in all have been reported. In 140,000 persons injected by New York 608 PRACTICAL APPLICATIONS OF SEBUM THERAPY City Health Department Inspectors there have been 2 deaths due to serum. About the same proportion is reported from Boston. The persons suffering severe symptoms have usually been subject to asthma, while the fatal cases usually have the pathological changes known as status lymphaticus. A few of these rare cases die almost instantly. As a rule, when death occurs it takes place within a few minutes after the development of symptoms. Usually the respiratory rather than the circulatory centre seems to be affected. Persons who have not reacted ' badly to a first injection do not need to fear a second. So far as known all fatal results have followed the first injection. Results from the Use of Antitoxic Globulin Solution. — ^The curative effect proved to be identical with that of the whole serum. Our tests showed clearly that not only the toxin, but also the poisons produced in the animal by injections with virulent bacilli are neutralized as com- pletely by the globulin solution as by the antitoxic serum from which it is separated. The injections of the globulin solution were found to be followed by decidedly less severe rashes than the whole serum, and it was especially noted that there were very few who had any constitu- tional disturbances even when the development of the rashes did occur. The following comparative table gives a siunmary of the constitu- tional and local reactions obtained in the treatment of 50 cases of diph- theria in young children, with a lot of antitoxic serum received from three horses which was found to produce an excessive amount of dis- turbance, and of an equal number of similar cases treated with a solution of the antitoxic globulins derived from a portion of the same lot of serum is as follows: Children who were treated Children treated with with the antitoxic the whole serum. globulins. Marked constitutional symptoms accompanied by a severe and persistent rash .... 28 per cent. per cent. Moderate constitutional symptoms accom- panied by a well-developed erythema or urticaria . 18 per cent. 4 per cent. Very slight constitutional disturbance accom- panied by a more or less general urticaria or erythema 20 per cent. 8 per cent. No appreciable constitutional disturbance but more or less general urticaria or erythema . 4 per cent. 34 per cent. No appreciable deleterious after-effects what- ever 30 per cent. 54 per cent. DuEATioN OF Rashes. Days ........1 2 3 4 5-6 7 8 Totals Antitoxic globulin cases ...5 7 5 2 3 .. 1 ..23 Whole serum cases 1 4 10 1 10 3 2 5 36 The concentration of antitoxin made possible by the elimination of the non-antitoxic substances is not only a convenience but is of a distinct importance, as it tends to encourage large doses. Some pro- ducers, however, supply a product which is so rich in protein as to be almost semisolid. This is not quite so well absorbed as the less concentrated product. The total solids in the globulin solution should not be more than twice those in the serum. TETANUS ANTITOXIN IN TREATMENT 609 The antitoxic globulin solution tends to become slightly cloudy when kept at moderate or high temperatures. This does not interfere with its potency. Substances such as solutions of carbolic acid and tricresol precipitate it but not in the quantity usually employed as preservatives. TETANUS ANTITOXIN IN TREATMENT AND IMMUNIZATION. While tetanus antitoxin has proved most eJ0Bcacious in the prevention of tetanus, its employment as a curative agent has been much less successful. This failure to produce more uniformly good results is due chiefly to its too late' administration. Insufficient dosage and the use of the subcutaneous method have also been important factors. While the subcutaneous use of antitoxin, which at first was the usual method employ ^id, may prove of value in large doses given within the first few hours after the onset of symptoms, nevertheless it frequently fails because of slow absorption and the time required to reach the tissues of the central nervous system. In the light of our present knowledge, this method can be justified only by an inability of the physician, for one i-eason or another, to give an intravenous or an intraspinal injection. The latter method is being more and more recommended, and when employed before the disease has made too much headway has given much better results than the subcutaneous or even the intravenous. J'rom time to time since 1903 single cases or small series of cases have been reported in which the antitoxin has been given intraspinally. Thus Neugebauer^ in 1905 reported 43 cases from Continental and American sources with a mortality of 22, or 51 per cent., and also 3 patients treated by him of whom 2 recovered. Many of these patients received also intravenous or subcutaneous injections or both. The results seemed to some observers to be better than when other methods were used without intraspinal dosage, but on the whole, this method of treatment has not made much headway. Experimental proof of the greater value of the intraspinal method has not until very recently been at all convincing. Permin, of Copenhagen, working on rabbits and dogs, showed that local tetanus could be prevented by the simultaneous injection of tetanus toxin intramuscularly and a dose of antitoxin given in the spine, whereas with the same dose of antitoxin given intravenously, local tetanus occm'red. . Four hours after the giving of the toxin alone neither method of preventing the occurrence of local tetanus was eflficacious, and when nine hours were allowed to elapse between the giving of toxin and antitoxin, the animals could not be saved. The following series of experiments were undertaken by Nicoll and us to determine to what extent tetanus antitoxin given in the spine has greater curative power, when the disease is actually established, than when given in the circulation. The comparative ineffectiveness of subcutaneous injections of tetanus antitoxin in developed cases is inci- ' Wiener klin. Wchnschr., 1905, xviii, 450. 39 610 PRACTICAL APPLICATIONS OF SEBUM THERAPY dentally brought out. The use of guinea-pigs for this purpose has not been recorded so far as we know. They possess, however, a marked susceptibility to tetanus together with comparative freedom from intercurrent diseases. The toxin in each case was inoculated intra- muscularly into the thigh. Weight, Condition Amount No. gms. of leg. Method. in units. 116 290 Fairly stiff Control 42 310 Fairly stiff Control 206 250 Slightly stiff Heart 100 237 275 Fairly stiff Heart 100 399 300 Fairly stiff Heart 100 216 255 Slightly stiff Nerve 200 ®7 255 Fairly stiff Nerve 200 289 280 Fairly stiff Nerve 200 306 285 Slightly stiff Nerve 200 69 255 Stiff Spine 10 304 275 Fairly stiff Spine 10 321 320 Fairly stiff Spine 10 Administkation and E,esdi.t of Tetanus Antitoxin. Result. D 3 days D 3 days D 8 days D 4 days D 5 days D 4 days D 3 days D 3 days D 3 days Discharged; normal Discharged ; drags leg Discharged; drags leg Six Given Antitoxin Twenty-two and a Half to Twenty-thkee Hours AFTER Inoculation of Toxin. D 5 days D 4 days D 4 days Discharged ; drags leg D 5 days Discharged; drags leg This would seem absolutely conclusive of the superiority of the intra- spinal method of giving antitoxin over the intravenous method. It will be noted that only those animals receiving the antitoxin in the spine were able to survive. This result is the more striking since the amount of antitoxin given them was only a fractional part of that given in the circulation. An attempt was made to give the antitoxin to four guinea-pigs intfaneurally. Under an anesthetic the sciatic nerve of the affected limb was cut down upon and freed from the surrounding tissues, as much antitoxin as could be introduced into the nerve sheath injected, and the remainder intramuscularly directly along the course of the nerve. Owing to the small caliber of the nerve, most of the antitoxin passed to the deep intramuscular tissues. The animals lived no longer than those receiving subcutaneous injections. In all of these experiments no guinea-pigs were discharged from observation until there was absolutely satisfactory evidence that the disease had long ceased to make any progress. While a number of the animals, perhaps most, on being discharged exhibited more or less stiffness in the inoculated leg lasting for weeks or even months, they were nevertheless in the best of health in other respects. This condi- 102 300 Stiff 10 325 Stiff 272 350 Stiff 263 285 Stiff 123 325 Stiff 294 350 Stiff H. 100 H. 200 H. 200 Sp. 50 Sp. 50 Sp. 50 TETANUS ANTITOXIN IN TREATMENT 611 tion appears to be identical with that seen in human tetanus, in which the stiffness in different groups of muscles is often very prolonged even when the patients are perfectly convalescent. Since beginning this work we have obtained records of twenty-four consecutive clinical cases of tetanus in which an intraspinal injection of antitoxin was given and eighteen patients recovered. In all of them in addition to the antitoxin used intraspinally, larger amounts were also given by other methods. On experimental and clinical grounds the following recommendation for the treatment of tetanus with antitoxin would seem to be amply justified: In every case strongly suspected of being tetanus, from three to five thousand units of tetanus antitoxin should be given at the first possible moment intraspinally, slowly by gravity, and always, if possible, under an anesthetic. In order to insure its thorough dissemination throughout the spinal meninges the antitoxin should be diluted if necessary, to a volume of from 3 to 10 c.c. or more, according to the patient's age. When fluid is drawn off previously to the giving of the antitoxin, an amount of the latter somewhat less than that of the fluid withdrawn should be given. A number of cases of "dry tap" have been observed in the disease by those so expert in spinal puncture as to'leave no room for doubt that the canal was properly entered. In such cases only a small amount of tetanus antitoxin should be injected (from 3 to 5 c.c). In brief, tetanus antitoxin should be used in precisely the same way as antimeningitis serum. Results of the Use of Antitoxin for Immunization. — The striking results which have been obtained, both in human and in veterinary practice, with the prophylactic injection of tetanus antitoxin, would seem to warrant the treating of patients with immunizing doses of serum — at least in neighborhoods where tetanus is not uncommon — when the lacerated and dirty condition of their wounds may indicate the possibility of a tetanus infection. 'Splendid results have followed this practice in many places. It is the custom at many dispensaries in New York City and elsewhere to immunize all Fourth-of-July wounds by injecting 1000 units. None of these have ever developed tetanus. Even the few cases of human tetanus reported as occurring after single injections of antitoxin prove the value of immunizing injections, for the mortality was low. They teach, however, that where tetanus infection is suspected the antitoxic serum should be given a second and even a third time at intervals of seven days. In the European armies it is compulsory to give a serum injection to every wounded soldier at the first possible moment after the injury. A second injection is given ten days later. Since these regula- tions were adopted there have been almost no cases of tetanus develop. CHAPTER XLVIII. THE BACTERIOLOGICAL EXAMINATION OF WATER, AIR, AND SOIL. THE CONTAMINATION AND PURIFICATION OF WATER. THE DISPOSAL OF SEWAGE. EXAMINATION OF WATER. The bacteriological examination of water is undertaken for the purpose of discovering whether, any pathogenic bacteria are liable to be present. The determination of the number of bacteria in water was for a time considered of great importance, then it fell into disrepute, and the attempt was made to isolate the specific germs of diseases which were thought to be water-borne. At first these attempts seemed very successful in that supposed typhoid bacilli and cholera spirilla were found. Further study revealed the fact that there were common water and intestinal bacteria which were so closely allied to the above forms that the tests applied did not separate them. When proper identification was carried out the very great majority of the suspected organisms were found to be non-pathogenic. The improbability of getting typhoid bacilli from suspected water except under unusually favorable conditions caused a return to the estimation of the number of bacteria in water and above all to the estimation of the number of intestinal bacteria. It is known that the group of colon bacilli have a somewhat longer existence than the typhoid bacilli, and as the colon bacilli come chiefly or wholly from the intestinal passages of men and animals, it was fair to assume that typhoid bacilli, dysentery or other pathogenic bacteria could not occur from fecal pollution without the presence of the colon bacillus except in rare cases. During the past few years the attention of sanitarians has been seriously devoted to the interpretation of the presence of smaller or larger numbers of colon bacilli in water, until at present, upon the quantitative analysis (measuring, within certain limits, decomposing organic matter) and the colon test (indicating more specifically pollution derived from intestinal discharges of man or animals) the bacteriological analysis of water is based. The determination of the number of bacteria is also of value. Technic for Quantitative Analysis. — ^The utmost care is necessary to get reliable results. A speck of dust, a contaminated dish, a delay of a few hours, an improperly sterilized agar or gelatin, a too high or too low temperature, may introduce an error or variation in results which would make a reliable test impossible. Collection of Samples. — ^The small sample taken must represent the whole from which it was drawn. If a brook water, it must be taken some distance from the bank; if from a tap, the water in the pipes BACTERIOLOGICAL EXAMINATION OF WATER 613 must first be run off, for otherwise the effect of metallic substances will invalidate the results; if from lake or pond, the surface scum or bottom mud must be avoided, but may be examined separately. The utensils by which the water is taken should be of a good quality of glass, clean and sterile. From a brook the water can be taken directly into a bottle, the stopper being removed while it fills, avoiding the surface fihn and its attending excessive numbers of bacteria; from a river or pond it can be taken from the bow of a small boat, or from a bottle properly fastened on the end of a pole so as to avoid con- tamination; from a well a special apparatus has been devised by Abbott, where a bottle with a leaded bottom is so held that when lowered to the proper depth a jerk will remove the cork and allow the bottle to fill. The same device or another accompUshing the same purpose can be rigged up readily by anyone. The sample of water should be tested as soon as possible, for the bacteria immediately begin to increase or decrease. In small bottles removed from the light, preda- tory microorganisms and many bacteria begin to increase, and among these are the members of the colon group. Thus, the Franklands record a case in which in a sample of well water kept during three days at a moderate temperature the bacteria increased from 7 to 495,000; while Jordan found that in a sample the bacteria in forty-eight hours fell from 535,000 to 54,500. In a sample we kept at room temperature the colon bacilli during twenty-fom- hours increased from 10 to 100 per c.c. The only safe way to prevent this increase is to plate and plant the water in fermentation tubes within a space of one or two hours or to keep it at a temperature under 5° C. (41° F.). If it cannot be kept cold, it is far better to make the cultures in the open field or in a house rather than to wait six to twelve hours for the conveniences and advantages of the laboratory. The third matter of great importance is the adding of proper amounts of water to the broth in the fermentation tubes and the media for plant- ing. Usually 1 c.c, 0.1 c.c, and 0.01 c.c. are added to the fermentation tubes and to 10 c.c of the melted nutrient agar or gelatin. If possible, duplicate tests should always be made. When it is desired to know whether colon baciUi are present in larger amounts than 1 c.c, quantities as great as 10 to 100 c.c. can be added to bouillon, and then after a few hours 1 c.c. added to fermentation tubes. Less than twenty colonies and more than two hundred on a plate give inaccxu-ate counts, the smaller number being too few to judge an average and the larger number interfering with each other. When as many as 10,000 colonies develop in the agar contained in one plate, it will be found that there will develop in a second plate containing but one-tenth the amount of water from 20 to 50 per cent, as many colonies. This shows that the crowding of the colonies had prevented the growth or caused a fusion of all but one-fifth to one-half of them. The chemical composition of the medium on which the bacteria are grown affects the result of the analysis. Nutrient 1.5 per cent, agar gives slightly lower counts than gelatin, but on account of its con- 614 BACTERIOLOGICAL EXAMINATION OF WATER venience in summer and its greater uniformity it is being more and more generally used for routine quantitative work. A uniform standard is a necessity to secure comparability of results. (See Media for Water.) At best only a certain proportion of bacteria develop, and it is only important that oiu* counts represent a section through the true bacterial flora which fairly represents the quick-growing sewage forms. Com- parability is the vitally essential factor. The temperature at which the bacteria develop is of great importance, and they should be protected from light. The access of oxygen which prevents the growths of anaerobes must also not be forgotten. As a rule, the plate cultures are developed at two temperatures, for four days at 20° to 21° C, and for twenty-four to forty-eight hours at incubator temperature. Some bacteria do not develop colonies in four days, but these are neglected. The number of bacteria growing at room tempera- ture is usually much greater than those growing at 37° C. As all the intestinal groups of bacteria grow at body temperature, while many of the water types do not, some investigators believe it important to develop the bacteria at both temperatures so as to compare the results. We have not found this to be of any advantage when tests are also made for the colon group of bacilli. The lactose broth, with indicator, is placed at 37° C. for growing the colon bacilli. The fermentation tubes not showing gas are recorded as negative and usually discarded. Those showing gas are suspected to contain colon bacilli. To a number of tubes containing melted litmus-lactose agar at about 44° C, are added 1, 0.1, and 0.01 loop of the culture fluid. Plates are poiu-ed and the whole placed in the incubator. The Bacillus coli ferments . lactose and thus produces acid, so that if colon bacilli are present we have a number of red colonies on a blue field. Later, if many colon bacilli were present, the whole medium becomes acid. At forty-eight hours, on account of alkali being produced by the formation of NH3, the blue may return. If after inspection red colonies are seen, four or five are picked and planted into lactose bouillon and other media. Litmus lactose agar is frequently used for the original plating of water samples, the absence or presence , of acid-producing colonies being thus immediately noted. The results of these tests are all more or less presumptive evidence of the bacilli belonging or not belonging to the intestinal colon bacillus group. Unfor- tunately there are some types of bacilli growing in the soil which resemble them. If it is necessary to be more accurate, the colon-like cultures should be subjected to the Vosges reaction (page 125), and should be kept for one month at 20° C. in gelatin before a decision is made. Colon bacilli do not liquefy gelatin nor give the Vosges reaction (page 125). There are a few colon-like bacilli in the intestinal tract that give the Vosges reaction. For a more complete understanding of the technic and the interpretation of results of the bacteriological examination of water see Elements of Water Bacteriology, by Prescott and Winslow. For the characteristics of the colon bacilli the Massachusetts State Board of Health uses six media— gelatin, lactose agar, dextrose broth. BACTERIOLOGICAL EXAMINATION OF WATER 615 milk, nitrate solution, and peptone solution, determining, respectively, absence of liquefaction, production of gas turbidity, coagulation with- out liquefaction of the coagulum, production of nitrite, and of indol. Lactose-bile-peptone solution has been much used. In badly con- taminated waters this has a distinct advantage in that ,the bile inhibits many varieties of bacteria more than those of the colon-typhoid group. In good waters the results are very similar from the lactose-peptone and lactose-blue-peptone solutions. Significance of the Colon Bacillus. — ^The colon test has been received by the majority of engineers and practical sanitarians with great satis- faction, and has been applied with confidence to the examination not only of water, but of shell-fish and other articles of food as well. On the other hand, some have denied its value. Bacteriologists have found bacilli like certain members of the colon group in apparently unpolluted well water. The discovery that animals have colon bacilli identical in the usual characteristics studied with those of man has complicated matters. Thus a fresh hillside stream may be loaded with colon bacilli from the washings of horse or cow manure put on the fields through which it runs or polluted by a stray cow or horse. Swine, hens, birds, etc., may contaminate in unsuspected ways. The number of colon bacilli rather than their presence in any body of surface water is therefore of importance. In well and spring water the presence of colon bacilli indicates contamination. The absence of colon bacilli in water proves its harmlessness so far as bacteriology can prove it. When the colon bacillus is present so as to be isolated from 1 c.c. of water in a series of tests, it is reasonable proof of animal or human pollution and the conditions should be investigated. Ten colon bacilli in 1 c.c. indicates serious pollution. Surface" waters from inhabited regions will always contain numerous colon bacilli after a heavy rain, storm, or shower. The washings from roads and cultivated fields con- tain necessarily large numbers. Winslow reports that in only two out of fifty-eight samples of presumably non-polluted well water did he get colon bacilli in the 1 c.c. samples. Even in twenty-one stagnant pools he only found colon bacilli in five of the 1 c.c. samples. The experience of all who have studied the -subject practically is that in delicacy the colon test surpasses chemical analysis: in con- stancy and defihiteness it also excels the quantitative bacterial count. All these tests must, however, be supplemented by inspection. Interpretation of the Quantitative Analysis. — The older experi- menters attempted to establish arbitrary "standards by which the sanitary quality of water could be fixed automatically by the number of germs alone. This has been largely given up. Dr. Sternberg con- siders that a water containing less than 100 bacteria is presumably from a deep source and uncontaminated by surface drainage; that one with 500 bacteria is open to suspicion; and that one with over 1000 bacteria is presumably contaminated by sewage or surface drain- age. Even this conservative opinion must be applied with caution. The source of the sample is of vital importance in the interpretation; 616 BACTERIOLOGICAL EXAMINATION OF WATER thus, a bacterial count which would condemn a spring or well might be normal for a river. In woodland springs and lakes several hundred bacteria per cubic centimeter are frequently found. In lakes the point at which the sample is taken is of great importance, as the bacterial count varies with the distance from the shore and with the depth. The weather also is an influence, since the wind causes currents and waves which stir up the bottom mud, bringing up organisms which have been sedimented. Rains greatly influence streams by flooding them with surface water, bringing a huge number of bacteria at times. The season of the year is an important factor. The counts are highest in the winter and spring months, and lower from April to September. The following figures illustrate this point: Water. _ Observers. Year. Jan. Feb. Mar. April. May. June New York City tap water . Houghton 1904 890 1100 650 240 350 370 New York City .... Noble 1916 13 17 50 33 6 13 Boston tap water . Whipple 1892 135 211 102 52 53 86 Merrimac River tap water Clark 1899 4900 5900 6300 2900 1900 3500 The winter and spring increases are not exceptions to the rule that high numbers indicate danger, but an indication of its truth; for it means a melting of the snow and a flow of surface water into the streams without the usual flltering soil filtration. A number of severe epidemics of typhoid fever have been produced in this way. It is only the fact that typhoid fever is at a minimum in winter that prevents more frequent pollution. Although, as a rule, a series of tests are necessary to pass judgment on a water, a single test may be very important. A large increase in the number in tap water a day after a storm points unerringly to surface pollution, and if towns exist in the water-shed, to street and sewer pollution. The Croton water frequently jumps from hundreds to thousands after such a storm. The low bacterial content in 1916 is due to the addition of chlorine to the water after its leaving the reservoir. In a typhoid epidemic at Newport, Winslow reports that a test of the water supply showed but 334 bacteria per cubic centimeter, but one from a well showed 6100. The suspicion aroused was justified by finding all the typhoid cases had gotten water from this well. The study of the bacterial efiluent from municipal water filters is the only way in which the efficiency of the filter and the accidents which occur can be determined. In Germany these regular tests are obligatory. The filter should remove about 99 per cent, of the bacteria. Elaborate studies have recently been made of the exact distribution of streams of sewage in bodies of water into which they flow, their disappearance by dilution and sedimentation, and their removal by death. Under peculiar conditions bacteria in water may increase for a time, but here the prevailing bacteria belong almost exclusively to one type. Streptococci in Sewage. — The varieties of streptococci found most often in polluted water correspond to the streptococci described by Houston. In some water in which these are found no B. coli have CONTAMINATION AND PURIFICATION OF WATER 61.7 been found and there is considerable doubt in such cases as to whether the streptococci imply serious pollution. The streptococci remain alive longer than the colon bacilli. In England the examination for streptococci in water is being less regularly done than formerly. Other Bacteria. — ^Most of the bacteria which develop in the intestines of man and animals necessarily occur in polluted water, and an examina- tion for some of these has been advocated by many, such as the B. {enteritidis) sporogenes, other anaerobic spore formers, the various members of the typhoid-colon group, and the proteus group. Isolation of the Typhoid Bacillus from Water. — If it were possible to readily obtain the typhoid bacilli from water, when they were present in small numbers, its examination for that purpose would be of much greater value than it is now; but we have to remember that we can only examine at one time a few cubic centimeters of water by bacterio- logical methods, and that although the typhoid bacilli may be suflSciently abundant in the water to give, in the quantity that we ordinarily drink, a few bacilli, yet it must be a very lucky chance if they happen to be in the small amount which we examine. Still further, although it is very easy to isolate typhoid bacilh from water when they are in considerable mmibers, yet when they are a very minute proportion of all the bacteria present it is almost impossible not to overlook them. Many attempts have been made to devise some method by which the relative number of the typhoid and other parasitic bacteria present in water could be iacreased at the expense of the saprophytic bacteria. Thus, to 100 c.c. of water 25 c.c. of a 4 per cent, peptone nutrient bouillon is added, and the whole put in the incubator at 38° C. for twenty-fom- hours. From this, plate cultures are made. As a matter of fact, the typhoid bacillus is rarely found, even in specimens of water where we actually know that it is or has been present because of cases of typhoid fever which have developed from drinking the water. From these facts we must consider our lack of finding the bacillus in any given cases as absolutely no reason for considering the water to be free from danger. Another serious drawback to the value of the examina- tions for the typhoid bacillus is that they are frequently made at a time when the water is really free from contamination, though both earlier and later the bacillus was present. It is hardly worth while, therefore, except in careful experimental researches, to examine the water for the typhoid bacillus, but rather study the location of the surrounding privies and sources of contamination. A number of ob- servers, resting on the agglutination test, have thought they have isolated typhoid bacilli from the soil and water, but these investigators had not considered suflBciently the matter of group agglutinins, and their results are not trustworthy. CONTAMINATION AND PURIFICATION OF DRINKING WATERS. Brook water and river water are contaminated in two ways: through chemicals, the waste products of manufacturing establishments, and 618 BACTERIOLOGICAL EXAMINATION OF WATER through harmful bacteria by the contents of drains, sewers, etc., the latter method being far the more dangerous. When water, which has been soiled by waste products of manu- factories only, becomes so diluted or purified that the contamination is not noticeable to the senses and shows no dangerous products on chemical analysis, it is probably safe to drink. When sewage is the contamination, this rule no longer holds, and there may be no chemical impurities and no pathogenic bacteria found, and yet disease be pro- duced. That river water which has been fouled by sewage will by oxidization, dilution, sedimentation, action of sunlight, and predatory microorganisms, become greatly pm'ified is an indisputable fact. The increase in bacteria which occurs from contamination is also largely or entirely lost after ten to twenty miles of river flow. Nevertheless, the history of many epidemics seems to show that a badly contaminated river is never an absolutely safe water to drink, although with the lapse of each day it becomes less and less dangerous, nor will sand filter-beds absolutely remove all danger. These statements are founded upon the results of numerous investigations; thus the marked disappearance of bacteria is illustrated by the following: Kummel found below the town of Rosbock 48,000 bacteria to the cubic centimeter; twenty-five kilo- meters farther down the stream only 200 were present — about the same number as before the sewage of Rosbock entered. On the other hand, the doubtful security of depending on a river purification is proved by such experiences as the following: In the city of Lowell, Massa- chusetts, an alarming epidemic followed the pollution of the Merrimac River three miles above by typhoid feces, and six weeks later an alarming epidemic attacked Lawrence, nine miles below Lowell. It is estimated that the water took ten days to pass from Lowell to Lawrence and through the reservoirs. Typhoid bacilli usually die in river water in from three to ten days, but they may live for twenty-five. days in other water; the Lawrence epidemic is easily explained. Newark-on-Trent, England, averaged 75 cases a year from moderately well-filtered water and only 10 when it was changed to deep-well supply. Purification of Water on a Large Scale. — For detailed information on this subject the reader is referred to works on hygiene. Surface waters, if collected and held in sufficiently large lakes or reservoirs, usually become so clarified by sedimentation, except shortly after heavy rains, as to require no further treatment so far as its appearance goes. The collection of water in large reservoirs allows not only the living and dead matter to subside, but allows time also -for the patho- genic germs to perish through the influence of light as well as of antagonistic bacteria and other deleterious influences, such as sand, or mechanical coagulant. Filtration of water exerts a very marked puri- fication, taking out 99.8 per cent, of the organisms in those best con- structed and at least 95 per cent, in those commonly used in cities. The construction of filters is too large a subject to enter on minutely here; sand filters consist, as a rule, of several layers, beginning with fine sand, and then smaller and larger gravel, and filially rough stones. CONTAMINATION AND PURIFICATION OF WATER 619 A certain time elapses before the best results are obtained; this seems to wait for the formation of a film of organic material on the sand, which is full of nitrifying bacteria. Even the best filters, only greatly diminish the dangers of polluted water. Spring water and well water are, in fact, filtered waters. Water which is subject to serious pollution must be submitted to a preliminary purification before it can be considered a suitable source for a drinking-water supply. The means employed for its purification depend to a large extent upon the character of the water and the nature of the pollution. Filtration through slow sand filters, three to five feet in depth, removes 98 to 99.5 per cent, of the bacteria and organic matter; so that effluents from the best constructed sand filtration beds constitute safe and reliable drinking waters. Five hundred thousand to one or two million gallons, depending somewhat upon the extent of pollution and the fineness of the sand, can be filtered daily per acre. Only the siu-face of the sand filter becomes in any way clogged and as thin a layer as can be scraped off is removed one or more times a month. This surface sand is washed with clean water and several scrap- ings replaced at one time. Sand filtration beds are very widely used abroad and are coming into extensive use in this country. The filter- beds at Lawrence, Mass., have been used over ten years with marked success; when properly managed, they render the highly polluted Merrimac River a fairly safe drinking water; the filter-beds there are scraped about thirteen times a year. Mechanical filtration plants find considerable favor where clarifica- tion as well as bacterial purification is desired. A coagulant such as sulphate of aluminum is employed and forms in the water a flocculent precipitate which carries down with it all suspended matter; 125,000,000 or more gallons of water can be filtered on an acre daily, but the filters must be washed daily by reversing the flow and cleansing the clogged filter with a stream of the pm-ified water. Chlorinated lime when added to drinking water to the extent of one-eighth to one-twelfth of a grain per gallon will destroy all intestinal bacteria of the typhoid- colon group within a few hours. This is a very useful means of purifi- cation. It does not injure the water and is being used very extensively. Under special conditions other methods, such as the passage of ozone, have proved successful. Domestic Purification. — ^Water which requires private filtering should not be supphed for drinking purposes. Unhappily, however, it often is. Domestic filters may be divided, roughly, into those for high and low pressure. The former are. directly connected with the water main, while the others simply have the slight pressure of the column of water standing in the filter. Many household filters contain animal charcoal, silicated carbon, etc., either in a pressed condition or in one porous mass. These filters remove much of the deleterious matter from the suspected waters, but the majority cannot be depended upon to remove all bacteria. Even those which are equipped for self-cleansing become foul in a little while, and, if not cleaned, unfit for use. The best of the 620 BACTERIOLOGICAL EXAMINATION OF WATER filters are of porous stone, such as the Poulton-Berkefeld and Pasteur filters. These yield a water, if too great a pressure is not used, almost absolutely free from bacteria, and if they are frequently cleansed they are reUable. A large Berkefeld filter will allow sixty gallons of water to pass per hour. The Pasteiu- filter is more compact and slower. From the best Pasteur filters sterile water may be passed for two or three weeks; from the Berkefeld usually only a few days. A single typical low-pressure filter is that of Bailey Denton. The upper compartment contains the filtering material, which may be sand or charcoal, and is fed from a cistern or hydrant. After a certain quantity of water has passed in, the supply is automatically cut off until the whole amount is filtered. A fairly efl&cient filter is the following: Take a large-sized earthenware pot and plug the hole in the bottom with a cork, through which pass a short glass tube. Upon the bottom place an inch of small pieces of broken flower-pot; upon this a couple of inches of well-washed small gravel, and upon this six to twelve inches of well-washed fine, sharp sand. Cover the sand with a piece of filter-paper and hold this down with a few small stones. Mount the pot on a tripod, and it is ready for use. The paper prevents the sand being distiu-bed when water is added, and as it also holds most of the sediment, this can be readily removed. Every few months the sand can be washed and replaced. Animal char- coal is not a good substance for permanent filters, as bacteria grow well in it. Whenever water is suspected, and there' is any doubt as to the filters, it should be boiled for ten minutes; this wUl destroy all bacteria. This precaution should always be taken in the presence of typhoid fever and cholera epidemics. THE DISPOSAL OF SEWAGE. The disposal of sewage is becoming a vital question with all towns and cities which are not situated near salt-water outlets, since the present tendency in legislation is to compel such towns to dispose of their waste so that it shall not be a menace to drinking-water streams, destructive to fisheries, or a nuisance to harbors. Methods of- sewage purification depend upon the character of the sewage and the kind of effluent desired. Two hundred thousand gallons of crude sewage may be filtered upon an acre of land daily and an effluent obtained which will compare favorably in every way known to the chemist and bacteriologist with the best mountain springs. This is, however, a slow process, and it is rare that such a piu-e effluent is required. Similar results may be obtained by utilizing the septic-tank method, running the sewage from the septic tank to' contact beds and thence to sand filter-beds; where, because of the partial "self-purification of the sewage" in the septic tank and contact beds, 2,500,000 gallons of sewage can be filtered daily on an acre of surface. In this process less land is required and both these effluents can be safely turned into drinking-water streams. If, however, a merely non-putrescible effluent is required, one which. BACTERIA IN AIR— EXAMINATION OF AIR 621 though high bacterially, will not be offensive in any way, or subject to further decomposition, it "may be obtained by passing crude sewage to septic tanks, thence to double contact beds, the resulting effluent having merely an earthy, humus-like odor and being non-putrescible. Where acid wastes, tannery wastes, dyestuffs, etc., from various factories enter into sewage, its disposal becomes a more complicated problem and chemical precipitation by the use of lime or other chemicals is generally employed for such sewage purification, which at best is only partial and is sometimes supplemented by sand filtration. Sea Water. — This is only feebly bactericidal. The salty tidal waters of rivers allow typhoid bacilli and other members of the typhoid-colon group to live for a number of days. BACTERIA IN AIR— EXAMINATION OF AIR. Saprophytic bacteria are always present in considerable numbers in the air except far out at sea or on high mountains. They reach the air from the earth's surface and are most plentiful nearest to it. They are more abundant where organic matter abounds and in dry and windy weather. The air is kept constantly in motion by winds so that fine particles are constantly being carried into it from the ground, especially in an inhabited area with its dusty streets. The rays of sunlight visibly reveal these particles to us. The bacteria in the dust of the fields and streets are carried along with these dust particles. They are usually the harmless soil bacteria or the almost equally harmless intestinal bacteria of animals. After a storm few bacteria are in the air, while on a dry windy day many thousands exist in a cubic meter. In warm weather rain carries down the bacteria of the air. The number of bacteria in the air of the country are much less than in the city air. Forests decrease the number of bacteria. On high mountains and on the sea far from land, bacteria are very scarce. The bacteria that multiply in the soil of street and country are almost entirely saprophytic types. Sunlight and drying rapidly destroy bacteria. In dwellings the bacterial content depends on many factors, of which the chief are the opening of windows to the outside dust-laden air, the cleanliness of the dwelling, and the amount of stirring up of the dust by sweeping. It is almost impossible to separate the effect of the bacteria which we inhale from that of the dust particles which they accompany. . Both probably act as slight irritants and so predispose to definite infections. Except in the air in direct contact with men and animals pathogenic bacteria are present in the air only under exceptional circum- stances and usually as spores, such as those of anthrax bacilli from the dust from the wool and hides of infected animals or of tetanus bacilli from tlie infected manure. The practical results obtained from the examination of air for pathogenic bacteria have been slight. We know that at times they must be in the air, but unless we purposely increase 622 BACTERIOLOGICAL EXAMINATION OF WATER their numbers they are so few in the comparatively small amount of air which it is practicable to examine that we rarely find them. It is estab- lished that in loud speaking, in coughing and sneezing many bacteria from the larynx, fauces and lips are expelled. Examination of dust in hospital wards and sick rooms, in places where only air infection was possible, has occasionally revealed tubercle bacilli and other pathogenic bacteria. Although very light they generally settle to the ground. It is now thought that the factor of air infection in the spread of infectious diseases is of very little importance, except when recent infection has occurred. With the tiny droplets of mucus most of these bacteria die quickly and do not disseminate, while alive, far from the place of their origin. The simplest method of searching for the varieties of bacteria in the air and their number in any place is to expose to the air for longer or shorter periods nutrient agar spread upon the surface of the Petri dish. After exposure the plates are either put in the incubator at 37° C, or kept at room temperature. When it is desired to obtain the pathogenic bacteria a little rabbit's blood is smeared over the agar. The more careful quantitative examination is made by drawing a given quantity of air through tubes containing sterile sand, which is kept in place by pieces of metal gauze. When the operation is completed the sand is poured into a tube containing melted nutrient gelatin or nutrient agar, and after thorough shaking, the mixture is poured into a Petri dish and the bacteria allowed to develop, either at 37° or 23° C, according as the growth of the parasitic or saprophytic varieties is desired. Instead of agar or gelatin, ascitic broth or animals may be inoculated. Such examinations are occasionally made of the air of theatres, crowded streets in cities, etc. They give the number of non- pathogenic bacteria only. BACTERIOLOGICAL EXAMINATION Or THE SOIL. The subject from its agricultural side is considered in Chapter LI. Specimens of deep soil can be gathered in sterile, sharp-pointed, sheet- iron tubes. Through the examination, we wish to learn either the number of bacteria or the important varieties of bacteria present. To estimate the number, small fractions of a gram are taken and planted in nutrient agar or in special media contained in Petri dishes. Anaerobic as well as aerobic cultures should be made. According to Houston, uncultivated sand soil averages 100,000 bacteria per gram, garden soil 1,500,000, and sewage-polluted 115,000,000. The most important bacteria to be sought for are bacilli of the colon group and streptococci. Both of these suggest fairly recent excremental pollution. Tetanus spores may also be found. The period during which typhoid bacilli remain alive in soil is variable, since it depends on so many unknown factors and difFers so in different places. The typhoid bacilli probably rarely increase in the soil and prob- ably rarely survive a month in it. The main danger of soil bacteria is their being washed into water supplied by rains or carried to them by the wind. CHAPTER XLIX. THE BACTERIOLOGY OF MILK IN ITS RELATION TO DISEASE. Feom the stand-point of the dairy many of the different varieties of bacteria found in milk are of importance which have Httle or no medical interest. We have space here only to consider the bacteri- ology of milk so far as it is related to health and disease. The fermented milks were considered in Chapter XXI. The saprophytic bacteria taken collectively have importance because one can determine from their number something as to the care taken in handling the milk and also because, when numerous, they produce chemical changes in the milk which are harmful for infants. The bacilli of the colon group are of little more importance than any of the saprophytic types, as in the case of milk they simply indicate pollution from the cow which is no more or less harmless than other forms. Numerical Estimation of Bacteria. — ^The number of bacteria in a cubic centimeter of milk is usually estimated from the colonies developing in nutrient agar plate cultures during a period of forty-eight hours when kept at a temperature of 37° C. This allows in market milk in which bacteria have developed at low temperature only a certain proportion of the varieties of bacteria to develop colonies. Sometimes fully twice as many colonies develop at 20° to 27° C. during three to five days as at 37° C. for two days. The advantages of the shorter time and of uniformity have led to the adoption of the technic given in Part I. Any cultiu^e method necessarily underestimates their number, as many of the bacteria remain after vigorous shaking in pairs or small groups. In order to overcome this and also to note the morphological types, the direct microscopic examination of smears of the sediment has been urged. A great practical objection to this is that, if a heated milk is examined, the dead as well as the living bacteria are counted, for no satisfactory method has yet been found by means of stains to differentiate between them. It is true that many varieties of bacteria stain less intensely after heating but others do not. This method has, however, great advantages at the creamery or farm in that one can immediately tell whether a sample has few or many bacteria and also note the presence of streptococci and leukocytes. The microscopic count gives from two to ten times as many bacteria as the routine culture method when the individual bacteria are counted. When clumps of bacteria are given the same value as single bacteria the microscopic counts agree closely with the colony counts. Smear Method for Direct Examination of Milk. — The Prescott-Breed method is the most accurate. That of Slack is also used. (See below.) 624 BACTERIOLOGY OF MILK IN RELATION TO DISEASE Advantages and Disadvantages of the Direct or Microscopic Method of Milk Analysis. — Advantages. — 1. The counting of each organism seen, while difficult and tune-consuming, makes it possible to ascertain approximately the actual number of bacteria in milk. The counting of the clumps as one organism approximates the colony count, but loses the value of an exact estimation. 2. It eliminates the greater portion of glassware required at present in the plate method, the preparation of agar, etc., which lessens the initial cost. These advantages, however, accrue mainly to the dealers employing a bacteriologist to make the milk analyses at the source. Disadvantages. — ^The disadvantages of the direct method are: 1. The very small quantity of milk used for counting leads to inaccuracy. 2. That as a resultant of the foregoing facts large factors have to be used in estimating the bacterial content of milk, with the consequent introduction of large factors of error. 3.. In estimating the bacterial content at all accurately of a very good milk, large numbers of fields have to be examined with a consequent expenditure of time, which increases the cost. 4. The individuality of the counter enters more largely into this method than in the plate method, and a better trained bacteriologist is necessary. Samples graded on colony count, bacterial content per c.c. No. of samples. Average plate count. Average micro- scopic count of individual organisms. Ratio. Less than 50,000 . 50,000 to 300,000 . 300,000 to 1,500,000 . 1,500,000 to 5,000,000 . 5,000,000 or over '. . Total .... . 594 . 133 42 13 8 . 790 14,491 103,075 ■ 648,452 2,538,461 6,575,000 171,181 92,034 708,681 5,254,018 21,541,534 27,519,531 1,100,997 6.29 6.87 8.1 8.48 4.18 6.43 Prescott and Breed explain their smear method briefly as follows: The sample of milk to be examined is shaken thoroughly and 0.01 c.c. is withdrawn by means of a specially constructed capillary pipette. The milk so obtained is spread evenly over an area of one square centi- meter on an ordinary glass slide. These areas may be easily determined by placing the glass slide over paper or glass on which areas of this size have been accurately ruled out. More satisfactory results can be obtained by using circular instead of square areas. Duplicate smears should be made on the same slide. The milk is then dried with gentle heat, the fat dissolved out with xylol or other fat solvent, the smear again dried, immersed in alcohol for a few minutes to fix the film, again dried and stained with methylene blue or other stain. Alkaline stains or others which attack the casein and loosen the smear must be avoided. The glassware used must be cleansed but need not be sterilized, as the bacteria have no chance to increase in number. The counting of the bacteria is done by a microscope and an oil- immersion lens. If the diameter of the field is so adjusted, by means of the draw tube, that it equals 0.16 mm., then each field of the micro- DIRECT OB MICROSCOPIC METHOD OF MILK ANALYSIS 625 scope covers approximately one-five-thousandth (0.0002) of a square centimeter. On this basis each bacterium seen in a field taken at ran-, dom represents 500,000 per c.c. if they are evenly distributed. However, it is impossible to distribute them evenly so that at least 110 fields of the microscope should be counted when there are many bacteria, 50 fields when the number is moderate and 100 fields if accurate results are required. The total number of bacteria seen in 10 fields multiplied by 50,000 or the total number seen in 100 fields multiplied by 5000 gives approximately the total number of bacteria per cubic centimeter. The counts vary considerably when individual bacteria are counted, even when made by the same examiner, because of one or more large clumps being in one series of fields and not in another. Even when clumps are counted as one the distribution varies in the different fields. COMPAKISON BETVEEN MICROSCOPIC CoUNTS BY DiFFEBENT EXAMINERS. Examiner. A Count of individual bacteria 164,000 Count of clumps . . . 65,000 B Count of individual bacteria 29,000 Count of clumps . . . 8,000 C Count of individual bacteria 133,000 Count of clumps . . . 38,000 Varieties. — Bacteria in milk can be divided into two great groups — those which get into the milk after it leaves the udder and those which come from the cow. The first group comprises bacteria from dust, hands, milking pails, strainers, etc. The extraneous bacteria are of importance because they indicate the conditions under which the milk was produced and cared for and because they produce changes in the chemical composition of the milk when they have developed in great numbers. The number of bacteria in any sample of milk depends on three factors: the number deposited in the milk from the cow's udder, from the air, and utensils; the time during which they have developed, and the temperature at which the milk has stood. The last is perhaps the most important factor. The attempt was made during a period of one year to connect illness in infants and children with special varieties of saprophytic bacteria in milk, but with negative results. From the milks altogether 239 varieties of bacteria were isolated and studied. These 239 varieties, having some cultural or other' differences, were divided into 31 classes, each class containing from 1 to 39 more or less related organisms. As to the sources of bacteria found in milk, we made sufficient experi- ments to satisfy us that they came chiefly from outside the udder and milk ducts. Bacteria were isolated from various materials which under certain conditions might' be sources of contamination for the milk, and the cultures compared with those taken from milk. Thus there were obtained from 20 specimens of hay and grass, 31 varieties of bacteria;; 40 626 BACTERIOLOGY OF MILK IN RELATION TO DISEASE from 15 specimens of feces, manure, and intestinal contents, 28 varieties; from 10 specimens of feed, 17 varieties. Of these 76 varieties there were 42 which resembled closely those from milk — ^viz., 11 from grass or hay; 26 from manure; 5 from feed. During the investigation a number of the varieties isolated from milk were shown to be identical with types commonly found in water. From the few facts quoted above and from many other observa- tions made during the course of the work it would seem that the term "milk bacteria" assumes a condition which does not exist in fact. The expression would seem to indicate that a few varieties, especially those derived in some way from the cow, are commonly found in milk, which forms having entered the milk while still in the udder or after its withdrawal, are so well fitted to develop in milk that they outgrow all other varieties. As a matter of fact it was found that milk taken from a nmnber of cows, in which almost no outside contamination had occurred, and plated immediately, contained, as a rule, very few bacteria, and these were streptococci, staphylococci, and other varieties of bacteria not often found in milk sold in New York City; the temperature at which milk is kept being less suitable for them than for the bacteria which fall into the milk from dust, manure, etc. A number of speci- mens of fairly fresh market milk averaging 200,000 bacteria per cubic centimeter were examined immediately, and again after twelve to twenty-fom- hoiu-s. In almost every test the three or four predominant varieties of the fresher milk remained as the predominant varieties after the period mentioned. The above experiments seem to show that organisms which have gained a good percentage in the ordinary commercial milk at time of sale will be likely to hold the same relative place for as long a period as milk is usually kept. After the bacteria pass the ten or twenty million counts a change occurs, since the increasing acidity inhibits the growth of some forms before it does that of others. Thus some varieties of the lactic acid bacteria can increase until the acidity is twice as great as that which inhibits the growth of many bacteria. Before milk reaches the curdling point, the bacteria may have reached over a billion to each cubic centimeter. For the most part specimens of milk from different localities showed a difference in the character of the bacteria present, in the same way that the bacteria from hay, feed, etc., varied. Even the intestinal contents of cows, the bacteriology of which might be expected to show common characteristics, contained, besides the predominating colon types, other organisms, which differed widely in different species and in different localities. Cleanliness in handling the milk and the temperature at which it had been kept were also found to have a marked influence on the predominant varieties of bacteria present. Pathogenic Properties of the Bacteria Isolated.— Intraperitoneal injection of 2 c.c. of broth or milk cultures of about 40 per cent, of the varieties tested caused death. Cultures of most of the remainder STREPTOCOCCI IN RELATION TO DISEASE 627 produced no apparent deleterious effects even when injected in larger amounts. The filtrates of broth cultures of a number of varieties were tested, but only one was obtained in which poisonous products were abundantly present. Death in guinea-pigs weighing 300 grams followed within fifteen minutes after an injection of 2 c.c. ; 1 c.c. had little effect. As bacteria in milk are swallowed and not injected under the skin, it seemed wise to test the effect of feeding them to very young animals. We therefore fed forty-eight cultures of 139 varieties of bacteria to kittens of two to ten days of age by means of a glass tube. The kittens received 5 to 10 c.c. daily for from three to seven days. Only one culture produced illness or death. Very young guinea-pigs were fed in the same manner with similar results. After five years of effort to discover some relation between special varieties of bacteria found in milk and the health of children the con- clusion has been reached that neither through animal tests nor the isolation from milk given sick infants have we been able to establish such a relation. Pasteurized or "sterilized" milk is rarely kept longer than thirty-six hours, so that varieties of bacteria which after long standing develop in such milk did not enter into oiu" problem. The harmlessness of cultures given to healthy young kittens does not, of course, prove that they would be equally harmless in infants. . Even if harmless in robust infants, they might be injurious when simimer heat and previous disease had lowered the resistance and the digestive power of the subjects. Streptococci in Relation to Disease. — In an investigation by Dr. D. H. Bergey connection between diarrhea and pus and streptococci was sometimes found. The results of this investigation appear to warrant the following conclusions : 1. The occurrence of an excessive number of leukocytes in cows' milk is probably always associated with the presence in the udder of some inflammatory reaction brought about by the presence of some of the ordinary pyogenic bacteria, especially of streptococci. 2. When a cow's udder has once become infected with the pyogenic bacteria, the disease tends to persist for a long time, probably extending over several periods of lactation. 3. Lactation has no causative influence per se upon the cellular and bacterial content of cows' milk, though it probably tends toward the aggravation of the disease when the udder is once infected. It is impossible to differentiate in routine milk examinations the pathogenic streptococci of diseased cows from saprophytic varieties. Thus it happens that a milk which contains great numbers of strepto- cocci may or may not be more dangerous than one which contains an equal number of other apparently less harmful bacteria. The identi- fication of the varieties present requires great care and is only done in the face of feared infection such as an epidemic of septic throats. Those that produce human diseases, except in infants, are probably always from cows in which the udder has befen infected from human scrurces. 628 BACTERIOLOGY OF MILK IN RELATION TO DISEASE The Deleterious Effect of Live Bacteria in Milk on Infants.^We have tested this ourselves in the following way: During each of the summers of 1902, 1903, and 1904 a special lot of milk was modified for a group of fifty infants, all of whom were under nine months of age, and dis- tributed daily. To one half a portion of the milk was given raw; to the other half a portion heated at 60° for twenty minutes. The modified milk was made from a fairly pure milk mixed with ordinary cream. The bacteria contained in the milk numbered on the average 45,000 per cubic centimeter, in the cream 30,000,000. The modified raw milk taken from the bottles in the morning averaged 1,20Q,Q00 bacteria per cubic centimeter, or considerably less than the ordinary grocery milk; the pasteurized, about 1000; taken in the late afternoon of the same day they had, respectively, about 20,000,000 and 50,000. ' Twenty-one predominant varieties of bacteria were isolated from six specimens of this milk collected on different days. The varieties represented the type^ of bacteria frequently found in milk. The infants were selected during the first week in June, and at first all were placed on pasteurized milk. The fifty infants which had been selected were now separated into two groups as nearly alike as possible. On the fifteenth of June the milk was distributed without heating to one half the infants, the other half receiving as before the heated milk. In this way the infants in the two groups received milk of identically the same quality, except for the changes produced by heating to 165° F. for thirty minutes. The infants were observed carefully for three months and medical advice was given when necessary. When severe diarrhea occurred barley water was substituted for milk. The first season's trial gave the following results: Within one week 20 out of 27 infants put on raw milk suffered from moderate or severe diarrhea; while during the same time only 5 cases of moderate and none of severe diarrhea occurred in those taking pasteurized milk. Within a month 8 of the 27 had to be changed from raw back to heated milk, because of their continued illness; 7, or 25 per cent., did well all summer on raw milk. On the other hand, of those receiving the pasteiu-ized milk, 75 per cent, remained well, or nearly so, all summer, while 25 per cent, had one or more attacks of severe diarrhea. There were no deaths in either group of cases. During the second summer a similar test was made with 45 infants. Twenty-four were put on raw modified' milk; 13 of these had serious diarrhea,, in 5 of whom it was so severe that they were put back upon heated milk; 10 took raw milk all summer without bad effects; 2 died, 1 from gross neglect on the part of the mother, the other from diarrhea. Of the 21 on pasteurized milk, 5 had severe attacks of diarrhea, but all were kept on this milk except for short periods, when all food was omitted; 16 did well throughout the summer. One infant, markedly rachitic, died. The third summer's results have not been tabulated, but were similar to those of the first two tests. BACTERIA IN MILK 629 The outcome of the these observations during the first two summers are summarized in the following table : Kinds of milk. Number of bacteria when consumed. Number of infants. Remained well for entire summer. Number having severe or moderate diarrhea. Average number days off milk during summer. Average weekly gain in weight, oz. Average number of days diarrhea. Deaths. Pasteurized millc, 1000-50,000 bacteria per e.c Raw milk, 1,200,000-20,000,000 bacteria per c.c 41 51' 31 17 10 33 3.0 5.5 4.0 3.5 3.9 11.5 1 2 Although the number of cases was not large, the results, almost identical during the three summers, indicate that even a fairly pure milk, when given raw in hot weather, causes illness in a much larger percentage of cases than the same milk given after pasteurization. A considerable percentage of infants, however, do apparently quite as well on raw as on pasteurized milk. Bacteria in Milk. Effect on Older Children. — The children over three years of age who received unheated milk, containing at different times from 145,000 to 350,000,000 bacteria per cubic centimeter, showed almost no gastro-intestinal disturbance. The conditions at three institutions will serve as examples. In the first of these an average grade of raw milk was used which, during the summer, contained from 2,000,000 to 30,000,000 bacteria per cubic centimeter. This milk was stored in an ice-box until required. It was taken by children unheated and yet no case of diarrhea of suf- ficient gravity to send for a physician occurred during the entire summer. This institution was an orphan asylum containing 650 children from three to fourteen years of age — viz., three to five years, 98; five to eight years, 162; eight to fourteen years, 390. A second institution used an unheated but very pure milk which was obtained from its own farm. This milk averaged 50,000 bac- teria per cubic centimeter. The inmates were 70 children of ages ranging from three to fourteen years. In this institution not a single case of diarrheal disease of any importance occurred during the summer. In a third institution an average grade of milk was used which was heated. This milk before heating contained 2,000,000 to 20,000,000 bacteria per cubic centimeter. This institution was an infant asylum in which there were 126 children between the ages of two and five years. There were no cases of diarrhea during the summer. These clinical observations taken in connection with the bacterio- logical examination at the laboratory show that although the milk may come from healthy cattle and clean farms and be kept at atem- 1 Thirteen of the fifty-one infants on raw milk were transferred before the end of the trial to pasteurized milk because of serioiis illness. If these infants had been left on raw milk it is believed by the writers that the comparative results would have been even more unfavorable to raw milk. 630 BACTERIOLOGY OF MtLtC tN BMLATWN fO DISeASS perature not exceeding 60° F., a very great increase in the number of bacteria may occm". Furthermore, this may occur without the accumulation in the milk of suflBcient poisonous products or living bacteria to cause appreciable injury in children over three years of age, even when such milk is consumed in considerable amount and for a period extending over several months. Milk kept at tempera- tures somewhat above 60° F. was not met with in our investigations, but the histories of epidemics of ptomain poisoning teach that such milk may be very poisonous. It is also to be remembered that milk abounding in bacteria on account of its being carelessly handled is also always liable to contain pathogenic organisms derived from human or animal sources. Results with Very Impure Milk Heated vs. Those with Pure or Average Milk Heated. — During the summer of 1901 we were able to observe a number of babies fed on milk grossly contaminated by bac- teria. In 1902 systematic supervision of all stores selling milk was instituted by the Health Department, so that the very worst milk was not offered for sale that suromer. The observations upon the impure milk of 1901 are of sufficient importance to be given in detail, although already mentioned in the report of the observations upon infants of both summers which were fed on "store milk." A group of over 150 infants was so divided that 20 per cent, were allowed to remain on the cheapest store milk which they were taking at the time. To about the same number was given a pure bottled milk. A third group was fed on the same quality of milk as the second, but sterilized and modified at the Good Samar- itan Dispensary. A fourth group received milk from an ordinary dau-y farm. This milk was sent to a store in cans and called for by the people. A few infants fed on breast and condensed milk were observed for control. In estimating the significance of the observations recorded in the tables, one should bear in mind that not only do different infants possess different degrees of resistance to disease, but that, try as hard as the physicians could, it was impossible to divide the infants into groups which secured equal care and were subjected to exactly the same conditions. It was necessary to have the different groups in somewhat different parts of the city. It thus happened that the infants on the cheap store milk received less home care than the aver- age, and that those on the pure bottled milk lived in the coolest portion of the city. Certain results were, however, so striking that their interpretation is fairly clear. It is to be noted that the number of infants included in each group is small. There is nothing in the observations to show that fairly fresh milk from healthy cows, living under good hygienic conditions and con- taining, on some days, when delivered, as many as 200,000 bacteria per cubic centimeter, had any bacteria or any products due to bac- teria that remained deleterious after the milk was heated to near the boiling-point. RESULTS WITH IMPURE AND PURE MILK HEATED 631 Table Showing the Results of Feeding during July and August, 1901, in Tenement Houses, of 112 Bottle-fed Infants under one year of Age, and of 47 Bottle- fed Infants between One and Two Years of Age with Milk from Different . Sources, and the Number of Bacteria Present in the Milk. Infants under one year. Infants over one year. Character of milk. 1^ Diarrhea. n 1 a 2"" Diarrhea. 1 1 1 CO 1 p 1. Pure milk boiled and modified at dispensary or stations; given out in smaU bottles. Milk be- fore boiling averaged 20,000 bacteria per c.c.; after boiling 2 per c.c. 2. Pure milk, 24 hours old, sent in in quart bottles to tenements, heated and modified at home, 20,600 to 200,000 bacteria per c.c. when delivered. 3. Ordinary milk, 36 hours old, from a sele(;ted group of farms, kept cool in cans during trans- port; 1,000.000 to 25,000,000 bacteria ptsr c.c, heated and modified at home before using. 4. Cheap milk, 36 to 60 hours old, from various small stores, de- rived from various farms, some fairly clean, some very dirty: 400,000 to 175,000,000 bacteria per c.c. before home heating. 5. Condensed milk of different brands. Made up with hot water. As given, contained bac- teria from 5,000 to 200,000 per c.c. 6. Breast milk 41 23 18 21 9 16 3 oz. 4Joz. 4 oz. ioz. } oz. 2ioz. 10 8 6 4 5 5 8 5 6 13 2 2 1' V 43 3 24 12 7 4 4ioz. 4 oz. i oz. 3Joz. 8 1 1 1 2 2 a 3 On the other hand, it is possible that certain varieties of bacteria may, under conditions that are insanitary, find entrance to milk and survive moderate heat or may develop poisonous products resistant to heat in sufficient amount to be harmful, even when they have accumulated to less than 200,000 per c.c. Turning now to the results of feeding with milk which has been heated and which before sterilization contained from 1,000,000 to 25,000,000 bacteria per cubic centimeter, averaging about 15,000,000, though obtained from healthy cows hving under fairly decent con- ditions and although the milk was kept moderp,tely cool in transit, we find a distinct increase in the amount of diarrheal diseases. Though it is probable that the excessive amount of diarrhea in this group of children was due to bacterial changes which were not neutralized by heat or to living bacteria which were not killed, yet it is only fair to consider that the difference was not very great and that the infants of 1 This infant died from enteritis and toxemia. 2 This infant died of pneumonia. There had been no severe intestinal disorder noted. =One of the four had pertussis, the remaining three died from uncomplicated enteritis. 632 BACTERIOLOGY OF MILK IN RELATION TO DISEASE this group were under surroundings not quite so good as those on the pure milk. Finally, we come in this comparisoji to the infants who received the cheap store milk after heating. This milk had frequently to be returned because it curdled when boiled, and contained, according to the weather, from 4,000,000 to 200,000,000 bacteria per cubic centi- meter. In these infants the worst results were seen. This is shown not only by the death-rate, but by the amount and by the severity of the diarrheal diseases, and the general appearance of the children as noted by the physicians. Although the average number of bacteria in the milk received by this group is higher than that received by the previous group, the difference in results between this group and the previous one can hardly be explained by the difference in the number of bacteria. The varieties of bacteria found in this milk were more numerous than in the better milk, but we were unable to prove that they were more dangerous. Probably the higher temperature at which the milk was kept in transit, and the longer interval between milking and its use, allowed more toxic bacterial products to accumulate. Bacterial Contamination of Milk — General Conclusions^ as to Rela- tive Importance.^ — 1. During cool weather neither the mortality nor the health of the infants observed in the investigation was appre- ciably affected by quality of the market milk or by the number of bac- teria which it contained. The different grades of milk varied much less in the amount of bacterial contamination in winter than in sum- mer, the store milk averaging only about 750,000 bacteria per cubic centimeter. 2. During hot weather, when the resistance of the children was lowered, the kind of milk taken influenced both the amount of illness and the mortality; those who took condensed milk and cheap store milk did the worst, and those who received breast milk, pure bottled milk, and modified milk did the best. The effect of bacterial contamination was very marked when the milk was taken without previous heating; but, unless the contamination was very excessive, only slight when heating was employed shortly before feeding. 3. The number of bacteria which may accumulate before milk becomes noticeably harmful to the average infant in summer differs with the nature of the bacteria present, the age of the milk, and the temperature at which it has been kept. When the milk is taken raw, the fewer the bacteria present the better are the results. Of the usual varieties, over 1,000,000 bacteria per cubic centimeter are certainly deleterious to the average infant. However, many infants take milk without apparently harmful results. Heat of 145° F. for thirty minutes or of 170° F. for a shorter period not only destroys most of the bacteria present, but, apparently, some of their poisonous products. No harm from the bacteria previously existing in recently heated ' These conclusions were drawn up by the writer in association with Dr. L. E. Holt, after a joint study of the results obtained in the studies above recorded. INFLUENCE OF TEMPERATURE UPON BACTERIA IN MILK 633 milk was noticed in these observations unless they had amounted to many millions, but in such numbers they were decidedly deleterious. 4. When milk of average quality was fed, pasteurized and raw, those infants who received milk previously heated did, on the average, much better in warm weather than those who received it raw. The difference was so quickly manifest and so marked that there could be no mistaking the meaning of the results. 5. No special varieties of bacteria were found in unheated milk which seemed to have any special importance, in relation to the sum- mer diarrheas of children. A few cases of acute indigestion were seen immediately following the use of pasteurized milk more than thirty- six hours old. Samples of such milk were found to contain more than 100,000,000 bacteria per cubic centimeter, mostly spore-bearing varieties. The deleterious effects, though striking, were neither serious nor lasting. 6. After the first twelve months of life infants are less and less affected by the bacteria in milk derived from healthy cattle and the air. According to these observations, when the milk had been kept cool, the bacteria did not appear to injiKe the children over three years of age at any season of the year, unless in very great excess. 7. While it is true that even in tenements the results with the best bottle feeding are nearly as good as average breast feeding, it is also true that most of the bottle feeding is at present very badly done; so that, as a rule, the immense superiority of breast feeding obtains. This should therefore be encouraged by every means, and not dis- continued without good and sufficient reasons. The time and money required for artificial feeding, if expended by the tenement mother to secure better food and more rest for herself, would often enable her to continue nursing with advantage to her child. Influence of Temperature upon the Multiplication of Bacteria in Milk. — Few, even of the well informed, appreciate how great a difference a few degrees of temperature will make in the rate of bacterial mul- tiplication. Milk rapidly and sufficiently cooled keeps almost unaltered for thirty-six hours, while milk insufficiently cooled deteriorates rapidly. The majority of the bacteria found in milk grow best at tem- peratures above 70° F., but they also multiply slowly even at 40° F.; thus, of 60 species isolated by us, 42 developed good growth at the end of seven days at 39° F. Our observations have shown that the bacteria slowly increase in numbers after the germicidal prop- erties of the milk have disappeared, and the germs have become accus- tomed to the low temperatiu-e. In fact, milk cannot be permanently preserved unaltered unless kept at 32° F. or less. The degree of cooling to which ordinary supplies of milk are subjected differs greatly in various localities. Some farmers chill their milk rapidly, by means of pipe coils over which the milk flows; others use deep wooden tanks filled with water into which the cans of milk are placed soon after milking. In winter these methods are very satisfactory, for the water runs into the pipes or tanks at about 38° F. In warmer weather they 634 BACTERIOLOGY OF MILK IN RELATION TO DISEASE are unsatisfactory, unless ice is used, as the natural temperature of the water may be as high as 55° F. A considerable quantity of milk is not cooled at all at the farms. It is sent to the creamery or railroad after two to six hours, and is then more or less cooled. These few hours in summer, when the milk is left almost at blood heat, allow an enormous development of bacteria to take place, as is shown in the table below. Table I. — Showing the Development of Bacteria in Two Samples of Milk Main- tained AT Different Temperatures for Twenty-four, Forty-eight, and Ninety-six Hours, Respectively. The First Sample op Milk was Obtained UNDER the best CONDITIONS POSSIBLE, THE SECOND IN THE UsUAL WaY. WhEN Received, Specimen, No. 1 Contained 3000 Bacteria per c.c, Specimen No. 2, 30,000 PER c.c. Time which elapsed before making test. Temperature. ^ N Fahrenheit. 24 hours. 48 hours. 96 hours. 168 hours. 32° 2400 2100 1850 1400 30,000' 27,000 24,000 19,000 39° 2500 3600 218,000 4,209,000 38,000 56,000 4,300,000 38,000,000 42° 2600 3600 500,000 11,200,000 43,000 210,000 5,760,000 120,000,000 46° 3100 12,000 1,480,000 80,000,000 42,000 360,000 12,200,000 300,000,000 50° 11,600 540,000 300,000,000 1,000,000,0002 89,000 1,940,000 1,000,000,000' 55° 18,800 187,000 3,400,000 38,000,000 60° 180,000 900,000 28,000,000 168,000,000 68° 450,000 4,000,000 500,000,000 1,000,000,0002 Observations on Bacterial Multiplication in Milk at 90° F., a Common in New York in Hot Summer Weather. Table II. — Number op Bacteria per c.c. Temperature Milk I. Milk II. Milk III. Fresh and of good Fair quality from Bad quality from quality. store. store. Original number 5,200 92,000 2,600,000 After two hours 8,400 184,000 4,220,000 After four hours . 12,400 470,000 19,000,000 After six hours . 68,500 1,260,000 39,000,000 After eight hours . 654,000 6,800,000 124,000,000 A sample of milk No. 1 removed after six hours and cooled to 50° F. contained 145,000,000 at the end of twenty-four hours. Some of this milk, kept cool from the beginning con- tained but 12,800 bacteria per cubic centimeter at the end of twenty-four hours. Pasteurization of Milk.— The two dominant factors which control the temperature and time at which the milk should be heated are (1) the thermal death-points of pathogenic bacteria, and (2) the ther- molabile food constituents of the milk. The first factor is almost equally important for milk used by persons of all ages, while the second factor is only important for milk used for very young children. The exposure of bacteria for a short time at a high temperature ' The figures referring to tests of the second sample are printed in heavy-face type. 2 These figures signify the maximum growth and are conservative estimates only. PasteVri^atIon of MiLli 635 IS equivalent to a longer time at a lower temperature. The ferments and other labile food constituents, on the other hand, are altered much more by the higher temperature. It is well, therefore, to choose the lowest possible temperature which will kill the non-spore-bearing patho- genic bacteria in a practicable length of time. Such an exposure is 60\C.i(140° F). for thirty minutes, 70° C. (158° F.) for five minutes. Very much shorter exposures, as one minute at 70° C, will kill the great majority of pathogenic and other bacteria in the milk and add much of safety, as seen in the tables below, but it is better to be on the safe side. Table Showing Effect of Heat tjpon Tubercle Bacilli in Milk. Degree of heat. Time exposed. Amount of milk. Result in guinea-piga. 60° C. 15.0 min. 1 c.c. Infection 60° C. 20.0 mm. 1 c.c. No infection 60° C. 30.0 min. 1 CO. No infection 70° C. . 5 min. 1 c.c. Infection' 70° C. 1.0 min. 1 0.0. No infection 70° C. 2.0 min. 1 c.c. No infection Control not heated 0.001 Infection This milk was infected by adding one-fifth of its quantity of sputum rich in tubercle baciUi. Development of Bacteria in Milk which has been Heated. — ^There is a common idea that bacteria develop much more rapidly in milk that has been heated than in raw milk. This is only true for freshly drawn milk which has slight bactericidal power. The table below shows the effect on bacteria in milk of heating to 70° C. for one-half and one minute. Not only the immediate reduction in number is seen to be great, but the difference continues when the milk is kept cold for two days. Two Samples Mixed from 100 Samples prom Inspectors. Pasteurized at 160° F. Plates Made Same Day. Sample I. Sample II. Raw milk 600,000 Raw milk .... 5,400,000 J minute pasteurized . 2,000 J minute pasteurized . 7,400 1 minute pasteurized . 1,000 1 minute pasteurized . 600 Same Samples Kept in Ice-box Twenty-four Hours at 45° F. (7° C). Raw milk 6,300,000 Raw milk .... 21,600,000 5 minujte pasteurized . 18,000 5 minute pasteurized . 12,000 1 minute pasteurized . 900 1 minute pasteurized . 3,600 In Ice-box Forty-eight Hours at 45° F. (7° C). Raw milk 16,200,000 Raw milk .... 63,000,000 I minute pasteurized . 120,000 | minute pasteurized . 276,000 1 miaute pasteurized . 10,000 1 minute pasteurized . 90,000 In Room at 71° F. (22° C). Raw milk 36,600,000 Raw milk .... 150,000,000 J minute pasteurized . 5,460,000 j minute pasteurized . 4,500,000 1 minute pasteurized . 5,400,000 1 minute pasteurized 3,600,000 ' Most of the guinea-pigs were not infected by the milk heated for one-half minute. 636 BACTERIOLOGY OF MILK IN RELATION TO DISEASE Number of Bacteria in Milk Produced under Different Conditions. 1. The number of bacteria present at the time of milking and twenty-four, fprty-eight, and seventy-two hours afterward in milk obtained and kept under correct conditions. No preservatives were present in any of the following specimens: Pure milk obtained where every reasonable means was taken to ensure cleanli- ness. The long hairs on the udder were clipped; the cows roughly cleaned and placed in clean barns before milking; the udders were wiped off just previous to milking; the hands of the men were washed and dried; the pails used had small (six-inch) openings, and were thoroughly cleaned and sterilized by steam before use. Milk cooled within one hour after milking to 45° F., and subsequently kept at that temperature. The first six specimens were obtained from individual cows; the last six from mixed milk as it flowed at different times from the cooler. Temperature of barns 55° F. Number of Bacteria in 1 c.c. of Milk. FROM SIX INDIVIDITAL COWS. 5 hours after milking. After 24 hours. After 48 hours. After 72 hours. 500 700 12,500 Not counted. 700 700 29,400 Not counted. 19,900 5200 24,200 Not counted. 400 200 8600 Not counted. 900 1600 12,700 Not counted. 13,600 3200 1933 19,500 Not counted. Average 6,000 17,816 FROM SAMPLES OP MIXED MILK OF ENTIRE HERD. 6900 12,000 19,800 494,000 6100 2200 20,200 550,000 4100 700 7900 361,000 1200 400 7100 355,000 6000 900 9800 445,000 1700 400 8700 389,000 Average 4333 2766 10,583 329,000 Twenty-five sam.ples taken separately from individual cows on another day and tested immediately averaged 4550 bacteria per cubic centimeter and 4500 after twenty-four hours. These twenty-five specimens were kept at between 45° and 50° F. 2. Milk taken during winter in well-ventilated, fairly clean, but dusty barns. Visible dirt was cleaned off the hair about the udder before milldng. Milkers' hands were wiped off, but not washed. Milk pails and cans were clean, but the straining cloths dusty. Milk cooled within two hours after milking to 45° F. Number of Bacteria in 1 c.c. of Milk. At time of milking. 12,000 13,000 21,500 Average 16,500 After 24 hours. 14,000 20,000 31,000 21,666 After 48 hours. 57,000 65,000 106,000 76,000 Number in City Milk. 3. The condition of the average raw city milk is very different, and is shown in the following tables. CLEANLINESS USED IN OBTAINING MILK 637 The twelve samples were taken late in March, 1912, by Inspectors of the Department of Health of New York City from cans of milk immediately upon their arrival in the city. Raw milk at present gives similar counts. The temperature of the atmosphere averaged 50° F. during the previous twenty-four hours. The temperature of the milk when taken from the cans aver- aged 45° F. Much of this milk had been carried over two hundred miles. From the time of its removal from the cans, which was about 2 a.m., until its distribu- tion in nutrient agar, at 10 a.m., the milk was kept at about 45° F. From New York and Hudson River Railroad. From Harlem Railroad. No. of bacteria No. of bacteria No. of sample. in 1 c.c. No. of sample. in 1 c.c. 50 -35,200,000 48 6,200,000 51 13,000,000 49 2,200,000 52 2,500,000 50 15,000,000 53 1,400,000 51 70,000 54 200,000 52 80,000 55 600,000 53 320,000 The results of the examination of many thousands of specimens last year indicated that most of the milk of grade A was of fair quality, but that the raw milk of the other grades contained excessive numbers of bacteria before pasteurization. It must be kept in mind that milk averaging 3,000,000 bacteria per cubic centimeter will, when kept at the temperature common in ■the homes of the poor who comprise the larger part of the population, soon contain very largely increased numbers and show its dangerous condition by turning sour and curdling. Cleanliness Used in Obtaining Milk, and Its Influence. — The present conditions under which much of the milk is obtained are not pleasant to consider. In winter, and to a less extent at other seasons of the year, the cows in many stables stand or lie down in stalls in the rear portion of which there is altogether too much manure and urine. When milked the hands of the milkers are not cleansed, nor are the under portions of -the cows, only visible masses of manure adhering to the hair about the udder being removed. Some milkers even moisten their hands with milk, to lessen friction, and thus wash off the dirt of their hands and the cow's teats into the milk in the pails. Some may regard it as ah unnecessary refinement to ask that farmers should thor- oughly clean the floors of their stalls once each day, that no sweeping should be done just before milking, and that the udders should be wiped with a clean damp cloth arid the milkers should thoroughly wash and wipe their hands before commencing milking. The pails and cans should not only be carefully cleansed, but afterward scalded out with boiling water. The washing of the hands would lessen the number of ordinary filth bacteria in the milk, and diminish risk of transmitting to milk human infectious diseases, like scarlet fever, diphtheria, and enteric fever, by the direct washing off of the disease germs from infected hands. It would also inculcate general ideas of the necessity of cleanli- ness and of the dahger of transmitting disease through milk. The value of cleanliness in limiting. the number of bacteria is demonstrated by the figures contained, in the tables.. ... 638 BACTERIOLOOY OF MILK IN RELATION TO DISEASE General Conclusions. — Because of its location and its hairy covering, the cow's udder is always more or less soiled with dirt and manure unless cleaned. On account of the position of the pail and the access of dust-laden air it is impossible to obtain milk by the usual methods without mingling with it a considerable number of bacteria. With suitable cleanliness, however, the number is far less than when filthy methods are used, there being no reason why fresh milk should contain in each cubic centimeter, on the average, more than 12,000 bacteria per cubic centimeter in warm weather and 5000 in cold weather. Such milk, if quickly cooled to 46° F., and kept at that temperature, will at the end of thirty-six hours contain on the average less than 50,000 bacteria per cubic centimeter, and if cooled to 40° F. will average less than its original niunber. With only moderate cleanliness such as can be employed by any- one without adding appreciably to his expense, namely, clean pails, with small openings, straining cloths, cans or bottles, and hands, a clean place for milking, and a decent condition of the cow's udder and belly, milk when first drawn will not average in hot weather over 30,000, and in cold weather not over 25,000 bacteria per cubic centimeter. Such milk, if cooled and kept at 50° F., will not contain at the end of twenty-four hours over 100,000 bacteria per cubic centimeter. If kept at 40° F. the number of bacteria will not be over 100,000 per cubic centimeter after forty-eight hours. If, however, the hands, cattle, and barns are filthy and the pails are not clean, the milk obtained under these conditions will, when taken from the pail, contain very large numbers of bacteria, even up to 1,000,000 or more per cubic centimeter. Freshly drawn milk contains a slight and variable amount of bac- tericidal substances which are capable of inhibiting bacterial growth. At temperatures under 50° F. these substances act eflBciently (unless the milk is filthy) for from twelve to twenty-four hom-s, but at higher temperatures their effect is very soon completely exhausted, and the bacteria in such milk will then rapidly increase. Thus the bacteria in fresh milk which originally numbered 5000 per cubic centimeter decreased to 2400 in the portion kept at 42° F. for twenty-four hours, but rose to 7000 in that kept at 50° F., to 280,000 in that kept at 65° F., and to 12,500,000,000 in the portion kept at 95° F. As we have seen, the milk in New York City before general pasteur- ization was adopted was found on bacteriological examination to con- tain, as a rule, excessive numbers of bacteria. During the cold weather the raw milk in the shops averaged over 300,000 bacteria per cubic centimeter, during cool weather about 1,000,000, and dming hot weather - about 2,000,000. The above statement holds for milk sold at the ordinary shops, and not that of the best of the special dairies, where, as previously stated, the milk contained only from 1000 to 30,000 bacteria, accord- ing to the season of the year. The committee on the bacteriological examination of milk of the A. P. H. A. at the meeting at Cincinnati in CLEANLINESS USED IN OBTAINING MILK 639 October, 1916, presented the following statement of the interpretation to be placed on milk counts and the number of bacteria permissible. 1. Where the analysis can be made immediately after the milking the number of bacteria enables conclusions to be drawn as to the clean- hness and care in the dairy and the thoroughness in the cleaning and sterilization of the milk vessels, or sometimes the presence of cows with diseased udders. With properly cleaned and sterilized milk vessels and proper care in the farm and dairy the numbers of bacteria should not exceed 10,000, and may be easily brought down to 5000. Numbers beyond these in milk analyzed immediately after the milking may be regarded as an indication of unclean dairy methods, dirty and unsterile milking vessels, or to diseased udders. Apart from diseased udders the factors in dairying that most noticeably increase the bacteria count are unclean udders, milking with wet hands, unsterile milk vessels, unsterile strainers, and failure to cool the milk promptly. 2. If the milk is properly cooled with ice the numbers should not materially increase in five to seven hours. Communities within five to seven hours of their dairies should be able therefore to obtain milk with nearly as low a count as above indicated. Hence in such communities bacterial counts above these numbers should not be found in properly guarded milk. A count of 50,000 in such a community is an indication either of unsatisfactory dairy conditions or of failure to properly cool the milk during transportation. Night's milk if properly cooled can also easily be brought within these limits if analyzed the next morning. A count of over 50,000 for a community close to the dairies must be regarded as unsatisfactory, and the number should not be much more than 10,000 for high grade milk. In hot summer weather the difficid- ties of keeping low counts are greater, but even then they need not surpass 30,000 if the milk is properly cooled. 3. Where milk must be a longer time in transportation from the dairy there will be inevitably an increase in bacteria, depending on the length of time and the temperature. Experience has shown, however, that even in these conditions the excessively high nimibers that have frequently been found in city milk are in reahty due to dirty dairy con- ditions, to dirty and unsterile milk utensils, or culpable neglect in cooling. Moreover, such high bacterial counts at the shipping station are frequently traceable to a few dirty dairies whose milk with an abnormally high count contaminates the rest of the supply. Dirty shipping cases and warm temperatures in shipping are responsible for most of the high bacterial counts in city milk. Where the milk from healthy cows reaches the city within twenty-four hours, however, the number should not be over 100,000 in winter or 200,000 in summer, and numbers in excess of this may be regarded as due either to improper dairy conditions, dirty milk vessels, or insufficient cooling. In larger cities where much of the milk is forty-eight hours in reaching the city higher numbers may naturally be expected. But even under these conditions there is no good reason why the number of bacteria should reach 1,000,000; and it may mostly be brought down to beloW 200,000. 640 BACTERIOLOGY OF MILK IN RELATION TO DISEASE In such cities, therefore, milk with more than 1,000,000 bacteria must be regarded as improperly guarded either at the dairy or on its transit. 4. For a Grade A milk higher demands should be made than for the ordinary grade. The standard set by the Milk Commission for Grade A, viz., of 200,000 for milk to be subsequently pasteurized or for 100,000 to be used raw, is stated by that Commission to be an extreme limit for the most unfavorable conditions. Cities situated near the supplying dairies should demand a much higher standard, which should not allow over 10,000 in bacterial content in Grade A milk in communities favorably situated. 5. For communities situated where ice is not available it may be necessary to accept a milk with a higher bacterial content; but as rapidly as possible the standard should be made to approach the limits above given. The question might be raised. Are even these enormous numbers of bacteria often found in milk during hot weather harmful ? Our knowledge is probably as yet insufficient to state just how many bacteria must accumulate to make them noticeably dangerous in milk. Some varieties are undoubtedly more harmful than others, and we have no way of restricting the kinds that will fall into milk, except by enforcing cleanliness. We have also to consider that milk is not entirely used for some twelve hours after being purchased, and that during all this time bacteria are rapidly multiplying, especially where, as among the poor, no provision for cooling it is made. Slight changes in the milk which to one child would be harmless, would in another produce disturbances which might lead to serious disease. A safe conclusion is that no more bacterial contamination should be allowed than it is practicable to avoid. Any intelligent farmer can use sufficient cleanliness and apply sufficient cold, with almost no increase in expense, | cent per quart, to supply milk twenty-four to thirty-six hours old which will not contain in each cubic centimeter over 50,000 to 100,000 bacteria, and no milk containing more bacteria should be sold. The most deleterious changes which occur in milk during its trans- portation are now known not to be due to skimming off the cream or to the addition of water, but to the changes produced in the milk by mul- tiplication of bacteria. During this multiplication, acids and distinctly poisonous bacterial products are added to the milk, to such an extent that much of the milk has become distinctly deleterious to infants and invalids. It is the duty of health authorities to prevent the sale of milk rendered unfit for use because of excessive numbers of bacteria and their products. The culture tests to determine the number of bacteria present in any sample of milk require at least forty-eight hours; so that the sale of milk found impure cannot be prevented. It will, however, be the purpose of the authorities gradually to force the farmers and the middle- men to use cleanliness, cold, and dispatch in the handling of their milk, rather than to prevent the use.of the small amount tested on any one day. TRANSMISSION OF CONTAGIOUS DISEASES THROUGH MILK 641 If the milk on the train or at the dealer's were found to contain excessive numbers of bacteria, the farmers would be cautioned and instructed to carry out the simple necessary rules furnished them on a printed form. Transmission of Contagious Diseases through Milk. — Pathogenic Bacteria in Milk. — ^Tuberculosis, typhoid fever, septic sore throat, scarlet fever, and diphtheria are the chief diseases transmitted by means of milk in this locality. In other countries cholera, Malta fever, and possibly other diseases may be due at times to milk infection. The obscure disease trembles is also believed to be due to milk. The tubercle bacilli are in the majority of cases derived from the cow, but may come from hmnan sources, the typhoid bacilli are entirely from man, the contagion of true scarlet fever conveyed in milk is probably always from man, but the contagion of a disease closely allied to it is certainly given off by cows suffering from certain septic diseases as yet not fully identified. Diphtheria bacilli are probably always of human origin, as animals, except cats, practically never suffer from the disease and these only under exceptional conditions. The streptococci exciting septic throats are probably usually from human cases of septic inflam- mation but may at times come from septic cows. As milk is usually kept below 60° F. the typhoid bacilli and the streptococci are the only pathogenic germs that we believe increase in any appreciable extent. The following epidemics and cases have been recorded in the bulletin of the Marine Hospital Service, as produced by cow's milk: Epidemics. Cases. Typhoid fever 179 6900 Scarlet fever 51 2400 Sore throat 7 1100 Diphtheria 23 960 Tuberculosis .... No case of measles, smallpox, chicken-pox, whooping-cough, mumps or poliomyelitis has been clearly traced to milk. The Relation of the Typhoid Carrier to Milk Infection. — Many epidemics of typhoid fever have until recently puzzled investigators because, though evidently milk-borne, yet no case of typhoid fever could be found. The discovery that about 2 per cent, of those who have recovered from typhoid fever remain infected and continue during thei rest of their lives to pass typhoid bacilli has cleared up the mystery. Epidemics due to these carriers have already been traced both in New York City and elsewhere. Many observers have already discussed the relation of typhoid cases to milk infection. Hands, water, flies, etc., may all aid in the transfer of the bacilli from the dejecta to the milk. Recently we traced over 400 cases to infection of a milk supply by a typhoid carrier who had the disease forty-seven years ago. The Conveyance of Scarlet Fever by Means of Milk. — ^As we do not know the organism which excites scarlet fever, we are not as clear as to the means by which it is spread as we are in the case of tuberculosis, typhoid fever, and diphtheria. We know, however, that the throat 41 642 BACTERIOLOGY OF MILK IN RELATION TO DISEASE secretions are dangerous. Where the infection has been traced it has usually been found that the milker has suffered from an unrecognized case or is convalescent. It seems as if the contagion must either increase in milk or be capable of infecting when greatly diluted, for cases have developed from milk after great dilution. A small number of epidemics have appeared to come from the milk of diseased cows. Many are skeptical about this, but after personal experience we think it probable. The history of one case was as follows: The milk from a septic cow was delivered to two schools. About thirty of the boys who drank the milk developed the disease while none of the day scholars who went home to lunch did. Some of the cases developed at first only sore throats, others only the rash. On the second day the cases resembled very closely scarlet fever. There was no scarlet fever in the town. The milk con- tained immense numbers of long-chained streptococci. Diphtheria and septic sore throats are occasionally produced by milk. The diphtheria bacilli usually originate from a mild case, the nature of which is not detected. Septic sore throats produced by milk are usually traced to contamination from a human source, but like cases developing the scarlet rash the infection may come from cows suffering from some acute septic udder disease. The Grading of Milk. — ^The appreciation of the importance of the bacteria in milk has led to the pasteurization of all milk entering New York City, except that produced and transported under the very best conditions. Together with the farm conditions the bacterial content of milk is used to grade it. All milk in New York City is divided into grades A, B, and C. Grade A is raw and pasteurized. The raw is from cows which have successfully passed the tuberculin test. Grade B is all pasteurized. Grade C is all pasteurized. The grade A may contain 60,000 bacteria per cubic centimeter. The grade B may contain 1,500,000 when raw and 50,000 after pasteurization. The grade C may contain any number within reason before and 100,000 after pasteurization. It is supposed to be used for cooking purposes only. The State of New York has adopted a similar grading except that it allows raw and pasteurized in all three grades. Many cities and States are adopting such grades. A powerful influence in this direction was the report of the national committee of experts on milk standards. This was a commission appointed by the New York Milk Committee. REFERENCES. Park and Holt: Arch, of Fed., 1903, xv, 705. Pkebcott and Beehd: Central, f. Bakt., Parasiten. u. Infektionskrankheiten, 1911, Band xxx, Abt. II, Heft 16/18, p. 337. CHAPTER L. THE BACTERIOLOGICAL EXAMINATION OF SHELLFISH. Of the shellfish commonly used as food, oysters are the most exten- sively eaten. According to the United States Deputy Commissioner of Fisheries (1913): "Economically, oysters are the most important of all cultivated water products." He estimates the entire oyster crop of the world as over 42,000,000 bushels, representing a money value of $25,000,000. The share of the United States in this industry is about 88 per cent, of the quantity and about 70 per cent, of the value. Three- fourths of this is controlled by the following States: New York, Virginia, Connecticut, Massachusetts, Maryland, New Jersey, Rhode Island and Louisiana. In their normal habitat, in sea waters free from pollution, shellfish are free from dangerous bacteria, but since cities and towns situated on or near the sea coast find it convenient and advantageous to use running water for the disposal of their sewage, oyster-breeding grounds may be subject to pollution and the oysters infected with organisms of the intestinal type. Only when the pollution is sufficiently remote is serious contamination of the beds avoided and this can only be determined by careful sanitary and bacteriological examinations. As a matter of fact, oysters should not be marketed or harvested from waters which are exposed to dangerous sewage pollution. Serious epidemics of typhoid due to the eating of infected shellfish have been reported at various, times. One of the first outbreaks which called attention to this danger was reported by Professor Conn, of Wesleyan University. Investigations showed that the oysters had been fattened at the mouth of a stream, that a house nearby contained two cases of typhoid fever and that drainage from the house entered the stream. The period of infection of oysters is transitory, however. If they are removed from polluted waters, they will cleanse themselves and be safe for food in from six to eight days. This is done at Concarneau, France. Sea water is raised mechanically, is passed through coarse filters, then through sand and finally into reservoirs containing the oysters. Here the water is slowly changed and the purification of the osyters results. Artificial purification of oysters from polluted beds has been reported by Wells, and the method is being tried out on an extensive scale in the harbor at New Haven, Conn. The oysters are placed in water-tight stationary floats containing sea water. Hypochlorite of lime is added in about the proportion of one part of available chlorine to one hundred thousand parts of water. The water is treated a second time with the 644 BACTERIOLOGICAL EXAMINATION OF SHELLFISH hypochlorite, about six hours later. This is to insure the disinfection of organisms which may have escaped the first treatment, by reason of the closing of the oysters when the disinfectant was added until such time as the water is no longer objectionable to them. In twenty-four hours, the total number of organisms in the oysters is greatly reduced and there are practically no bacteria of the B. coli group. It has been noticed by several observers that there is a seasonal variation in the bacterial content of oysters. Gorham (1912) in a series of examinations, found that results obtained from oysters examined in the summer did not agree with those obtained from oysters from the same beds examined in the winter. He concluded, therefore, that oysters hibernate. Experiments carried on in the Research Laboratory of the New York City Health Department (1912) showed that oysters placed in typhoid- infected sea water did not become infected with typhoid bacilli while the temperature was maintained at 3° C. The surrounding water con- tained 13,000 typhoid bacilli per cubic centimeter. Further experi- ments along the same lines were made by Pea,se. He placed oysters and sea water in separate containers in the ice-box and held them at a temperature of 36° F. overnight. Then fuchsin was added to the water and the oysters were placed in the solution. The whole was left in the ice-box all day. Another lot of oysters was placed in fuchsin sea water and kept at a temperature of 65° F. Both sets of oysters were washed in salt water to remove the fuchsin from the shells. When opened the oysters which had been kept at 34° F. showed no trace of fuchsin, while the gills of the oysters kept at 65° F. were turned a dark fuchsin color. From these experiments he concluded that oysters kept at a tempera- ture of 34° to 36° F. will remain closed, and that since particles of soluble dyes in aqueous solution are much smaller than the bodies of bacteria, bacteria in the waters surrounding the oysters are totally excluded. Furtherniore, he says, "Those organisms which have pre- viously gained access to the oyster are destroyed or gradually eliminated so that the total number of bacteria in the oysters is greatly reduced and the oysters become practically free from colon bacilli. Oysters gathered during the hibernating season are more easily handled than during the early part of the oyster season." Standard Methods for the Examination of Shellfish Adopted by the Ameri- can Public Health Association.— Oysters in the Shell.— SeZeciion of Sample.— Twelve oysters of the average sizes of the lot under examination, with deep bowls, short lips and shells tightly closed, shall be picked out by hand and pre- pared for transportation to the laboratory. As complete a record of such data as is possible to obtain shall be made covering the following points: The exact location of the bed from which the sample has been selected. The depth of the water over the bed at time of collection. The state of the tide. The direction and velocity of the wind. Other weather conditions. The day and hour of the removal of the stock from the water. The conditions under which the stock has been kept since removal from the water and prior to the taking of the sample. The day and the hour of the taking of the sample. Transportation of- the Sample.— The oysters so selected shall be packed in suitable metal or pasteboard containers of such size and shape that a number BACTERIOLOGICAL EXAMINATION OF SHELLFISH 645 of them can be enclosed in a shipping case capable of satisfactory refrigeration by means of ice. The important points in this connection are: A. The prevention of the mixing of the oyster liquor of different samples, and of the mixing of the ice-water with the oysters. B. The icing of the samples if they are not to arrive at the point of laboratory examinations inside of thirty-six hours or if the outside temperature is above 50° F. It is not necessary to enclose the oysters in an absolutely tight container providing the above conditions are maintained. Condition of Sampks. — Record shall be made of the general condition of the oysters when received, especially whether the shells are open or closed; of the presence of abnormal odors and of the temperature of the stock. Technical Procedure. — The bacteriological examination shall be started as soon as possible after the receipt of the sample. The oysters shall be thoroughly cleaned with a stiff brush and clear running water and then dried. The edges of the shell shall be passed through the flame or burned with alcohol. The opening of the shell shall be accomplished by either of the following methods : A. By the use of a sterile oyster knife in the usual manner. B. By drilling through a flamed portion of the shell near the hinge with a sterile drill. The drill shall be sterilized and the site of the operation the shell be flamed at least once during the drilling process. Bacterial Counts. — Bacterial counts shall be made of the composite sample of each lot obtained by mixing the shell liquor of five oysters. Agar shall be used for the culture medium and in general the procedure shall be in accordance with the method recommended for examination of water by the Committee of Standard Methods of Water Analysis of the American Public Health Association. The water used for dilution purposes shall contain 1 per cent, sodium chloride, in order to approximate the natural salinity of oyster liquor. The agar plates shall be incubated at 20° C. for three days and the colonies then counted. A satisfactory method of opening an oyster is to strike it a sharp blow with a hammer directly on the large muscle which holds the shells together. This injm-es the oyster just enough to cause the shells to spring apart. The edges of the shell are passed through the flame and the liquor is poured into a sterile test-tube. Determination of Bacteria of the BacHlus Coli Group. — The quantitative determination of the presence of B. coli shall be in accordance with the following procedure : Measured quantities (1.0, 0.1, 0.01 c.c, etc., or their equivalents in dilutions) of the shell water of each of five oysters selected from the dozen, shall be placed in fermentation tubes containing lactose peptone bile, prepared according to the method reconunended by the Committee on Standard Methods of Water Analysis. These shall be incubated for three days at 37°- C, and the presence or absence of gas noted daily. For all ordinary purposes of routine work, a development of 10 to 85 per cent, of gas during this time period shall constitute a positive test indicating a presumption of the presence of at least one bacterium of the Bacillus coli group in the quantity of shell water tested. But no final B. coli rating based on these results shall be used for official approval or con- demnation unless positive confirmatory tests for the presence of organisms of the B. coli group shall have been obtained from the tube of highest or next highest dilution from each oyster; showing the presence of gas. These confirma- ' There seems to be no special object gained in confirming the presumptive tests in oysters showing low scores. A low score is approved anyway and confirmation of B, coli will not raise the score. 646 BACTERIOLOGICAL EXAMINATION OF SHELLFISH tory tests shall be begun immediately upon noting the formation of gas, and carried out in accordance with the procedure recommended by the Committee on Standard Methods of Water Analysis. Statement of Results. — ^The results of the bacterial counts shall be expressed as number of bacteria per cubic centimeter. The results of the lists for B. coli shall be expressed either in the form of the following arbitrary numerical system to be known as "The American Public Health Association Method of Rating Oysters for B. coli; or in Estimated Number of Bacteria of the B. coli Group per Cubic Centimeter of the Sample." The presence of B. coli in each oyster of the five examined is to be given the following values, which represent the reciprocals of the greatest dilutions in which the test for B. coli is positive: If present in 1 c.c. but not in 0.1 c.c, the value of 1. If present in 0.1 c.c. but not in 0.01 c.c, the value of 10. If present in 0.01 c.c. but not in 0.001 c.c. the value of 100, etc. The addition of these values for the five oysters would give the total numerical value for the sample, and this figure would be the score for B. coli. The results should be expressed in the following tabular form: Results op Tests for B. Coli in Dilutions Indicated. Oysters. 1 0.0. 0.1 0.0. 0.01 o.c. Numerical value. 1 + + 10 2 + + 10 3 + 1 4 + 1 S + 1 Total or score for B. coli = 23 + = Presence of B. coli group in fermentation tube test with lactose bUe where sub- sequent isolation tests have confirmed the results of the presumptive test or other satis- factory test. = Failure to demonstrate presence of B. coli group. Estimated Number of B. Coli -per Cubic Centimeter.^ — It will be seen that if the B. coli score is divided by 5, the standard number of oysters tested, the results wiU approximate the number of B. coli per cubic centimeter of shell water. Partly because it does not do this exactly but also for simplicity and the avoid- ance of fractions, the method of stating results as an arbitrary "score" is preferred by the committee. Practical experience with the method has also appeared to justify this conclusion. Sometimes results similar to the following are obtained, that is, one or more oysters may show positive results in small quantities of shell water while an equal number may show negative results in larger quantities. In this case the next lower numerical value shodd be given to the positive results in the high dilutions and such positive results should be considered as being transferred to a lower dilution giving negative results in another oyster. This is done in order to avoid the unnatural result that could follow from what is probably an unequal distribution of the bacteria in the shell water. This recession of numeri- cal values, however, should not be carried beyond the point where the number of such recessions is gp-eater than the number of instances where other oysters in the series failed to give positive B. coli results. As an example of the method of obtaining the score for B. coli, the following illustration is given. . 1 Where the term B. coli is used it refers in all cases to bacteria of the B. coli group and not to the specific prototype. . . BACTERIOLOGICAL EXAMINATION OF SHELLFISH 647 Results of B. Coli Tests in Dilutions Indicated. Oysters. 1 CO. 0.1 0.9. 0.01 CO. Numerical value. 1 + + 10 2 + + 10 3 + + 10 4 + 10 (Not 1) 5 + + + Score, 10 (Not 100) 50 Examination of Oysters Removed from the SheU, or Shucked Stock. — ^The procedure specified for oysters in the shell shall be followed, but attention is called to the fact that higher dilutions than 100 c.c. are usually required. Triplicate fermentation tubes shall be inoculated from end dilutions of the sample. Statement of Results. — ^The results of the bacteriological examination of the opened oysters or shucked stock shall be expressed in the same way as that specified for oysters in the shell, except that in the calcula- tions of B. coli, rating the values for the results of the fermentation tests after confirmation shall be recorded for each of the inoculations of each dilution. In order that the rating from these triplicate tests may be compared with that obtained from testing five oysters in the shell, the sum of the values for the triplicate tests shall be multiplied by |-. If, instead, the sum is divided by 3, the result will give approximately the number of B. coli per cubic centimeter. Clams and Other Shellfish. — The methods for examining clams and shellfish other than oysters shall be those given above. Certain modi- fications are necessary in the method of handling the sample and the opening of the shells, etc. Clams are more likely to lose water during transportation than oysters. It is therefore necessary to take greater precautions to separate differ- ent samples of clams from each other than in the case of oysters. REFERENCES. Smith, H. M.: Oysters, The World's Most Valuable Water Crop, National Geographic Magazine, March, 1913, p. 257. Pkescott and Winslow: Elements of Water Bacteriology, 1913, 3d ed., p. 245. Wells: Public Health Reports, July 14, 1916, pp. 1848, 1852. Gobham: Seasonal Variation in the Bacterial Content of Oysters, Am. Jour. Pub. Health, 1912, u, 24. Pease : Hygienic Results of Refrigeration in the Conservation of Fish and MoUusks, Jour. Am. Pub. Health, November, 1912, ii, 849. CHAPTER LI. THE SOIL BACTERIA AND THEIR FUNCTIONS. SEWAGE BACTERIA. BACTERIA IN INDUSTRIES. son. BACTERIA. The bacteria in the soil belong to many varieties. Some varieties are only accidentally present, being due to the contamination of the earth with the bacteria contained in animal feces and other waste products. The majority, however, pass their life and reproduce them- selves chiefly or wholly in the soil. Many of these varieties have most .important functions to perform in continuing the earth's food supply. Without them plant food, and therefore animal food, would cease to exist. Some make available for plants the carbon, nitrogen, hydrogen, and other compounds locked up in the dead bodies of animals and plants. Others construct food for plants from the gases of the air and the inorganic elements of the earth which in their simpler forms were not available. The bacteria, together with the other somewhat less important microscopic plants and animals, thus form a vital link in the earth's life cycle of plants and animals. The bacteria in the soil require for their activities food, moisture, and a proper temperature^ They may be present to the extent of many millions in a single gram of rich loam, while in an equal quantity of sand they may be almost absent. The various species associated together in the soil flora influence each other. Thus anaerobic bacteria are enabled to grow because of associated aerobes using up the free oxygen, while other species make assimilable substances not usable by others. The Splitting up of Carbon Compounds. — The plants form starch, I and, from it cellulose, wood, fats, and sugar. These substances once formed cannot be utilized by other generations of plants. Some of these are transformed in the bodies of animals, but the largest per- centage await the activities of the microorganisms. The sugars and starches usually undergo an alcoholic fermentation excited by the yeasts and molds with the production of alcohol and carbon dioxide, or an acid fermentation excited by bacteria with the production of acids and frequently of carbon dioxide. Cellulose which is so resistant to decay is attacked by certain varieties of bacteria which are abundant in the soil. They act both in the presence and absence of free oxygen. Molds also act on cellulose. Carbon dioxide, marsh gas, and other products are produced. Wood is appar- ently first attacked by the fungi and only later by other microorganisms. These bacteria are carried into the intestines and act upon cellulose and other substances. DECOMPOSITION 649 The Decomposition of Nitrogenous Compounds. — Plants obtain their nitrogen chiefly in the form of nitrates. The small amount of usable nitrogen in the soil must be constantly replenished. This must either come from the nitrogen forming a part of protein materials or from the free nitrogen in the air. The animals utilize the plant proteins and reduce them to much simpler compoimds, such as urea, but even these are not suitable for plant use. We now know that microorganisms are employed to break compounds into simpler compounds and also to utilize the nitrogen of the air. Decomposition. — This process is to some extent carried out through the agency of molds and other fungi but it is chiefly due to the activities of bacteria. When this process is carried on in the absence of oxygen it is incomplete, giving rise to substances with unpleasant odors, such as HaS, NH3 and CH4. This is called putrefaction. When oxygen is freely accessible more complete decomposition occurs with such end-products as CO2, N and H2O. These two processes, putre- faction and complete decay, cannot be sharply separated, as the second usually follows the first. The varieties of organisms causing these changes are many. Some groups will be found chiefly in decaying vegetable substances, others in animal tissues. They include all mor- phological forms of bacteria as well as yeasts and higher fungi. These forms exist everywhere in nature, although in various degrees, so that every bit of dead organic matter is sure to be decomposed if only moisture and warmth are present. The B. subtilis and B. proteus vulgaris groups are well-known laboratory bacteria that are commonly found among decomposing materials. B. proteus is described under Pathogenic Bacteria. B. subtilis, or hay bacillus (type form), has the following characteristics (Fig. 208): Source and Habitat. — ^Hay, straw, soil, dust, milk, etc. Morphology. — Short, thick rods with round ends, sometimes form threads; sometimes also chains of long rods, short rods, and coccus forms. 0.8m to 1.2,11 broad, 1.3m to 3m long. Often united in strings and threads. Staining Reaction. — Stains by Gram's method. Capsule, Flagella, Motility. — ^Bacilli possess a thin capsule and many flagella which are long and numerous; short forms actively motile; threads immotile. Spore Formation. — Oval spores formed in presence of air germinating at right angles to long diameter. Spores are set free in about twenty-four hours, size 1.2m by 0.6m; widely distributed in nature, dust, air, excreta, etc. (Plate III, Fig. 23). Biology: Cultural Characters (Including Biochemical Features). — ^Bouillon. — Uniformly cloudy growth with marked pellicle, wrinkled and thick; copious spore formation. Gelatin Plates and Tubes. — Saucer-like depressions; colonies have granular centres and folded margins. Surface growth in stab cultures is whitish-gray; colonies sink on liquefaction of medium ; liquefaction progresses in a cylindrical form, and a thick white scum is formed. Agar Plates and Tubes. — Small, irregular, grayish-white colonies; moist glistening growth along needle track in stab cultures. The bacteria in taking certain atoms from the molecules utihzed in their growth leave the other atoms to enter into new relations and 650 SOIL BACTERIA AND THEIR FUNCTIONS form new compounds. The actual products will depend on the decaying substance, the variety of bacteria and the conditions present. Nitrification.— This is a process of oxidation by which through bacterial activities ammonia compounds are changed to nitra.tes and thus rendered utilizable by plants. This change is accompUshed in two stages: first, the ammonia is oxidized to nitrite and, second, to nitrate. The nitrates are taken up by the plant roots from the soil. The bacterial nature of these changes were discovered in 1877 by two French investigators, Schlosing and Muntz. They noted that fer- menting sewage after a time lost its ammonia and gained in nitrates, but that if the sewage was treated with antiseptics, so that fermentation ceased, no such change occurred. Warrington first and Winogradsky later more thoroughly investigated the bacterial cause of these changes. The latter by means of silica jelly, which contained no organic matter, was able to isolate two varieties of cocci, one in Europe and the other in America, which were able to change ammonia to nitrites. He called the one nitrosomonas and the other nitro- sococcus. They are capable of acting on almost any ammonia salt. One variety of organisms capable of changing nitrites to nitrates was isolated, and this bacillus he called nitrobacter. These are small, slightly elongated baciUi. These bacteria are remarkable in that in pure cultiu-es very small amounts of organic matter in the media act as antiseptics. They appear to be able to depend on mineral substances for their food. These bacteria are cxt Fig. 207.— Bacillus subtilis with tremely important, for the plants take up Sentii'li^f "x%ooo*r;- most of their nitrogen in the form of ni- eters. (Franke.) tratcs. Thcsc changes are mostly produced in the surface soil. If the reaction of the soil becomes acid growth ceases. Soil bacteriologists are studying the nitrifying power of different types of soil under identical conditions. The process being one of oxidation, the access of air is necessary. Denitrification. — ^This is a reducing process. The nitrate is made to yield up a part or all of its oxygen and thus becomes changed, to nitrites and to ammonia and even to free nitrogen. The partial change does not rob the soil of its available nitrogen as does the total change, for the nitrites and ammonia may be changed by the nitrifying bacteria to nitrates. These bacteria exist normally in most soils and are especially abundant in manure. There are three different types of nitrogen reduction: (1) The reduction of nitrates to nitrites and ammonia. (2) The reduction of nitrates and nitrites to gaseous oxides of nitrogen. (3) The reduction of nitrites with the development of free nitrogen gas. Nitrogen-fixing Bacteria. — Helbrigel, in 1886, demonstrated that cer- tain plants were able to use the nitrogen of the air and this apparently through the aid of bacteria growing in their roots. These root bacteria NITRIFICATION 651 are named B. radidcola. They produce enlargement (tubercles) on the roots. According to Ball, there is no reasonable doubt but that B. radidcola can and usually does remain active for very long periods in soil devoid of leguminous vegetation. Furthermore, the bacterium diffuses at a very considerable rate through soils that are in proper condition; therefore if a soil should be found lacking the organism, it is illogical to attempt to introduce it artificially without having first made the soil fit for the development of the bacteria. It has not been shown by anyone that increased powers of resistance to unfavorable conditions of certain varieties are at all correlated with their enhanced "greed for nitrogen." Moreover, it is far from being proven that any one race or "physiological species" is really more virile than another. Greig-Smith has shown that as many as three races are sometimes present in one and the same tubercle. Possibly, therefore, fixation of nitrogen may occur most ra,pidly only when two or more of these races are growing together. Buchanan has recently made a minute morphological study of B. radidcola. Some of his conclusions are as follows: 1. Considerable variation in the morphology of B. radidcola may be induced in artificial media by the use of appropriate nutrients. Of the salts of the organic acids, sodium succinate brings about the most luxuriant development and the production of the greatest variety of bacteroids. 2. B. radidcola in the roots of the legumes shows the same type of bacteroids as may be found in suitable culture media. On the other hand, there is little or no correspondence between the type of bacteroid produced in culture media by a certain organism and that produced in the nodule by the same form. 3. It is probable that the term B. radidcola includes an entire group of closely related varieties or species, which differ from each other to some degree in morphological characters. 4. The nodule organism resembles morphologically both the yeasts and the bacteria. The difference between this form and those ordi- narily included under the terms Bacillus and Pseudomonas justify the use of a separate generic name, Bhizobium. In 1893 Winogradsky furnished proof that there are in the soil, bacteria which are outside of the plant roots performing the same function as those within the roots. These bacilli he called Clostridium pasteur- ianum. They are anaerobic and produce spores. Their power to fix nitrogen is increased in presence of sugar and lessened in presence of nitrogenous substances. Beijerinck, in 1901, described two aerobic species of nitrogen-fixing bacteria. Later Bailey described three additional species. These were called Azotobader. These studies have abeady led to the inoculation of soils and to the investigation of the kind of soils and crops best fitted for the growth of these bacteria. Many impoverished soils have already been greatly improved. There are probably many other varie- 652 SOIL BACTERIA AND THEIR FUNCTIONS ties of bacteria capable of fixing nitrogen, because one can hardly examine the roots of any leguminous plants without finding different kinds of tubercles. The use of seed inoculated with the special variety of bacteria suitable for the plant and the soil is already largely practised. Bacteria and Soil Minerals. — Some of the bacterial products act upon the inorganic constituents of the soil. The carbon dioxide and the organic acids act upon compounds of lime and magnesia, practically insoluble in water, to form more soluble substances. The same is true of the rock phosphates, the silicate of potassium, sulphates, etc. Scientific farming is beginning to make use of the knowledge already acquired, and there is reason to hope that great practical advantages will flow from the investigation of the relation of bacteria to soil exhaustion and replenishment. The effect of excessive bacterial development appears at times to be harmful to the soil. Each crop seems to favor the growth of certain varieties, and the exhaustion of the soil which follows the constant raising of the same crop is now suspected to be due in part at least to the continuance of a few restricted species of bacteria in the soil, which failing to produce all the necessary substances for the nutrition of the special crop, vegetation suffers, or again the bacteria finally entirely dissipate substances already in the soil necessary to growth. The application of manure not only adds food for plant life, but also countless numbers of bacteria which make the food more available. , The greatest number of bacteria are contained a little below the surface of the soil, where they are protected from d'rying and sunlight and are in contact with oxygen and with the roots and other food of the superficial soil. BACTERIA IN SEWAGE. The materials which flow from our sewers are a menace to public health, mainly because they so frequently contain pathogenic bacteria. The other products of men and animals are offensive, but rarely con- centrated enough in drinking water to be appreciably deleterious. Sewage can be made harmless by being sterilized, but can be freed from offense only by the destruction of organic matter. This, except when chemical precipitants are used, is almost wholly obtained through bacterial pro- cesses. The purifying value of soil has long been recognized. This is largely due to the action of the soil bacteria. In 1895 the Enghshman, Cameron, introduced the "septic tank," which was a covered cemented pit. The sewage admitted at the bottom flowed out at the top, after about twenty-four hours' subjection to anaerobic conditions. The anaerobic bacteria during this time ferment the organic matter energetically, liquefy it, and develop abundant gas. The knowledge that soil and sand filters act not only mechanically but also and perhaps chiefly bacteriologically, having been acquired, intermittent soil filtration was established as one of the best means of bacteriologically purifying sewage. The sewage is conducted to the beds, allowed to pass through, and then after a few hours again poured THE PRESERVATION OF FOODS 653 on. The purification is based chiefly on the action of the aerobic bacteria in the upper layers of the soil or sand. The best practical results are obtained by combining the two processes: first the anaerobic treatment is used to break down the solid materials, and then the intermittent sand filtration, to oxidize the compounds and render these products harmless. With low temperatures the chemical changes are very much lessened and the filter beds act more like pure mechanical filters. The anaerobic bacteria change the protein substances into simple chemical compounds, among which is ammonia. The carbohydrates are changed into gaseous compounds, acids, etc. The gases are mainly nitrogen, carbon dioxide, and marsh gas. The bacterial changes pro- duced in sewage pom-ed on contact beds made of coarse coke, clinkers, or other material act much as in the sand filters after the filtration. Varieties of Bacteria in Filter Beds and Septic Tanks. — ^The septic tanks all contain spore-bearing bacilli which destroy cellulose, others that attack nitrogenous compounds. The cocci are in a minority. The filter beds have a number of small, non-spore-bearing bacilli; some of these change ammonia into nitrites and nitrates. There are also denitrifying bacteria. As before mentioned, the bacterial efficiency of the bed is increased with suitable temperature and much lessened with low temperature. Sewage Farming. — ^The action of bacteria is utilized in the breaking up of sewage which has been distributed over fields. The amount of sewage which can be poured on a certain area is limited. One acre of land can usually take care of the sewage from one hundred persons. If too much is poured on, it runs off impurified or clogs the soils and prevents the access of oxygen to aerobic bacteria. In warm weather evaporation and bacterial activities are much greater than in cold weather. So far as experience shows, those who eat vegetables from these small farms contract no disease from them. THE PRESERVATION OF FOODS AGAINST DECOMPOSITION BY MICROORGANISMS. The preservation of foods against decomposition by bacteria, yeasts, molds, and higher fungi is obtained by using processes which will prevent the growth of microorganisms. Drying, exposure to wood smoke with consequent absorption of creosote, the addition of salt and sugar, of acids such as vinegar, spices, germicides such as boracic acid, formaldehyde, all are familiar methods of making foods unsuitable for bacterial growth. Instead of using food preserved by drying or chemicals, products may be kept at temperatures too low for bacterial growth. Cold storage of meats, eggs, vegetables, etc., is now common. The sterilization of food substances by heat with protection from infection afterward is made use of extensively in the canning of fruits and vegetables. Care must be taken that absolutely all bacteria are killed, for otherwise decomposition will finally occur. 654 SOIL BACTERIA AND THEIR FUNCTIONS Bacterial Fermentation in Relation to Miscellaneous Products. — Pasteur, in 1857, explained the process of fermentation as due to the action of microorganisms. He demonstrated that the change of sugar into lactic acid only occurred when living bacilli were present. If the fluid was sterilized the fermentation ceased. He stated that "organic liquids do not alter until a living germ is introduced into them." When the action is direct we speak of an organized ferment; when it is indirect, that is, due to the cell product, we call it an unorganized soluble ferment or enzyme. Similar enzymes are produced by the cells of the animal tissues, such as ptyalin, pepsin, and trypsin. Pasteur's work led to the conclusion that the different fermentations were due to different varieties of organisms. The major_part of fermentation is due to yeast. Some important fermentations are due to bacteria and a few to the molds. Wines and Beers. — Alcoholic Fermentation. — If there is a develop- ment of the yeast cells in a solution of grape-sugar we have a fermenta- tion of the sugar with a final development of alcohol and carbon dioxide. It is thus that beers and wines are developed. When the carbohydrate is in the form of starch this is first converted into sugar and then later into the final products. If the sugar is in the form of saccharose, it is first changed by the yeast ferments to glucose. In all these three forms of fermentation the sugar is changed into alcohol and carbonic acid. When the alcohol reaches about 13 per cent, it stops further fermentation. These yeasts called saccharomyces comprise a number of distinct varieties, some of which are cultivated while others, called "wild yeasts," propagate themselves. The distillery, brewery, and wine industries each makes use of special yeasts and special conditions. The rising of bread is one of the most common uses of fermentation by yeast. The yeast acts upon the sugar made by the diastase from the starch. The resulting CO2 and alcohol create myriads of .little bubbles in the dough. Diseases of Beer and Wines. — ^Hansen, Pasteur, and others demon- strated that the spoiling of beers and wines was due to the development of varieties of bacteria and yeasts which produce different kinds of fermentation from that desired. These produce alterations in flavor, bitterness, acidity. Vinegar Making. — ^Vinegar is made from some weak alcoholic solution by the union of alcohol with oxygen. This oxidation can be brought about by a purely chemical process. When vinegar is formed in the usual way bacteria are essential. The scum on the surface of the fermenting alcohol is a mass of microorganisms. The mother of vinegar was named mycoderma by Pearson. Kutzing showed that this was composed of living cells. Hansen proved these to be bacteria. We now know there are many varieties of bacilli capable of producing this fermentation. Each variety has its own optimum temperature and differs in the amount of acid it produces. Most of these have the peculiarity of growing at high temperatures into long threads without any traces of division. At low temperatures they produce long threads BACTERIAL DISEASES OF PLANTS 655 with swollen centres. The usual vinegar is made by u^ing the variety of bacilli prevalent in the surroundings, but the custom is growing of adding to the pasteurized alcoholic solution the special variety desired in pure culture. Sauerkraut.^ — ^This is cabbage leaves shredded, slightly fermented, and prevented from decay by the lactic acid bacteria. At first both yeasts and bacteria increase together, but with the increase in acidity all growth ceases. Putrefaction is prevented by the same cause. The lactic acid bacteria are the same as those found in sour milk. Ensilage. — ^The fermentation is believed to be due partly to enzymes in the corn tissues and partly to bacterial action. The first changes are due chiefly to the enzymes. The Curing of Tobacco. — The curing of tobacco is apparently due partly to bacterial processes and partly to the action of leaf enzymes. The Bacterial Diseases of Plants. — ^These are probably as serious and varied for plants as for animals. The pear blight, the wilt disease of melons, the brown rot of tomatoes, the block rot of cabbages are examples. These plant diseases can be communicated by means of the pathogenic pure cultures of bacteria experimentally just as readily as animal diseases by their specific bacteria. REFERENCES. Bailey L. H.: Bacteria in Relation to Country Life. Bail, O. M.: A contribution to the Life History of B. radicicola Beij., Centralbl. f. Bakt., etc., 1909, II Abt., xxiii, 47. Buchanan, R. E.: The Bacteroids of Bacillus radicicola. Centralbl. f. Bakt., etc., 1909, II Abt., xxiii, 59. Greig-Smith: Jour. Soc. Chem. Indust., 1907, No. 7. CHAPTER LII. THE DESTRUCTION OF BACTERIA BY CHEMICALS. PRACTICAL USE OF DISINFECTANTS. Many substances, when brought in contact with bacteria, combine with their cell substance and destroy the life of the bacteria. While in the vegetative stage bacteria are much more easily killed than when in the spore form, and their life processes are inhibited by substances less deleterious than those required to destroy them. Bacteria, both in the vegetative and in the spore form, differ among themselves considerably in their resistance to the poisonous effects of chemicals. The reason for this is not wholly clear, but it is connected with the structure and chemical nature of their cell substance. Chemicals in sufficient amount to ' destroy life are more poisonous at temperatures suitable for the best growth of bacteria than at lower temperatures, and act more quickly upon bacteria when they are sus- pended in fluids singly than when in clumps, and in pure water rather than in solutions containing organic matter. The, increased energy of disinfectants at higher temperatures indicates in itself that a true chemical reaction takes place. In estimating the extent of the destruc- tive or inhibitive action of chemicals the following degrees are usually distinguished : L The growth is not permanently interfered with, but the pathogenic and zymogenic functions of the organism are diminished — attenuation. This loss of function is usually quickly recovered. 2. The organisms are not able to multiply, but they are not destroyed — antiseptic action. When transferred to a suitable culture fluid free of the disinfectant these bacteria are capable of reproduction. 3. The vegetative development of the organisms is destroyed, but not the spores — incomplete or complete sterilization or disinfection, according as to whether spores are present in the organisms exposed and as to whether these spores are capable of causing infection. 4. Vegetative and spore forms are destroyed. This is complete sterilization or disinfection.^ The methods employed for the determination of the germicidal action of chemical agents on bacteria are, briefly as follows: If it is desired to determine the minimum concentration of the chemical substance required to produce complete inhibition of growth we proceed thus: A 10 per cent, solution of the disinfectant is prepared and 1 c.c, 1 Disinfection strictly defined is the destruction of all organisms and their products which are capable of producing disease. Sterilization is the destruction of all saprophytic as well as parasitic bacteria. It is not necessary in most cases to require disinfectants to be capable of sterilizing infected materials containing spores, for there are but few varieties of pathogenic bacteria which produce spores. DESTRUCTION OF BACTERIA BY CHEMICALS 657 0.5 c.c, 0.3 c.c, etc., of this is added to 10 c.c. of liquefied gelatin, agar, or bouillon, or, more accurately 10 c.c. minus the amount of solution added, in so many tubes. The tubes then contain 1 per cent., 0.5 per cent., 0.3 per cent., and 0.1 per cent, of the disinfectant. The fluid medium in the tubes is then inoculated with a platinum loopful of the test bacterium. The melted agar and gelatin may be simply shaken and allowed to remain in the tubes, and watched for any growth which takes place, or the contents of the tubes may be poured into Petri dishes, where the development or lack of development of colonies and the number can be observed. If no growth occurs in any of the dilutions, lower dilutions are tested. Bacteria that have been previously injured in any way will be inhibited by much weaker solutions of chemicals than will vigorous cells. The same test can be made with material containing only spores. If it is desired to determine the degree of concentration required for the destruction of vegetative development, the organism to be used is cultivated in bouillon, and into each of a series of tubes is placed a definite amount of diluted culture from which all clumps of bacteria have been filtered; to these a definite amount of watery solution of different percentages of the disinfectant is added. At intervals of one, five, ten, fifteen, and thirty minutes, one hour, and so on a small platinum loopful of the mixture is taken from each tube and inoculated into 10 c.c. of fluid agar or gelatin, from which plate cultures are made. When- ever it is probable that the antiseptic power of the disinfectant approaches somewhat the germicidal, it is necessary to inoculate a second series of tubes from the first so as to decrease still further the amount of anti- septic carried over. The results obtained are signified as follows: x per cent, of the disinfectant in watery solution and at y temperature kills the organism in twenty minutes, z per cent, at the same temperature kills in one minute, and so on. If there be any doubt whether the trace of the disinfectant carried over with the platinum loops may have rendered the gelatin unsuitable for growth, thus falsifying results, control cultures, if extreme accuracy is desired, should be made by adding bacteria which have been somewhat enfeebled by slight contact with the disinfectant to fluid to which a similar trace of the disinfectant has been added. If the strength of the disinfectant is to be tested for different substances it must be tested in these substances or their equivalent, and not in water. The disinfectant to be examined should always be dissolved in an inert fluid, such as water; if on account of its being insoluble in water it is necessary to use alcohol for its solution, control experiments may be required to determine the action of the alcohol on the organism. Sometimes, as in the case of corrosive sublimate, the chemical unites with the cell substance to form an unstable compound, which inhibits the growth of the organism for a time before destroying it. If this compound is not broken up in the media it will probably not be in the body. In some tests it is of interest to break up this union and note then whether the organism is alive or dead. With corrosive sublimate the bacteria probably die within thirty minutes after the union occurs. 42 658 DESTRUCTION OF BACTERIA BY CHEMICALS In the above determinations the absolute strength of the disinfectant required is considerably less when culture media poor in albumin are employed than when the opposite is the case. Cholera spirilla grown in bouillon containing no peptone or only 0.5 per cent, of peptone are destroyed in half an hoiu- by 0.1 per cent, of hydrochloric acid; grown in 2 per cent, peptone bouillon, theu- vitality is destroyed in the same time on the addition of 0.4 per cent. HCl. In any case the organisms to be tested should all be treated in exactly the same way and the results accompanied by a statement of the conditions under which the tests were made. It is becoming the custom to state the power of a disinfectant in terms' of comparison with pure carbolic acid. A sub- stance which had the same destructive power in a 1 to 1000 solution as carbolic acid in a 1 to 100 solution would be rated as of a strength ten times that of carbolic acid. The following table gives the results and methods used in an actual experiment to test the effect of blood serum upon the disinfecting action of bichloride of mercin-y and carbolic acid upon bacteria: Test fob the Difference of Effect of Bichloride of Mercury and Carbolic Acid Solution on Ttphoid Bacilli in Serum and in Bouillon. Time. 1' 3' 5' 10' 20' 30' 45' hi. lil. 2 hrs. Strength of solution. A. Serum . . 2.5 c.c' HgCU sol. 1 : 1000 2.5 c.o. Typhoid broth culture. B. Bouillon . 2.5 c.o." HgCl2 sol. 1:1000 2.5 c.c. Typhoid broth culture. C. Serum . . 2.5 c.c' CarboUc sol. 5% 2.5 c.c. Typhoid iroth culture. D. Bouillon . 2.5 c.o.' Carbolic sol. 5% 2.5 o.c. Typhoid broth culture. + + + + + + - - - _ - , 'Equals 1:2000 bi- , chloride. Same. ("Equals 2J% car- \ bolic acid. Same. — Indicates total destruction of bacteria with no growth in media. + Indicates lack of destruction of bacteria with growth in media. THE STANDARDIZATION OF DISINFECTANTS. Rideal and Walker were the first to urge a useful method for standard- izing disinfectants. In carrying out the test the various factors must be carefully con- trolled, thus: Tivie: this should be constant, the strength of the disinfectant being the variant. Test organisms: a standard culture of the typhoid bacillus (Hopkins' strain) is used to avoid any variations due to the different degrees of resistance of various strains. The culture should be subinoculated three days before used. Medium: a standard STANDARDIZATION OF DISINFECTANTS 659 meat-extract broth 1.5 per cent, acid to phenolphthalein; 10 c.c. to a tube is employed. Temperature: this test is done at 20° C. This is important, as the germicidal activity increases with the temperature. Constant amount of culture used: 0.1 c.c. of the twenty-four-hour broth culture is added, to 5 c.c. of the disinfectant solution. This is more accurate than the drop method. Amount inoculated: it is essential that the same amount be inoculated from each dilution. Platinum loops made of 23 United States gauge wire, the loops being 4 mm. in diameter are employed. Several are used, being left on a rack after sterilization so as to be cold when needed. The loop is bent at an angle of 45° to the shank. The actual test is carried out as follows: A 5 per cent, carbolic solu- tion (phenol C. P.), is prepared and standardized by bromine titration. From this freshly prepared 1 to 90 to 1 to 100 and 1 to 110 dilutions are made as needed. The necessary dilutions of germicide are then prepared. Wide jumps in the dilutions are made and then narrowed as the limits of the disinfectant to be tested are determined. Five test-tubes are arranged in a row in a water -bath at 20° C, and the solutions added in 5 c.c. amounts. Time must be allowed for the solution to reach 20° C. If the bath be large enough, little attention is needed to keep the temperature constant. The culture having been brought to 20° C. is then added in 0.1 c.c. amounts and the tubes shaken, an interval of thirty seconds allowed between each tube. Subinoculation of the first tube is then made after thirty seconds, which gives an interval of two and a half minutes after inoculation. The tubes are then sub- inoculated in order at thirty-second intervals, giving an interval for each of two and one-half minutes after inoculation, and starting at the first, gives an interval of five minutes, etc. It is possible to use ten tubes, as each step can be done in less than fifteen seconds if properly arranged, allowing a much wider range for each test. It is not necessary to keep the cotton plugs in the tubes during the operation nor to remove the tubes from the bath to obtain the loopful for inoculation. The loop is plunged to the bottom, care being taken not to touch the sides of the tubes, and care should also be taken that a loopful is carried away each time. The broth tubes are incubated for forty-eight hours and then examined for growth. The following are given as two examples: Sample. Dilution. 2.5 min. Time of i 5 7.5 min. min. Bxposure: 10 min. 12.5 min. 15 min. Phenol coefficient. Phenol . . . . 1 to 90 1 to 100 + + + + — — — 100)550 5.5 Disinfectant A. . . 1 to 450 1 to 550 1 to 600 + + + + + + + + + - Phenol . . . . 1 to 100 1 to 110 + + + + + + + -1- _ 110)650 5.191 Disinfectant A. . . 1 to 600 1 to 650 1 to 700 + + + + + + + + + + + + + 660 DESTRUCTION OF BACTERIA BY CHEMICALS These tables not only serve as an example, but also show that unless many repetitions of the tests are made and averaged, considerable variations in the results may be obtained. A report of 5.5 or 5.1 is equally accurate in the test here reported of the Rideal-Walker method, even with certain improvements added. With practice, and by selection of the dilutions to be employed, the operator evidently can regulate the time factor so that fairly uniform results are obtainable. On the other hand, it not infrequently happens that if more than one carbolic dilution is employed, more than one time period is open to comparison. For these reasons Anderson and McClintic have modified the test by setting two time limits two and one-half and fifteen minutes and taking the average. The following is an example: Time of exposure: 2.5 5 7.5 10 12.5 15 Phenol Sample. Phenol . . Dilution. . 1 to 80 1 to 90 min. + min. min. min. min. min. coefficient. 375 650 1 to 100 + + + — — — 80 110 1 to 110 + + + + + — Disinfectant A. . 1 to 350 _ _ _ - 1 to 375 — — — 4.69 + 5.9 1 to 400 + — — — 1 to 425 + + — — — — 2 1 to 450 + + — — — — 1 to 500. + + — — — — 5.30 1 to 550 + + + — — — 1 to 600 + + + + — — 1 to 650 + + + + + — 1 to 700 + + + + + + * 1 to 750 + + + + + + Disinfectants vary widely in their germicidal properties, depending on whether organic matter is present or not. As under practical condi- tions organic matter is usually present, it is of some importance to know how far organic matter decreases the efficiency. For the purpose of obtaining comparable results, Anderson and McClintic have suggested the use of peptone 10 per cent, and gelatin 5 per cent, in distilled water. One part of the culture is mixed with 10 parts of the organic solution, 1.1 c.c. being then added to a series of dilution tubes containing 4 c.c. In determining the coefficient allowance must be made for the added amount of organic matter. The modified methods of Anderson and McClintic^ are called the "hygienic laboratory phenol coefficient," with or without organic matter. Any organic matter may be used in the test to approach the special conditions under which a disinfectant is to be used. In comparing the value of disinfectants the cost as well as the coefficient must be considered. This is best stated in terms of the 1 See Hygienic Laboratory Bulletin No. 82, for further details and apparatus for sim- plifying the steps of the test. STANDARDIZATION OF DISINFECTANTS 661 relative cost of 100 units of efficiency as compared with pure phenol = 100, thus: Cost of disinfectant per gallon CoefBcient of disinfectant. ( = Cost ratio) -^ Cost of phenol per gallon Coefficient of phenol (1). ( = the efficiency) , X 100 = relative cost per 100 units. Antiseptic Value. — With certain disinfectants there is sufficient of the disinfectant carried over by the loop to exert antiseptic action and growth does not occur. If this is not taken into consideration a disinfectant will be given an excessively high coefficient. No satis- factory method has been devised to avoid this diflficulty. The inocu- lated broth tube may be shaken and a loop or more inoculated from it to second broth tube, in this way diluting the disinfectant still further. Chick^ has attempted to overcome the difficulty in the case of mercury-containing disinfectants by adding 0.2 c.c. of a saturated watery solution of hydrogen sulphide to each tube of broth. Many substances which are strong disinfectants become altered under the conditions in which they are used, so that they lose a portion or all of their germicidal properties; thus, quicklime and milk of lime act by means of their alkali and are disinfecting agents only so long as sufficient calcium hydroxide is present. If this is changed by the carbon dioxide of the air into carbonate of lime it becomes harmless. Bichloride of mercury and many other chemicals form compounds with many organic and inorganic substances, which, though still germi- cidal, are much less so than the original substances. Solutions of chlorine, peroxides, etc., when in contact with an excess of organic matter soon become inert because of the chemical compounds formed. The Disinfecting Properties of Inorganic Compounds. — ^Bichloride of Mercury. — ^This substance, which dissolves in 16 parts of cold water, when present in 1 part in 100,000 in nutrient gelatin or bouillon, inhibits the development of most forms of bacteria. In water 1 part in 50,000 will kill many varieties in a few minutes, but in bouillon twenty-four hours may be needed. With organic substances its power is lessened, so that 1 part to 1000 may be required. Most spores are killed in 1 to 500 watery solution within one hour. Corrosive sublimate is less effective as a germicide in alkaline fluids containing much albumin- ous substance than in watery solution. In such fluids, besides loss in other ways, precipitates of albuminate of mercury are formed which are at first insoluble, so that a part of the mercuric salt does not really exert any action. In alkaline solutions, such as blood, blood serum, pus, sputum, tissue fluids, etc., the soluble compounds of mercury are converted into oxides or hydroxides. For ordinary use, where corrosive sublimate is employed, solutions of 1 to 500 and 1 to 2000 will suffice, when brought in contact with bacteria, to kill the vegetative forms within from one to twenty minutes, the stronger solution to be used when much organic matter is present. 1 Journal of Hygiene, 1908, viii, 654. 662 DESTRUCTION OP BACTERIA BY CHEMICALS Mercuric chloride \'olatilizes slowly and it is better to wash ofF walls after use of bichloride solutions. Solutions of this salt should not be kept in metal receptacles. Mercuric chloride solution has disadvantages in that it corrodes metals, irritates the skin, and forms almost inert, compounds with albuminous matter. In order to avoid accidents, solutions of this odorless disinfectant should be colored by some dye. Biniodide of Mercury.— This salt is very similar in its efFect to the bichloride. Nitrate of Silver. — ^Nitrate of silver in watery solution has about one-fourth the value of the bichloride of mercury as a disinfectant, but nearly the same value in inhibiting growth. In albuminous solutions it is equal to bichloride of mercury. Compounds of silver nitrate and albuminous substances have been used because of the absence of .irri- tative properties combined with moderate antiseptic power. Sulphate of Copper. — ^This salt has about 50 per cent, of the value of mercuric chloride. It has a quite remarkable affinity for many species of algse, so that when in water 1 to 1,000,000 it destroys many forms; 1 to 400,000 destroys typhoid bacilli in twenty-four hours when the water has no excessive amount of organic material. It is not known to be poisonous in this strength, so that it can be temporarily added to water supplies. Sulphate of Iron. — ^This is a much less powerful disinfectant than sulphate of copper. A 5 per cent, solution requires several days to kill the typhoid bacilli. It can only be considered as a mild antiseptic and deodorant. Zinc Chloride. — This is very soluble in water, but is a still weaker disinfectant than copper sulphate. Sodium Compounds. — A 30 per cent, solution of NaOH kills anthrax spores in about ten minutes, and in 4 per cent, in about forty-five minutes. One per cent, kills vegetative forms in a few minutes. Sodium carbonate kills spores with difiiculty even in concentrated solution, but at 85° C. it kills spores in from eight to ten minutes. It is used fre- quently to cover metallic instruments. A 5 per cent, solution kills in a short time the vegetative forms of bacteria. Even ordinary soapsuds have a slight bactericidal as well as a marked cleansing effect. The bicarbonate has almost no destructive effect on bacteria. Calcium Compounds. — Calcium hydroxide, Ca(0H)2, is a powerful disinfectant; the carbonate, on the other hand, is almost without effect. The former is prepared by adding one pint of water to two pounds of lime (quicklime, CaO). Exposed to the air the calcium hydrate slowly becomes the inert carbonate. A 1 per cent, watery solution of the hydroxide kills bacteria which are not in the spore form within a few hours. A 3 per cent, solution kills typhoid bacilli in one hour. A 20 per cent, solution added to equal parts of feces or other filth and mixed with them will completely sterilize them within two hours. Effect of Acids. — An amount of acid per liter which is equivalent to 40 c.c. of normal hydrochloric acid is sufficient to prevent the growth of all STANDARDIZATION OF DISINFECTANTS 663 varieties of bacteria and to kill many. Twice this amount destroys most bacteria within a short time. The variety of acid makes little difference. Bulk for bulk, the mineral acids are more germicidal than the vegetable acids, but that is because their molecular weight is so much less. A 1 to 500 solution of sulphuric acid kills typhoid bacilli within one hour. A similar solution of hydrochloric acid is about one- third weaker, and acetic acid somewhat weaker still. Citric, tartaric, malic, formic, and saUcyhc acids are similar to acetic acid. Boric acid destroys the less resistant bacteria in 2 per cent, solution and inhibits the others. Gaseous Disinfectants. — The germicidal action of gases is much more active in the presence of moisture than in a dry condition. Sulphur Dioxide (SO2). — Niunerous experiments have been made with this gas owing to the fact that it has been so extensively used for the disinfection of hospitals, ships, apartments, clothing, etc. This gas is a much more active germicide in a moist than in a dry condition; due, no doubt, to the formation of the more active disinfecting agent — sulphurous acid (H2SO3). In a pure state anhydrous sulphur dioxide does not destroy spores, and is not certain to destroy bacteria in the vegetative form. Sternberg has shown that the spores of the Bacillus anthracis and Bacillus subtilis are not killed by contact for some time with liquid SO2 (liquefied by pressure). Koch found that various species of spore-bearing bacilli exposed for ninety-six hours in a dis- infecting chamber to the action of SO2, in the porportion of from 4 to 6 per cent, by volume, were not destroyed. In the absence of spores, however, the anthrax bacillus in a moist condition, attached to silk threads, was found by Sternberg to be destroyed in thirty minutes in an atmosphere containing 1 per cent, by volume. As the result of a large number of experiments with SO2 as a disin- fectant it has been determined that an "exposure for eight hours to an atmosphere containing at least 4 volumes per cent, of this gas in the presence of moisture" will destroy most, if not all, of the pathogenic bacteria in the absence of spores. Four pounds of sulphur burned for each 1000 cubic feet will give an excess of gas. Peroxide of Hydrogen (H2O2). — ^This is an energetic disinfectant, and in 2 per cent, solution (about 40 per cent, of the ordinary commercial article) will kill the spores of anthrax in from two to three hours. A 20 per cent, solution of a good commercial hydrogen peroxide solution will quickly destroy the pyogenic cocci and other spore-free bacteria. It combines with organic matter, becoming inert. It is prompt in its action and not poisonous, but apt to deteriorate if not properly kept. Chlorine. — Chlorine is a powerful gaseous germicide, owing its activity to its affinity for hydrogen and the consequent release of nascent oxygen when it comes in contact with microorganisms in moist condition. It is therefore a much more active germicide in the presence of moisture than in a dry condition. Thus, Fischer and Proskauer found that dried anthrax exposed for an hour in an atmosphere containing 44.7 664 DESTRUCTION OF BACTERIA BY CHEMICALS per cent, of dry chlorine were not destroyed; but if the spores were previously moistened and were exposed in a moist atmosphere for the same time, 4 per cent, was effective, and when the time was extended to three hoiu-s 1 per cent, destroyed their vitality. The anthrax bacillus, in the absence of spores, was killed by exposure in a moist atmosphere containing 1 part to 2500 for twenty-four hours. In watery solutions 0.2 per cent, kills spores within five minutes and the vegetative forms almost immediately. One part in one million is suflBcient to destroy typhoid bacilli, in a water containing little organic matter, in the course of a few hours. In water with much organic matter a much greater amount is required. Chlorinated Lime (Called "Chloride of Lime"). — Chlorinated lime is made by passing nascent chlorine gas over unslaked lime. It should not contain less than 10 per cent, of available chlorine, and can now be obtained containing 30 per cent. It should have a strong odor of chlorine. Its efficacy depends on the chlorine it contains in the form of hypochlorites. The calcium hypochlorite is readily broken up into hypochlorous acid. A solution in water of 0.5 to 1 per cent, of chlorin- ated lime will kill most bacteria in one to five minutes, and 1 part in 100,000 will destroy typhoid bacilli in twenty-four hours. A 5 per cent, solution usually destroys spores within one hour. Chlorinated lime not only bleaches, but destroys fabrics. The Hypochlorites (Labarraque's Solution). — Solutions of hypochlorites are practically the same as solutions of chlorinated lime and are much more expensive. Bromine and iodine are of about the same value as chlorine for gaseous disinfectants, in the moist condition; but, like chlorine, they are not applicable for general use in house disinfection, owing to their poisonous and destructive properties; they have a use in sewers and similar places. Trichloride of iodine in 0.5 per cent, solution destroys the vegetative forms of bacteria in five minutes. Organic Disinfectants. — Alcohol in 10 per cent, solution inhibits the growth of bacteria; absolute alcohol kiUs bacteria in the vegetative form in from several to twenty-four hours. According to Epstein, 50 per cent, alcohol (in water) has more germicidal power than any other strength, the power gradually diminishing with both stronger and weaker solutions. Formaldehyde. — Formaldehyde, or formic aldehyde, was isolated by von Hoffmann in 1867, who obtained it by passing the vapors of methyl alcohol mixed with air over finely divided platinum heated to. redness. The methyl alcohol is oxidized and produces formaldehyde as follows: CH3OH + = CH2O + H2O. Formaldehyde is a gaseous compound possessed of an extremely irritating odor. At a temperature of 68° F. the gas is polymerized — that is to say, a second body is formed, composed of a union of two STANDARDIZATION OF DISINFECTANTS 665 molecules of CH2O. This is known as a paraformaldehyde, and is a white, soapy body, soluble in boiling water and in alcohol. Formal- dehyde is sold in commerce as a clear, watery liquid containing from 33 to 40 per cent, of the gas and 10 to 20 per cent, of methyl alcohol, its chief impurity.. If the commercial solution — ordinarily known in the trade as "formalin" — is evaporated or concentrated above 40 per cent., paraformaldehyde results; and when this is dried in vacuo over sulphuric acid a third body — trioxjonethylene — is produced, consisting of three molecules of CH2O. This is a white powder, almost insoluble in water or alcohol, and giving off a strong odor of formaldehyde. The solid polymers of formaldehyde, when heated, are again reduced to an active gaseous condition; ignited, they finally take fire and burn with a blue fiame, leaving but Uttle ash. When burned they have no" germicidal properties. Formaldehyde has an active aflBnity for many organic substances, and forms with some of them definite chemical combinations. It combines readily with anunonia to produce a compound called hexa- methylene tetramine, which possesses neither the odor nor the anti- septic properties of formaldehyde. This action is made use of in neutral- izing the odor of formaldehyde when it is desired to dispel it rapidly after disinfection. Formaldehyde also forms combinations with certain aniline colors — viz., fuchsin and safranin — the shades of which are thereby changed or intensified. These dyes are tests for aldehydes. These are the only colors, however, which are thus affected, and as they are seldom used in dyeing, owing to their liability to fade, this effect is of httle practical significance. The most delicate fabrics of silk, wool, cotton, fur, etc., are unaffected in texture or color by formaldehyde. Iron and steel are attacked, after long exposure, by the gas in combination with watery vapor; but copper, brass, nickel, zinc, silver, and gilt work were not at all acted upon. Formaldehyde unites with nitrogenous products of decay — ^fermen- tation or decomposition — forming true chemical compounds, which are odorless and sterile. It is thus a true deodorizer in that it does not replace one odor by another more powerful, but forms new chemical compounds which are odorless. Formaldehyde has a peculiar action upon albumin, which it transforms into an insoluble and indecomposable substance. It renders gelatin insoluble in boiling water and most acids and alkalies. It is from the property of combining chemically with the albuminoids forming the protoplasm of bacteria that formaldehyde is supposed to derive its bactericidal powers. Formaldehyde is an excellent preserva- tive of organic products. It has been used for the preservation of meat, milk, and other food products; but, according to Trillat and other investigators, formaldehyde renders these substances indigestible and unfit for food. It has been successfully employed as a preservative of pathological and histological specimens. There are no exact experiments recorded of the" physiological action of formaldehyde on the human subject when taken internally. A 1 per cent, solution has been taken in considerable quantity without 666 DESTRUCTION OF BACTERIA BY CHEMICALS serious results; and trioxymethylene has been given in doses up to 90 grains as an intestinal antiseptic. Accorduig to Aronson rabbits and guinea-pigs allowed to remain for twelve and twenty-four hours in rooms which were being disinfected with formaldehyde gas were found to be perfectly well when the rooms were opened. On autopsy the animals showed no injurious effects of the gas. Others have noticed that animals, such as dogs and cats, which have accidentally been confined for any length of time in rooms undergoing formaldehyde disinfection occasionally died from the effects of the gas. Many observers, how- ever, have reported that insects, such as roaches, flies, and bedbugs, are not, as a rule, affected. The result of these observations would seem to indicate that although formaldehyde is comparatively non-toxic to the higher forms of animal life, nevertheless a certain degree of caution should be observed in the use of this agent. It is important to remember that formaldehyde in gaseous form is practically inert as an insecticide except in extremely great concentrations. The researches of Pottevin and Trillat have shown that the germi- cidal power of the gas depends not only upon its concentration, but also upon the temperature and the condition of the objects to be steril- ized. As with other gaseous disinfectants — ^viz., sulphur dioxide and chlorine — it has been found that the action is more rapid and complete at high temperatures— i. e., at 35° to 45° C. (95° to 120° F.)— and when the test objects are moist than at lower temperatures and when the objects are dry. Still, it has been repeatedly demonstrated by actual experiment in rooms that it is possible to disinfect the surface of apartments and articles contained in them, under the conditions of temperature and moisture ordinarily existing in rooms even in winter, by an exposure of a few hours to a saturated atmosphere of formalde- hyde gas. The results of numerous experiments have shown that in the air 2.5 per cent, by volume of the aqueous solution, or 1 per cent, by volume of the gas, are sufficient to destroy fresh virulent cultures of the common non-spore-bearing pathogenic bacteria in a few minutes. Stahl has shown that bandages and iodoform gauze can be kept well sterilized by placing in the jars containing pieces of a prepara- tion of paraformaldehyde in tablet form containing 50 per cent, of formaldehyde. The same experimenter has also succeeded in making carpets and articles of clothing germ-free by spraying them with 0.5 to 2 per cent, solution of formaldehyde for fifteen to twenty minutes without the color of the fabrics being in any way affected. The investigations of Trillat, Aronson, Pottevin, and others have shown that a concentration of TTTTTrTr of the aqueous solution (40 per cent.), equal to 2^TffTni" of pure formaldehyde, was safe and sufficiently powerful to retard bacterial growth. A 2 per cent, watery solution of formalin destroys the vegetative forms of bacteria within five to thirty minutes. In our experiments formalin has upon the vegetative forms about one-half the strength of pure carbolic acid. STANbAnbl^ATldN OP biStNPECTANfS 66? Chloroform (CHCI3). — ^This substance, even in pure form, does not destroy spores, although it kills bacteria in vegetative form, even in 1 per cent, solution. Chloroform is used practically as an antiseptic in antitoxic sera and in blood serum for culture purposes. The chloro- form is expelled from the serum by heating it to 55° C. Iodoform (CHI3). — ^This substance has but very little destructive action upon bacteria; indeed, upon most varieties it has no appreciable effect whatever. When mixed with putrefying matter, wound dis- charges, etc., the iodoform is reduced to soluble iodine compounds, which act partly by destroying the bacteria and partly by uniting with poisons already produced. Carbolic Acid (CgHsOH). — Pure phenol crystallizes in long, colorless crystals. In contact with air it deliquesces. It has a penetrating odor, a burning taste, and is a corrosive poison. It is soluble at ordinary temperatures in about 15 parts of water. Carbolic acid dissolves in water with some difficulty and should be therefore thoroughly mixed. It is not destructive to fabrics, colors, metals, or wood, and does not combine as actively with albuminous matters as bichloride of mercury. It is therefore more suitable for the disinfection of feces, etc. A solution having 1 part to 1000 inhibits the growth of bacteria; 1 part to 400 kills the less resistant bacteria, and 1 part to 100 kills the remainder. A 5 per cent, solution kills the less resistant spores within a few hoiu-s and the more resistant in from one day to four weeks. A slight increase in temperature aids the destructive action; thus, even at 37.5° spores are killed in three hours. A 3 per cent, solution kills streptococci, staphylococci, anthrax bacilli, etc., within one minute. Carbolic acid loses much of its value when in solution in alcohol or ether. An addition of 0.5 HCl aids its activity. Carbolic acid is so permanent and so comparatively little influenced by albumin that it is widely used in practical disinfection even in place of more powerful substances. Cresol. — Cresol [C6H4(CH3)OH] is the chief ingredient of the so-called "crude carbolic acid." This is almost insoluble in water, and therefore of restricted value. Many methods are used for bringing it into solution so as to make use of its powerful disinfecting properties.- With equal parts of crude sulphiu-ic acid it is a powerful disinfectant, but it is, of course, strongly corrosive. An alkaline emulsion of the cresols and other products contained in "crude" carbolic acid with soap is called creolin. It is used in 1 to 5 per cent, emulsions. It is fully as powerful as pure carbolic acid. Lysol is similar to creolin, except that it has more of the cresols and less of the other products. It and creolin are of about the same value. Tricresol. — ^Tricresol is a refined mixture of the three cresols (metar cresol, paracresol, and orthoscresol) . It is soluble in water to the extent of 2.5 per cent, and its disinfecting power is about three times as great as that of carbolic acid. Creolin. — Creolin contains 10 per cent, of cresols held in solution by soap. 668 DESTRUCTION OF BACTERIA BY CHEMICALS Lysol. — ^Lysol contains about 50 per cent, of cresols. It mixes with water in all dilutions. Oil of Turpentine, 1 to 200, prevents the growth of bacteria. Camphor has very slight antiseptic action. Creosote in 1 to 200 kills many bacteria in ten minutes; 1 to 100 failed to kill tubercle bacilli in twelve hours. Essential Oils. — Cardiac and Meumir found that the essences of cinnamon, cloves, thyme, and others killed typhoid bacilli within one hour. Sandalwood required twelve hours. Thymol and eucalyptol have about one-fourth the strength of car- bolic acid (Behring). Oil of peppermint in 1 to 100 solution prevents the growth of bacteria. Tables op Antiseptic Values.' , Alum .... AluWnum acetate Ammonium chloride Boric acid . Calcium chloride . Calcium hypochlorite Carbolic acid . Chloral hydrate Cupric sulphate Ferrous sulphate . Formaldehyde (40%) Hydrogen peroxide 1 to 222 Mercuric chloride . . . 1 to 14,300 1 to 6000 Mercuric iodide 1 to 40,000 1 to9 Potassium bromide 1 to 10 1 to 143 Potassium iodide . 1 to 10 1 to 25 Potassium permanganate 1 to 300 1 to 1000 Pure formaldehyde 1 to 25,000 1 to 333 Quinine sulphate . 1 to 800 1 to 107 Silver nitrate .... 1 to 12,500 1 to 2000 Sodium borate 1 tol4 1 to 200 Sodium chloride . 1 to6 1 to 10,000 Zinc chloride .... . 1 to 500 1 to 20,000 Zinc sulphate .... . 1 to 20 1 These figures are approximately correct, and represent the percentage of disinfection required to be added to a fluid containing considerable organic material, in order perma- nently to prevent any bacterial growth. Solutions of half the given strength will inhibit the growth of most bacteria and prevent the growth of many varieties. STANDARDIZATION OF DISINFECTANTS 669 OOO'OI 0!» I 'ujtasjoiqQ ■ooe °% I '[osaaoux OOZ o* I 'aq^qdins jaadoo ■OBZ 01 I 'auipoi JO JO spuofqoijj, ■OS 0% I OOZ 0% I 'apixoiad uaaoipjtjj S lOO "O rt lO oo lOO "50 (N^ CO rH o o3 O O lOO lOO »o i-IC<3 ' r-l on O o 03 O pq ^ S3 ^ e- P. B 3 Ph o 3 ^1 c3 bll ^ ri a ■ o3 o .CI El a p. 3o a? £ >>Q. ft a ■+-' o ,D r^ h-l "pi S ^ O ■s 3.1 £| -35 ag -rt S I • p c3 o 1=1 S^ -, CQ -eight-hour broth culture of B. pyocyaneus. They are left for two or three minutes or until they are thoroughly saturated, then removed to filter-paper in another covered Petri dish and left to dry. When dry they are placed in tissue-paper envelopes, which are stamped with all necessary data. Each envelope is dated and sealed and sent to the disinfector who places it in the room which is to be disinfected. The driver who calls for the bedding takes up the tests, placing them in a manilla envelope and entering them upon his card. The envelopes are then returned to the laboratory where the tests and receipt card are compared and any discrepancy noted. The test envelopes are then stamped with date of receipt, and the threads are removed and placed in a modified Ayer's medium, which is a synthetic medium and consists of the following: Asparagin 4 Neutral sodium phosphate 2 _ Sodium lactate 6 Sodium chlorate 5 Water 1000 Add enough NaOH to render the medium alkaline to litmus. This culture medium may be depended upon to give bright green color reaction in twenty-four to forty-eight hours. The tubes are incubated for forty-eight hours and the color reac- tion noted and entered upon test envelope. At the end of the week a bacteriologist's report is compiled which shows at a glance the work of each disinfector, the number of cases of each disease for which disinfection was performed, the number of successful disinfections, the number of tests lost, etc. Sulphur Dioxide in House Disinfection. — Four pounds of sulphur should be burned for every 1000 cubic feet. The sulphur should be broken into small pieces and put into a pan sufficiently large not to allow the melted sulphur to overflow. This pan is placed in a much larger pan holding a little water. The cracks of the room should be carefully pasted up and the door, after closing, also sealed. Upon the broken sulphur is poured three or four ounces of alcohol and the whole lighted by a match. The alcohol is not only for the purpose of aiding the sulphur to ignite, but also to add moisture to the air. An exposure of eight to twelve hours should be given. Sulphur fumigation carried out as above indicated is not as efficient as formaldehyde fumigation, but suffices for surface disinfection for diphtheria and the exanthemata. All heavy goods should be removed for steam disinfection if there is any possibility of the infection having penetrated beneath their surface. If there is no place for steam disin- fection their surfaces should be thoroughly exposed to fumigation and then to the air and sunlight. In many cases when cleanliness has been observed, surface disinfection of halls, bedding, and furniture may be all that will be required. 680 PRACTICAL DlSlNfECflON AHD Sf^BILII^AflOM There is always a very slight possibility of a deeper penetration of infection than that believed to have occurred; it is therefore better to be more thorough than is considered necessary rather than less. Sulphur dioxide without the addition of moisture has, as already stated under the consideration of disinfectants, very little germicidal value upon dry bacteria. Advantages of Formaldehyde Gas over Sulphur Dioxide FOR Disinfection of Dwellings. — Formaldehyde gas is superior to sulphur dioxide as a disinfectant for dwellings; first, because it is more efficient in its action; second, because it is less injurious in its effects on household goods; third, because when necessary it can easily be supplied from a generator placed outside of the room and watched by an attendant, thus avoiding, in some cases, danger of fire. Apart from the cost, of the apparatus and the greater time involved, formaldehyde gas, generated from commercial formalin, is not much more expensive than sulphur dioxide — ^viz., twelve to twenty cents per 1000 cubic feet against ten cents with sulphur. Therefore, we believe that .formaldehyde gas is the best disinfectant at present known for the surface disinfection of infected dweUings. For heavy goods it is far inferior in penetrative power to steam; but for the disinfection of fine wearing apparel, furs, leather, upholstery, books, and the like, which are injured by great heat, it is, when properly employed, better adapted than any other disinfectant now in use. Public Steam Disinfecting Chambers. — ^These should be of sufficient size to receive all necessary goods, and may be either cylindrical or rectangular in shape, and are provided with steam-tight doors opening at either end, so that the goods put in at one door may be removed at the other. When large the doors are handled by convenient cranes and drawn tight by drop-forged steel eye-bolts swinging in and out of slots in the door frames. The chambers should be able to withstand a steam pressure of at least one-half an atmosphere, and should be constructed with an inside jacket, either in the form of an inner and outer shell or of a coil of pipes. This jacket is filled with steam during the entire operation, and is so used as to bring the goods in the disin- fecting chamber up to the neighborhood of 220° F. before allowing the steam to pass in. This heats the goods, so that the steam does not condense on coming in contact with them. It is an advantage to dis- place the air in the chamber before throwing in the steam, as hot air has far less germicidal value than steam of the same temperature. To do this a vacuum pump is attached to the piping, whereby a vacuum of fifteen inches can be obtained in the chamber. The steam should be thrown into the chamber in large amount, both above and below the goods, and the excess should escape through an opening in the bottom of the chamber, so as more readily to carry off with it any air still remaining. The live steam in the chamber should be under a pressure of two or three pounds so as to increase its action. To disinfect the goods we place them in the chamber, close tight the doors, and turn th/e steam into the jacket. After about ten minutes. biBlNFECTANfS AND METHODS OF DISINFECTION 681 when the goods ha\'e become heated, a \acuum of ten to fifteen inches is produced, and then the hve steam is thrown in for twenty minutes. The steam is now turned oft", a vacuum is again formed, and the chamber again superheated. The goods are now thoroughly disinfected and dry. In order to test the thoroughness of any disinfection, or any new chamber, maximum thermometers are placed, some free in the chamber and others surrounded by the heaviest goods. It will be found that, even under a pressure of three pounds, live steam will require ten minutes to penetrate heavy goods. Practical Points on Heat Disinfection.' — ^In the practical application of steam for disinfecting purposes it must be remembered that while moist steam under pressure is more effective than streaming steam, it is scarcely necessary to give it the preference, in view of the fact that most known pathogenic bacteria produce no spores and the spores of the few that do develop them are quickly destroyed by the temperature of boiling water, and also that "superheated" steam is less effective than moist steam. When confined steam in pipes is "superheated" after its generation it has about the same germicidal power as hot, dry air at the same temperature. Esmarch found that anthrax spores were killed in streaming steam in four minutes, but were not killed in the same time by superheated steam at a temperature of 114° C. It should also be remembered that dry heat has but little penetrating power, and that even steam requires time to pass through heavy goods. Koch and Wolffhiigel found that registering thermometers placed in the interior of folded blankets and of other large packages did not show a temperature capable of killing bacteria after three hours' exposure in a dry hot-air oven at 133° C. and over. We have put a piece of ice in the middle of several mattresses and recovered it after exposing the goods to an atmosphere of live steam for ten minutes. The Disinfection of Hands, Instruments, Ligatures, and Dressings for Surgical Operations. — ^Instruments. — AH instruments, except knives, after having been thoroughly cleansed, are boiled for three minutes in a 1 per cent, solution of washing soda. Knives, after having been thoroughly cleansed, are washed in sterile alcohol and wiped with sterile gauze and then put into boiling soda solution for one minute. This will not injure their edges to any great extent. Gauze. — Gauze is sterilized by moist heat either in an Arnold steam sterilizer for one hour or in an autoclave for thirty minutes. It is placed in a perforated cylinder or wrapped in clean towels before putting in the sterilizer, and only opened at the operation. Iodoform gauze is best made by, sprinkling sterile iodoform on plain gauze sterilized as described above. Ligatures — Catgut. — Boil for one hour in alcohol under pressure at about- 97° C. It is often put in sealed glass tubes, which are boiled under pressure. These remain indefinitely sterile. The alcohol does not injure the catgut. If desired,- the catgut can be washed in ether and then soaked a short time in bichloride before heating in alcohol. Bookman, of- St. Paul, suggested wrapping the separate strands of 682 PRACTICAL DISINFECTION AND STERILIZATION " catgut in paraifin paper and then heating for three hours at 140° C. This procedure prevents the drying out of the moisture and fat from the catgut, so that it remains unshri veiled and flexible after its exposure. Darling, of Boston, tested this method and found it satisfactory. Dry formaldehyde gas does not penetrate sufficiently, and is not reliable. Silver wire, silk, silkworm gut, rubber tubing, and catheters are boiled the same as the instruments. Hand-brushes. — These should be boiled in soda solution for ten minutes. The Skin of the Patient. — It is impossible absolutely to sterilize the deeper portions of the skin, but sufficient bacteria can be removed to render infection rare. The skin is washed thoroughly with warm green soap solution, then with alcohol, and finally with 1 to 1000 bi- chloride. A compress wet with a 25 per cent, solution of green soap is now placed on, covered with rubber tissue, and left for three to twelve hours; and after its removal the skin is washed with ether, alcohol, and bichloride solution, and then covered with a gauze com- press previously moistened with a 1 to 1000 bichloride of mercury solu- tion. At the operation the skin is again scrubbed with green soap solu- tion followed by ether, alcohol, and then with the bichloride of merciu-y solution. In some places the bichloride compress is replaced one hour before the operation by a pad wet in 10 per cent, solution of formalin. The Hands. — Fiirbinger's method, shghtly modified, is now much used, and gives good results. . The hands are washed in hot soap and water for five minutes, using the nail-brush. They are then soaked in 85 per cent, alcohol for one minute and scrubbed with a sterile brush. They are finally soaked in a 1 to 1000 bichloride of mercury solution for two minutes. The alcohol and bichloride of mercury are sometimes combined and used together. Another method which gives good results is as follows: Skin of operator is scrubbed for five minutes with green soap and brush, then washed in chlorinated lime and car- bonate of soda in proportions to make a good lather; washed off in sterile water, and then scrubbed with brush in warm bichloride solution 1 to 1000. Owing to the risk of leaving untouched bacteria under the nails and in cracks of the skin, sterilized rubber gloves are now being used more and more in operations. Some surgeons prefer sterilized cotton gloves frequently changed. The gloves can be sterilized by steam. Mucous Membranes. — Here absolute sterilization cannot be achieved without serious injury to the tissues. Those of the mouth and throat are cleansed by a solution consisting of equal parts of peroxide of hydro- gen and lime-water. In the nostrils it is better to employ the milder solutions, such as diluted Dobell's or Listerine. These are also used in the mouth instead of the peroxide. Wadsworth urges the use of prepara- tions containing about 30 per cent, of alcohol as being very efficient. Dakin's solution (see below) may also be tried. The vagina is swabbed out thoroughly with sterile warm soap and water, and then irrigated with a 2 per cent, carbolic acid or a 1 to 1000 bichloride of mercury solution. DISINFECTANTS AND METHODS OF DISINFECTION 683 Disinfection of Wounds. — The immense number of wounded in the European War has led to a further study of the use of antiseptic solutions. The great efficiency of chlorine and some of its less stable compounds as purifying, deodorizing and bleaching agents has for a long time been known. Its action was attributed to the strong indirect oxidizing influ- ence exerted by it. Chlorine gas and calcium hypochlorite, "bleaching powder" or "chloride of lime" were recognized as among the most potent, though destructive disinfectants. Potassium and sodium hypo- chlorite were also employed as disinfectants and to some extent as antiseptics, though the latter use was restricted by the irritating action of these substances. , It was not until the present war in Europe created a demand for more efficient antiseptics than those in general surgical use that recourse was made to chlorine compounds in which the chlorine existed in a combina- tion from which it could be readily liberated in active form according to requirements. The researches of Dr. H. D. Dakin, at first associated with Dr. Alexis Carel in France and subsequently commissioned by the British Govern- ment, demonstrated that sodium hypochlorite in strictly neutral solu- tion and not exceeding 0.5 per cent, in concentration, could be used freely in surgical practice without occasioning discomfort.^ His studies on the action of this antiseptic showed that when it acted on protein substances it conferred antiseptic properties on them and that the chlorine presumably became linked to a nitrogen of one or more of the amino acids composing the protein. This led to the inference that soluble substances possessing a nitrogen-chlorine linkage would be likely to have antiseptic properties. This was verified by experiment. The group of substances, called chloramines, which have this linkage, contain many preparations which are soluble and very slightly, if at all, toxic. The cheapest of these, because it can be manufactured from a waste substance obtained in the production of saccharin, is paratoluene- sodiumsulphochloramide, C6H6S02NaNC1.3H20. This substance is ob- tained in white crystals, readily soluble in water and singularly stable both dry and when in solution. It readily parts with chlorine when 1 A neutral sodium fiypoehlorite solution of proper strength, "Dakin's solution," can be prepared as follows: Dissolve 14 grams of dry sodium carbonate (or the equivalent of the hydrated crystals) in 1 liter of water, add 20 grams of fresh bleaching powder with not less than 30 per cent, available chlorine and shake the mixture vigorously for at least five minutes. Allow it to stand for half an hour and then filter. To the clear filtrate, add 2 grams of boric acid, which should so completely neutralize the solution that a particle of solid phenolphthalein will not become pink when moistened with it. It is also possible to prepare a neutral solution of sodium hypochlorite without the addition of boric acid, if sodium bicarbonate is employed. With the bleaching powder of the quality mentioned, the proportions may be chosen as follows: 45 grams dry sodium carbonate and 45 grams of sodium bicarbonate are dissolved in 1 liter of water. One hundred grams of bleaching powder are well shaken with 1 liter of water and then the alkaline solution added and the whole again thoroughly shaken. The precipitate is then removed by filtration. The result- ing clear solution should react neutral to phenolphthalein. Should this not be the case, a little carbon dioxide may be passed into the solution to neutralize the residual trace of acid. This solution is three times the strength which should be used. When required, it should be diluted with two volumes of water. Sodium hypochlorite solutions deteriorate. They should not be used when over a week old. 684 Practical disinpection and sterilization brought into contact with easily oxidizable substances, and is a powerful antiseptic. Proteins are not precipitated by this chloramine, and its penetration when applied to the tissues is unusually great. As it is decomposed in exerting its action, it should be frequently renewed, but its non-toxic and unirritating characters permit its free and abundant use. Besides Dakin's solution other antiseptic irrigation fluids are being experimented with in the treatment of the wounded, and also ointments and powders. It is believed that for certain infections special anti- septics should be selected. Hypodermic and Other Syringes. — ^These when not boiled are steril- ized by drawing up into them boiling water a number of times and then finally a 5 per cent, solution of carbolic acid, the acid after three minutes to be washed out by boiling water. If cold water is used the carbolic solution should remain in the barrel for ten minutes. Great care should be taken to wash out all possible organic matter before using the carbolic acid or boiling to sterilize. Syringes made entirely of glass or of glass and asbestos can be boiled in soda solution. The Sterilization of Milk. — Complete sterilization destroys all the germs in milk, and so, if no new ones gain entrance, prevents per- manently fermentative changes. This requires boiling for fifteen to forty-five minutes on two or three consecutive days, according to the presence or absence of certain spores. Milk is best sterilized by heat, for nearly all chemicals, such as boric acid, salicylic acid, and formalin, are not only slightly deleterious themselves but also make the milk less digestible, and therefore less fit for food. Formalin is the least objectionable of the three. Milk may be sterilized at a high or low temperature — that is, at the boiling temperature — or at a lower degree of heat, obtained by modifying the steaming process. Milk heated at as high a temperature as 100° C. is not altogether desirable for prolonged use for infants, as the high tem- perature causes certain changes in the milk which make it less suitable as a food for them. Pasteurization. — ^These changes are almost altogether avoided if a temperature below 70° C. is used. It is recommended, therefore, that the lowest temperature be used for partial sterilization which will keep the milk wholesome for twenty-four hours in the warmest weather and kill the tubercle, typhoid, and other non-spore-bearing bacilli. Raising the milk to a temperature of 60° G. for twenty minutes, 65° G. for fifteen, 70° for five, 75° for two, or 80° for one will accomplish this. Exposure for even one minute at 70° destroys 98 per cent, of the bacteria which are not in the spore form. Fully 99 per cent, of tubercle bacilli are destroyed. This subject is considered more fully in the chapter on Milk. One of the many forms of apparatus is the following: (a) A tin pail or pot, about ten inches deep by nine inches in diameter, provided with the ordinary tin cover which has been perforated with eight holes each an inch in diameter. (b) A wire basket, with eight nm-sing bottles (as sold for this pur- pose in the shops). GENERAL CONCLUSIONS ON DISINFECTION 685 (c) Rubber stoppers for bottles and a bristle brush for cleaning the bottles. Directions (Koplik). — Place the milk, pure or diluted (as the physi- cian may direct), in the nursing bottles and place the latter in the wire basket. Put only sufficient milk for one nursing in each bottle. Do not cork the bottles at first. Having previously poured about two inches of water in the tin pail or pot and brought it to the boiling-point, lower the basket of nurs- ing bottles slowly into the pot. Do not allow the bottles to touch the water or they will crack. Put on the perforated cover and let the steaming continue for ten minutes; then remove the cover and firmly cork each bottle. After replacing the cover, allow the steaming to continue for fifteen minutes. The steam must be allowed to escape freely or the temperature will rise too high. The process is now complete. Place the basket of bottles in a cool, dark place or in an ice-chest. The bottles must not be opened until just before the milk is to be used, and then it may be warmed by plunging the bottle in warm water. If properly prepared the milk will taste but little Uke boiled milk. The temperature attained under the conditions stated above will not exceed in extreme cases 87° C. (188° F.). A different but admirable method is the one devised by Dr. Free- man. Here a pail is filled to a certain mark with water, and then placed on the stove until the water boils. It is then removed, and immediately a milk-holder, consisting of a series of zinc cylinders, is lowered with its milk bottles partially full of milk. The cover is again applied. The heatof the outside water raises the temperature of the milk in ten minutes to about 65° C. (150° F.), and holds it nearly at that point for some time. After twenty minutes the milk is removed, placed in cold water, and quickly cooled. The milk is kept in the ice- chest until used. When milk is pasteurized in great quantities it should always be done by the "holding process," as "flash" pasteurization is unreliable. Milk should be pasteurized when it is as fresh as possible, and only sufficient milk for twenty-four hours should be pasteurized at one time. If after nursing the infant leaves some milk in the bottle this should be thrown away. Care of the Bottles. — After nursing, the bottles should be filled with a strong solution of washing soda, allowed to stand twenty-four hours, and then carefully cleaned with a bristle (bottle) brush. The rubber stoppers and nipples after using should be boiled in strong soda solution for fifteen minutes and then rinsed and dried. After sterilizing, milk should never be put into unsterilized bottles, as this will spoil it. GENERAL CONCLUSIONS ON DISINFECTION. The previous pages have shown that it is comparatively easy to destroy microbes by germicides or heat when access to them is attain- able. Foods, instruments, clothing, bedding, the excreta, the surface 686 PRACTICAL DISINFECTION AND STERILIZATION of the body, etc., can be readily disinfected, but when we try to disin- fect the mucous membranes of the hving person we fail. The Importance of Disinfection of Surroundings after Recovery or Death from Infectious Diseases. — ^Year by year knowledge is accumulating which indicates that nearly all cases of spread of infection are due to the imme- diate transfer from a living carrier to the person who contracted the disease. The carrier may be diseased or may simply harbor the germs. If cleanliness is maintained throughout the disease there is, as a rule, little need of specific disinfection after recovery. REFEEENCES. Dakin: British Med. Journal, 1915, ii, 318. Dakin: British Med. Journal, Jan. 29, 1916, p. 160. Wadsworth: Mouth Disinfection, Jour. Infect. Dis., 1906, p. 779. INDEX. A Abbe condenser, 123 Abrin production of antitoxins, 163 Achorion Schoenleinii (Favus), 237. See under Molds, pathogenic. Acid-fast bacilli, 394-398 chromogenic, 394 non-chromogenic, 395 staining of, 79 Acids, formation of, from alcohols, 64 from carbohydrates, 64 Actinomyces, 41, 467—471 isolation of, 470 microscopic appearance of, 469 occurrence of, 470 pathogenesis of, in animals, 470, 471 in man, 471 in tissues, 469 Actinomycosis, 470, 471 treatment of, 471 Aedes calopus, 580 Agar media, 97. See Media. Agar-stick cultures, mioro-aerophilic or- ganisms, 53 Agglutination, 202-215. See also- Agglu- tinin and Agglutinogen. BaoiUus dysenterise, 360 mallei, 403 incubator, 403 rapid, 404 tuberculosis, 387 tjrphosus, 342 bacteria, loss of capacity, 208 non-agglutinable, 209 bacteriolysis, relation to, 209 group, 204 pseudoreactions, 212 reaction, comparison of tube and slide methods, 215 macroscopic, slide, 211 tube, 210 microscopic, 211 pseudo-, 212 serum, to obtain, 210 Agglutinin, absorption methods of, 206 accuniulation of, group and specific, 206 characteristics of, 203 cholera, 461 development of, 203 specific and group, 205 variation in strength of, 209 Agglutinogen, characteristics of, 203 Aggressins, 155 Air, bacteria in, 621 Aleuronat powder, use of, 129 Alexin, 179. See Complement. Alkali, dilute, action of, on bacteria, 151 Amboceptor, 179 content titration, 195 multipartial, 182 origin of, 182 polyvalent, 182 preparation of, 188 relation of, to virulence, 182 unit, 195 Ameba, 522-532. See also Protozoa. binucleata, 48 cultivation of, 525 diagnosis of, differential, 531 found in feces, 135 in lungs, 135 in sputum, 135 historical note on, 522 immunity from, 530 isolation of, 525 media for. 111 morphology of, 525 pathogenesis of, 527-530 in animals, 527 in man, 528 prognosis of infection, 530 in pyorrhea alveolaris, 531 reproduction of, 526 sites of, in human body, 524 source of, 530 staining of, 524 treatment of infection, 530 viabihty of, 527 Amebae, classification of, 20 medium for pure culture, HI Amphitricha, 36 Anaerobes found in feces (Gram-positive), 135 in pelvic organs, 135 Anaerobic bacteria, 434—454 conditions, to obtain, 120 methods, fluid media, 122 solid media, Buchner, 121 Kjumwiede, 121 Wright, 121 Zinsser, 121 organisms, media for, 105 Anaphylaxis, 227-232 antianaphylaxis, 228 anaphylotoxins, 231 incubation period, 228 688 INDEX Anaphylaxis, to sensitize, amount re- quired, 228 specificity of proteins, 228 Anaphylotoxins, 231 Andrade indicator, 100 Aniline dyes, 75 water, to make, 77 Animals, autopsies of, 129 inoculation of, 127 blood, intravenous, 128 body, cavities, 128 brain substance, 128 cutaneous, 127 eye, anterior chamber, 128 by inhalation, 128 intestines, 128 intracutaneous, 127 intravenous, 128 lungs, 128 peritoneum, 129 skin, 128 subcutaneous, 128 trachea, 128 ventricle, 128 Koch's postulate, use of, 123 use of, for diagnosis and test, 127 Anthrax bacillus, 426. See B. anthracis. internal, 431 % intestiiial, 431 ' maUgnant edema, 431 symptomatic, 444. See B. anthracis symptomatici. Antiamboceptors, 181 Antianaphylaxis, 228 Antibodies, action of, 165 types of, 162 Antibody, 164. See also Complement- fixation, agglutination. Ehrhch theory of, 166 unit, 195 Anticomplementary dose, 193 Anticomplements, 181 Antiferments, 230 . Antiformin method for tubercle bacilli, 391 Antigen, 164 erythrocytes, 191 preparation of, 189 requirements of, 193 unit, 194 Antiseptic action, 656 solutions in lessening introduction of infection, 159 values, table of, 668 Antiseptics, 158 Antitoxin, bacillus diphtheria, nature of, 173 persistence of, 306 presence of, in living body, 607 production of, 174 toxin-, immunization, 607 treatment with, 604 tetanus, 440 nature of, 173 production of, 175 Antitoxin, baciUus tetanus, treatment with, 609 unit, 440 concentration of, 176 Ehrlich theory of, 168-171 globulins, isolation of, 177 intracutaneous test of, 176 neutralization of toxin in, 171 neutralizing value of, 172 serum, collection of, 175 standardizing of, 175 unit, diphtheria, 176 hsts of, 171 tetanus, 176 Antituberculous serum, 387 Ascomycetes, 27 Auto-infection, microbal, 149 Autolysates, 586 Autopsy of aninials, 129 technic for obtaining material, 129 Azobacter, 651 AzoUtmin (Kahlbaum), 100, 109 B Babesia, 548-552 cultivation of, 552 morphology of, 549 prophylaxis of, 551 Texas fever, symptoms of, 551 Bacilli, characteristics of, shape, 31 size, 31 BaciUus abortus (Bang), 407 aceti, 61 formation of, etc., 64 acidi lactici, 324 found in feces, 135 acidophilus, 317 acne, 303 vaccine for, 590 aerogenes, 324 capsulatus, 332, 447. See Bacil- lus welchii vaccine for, 594 alkaligenes, 350 classification of, 323 found in feces, 135 reaction of, differential, 324 anaerobic, 434r^54 anthracis, 426-433 biology of, 428 cultivation of, 429 diagnosis of, bacteriological, 432 differential, 433 historical notes on, 426 infection, prevention of, 159 leukocyte extract, 225 McFadyean-Heine methylene- blue reaction, 428 morphology of, 426 pathogenesis of, 157, 429-432 in animals, 429-431 in man, 431-4-32 prophylaxis against infection,432 serum treatment, 432 INDEX 689 Bacillus anthracis, spore formation, 427 stain, differential, 428 staining of, 428 symptomatici, 444-446 antitoxins, production of, 163 biology of, 445 cultivation of, 445 distribution of, 445 inoculations, preventive,446 morphology of, 444 pathogenesis of, 445 toxin, 445 avisepticus, 421 bifidus, 317 of Bordet-Gengou, 417 botulinus, 449-450 diagnosis of botulism, 450 poison of, 153 poisoning due to, 155 prophylaxis of, 450 resistance of, 449 toxin production of, 449 bronchisepticus, similar to Bacillus pertussis, 419 of bubonic plague, 421. See Bacillus pestis. bulgaricus, 319 of chicken cholera, 421 cloacse, 333 coli acids, effect of, 325 characteristics of, 325 classification of, 323 gall-stones, cause of, 330 gas production of, 326 groups of, 324 found in feces, 135 in peritoneal fluid, 135 growth with other bacteria, 328 immunity from, 331 indol formation of, 327 intestinal juices, effect of, 327 intracellular poison, 151 isolation of, 332 media for, 106-108 meningitis, 283 pathogenesis of, in animals, 328 in man, 328 in pneumonia, 266 reaction, differential, 324 fermentation, 326 resistance of, 328 significance of, in water, 616 temperature of, influence of, 328 vaccine, 594 commimior, 324 communis, 324, 325 diphtheria;, 292-315 antiseptic solutions in, 159 antitoxin formation, 163 nature of, 173 persistence of, 306 treatment, 604 bacteria associated with, 306 biology of, 295 cultivation of, 295-297 44 Bacillus diphtheriae cultures, examina^ tion of, 310 inoculation of, 310 diagnosis of, bacteriological, 309 technic of, 309 relation of, to bacteriology, 308 diphtheria-like bacilli, 301 diphtheroids, 302. See Bacillus hofmanni. exudate, microscopic examina- tion of , 311 found in eye, 136 in nose, 135 in sputum, 135 in throat, 135 historical notes on, 292 immunity against, 306 infection, mixed, 306 localization of, 139 media for toxin production, 109- 110 in milk, 641 morphology of, 292 on serum-free media, 295 mutation of, 304 non-toxin producing, 301 pathogenesis of, in animals,. 297 in humans', 298 in pneumonia, 266 resistance of, 305 Schick reaction, 312-315 staining of, 293 susceptibility to, 306 swab for cultures, 309 toxicity test, 312 toxin, 298-302 antitoxin, 607 specific soluble, 151, 153 variation in production of, 137 transmission of, 305 diplobaciUi, 32 of Ducrey, 419 dysenteriae, 356-360 ag^utination of, 360 antiaggressin, 156 carriers of, 369 classification of, 323 communicability of, 358 cultivation of, 357 diagnosis of,"359 differential, 359 from ameba, 531 distribution of, 368 found in feces, 135 historical note on, 356 immunity, 359 media for, 106-108 morphology of, 356 motility of, 356 pathogenesis of, in animals, 357 in man, 358 reaction of, differential, 374 serum treatment, 602 susceptibility to, 359 690 INDEX Bacillus dysenterise toxins, 358 vaccine, 593 variability of, 360 viability of, 358 edematis maHgni, 446-447 biology of, 447 cultivation of, 447 immunity to, 447 morphology of, 446 pathogenicity of, 447 resistance of, 447 stain for, 446 enteritidis, 353 diagnosis of, differential, 354 methods of, 355 historical note on, 351 in rats, 354 reaction, differential, 324 sporogenes, 448. See Bacillus welchii. variation of, 354 fecalis alkahgenes, 350. See BacUius alkaUgenes. food poisoning, 450 fusiformis, 450-451 hofmanni, 302 of hog cholera, 355 icteroides, 579 , influenzse, 408-413 agglutination of, 413 baciUi resembling, 412 biology of, 409 complement-fixation in, 410 in conjunctivitis, 411 cultivation of, 409 diagnosis of, bacteriological, 411 epidemiology of, 411 examination of sputum for, 412 group found in brain, 136 in cerebrospinal fluid, 135 in eye, 135 in lungs, 135 in nose, 135 in pleural cavity, 135 in sputum, 135 in throat, 135 hemoglobin requirement of, 138 historical note on, 408 length of life of, 409 • in meningitis, 283, 411 morphology of, 408 pathogenesis of, in animals, 410 immunity to, 410 in man, 410 in pneumonia, 262 pseudo-influenza bacilM, 412 resistance of, 409 serum therapeutics, 413 staining of, 409 symptoms of, clinical, 413 in tuberculosis, 411 vaccine, 413, 595 of Koch- Weeks, 413-417 in trachoma, 415 leprse, 394 of leprosy, staining reaction of, 76 Bacillus of Lustgarten, 398 of mahgnant edema (Bacillus edema- tis maUgni), 446 diagnosis of,- differential, 433 mallei, 399-407 agglutination reaction of, 403 incubator, 403 rapid, 404 biology of, 399 complement-fixation of, 199, 403 cultivation of, 400 glanders, diagnosis of, 402-406 mallein, 404 physical, 402 postmortem, 400 Straus reaction, 406 immunity to, 402 medium for mallein (eye and subcutaneous), 110 mallein, preparation of eye, sub- cutaneous, 406 reactions, 404 eye, 404 subcutaneous, 405 morphology of, 399 pathogenicity of, 400 spread of, 401 staining of, 399 vaccine, 594 mesentericus, 318 Morax-Axenfeld, 414 found in eye, 135 mucosus (capsulatus), found in lungs, 135 in sputum, 135 immunity to, 333 serum reaction, 333 ozense, 333 immunity to, 333 serum reaction, 333 vaccine, 594 paratyphosus A, 351-352 in animals, 355 B, 351 C, 355 carriers of, 352 classification of, 323 communicability of, 353 diagnosis of, differential, 354 methods of, 355 infections of, 353 in mares, 355 media for, 106-108 pseudotuberculosis in guinea- pigs, 355 reaction to, differential, 324 vaccine, 593 variations of, 354 pasteurianum, 61 perfringens, 448. See B. welchii. pertussis, 417-419 agglutination of, 418 complement-fixation of, 199, 418 diagnosis of, agglutination, 418 complement-fixation, 418 culture, 417 INDEX 691 Bacillus pertussis, diagnosis of, differen- tial, 417 found in lungs, 135 in sputum, 135 pathogenesis of, 418 stain for, 78 vaccine, 594 pestis, 421-425 biology of, 423 epidemiology of, 424 diagnosis of, bacteriological, 425 immunity to, 425 in meningitis, 283 morphology of, 422 pathogenesis of, 423 in pneumonia, 266 resistance of, 425 staining of, 423 vaccine, 593 viability of, 425 phlegmonis emphysematosse, 448. See B. welchii. pneumoniae, Friedlander, 332 immunity to, 333 in pneumonia, 266 serum reactions, 333 prodigiosus, chemical composition of, 50 pigment, 62 proteus (vulgaris), 363, 364 pathogenesis of, 364 vaccine, 594 pseudotetanicus, aerobius, 444 anaerobius, 444 psittacosis, 355 putrificus, 318 pyocyaneus, 361-363 diagnosis of, differential, 363 ' in meningitis, 283 pathogenesis of, 362 pigment, 62 vaccine, 594 radicicola, 651 of rhinoscleroma, 333 immunity to, 333 serum reactions, 333 of smegma, 398 of soft chancre, 419 sporogenes in intestines, 617 streptobaciUi, 32 subtilis, characteristics of, 649 diagnosis of, differential, 433 oxygen absorber, 120 suipestifer, 355 suisepticus, 355, 421 of swine plague, 421 of symptomatic anthrax, 444. See B. anthracis symptomatici. symptomatici anthracis, 444. See B. anthracis symptomatici. tetani, 434-^44 antitoxin, 440 absorption of, 443 formation of, 163 nature of, 173 persistence of, in blood, 441 Bacillus tetani antitoxin, production of, 440 testing serum, 440 treatment, 609 unit, 440 biology of, 435 cultivation of, 435 diagnosis of, differential, 444 examination of case, 444 infection, natural, 437 isolation of, 436 media for toxin production, 110 morphology of, 435 occurrence of, 434 oxygen requirement of, 137 pathogenesis of, 436 in man, 437 resistance of spores, 435 staining of, 435 toxins, 151, 153, 438 action of, 439 potency test, 439 presence in blood, 440 production of, 438 theories of action of, 441 union with nerve substance, 442 of timothy grass, 398 of tuberculosis, 365-393 adaptation of, to soil, 139 in animals, cold-blooded, 390 antiformin method for concen- trating, 391 antituberculous serum, 387 attenuation of, 379 avian (bird), 389 biology of, 366 parasite, 368 chemical composition of, 366 collection of sputum in, 390 complement-fixation in, 200 concentrating methods in, 391- 392 cultivation of, 368, 369 from feces, 393 from urine, 393 diagnosis of, agglutination, 387 complement-fixation, 200 distribution of, 365 examination of, 392 in feces, 392 methods of, 390-393 in milk, 376 in urine, 392 found in brain, 136 in cerebrospinal fluid, 135 in eye, 135 in lungs, 135 in nose, 135 in pelvic organs, 136 in peritoneal fluid, 135 in pleural fluid, 135 in sputum, 135 in throat, 135 in urine, 135 historical note on, 365 692 INDEX Bacillus of tuberculosis, immunity from, 380 immunization against, 382 infection, bovine, in man, 376 conveyed by milk, 375 of fetus, 379 mixed, 379 with dried, 373 moist, 373 inoculation with feces, 393 with urine, 393 isolation of, 370 animal, 370 antiformin, 370 Petrofif, 370 Kinyoim method for concen- trating, 392 localization of, 140 in meningitis, 282 in milk, 641 morphology of, 365 pathogenesis of, 371 in animals, 387 Petroff method, 370 in pneumonia, 266 poison of, 151 prophylaxis of, 387 resistance of, 366 sputum examination in, 390 staining of, 366 reaction of, 76 susceptibility to, 379 toxin, 151, 372 action of, 372 - tuberculin, 383. See Tuberculin, types of, 388 bovine, 388 cultural differences, 388 human, 388 morphological differences, 389 stability of, 389 virulence in calves, 388 in rabbits, 388 vaccine, 595 tularense, 425 typhi-exanthematici, 451-454 antibodies, 453 etiology of, 484 isolation of, 452 morphology of, 451 non-filtrable, 485 pathogenicity of, 453 prophylaxis of, 454 transmission of, by body louse, 454 vaccine, 591 typhi-murium, 355 typhosus, 334r-350 antiaggressin, 156 bactericidal serum for, 162 biology of, 334 in blood, 136 carriers of, 339 classification of, 323 communication of, 340 Bacillus typhosus, complement-fixation in, 201 cultivation of, 335 cultures from blood, 345 diagnosis of, differential, 350 elimination of, from body, 338 ^- in feces, 135, 346 fermentation reactions of, 337 grape-sugar, action on, 64 growth of, inhibited by, 336 in ice, 348-350 immimization of, 342 indol reaction of, 336 isolation of, from blood, 345 ■ — from feces, 347 ^^ from urine, 348 ^ from water, 617 locaUzation of Peyer patches, in, 139-140 . media for, 106-108 in meningitis, 283 in milk, 641 morphology of, 334 "motihty of, 335 neutral red, 336 paratyphosus group, 351-355 pathogenesis of, 337 in animals, 337 human, 337 lesions, 338 in pneumonia, 266 poison of, toxin of, 157 reaction of, differential, 324 resistance of, 334 in rose spots, 348 in spleen, 348 staining of, 334 susceptibility to, 342 in typhoid fever, 337 leukocyte extract, 225 in urine, 135, 347 vaccine, 591 viabihty of, 340 in water, 348 Widal reaction, 342 of typhus fever, 451. See B. typhi- exanthematici. of Vincent's angina, 450 welchii, 447-449 , biology of, 448 cultivation of, 448 isolation of, 448 morphology of, 448 pathogenicity of, 449 xerosis, 303 of Zur Nedden, 415 Bacteremia, 145 Bacteria, adaptation to medium, 139 aerobic, 53 facultative, 53 in air, 621 anaerobic, 63 facultative, 53 bacillus. See Bacillus, capsule, 33. See Capsule, carbonic acid gas atmosphere, 53 INDEX 693 Bacteria cell membrane, 33 substance, 34 characteristics of, general, 27 chemical products of, 60 chemotaxis, 59 negative, 69 positive, 59 classification of, 24, 26 coccus. See Coccus, counting colonies of, 114 definition of, 29 effect of agitation on, 57 carbonic acid under pressure on, 58 of chemicals on, 59. See g,lso Disinfection, of drying on, 58 of electricity on, 57 of hglit on, 56 blue rays, 56 formation of H2O2, 66 ultraviolet, 66 violet, 56 of pressure on, 57 of pure water on, 58 of radium on, 57 of s-rays on, 57 elimination of, from tissues, 147 entrance of, in tissues of host, 142- 146 fermentation, 60 forms of, basic, 31 higher, 41 genera, 24 growth of, 37 poisonous products of. See Toxin, reduction processes, 63 sulphuretted hydrogen, 63 habitat of, natiual, 29 historical sketch of, 17-23 hydrogen atmosphere, 53 identification chart, 124-125 infection, influence of quantity on, 139 mixed, 141 influence of light on, 56 intestinal, in adults, 318 ' Bacillus acidophilus, 317 bifidus, 317 coh, 323 mesentericus, 318 putrificus, 318 development of, 316 enterococcus, 317 food in, influence of, 322 importance of, 316 in infants, artificially fed, 318 breast-fed, 318 Micrococcus ovaMs, 317 variations of, 318 invasion of living host, 142 local effects of, 146 symptoms of, 146 localization of, 139 in milk, 623 Bacteria, morphology of, 30 motility of, 36 organs of, 35 mutations of, 25 nitrifying, 63 nitrogen atmosphere, 53 nomenclature of, 26 nucleus, 34 pathogenic power of, 138 hmitations of, 139 physiological characteristics of, 36 pigment production of, 62 • production of alkaU, 61 of enzymes, 59. See also En- zymes, of ferments, 59 of heat, 59 of Ught, 59 of putrefactive products, 59 of waste products, 59 properties of, for classification, 24 protista, 24 regional distribution of, 135 relation of, to disease, 132, 137-150 reproduction of, 37 in sewage, 662 shape of, 31 constancy of, 31 size of, 30 of soil, 622, 648-652. See Soil bac- teria, sour milk, 319 Bacillus bulgaricus, 319 species of, 24 permanence of, 25 spirilla, 32 structure of, 33 sulphuretted hydrogen, 54 symbiosis, 52 temperature of, influence of, 54 toxicity, decrease of, 141 increase of, 141 variation of, 140 virulence of, decrease of, 141 increase of, 141 variation of, 140 Bactericidal germs, protective value of, 162 Bacteriolysis, relation of, to agglutination, 209 Bacteroids, 64 Balantidium coh, 536 minutum, 636 Barber's method of isolation, 114 Bartonia bacilliformis, 552 Basidiomycetes, 27 Beggiatoa sulphur, 50 Blackwater fever, 548 Blastomycetes, 28 pathogenic, 242. See also Yeasts. Blastomycosis, 242 BMstei" fluid, to obtain, 132, 210 Blood, convalescent, therapeutic use of, 603 obtaining of, for media, 103 organisms found in, 136 694 INDEX Bodo, 518 lacerta;, 518 Bordet theory of hemolysis, ISO Bordet-Gengou bacillus, 417 Botulism, diagnosis of, 450 Bouillon. See Media. Brain, organisms found in, 136 Brilliant green agar.