>A ra 7; -4 4 1 •■ f. '" III 'ifiittiiJiii d CORNELL UNIVERSITV LIBRARY 3 1924 073 264 123 apcRT^ramLiPM comcLL^RiwEiijn The original of this book is in the Cornell University Library. There are no known copyright restrictions in the United States on the use of the text. http://www.archive.org/details/cu31924073264123 Ij I Li? '■ '- ' li^ .•J I. EXPERIMENTS ON THE EFFECT OF FREEZING ^y? .. OTHER LOW TEMPERATURES UPON i'fHE Vl/ilrillTY O^ THE BACILLUS OF TYPHOID FEVER, WITH CONSILFi'ATIONS REGARDING ICE AS A VEHICLE OF INFECTIOUS DIS ' aS,;^ 7l^ (4 "^ iU. II. STATISTICAL STUDIES ON TH:^ SEASONAL PREVALENCE OF TYPHOID FEVER IN VARIOUS COUNTRIES AND ITS RELATION TO' SEASONAL TEMPERATURE. », BY TLLIAM T. SEDGWICK, Ph. D., and CHARLES-EDWARD A. WINSLOW, S. M., Professor of Biology, -.^ Instructor in Biology, In the MA88i£«?|iij setts Institute of Technology, Bd'STON, MASSACHUSETTS. ,^ WITH EIGHT PLATES. Pbf^jEnted Makch 12, 1902. (Prelir- ■'nary Communication, December 13, 1899.) TABLE OF CONTENTS. PART I. FAOE I. INTRODUCTORY 471 II. A REVIEW OF THE LITERATURE RELATING TO ICE AS A VEHICLE OF DISEASE AND TO THE BACTERIOLOGY OF ICE 472 A. Infectious Diseases attributed to Polluted Ice and Ice-cream 472 B. Bacteria in Natural Ice, Snow and Hail, and in Ice-cream 475 C. Experiments on the Effect of Freezing and other Low Temperatures upon THE Viability of Bacteria 478 D. Quantitative Studies upon the Destruction of Bacteria by Freezing and other Low Temperatures 483 III. EXPERIMENTS BY THE AUTHORS ON THE EFFECT OF COLD UPON THE BACILLI OF TYPHOID FEVER 487 A. Experiments on the Percentage Reduction of Typhoid Fever Bacilli effected BY Freezing for Different Periods of Time 487 B. Experiments on the Effect of Alternate Freezing and Thawing upon the Bacilli of Typhoid Fever 499 C. Experiments on the Effect of Temperatures slightly above the Freezing- PoiNT upon Typhoid Bacilli in Water 501 D. Experiments on the Viability of Typhoid Bacilli in Earth at Various Tem- peratures 508 E. Experiments on the Effect of Sedimentation and Crystallization during the Freezing of Typhoid Fever Bacilli in Water 516 IV. DEDUCTIONS FRO.M THE EXPERIMENTS CONCERNING ICE AS A VEHICLE OF INFECTIOUS DISEASE, WITH SPECIAL REFERENCE TO THE PROB- LEMS OF ICE-SUPPLY AND THE PUBLIC HEALTH 519 470 CONTENTS. PART II. PAGE I. A REVIEW OF THE LITERATURE RELATING TO THE SEASONAL PREVA- LENCE OF TYPHOID FEVER 521 II. STATISTICAL STUDIES BY THE AUTHORS ON SEASONAL VARIATIONS IN TEMPERATURE, AND IN THE PREVALENCE OF TYPHOID FEVER IN VARIOUS COUNTRIES 537 III. INTERPRETATION OF THE STATISTICAL RESULTS 567 IV. CONCLUSION OF THE AUTHORS THAT THE SEASONAL PREVALENCE OF TYPHOID FEVER DEPENDS MAINLY UPON SEASONAL TEMPERATURE 669 PART III. BIBLIOGRAPHY 573 A. On Disease atteibdted to Polluted Ice and Ice-ckeam 573 B. On the BacteeiOlogt of Natural Ice, Snow, and Hail, and of Ice-cream . . 573 C. On the Effect of Freezing and other Low Temperatures upon Bacteria . . 574 D. On Quantitative Studies of the Destruction of Bacteria bt Freezing . . . 576 E. On the Seasonal Prevalence op Typhoid Fevbk and its Relation to Seasonal Temperature 576 QA appears to have been the first European to give the matter marked attention, although a recent French writer"' mentions an ice epidemic at "Eveshem," in 1882, of which we have found no other account. Duclaux enlarged at length upon the danger from ice, especially the artificial ice made in Paris from the water of certain highly polluted canals. In 1893 Professor Riche<^' made a long report to the Gonseil d'hygiene et de salubrite de la Seine upon the dangers to the inhabitants of Paris from the sale of highly polluted ice. He quoted a letter from Pasteur as follows : " Le docteur Roux vous a dit son opinion, et c'est aussi la mienne, que toute eau impropre ^ la boisson Test egalement pour preparer, en hiver, de la glace pour 1' alimentation. Les microbes inoffensifs ou pathogenes rdsistent presque tons a des tempi^ratures raeme tres basses." M. Riche showed that much of the Paris ice came from contaminated sources, and recommended strong legal restrictions upon its sale. Finally, Dr. Dorange, in the Revue d' Hygiene,''''^ described a supposed ice-epidemic of typhoid fever at the military post of Rennes in the autumn of 1895. Eight lieutenants of the regiment there stationed were taken ill between the twelfth and the twenty-fifth of December. The fact that these ofiicers did not habitually live in common but had all been present at a regimental banquet upon the fourth of December, pointed to that occasion as the moment of infection. The higher ofiicers dined in a separate room, and used no water but the town supply, which was excel- lent. The lieutenants, on the other hand, drank a " tisane " of champagne mixed with chilled water.- The man who provided this claimed that it also was derived from the regular town-supply. The fact that the town water could be obtained by him only from a considerable distance and under strict police regulations, led Dr. Dorange to suspect that he had made use of the water in a reservoir which stood in the room 474 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVER. where he cooled his decanters and which received the meltings from his stock of ice. The ice supply of the town was considered highly polluted. The additional facts are cited that the menus of the different classes of officers were the same, and that certain of the petty officers who did not drink from the " tisane " but made use of beer instead, escaped the disease. Altogether it appears probable that the milder intestinal disorders, caused by mere decomposing organic matter and not by specific germs, have at times been caused by polluted ice. The Rye Beach epidemic was carefully and thoroughly studied, and leads directly to that conclusion. With respect to typhoid fever the case is different. The only ice-epidemic of typhoid fever which has come to our notice, viz., that at Rennes, rests on a doubtful chain of circumstances, and lacks the con- firmation of a complete exclusion of all possible factors other than ice. We have been unable, then, to find any conclusive evidence that typhoid fever has been caused by polluted ice-supply. A number of English epidemics of typhoid fever, more or less clearly traced to ice-cream, should be noticed here, although the conditions are quite different from those which obtain in the case of ice. The first of these epidemics occurred in the English sanitary districts of Greenwich and Rotherhithe in 1892.<*' During the last week of September and the two months next following 511 cases were reported, the beginning of the attack in 15 per cent of the cases falling on October 1 and in 57 per cent of the cases falling in the fortnight preceding October 3. A remarkably large proportion of the victims were young children. The water supply and sewerage of the four separate foci of infection were different and apparently all in good condi- tion. The milk supply of the households attacked came from seven dairy farms, and in many cases consisted only of condensed milk. Suspicion was then directed to the ice-cream sold by Italians from barrows in the street. A careful canvass of one neighborhood in which 56 cases of typhoid fever had occurred showed that 924 persons lived in houses where ices had not been eaten, 232 lived in houses where ices had been obtained from shops, and 395 in houses where ices had been obtained from a certain ice-cream vendor. All the cases of typhoid fever were in this latter class. A detailed examination of the cases in all the infected areas showed that 88.9 per cent of the sufferers had eaten ices, and that, of these, 91.4 per cent had obtained their supply from ten Italian vendors living in a certain Mill Lane, of whom one was the dealer above mentioned. The sanitary conditions in Mill Lane were found to be abominable ; and in the family of one of the purveyors of ice-cream two children had sickened with typhoid fever on July 29 and August 5 respectively. SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVER. 475 An epidemic of typhoid fever which attacked over 800 persons in the county of Renfrew, in Scotland, in 1893, was attributed by Dr. A. C. Munro partly to ice-cream and partly to the public water-supply. <«) Out of the first 180 cases 63 were shown to have eaten ice-cream prepared by a dealer in whose family a case of typhoid fever had occurred during the previous month. The patient had been in intimate contact with the ice-cream business during the greater part of her illness. Vaughan and Perkins, in 1895/1") ascribed two epidemics of severe, but not fatal, intestinal disease to a new pathogenic bacillus which they isolated from ice-cream in one case and from cheese in the other. The germ belonged to the colon group, and the authors note that neither twenty-nine days of continuous freezing nor alternate freezing and thawing could destroy its vitality. Dr. Hope, in 1898,'"' studied an epidemic affecting 27 school children in Liverpool in which the only clue appeared to be the presence of all the patients at a fair just at the time of infection. Here 24 of the children had eaten ice-cream and two more had partaken of "chip" potatoes sold by an Italian in whose house there had been two cases of typhoid fever. In these cases of infection from ice-cream there is, of course, no certainty that the disease germs were actually frozen. The possibility of contamination from spoons, vessels, and the hands of the vendor might easily account for all the phenomena. Even if the infection was really carried in the ice-cream the exposure to a low temperature must have been a relatively short one. The same reasoning applies to the famous Plymouth, Pa., epidemic of typhoid fever. This little mining town had 1200 cases of the disease and 130 deaths among its 8000 inhabitants in 1885, and the investigation'*^' clearly traced the infection to the dejecta of a single typhoid fever pa,tient which were thrown out on the snow on the banks of the brook supplying the town with water, and which had been washed in by the first general thaw of the spring. It may easily have been that the discharges thrown out during the day or two preceding the thaw were never really frozen at all. In any case the conditions affecting germs imbedded in a solid mass of rich food material are quite different from those which obtain in the formation of ice iipon a stream or pond. B. BACTERIA IN NATUEAL ICE, SNOW, AND HAIL, AND IN ICE-CREAM. In spite of the absence of epidemiological evidence, it has been the common opinion of sanitarians that ice might be an important source of infection for typhoid fever or any other germ disease. Its apparent purity was shown by the earliest bacteriologists to be deceptive. Burdon-Sanderson,'*^* in 1871, found that liquid 476 SEDGWICK AND WINSLOW. — BACILLUS OP TYPHOID FEVER. culture media showed bacterial growth when inoculated with melted ice or with snow. In the next year, Cohn ^"> described experiments in which nutrient solutions containing bacteria were not sterilized by exposure to a temperature ranging as low as —18° C. for about 6 hours or by a temperature with a minimum of —7° C. for 18 hours. Professor Joseph Leidy, in 1884,<*=) exhibited, at a meeting of the Academy of Natural Sciences at Philadelphia, snow water derived from melted ice, containing not only Infusoria but also Rotifers and Worms. Pohl, in the same year,<^*> recorded the finding of many bacteria in snow and ice, 110 per centimeter in Neva ice, and 20,774 in one sample of bubbly ice. He also found bacteria in falling snow, the number decreasing with the continuation of the storm. A report on the ice supply of the city of Syracuse ^"^ was made to the New York Board of Health in 1886 in which the presence of a great number of bacteria was noted in ice from Onondaga Lake and the Erie Canal. In 1888 Breunig<'*' found 1310-2760 germs in ice, and Kowalski^"' analyzed sixty samples of natural ice, and found from 10 to 1000 germs per cubic centimeter, no sample being sterile. Still another paper was published at this period, 1888-89, by Heyroth,'^^ who studied the Berlin ice-supply, and, in 25 samples, found from 2 to 133,000 bacteria per cubic centimeter, the highest figures corresponding to chemical analyses, which showed the most marked pollution. An elaborate report was made by the State Board of Health of Massachusetts in 1889,^^^ in which 238 samples of natural ice from the ponds and streams of this State were analyzed bacteriologically. The figures for ice from different portions of the cake were as follows : — Number of Samples. Bacteria per c.c. Transparent Ice Clear Ice . . Bubbly Ice . . Snow Ice . . 27 75 113 23 Haximum. 893 370 1950 2968 Minimum. ATenge. 106 16 111 622 A " Lancet " analytical sanitary commission made an examination, of some ice sold in London in 1893, and found that while all the specimens gave good chemical analyses, two out of the six examined contained 400 to 700 bacteria per cubic centime ter.^^^ Girard and Bordas'^' published some startling analyses of the Paris ice-supply also in 1893. They found a minimum of 23,000 colonies and a maximum of 100,000 colonies per cubic centimeter, including the Bacillus coli communis and a patho- SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVER. 477 genic vibrio. These quantitative results are so large as to suggest that the sam- ples were probably not planted promptly after melting. Christomonas<^' has recently studied artificial ice, and reports that when water contammg 71 bacteria per centimeter was frozen, 450 germs per centimeter were found in the central core and 8-10 in the clear ice at the sides. The bacteria of snow and hail have also received considerable attentifOn. Soon after the work of Pohl,"«' Janowsky'^^) made analyses of old and of freshly fallen snow in the neighborhood of Kiew, and found bacteria in both, less in the former than in the latter. Schmelk^^' studied the bacterial life in the snow of a Norwegian glacier and in the chill streams flowing therefrom ; and in a later paper ^^'^ he recorded small numbers in both snow and ice at Christiania. Bujwid<2»> found 21,000 bacteria per cubic centimeter in the analysis of a melted hail-stone; and Foutin<^> in Russia obtained similar, though smaller, figures. Giacosa'^^ found bacteria present in small numbers in snow lying at an elevation of 3800 meters above the sea, and Abbott ^'^' noted 703 colonies per cubic centi- meter in hail. Dominguez,*^) in 1892, published a paper on the bacterial content of hail ; and finally, Scofone/^' who accompanied a scientific expedition to Monte Rosa in 1894-95, recorded the presence of small numbers of bacteria in melted snow obtained froih high altitudes. In the following year he gave the results of some examinations made on a plateau 2460 meters above the sea, which confirmed his previous conclusion that the bacteria in the deeper layers of the snow were somewhat more numerous than in the superficial layers.^^*^ The number of bacteria present in ice-cream has been shown at times to be enormous. Klein ^*^) found the germ content of London ice-cream very high, and B. coli communis frequently present. Nield-Cook'^' recorded from 5,000,000 to 14,000,000 germs per cubic centimeter in ice-cream from the same source, the majority being colon bacilli. Stevenson ^*'> testified, at the trial of an Italian ice- cream vendor, that he had found over 4000 germs per cubic centimeter, of which three proved to be B. coli communis. Wilkinson ^**^ reached similar results, and quoted, without reference, the following results of other observers : — Macfadyen 119,000 — 7,000,000 bacteria per culjjc centimeter. Kanthack 8,000,000 - 13,000,000 " " « « Foulerton 600,000- 7,000,000 " " " « In this connection it may be interesting to note the very small numbers of bacteria present in the air and water of the Arctic regions. Nystrom '■^^ discovered this fact in 1868 by the exposure of a number of flasks of putrescible matter, after the 478 SEDGWICK AKD WINSLOW. — BACILLUS OF TrPHOID FEVEE. manner of Pasteur. Couteaud^*' found but one colony in 19 flasks exposed to Arctic air, the experiment being carried on, however, on the open sea, so that the result is not surprising. He also found but few species present in some analyses of water and of soil. In the Nansen expedition the poverty of the bacterial flora of the air was noted. Finally, Dr. Levin^*^' of Stockholm made an elaborate study of the subject with the Natthorst expedition. In 21,600 liters of air examined at twenty different places 3 germs alone were found, all in one sample. In sea water, at the sur- face, 11 germs per centimeter occurred, belonging apparently to two characteristic species. Fresh water and melted ice and snow gave similar small numbers. Samples from considerable depths in the ocean showed somewhat higher numbers than were obtained at the surface. Finally, tests of the alimentary canals of various Arctic animals and birds showed many of them to be completely sterile. C. rXPEEIMENTS ON THE EFFECT OF FREEZING AND OTHER LOW TEMPERATURES UPON THE VIABILITY OF BACTERIA. Laboratory experiments have confirmed the conclusion, drawn from the examina- tion of natural ice, that freezing is by no means always fatal to germ life. Von Frisch^*^^ froze putrefying solutions and reduced the frozen mass to a temperature of —87' C, and after some hours found that sterilization had not ensued. Pictet and Young ^**' subjected bouillon cultures of several species to a tempera- ture below —70° C. for 108 hours, during twenty hours of which time the temperature was below —130°. After this treatment B. anthracis and the bacillus of "charbon symptomatique " were alive and virulent ; B. subtilis and B. ulna grew readily ; half the inoculations made from the cultures of two species of micrococci grew and half did not. Finkler and Prior ^**^ stated that the vibrio described by them could survive a temperature of —4' C. for many days. McKendrick,^*^' in a communication to the British Association in 1885, noted that putrescible liquids were not sterilized by a temperature of —84° C. Forster^*"? found that the phosphorescent bacteria which he isolated from fish preserved by cold storage grew vigorously at 0° C. Fischer ^*^^ isolated 6 species of bacteria from the water of the harbor at Kiel, and 9 other forms from the soil, all.capable of multiplying at 0°. In the research already cited,''*) Hey- roth froze gelatine stick-cultures of various species for from seven to ten days, and then placed them once more xmder favorable conditions ; out of 30 species, thus treated, 26 showed growth, though 5 of these had partially lost their liquefying power. D'ArsonvaV**' in 1891, recommended liquefied carbonic acid for use in steriliz- ing organic extracts, and stated that when the treatment is prolonged, especially SEDGWICK AND WINSLOW. BACILLUS OP TYPHOID FEVER. 479 if broken by a return to 40° for a time, " nothing living can resist it," but his own and other later researches showed the error of this conclusion. Forster, in 1892,**'^ examined various natural waters, foods, wastes, sweepings, and soils for bacteria capable of growth at 0°, and found a few such forms in water, earth, and street sweepings. When present at all they occurred in great nmnbers. Forster also demonstrated the multiplication of bacteria and the progress of decomposition in butcher's meat chopped up and kept in an ice calorimeter. Fischer ^^> noted that Miller's vibrio and the vibrio of Finkler and Prior could withstand a freezing tempera- ture for some days. Pictet, in 1893,'®^' studied the effect of cold on plants and animals of the most widely separated classes. Of the bacteria he subjected 30 to 35 species to tempera- tures ranging as low as —200° C. by immersing them in liquid air, but the viability of the germs used appeared unaffected after " prolonged " treatment of this sort. D'Arsonval and Charrin'^^* subjected cultures of Bacillus pyocyaneus to a temperature of —40° to —60° C. with the result that, in six out of eight instances, the germs remained alive. In another paper ^®*' these authors mentioned that Bacillus pyocyaneus after exposure to —40°, —60°, and —95° C. exhibited profound changes in morphology and physiology. For some generations the descendants of the frozen germs showed elongated, ovoid, and other abnormal forms, and their colonies on gelatine were also of unusual character. Weber '^' noted that Hofer's bacillus, producing a contagious disease among Crustacea, can endure a temperature of —40° C. for four hours, as well as repeated thawings and freezings. Professor Mason '^' recorded the exposure of cultures of "ordinary bacteria" to the temperature of solid carbon dioxide for many hours without causing their destruction. Still more recently Ravenel '^> submitted cultures of the anthrax, diphtheria, and typhoid bacilli, and of Bacillus prodigiosus to the temperature of liquid air, 191° below zero Centigrade, for periods of three hours, thirty minutes, one hour, and one hour respectively ; in no case could any weakening of the vegetative power of the culture be detected. Besides Pictet and Young ^*^^ and Ravenel ^^ a number of other observers have tested the effect of low temperatures upon specific pathogenes. Cad^ac and Malet'"' found that tuberculous matter kept frozen for four months still produced charac- teristic symptoms in guinea pigs. In some work on the spores and vegetative forms of Bacillus anthracis carried out by one of the Franklands and Dr. Templeman,^'*^ it was found that a single freezing at —20° C. reduced the numbers present in water from 480 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEYEK. 15,000 to 3500 per cubic centimeter, and after 29 successive freezings, extending over a period of three months, 3000 germs per centimeter could still develop. Evi- dently the vegetative forms vi^ere killed by one freezing, and the spores, not at all. Another culture which was spore-free showed reduction from 8000 germs per centimeter to 2 per centimeter after one freezing, sterilization following the second freezing. Gabritschewsky, Wladimiroff, and Kressling and Gladin quoted by Kasansky^®*^ found that the plague germ could bear an artificial cold of —22° C. for two hours and natural cold ranging from 0° to —20° C. for from twelve to forty days. Kasansky himself in 1897-98 made some interesting experiments on the resistance of the specific organisms of plague and diphtheria against cold. The cultures were placed outside the window of the laboratory at Kasan, sheltered from light but exposed to the winter's cold, which ranged from a maximum of 5° C. to —34° C. Bouillon cultures of the plague germ showed life after thirty-two days ; four months' exposure sterilized most of the tubes, but in one case growth was obtained after six months. Of the agar cultures tested some died in four months, and others contained living germs after five months and a half. Sixteen bouillon tubes of the diphtheria bacillus were kept for six months under similar conditions, and one tube only showed growth at the end of that time ; two of the others, however, still gave positive results on the fifty-third and one hundred and eighteenth day, respectively. Abel'™^ exposed cultures of the diphtheria germ on blood serum and on dried threads to the winter's cold at Greifswald, and compared them with cultures kept in the room in the same condition. The first race used persisted on the blood serum for the whole period of eighty-six days both in the room and out of doors, although in the second case the growth obtained was meagre after the fiftieth day. The dried germs had disappeared by the sixty-eighth day out of doors and by the seventy- fourth indoors. Of the second race the serum culture remained alive in the room all through the experiment; the frozen one showed no growth after the seventy- fourth day. The threads gave living germs up to the seventy-fourth day in-doors and up to the fifty-sixth day out-doors. The threads of the third race gave precisely the same result; the serum cultures kept in the room gave vigorous growths up to the end of the experiment, while only two colonies developed from the inoculation of the frozen tube. The out-door temperature during the experiment varied from 12° C. to - 20° C. With regard to the behavior of the typhoid bacillus in ice, there is more evidence available. Dr. Carl Seitz(«>> noted in 1886 that cultures of this organism in gelatine, bouillon, and milk were not rendered sterile by the continuance of a temperature SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVER. 481 below 3° C, although the growth on gelatine at the low temperature was very much retarded. Dr. Billings, in this country/®^) described a single experiment in which five cubic centimeters of sterile water were inoculated with the typhoid germ and frozen by the out-door cold. On the next day the frozen mass was thawed, and three gelatine tubes and one agar tube were inoculated with portions of it. Three of the four tubes showed typical growths. Chantemesse and Widal '■^^ recorded the freezing of bouillon cultures of the same microbe without sterilization. Bashenow <^^ stated that typhoid germs survived exposure for thirteen days to a temperature between —8° and —15° C. Janowsky published in 1890 some very extended researches '^> in which he used pure cultures of the typhoid bacillus in bouillon and froze them by means of ice and salt, ice and chloride of calcium or carbon dioxide and ether. He made no quantitative estimations ; but bouillon frozen by each of the above methods could still produce growth in Esmarch roll-tubes. Janowsky tried also the effect of successive freezings, using the calcium-chloride mixture. After the culture had solidified, it was left in the freezing mixture for fifteen minutes, then thawed in a wfiter bath at 25°— 30° C, a sample taken, and the cycle repeated. This was done three times a day ; and during the night the culture was kept at 2°— 5° C. After twelve such freezings sterilization had not been accomplished; the development of the frozen bacilli was, however, much retarded. To imitate more closely the conditions in nature, Janowsky placed a bouillon culture and two flasks in which were threads bearing the germ in a dried condition, in a wire cage out of doors. Four sets of experiments were conducted, in three of which periods of seven, ten, and twelve days, respectively, did not suffice for sterilization. In the fourth set of cultures the bouillon tube showed no growth after nineteen days ; the minimum temperature during the period had been —17° C. and the maximum 4°, the culture thawing and freezing three timesJ Finally, among ex- periments on the typhoid bacillus must be mentioned a remarkable paper by Rem- linger,^^^ in which he states that he used a culture of B. typhi of such virulence that .5 c.c. would kill a guinea pig in 36-48 hours. He took agar cultures of this germ out of the incubator every two or three hours to immerse them in water, cooled down to 22°-23°, for ten minutes. After ten days of this treatment the cultures had entirely lost their virulence, and after thirty-five days their power of growth as well. The author does not state whether control experiments were made or not. Even more extensive is the literature with respect to the effect of cold on the cholera vibrio. Koch, the discoverer of the organism, stated that it was not destroyed by a temperature of -10° C. in ten hours.^*'> Raptschewski^^' found that cholera germs could endure for a month severe cold, ranging as low as -15° C, but that a tempera- si 482 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVER. ture of —21° C. was fatal. Von Babes'®'^ succeeded in keeping a series of agar cultures of the vibrio alive, though exposed to the cold of a Berlin winter (1884-85) ranging as low as —14° C. In the year 1893 no less than eight papers were published dealing with the relation of the cholera germ to cold. SchrufE '™^ found that a broth culture made from fresh choleraic faeces was not sterilized by eight months' exposure to the winter's cold ranging as low as— 12.5° C. Finkelnburg^"' noted that cultures of an old laboratory race were killed out in ten days, while cultures of fresher races were not. Karschinski '■''^^ stated that a cholera culture with which he worked was sterilized in four days by an average cold of —12.7° C. with a minimum of —17.6° C. Renk ^^^^ froze the germs in sterilized river water at —5° C. to —7° C. and kept the flasks at that tem- perature, removing one each day for examination. Growth resulting from the melted ice was tested by cover-glass examination and by the Indol reaction. Aftei* five days' uninterrupted freezing the cholera germs disappeared, but when the period was broken by the melting of the contents of a flask for analysis and its re-freezing, a little longer period was necessary. When imsterilized river water was inoculated and frozen, the bacteria present fell off from 1,483,000 per centimeter to 62,445 in twenty-four hours, and to 4480 after three days. The cholera germs in this case could not be de- tected after seventy-two hours, and in one case not after thirty-nine hours. Ufielmann '•''*^ found that cholera germs died out in five days at— 15.5° C. and in three days at —24.8° C. Wnukow,^*^ on the other hand, stated that gelatine stick cultures of the same micro- organism were subjected for forty days to an outdoor temperature between —1° C. and —32° C. without sterilization. Double thawing and freezing also failed to destroy their power of growth. Montefusco '■''^ tested the pathogenicity of chilled cholera cultures for guinea pigs, and recorded that a temperature of —10° to —15° C. entirely destroj'^ed their virulence in half an hour, while a temperature between 0° and -5° only weakened it. Cultivation at 37.5° soon restored the powers of the germs, but in the chilled and attenuated condition they produced a state of immunity in the animals injected. Abel^' also mentions experiments in which cholera vibrios frozen in bouillon died out completely in from three to eight days. Kasansky,"*> in 1894, found that cholera cultures withstood for four months the winter's cold at Kasan, where the temperature fell to —31.8° C. One culture gave growth after twenty days of freezing. Some were thawed and refrozen as many as twelve times. After longer exposure, for five months, the cultures gave no growth. Kasansky demonstrated nearly as great a resistance to cold in the case of the vibrios of Finkler-Prior, Miller, Deneke, and Metschnikoff. Finally, some light was thrown on the discordant results of previous observers by the work of Weiss,^'' who inoculated tubes of broth and water from SEDGWICK AND WINSLOW, — BACILLUS OF TYPHOID FEVEE. 483 the Spree with cholera cultures and froze them, thawing, sampling, and refreezing the tubes daily. In broth the germs persisted for twenty-one days, but in river water only for five days, the addition of a little broth to the water prolonging the time to eight days. Fresh intestinal contents of a cholera patient showed no vibrios after two or three freezings.^ From this long series of experiments it is evident that sterilization of rich cultures of bacteria cannot always be secured by the action of even very extreme cold. Hence the conclusion was drawn that the freezing of water could not be trusted at all to remove its bacterial impurities. There are, however, two objections to this line of reasoning. In the first place, the effect of cold on germs suspended in water may diflFer materially from its action on similar organisms when in a richly nutrient medium. In the second place, even if sterilization does not result from freez- ing in cultures containing millions of bacteria, it is conceivable that such a large proportion of the microbes may perish as to render very slender the chance of danger from ice formed under natural conditions. Experiments have shown that easily detected germs like B. prodigiosus can pass through a sand filter when applied to the surface in large numbers under certain conditions; yet a sand filter, in prac- tice, is regarded as an efficient protection. A quantitative determination of the per- centage reduction actually effected by freezing is required before drawing conclusions as to the sanitary significance of ice-supply in relation to the public health. D. QUANTITATIVE STUDIES UPON THE DESTKUCTION OF BACTERIA BY EEEEZING AND OTHER LOW TEMPERATURES. The quantitative studies of Frankland <^' on B. anthracis, of Eenk "'^ on river- water bacteria, and of Christomonas,'^' on artificial ice, have already been mentioned. Work on the disappearance of bacteria in the freezing of natural water had, however, been undertaken at a much earlier period. Pengra,^*'> in 1884, made an actual microscopic count of the organisms present, working with bacteria (species not stated), and other micro-organisms from decomposing meat juice, infusion of hay, and stag- nant pools. His freezing was done by the winter's cold, and his figures were obtained by counting the contents of ten drops and taking an average. He found 1" Macfadyen (Lancet, I, 1900, p. 849) has recently exposed cultures of Bacillus typhi, BaciUus coli commu- nis Bacillus diphtherias. Spirillum choleras asiaticas, Bacillus proteus vulgaris, Bacillus acidi lactici, Bacillus anthra- cis (spore bearing), Staphylococcus pyogenes aureus, BaciUus phosphorescens, and Photobacterium balticum in solid and liquid cultures to the temperature of liquid air (-182° C. to -190° C), for twenty hours without sterilization and without impairing the properties of the organisms in any degree. Macfadyen and Rowland (Lancet, Vol. I, 1900, p. 1130) treated the same organisms in broth emulsions in fine quill tubes with liquid air for seven days with the same results, except that a slightly delayed growth was noticed in some instances. 484 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVER. in the upper part of the ice 16 bacteria; in the lower part, only partially frozen, 250; in the upper and lower parts of a duplicate unfrozen vessel of water, 160 and 170, respectively. He obtained similar results with three species of Infusoria, and concluded that 90 per cent of the organisms were removed by freezing. His experi- ments appear, however, to show crystallization effects principally. The first careful work on this subject was done by Fraenkel in Berlin.^^^^ He collected river water, and after planting samples, froze them artificially at —8° to —12° C, thawing after different periods. In two days 83 per cent of the water bacteria present were killed ; in three days 99 per cent ; in five days, 90 per cent ; in six days, 80 per cent ; in six days, in another case, 93 per cent; and in nine days, 99 per cent. The different samples evidently varied greatly. Fraenkel also analyzed the regular Berlin ice-supply, and got results ranging from 21 to 9700 bacteria per cubic centimeter. He concluded that the ice was highly polluted and should not be taken into the system. About the same time Wolff huegel and Riedel'*^' gave an account of some experiments in which flasks of tap-water were kept in the ice-chest without freezing, and showed the following reductions : after one day, from 148 germs per cubic centimeter to 126 and from 150 to 115 ; after two days, from 123 to 69 and from 158 to 101 ; after three days, from 123 to 29 and from 156 to 33. In 1887 Dr. Prudden of New York published the most exhaustive review hitherto attempted of the subject of quantitative reduction, and the first in which specific pathogenic germs were used.<^> His tubes, in the experiments with the latter organisms, were inoculated from pure cultures and frozen at —10° to —1° C, and his results were as follows, the numbers in each case referring to bacteria per cubic centimeter : — B. prodigiosus. In water, 6300 ; in ice after 4 days, 2970 ; after 37 days, 22 ; after 51 days, 0. Proteus vulgaris. In water, 8320 ; in ice after 18 days, 88 ; 51 days, 0. Staphylococcus pyogenes aureus. In water, innumerable ; in ice after 18 days, 224,598; 20 days, 46,486; 54 days, 34,320; 66 days, 49,280. Species unnamed. In water, innumerable ; in ice after 4 days, 571,450 ; 11 days, 520,520; 51 days, 183,040; 65 days, 10,978; 77 days, 85,008. Species unnamed. In water, 800,000; in ice after 7 days, 0. B. typhi. In water, innumerable; in ice after 11 days, 1,019,403; 27 days, 336,457 ; 42 days, 89,796 ; 69 days, 24,276 ; 77 days, 72,930 ; 103 days, 7348. Same. In water, 378,000; in ice after 12 hours, 164,780; after 3 days, 236,676; 5 days, 21,416; 8 days, 76,032. Dr. Prudden then made certain experiments to determine the effect of alternate SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVER. 485 freezing and thawing, and obtained the following results. The tubes were here im- mersed in ice and salt at -20° C. B. TYPHI. In water . . . 40,896 Frozen 24 hours 29,780 Eefrozen 3 times 90 " 3 days 1,800 " 5 " « 4 " 950 " 6 " " 5 « 2,490 " 6 « B. PKODIGIOSUS. In water . . . 339,616 Frozen 24 hours 36,410 Eefrozen once 2,570 " 30 " 41,580 " 2 times 275 " 48 " 14,440 " 3 " 15 " 96 « 4,850 « 4 « STAPHYLOCOCCUS PYOGENES AUREUS. In water . . . 111,782 Frozen 15 minutes 52,500 " 2 hours 21,300 " 24 " 22,690 Eefrozen once 13,495 " 48 « 6,460 " 3 times 110 " 96 " 6,155 • « 4 « Dr. Prudden found that, with fresh, active agar cultures of this staphylococcus 49,280 germs remained alive, out of innumerable germs originally present, after sixty days ; when cultures from old and dried agar were used, 162,000 germs dis- appeared entirely after five days. He ultimately drew the following conclusions from these experiments with pathogenic germs : 1. Many bacteria are killed by freez- ing. 2. The vitality of the original culture affects the number so killed. 3. The number killed varies with the species. 4. The number killed increases as the time of freezing is prolonged. 5. The resistance to cold varies with the individual bac- terium. 6. Alternate freezing and thawing is very generally fatal. Dr. Prudden also froze natural waters with their native bacteria for varying periods, and obtained somewhat similar results. He analyzed 270 samples of New York ice, and found an average of 2033 bacteria per cubic centimeter. The numbers were highest in the upper layers of snow ice and bubbly ice, and in ice cut in the immediate vicinity of Albany, falling off rapidly in ice five or six miles down the river. He concluded that this highly polluted ice probably contained the germs of typhoid fever and should not be taken into the human body. 486 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVEK. Later in the same year Bordoni-Uffreduzzi '**' published a paper in which he took issue with Prudden on several points. He contended that the changes of temperature in the latter's experiments were too abrupt, that the resistance of the germs worked with had been weakened by cultivation on artificial media, and that the effect had been abnormally severe on account of the small size of the tubes frozen. He himself analyzed the natural water in one of the municipal basins of Berlin, just before a frost, and then kept a large lump of the ice in a double- walled zinc chest, break- ing off samples for analysis every month. He found that about 90 per cent of the bacteria were killed, and thought the duration of the freezing did not make any material difference. His results, of course, varied very widely on account of the unequal distribution of the bacteria in the ice. Kussell ^^^ a little later made similar experiments at Madison, Wisconsin, in which he found that the ice formed on Lake Mendota contained about 40 per cent of the germs present in the water itself A report already cited '^'^ was made by the State Board of Health of Massachusetts in 1889 in which ice from fifty-eight sources was analyzed in comparison with the water on which it had formed. Averaging all results, there were 81 per cent as many bacteria present in the snow ice as in the water, 10 per cent in all the rest of the ice, and only 2 per cent in the clear ice. In the report of the Board for the next year,^^^ Mr. Hiram F. Mills noted an isolated but significant experiment in which sterilized tap water was inoculated with the typhoid germ, kept in a bottle surrounded by ice and sampled at intervals. The results were as follows : — Day 1 5 10 Number of Typhoid Bacilli. 6120 3100 490 Day 15 20 25 Number of Typhoid Bacilli. ■100 17 Taken altogether, more exact studies confirm the rough estimate of Pengra that some 90 per cent of ordinary water bacteria are eliminated by the process of freezing. As to the percentage reduction of specific pathogenes and, in partic- ular, of the typhoid bacillus, probably the only form of great practical importance, the evidence is very meagre. The only results hitherto, as far as we have been able to discovei', which fix quantitatively the effect of cold on this organism, are the three experiments of Dr. Prudden and the single experiment of the biologists of the Massachusetts State Board of Health. These certainly appear to form a slender basis for conclusions relative to the importance of ice-supply as a possible source of typhoid fever. SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVER. 487 III. EXPERIMENTS BY THE AUTHORS ON THE EFFECT OF COLD UPON THE BACILLI OF TYPHOID FEVER. A. EXPERIMENTS ON THE PEECENTAGE EEDUCTION OF TYPHOID FEVER BACILLI EFFECTED BY FREEZING FOR DIFFERENT PERIODS OF TIME. Methods Employed. The following investigation was undertaken in order to so extend and amplify the work of Prudden as to obtain some idea of the average fatality occurring among typhoid bacilli in ice, and of the special conditions which affect such fatality. Pure cultures alone were used, as it is obvious that figures, to be of much value, must be determined separately for each specific germ. Great pains were taken to preserve, as far as possible, the vigor of the culture used, and new cultures from recent post-mortem examinations were obtained at intervals during the work. Finally, a large number of determinations were made for each set of conditions, in order to obtain average results free from the errors which may beset any individual case. Our experiments on the percentage reduction effected by freezing were carried on by freezing small tubes of infected water, as only, in this way can the con- ditions of the experiment be rigidly controlled. Ordinary test-tubes, containing about 10 cubic centimeters of sterilized tap water, were inoculated from a two or three day bouillon culture, and duplicate samples were at once planted. The ten tubes of the set under experiment were then placed in a double-walled tin vessel in which they were to be frozen. The inner vessel was a cylinder about 8 inches deep, nearly filled with a mixture of equal parts of glycerine and 95 per cent alcohol ; in this solution the tubes were immersed, being supported by a disc per- forated with holes to receive them. The solution served to make the lowering of temperature equal and gradual, and also acted as an antiseptic when the tubes broke, which sometimes happened when they contained too much water, or when the temperature went down too rapidly. In the outer vessel, which was jacketed with felt, was placed cracked ice which reduced the temperature of the glycerine-alcohol mixture to about 10°— 15° C. in from an hour to an hour and a half The ice was then replaced by a mixture of ice and salt which completed the freezing 488 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVER. in a half or three-quarters of an hour more. The time occupied by the whole process of freezing is recorded in the tabulation of each experiment. The tem- perature, in the first set of experiments with " Race A," was observed by means of three mercury thermometers inserted in different parts of the liquid, and at the time when the tubes froze the thermometers registered 6°— 7° below zero, C. In later experiments the temperature was observed by means of a minimum regis- tering spirit thermometer fastened to the inside of the cover of the inner cylinder, which recorded the temperature of the air just above the liquid in which the tubes were immersed. Partly on this account and probably partly because of its greater quickness of response, this thermometer gave lower records than did the mercury instruments in the first experiments. The readings of the spirit thermometer are given in the tables for each set of tubes. As soon as the tubes froze, they were removed from the freezer and either thawed at once or kept frozen in an ice-chest for a few hours, or placed in a cold-storage ware- house where they were kept for the longer periods at a temperature one or two degrees below zero, C. After the frozen condition had been maintained for the de- sired length of time, the contents of the tubes were thawed, shaken up, and sampled, again in duplicate. As a rule the samples taken from the thawed tubes were planted directly, while those made before freezing were diluted, one to ten, with sterilized ~ water. All plates, for these quantitative determinations, were planted with common nutrient agar-agar, containing 1.25 per cent agar, 1.00 per cent Witte's peptone, and .25 per cent salt, and having an acidity equal to 1.50 per cent. As the counts to be made were chiefly comparative, agar was preferred to any other medium, on account of its freedom from liquefaction. The plates were allowed to develop at the room temperature except in certain special cases to be noted later. Those made from the unfrozen water showed their maximum growth in three days and were counted after that intervaL Those made from the thawed ice, however, were found to develop more slowly ; for them five days was generally found sufficient, although after the longer periods of freezing as much as ten days was sometimes allowed. The plates were finally counted with the aid of a hand lens. In many of the sets of experiments a control tube was included, which was treated just like the others except that it was not inoculated. Each series of tubes includes two lots of eight or ten each, frozen on two different days. The cultures were grown in bouillon (containing 1.00 per cent peptone, .25 per cent salt, and 1.00 per cent acid), and were changed twice or three times a week. In the earlier experiments the tubes were inoculated from a culture grown at the room SEDGWICK AND WINSLOW. BACILLUS OP TYPHOID FEVEK. 489 temperature, itself inoculated from one grown at 37.5° C. In the later work the cul- tures were all kept at the room temperature. When experiments made on the culture obtained in November, 1898, gave results somewhat diflferent from those given by the culture used in February, it was decided that still a third culture from a different source must be compared with the first two. The results showed that the descendants of these different stocks exhibited slight though constant and persistent differences in their reaction to cold. We have called the cultures derived from these original sources " Races," for physiological races they apparently must be considered. The first culture used, Race A, was obtained from the Boston City Hospital as a forty-hour-old blood-serum culture on February 23, 1898. Unfortunately, the history and tests applied to this culture in the Hospital were not recorded, beyond the fact that it had been isolated from an autopsy about two weeks previously, by the usual differential methods. Race B was obtained by the kindness of Dr. M. W. Richardson of the Massachu- setts General Hospital in the middle of November, 1898, with the following history. It had been isolated from the spinal canal, in a case of typhoid meningitis. It gave typical reactions in media as follows : bouillon, very motile; litmus milk, no coagulum, slight acid production; sugar-agar, no gas; peptone solution, no indol ; gelatine slant stab, typical growth, no liquefaction ; arsenic bouillon (Thoinot), no growth ; Capaldi- Proskauer sol. No. 1, no growth ; potato, no visible growth ; tube medium of His, clouding without gas production ; typhoid serum, perfect reaction. Race C was obtained, January 14, 1899, by the courtesy of Dr. Pratt of the Boston City Hospital. It had been isolated, December 30, from the peritoneal cavity in a case of peritonitis following typhoid fever. It gave typical growths on the ordinary media, gelatine, bouillon, and glycerin-agar ; it was motile in the hanging drop ; it gave no indol and no gas in glucose solution ; it was decolorized by the Gram method and reacted to typhoid serum. Race D was isolated in the laboratory of the City Hospital, March 26, 1899, from the urethra. It was identified by the same tests used for Race C. Results Obtained. The percentage reductions recorded in the subjoined tables (pp. 492-498), sum- marized in final form, are as follows : — 490 SEDGWICK AND WINSLOW. — BA.CILLUS OP TYPHOID FEVEK. PERCENTAGE REDUCTION OBSERVED IN EXPERIMENTS ON THE VIABILITY OF TYPHOID BACILLI IN ICE. Race A. B. C. D. Frozen 15 minutes 59.4 13.8 63.7 " 30 <( " 1* hours 2 « 3 » 6 It " 12 It " 15 it " 24 tt 3 days 7 (( 2 weeki 4 (( " 8 tt « 12 tt 32.2 73.6 77.8 99.8 99.8 99.8 99.8 41.4 99.5 74.8 97.0 38.6 98.0 84.4 53.8 82.7 99.0 98.4 99.9 93.3 99.5 99.4 99.9 Conclusions. 1. Evidently we may reaffirm for the bacillus of typhoid fever the first of Prudden's conclusions as to the various pathogenes with which he worked, namely, that many bacteria are killed by freezing. After two weeks' exposure to the freezing tempera- ture an average of considerably over 99 per cent of the germs perished. Of the 140 tubes inoculated with Races A, B, and C, and frozen for periods of two weeks and over, all but nine showed a reduction of over 99 per cent ; and of the nine, all but one showed a reduction of 98 per cent or over. We may safely conclude that less than 1 per cent of the typhoid germs present in water can survive fourteen days of freezing. 2. During the first half-hour of freezing a heavy reduction takes place, amount- ing, perhaps, to 50 per cent. The tubes exposed for such short times to the un- favorable conditions exhibit a remarkable variability among themselves. In the same set one tube may show no reduction, while its neighbor is rendered almost sterile. Whether these differences are due to the varying physical conditions in the individual tubes, or to variations in the biological character of the loopful of bacteria used for inoculation, is uncertain. From the general harmony of the results obtained it appears that this factor of variability, whatever it may be, is practically eliminated by the averaging of 20 tubes. After this brief period of sudden but uncertain reduction, the destruction of the germs proceeds pretty regularly as a function of the time. Although the different races vary, there is in each race a steadily increased reduction, with slight variations, as the time of freezing is prolonged. After 14 days, even with the most resistant SEDGWICK AND WINSLOW. BACILLUS OF TYPHOID FEVER. 491 stock, Race B, the reduction was over 99 per cent. The reduction now proceeds, however, with increasing slowness ; the two or three germs per thousand which have survived thus far appear to possess special powers of resistance. Even after 12 weeks few of the individual tubes were rendered sterile. These results appeared so remark- able that special experiments were conducted to test their accuracy, as it was felt that perhaps the few germs developing from the thawed ice might have been intro- duced from the air, as was obviously the case in some instances. Fifty tubes of Races B and C were therefore frozen for periods of a week and a month ; plates were planted from them, with special precautions, and incubated at 37.5°; and the developing colonies were examined individually. The results, as the appended tables show (see p. 492), confirm those of the general investigation. Of the 20 tubes inoculated with Race B and frozen for a month, 10 were sterile ; 9 gave one sterile plate, and one with one or two colonies of what proved to be extraneous germs ; tube IV. alone gave, on one plate, 7 germs per cubic centimeter, which examination in the hanging drop, and growth on gelatine, and potato, in milk and glucose solu- tion, showed to be the original typhoid culture. So of the 30 tubes of Race C frozen for a week, 17 were sterile ; 9 showed contamination, one or two germs per plate ; the other four showed 15, 4, 1, and 267 typhoid bacilli per cubic cen- timeter. These experiments confirm the results of those observers who froze typhoid cultures containing millions of germs without effecting sterilization. 3. Prudden's statement that the number of bacteria killed by freezing varies with the species may be extended. It is evident that within the species B. typhi abdominalis there are races, each having a power of resistance of its own, depen- dent upon its history within and without the body. A comparison of the tables for the shorter periods of freezing shows clearly that Race C succumbed with much greater readiness to the influence of cold than did Race B ; while Races A and D occupied an intermediate position. These differences appear constant through the various sets, so that in each race the progressively increased reduction with more prolonged freezing follows a parallel course. The facts cannot, we think, be at- tributed to differences in the immediate environment of the germs ; such differ- ences do produce their effect, cultivation for a time on agar, for example, causing a decrease in resistance. The last sort of change is, however, temporary and may be quickly reversed by cultivation in bouillon ; while the race differences were permanent during the period of experimentation. Correlated with them were cer- tain minor characters ; for instance, the weakest race. Race C, grew more slowly than either of the others, and took perceptibly longer to produce a definite clouding in a liquid medium. 492 SEDGWICK AND WINSLOW. •BACILLUS OF TYPHOID FEVEK. Race A. Series I. Race A. Series II. Number of Tube. Average Number Bacteria per c.c. Reduction per cent. Unfrozen Water. Thawed Ice. 1 8750 21 99.8 2 910 4 99.5 3 4910 1 99.9+ 4 1465 1 99.9 5 900 - 6 2475 4 99.8 7 1260 3 99.8 8 1360 10 99.2 9 1535 1 99.9+ 10 1030 7 99.3 11 35210 100.0 12 22575 3 99.9+ 13 63060 1 99.9+ 14 8575 — 15 94580 - 16 116235 1 99.9+ 17 140175 - 18 95725 4 99.9+ 19 4602 3 99.9 20 229950 2 99.9+ Average . . . 99.8 Tubes 1-10, frozen Marcli 2, 1898, in IJ hours ; thawed May 25, after IS weeks. » Tubes 11-20, frozen March 4, 1898, In 2 hours; thawed May 27, after 12 weeks. Number of Tube. Average Number Bacteria per c.c- Beduction percent. Unfrozen Water. Thawed Ice. 21 10655 17 99.8 22 7695 42 99.4 23 3170 2 99.9 24 4265 2 99.9 25 90825 1 99.9+ 26 79625 2 99.9+ 27 5920 6 99.9 28 275 1 99.6 29 6400 1 99.9+ 30 2085 3 99.9 31 11480 2 99.9+ 32 24637 12 99.9 33 214200 9 99.9+ 34 2760 . 7 99.7 35 10430 113 98.9 36 32110 4 99.9+ 37 12757 7 99.9 38 26547 4 99.9+ 39 15155 8 99.9 40 19890 1 99.9+ Average . . . 99.8 Tubes 21-30, frozen March 7, 1898, in 2i hours ; thawed May 2, after 8 weeks. Tubes 31-40, frozen March 12, 1898, in 2^ hours; thawed May 7, after 8 weeks. Kace a. Series III. Race A. Series IV. Number of Tube. Average Number Bacteria per c.c. Reduction per cent. Unfrozen Water. Thawed Ice. 41 3730 6 99.8 42 7880 5 99.9 43 2810 6 99.8 44 710 7 99.0 45 4470 4 99.9 46 9626 3 99.9+ 47 10482 2 99.9+ 48 3035 12 99.6 49 2085 11 99.5 50 5710 5 99.9 61 136710 5 99.9+ 62 41230 3 99.9+ 63 82215 1 99.9+ 64 26285 5 99.9 65 22225 1 99.9+ 66 19145 3 99.9+ 67 Control Control — 68 12.320 2 99.9+ 69 10850 4 99.9+ 70 10920 3 99.9+ Average . . . 99.8 Tubes 41-50, frozen March 16, 1898, in 2 hours ; thawed April 13, after 4 weeks. Tubes 61-70, frozen March 19, 1898, in 1 J hours ; thawed April 16, after 4 weeks. Number of Tube. Average Number Bacteria per c.c Reduction per cent. Unfrozen Water. Thawed Ice. 51 24640 25 99.9 52 49000 10 99.9+ 53 48930 30 99.9 54 40450 60 99.8 55 29340 30 99.9 56 282240 65 99.9+ 57 44380 110 99.7 58 132300 50 99.9+ 59 24185 25 99.9 60 93555 75 99.9 71 55650 _ 72 Control Control 73 52395 35 99.9 74 9230 70 99.2 75 86870 60 99.9 76 46025 25 99.9 77 1740 25 98.6 78 41825 5 99.9+ 79 33165 35 99.9 80 23250 30 99.9 Average . . . 99.8 Tubes 51-60, frozen March 18, 1898, in 1} hours ; thawed April 1, after ^ weeks. Tubes 71-80, frozen March 21, 1898, in 2^ hours ; thawed April 4, after S weeks. SEDGWICK AND WINSLOW. BACILLUS OF TYPHOID FEVER. 493 Race A. Series V. Number of Tube. Aveiiage Number Bacteria per c.o Beduction per cent. Unfrozen Water. Thawed Ice. 171 628425 48090 92.3 172 5355 2040 61.9 173 520380 8610 98.3 174 355950 122536 65.6 175 354690 36540 89.7 176 206010 19775 90.4 177 474390 4000 99.0 178 402020 191 3365 15 99.5 192 3300 40 98.8 193 103320 30 99.9+ 194 133875 275 99.8 193 348655 315315 9.6 196 40 70 0.0 197 214200 360 99.8 198 169155 64675 61.8 Ave rage . . . 77.8 Race A. Series VI. Tubes 171-178, frozen May 9, 1898, in 2 hours ; tliawed same day, after 6 hours. Tubes 191-198, frozen May 13, 1898, in ^ hours ; thawed same day, after 6 hours. Number of Tube. Averi^e Number Bacteria per c.c. Reduction per cent. Unfrozen Water. TJiawed Ice. 151 40776 21980 46.1 152 39235 17080 56.5 153 46465 14770 67.5 154 26530 15190 42.7 155 36295 14385 60.4 156 10710 5110 52.3 157 23520 3800 83.8 158 127260 40005 68.6 181 300 15 95.0 182 51030 9660 81.1 183 13265 1410 89.4 184 20475 2955 85.6 185 14596 1145 92.2 186 23416 805 96.6 187 22366 2915 87.0 188 2260 — — Average . , . 73.6 Tubes 151-158, frozen April 30, 1898, in 2\ hours; thawed, same day, after 2 hours. Tubes 181-188, frozen May 11, 1898, in 2\ hours; thawed same day, after '2 hours. Race A. Series VII. Race A. Series VIII. Number of Tube. Average Number Bacteria per c.c. Reduction per cent. Unfrozen Water, Tliawed Ice. 81 1820 990 46.9 82 2796 40 98.6 83 1265 25 98.1 84 820 100.0 85 355 15 95.8 86 430 15 96.5 87 2615 2075 17.5 88 1286 100.0 89 755 10 98.7 90 166 5 97.0 101 25970 11340 56.3 102 11665 8016 31.3 103 16955 4555 73.1 101 30730 26356 14.2 105 Control Control — 106 8760 6510 25.6 107 9206 5525 40.0 108 9345 3380 63.8 109 20090 11410 43.2 110 14315 9170 35.9 Ave rage . . . .... 63.7 Tubes 81-90, frozen March 25, 1898, in If hours ; thawed same day, after SO minutes. Tubes 101-110, frozen April 9, 1898, in 2 hours; thawed same day, after 30 minutes. Number of Tube. Average Number Bacteria per c.c. Reduction per cent. Unfrozen Water. Thawed Ice. 121 500220 716 99.9 122 492345 262640 48.7 123 57420 27755 51.7 124 53795 705 98.7 125 Control Control — 126 5705 966 83.2 127 124110 7175 94.2 128 77490 9800 87.3 161 33810 17640 47.9 162 276900 275940 .3 163 349020 120960 65.3 164 246645 111930 54.6 165 120775 62060 48.6 166 472500 236880 49.9 167 756660 605576 33.2 168 170100 123796 27.2 Average . . . .... 59.4 Tubes 121-128, frozen April 23, 1898, in IJ hours ; thawed same day, after 15 minutes. Tubes 161-168, frozen May 4, 1898, in IJ hours; thawed same day, after 15 minutes. 494 SEDGWICK AND WINSLOW. BACILLUS OF TYPHOID FEVEB. Race B. Series I. Race B. Series II. Number of Tube. Average Number Bacteria per c.c. Reduction per cent. Unfrozen Water. Thawed Ice. 71 37275 462 98.8 72 45990 27 99.9 73 41685 189 99.5 74 63210 382 99.4 75 26250 773 97.1 76 34230 599 98.3 77 18800 378 98.0 78 40110 467 98.8 79 42525 613 98.6 80 50295 47 99.9 81 144325 23 99.9+ 82 108360 11 99.9+ 83 123165 8 99.9+ 84 89775 7 99.9+ 85 83790 9 99.9+ 86 58275 10 99.9+ 87 104895 21 99.9+ 88 83475 11 99.9+ 89 187110 51 99.9+ 90 56595 15 99.9+ Ave ratje . . . 99.4 Tubes 71-80, frozen December 16, 1898, in 1^ hours ; thawed December 30, after S weeks. Minimal tempera- ture, (-14° C ). Tubes 81-90, frozen December 17, 1898 in 2 hours; thawed December 31, after $ weeks. Minimal tempera- ture, (-8° C). Race B. Sekies III. dumber ol Tube. Average Number Bacteria per c.c. Reduction per cent. Unfrozen "Water. Thawed Ice. 91 34965 180 99.5 92 25445 65 99.8 93 28560 60 99.8 94 29085 165 99.4 95 33810 365 98.9 96 32745 25 - 99.9 97 26880 5705 78.8 98 15855 15 99.9 99 22330 75 99.7 100 90300 30 99.9+ 151 2560 2 99.9 152 1595 4 99.7 153 1555 — — 154 — — — 155 225 1 99.6 156 1195 4 99.6 157 95 2 97.9 158 80 1 98.8 159 30 100.0 160 25 100.0 Ave rat/e . . . .... 98.4 Tubes 91-100, frozen December 20, 1898, in 2 liours; thawed December 23, after S days. Minimal tempera- ture, (-12° C). Tubes 151-160, frozen January 3, 1899, in 2 hours; thawed January 6, after 3 days. Minimal temperature, (-12° C). Number of Tube. Average Number Bacteria per c.c. Beduction per cent. Unfrozen Water. Thawed Ice. Ill 52605 3622 93.1 112 88200 1386 98.4 113 95235 4018 95.8 114 63065 1270 98.0 115 31080 1166 96.3 116 43470 1470 96.6 117 47040 896 98.1 118 37065 511 98.6 119 32890 441 98.7 120 54496 2935 94.6 121 10290 2373 76.9 122 54705 4106 92.6 123 69990 1466 97.9 124 21176 2993 85.9 125 46150 — 126 61005 4452 92.7 127 61950 3633 94.1 128 114030 17042 86.1 129 90090 8127 91.0 130 6650 805 87.9 Average . . . 93.3 Tubes 111-120, frozen December 23, 1898, in 2 liours ; thawed December 30, after 1 week. Minimal temperature, (-10° C). Tubes 121-130, frozen December 24, 1898, in IJ hours; thawed December 31, after 1 week. Minimal temperature, (-12° C). Race B. Series IV. Number of Tube. Average Number Bacteria per c.c. Reduction per cent. Unfrozen Water. Thawed Ice. 41 70560 38535 45.4 42 62290 31605 39.6 43 38640 28665 26.8 44 48405 10589 78.1 45 71505 14458 79.8 46 44100 10822 76.4 47 63945 21641 66.2 48 28246 13641 62.1 49 91036 19845 78.3 50 27300 11340 68.3 51 14140 5740 59.4 52 37800 25830 31.7 53 29925 15995 46.6 54 14280 5810 59.3 55 39710 16870 57.5 56 27826 9486 65.9 57 13686 6390 60.6 58 12565 5565 55.7 59 32760 33075 0.0 60 24570 9346 62.0 Ave rage . . . .... . 63.8 Tubes 41-50, frozen December 1, 1898, in 2 hours; thawed December 2, after Si hours. Minimal tempera- ture, (-7° C). Tubes 61-60, frozen December 8, 1898, in 2\ hours; thawed December 9, after S4 hours. Minimal tempera- ture, (-10° C), SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVEK. 495 Race B. Series V. Number of Tube. Average Number Bacteria per c.c. Seduction per cent. Unlrozen Water. Thawed Ice. 61 30135 13510 55.2 62 23625 8505 64.0 63 19635 10430 46.9 64 13055 12600 3.4 65 21840 10600 51.9 66 13685 6720 50.9 67 16800 10535 37.3 68 12075 8436 30.1 69 13230 11130 15.9 70 18025 12740 29.3 101 32865 18516 43.7 102 31710 37275 0.0 103 42625 5670 86.7 104 32865 36225 0.0 103 4585 65 98.5 106 22050 9380 57.6 107 5280 184690 0.0 108 5267 206010 0.0 109 15155 100.0 110 4585 107740 0.0 Ave rage . . . 38.6 Tubes 61-70, frozen December 9, 1898, in 2J hours ; thawed December 10, after 1^ hours. Minimal tempera- ture, (-6° C). Tubes 101-110, frozen December 21, 1898, in IJ hours ; thawed December 22, after 1^ hours. Minimal tempera- ture, (-8° C). Race B. Series VII. Numlwr of Tube. Average Number Bacteria per c.c. Reduction per cent. Unfrozen Water. Thawed Ice. 1 3515 2080 40.8 2 2180 3000 0.0 3 3635 2620 25.9 4 4465 3106 30.3 5 Control Control — 6 4300 4326 0.0 7 4975 3626 29.1 8 3405 3460 0.0 9 4305 6970 0.0 10 4615 3226 30.1 11 7960 6300 20.8 12 16380 14490 11.6 13 7560 6860 9.2 14 19460 21660 0.0 U 12215 10080 17.6 16 21700 15085 30.5 17 7665 8400 0.0 18 13300 11060 16.8 19 10920 11340 0.0 20 10360 14770 0.0 Ave . 13.8 Tubes 1-10, frozen November 19, 1898, in 2J hours ; thawed, same day, after i5 minutes. Tubes 11-20, frozen November 21, 1898, in IJ hours ; thawed, same day, after IS minutes. Race B. Series VI. Number of Tube. ATsrage number Bacteria per c. c. Reduction per cent. Unfrozen Water. Tliawed Ice. 21 10 200 0.0 22 75 170 0.0 23 190 70 63.2 24 2695 3360 0.0 25 100 260 0.0 26 210 375 0.0 27 1605 1505 6.2 28 180 350 0.0 29 1875 1826 2.7 30 3400 620 81.8 31 22906 12040 47.4 32 32655 8296 74.6 33 18660 6300 66.1 34 22226 6125 72.4 35 13766 4166 69.7 36 16675 3972 74.6 37 15750 7490 62.4 38 15470 3920 74.7 39 19216 5706 70.3 40 9590 2610 72.8 Average . . . 41.4 Tubes 21-30, frozen November 28, 1898, in IJ hours ; thawed same day, after S hours. Minimal temperature, (-8° C). Tubes 31-40, frozen November 29, 1898, in 1^ hours ; tliawed same day, after 3 hours. Minimal temperature, (-8° C). Race C. Series I. Number of Tube. Average Number Bacteria per c.c. Reduction per cent. Unfrozen Water. Thawed Ice. 1 6680 _ . 2 13475 5 99.9+ 3 4795 7 99.9 4 9310 4 99.9-f 5 10005 6 99.9-f 6 10886 2 99.9+ 7 6230 102 98.4 8 6216 100.0 9 10326 - — 10 11650 6 99.9+ 21 120645 12 99.9+ 22 142065 16 99.9+ 23 16695 100.0 24 — 25 1 — 26 13755 12 99.9 27 378945 1 99.9+ 28 101115 100.0 29 4370 2 99.9+ 30 128520 88 99.9 Average . . . 99.9 Tubes 1-10, frozen January 16, 1899, in If hours ; thawed January 30, after $ weeks. Minimal temperature, (-13° C). Tubes 21-30, frozen January 18, 1899, in IJ hours; thawed February 1, after Z weeks. Minimal temperature, (-10° C). 496 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVER, Race C. Series II. Number of Tube. Average Number Bacteria per c.c. Reduction per cent. Unfrozen Water Thawed Ice. 51 1920 •> 99.9 52 2675 2 99.9 53 2200 1 99.9+ 54 2510 3 99.9 55 2065 33 98.4 56 1605 10 99.4 57 1685 1 99.1 58 835 13 98.4 59 460 15 96.7 60 1820 — — 71 6680 100.0 72 7700 10 99.9 73 2485 1 99.9+ 74 6440 1 99.9+ 75 6145 3 99.9 76 4130 1 99.9+ 77 3920 1 99.9+ 78 3080 4 99.9 79 3535 100.0 80 640 100.0 Average . . . 99.5 Tubes 51-60, frozen January 2.3, 1899, in IJ hours ; tliavved January 30, after 1 week. Minimal temperature, (-12° C). Tubes 71-80, frozen January 25, 1899, in IJ hours; thawed February 1, after 1 week. Minimal temperature, (-14° C). Race C. Sekies III. Number of Tube. Average Number Bacteria per c.c. Reduction per cent. Unfrozen Water. Thawed Ice. 11 16765 9 99.9 12 17220 100.0 13 14315 2 99.9+ U 900 2 99.8 15 18270 3 99.9+ 16 9170 1 99.9+ 17 6930 100.0 18 7385 100.0 ■ 19 2925 100.0 20 9555 1 99.9+ 41 83476 6 99.9+ 42 83160 6 99.9+ 43 64890 2 99.9+ 44 66670 4 99.9+ 45 11200 1 99.9+ 46 21350 23 99.9 47 2030 3 99.9 48 700 1 99.9 49 186 2 98.9 50 1625 2 99.9 Average . . . 99.9 Tubes 11-20, frozen January 17, 1899, in 1 J hours ; thawed January 20, after 3 days. Minimal temperature, (-13° C). Tubes 41-50, frozen January 20, in 2 hours; thawed January 23, after S daijs. Minimal temperature, (—10° C. ) Race C. Series IV. Race C. Series V. Number of Tube. Average Number Bacteria per c.c. Reduction per cent. Unfrozen Water. Thawed Ice. 61 3335 6 99.9 62 3520 25 99.3 63 195 10 94.9 64 885 65 93.8 i 65 235 25 89.4 ; 66 215 60 72.1 67 2106 10 99.5 68 665 20 96.4 ' 69 40 20 50.0 70 500 15 97.0 81 1855 85 95.4 82 1830 20 98.9 83 260 65 78.8 84 935 35 96.3 85 110 95 13.6 86 3595 30 99.2 87 4480 36 99.2 88 315 70 77.8 89 60 40 20.0 90 40 — Average . . . 82.7 Tubes 61-70, frozen January 24, 1899, in \\ hours; thawed January 26, aiteTS4 hours. Minimal temperature, (-12° C). Tubes 81-90, frozen January 26, 1899, in IJ hours; thawed January 27, after Z4 hours. Minimal temperature, (-13° C). Number of Tube. Average Number Bacteria per c.c. Reduction per cent. Unfrozen Water, Thawed Ice, 91 10710 20 99,8 92 7280 76 99.0 93 ; 9555 90 99.1 94 4645 5 99.9 95 7735 35 99.5 96 1570 355 77.4 97 1325 20 98.5 98 — — 99 6440 690 90.8 100 13090 10 99.9 111 143640 5 99.9+ 112 234360 105 99.9+ 113 106626 10 99.9+ 114 41265 135 99.7 115 11655 5 99.9+ 116 36855 20 99.9 117 27195 40 99.9 118 119070 50 99.9+ 119 45360 5 99.9+ 120 16856 — — Ave rage . . . 98.0 Tubes 91-100, frozen January 27, 1899, in 2 hours; thawed January 28, after IB hours. Minimal temperature (-14° C). Tubes 111-120, frozen February 3, 1899, in 2 liours ; thawed February 4, after 16 hours. Minimal temperature' (-15° C). SEDGWICK AND WINSLOW. — BACILLUS OF TTPHOID FEVER. 497 Race C. Series VI. Number of Tube. Average Number Bacteria per c.c. Reduction per cent. Unfrozen Water. Thawed Ice. 31 350 10 97.1 32 270 100.0 33 185 100.0 34 90 5 94.5 33 5 100.0 36 — , 37 5 100.0 38 20 100.0 39 5 100.0 40 101 172080 1 99.9+ 102 61110 9 99.9+ 103 56700 1 99.9+ 104 40005 4 99.9+ 105 16660 100.0 106 146476 1 99.9+ 107 8865 1 99.9+ 108 9346 12 99.9 109 6930 2 99.9+ 110 6075 100.0 Ave rar/e . . . . ... . 99.6 Eace D. Series I. Tubes 31-40, frozen January 19, 1899, in 2 liours ; thawed same day, after 3 hours. Minimal temperature, (-8° C). Tubes 101-110, frozen February 2, 1899, in 2 hours ; thawed same day, after S hours. Number of Tube. Average Number Bacteria per c.c. Reduction per cent. Unfrozen Water. Thawed Ice. 1 5355 3080 42.5 2 5915 3265 44.8 3 6090 2465 59.5 4 5670 670 88.2 5 3010 1615 46.3 6 4410 780 82.3 7 3745 365 90.3 8 3290 1000 69.6 9 4375 480 89.0 10 11 12 13 6580 3640 44.7 2380 95 96.0 14 U 7210 66 99.1 16 1855 40 97.8 17 18 3675 90 97.5 19 20 "^ Average . . . 74.8 Tubes 1-10, frozen April 27, 1899, in 2 hours ; thawed same day, after S hours. Minimal temperature, (—10° C). Tubes 11-20, frozen April 28, 1890, in 2 hours ; thawed same day, after 3 hours. Minimal temperaliire, {—14° C). Race D. Series II. Race D. Series III. Number of Tube. Average Number Bacteria per c.c. Reduction per cent. Unfrozen Water. Thawed Ice 31 62605 318 99.4 32 63235 5072 90.5 33 77175 52 99.9 34 6565 927 83.3 35 184275 7339 96.0 36 6580 420 93.6 37 1890 — — 38 62055 6467 89.6 39 3265 87 97.3 40 6020 134 97.8 71 24360 2 99.9+ 72 29505 2 99.9+ 73 8925 22 99.8 74 2430 100.0 75 12810 4 99.9+ 76 24355 3 99.9+ 77 9450 210 97.8 78 2065 1 99.9+ 79 3160 1 99.9+ 80 2185 1 99.9+ 97.0 Tubes 31-40, frozen Mny 1, 1899, in 2 liours; thawed same day, after 6 hours. Minimal temperature, (—10° C). Tubes 71-80, frozen May 8, 1899, in 2 hours; thawed same day, after 6 liours. Minimal lempcrature, (—10° C.) Number of Tube. Average Number Bacteria per c.c. Reduction per cent. Unfrozen Water. Thawed Ice. 21 3990 15 99.6 22 3675 20 99.5 23 670 15 97.8 24 180 100 44.4 25 596 45 92.4 26 2276 15 99.3 27 180 20 88.9 28 140 25 83.6 29 25 25 0.0 30 240 100.0 41 515 33 93.6 42 1575 162 90.3 43 495 39 92.2 44 — — — 45 1856 88 96.3 46 47 2626 409 84.4 48 49 7176 611 92.9 50 1026 192 81.3 A.ve i'ane . 84.4 32 Tubes 21-30, frozen April 28, 1899, in 2 hours: thawed April 29, after 13 hours. Minimal temperature, (—11° C). Tubes 41-50, frozen May 1, 1899, in 2 hours; thawed May 2, after IS hours. Minimal tcmporature, (—10° C). 498 SEDGWICK AND WINSLOW. •BACILLUS OP TYPHOID FEVER. Race D. Series IV. Race B. Special Series. Number of Tube. Average Number Bacteria per c.c. Reduction per cent. Dnirozen Water. Thawed Ice. 51 33915 52 24570 106 99.6 53 60795 33 99.9 54 15960 35 99.8 55 24805 89 99.6 56 8820 103 98.8 57 6860 6 99.9 58 8960 55 99.4 59 2660 29 98.9 60 130410 199 99.8 61 21735 32 99.9 62 3200 71 97.8 63 5215 5 99.9 64 6160 63 99.0 65 955 10 99.0 66 2085 40 98.1 67 10885 65 99.4 68 250 3 98.8 69 790 34 95.7 70 1150 23 98.0 Ave raae 99.0 Number of Tube. Number Colonies per c.c. Un6-ozen Water. Number Colonies per CO. Thawed Ice. I 56070 83790 II 55440 53550 III 62370 61110 1 IV 19320 21210 7 • V 42210 29190 VI 28980 30030 vn 28770 18060 1 vni 23730 33390 1 IX 46410 42630 1 2 X — 7420 1 XI 13720 12180 XII 22050 28980 1 XIII 11830 7700 XIV 7980 7660 XV 7070 6020 1 XVI 6020 4690 xvn 6810 4690 XVIII 1840 1850 1 XTX 1260 1610 1 XX 3430 4200 Tubes 51-60, frozen May 2, 1899, in 2 hours ; thawed May .3, after S4 hours. Minimal temperature, (—13° C). Tubes 61-70, frozen May 3, in 2 hours ; thawed May 4, after S4 hours. Minimal temperature, (—12° C). Tubes I-X, frozen February 10, 1899, in 4 hours ; thawed March 10, after 4 weeks. Minimal temperature, (-10° C). Tubes ZI-XX, frozen February 15, in 1\ hours ; thawed March 15, after 4 weeks. Minimal temperature, (-10° C). * Colonies in ice of Tube IV proved to be typhoid. Colonies in ice in tubes not starred proved to be contaminations. Race C. Special Series. KtDnber of NninbeF Cdloniea per c.c. Number Colo- Number of Number Coloniea per c.c. Number Colo- Tube. Unfrozen Water. Tbawed Ice. Tube. tTnfrozen Water. Thawed Ice. I 4690 5390 XVI 9 n 6440 5390 1 xvn 21840 17640 ni 7140. 5320 1 xvin 9620 14070 IV 9870 12880 18 12 « XIX 15960 8680 V 5560 7210 3 4 * XX 7910 16170 VI 10080 11060 1 XXI 26410 30450 3 1 vn 4060 3710 1 xxn 34020 31920 _ VIII 4480 3990 2 xxm 3 IX 1 1 XXIV 20790 17640 _ X 1 2 1 — XXV 107730 103950 272 262 * XI 147420 148050 _ XXVI 380 560 1 XII 2 XXVII xni 63630 71190 xxvin 330 210 1 * XIV 57960 48510 XXIX 150 280 XV 87570 86940 1 XXX 1330 1440 Tubes I-X, frozen February 17, 1899, in 3 hours ; XI-XZ. February 23, in 2 hours ; XXI-XXX, March 3, in If hours; thawed after 1 week in each case. Minimal temperature, —10^ C. * Colonies in ice proved to be typhoid. Colonies in ice in tubes not starred proved to be contaminations. SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVEB. 499 B. EXPEEIMENTS ON THE EFFECT OF ALTERNATE FREEZING AND THAWING UPON THE BACILLI OF TYPHOID FEVEE. Dr. Prudden, as we Jiave seen, considered intermittent more fatal than uninter- rupted freezing, and, indeed, succeeded in one case in entirely sterilizing a tube inoculated with B. typhi by this method. Our four series of experiments on this subject were conducted by freezing tubes in the freezer as described in the previous section. The tubes of Series I, Race A, were frozen daily for five days and allowed to thaw each time after about eighteen hours, samples being planted after each thawing. Those of Series I, Race B, were frozen three times, on alternate days, remaining frozen for twenty-four hours each time and kept below 2° for the rest of the time. The two series in Race D were treated like the tubes frozen for three hours and six hours in the last section, except that instead of remaining frozen they were thawed and refrozen once and twice respectively during that time. The results of these experiments with the results of simple freezing directly com- parable are as follows : — Race A. Reduction. Frozen once in one day 96.1 Frozen twice in two days 98.9 Frozen three times in three days 99.5 Frozen four times in four days 99.8 Race B. Kept frozen for three days (see previous section, Race B, Series III) 98.4 Frozen twice in four days 99.G Kept frozen for seven days (see previous section, Race B, Series II) 93.3 Frozen three times in six days 99.8 Race D. Kept frozen for three hours (see previous section, Race D, Series I) 74.8 Refrozen once in three hours 97.4 Kept frozen for six hours (see previous section, Race D, Series II) . 97.0 Refrozen twice in six hours 99.5 Conclusion. Thawing and ref reezing are somewhat more fatal than simple freez- ino- in its effect on the typhoid bacillus. Four successive freezings and thawings do not, however, suffice to kill off the most resistant bacilli. 500 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVER. Race A. Sebees I. Number of Tube. Average before Freezing. After One Freezing- After Two Freezings. Alter Three Sneezings. After Four TreezingB. Average. Reduction per cent. Average. Rednction per cent. Average. Reduction percent. Average. Reduction per cent. 91 49910 _.. 92 22786 175 99.3 6 99.9 1 99.9 3 99.9 94 348390 26320 92.5 2495 99.3 644 99.9 147 99.9 95 308385 1735 99.6 171 99.9 59 99.9 2 99.9 96 167680 635 99.7 6 99.9 2 99.9 2 99.9 97 277515 40 99.9 1 99.9 3 99.9 2 99.9 98 60820 745 98.6 290 99.4 200 99.6 120 99.8 99 600 180 70.0 4 99.3 1 99.8 — — 100 34090 190 99.4 55 99.9 • 16 99.9 8 99.9 111 76895 380 99.5 80 99.9 22 99.9 20 99.9 112 23875 685 97.1 356 98.5 33 99.9 13 99.9 113 29750 — — — — 114 38290 265 99.3 40 99.9 3 99.9 100.0 115 31500 1785 94.3 1232 96.1 416 98.7 427 98.6 116 46585 75 99.8 3 99.9 23 99.9 3 99.9 117 21 17 15 2 — 2 — 118 48335 4450 90.8 3895 91.9 2440 95.0 — 119 38430 155 99.6 11 99.9 2 99.9 2 99.9 120 25200 215 99.1 14 99.9 2 99.9 3 99.9 Averages 96.1 98.9 99.5 99.8 Tnbes 91-100, frozen March 28, 1898; thawed and sampled and refrozen, on each of the four days gncceeding. Tabes remained frozen 18 hours each time. Tubes 111-120, treated in same manner week of April 11, 1898. Race B. Series I. Average before Freezing. After One Freezing. After Two Freezings. After 'Threi 3 Freezings. of Tube. Average. Reduction per cent. Average. Reduction per cent. Average. Rednction per cent. 1 46950 420 99.1 67 99.9 38 99.9 2 22260 155 99.3 24 99.9 23 99.9 3 12810 315 97.5 21 99.8 19 99.9 4 12145 215 98.2 62 99.6 29 99.8 5 10640 20 99.8 10 99.9 7 99.9 6 8715 120 98.6 10 99.9 5 99.9 7 7945 80 99.0 13 99.8 9 99.9 8 4190 65 98.4 13 99.7 7 99.8 9 2620 85 96.6 9 99.6 4 99.8 10 2450 95 96.1 7 99.7 1 99.9+ 11 115290 775 99.3 280 99.8 161 99.9 12 142695 1980 98.6 1008 99.3 366 99.8 IB 183385 315 99.8 96 99.9 85 99.9+ 14 74970 1140 98.5 595 99.2 354 99.5 15 138915 480 99.7 276 99.8 116 99.9 16 227745 11866 94.8 5733 97.6 458 99.8 17 104265 670 99.4 198 99.8 129 99.9 18 107730 1250 98.8 403 99.6 269 99.8 19 163485 660 99.6 139 99.9 65 99.9+ 20 120015 390 99.7 171 99.9 75 99.9 Averages 98.5 99.6 99.8 Tubes 1-10, frozen April 10, 189!) ; kept frozen for 24 hours, and below 2° for 24 hours more ; refrozen April 12 and April 14. Samples piiintcd before each freezing and April 15. Tubes 11-20, treated in same way, April 17, and following days. SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVER. 501 Race D. Series I. Number of Tube. Average Number Bacteria per c.c. Beduction per cent. Unfrozen "Water. Thawed Ice. 101 78435 147 99.8 102 76230 451 99.4 103 1765 23 98.7 104 2730 47 98.3 105 275 15 94.5 106 7735 26 99.7 107 1120 49 95.6 108 11690 134 98.9 109 6895 235 96.6 110 131 132 133 3500 175 95.0 134 — 135 29190 • 77 99.7 136 20160 371 98.2 137 6055 192 96.8 138 5710 388 93.2 139 — 140 3885 144 96.3 Ave rage 97.4 Race D. Series II. Number of Tube. Average Number Bacteria per c.c. Reduction per cent. Unfrozen Water. Thawed Ice. Ill 34020 71 99.8 112 59090 152 99.7 113 13230 7 99.9 114 32525 274 99.2 115 1545 9 99.4 116 4270 29 99.3 117 2095 1 99.9+ 118 6685 5 99.9 119 6250 68 98.7 120 13685 210 98.5 121 — — 122 — — — 123 203175 3 99.9+ 124 40005 1 99.9+ 125 56070 2 99.9+ 126 170100 297 99.8 127 — — — 128 — — — 129 — — — 130 925 1 99.9 Average . . . 99.5 Tubes 101-110, frozen May 5,1899; thawed and re- frozen in 3 hours. Minimal temperature, (—13° C). Tubes 131-140, frozen May 13, 1899 ; thawed and re- frozen in 3 hours. Minimal temperature, (—19° C). Tubes 111-120, frozen May 9,1899; thawed and refrozen twice in next 6 hours. Minimal temperature, (—14° C). Tubes 121-130, frozen May 10,1899; thawed and refrozen twice in next 6 hours. Minimal temperature, (—10° C). C. EXPERIMENTS ON THE EFFECT OF TEMPERATURES SLIGHTLY ABOVE THE FREEZING-POINT UPON TYPHOID BACILLI IN WATER. In these experiments sterilized test tubes were inoculated with pure cultures as in all the preceding work. Afterward they were treated in one of three ways, — either placed in an incubator at the room temperature, 20° C, or in an ice-chest ranging from 8°-12°, or cooled in the freezer to a point just above freezing. This last was effected by filling the outer chamber with ice without salt. In the three sets of tubes treated by the last method at 1°, the duration of expo- sure and the reduction were as follows : Race A in two hours was reduced 47.8 per cent ; Race B in one and one-half hours was reduced 32.9 per cent ; Race C in three hours was reduced 80.1 per cent. The reductions for the same races actually frozen for the nearest corresponding periods, were 73.6 per cent, 41.4 per cent, and 99.5 per cent, respectively. Each race maintains its relative position of resistance. The reduc- tion in the chilled water is very nearly as great as in the ice ; and the difference is only what the temperature difference might be expected to produce. Evidently there 502 SEDGWICK AND WINSLOW. BACILLUS OF TYPHOID FEVER. is nothing mysterious about the act of freezing, no mechanical crushing of bacteria ; the process of destruction is continuous above and below the freezing-point, depending upon the two main factors of time and temperature. Series II and III of Races B and C cover longer periods of time and higher tem- peratures. Half of the tubes in each series were kept at 10° and half at 20°, but no marked differences appeared as the result of these two modes of treatment, and the two sets are averaged together in each series. The tubes wei'e kept in these experiments for two weeks, one-half of them being sampled on the second and the seventh day, the others on the third day and the fourteenth. The tubes were, of course, protected from the action of light. Race B. Race C. Reduction per cent. Reduction per cent. Time. Series 11. Series III. Series IL Series m. 1 day 92.2 78.7 88.4 71.0 3 days 99.3 86.7 90.4 83.8 7 days 99.9 99.3 94.1 89.6 14 days 99.9 98.8 99.9 • 99.9 It will be noted that with each race the second series shows a greater reduction than the third. The explanation for this lies in the fact that these experiments were carried on some time after the regular experiments on freezing their respective races. During the intervening period the germs had been grown on agar, and the first new series of experiments with each race showed an extraordinary reduction, over 99 per cent in a day, etc. The results of this series have not been tabulated. The second series of each race, Series II above, showed more moderate, but still high reductions ; while by the time the third series was inoculated, a week later, the cultures, by culti- vation in bouillon, had regained their normal condition. The tubes inoculated with Race D were kept for twenty-four hours only, samples being planted after 3, 6, 12, and 24 hours. Series I was kept at 20° and Series II at 10°. After 3 hours. 6 hours. 12 hours. 24 hours. Series I (20°) . . . . . 70.8 72.6 85.7 88.4 Series II (10°) . . . . . 63.1 74.0 87.4 95.5 Conclusions. From these experiments it appears that typhoid fever bacilli behave in water much as they do in ice. A large proportion of them are killed by a few minutes' exposure to the unfavorable conditions ; during the next few hours the reduction proceeds pari passu with the duration of the experiment; while a few germs persist for some time. SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVEK. 503 The results differ from those obtained by actual freezing in two respects. We have seen that freezing for short periods produced varying and uncertain results, while ice over twenty-four hours old showed a constant reduction of over 90 per cent. The tubes of water which were not frozen remained subject to this uncertainty for a much longer period. Inspection of the tables will show that individual tubes contained some- times half of their original germ content after a week, or four-fifths of it after three days. On the other hand, complete sterilization ensued more often than in the frozen tubes. A second characteristic of the viability of the germs in water is the fact, closely allied to the fii-st, that an increase seems sometimes to occur. The successive sam- plings of the same tube show in certain instances a slight multiplication. The reduction in water at 10° does not seem to be any greater than at 20°. 504 SEDGWICK AND WINSLOW. BACILLUS OF TYPHOID FEVER. Race B. Series I. Race A. Series I. Number of Tube. Average Number Bacteria per c.c. Reduction per cent. Uufrozen Water. Thawed Ice. 131 140490 51450 63.4 132 505 10 98.0 133 10640 11410 0.0 134 12390 6965 43.8 lU 31465 165 99.5 136 273105 87885 67.8 137 17745 9870 44.4 138 112770 63000 44.1 141 254205 105840 58.4 142 157815 92610 41.3 143 72135 42490 41.1 143 302715 302715 0.0 146 141750 45990 67.6 147 17360 21735 0.0 Ave rage . . . 47.8 Tubes 131-138, cooled down to V C. in 1 J hours, April 25, 1898. Kept at that temperature for ^ hour more. Tubes 141-147, cooled down to 1° C. in 1| hours, April 29, 1898. Kept at that temperature for ^ hour more. Number of Tube. Average Number Bacteria per c.c. Reduction per cent. Unfrozen Water. Thawed Ice. 131 1946 1690 13.2 132 1610 1001 37.8 133 1848 1165 37.0 134 1571 1183 24.7 135 1379 1155 16.1 136 1291 1506 0.0 137 1232 1438 0.0 138 874 962 0.0 139 892 1473 0.0 140 1022 1051 0.0 141 4096 6 99.9 142 260 60 80.8 143 205 90 66.1 144 270 55 79.6 145 40 5 87.6 146 215 146 32.6 147 290 275 6.2 148 225 140 37.8 149 216 120 44.2 150 80 76 6.3 Average . . . 32.9 Tubes 131-140, cooled down to 0°, without freezing, and kept at that temperature for 1 1 hours. Date, December 29, 1898. Tubes 141-150, cooled down to 0°, without freezing, and kept at that temperature for 1^ hours. Date, January 2, 1899. Race C Series I. Number of Tube. Average Number Bacteria per c.c. Reduction per cent. Unfrozen Water. Thawed Ice. 121 66125 14910 73.0 122 244755 217350 11.2 123 66465 11045 83.4 124 62686 69010 6.9 125 211050 32760 84.5 126 269955 75915 71.9 127 103005 7385 92.8 128 106840 8086 92.4 129 67725 20685 69.6 130 23100 4725 79.5 131 139860 9660 93.1 132 76645 3675 96.2 133 68275 6580 88.7 134 219135 23205 89.4 135 82530 3675 95.6 136 84106 3160 96.3 137 1290 100,0 138 38850 2105 94.6 139 30030 860 97.1 140 5630 640 88.4 Average . . 80.1 Tubes 121-130, cooled down to 0° in J hour, February 6, 1899 j kept at that temperature (not frozen) for three hours. Tubes 131-140, cooled down to 0° in J hour, February 7,1899; kept at that temperature (not frozen) fur 3 hours. SEDGWICK AND WINSLOW. — BACILLUS OP TYPHOID FEVER. 505 Race B. Series II. Number Average after After One Day. After Three Days. After Seven Days. After Fourteen Days. of Xube. Inoculation. Average. Seduction per cent. Average. Reduction per cent. A .......... 1 Reduction Average. | j^^^^t. Average. Reduction per cent. 201 232155 20 99.9+ 2 99.9+ 202 93870 6755 92.8 100.0 203 104895 4615 95.7 100.0 204 9345 1180 87.4 • 100.0 205 72135 60 99.9 1 99.9+ 206 61660 100.0 100.0 207 56070 1837 96.3 3 99.9+ 208 1615 100.0 100.0 209 216406 2972 98.8 21 99.9+ 210 Control Con trol Con trol 211 11026 100.0 1 99.9+ 212 17886 6076 71.6 61 99.7 213 20790 90 99.6 100.0 214 10420 30 99.7 100.0 215 81270 2020 76.2 73 99.9 216 10640 1 99.9+ 1 99.9+ 217 826 100.0 8 99.0 218 170 1 99.4 100.0 219 22330 5 99.9+ 100.0 220 74340 623 99.2 3 99.9+ Averages 92.2 99.3 99.9 99.9 Tubea 201-210, inoculated March 17, 1899 ; kept in ice-chest at about 10° C. Tubes 210-220, inoculated March 17, 1899 ; kept in room at about 20° C. Race B. Series in. Number of Tube. Average after Inoculation. After One Day. After Three Days. After Seven Days. After Fourteen Daye. Average. Reduction per cent. Average. Reduction per cent. Average. Reduction per cent. Average. Reduction per cent. 221 113400 21630 80.8 2467 97.8 222 74340 386 99.6 3 99.9+ 223 74340 13406 82.0 1043 98.6 224 48610 7840 83.8 1043 97.8 225 137026 2426 98.2 100.0 226 17535 1389 92.1 100.0 227 103635 40446 61.0 6835 93.4 228 74656 27406 63.3 1603 97.9 229 26200 2520 90.0 28 99.9 230 86050 7465 91.2 112 99.9 231 34660 5366 30 99.9 232 18795 6930 5 99.9+ 233 7320 1206 39 99.5 234 6146 3420 9 99.8 235 6566 1196 16 99.7 236 507S 84.6 22 , 99.6 1 99.9+ 237 10636 63.1 .312 97.0 2 99.9+ 238 4340 83.5 28 99.3 3 99.9 239 17956 33.5 4284 76.1 486 97.3 240 6090 78.5 173 97.2 3 99.9+ Averages 78.7 86.7 99.3 98.8 Tubes 221-230, inoculated March 24, 1899 ; kept in ice-chest at about 10° C. Tubes 231-240, inoculated March 24, 1899; kept in room at about 20° C. 506 SEDGWICK AND WINSLOW. — BACILLUS OF Ti'PHOID FEVEE. Race C. Series II. Averi^e after Inoculation. After One Day. Alter Tliree Days. After Seven Days. After Fourteen Days. of Tube. Average. Reduction per cent. Average. Reduction per cent. Average. Reduction, per cent. Average. Reduction per cent. 141 68985 1596 97.7 30 99.9+ 142 81270 10206 87.4 1043 98.7 143 143640 1141 99.2 56 99.9+ 144 198450 107730 45.7 125956 36.5 145 132300 1883 98.6 38 99.9+ 146 198450 13041 93.4 196 99.9 147 210735 2142 99.0 30 99.9+ 148 80325 238 99.9+ 2 99.9+ 149 82215 479 99.9 2 99.9+ 150 79065 228 99.9+ 100.0 151 51345 1176 97.7 2 99.9+ 152 66780 14238 78.7 501 99.2 153 349650 11970 96.9 128 99.9+ 154 73395 7 99.9+ 100.0 155 230680 40761 82.3 8347 96.4 156 168210 135229 19.6 54 99.9+ 157 41265 1400 96.6 22 99.9 158 17395 16 99.9 1 99.9+ 159 83790 3402 95.9 5 99.9+ 160 120015 42 99.9+ 3 99.9+ Averages 88.4 90.4 94.1 99.9 Tubes 141-150, inoculated March 20, 1899; kept in ice-chest at ]0°. Tubes 151-160, inoculated March 20, 1899; kept in room at about 20° C. Race C. Series III. Number of Tube. Average af t(tr Inoculation. After One Day. After Three Days. After Seven Days. After Fourteen Days. Average. Reduction per cent. Average. Reduction per cent. Average. Reduction per cent. Average. Reduction per cent. 161 10535 15 99.9 1 99.9+ 162 26985 5 99.9+ 1 99.9+ 163 44205 30660 30.6 242 99.5 164 5705 30 99.4 100.0 165 340 230 32.4 166 41685 7497 82.0 119 99.7 167 1080 45 95.8 1 99.9 168 24465 2709 88.9. 14 99.9 169 15330 15876 0.0 170 Control Control Con trol 171 420 40 90.5 100.0 172 1205 805 33.2 1 99.9 173 305 100.0 1 99.7 174 19740 9275 53.0 609 96.9 175 1065 310 70.9 2 99.8 176 3255 46. 98.6 1 99.9+ 177 4105 271 93.4 2 99.9+ 178 1725 1 99.9 100.0 179 620 100.0 2 99.7 180 17850 1596 91.1 48 99.7 Averages 71.0 83.3 89.6 99.9 Tubes 161-170, inoculated March 27, 1899 ; kept in iee-chest at 10°. Tubes 171-180, inoculated March 27, 1899 ; kept in room at about 20° C. SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVEE. 507 Race D. Series I. Number Alter Three HouiB. After Six Hours. Alter Twelve Hours. After Twenty-four Hours. of Tube. Inoculation. Average. Reduction per cent. Average. Eeduction per cent. Average. Reduction per cent. Average. Reduction per cent. 141 2590 30 98.8 100.0 5 99.8 5 99.8 142 5670 25 99.6 20 99.6 100.0 6 99.9 143 2316 5 99.8 226 90.3 100.0 10 99.6 144 3185 140 95.6 90 97.2 20 99.4 100.0 143 15 10 33.3 15 0.0 10 33.3 -— - 14« 340 70 79.4 60 82.4 10 97.1 25 92.6 147 745 30 96.0 210 71.8 5 99.3 10 98.7 148 50 35 30.0 100.0 100.0 20 60.0 149 45 30 33.3 60 0.0 100.0 6 88.9 150 230 40 82.6 5 97.8 6 97.8 5 97.8 151 52920 22890 66.7 18690 64.7 7969 84.9 770 98.5 152 8680 6300 27.4 3780 56.6 1704 80.1 483 94.4 153 68670 28360 58.7 26885 60.8 21168 69.2 20097 70.7 154 43155 716 98.3 386 99.1 262 99.4 119 99.7 155 6650 1025 84.6 380 94.3 273 95.9 158 97.6 156 41895 7000 83.3 7106 83.0 5323 87.3 4410 89.5 157 74025 12366 83.3 14700 80.1 11056 85.1 6647 91.0 158 53235 7980 85.0 13125 75.3 14647 72.5 5386 89.9 159 23206 6090 73.8 9695 58.2 8064 65.2 10206 56.0 160 3256 2746 16.0 1925 40.9 1669 48.7 86.7 1480 54.5 Averages 70.8 72.6 88.4 Tubes 141-160, inoculated May 11, 1899. Tubes 151-160, inoculated May 15, 1899. Kept at room temperature. Kept at room temperature. Race D. Series II. Average after Inoculation. After Three Hours. After Six Hours. After Twelve Hours. After Twenty-four Hours. ol Tube. Average. Reduction per cent. Average. Reduction per cent. Average. Reductiou per cent. Average. Reduction per cent. 161 27616 7465 73.9 5810 79.0 6268 77.3 162 291060 47146 83.8 21000 92.8 3896 98.7 163 200340 11235 43.9 205 99.9 — . 1 99.9+ 164 208530 24045 88.5 10500 95.0 — 6261 97.5 165 178605 1860 99.0 355 99.8 21 99.9+ 166 64810 7000 87.2 — — 167 106786 11656 89.1 2275 97.9 1643 86.4 168 212626 4130 98.1 1386 99.4 2001 99.1 169 26096 1960 92.2 745 97.1 — — 529 97.9 170 2660 145 94.5 36 98.7 — 129 95.2 171 236260 95446 69.6 78435 66.8 25578 89.2 3643 98.5 172 240976 120330 60.1 144685 40.0 46360 81.3 20128 91.8 173 199396 98595 60.6 41840 79.0 16065 91.9 196 99.9 174 299566 160266 49.8 156200 47.8 29678 90.1 6691 98.1 175 179660 104896 41.6 54180 69.8 9670 94.4 2992 98.3 176 107730 64260 40.4 72265 32.9 17010 84.2 2396 97.8 177 133235 : 69930 47.6 81900 38.5 17671 86.7 7140 94.6 178 141760 89615 36.8 35700 74.8 20790 85.3 1101 99.2 179 211680 226800 0.0 158760 25.0 56891 73.1 31468 85.1 180 47565 30765 36.3 8610 71.9 1141 97.6 288 99.4 Averages 63.1 74.0 87.4 95.5 Tubes 161-170, inoculated May 17, 1899. Tubes 171-180, inoculated May 19, 1899. Kept in ice-cliest at 10°. Kept in ice-chest at 10°. 508 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FBVEK. D. EXPERIMENTS ON THE VIABILITY OP TYPHOID FEVER BACILLI IN EARTH AT VARIOUS TEMPERATURES. These experiments were carried on in order to compare the conditions affecting a reduction of the number of typhoid bacilli in soil, with those operating on them in water and ice. The general method pursued was the same, the inoculation of numerous small portions of a sterile medium with a pure culture of the micro- organism. In each series of experiments about one hundred grains of sifted clayey soil were sterilized by baking for sixteen hours, on two successive days. The whole of the earth was then inoculated by mixing with it a bouillon culture two or three days old, of B. typhi. Race B; and an even distribution was accomplished by stirring and kneading with a spatula. The earth, having been dried by the previous baking, absorbed the bouillon culture without becoming visibly damp. Fifty portions of the inoculated earth of one gram each were then weighed out and placed in fifty sterile empty test-tubes. Of these fifty portions, ten were at once mixed with sterile water and two check plates made from each flask. The remaining forty tubes were carried to the cold storage warehouse or kept at the room temperature, as the case might be, in either condition being protected from the action of light. After one day, three days, one week, and two weeks, ten tubes were removed and planted. In every case the entire gram of earth mas mixed with ten, one hundred or nine hundred cubic centimeters of sterile water ; and two check plates were made from the dilution. The inoculation, weighing and tubing of the earth, were conducted in a glass chamber some three feet square, with a sliding door raised only sufficiently to admit the arms of the operator. Control plates were made from four portions of the earth before inoculation, the portions of a gram apiece being tubed and planted exactly like the regular tubes. The following were the results per gram : — Colonies pee Geam. 1 a 3 4 14 1 3 360 The tubes of the first three series were kept at the cold storage warehouse during the period of the experiment, at 0° C. Those of the fourth series were kept at the room temperature. The summarized average results of these four series are as follows : — SEDGWICK AND WINSLOW. — BACILLUS OF TTPHOID FEVJ:R. 509 Typhoid Bacilli in Earth. Average Number per Gram. After Inoculation. After 1 day. 3 days. 7 days. 14 days. Series 10° 180776 4635 705 25 9 II 0° 4846856 95017 1395 525 588 III 0° 7778595 324588 4656 1304 1160 IV 20° 4673683 2565 450 95 92 Two more series of experiments with earth were carried out to throw light on the part played by dryness in the reduction manifest in the first experiments. In these latter researches the sets of fifty tubes were inoculated just as before, and ten of them were planted at once. The remaining forty were divided into two portions. The gram of earth in each of twenty of the tubes was moistened by the addition of about one-third cubic centimeter of sterilized tap water ; while the earth in the other twenty tubes was left in its comparatively dry condition. The tubes were all kept at the room temperature. Thus a comparison may be drawn as to the viability of the germ in damp and in dry earth. The results were as follows : — Ttphoid Bacilli in Dry and Damp Earth. Average Number per Gram. After Inoculation. Series V 939115 i^^^ VI 1198260 I-^''y I Damp jDiy I. Damp After 1 day. 2070 3 days. 50 7 days. 2 14 days. 47 225 7010 1295 8 566 71 12 4 1699110 — — 29587 Conclusions. 1. The typhoid bacilli in dry earth behave just as in water and in ice. They die out, rapidly at first, and their numbers are progressively reduced as the treatment is prolonged. A fraction of one per cent persists for some time. 2. Cold alone does not materially affect the reduction of typhoid germs in dry earth. 3. In moist earth, although the main phenomena are the same, the destruction of the bacteria is much less rapid. With the liberal food supply introduced with the bouillon in these experiments, they appear sometimes to hold their own entirely. 510 SEDGWICK AND WlXbLO\V. BACILLUS OF TYPHOID FEVER. TYPHOID BACILLI IN EARTH. Series I. Febrnaiy 13, 1S99. February 14, 1899. Number of Tube. Number of Tube. Bacteria per gram. Bacteria per gram. 1 233000 243000 11 5400 1800 2 217000 — 12 4500 4500 3 114000 112000 13 7200 1800 4 123000 120000 U 6300 4500 5 504000 207000 15 2700 3600 6 107000 126000 16 900 5400 7 178000 141000 17 1800 13500 8 157000 153000 18 8100 9900 19 3600 1800 20 2700 2700 Average 180776 Average 4635 Febraary 16, 1899. February 20, 1899. February 27, 1899. Number of Tube. Number of Tube. Number of Tube. Bacteria per grant. Bacteria per gram. Bacteria per gram. 21 900 600 31 30 10 41 20 22 1000 900 32 10 42 30 40 23 900 200 33 90 50 43 10 24 600 400 34 10 44 25 900 1600 35 45 20 26 900 400 36 30 46 20 27 400 700 37 80 47 28 300 400 38 30 70 48 30 29 900 800 39 10 40 49 30 400 700 50 Average 705 Average 25 Avei-age 9 Tubes kept at 0° 0. SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVEE. 511 Series II. Number of Tube. February 20, 1899. Number of Tube. February 21, 1899. Bacteria per gram. Bacteria per gram. 1 4932900 5896800 ■ 11 130500 176500 2 5046300 4649400 12 103500 3 5216400 4876200 13 119700 81900 4 2286900 4706100 14 65800 70200 5 5963600 7030800 15 64000 51300 6 2570400 3628800 16 102600 7 5159700 4309200 17 41400 37800 8 6860700 5443200 18 117900 140400 9 3686500 4989600 19 114300 116200 20 126900 74700 Average 4846856 Average 95017 February 23, 1899. February 27, 1899. Marcb 6, 1899. Number of Tube. Number of Tube. Number of Tube. Bacteria per gram. Bacteria per gram. Bacteria per gram. 21 1200 1300 31 30 110 41 340 400 22 1400 1900 32 210 310 42 160 110 23 700 1700 33 60 240 43 1770 750 24 600 1200 34 660 470 44 1070 1690 25 1100 800 35 110 90 45 160 200 26 300 700 36 240 290 46 150 900 27 2200 4000 37 310 500 47 120 800 28 2600 3100 38 820 210 48 360 430 29 1100 700 39 190 240 30 400 1000 40 2890 2620 Average 1396 Average 526 Average 588 KeptatO°C. 512 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVEK. Semes III. February 24, 1899. February 25, 1899. Number of Tube. Number of Tube. Bacteria per gram. Bacteria per gram. 1 8541900 7144200 ■ 11 466200 522900 2 6747300 7200900 12 529200 229500 3 7314300 6066900 13 409500 621100 4 10432800 10092600 14 415800 371700 5 7711200 6917400 15 270900 258300 6 6860700 7597800 16 289800 216900 7 7881300 8278200 17 573300 346500 8 8731800 6577200 18 144900 171100 9 6463800 7711200 19 5400 3600 10 9695700 7994700 Average 7778595 Average 324588 Pebraary 27, 1899. March 3, 1899. March 10, 1899. Number olTnbe. Number of Tube. Number of Tube. Bacteria per gram. Bacteria per gram. Bacteria per gram. 21 1600 1400 31 220 370 41 1710 420 22 2400 1500 32 380 250 42 1390 350 23 6900 7300 33 3710 370 43 1770 — 24 1500 900 34 310 — 44 — — 25 1600 2600 35 3290 240 45 2870 2150 26 4000 3600 36 5530 2310 46 660 420 27 1900 1600 37 930 270 47 160 210 28 10500 25200 38 690 480 48 1510 40 39 1020 4410 49 1360 1450 40 190 140 50 2420 1220 Average 4656 Average 1304 Average 1160 Kept at 0° C. SEDGWICK AND WINSLOW.. — BACILLUS OF TYPHOID KEVER. 513 Series IV. Kumber of Tube. February 28, 1899. Number of Tube. March 1, 1899. Bacteria per gram. Bacteria per gram. 1 2 3 4 5 6 7 4139100 3742200 3798900 3685500 5896800 5216400 4025700 6010200 4876200 2721600 7144200 5556600 3628800 11 12 13 14 15 16 17 18 19 20 10800 2700 2700 900 1800 900 1800 900 5400 1800 4500 900 1800 900 2700 900 900 7200 1800 Average 4673683 Average 2565 March 3, 1899. March 7 1899. March 14, 1899. Number of Tube. Number of Tube. Number of Tube. Bacteria per gram. Bacteria per gram. Bacteria per gram. 21 900 1800 31 20 10 41 10 22 900 32 20 30 42 140 380 23 900 33 60 20 43 20 170 24 1800 34 30 10 44 130 110 25 900 35 90 10 45 350 30 26 36 690 60 46 160 60 27 37 340 100 47 40 28 900 38 120 20 48 40 20 29 39 250 49 190 30 900 40 10 10 Average 450 Average 95 Average 92 Kept at 20° 0. 33 514 SEDGWICK AND WINSLOW. BACILLUS OF TYPHOID FEVER. Series V. AFTER INOCULATION. March 15, 1899. Number ot Tube. Bacteria per gram. 1 1045800 863100 2 1140300 989100 3 1663200 1499400 4 673300 938700 5 592200 686700 6 1297800 863100 7 1348200 — 8 1004400 919800 9 1026900 636300 10 888300 875700 Average 939115 DAMP EARTH. DRY EARTH. of Tube. Bacteria per gram. Averages. of Tube. Bacteria per gram. Averages. 11 16 900 900 12 900 17 9000 1800 March 16 13 900 March 16 18 2700 14 19 900 3600 225 20 900 2070 21 26 200 22 100 27 100 March 18 23 21100 25100 March 18 28 100 100 24 8400 15000 29 25 400 7010 30 50 31 220 180 36 32 10 37 - March 22 33 3360 6580 March 22 38 34 10 39 10 35 — — 1295 40 10 2 41 40 40 46 10 30 42 47 260 100 March 29 43 March 29 48 20 44 49 20 45 8 50 10 20 47 SEDGWICK AND WIKSLOW. -- BACILLUS OF TYPHOID FEVEE. 515 Series VI. AFTER INOCULATION. March 29, 1899. Number ol Tube. Bacteria per gram. 1 1455300 1379700 2 1682100 1455300 3 1152900 1228500 4 1083600 825300 5 926100 1152900 6 1304100 1020600 7 1152900 1568700 8 1115100 1266300 9 926100 1096200 10 1398600 774900 Average 1198260 DAMP EARTH. DRY EARTH. Number ol Tube. Bacteria per gram. Averages. Number of Tube. Bacteria per gram. Averages. March 30 11 12 13 14 15 1341900 2060100 1436400 2891700 963900 1266300 1719900 1247400 3364200 699300 1699110 March 30 16 17 18 19 20 200 600 50 60 60 300 4200 130 40 20 566 April 1 21 22 23 24 25 April 1 2G 27 28 29 30 30 320 70 40 20 40 50 71 April 5 31 32 33 34 35 April 5 36 37 38 39 40 10 10 10 10 30 10 40 12 April 12 41 42 43 44 33300 900 1800 88200 41400 1800 69300 29587 April 12 46 47 48 49 50 10 6 1 1 12 5 2 4 4 516 SEDGWICK AND WINSLOW. BACILLUS OF TYPHOID FEVER. E. EXPERIMENTS ON THE EFFECTS OF SEDIMENTATION AND CRYSTALLIZATION DURING THE FREEZING OF TYPHOID FEVER BACILLI IN WATER. In the experiments under Section I, the reduction effected represented simply the death-rate among the bacteria due to tlie adverse conditions. All the bacteria in the unfrozen water which did not perish must, from the nature of the case, be present in the thawed ice. In nature, however, the conditions are widely different. Ice is formed immediately over and in immediate contact with a large body of water. In the water, before and during the process of freezing, the bacteria, being particles somewhat heavier than water, continually tend to settle out from the region where ice is to form and fall gradually to the bottom. And when the ice formation actually takes place, a still more powerful force comes into play. In the process of crystalli- zation there is a strong tendency to throw out all substances other than the pure compound chiefly concerned. If, then, soluble chemical compounds, other than hydrogen monoxide are excluded to a large extent when water freezes, this must be still more the case with suspended particles like the bacteria. These a priori conclusions are strengthened by the work of Pengra and of the Massachusetts State Board of Health as well as by common scientific knowledge. To test them more carefully with respect to Bacillus typhi abdominalis and Bacil- lus coli the following experiments were made. A new wine-cask, of about ten gallons capacity, was allowed to stand full of water for a few days in order to remove any extractives present. Four pet-cocks were then screwed in, on opposite sides of the cask, two about four inches from the top and the others an inch or so from the bottom. The whole cask was jacketed with felt so that when placed at a low temperature it would freeze from above down and not from the sides inward. It was then filled with water, at about the boiling-point, drawn from an ordinary water- heater. This water was then allowed to stand for twenty-four hours, when it was found cool and still very nearly sterile, containing three or four germs per cubic centimeter. The barrel of water was then inoculated by pouring into it a bouillon culture of the germ used, the common colon bacillus in the first four experiments, the typhoid bacillus. Race B, in the last two. Dining the course of the experiments no sterilization was attempted beyond that partially effected by the boiling water. After adding the culture and stirring with a sterile rod, samples were taken from the four pet-cocks and planted. The covered cask was then set aside in the room or placed on a broad sill just outside the window of the laboratory, where it was exposed SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVER. 517 to the winter's cold. After twenty-four hours of this treatment a thin sheet of ice a quarter to half an inch thick was found covering the surface. Samples were again taken from the upper cocks just under the ice, and from the lower cocks at the bottom of the barrel, and portions of the ice were also planted, being melted in sterile bottles, after washing with the water produced by their own melting, according to the usual technique. Conclusions, 1. These experiments indicate that sedimentation does not produce marked or constant effects on colon and typhoid bacilli in water during as short a period as twenty-four hours. 2. On the other hand, the experiments show that ice formed on the surface of a quiet body of water contains only about ten per cent of the bacteria present in the water just below. This difference is probably due to the physical exclusion by the process of crystallization, and not to any germicidal action, as the temperature of the ice can only differ from that of the adjacent water by a very slight amount. There are two distinct forces at work, — the low temperature, killing out germs in the ice and water nearly equally, and the crystallizing process extruding germs from the ice into the water below. REDUCTION OF BAGTEEIA BY SEDIMENTATION. B. CoLi. Series I. Bacteria per c.c. in samples taken from top and bottom of cask. December 29, 1898. Averages, Top Bottom 60270 3570 51870 3680 19320 4650 18900 4310 42590 4028 December 30, 1898. Top Bottom 11200 51030 15610 44730 12390 13020 10095 13580 12324 30690 December 31, 1898. Top Bottom 7070 51870 6860 8120 5110 5845 5495 40740 6132 26640 Kept in room. Series II. January 3, 1899. Averages. Top Bottom 120960 114030 110880 97650 114660 103320 101430 85050 114480 100012 January 4, 1899. Top Bottom 54180 52920 42840 47880 60910 60270 66070 62160 53500 56050 Put outdoors. Temperature —6° to —10° C. Surface did not freeze. 518 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVEK. REDUCTION OF BACTERIA BY SEDIMENTATION AND BY FORMATION OF ICE ON FREE SURFACE. B. CoLi. Series III. January 9, 1899. Averages. Top Bottom 28560 25620 21630 10010 23100 32760 20370 12180 23415 20142 January 10, 1899. Ice I/Top J\ Bottom 370 4620 4410 250 4900 10360 550 7490 670 7700 460 4380 7490 Put outside. Temperature —1° C. J inch ice formed. Series IV. - January 11, 1899. Averages. Top Bottom 69930 57330 62370 61110 45990 68670 76860 77490 63787 66150 Jaraiary 12, 1899. Ice |(Top ^ ( Bottom 1240 16720 8820 960 11760 10920 1890 9870 13090 780 8410 13020 1215 11440 11462 Put outside. Temperature, —15° C. J inch ice formed. B. Typhi. Series I. January 18, 1899. Averages. Top Bottom 147420 247590 226800 211680 198450 245700 204120 163090 194197 214515 January 19, 1899. Ice |/Top f \ Bottom 21840 234360 209790 28350 194670 176660 27090 147420 232470 23940 145530 181440 25305 180495 200090 Put outside. J inch ice formed. Series II. January 19, 1899. Avenm^. Top Bottom 202230 154980 198460 218610 156870 302400 171990 254520 182385 232627 January 20, 1899. Ice |fTop ^\ Bottom 68040 270270 307180 75600 404460 257040 18480 578340 386820 17430 319960 238140 44887 393257 297295 Put outside. J inch ice formed. SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVER. 519 IV. DEDUCTIONS FROM THE EXPERIMENTS CONCERNING ICE AS A VEHICLE OF INFECTIOUS DISEASE, WITH SPECIAL REFERENCE TO THE PROBLEMS OF ICE-SUPPLY AND THE PUBLIC HEALTH. Reviewing the several series of experiments described in detail above, and keep- ing carefully in mind the conditions under which natural ice is formed, cut, harvested, stored, delivered, and finally consumed, as well as those pertaining to the manufac- ture, distribution, and consumption of artificial ice, certain conclusions appear to be justified concerning ice as a vehicle of disease; and these conclusions are, on the whole, decidedly reassuring. The conditions which tend naturally to purify polluted waters, are now well under- stood. Light, cold and poor food-supply are antiseptic or disinfectant agents of con- siderable power ; hostile infusoria may devour the living germs of infectious disease ; the chemical composition of the water may be unfavorable to their survival ; and gravity may cause them to settle to the bottom, where they may soon perish for want of air. The main factor determining the reduction of germs in water is, however, the time, — the time during which these and other forces are left to act. Epidemiology shows clearly that disease follows best a direct, quick transfer of infectious material from patient to susceptible victim; and, if storage of water for some months could be insured, many sanitarians would consider such storage a sufficient purification. In ice we have this condition realized, — a forced storage of at least weeks and at best of many months. At the same time the other effective conditions are also heightened. It is no wonder, then, that our experiments show a reduction of over 99 per cent in typhoid bacilli frozen ; and we may be sure that in nature the destruction would exceed, rather than fall short of, such a limit. This reduction obtains in tubes which are frozen solid, where there is no chance for mechanical exclusion. In natural ice there is another purifying influence. Of the germs remaining in the water at the time of freezing, 90 per cent are thrown out by the physical phenomena of that process. This reduction is separate from, and supplementary to, the disinfecting action of the cold. Accordinglj', when both factors work together, it is obvious that only one out of a thousand typhoid germs present in a polluted stream has a chance of surviving in the ice. Under natural conditions the pathogenic germs present in the most highly pol- luted stream are comparatively few. Of these few, one-tenth of one per cent may be present in ice derived therefrom. But even these scattered individuals are weakened by their sojourn under unfavorable conditions, so that, as we have seen. 520 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVER. they require nearly twice as long for their development as do the normal germs, and these few and weakened germs very likely could not produce many, if any, cases of typhoid fever, for vitality and virulence in disease germs are probably closely related. With artificial ice the case is somewhat different, for such ice is made from water frozen solid, and is, as a rule, quickly consumed. Artificial ice, if made from pure water, should be above reproach ; but if it be made from water that is impure it may contain the germs of infectious disease ; and inasmuch as artificial ice is used quickly after its manufacture, the possibility of purification by time is excluded, and such ice might therefore conceivably be a menace to the public health. With natural ice, as long as absolute sterilization is not effected, there must always remain a certain element of doubt, as in the use of sand filters, alluded to above, or in the practice of room-disinfection after contagious diseases. The thickness of a layer of ice is often artificially increased by cutting holes in it and flooding that already- formed with the water of the pond. In such a case the effects of crystallization are excluded, as in the laboratory tubes. Ice thus formed might be cut at once, and served within a week or two ; and in such an exceptional case we cannot say that sufficient of the virus might not persist to excite the malady. Yet such an instance must be very exceptional; and the general result of human experience, the absence of epidemics of typhoid fever traced conclusively to ice, the fact that cities like New York, and Lowell and Lawrence in Massachusetts, have used the ice of polluted streams, and have yet maintained low death-rates from typhoid fever, all tend to support the conclusion at which we have arrived, namely, that natural ice can very rarely be a vehicle of typhoid fever. PART II. STATISTICAL STUDIES ON THE SEASONAL PREVALENCE OF TYPHOID FEVER IN VARIOUS COUNTRIES AND ITS RELATION TO SEASONAL TEMPERATURE. I. A REVIEW OF THE LITEEATURE ON THE SEASONAL PREVALENCE OF TYPHOID FEVER. The variations in the prevalence of typhoid fever with the changing seasons was one of the characteristics of that remarkable disease which struck the very earliest observers. Elisha Bartlett, in 1842/*'^' wrote of it as follows: " It is not settled whether typhoid fever occurs, with any degree of uniformity, more frequently in one season of the year than in another. ... I am sure, however, that, as a general rule, its annual prevalence is greatest in the autumn. In New England it is not unfrequently called the autumnal or fall fever." Dr. Flint, in 1855,^*^^ pointed out as one of the points of distinction between typhus and typhoid fever that while the former is unaffected by season, the latter " manifests a predilection for the autumnal months, although it is by no means restricted in its occurrence to the latter." Griesinger, a little later, '*^' noted that in middle Europe and North America the majority of cases as well as the epidemic out- breaks occurred most abundantly in autumn, and that the winter typhoid stood next in relative intensity, followed by that of summer, while the fewest cases occurred in the spring. He quoted Lombard as authority for the fact that in Geneva the month of October shows seven times as many typhoid cases as the month of March. In 1860, Dr. Tweedie^*^ published a table of the admissions of the different forms of continued fever into the London Fever Hospital for ten years and brought out an interesting contrast between typhoid and typhus fevers. His monthly figures for typhoid were as follows : — J F M A M J J A s N D 113 85 77 60 79 119 157 233 260 253 223 161 522 SEDGWICK AND WINSLOW. — BACILLUS OP TYPHOID FEVEE. By quarters the difference between the two forms of fever, then just beginning to be clearly distinguished, was shown very markedly. QuARTEKLY Admissions. Typhus Fever. First Quarter 1074 Second " 1088 Third " 725 Fourth " 619 Typhoid Fever. 275 258 650 637 Dr. Tweedie concluded that " typhus is most prevalent in spring, and the least so in autumn, while enteric fever is least prevalent in spring, and most prevalent in autumn." In the same year, Hirsch, in the first edition of the " nistorisch-geograpM- scJien PatJwlogie," ^^' gave an extensive resum^ of current opinion on the subject. He quoted statistics to show that of 519 typhoid epidemics, 168 occurred in autumn, 140 in winter, 132 in summer, and only 79 in spring. He also printed a table of typhoid cases at the hospitals of Lausanne and Geneva, in Lowell and Nassau, and of typhoid deaths in the canton of Geneva and the State of Massachusetts, showing an autumn maximum and a spring minimum in every case. Summer occupied the second place except at Nassau and the canton of Geneva. As to the weather influences controlling this prevalence of the disease he quoted very conflicting opinions. While Drake and Huss attributed the autumnal fever largely to the summer temperature, Davidson and Lombard considered a relatively high humidity as of prime significance. Thomson maintained that both factors were of importance, and Seitz, Cless, and Franque denied any effect of meteorological conditions. Another review of the seasonal variations of typhoid fever was published by Murchison in 1862.'^^' He quoted nine English and continental authorities as recording the autumnal maximum, and added a table of the admissions into the London Fever Hospital which showed a steady rise from April to October. Fiedler, in the same year,^®*' noted that typhoid fever in Dresden was much more abundant in the second half of the year than in the first, and gave the following table of typhoid admissions for eleven years. Admissions to the Dresden Hospital, 1850-60. J F M A M J J A s N d 123 76 114 82 83 105 113 191 189 132 143 146 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVER. 523 Tlie first systematic attempt to show a relation between typhoid fever and defi- nite meteorological conditions was made by Haller in 1860/^'^ This author main- tained that the seasonal curve of typhoid corresponded to that of air pressure, and that the greatest prevalence was at periods of low temperature, noting, in that con- nection, the alleged fact that typhoid fever does not occur autochthonously south of the isotherm of 22° C. Haller's results, however, were not confirmed by other observers ; and a new theory as to the aetiology of typhoid fever soon took almost complete possession of the field. This was the famous ground-water theory of Petten- kofer and the Munich school. As applied to typhoid fever this theory was launched by Ludwig Buhl in the first article of the first number of the " Zeitschrift fiir Biologie." '®^' The author dealt with eight hundred and ninety-nine typhoid deaths in a Munich hospital during the period 1855-64, and compared, by the graphic method, the monthly and yearly variations with the changes in temperature, precipitation, and ground-water level. . The seasonal curve showed a maximum between December and March, culminating in February, and a minimum in August and October. These monthly variations, and the fluctuations from year to year, did not correspond to the temperature or the precipitation, but did show a certain inverse relation to the height of the ground water. Seidel ^^^ analyzed the figures given by Buhl in a more elaborate manner. He compared for each of the one hundred and eight months, from 1856 to 1864, the typhoid cases and the ground-water level, using in each case the difference between the value for the individual month and the average value for that month during the whole period. In 73.5 cases an excess of typhoid fever corresponded with an excessive fall of the ground water, and in 34.5 cases the reverse relation obtained. Seidel estimated the probability of this preponderance being due to chance alone as one to thirty-six thousand. His monthly averages for morbidity are as follows : — Typhoid Cases. Munich Hospital. Average, 1856-64. J p M A M J J A s N D 14.1 12.0 6.9 5.2 5.2 6.0 4.8 6.8 4.2 7.6 12.2 13.1 In the next year, Seidel '"^ analyzed Buhl's figures in relation to the monthly precipitation, again excluding any difference of season per se, by using only the differences between the value for a month and the average value for the same month 524 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVER. during the nine years considered. He demonstrated a certain inverse relation between an excess of precipitation and the prevalence of typhoid fever just as in the case of the variation in ground-water level, and considered both factors as of impor- tance. Of the fifty-six months in which precipitation and ground-water level varied in the same sense, forty-six showed a variation of typhoid morbidity in the opposite sense. The studies relating to the cases at the Munich Hospital were extended to the whole city by Pettenkofer in ISBS.'*^' He reproduced a chart prepared by F. Wagus, which gives by months the typhoid mortality for the whole city from 1850 to 1867 in comparison with the precipitation and the height of the ground water. The seasonal distribution of the disease coincided with that observed at the hospital, the average number of typhoid deaths for the whole city being as follows : — J p M A M J J A s N D 33.5 36.8 31.8 23.1 17.6 15.2 15.8 16.7 16.1 15.0 19.0 28.5 A long series of polemical papers on the relation of typhoid, and more particularly of cholera, to the ground water was contributed by Pettenkofer to the " Archiv fur Hy- giene" a.nA the " Zeitschrift fiir Bioloffie," nnd his conclusions were finally summarized in pamphlet form.^"^'' '"*' For a time the theories of the Munich school appeared to hold the field. Virchow '^"^^ studied the typhoid mortality in Berlin for the period 1854-71, and concluded that there was a striking inverse relation with the ground- water level. Virchow and Guttstadt ^"*' published curves for Berlin from 1883 to 1885, which showed a direct relation to the temperature and an inverse relation to the ground-water level. Finally, a most elaborate presentation of the facts was made by Dr. Soyka in 1887.^"^^ Like his confreres, this author rested his case in large part on the variations in the intensity of the disease and the height of the ground water from year to year ; but he also treated of the seasonal variations at some length. Although his table of the monthly distribution of the disease in seventeen cities, reproduced below, showed an autumnal maximum in all but four cases, he considered that these exceptions, Augsburg, Munich, Prague, and Vienna, proved the temperature relation to be an indirect one. SEDGWICK AND WINSLOW. — BACILLUS OF TTPHOID FEVER. 525 Percentage Monthly Distkibtjtion of Typhoid. After Soyka. Place. Period. Total No. J. F. u. A. M. J. J. A. 8. 0. N. D. Berlin Neufchatel ) Lausanne j Breslau Frankfort-a-M. Hanover Basel Paris Augsburg Bern Munich Prague Vienna Basel* Leipzig * Copenhagen * Bremen* Chemnitz * Christiania * 1854^85 1835-52 1863-78 1853-85 1874-85 1826-73 1867-78 1856-78 1871-80 1851-85 1873-84 1871-85 1875-85 1851-65 1842-58 1872-84 1838-82 1845-64 16660 933 2521 1496 397 2213 4152 1092 340 7530 998 4992 3599 1052 3198 1648 1455 4550 6.5 8.6 7.7 7.9 7.6 8.6 6.2 11.0 9.7 11.5 10.5 8.2 10.3 9.4 6.1 7.6 6.2 11.3 6.0 5.2 7.5 7.1 5.1 6.4 5.7 6.7 6.8 11.9 9.9 7.1 7.1 5.7 3.3 7.0 6.4 7.3 5.4 5.5 7.5 6.2 6.4 6.1 4.6 8.1 7.3 11.2 10.2 11.8 8.0 5.1 3.2 6.6 7.3 6.1 5.9 2.6 6.4 5.6 6.1 5.4 4.9 5.3 9.1 9.0 8.5 10.1 6.7 4.3 2.8 4.8 5.2 4.3 5.9 4.0 6.1 5.8 10.0 7.2 4.2 5.1 6.1 7.5 9.3 9.9 8.0 3.8 3.1 4.9 5.1 4.0 5.6 6.1 7.4 6.2 7.4 7.6 4.9 5.2 7.3 6.9 9.6 8.0 8.2 6.0 5.0 4.7 6.9 3.3 8.1 7.7 8.5 8.4 5.0 8.4 6.9 7.3 5.8 6.4 9.8 8.1 10.1 9.3 7.9 8.1 7.4 6.1 11.0 10.1 9.6 10.6 9.1 9.1 12.3 8.9 6.1 6.5 6.9 7.5 14.8 13.0 13.3 96 9.3 8.8 12.6 13.4 11.3 11.8 12.1 10.7 13.4 9.7 10.0 6.3 7.1 7.3 8.6 12.9 18 3 13.8 13.2 8.6 13.5 17.0 10.5 12.1 13.6 10.7 12.5 9.7 7.9 5.8 5.0 7.3 6.9 13.2 16.4 16.3 13.2 9.6 10.6 9.8 8.7 9.0 9.6 10.6 13.5 10.6 12.9 6.9 6.2 6.9 5.7 9.4 9.9 9.1 10.8 16.8 8.2 9.4 8.0 8.7 8.0 8.7 10.3 10.8 10.6 9.6 6.« 7.7 4.9 7.2 10.2 7.0 8.0 13.2 * Morbidity. Other figures refer to mortality. Soyka finally plotted the typhoid fever and ground-water level in Berlin, Frank- fort, Bremen, and Munich, and obtained quite regular complementary curves. His final conclusion was that '• the rhythm of typhus abdominalis is in general the inverted rhythm of the ground-water fluctuations." Unfortunately other researches did not harmonize with these results. Socin at Basle ^™^ and Fodor at Buda-Pesth ^"*" found quite different relations between typhoid and ground-water level. Later examinations of the yearly variations, even in Munich, failed to show the correspondence noted prior to 1881. Most potent of all, however, in overthrowing the ground-water theory was the gradual substitution of zymotic for miasmatic conceptions of disease which robbed it of any rational, aetiological basis. The only plausible explanation of the connection between, ground water and typhoid fever, on the basis of the germ theory, had been furnished by Lieber- meister,^'*^ who suggested in 1860 that the phenomena observed by Buhl might simply be due to the concentration of soil impurities in wells at the time of low water and their transmission in unusually large doses to those who drank therefrom. A simple modification of Liebermeister's idea, including a recognition of the fact that a well in use drains a wider area when the ground water is low and is thus liable to pollution from more distant sources, has been strongly advocated in this country by Dr. H. B. Baker of Michigan. As early as 1878 Dr. Baker ('"^^ published curves showing the 526 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVEK. seasonal distribution of the more important diseases, and pointed out the contrast be- tween such diseases as bronchitis, pneumonia, and croup which culminate in the winter and the fevers and diarrhceal diseases which attain a maximum in the hot months. His curves showed a slight rise in October for typhoid fever and much more marked rises for the classes of " Typho-malarial," " Remittent," and " Intermittent " fevers, the figures for which in absolute value greatly exceeded those for the former disease. Similar tables were published in the succeeding annual reports ; and in 1882 it was stated that " more than the average per cent of weekly reports stated the presence of typhoid fever in months when the average daily temperature, the average daily range of temperature, the absolute humidity of the atmosphere,, the monthly and the average daily range of the barometer and the average daily pressure of the atmos- phere were greater than the average for the year ; and less than the average per cent of reports stated the presence of typhoid fever in months when these conditions were less than the average for the year." These curves and conclusions have been repeated year by year in each annual report, the only change being the gradual increase of "typhoid fever" relative to the "typho-malarial" and "remittent" fevers with improvement in diagnosis. In 1884, Dr. Baker ^"*^ treated typhoid fever in more detail, comparing the seasonal variations of the disease for five years with the height of the ground water in Michigan and showing that the disease increased quite regularly with the number of inches of earth above the water in the wells. He concluded that " in summer when vegetation is active and not decaying, a lowering of the water is imiformly followed by increased prevalence of typhoid fever ; with the advent of colder weather there is a rise in the water level which is uniformly followed by a decreased prevalence of the fever ; that this decrease continues through the winter and spring even though the level of the well water is lowered, provided the surface of the earth is deeply frozen ; that on the contrary high-water level in wells in winter and spring coincident with ground not thoroughly frozen is followed by increased prevalence of the fever." The relation to ground water was again studied in the Report of the Michigan State Board of Health for 1888 (p. Iv.), and 1890 (p. 247); and in the Report for 1894 (p. 300) and succeeding reports, new diagrams were published and the following conclusions were added : " The evidence is conclusive that there is a necessary relation between the low water in wells and the sickness from typhoid fever. The fluctuations in the sickness from typhoid fever and the depth of the water in wells are nearly coincident throughout the several months. The maximum of sickness and the minimum of water are coincident in October." Finally, in 1897, Dr. Baker ^'^i SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVER. 527 printed a new diagram exhibiting the curves of typhoid fever and ground water for fourteen years, and suggested in support of his explanation of the inverse relation shown that another factor of less universal importance than the pollution of wells by distant privies might be the infection of air, food, and drink by germs blown from the surface of the ground, which must be dryer and more exposed to such action when the ground water is low. Dr. Baker's theory regarding the pollution of wells at times of low water seems quite insufficient to account for such a universal phenomenon as the autumnal maximum of typhoid fever, even with the additional suggestion as to air contagion. Well water is by no means the most important source of the disease ; and even as to wells the theory does not take all the facts into account. Other observers have attempted to trace with some success an almost exactly opposite relation between typhoid fever and excessive precipitation. Dr. F. H. Welch,<"^> for example, who noted that the maximum of typhoid fever occurred in the last quarter of the year in Malta and in Bermuda, in the latter half of the year at Gibraltar, during the autumnal months, — from March to May, — at the Cape of Good Hope, and in the warm season in India, finally concluded that " the great natural assistant (in the spread of the disease) is the rainfall in giving moisture for growth and putrefaction, in causing water circulation on the surface and in the subsoil, in its mechanical removal of material from drains and hidden receptacles." Whatever the explanation, it seems to be proven that at Munich in the period studied by Pettenkofer and his followers a real relation did exist between ground-water level and typhoid. In no other case, as far as we are aware, has another factor been excluded which normally varies inversely with the ground-water level and which does bear a plausible relation to the distribution of the typhoid germ. This factor is the temperature ; and the seasonal curve in many places, Michigan, for example, and Berlin, can be more satisfactorily explained by a direct relation to the temperature than by an inverse relation to the ground-water level. The first author forcibly to call attention to the importance of the temperature factor was Murchison. In the second edition of his work on the continued fevers,^^"^' he gave a table of the monthly admissions into the London Fever Hospital from 1848 to 1870, of which the totals were as follows : — J F M A M J J A s N D 433 306 318 209 232 335 434 721 803 839 819 539 528 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVER. Murchison pointed out that a " great increase of enteric fever in the autumn months was observed in each of the twenty-three years, with one noteworthy exception (I860)." He also noted that the autumnal increase did not subside immediately on the advent of winter, and concluded that " it would seem as if the cause of the disease were only exaggerated or called into action by the protracted heat of summer and autumn, and that it required the protracted cold of winter and spring to impair its activity or to destroy it." He quoted numerous observers, Todd and Burne in England, Stewart in Scotland, Lombard and Rilliet and Barthez in Switzerland, Piedvache, de Claubry and Druher in France, Forget and Quincke in Germany, and Bartlett, Wood, and Flint in the United States, as recording the autumnal character of the disease. Finally he added, " Not only does enteric fever increase in autumn, but it has been found to be unusually prevalent after summers remarkable for their dryness and high temperature, and to be unusually rare in summers and autumns which are cold and wet." The references to the early authorities quoted by Murchison will be found in his elaborate bibliography. Liebermeister also had a clear conception of the possible effect of temperature upon the prevalence of typhoid fever. In his article on typhoid fever in Ziemssen's Cyclopedia,'^"^^ he plotted the monthly deaths in Berlin and hospital admissions in London and Basle, compared with curves of the monthly variations in temperature, and commented on the results as follows : " The general bearing of these curves is evident. The curves representing the frequency of typhoid correspond to the curves of average temperature, only with this difference. The different points of the typhoid curve follow those of the temperature curve by an interval of some months. The minimum of temperature falls in January, that of typhoid in February or April ; the maximum of temperature falls in July, that of typhoid in September and October. It appears, therefore, that the development and spread of typhoid fever is favored by the high summer temperature and checked by the low winter temperature. The interval of two or three months between the temperature and the typhoid curves correspond to the time which is necessary for the changes of temperature to penetrate to the places where the typhoid poison is elaborated, for the development of the poison without the human body, for the period of incubation, and for the time between the commencement of the attack and that of the patient^s admission to the hospital, or that of his death." Cousot,'^"*' in France, about the same time, noted that the month of October always showed a maximum of typhoid, that the intensity then diminished till spring, and SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVER. 529 that the summer was marked by unimportant oscillations. This influence of the season he attributed to the effect of temperature and moisture, and he concluded that a moderate temperature accompanied by humidity furnished the conditions most favorable for the spread of the disease. Further evidence was contributed by Buchan and Mitchell/^"^' who tabulated deaths by weeks from all causes distinguished by the Registrar-General in London, for thirty years, 1845-74, and for each disease plotted a curve showing the average weekly deviation from the general weekly mean. For typhoid fever only the six years, 1869-74, were available as prior to 1869 typhus, typhoid and continued fevers were not distinguished. The curve showed a maximum in October and November and a minimum from the middle of May to the end of June, the rise beginning only at the beginning of July, " when the heat of summer has fairly set in." Pistor,'""' who compared the typhoid cases and deaths for 1883-85 in Berlin, with the height of the ground water and of the river Spree, the precipitation, the height of the barometer, and the temperature of the air and the earth, differed from Virchow and Guttstadt (see above) in finding no mai'ked correspondence with the ground-water variations. As regards temperature, he concluded that " typhoid is in general more abundant in the hot months than in the cold ; it appears, however, that mild and damp spring, autumn, and even winter months favor its spread, although not in the same degree as the hot season." Almqiiist,'™^ who studied in detail the seasonal prevalence of fourteen diseases in Gdteborg, concluded with regard to typhoid fever that an annual increase in summer or autumn is characteristic, but that this increase is sometimes postponed till the end of the year or the beginning of the next year. A second maximum in January is sometimes combined with the summer maximum. Dryness and the variation in the ground-water level, and above all the warmth in summer and autumn, appeared to him to be operative. Goldberg,'^""' in 1889, made an elaborate study of -the seasonal prevalence of a large number of diseases in relation to various meteorological conditions, and arrived at the conclusion that the weather influences the mortality from the infectious diseases both by its effect on the multi- plication of the germs and their facilities for entrance into the body and by its effect on the vital resistance of the human body in its reaction against the invading organ- isms. With regard to typhoid fever he analyzed the statistics for Berlin, Hamburg, and Cologne, and summed up his results as follows : — A. As regards individual disposition, the extremes of air temperature weaken the resistance against typhoid. B. As regards time-and-place disposition : 34 530 SEDGWICK AND WINSLOW. BACILLUS OF TYPHOID FEVEE. 1. The rise of typhoid morbidity and mortality in Berlin regularly follows the rise in the temperature of the earth one-half to one meter below the surface. 2. The very different annual periods and annual variations in Berlin, Hamburg, and Cologne correspond throughout to the rhythm of the movements of the ground water. 3. The distribution of rainfall in Berlin and Hamburg, if allowance be made for evaporation, explains satisfactorily the variations both in the height of the ground water and the frequency of typhoid fever, Goldberg noted what so many other observers have failed to consider that not only the temperature of a given month but also the course of the temperature curve during the months immediately preceding, must be considered ; thus the same mean monthly temperature in May and October need not correspond to the same amount of typhoid. He saw that a high temperature favored the spread of typhoid fever, and believed that this was due to a lowering of the vital resistance of the human body by extremes of temperature. The most important evidence bearing upon the relation of heat to the prevalence of typhoid fever was that collected by Davidson in his " Geographical Pathology,'' published in 1892.^^^^^ This author strongly emphasized the seasonal character of the disease and considered the temperature to be the one factor of prime importance. He stated that in South Australia, Victoria, and New South Wfiles typhoid attains its maximum in the autumn months of March, April, and May, and its minimum in September, October, and November. In Queensland the maximum seems to fall upon the hot season, from November to February. For India, he concluded that in the Bengal Presidency the disease attains its maximum in the second quarter and in Central India, Bombay, and Madras in the third quarter. In considering England and Germany, he mentioned the usual autumnal maximum; and for several countries as quoted below, he gave specific figures as to monthly prevalence. MoKTHLY Prevalence of Typhoid Fever. Compiled from figures given by Davidson. Place. Period. Number of Gases. Monthly Percentage of Total for Year. J. p. M. A. 2.7 M. 5.3 J. J. A. s. 0. N. D. Finland 1889 639 3.1 4.2 2.8 3.1 11.9 20.5 11.7 12.4 13.6 8.6 France ) (Paris) i 1868-78 — 6.2 5.7 4.6 4.9 4.2 4.9 6.9 12.3 13.5 12.5 13.6 10.3 France ) (Marseilles) J Italy 1886-87 three years — 6.7 6.7 4.2 6.5 4.4 6.8 4.5 7.2 6.5 7.3 7.0 7.2 10.4 9.4 14.6 11.2 14.6 11.1 11.0 10.6 8.2 8.6 7.8 7.4 Norway 1886-87 3138 11.3 7.3 8.9 8.4 5.8 6.1 7.0 8.1 9.5 10.5 8.7 8.4 Scotland (principal towns) Sweden I 1876-85 1886-87 3548 10743 8.5 8.9 7.7 6.5 7.4 6.8 7.4 5.9 8.8 6.3 7.4 5.7 5.9 8.1 7.4 10.3 9.6 11.5 11.7 10.0 8.7 11.2 9.4 8.7 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVER. 531 Davidson also attempted to show the causal relation between typhoid fever and temperature variations from year to year after the method adopted by Soyka in treating of the ground-water theory. In the case of New South Wales he took the figures for the period 1877-87, with a mean summer temperature (December to February) of 71.14 F., and a mean typhoid death rate of 5.02 per 10,000, and divided them to form the two following tables. Six Yeaks with Temperature and Typhoid Kate above the Mean foe the whole Period. i8»r. isrs. issa. 18S4. 18S5. 1886. Mean Summer Temperature Mean Typhoid Death Rate 71.40 5.96 72.00 6.70 71.17 5.66 71.47 5.86 71.87 5.40 72.10 6.03 Five Years with Temperature and Typhoid Rate below the Mean for the whole Period. isro. 1880. 1881. 1883. 1887. Mean Summer Temperature Mean Typhoid Death Rate 71.00 3.84 70.17 3.31 70.03 3.50 70.07 4.76 71.10 4.24 Again, in the case of England, Davidson separated from the period 1863-87, four years in which enteric fever was unusually prevalent, and five years which were rema,rkably free fronj that disease, and tabulated the relative mean temperatures for those years as follows : — Pour Tears with Maximum Typlioid. Five Tears with Minimum Typhoid. Tear. Difference between Temperature and Mean Temperature, 18C3-87. Tear. Difference between Temperature and Mean Temperature, 1863-87. For the Tear. For the Third Quarter. For the Tear. For the Third Quarter. 1865 1878 1880 •■. 1884 +1.0 +0.3 +0.1 +1.4 +2.1 +0.4 +1.0 +2.3 1867 1877 1879 1881 1885 -0.7 +0.1 -3.1 -0.6 -0.7 -0.7 -1.9 -2.3 -0.4 -1.3 These investigations of the yearly variations in typhoid fever are of considerable interest and should be extended ; but the differences shown by Davidson are so small and the material so limited as to preclude the drawing of any general conclusions. The clearest and most definite statement of the effect of temperature upon the spread of typhoid fever that we have seen was made by Professor Woodhead in testifying before the Royal' Commission on Metropolitan Water Supply in 1893.^'^^ Having spoken of the importance of spring floods in carrying infection into 532 SEDGWICK AND WINSLOW. — BACILLUS OP TYPHOID FEVER. water-supplies, he was asked why the maximum of typhoid occurred in autumn instead of at the time of the greatest floods, and his reply was as follows : — "You were speaking just now of the conditions under which the typhoid bacillus develops, and you were speaking of it as being a pathogenic organism, and therefore as not competing on equal terms with the saprophytic organisms ; and here the matter of temperature alone plays such a very important part that it cannot be left out of consideration. Although you have in February the highest point of floods, you have the temperature so low that the typhoid bacillus could scarcely develop under any conditions, whereas when you come to August, when the temperature is much nearer that of the body, that is, the temperature under which the typhoid bacillus can exist, then the conditions become so much more favorable that the organism can live more readily, more easily, and become more virulent outside the body than it can when the temperature is put very much lower, and, therefore, although at flood times the highest flood points one would expect (if you leave out the temperature) the typhoid bacillus to do the greatest amount of damage, still the temperature is so low that the presence of the bacillus is practically a matter of no importance at that period, and it is only when you get to the flood periods when the temperature is higher that you can take these statistics as bearing on the point. But beyond this, should there be a sporadic case of typhoid due to the use of contaminated water, the conditions for the propagation of the disease are not nearly so favorable during the cold months of February as they are in the hotter months of the year, and therefore the health returns and the tables would be much less affected, not only at the time of the primary outbreak but for some little time afterwards." Plausible as the conclusions of Murchison, Davidson, and Woodhead appear, they have not gained wide acceptance, and in Germany have been utterly ignored, except by Liebermeister in the passage quoted above. In the same year that his statement appeared, Oesterlen ^*°*' published some figures on the quarterly prevalence of typhoid as given below, and concluded : " That temperature exerts no, or at least a very secondary, influence, is obvious from the very small difference which often appears between the different seasons, and from the circumstance that typhoid epidemics may arise and culminate at the extremes of temperature, in great cold as well as great heat." Quarterly Prevalence op Typhoid. After Oesterlen. Place. Period. Winter. Spring. Summer. Autumn. Geneva Tjondon . . 1849-53 1818-56 1845-49 1840-47 1830-38 180 2813 670 429 130 27 109 2527 470 259 102 18 105 2916 486 528 163 23 203 3306 863 1132 250 41 Nassau Massachusetts Berlin (average monthly deaths) . . . SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVER. 533 A little later, Sanrler'^*^' gave a table showing the quarterly distribution of typhoid fever in Berlin, Munich, Halle, Hamburg, Schleswig-Holstein, Dresden, Leipsic, and Chemnitz, and stated that the winter in Munich and the autumn in most other places is the period of special incidence, while May and June are always the months which are most exempt. In 1881, OldendorfE ^^'^^ published a few figures as to quarterly prevalence, and repeated Oesterlen's conclusion as to the limited importance of the temperature factor. In the second edition of the " Geographical and Historical Pathology, "'"^^ Hirsch devoted considerable space to a consideration of the meteorological factors affecting the spread of typhoid fever. He quoted first numerous earlier observers, to whom references are given in his bibliography. Zulzer at Berlin and Trier at Copenhagen thought that hot and dry weather favored the disease, while others held a wet summer to be a contributory cause. Schiefferdecker at KOnigsberg, Pribram and Popper at Prague, and Jacoby at Breslau believed they had traced a connection between typhoid and the ground-water level. Hirsch then gave the very valuable tables of seasonal prevalence reproduced below, and in comment remarked, " The result obtained from these tables, that the amount of the sickness touches its highest point in autumn, is fully borne out by the facts as to the season of greatest prevalence of typhoid in many other localities." He cited Schwerin, Bremen, Iceland, Malta, Italy, the Cape, Greenland, and Newfoundland ; and added, " All the more noteworthy is the circum- stance that, in tropical and subtropical regions, it is chiefly the hot months that form the typhoid season," quoting Algiers, Tunis, Japan, India, Cochin China, Bermuda, and Cuba. An analysis of the typhoid statistics of Berlin from 1871 to 1878 failed to show any correspondence between the amount of typhoid in any given year and the excess of temperature compared with the mean for the whole period ; and the author concluded his consideration of the subject as follows : " That no special importance in this connection can be ascribed to the temperature of the air — high or low — % itsdf, follows from the fact that the acme of the disease falls variously in various regions within higher latitudes, either in autumn or in winter ; while, in the tropics, it falls mostly at the time of the greatest heats." 534 SEDGWICK AND WINSLOW. BACILLUS OF TYPHOID FEVEK. MoNTHLT Distribution of Typhoid Fever. After Hirsch. Period. Months. J. J. A. B. 0. N. D. J. F. ■ H. A. M. Christiania* . 1845-64 154 281 402 393 437 768 602 517 336 283 196 182 Draramen* . . 1861-67 46 100 149 180 263 251 202 141 92 88 56 56 Copenhagen* . 1842-58 162 254 428 588 526 317 328 195 106 103 92 100 Hamburg . . 1873-80 82 82 122 116 147 127 158 146 149 125 90 102 •Berlin . . . 1854-79 860 1169 1616 1879 1965 1540 1184 997 919 854 921 910 Breslau . . . 1863-78 187 216 244 287 267 220 202 197 192 192 164 164 Leipzig * . . 1851-65 64 98 137 136 144 99 76 100 60 54 44 41 Chemnitz* . . 1837-75 171 208 303 300 245 185 241 148 166 121 112 164 Prague * - . . 1874-76 78 90 69 79 76 84 115 191 122 119 106 110 Nassau * . . 1818-59 1118 1406 1742 2093 2350 2207 1946 1860 1684 1428 1060 848 Frankfort-a-M. 1863-80 52 74 91 106 113 93 76 60 58 50 60 43 Stuttgart . . 1852-77 69 76 83 87 88 108 122 106 84 90 73 06 Munich . . . j 1852-68 1873-79 |-408 377 379 366 363 426 619 718 783 699 548 444 Neufchatel ) . Lausanne I . Basel . . . 1835-52 67 72 95 126 159 92 88 81 49 52 25 38 1824-73 169 186 202 237 237 236 193 192 143 137 121 160 London * . . 1848-62 163 220 333 361 377 334 222 197 122 136 89 103 Glasgow * . . 1871-79 12 16 30 43 36 31 20 23 18 29 18 17 Paris .... 1867-78 205 289 611 569 522 565 429 259 240 192 205 176 Boston * . . 1840-47 30 47 86 92 98 60 48 39 43 40 21 41 Pittsburg . . 1873-77 27 32 66 64 90 65 62 53 37 43 44 63 * Hospital admisBioDg. Other figares refer to reported deaths. Seasonal Ratio of Ttphoid. After Hirsch. Place. Autunui. ■Winter. Summer. Place. Autnnm. winter. Summer. Copenhagen . . 4.9 2.1 2.9 Geneva .... 1.9 1.7 1.0 Drammen . . . 3.4 2.2 1.5 Chemnitz 19 1.4 1.8 Lausanne . . . 3.3 1.9 1.9 Basel 1.7 1.3 1.3 London .... 3.2 1.7 2.2 Glasgow 1.7 .9 .9 Paris .... 2.9 1.6 1.8 Pittsburg 1.5 1.0 .9 Massachusetts 2.8 1.3 1.6 Breslau . 1.5 1.2 1.3 Leipzig .... 2.7 1.7 2.1 Sweden . 1.2 1.2 1.1 Christiania . . . 2.4 2.2 1.3 Hamburg 1.2 1.3 .9 Boston .... 2.4 1.2 1.6 Stuttgart 1.2 1.3 1.0 Frankfort-a-M. . 2.2 1.3 1.5 Munich . .7 1.3 .7 Berlin .... 2.0 1.2 1.4 Prague . .7 1.3 .7 Nassau .... 2.0 1.6 1.3 These ratios refer to a value of 1 for the Spring Typlioid. Spring is considered to begin with March. SEDGWICK AND WINSLOW. UACILLUS OF TYPHOID FEVEK. 535 The work which has been done upon the seasonal prevalence of typhoid fever within the last ten years has, if anything, only made the subject more obscure. Magelssen, in his classic brochure ^*^*' on the dependence of diseases upon the weather, in which he showed so clearly the unfavorable influence of extreme low temperatures upon the general mortality, only alluded to typhoid in passing, stating that it is most abundant in the latter months of the year. KOrOsi, in 1894,^*^^ made an elaborate comparison of the reported cases of the infectious diseases in Berlin with the moisture and temperature by periods of five days, a week and a month, according to the incubation period of the disease. He criticised those observers, especially Haller, who have studied the relation of disease to season, in general, on the ground that such a comparison can throw no light on the causation of disease as the phenomena involved are too complex. His method consisted in the division of his pentads and months into five groups, designated as very cold, fairly cold, fairly warm, warm, and hot, and the calculation of the relative prevalence of the disease in each group of periods. He thus eliminated all the effects of the weather preceding the period con- sidered and obscured the facts. When analyzed into his five temperature groups, two maxima appeared, — one in the hot, one in the fairly cold months, — and he concluded that no positive relation is shown. Moisture, on the other hand, appeared to exert an appreciable eflFect, and he finally concluded that the maximum of morbidity occurred in dry weather with medium warmth, while the minimum was reached when a medium temperature coincided with an excess of moisture. Fodor, in 1896,*''®' declared that "the striking dependence on the warmth, and on the season which is so characteristic of cholera is almost entirely wanting in typhoid fever." In the same year, Jessen^^^' published curves which showed the monthly prevalence of measles, croup, and diphtheria, typhoid fever, cholera, pneumonia, phthisis, and diarrhceal diseases of children in comparison with variations in wind, temperature, humidity, and rainfall. With regard to typhoid fever he concluded that temperature was the only factor which aflected the disease, and that this was only of slight importance, as typhoid fever, though occurring principally in the cold months (!), sometimes attained a maximum when the temperature was high. Knoevenagel ^*^^ noted the increased prevalence of typhoid fever in Mecklenburg-Schwerin at the end of July and in August and September. Berger^'^^ and Ruhemann,"^^ in 1898, emphasized the importance of atmospheric conditions in aetiology, and criticised the exclusive attention paid to the bacteriological factors in disease. The former author, after an excellent review of literature on the influence of weather on various diseases (tuberculosis, pneumonia), published curves of morbidity from diphtheria, scarlet fever, measles, and typhoid 536 SEDGWICK AND WINSLOW. BACILLUS OF TYPHOID FEVEE. fever in a rural district for a period of four years. Typhoid fever, although the total number of cases was only twenty-two, showed a maximum in August and a minimum between November and February. Berger concluded that typhoid fever is most prevalent with a falling barometer and a rising thermometer, hy- grometer, and dew point, and that its occurrence is favored by damp and cloudy weather. Ruhemann alluded only in passing to typhoid fever, mentioning its summer maximum. Finally, in 1899, Weichselbaum ^'*^^ concluded that "no seasonnl distribution of typhoid, no preference of that disease for any special time of year, at least in the marked sense in which it has been shown for cholera, has been, or will be demonstrated."* Curschmann, in the latest monograph on typhoid fever, ^*^' notes that this disease shows a " constant and for many countries a uniform relation to the seasons." '• Everywhere the increased frequency occurs during the late summer and autumn months." " The period of least prevalence of typhoid fever is everywhere the spring and the beginning of the summer, especially the months of March, April, and May." He quotes the figures for' London (Murchison), Dresden (Fiedler), and the Hamburg epidemic of 1886-87, and gives a table for Leipsic which is reproduced below. The London and Leipsic figures, when plotted, show very regular curves. Cases of Typhoid Fevek keceived into Jacobsspital, Leipsic, feom 1880 to 1892. J F M A M J J A s n i>y 122 96 97 78 71 75 136 252 240 193 150 ■ '88., In commenting on these facts Curschmann says: "The causes for this remarkable uniformity in the relations of typhoid fever to season are as yet wholly unknown. * Behrens (Einjluss der Wilterung auf DiphtJierie, Scharlach, Maseru und Typhus, Arch. f. Hyg., XL.^ 1901, 1) has recently published an exhaustive study on the influence of weather on the prevalence of diphtheria scarlet fever, measles, and typhoid. His method consists in the arrangement of the individual months for a period of five years in classes according to temperature, humidity, and precipitation, and the tabulation of the morbidity and mortality for. the various classes of months. The cities treated are Carlsruhe, Berlin, Bremen, and Breslau. A series of tables is appended of morbidity in Carlsruhe from the four diseases treated by five-day periods with an elaborate analysis of the meteorological conditions. The results of the investigation are conflicting and incon- clusive. With reference to typhoid fever. Dr. Behrens sums up the evidence from his own work and that of Jessen and Korosi as follows: " Typhoid reaches its maximum in hot weather at Carlsruhe, Berlin, and Breslau, in cold weather at Hamburg, and in weather of medium warmth at Budapest. At Bremen no influence of temperature can be shown. Carlsruhe, Berlin, Breslau, and Budapest agree in the fact that the number of typhoid cases is greatest when the humidity is least; in Bremen, on the other hand, the maximum occurs when the hygrometer is highest. A heavy precipitation and a maximum of rainy days favor the disease in all cases." His final conclusion with regard to this disease is as follows: "Typhoid cases are as numerous with a warmer as with a cooler temperature, but are markedly favored in their occurrence by cloudy and rainy weather." SEDGWICK AND WINSLOW. BACILLUS OF TYPHOID FEVEK. 537 The universality of the relation, its recurrence in all possible, remotely situated regions, indicate that it is dependent not upon local, but upon general conditions, possibly such as are responsible for the power of multiplication and the vital activity of the typhoid germ itself. Although much is known with regard to the details in this con- nection, an insight into the solution of general questions is wanting, particularly the relation of the poison to important cosmic conditions. It is, therefore, better for the present to leave a glaring deficiency rather than to bridge it over with unstable theories." II. STATISTICAL STUDIES BY THE AUTHORS ON SEASONAL VARIATIONS IN TEMPERATURE AND ON THE PREVALENCE OF TYPHOID FEVER IN VARIOUS COUNTRIES. It appears, then, from a review of the literature that, although most observers have noted a characteristic seasonal distribution of typhoid fever, others, including some of those who have written most recently, have denied the existence of such regular variations. Of those who realized that the variations did exist, a few sought an explanation in the factor of temperature. Their views did not, however, gain acceptance, as the evidence furnished was insufficient; and the common view, among medical men and sanitarians, has been that the fall maximum of typhoid fever was an unexplained phenomenon. The bacteriological work on the effect of low temperatures upon th,e bacillus of typhoid fever, reported in the first section of this paper, lent force to the idea that the temperature really might in itself exercise a direct effect upon the aetiology of this disease. We therefore determined to see whether the relation shown by Mur- chison, Liebermeister, and Davidson for a few places could be demonstrated by a more exact examination of statistics collected from a wider field. We have, accordingly, brought together statistics of the monthly variations in tem- perature and in the prevalence of typhoid fever for thirty communities, as follows : The States of New York and Massachusetts, the District of Columbia, and the cities of Atlanta, Baltimore, Boston, Charleston, Chicago, Cincinnati, Denver, Mobile, Newark, New Orleans, New York, Oakland, Philadelphia, St. Paul, and San Francisco, in the United States ; the city of Montreal in Canada ; the cities of Berlin, Dresden, Leipsic, London, Munich, Paris, and Vienna in Europe ; the Empire of Japan, and the British Army in India, in Asia ; and the cities of Buenos Ayres and Santiago de Chile in South America. Four continents and both hemispheres are thus represented, and a very wide range of climate. (See pp. 540-566.) 538 SEDGWICK AI^D WINSLOW. — BACILLUS OF TYPHOID FEVEK. The mean monthly temperatures for the American cities were obtiiined from the reports of the United States Weather Bureau; those for the German cities, from the publications of the astronomical observatories in their respective districts; and those for London, Paris, Montreal, Buenos Ayres, and Santiago from special local publications mentioned in connection with the tables. For the States of New York and Massachusetts, it was assumed that the temperature of New York City and Boston would serve without serious error. For Japan, where the range of temperature is rather wide, an average was taken of the record of ten stations in different parts of the Empire, as given by the Central Meteorological Observatory. In the case of India, it appeared inadvisable to attempt to calculate an average for the whole empire, as the seasons in the different districts are so very different. The typhoid figures are, there- fore, compared with two sets of temperature values, for Central India, and for the Punjab, taken from Hann's " Klimatohgie" which give a fair idea of the two most important meteorological zones. For each of the cities and stations, with one or two exceptions, the figures for ten years have been used in order to secure a reliable average ; and the mean monthly temperatures finally obtained have all been reduced to the Fahrenheit scale for uniformity and convenience in plotting the curves. The typhoid statistics include records of hospital admissions at the two hospitals of Santiago de Chile, of hospital admissions in the British Army in India, of reported cases at Newark and of deaths in all other instances. The figures for the American States and cities, for Montreal, London, and Paris, were obtained from the published reports of the local Departments of Health, supplemented in some cases by informa- tion furnished in reply to correspondence ; the German statistics were taken from the " Veroffentlichungen des KaiaerKchen GesundheUsamtes ; " for Japan, the Annual Reports of the Central Sanitary Bureau, for India, the Parliamentary blue-books, and for the South American cities, local sanitary periodicals referred to in the tables, were con- sulted. The figures for ten years were averaged in each case except as follows: for Vienna and Japan the period was five years ; for Atlanta, six years ; for Montreal and New Orleans, eight years ; for Denver and Paris, nine years ; for the Army in India, eleven years; for Buenos Ayres, twenty-two years. In each case the average number of deaths per month has been reduced to a ratio of one hundred deaths per year, the final figure for each month representing the number that occur in that month for every hundred deaths in the year. Thus the absolute amount of the disease is entirely eliminated, and only its seasonal distribution considered. The value of the statistics will not therefore be impaired by errors of registration, which it may be assumed will not vary from month to month. SEDGWICK AND WINSLOW. — BACILLUS OP TYPHOID FEVER. 539 Finally, the monthly values for temperature and typhoid prevalence have been plotted on the appended plates in order to show graphically the relation of the two curves. For each locality the abscissae represent the successive months, and the ordi- nates the monthly temperature and percentage of annual typhoid. We should not, however, expect the effect of January temperatures to be manifest in the typhoid death-rate until March, as about two months will be taken up in the transfer of the infection to the victim, in the incubation of the disease, and in its course toward a fatal termination. Accordingly, in order to make the relation of the two curves more striking, the typhoid curve has in each case been shifted along to the left by just two months, so that March typhoid comes just above January temperature, and so on. Where cases and not deaths have been considered (Santiago, Newark, India) the curve has been only moved along by one month. This transposition does not, of course, alter the shape of the curves or their relation to each other, but only makes that relation clearer to the eye. (See Plates I.- VIII.) 540 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVER. BOSTON. Monthly Typhoid Deaths. From Reports, Local Department of Health. Tear. J. F. M. A. M. J. J. A. s. 0. N. D. 1888 7 5 5 11 3 11 11 19 31 42 17 18 1889 6 7 7 7 9 12 17 35 33 23 17 13 1890 7 5 7 7 7 8 9 20 27 20 19 19 1891 8 4 11 9 8 4 7 14 29 29 15 16 1892 2 5 7 7 9 6 6 15 18 29 18 15 1893 13 9 6 10 13 12 7 15 14 26 17 6 1894 3 5 5 7 7 4 4 18 30 27 20 11 1895 8 3 6 7 11 8 9 26 28 26 13 18 1896 14 6 2 5 6 7 8 13 30 34 23 14 1897 14 7 9 11 8 9 10 25 27 22 18 13 Average 8.1 5.6 6.5 8.1 8.1 8.1 8.8 20.0 26.7 27.8 17.7 14.3 Ratio of 100 6.1 3 5 4.1 5.1 5.1 5.1 5.5 12.5 16.7 17.4 11.1 8.9 Mean Monthly Temperature. From " Monthly Weather Review," U. S. Weather Bureau. Tear. J. V. M. A. M. J. J. A. s. 0. N. D. 1888 20 28 32 42 52 67 68 69 59 47 43 34 1889 36 26 38 48 60 69 69 67 63 48 45 38 1890 32 33 35 46 57 64 71 ' 70 63 51 42 26 1891 31 32 34 48 56 65 69 70 67 52 41 40 1892 28 28 33 48 56 70 73 70 62 53 41 30 1893 21 27 34 44 56 65 71 70 60 55 42 30 1894 30 27 42 47 58 69 74 68 65 54 38 32 1895 29 25 35 46 60 67 69 71 66 60 45 36 1896 25 29 32 47 60 66 72 71 62 50 46 30 1897 28 31 37 49 58 62 72 70 63 54 41 34 Average 28 29 35 46 57 66 71 70 63 51 42 33 SEDGWICK AND WINSLOW. BACILLUS OF TYPHOID FEVER. 541 NEW YORK CITY. Monthly Typhoid Deaths. From Reports, State Board of Healtli. Tear. J. F. M. A. M. J. J. A. s. 0. N. D. 1887 28 13 21 11 11 16 33 51 53 38 26 22 1888 12 14 13 11 23 11 35 42 82 52 37 33 1889 27 15 21 18 15 19 31 71 57 57 40 21 1890 20 28 13 12 11 11 31 49 64 49 34 29 1891 14 11 17 13 20 23 28 57 65 56 51 29 1892 15 . 25 17 19 23 23 52 53 57 55 31 30 1893 22 19 29 25 29 23 21 35 42 70 41 26 1894 22 11 17 18 11 14 28 42 57 46 32 28 1895 17 16 8 14 13 23 27 37 46 48 37 36 1896 20 17 11 12 10 13 25 42 38 39 34 36 Average 19.7 16.9 1G.7 15.3 16.6 17.6 31.1 47.9 56.1 51.0 36.3 29.0 Ratio of 100 5.6 4.8 4.8 4.2 4.8 5.1 8.7 13.5 15.8 14.4 10.1 8.2 Monthly Temperature. From " Monthly Weather Reyiew," U. S. Weather Bureau. Tear. J. F. u. A. M. J. J. A. s. 0. N. D. 1888 26 32 32 48 58 71 70 72 63 49 45 34 1889 38 28 41 52 62 70 73 71 66 52 47 41 1890 40 40 37 51 61 70 73 72 67 55 46 31 1891 35 37 38 52 60 70 71 74 70 54 44 42 1892 30 33 35 50 59 72 75 74 66 55 43 31 1893 23 30 36 48 59 69 75 74 64 58 44 35 1894 35 30 44 50 61 71 76 73 70 57 42 37 1895 30 25 36 48 59 70 71 74 70 51 46 37 1896 28 30 32 50 64 66 73 73 65 52 48 32 1897 29 33 39 49 59 65 73 71 65 56 44 36 Average 31 32 37 50 60 69 73 73 67 54 45 36 542 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVER. MASSACHUSETTS. Average Weekly Typhoid Deaths fob each Month. From Reports, State Board of Health. Tear. J. F. H. A. M. J. J. A. 8. 0. N. D. 1886 6 5 4 3 3 4 10 15 16 11 10 1887 4 8 8 7 6 7 5 14 22 16 12 7 1888 6 5 6 7 5 6 6 10 16 26 11 8 1889 6 8 7 5 6 6 7 15 18 16 13 8 1890 6 7 5 4 5 5 4 9 16 14 18 15 ■ 1891 15 11 7 7 4 2 4 6 14 15 11 9 1892 6 5 7 4 5 5 6 9 11 37 11 12 1893 9 8 5 6 5 5 4 9 13 17 11 10 1894 5 7 4 5 6 2 4 7 16 15 15 9 1895 4 2 5 6 5 5 5 12 16 12 10 11 Average 6.7 6.6 5.8 5.6 5.0 4.6 4.8 10.1 15.7 18.4 12.3 9.9 Katio of 100 6.4 6.3 5.5 5.3 4.7 4.4 4.6 9.6 13.9 17.4 11.7 9.4 NEW YORK STATE. Monthly Typhoid Deaths. Trom Reports, State Board of Health. Year. J. F. M. A. H. J. J. A s. 0. N. D. 1887 72 57 72 56 37 54 102 194 248 182 149 104 1888 64 84 81 45 59 45 73 174 279 288 153 138 1889 89 71 69 78 63 45 117 224 247 261 169 117 1890 117 94 72 73 72 69 101 167 234 240 216 157 1891 138 127 121 103 88 90 97 171 287 290 241 183 1892 116 98 96 77 71 75 131 182 282 205 184 147 1893 120 101 115 111 93 83 87 157 227 253 180 158 1894 105 86 131 94 85 72 93 183 229 234 189 139 1895 108 99 99 115 92 81 108 156 220 265 204 169 1896 158 121 103 87 59 66 103 171 221 195 132 126 Average 109 94 96 84 72 68 101 178 247 241 182 144 Ratio of 100 6.7 5.8 5.9 5.2 4.5 4.2 6.3 11.0 15.3 14.9 11.3 8.9 iEDGWlCK AND WINSLOW. ■BACILLUS OF TYPHOID FEVEK. 543 ST. PAUL. Monthly Typhoid Deaths. From Report s, Local Board of Health. Tear. J. F. H. A. u. J. J. A. s. 0. N. D. 1888 7 8 4 5 6 4 6 14 27 29 22 10 1890 7 4 2 5 2 2 17 11 6 6 3 1891 3 6 4 1 2 3 2 6 12 10 7 5 1892 2 1 6 1 2 1 4 12 7 11 1893 3 2 1 2 3 1 11 8 9 5 6 1894 1 1 1 2 2 4 6 5 6 4 1895 3 5 3 1 1 3 4 5 2 8 1 2 1896 7 6 3 3 1 1 5 4 5 2 1897 2 2 2 1 1 1 3 3 6 1 Average 3.6 3.9 2.9 2.1 1.4 2.1 2.1 7.1 8.1 8.4 7.2 4.9 Ratio .of 100 6.6 7.2 5.4 3.9 2.7 3.9 3.9 13.2 15.1 15.7 13.4 9.1 Mean Monthly Temperature. From " Monthly Weather Review," TJ. S. Weather Bureau. Tear. J. F. H. A. H. J. J. A. 8. 0. w. D. 1888 -1 12 18 40 50 67 72 66 55 43 33 24 1889 20 10 37 48 56 64 71 70 59 45 29 29 1890 10 18 22 48 52 70 72 65 58 46 36 24 1891 21 11 23 48 58 65 66 67 66 48 26 27 1892 10 21 28 42 51 65 71 69 63 51 28 15 1893 3 9 23 39 54 71 73 69 62 49 30 12 1894 10 14 35 49 58 72 76 72 64 49 27 27 1895 6 11 28 52 59 67 70 70 65 44 31 21 1896 16 21 25 47 63 68 71 70 56 45 22 23 1897 9 19 24 46 57 64 74 66 68 53 29 15 Average 10 15 „ .27. .46 56 .67 . 72 68 62 47 29 22 544 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVEE. DENVER. Monthly Ttphoid Deaths. From Reports, Local Department of Health. Tear. J. F. M. A. H. J. J. A. s. 0. N. D. 1888 8 1 3 2 5 14 22 24 31 21 3 1889 4 1 1 4 1 14 23 61 55 22 12 1890 7 5 2 1 9 7 17 31 56 72 50 30 1891 13 9 4 3 2 3 6 11 15 17 9 7 1892 2 1 2 3 2 6 2 12 9 9 15 1 189.3 4 4 5 8 5 8 4 5 10 15 3 1894 4 2 1 1 3 6 3 8 8 7 48 8 1895 5 1 2 1 2 2 2 5 8 6 8 2 1896 5 2 1 4 6 13 28 17 12 3 Average 5.8 2.6 1.9 1.8 4.0 3.9 8.0 1.5.4 22.7 24.9 22.2 7.7 Ratio of 100 4.8 2.1 1.6 1.5 3.3 3.2 6.7 11.9 18.9 20.7 18.5 . 6.4 Mean Monthly Temperature. From " Moutlily Weather Review," TJ. S. Weather Bureaa. Tear. J. F. H. A. H. J. J. A. B. o. N. D. 1888 27 39 33 53 53 68 71 65 61 48 34 34 1889 27 30 43 51 55 64 72 73 60 52 32 40 1890 28 34 41 48 58 68 72 69 62 49 40 39 1891 25 27 32 48 56 63 70 69 64 62 38 31 1892 26 33 36 46 51 65 72 71 66 60 43 27 1893 38 31 38 45 54 69 73 70 63 51 39 38 1894 31 25 40 50 59 66 72 71 63 54 45 32 1895 28 27 37 50 56 62 67 70 66 51 38 34 1896 37 38 37 50 59 68 72 72 61 50 36 39 1897 27 31 36 47 61 65 70 70 66 51 41 28 Averj^e 29 31 37 49 66 66 71 70 63 $1 39 34 )GWICK AND WINSLOW, — BACILLUS OF TYPHOID FEVER. 645 MONTREAL. Monthly Typhoid Deaths. From Reports, Local Department of Health. Tear. J. F. u. A. M. J J. A. s. 0. N. D. 1888 5 2 4 2 4 4 4 20 24 14 6 5 1889 3 3 2 3 2 2 3 15 19 8 6 6 1891 8 1 2 2 4 7 13 10 8 7 1892 4 6 3 1 2 4 6 8 12 15 4 1893 6 3 2 4 4 3 6 2 6 8 2 5 1894 6 3 4 5 3 1 6 6 1 6 1 1895 1 2 1 2 5 2 4 3 10 6 5 3 1896 3 3 2 2 3 1 7 4 4 9 4 4 Average 3.5 2.7 3.2 2.4 3.0 2.0 4.0 7.9 11.2 8.5 6.5 4.4 Ratio of 100 5.9 4.6 5.5 4.0 5.1 3.4 6.7 13.3 18.9 14.3 10.9 7.4 Mean Monthly Tempeeatuke. From Reports, Local Department of Health. Year. J. F. M. A. M. J. J. A. s. 0. n. D. 1888 4 12 23 37 54 66 68 64 55 40 33 23 1891 15 19 27 40 52 64 69 65 58 46 32 7 1892 15 17 26 42 52 65 66 67 62 45 35 30 1893 15 18 23 41 53 66 70 66 57 46 33 19 1894 13 13 32 45 56 66 69 63 60 49 30 23 1895 15 14 22 41 58 70 67 66 60 41 34 22 1896 12 15 20 41 58 65 69 67 57 43 35 18 Average 13 15 25 41 55 66 68 65 58 44 33 20 35 546 SEDGWICK AND WINSLOW. BACILLUS OF TTPHOID FEVEE. BALTIMORE. Monthly Ttphoid Deaths. From Reports, Local Department of Health. Year. J. F. H. A. H. J. J. A B. 0. N. D. 1888 7 8 6 6 5 10 4 26 84 21 17 17 1889 15 7 14 4 12 16 8 30 26 14 19 26 1890 10 12 15 19 13 13 29 36 30 34 25 11 1891 15 8 3 5 9 6 9 14 22 29 17 13 1892 13 9 8 9 11 8 16 30 26 29 21 13 1893 20 5 11 10 4 13 23 33 32 27 34 12 1894 12 8 6 14 14 8 18 39 28 31 21 23 1895 11 11 6 9 7 3 24 12 27 31 19 13 1896 7 11 4 11 11 13 19 23 29 28 22 10 1897 7 8 6 6 6 8 13 36 36 27 19 17 Average 12.7 8.9 7.9 9.3 92 9.8 16.3 27.9 29.0 27.1 21.4 155 Ratio of 100 6.6 4.6 ; 4.1 4.8 4.8 5.1 8.4 14.4 15.0 14.0 11.1 8.0 Mean Monthly Tempeeatoee. From " Monthly Weather Review," V. S. Weather Bureau. Tear. J. F. u. A H. J. J. A. a. 0. N. D. 1888 29 35 37 53 63 73 74 75 64 51 47 36 1889 39 31 43 55 66 71 77 74 66 54 48 46 1890 44 43 42 54 64 75 75 74 68 57 48 35 1891 38 41 39 56 62 71 72 74 71 55 44 44 1892 32 37 37 52 63 76 76 76 66 56 44 33 1893 25 34 40 53 61 72 77 75 67 57 44 39 1894 37 34 48 52 65 73 78 73 71 57 43 38 1895 31 26 41 53 62 74 73 77 72 53 47 39 1896 34 36 38 57 69 71 78 76 68 55 61 36 1897 32 37 45 53 63 70 77 74 69 58 46 39 Average 34 35 41 54 64 73 76 75 68 55 46 38 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVER. 547 LONDON. Weekly Typhoid Deaths and Average Mean Temperatdke. From the Weekly Returns of the Registrar-General. 1. 8. 3. 4. 6. 6. 7. 8. 9. 10. 11. 18. 13. Deaths . . . Temperature 13 38 14 38 13 38 11 39 12 40 7 39 10 40 9 40 9 41 9 41 9 42 7 42 8 45 14. 15. 16. 17. 18. 19. 80. 81. 88. 83. 84. 86. 86. Deaths . . . Temperature 9 46 6 46 8 48 7 48 8 50 7 52 9 54 8 56 8 57 7 58 9 59 9 60 9 61 87. 88. 89. 30. 31. 38. 33. 34. 35. 36. 37. 38. 39. Deaths . . . Temperature 8 62 10 63 7 63 9 62 10 62 9 63 13 62 12 61 19 60 15 59 16 58 17 56 17 55 40. 41. 48. 43. 44. 45. 46. 47. 48. 49. 50. 61. 58. Deaths . . . Temperature 17 53 19 51 19 49 18 47 20 47 19 45 19 42 20 41 17 41 16 41 19 40 15 39 16 38 Weekly typhoid rate is average for ten years, 1888-1897. Temperature is average for years, 1840-1890. Average Weekly Typhoid Deaths for each Month. J. F. M. A. M. J. J. A. B. 0. N. D. Deaths Ratio of 100 ... Temperature . . . 13.0 8.9 38.0 9.0 6.2 40.0 8.0 5.5 42.0 7.0 4.8 47.0 8.0 5.5 53.0 8.0 5.5 59.0 8.0 5.5 62.0 16.0 11.0 62.0 16.0 11.0 57.0 18.0 12.3 50.0 19.0 13.0 43.0 16.0 11.0 40.0 LEIPSIC. Monthly Typhoid Deaths. From " Veroffentlichungen des Kaiserlichen Gesundheitsamtes." J. F. M. A. M. J. J. A. B. 0. N. D. 1888 1 2 1 1 1 3 2 1 1 4 2 1889 2 2 2 2 1 4 6 6 6 3 2 1890 6 1 1 1 5 2 7 6 4 3 6 1891 5 5 4 6 5 1 6 5 6 4 3 4 1892 3 1 1 1 4 3 4 7 3 2 1893 2 2 3 4 1 6 2 1 6 1894 1 2 2 1 5 5 4 3 2 4 5 4 1895 3 1 1 2 2 3 8 5 6 2 1896 2 3 2 5 1 1 1 3 2 2 3 7 1897 3 5 3 1 2 1 2 5 8 3 4 Average 2.2 2.8 1.7 1.7 1.9 2.2 2.9 3.7 4.9 3.7 3.5 3.5 Ratio of 100 6.3 8.1 4.9 4.9 5.5 6.3 8.4 10.7 14.1 10.7 10.1 10.1 Mean Monthly Temperature. 1864-1890. From "Amtliche Publication des Konigl. sachsischen meteorologischen Institutes. Das Klima des Eonigreiches Saclisen." Heft III, 1895. J. F. M. A. M. J. J. A. s. 0. N. D. Centigrade Fahrenheit -1 30 32 3 37 8 46 13 "55 17 63 18 64 17 63 14 57 8 46 3 37 32 548 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVER, BERLIN. Monthly Typhoid Deaths. From " Veroffentlichnngen dea Kaiserlichen Gesundheitsamtes." Tear. J. F. M. A. M. J. J. A. s. 0. N. D. 1888 38 19 10 11 8 10 18 22 13 15 11 13 1889 11 21 58 23 14 11 28 20 23 18 36 27 1890 14 15 11 9 10 8 10 16 18 18 9 5 1891 9 7 16 7 9 9 7 20 19 31 20 12 1892 12 6 15 7 10 10 7 9 23 15 10 13 1893 7 6 11 8 13 8 7 19 42 16 18 5 1894 7 9 8 7 7 5 7 5 10 10 5 12 1895 6 7 8 2 4 14 8 16 22 17 8 14 1896 9 6 6 11 8 6 11 14 17 11 4 5 1897 3 1 8 8 5 4 4 20 11 10 7 9 Averse 11.6 9.7 15.1 9.3 8.8 8.5 10.7 16.1 19.8 16.1 12.8 11.5 Ratio of 100 8.0 6.7 10.0 6.0 6.0 5.3 7.3 10.7 13.3 10.7 8.7 7.3 Mean Monthly Temperature. From " Ergebnisse der meteorologischen Beobachtungen yon dem KonigUch. Preiissischen meteorologischen Instltut" Year. J. F. M. A. M. J. J. A. s. 0. N. D. 1888 —1 -2 7 14 17 17 17 15 8 4 2 1889 —2 —1 1 9 19 22 18 17 13 9 4 1890 3 —1 6 9 16 16 18 19 15 9 4 -4 1891 -3 1 4 6 15 16 18 17 16 11 4 3 1892 -1 1 2 8 13 17 18 20 16 9 2 —1 1893 -7 2 5 9 13 17 19 18 13 11 3 1 1894 -1 3 6 11 13 16 20 17 12 9 5 1 Average -2 3 8 15 17 18 18 14 9 4 Fahrenheit 28 32 37 46 59 63 64 64 57 48 39 32 GWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVER. 549 EMPIRE OF JAPAN. Monthly Typhoid Deaths. From Annual Beports of the Central Sanitary Bureau of Japan. Year. J. F. M. A. M. J. J. A. s. 0. N. D. 1890 568 386 380 402 540 527 603 838 1159 1309 977 775 1891 556 285 264 392 724 1038 1028 940 1255 1286 1009 837 1892 541 382 366 405 468 628 734 938 1165 1252 921 729 1893 508 361 368 340 450 520 646 827 1190 1262 1016 695 1894 515 319 226 256 338 515 681 1068 1298 1141 995 702 Average 538 347 321 359 504 646 738 922 1203 1250 984 748 Ratio of 100 6.3 4.1 3.8 4.2 5.9 7.5 8.6 10.8 14.1 14.6 11.5 8.8 Mean Monthly Tempebatuke. (10 stations.) (3-6 years.) From " The Climate of Japan," Central Meteorological Observatory, Tokio, 1893. Stations. J F. M. A. M. J. J. A. s. 0. ». D. Kumamoto Matsuyama Hiroshima . Ozaka . . Wakayama Nagano Tokio . . Hakodate . Sapporo . Nemuro 3 4 3 4 5 -2 3 -4 -7 -6 7 6 5 5 5 4 -2 -5 -5 10 8 8 9 9 4 7 3 -1 16 13 13 14 14 11 13 7 5 4 19 17 19 18 18 14 16 11 11 7 22 21 22 22 22 19 21 14 15 10 26 25 25 26 26 23 24 18 19 15 27 26 27 27 27 24 26 21 21 18 25 23 23 24 23 20 22 18 17 16 18 17 17 17 17 12 16 11 9 10 12 12 11 12 12 7 11 5 3 4 8 9 7 7 8 4 6 1 -1 Average . Fahrenheit ' 32 2 36 6 43 11 52 15 59 19 66 23 74 24 75 21 70 14 58 9 48 5 41 550 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVEE. SAN FRANCISCO. Monthly Typhoid Deaths. From Reports, Local Department of Health. Tear. J. F. M. A. H. J. J A. 8. 0. ». D. 1888 12 10 18 13 15 12 1889 a 10 8 13 12 9 1890 17 6 7 6 4 17 17 13 11 21 14 10 1891 13 6 10 5 9 8 18 16 7 8 11 12 1892 8 6 8 4 4 1 13 14 5 13 11 7 1893 4 5 3 4 3 12 10 11 10 9 16 10 1894 11 7 5 5 9 6 8 13 12 9 10 20 1895 14 11 4 6 5 11 16 5 12 8 7 9 1896 10 6 6 5 7 10 8 7 10 7 9 1897 13 2 7 5 3 4 3 4 — 5 4 4 Average 10.7 6.7 6.4 5.9 6.2 8.7 10.8 10.4 10.2 10.7 10.6 10.3 Ratio of 100 9.9 6.1 6.0 5.5 5.8 8.1 10.0 9.7 9.4 9.9 9.8 9.6 Mean Monthly Temperature. From " Monthly Weather Beview," U. S. Weather Bareau. -Tear. J. F. u. A. M. J. J. A. s. 0. N. D. 1888 46 53 52 56 55 61 59 58 59 59 55 52 1889 50 54 57 59 59 60 59 60 65 62 59 51 1890 46 49 54 55 60 59 60 61 60 62 59 50 1891 52 51 55 53 56 60 59 62 62 60 59 50 1892 52 52 54 53 58 57 58 59 60 60 57 51 1893 47 50 51 52 56 56 57 57 59 58 56 52 1894 48 48 51 55 55 56 56 59 63 60 59 50 1895 49 54 52 55 58 59 58 58 61 59 56 49 1896 52 55 54 52 56 57 59 59 60 59 53 53 1897 49 51 49 57 57 59 58 58 61 58 53 51 Average 49 52 53 55 56 58 58 59 61 60 57 51 3EDGWICK AND WINSLOW. BACILLUS OF TTrHOID FEVER. 551 CINCrNNATI. MoNTHLX Ttphoid Deaths. From Reports, Local Department of Health. Tear. J. F. M. A. M. J. J. A. s. 0. N. D. 1888 41 34 16 11 6 7 6 12 17 16 22 15 1889 11 14 11 19 7 9 12 14 14 11 12 9 1890 18 11 17 9 14 14 23 24 20 23 28 9 1891 10 17 14 21 14 21 10 16 7 22 22 12 1892 17 10 8 4 4 ■ 7 6 10 12 9 11 23 1893 10 14 8 4 14 6 8 15 14 12 12 17 1894 18 11 15 10 10 8 12 6 10 21 11 37 1895 22 12 7 6 5 5 7 7 8 10 8 23 1896 34 22 15 11 11 5 6 14 9 11 U 15 1897 9 8 5 5 10 3 17 9 9 9 6 11 Average 19.0 15.3 12.6 10.0 9.5 8.5 10.7 12.7 12.0 14.4 13.8 17.1 Ratio of 100 12.3 9.9 8.2 6.5 6.2 5.5 6.9 8.2 7.8 9.4 8.9 11.1 Mean Monthly Temperature. From " Monthly Weather Review," V. S. Weather Bureau. Tear. J. F. M. A. M. J. J. A. 8. 0. N. D. 1888 29 35 39 55 63 74 76 73 63 50 45 36 1889 37 . 30 46 54 63 70 75 72 66 52 43 48 1890 41 43 40 66 64 78 77 73 66 56 48 36 1891 36 40 38 56 60 74 71 72 70 55 43 42 1892 26 39 38 53 62 75 76 75 68 56 40 32 1893 21 34 42 54 61 73 79 75 70 56 42 36 1894 38 33 49 54 63 75 77 77 72 57 41 37 1895 27 24 41 55 64 76 75 77 73 51 44 37 1896 34 35 37 62 71 73 76 75 65 53 48 38 1897 29 36 46 52 59 72 78 74 71 63 46 36 Average 32 35 42 55 63 74 76 74 68 55 44 38 552 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVEK. DISTRICT OF COLUMBIA. Monthly Typhoid Deaths. From Reports, Local Department of Health. Year. J. F. M. A. M. J. J. A. s. 0. N. D. 1887 18 32 22 20 18 15 1888 8 7 8 7 3 10 12 23 27 34 19 7 1889 14 7 9 5 6 7 23 18 29 15 18 29 1890 9 6 19 11 10 21 33 26 29 30 21 17 1891 12 6 12 9 5 8 6 22 21 36 .26 12 1892 13 13 8 7 8 11 19 21 30 22 25 18 1893 6 7 6 11 11 10 21 24 28 23 23 21 1894 10 5 5 6 5 20 33 30 26 30 24 16 1895 3 8 1 1 1 1 12 27 56 55 24 20 1896 9 8 3 3 4 7 8 15 25 25 18 16 1897 13 4 4 4 6 9 Average 9.7 7.1 7.5 6.4 5.9 10.4 18.5 23.8 29.3 29.0 21.6 17.1 Ratio of 100 5.2 3.8 4.1 3.5 3.2 5.6 10.0 12.9 15.8 15.7 11.7 9.2 Mean Monthly Tempekatuee. From Keports, Local Department of Health. Year. J. F. M. A. M. J. J. A. s. N. D. 1887 80.5 73.2 65.0 55.4 44.9 37.2 1888 29.2 35.7 37.5 52.9 62.7 73.0 72.9 73.9 63.2 50.5 45.8 35.2 1889 36.8 29.4 42.3 53.2 63.8 69.8 74.2 70.6 65.6 52.5 46.2 45.6 1890 44.2 43.4 41.4 53.7 63.8 74.9 75.1 73.5 67.7' 56.2 47.8 34.2 1891 37.3 41.5 38.5 55.4 61.3 71.4 72.0 74.5 79.2 54.4 43.9 43.1 1892 31.7 36.9 37.7 51.5 63.8 76.2 75.7 76.2 66.2 55.0 43.6 33.0 1893 24.0 34.9 41.0 54.0 61.6 72.0 77.0 74.7 66.0 56.4 43.6 38.4 1894 37.7 35.2 48.6 53.2 64.8 73.7 78.0 73.9 71.4 57.8 43.8 37.4 1895 31.6 26.2 41.8 53.8 62.6 74.6 72.7 77.3 72.4 52.1 46.4 38.7 1896 33.3 36.6 38.6 66.5 68.8 71.3 76.6 75.7 67.7 54.0 50.6 35.5 1897 30.9 36.5 46.0 53.0 62.5 69.7 Average 34 36 41 55 64 73 75 74 68 54 46 38 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVEK. 553 MOBILE. Monthly Typhoid Deaths. Obtained, in correspondence, by courtesy of Local Department of Health. Tear. J. p. M. A. M. J. J. A. B. 0. N. D. 1889 2 2 2 2 3 1 1 1 2 1890 2 1 1 2 6 2 1 1 1891 2 3 4 3 2 1892 1 1 1 4 3 1 2 1 1893 1 1 1 1 4 3 1 2 2 1894 1 2 1 2 4 1 1 1 1 1 1895 3 2 1 2 3 4 4 1 2 1896 1 2 1 5 1 2 3 1 1897 1 1 1 3 4 2 5 1 2 1898 1 2 1 1 2 6 4 2 2 1 1 Average .8 .7 .5 .5 1.1 1.8 3.9 2.0 2.0 1.9 1.1 .9 Ratio of 100 4.6 4.1 2.9 2.9 6.4 10.4 22.6 11.6 11.6 11.0 6.4 0.2 Mean Monthly Temperature. From " Monthly Weather Review," U. S. Weather Bureau. Tear. J. F. M. A. H. J. J. A. s. 0. N. D. 1889 51 51 59 68 70 77 81 79 77 66 56 61 1890 62 61 57 68 73 80 80 80 77 67 61 54 1891 49 59 59 66 72 80 80 80 77 65 57 53 1892 47 57 55 66 72 79 79 80 75 69 58 52 1893 46 58 57 69 74 79 82 81 78 66 58 55 1894 55 53 60 69 74 78 79 80 78 68 57 54 1895 49 43 58 66 72 79 81 81 81 65 58 50 1896 49 53 57 69 76 79 81 82 77 68 62 51 1897 48 55 66 66 71 81 82 80 78 71 60 54 1898 55 53 63 62 75 80 81 80 78 65 56 49 Average 51 54 59 67 73 79 81 80 78 67 58 53 554 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVER OAKLAND. Monthly Typhoid Deaths. Obtained, in correspondence, by courtesy of Local Department of Health. Year. J. p. H. A. M. J. J. A. s. 0. N. D. 1889 1 1 4 2 2 1 1 3 3 1 1890 2 5 1 1 2 1 2 3 2 3 1891 2 2 1 3 4 6 2 3 3 1892 2 1 2 3 1 1 2 5 1 2 1893 2 1 4 22 4 7 2 3 1 1894 1 2 3 1 1 2 2 1 1 1895 2 3 3 2 3 1 2 1 1 1896 1 3 1 1 2 2 3 3 2 1897 1 1 1 1 2 1 1 1 1898 2 1 3 2 1 1 1 1 Average. 0.8 1.4 1.4 1.1 1.4 1.1 3.7 1.7 2.3 2.1 1.8 1.6 Ratio of 100 3.9 6.9 6.9 5.4 6.9 5.4 18.1 8.3 11.3 10.3 8.8 7.8 Mean Monthly Temperature. From " Monthly Weather Review," V. S. Weather Bnreaa. Yeai J. F. M. A. M. J. J. A. 8. o N. D. 1889 48 50 57 59 59 61 60 61 63 61 57 50 1890 44 48 54 55 60 59 62 62 61 62 57 49 1891 51 49 63 63 55 60 61 63 62 69 57 49 1892 52 50 53 53 58 62 64 64 63 58 63 49 1893 49 51 54 56 58 62 62 61 62 58 54 61 1894 45 48 52 57 59 61 59 61 62 59 66 49 1895 47 52 51 56 59 60 63 59 62 56 64 47 1896 51 53 55 54 58 61 64 63 — 58 61 49 1897 46 49 49 59 61 64 63 61 63 58 51 47 1898 44 51 51 57 57 64 62 62 61 60 53 47 Average 48 50 53 56 58 61 62 62 62 59 54 49 QDGWIGK AND WINSLOW. — BACILLUS OF TYPHOID FEVEK. 655 DRESDEN. Monthly Typhoid Deaths. From " Veroflentlichungen des Kaiserlichen GesundheitsamteB.' Year. J. F. u. A. M. J. J. A. s. 0, N. D. 3 1888 4 2 2 1 1 6 4 1 2 1889 4 2 1 3 1 2 4 1 2 1 1890 1 3 4 1 1 1 2 1 3 3 2 1891 3 1 3 1 2 2 2 3 3 5 2 1892 4 1 1 2 1 1 3 1 2 1893 1 1 3 1 1 3 2 1894 1 8 3 2 1 3 5 2 1895 1 1 2 1 4 3 1 2 1 1896 4 2 2 1 1 1 3 1897 1 1 1 3 2 2 1 Average 1.4 .9 1.6 2.0 1.6 1.0 1.6 2.3 1.9 1.4 2.2 1.6 Ratio of 100 7.2 4.6 8.2 10.3 8.2 5.1 8.2 11.8 9.7 7.2 11.3 8.2 Mean Monthly Temperature. Average 1864-1890. From "Amtliche Publication des Konigl. sachsischen meteorologischen Institutes. Das Elima des EonigreicheB Sachsen." Heft III, 1896. J. F. M. A. M. J. J. A. B. 0. N. D. Centigrade Fahrenheit 32 1 34 3 37 8 46 13 55 16 61 18 64 17 63 14 57 9 48 4 39 32 556 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVEK. MUNICH. Monthly Typhoid Deaths. From " Veroffentlichungen des Kaiseriichen GesundheitBamteB.'' Year. J. p. u. A. M. J. J. A. B. 0. N. D. 1888 4 1 5 3 3 4 3 2 2 2 2 1889 3 2 2 2 3 2 6 1 1 6 2 1 1890 2 1 3 2 2 2 4 2 5 4 1 1891 2 3 3 3 2 3 1 1 1 3 2 1892 2 1 1 3 1 2 1 1893 3 3 1 1 20 15 . 9 1 3 1 1894 1 2 1 3 2 ■ 0- 1 1895 1 2 1 1 2 3 1 4 1896 2 2 1 1 1 3 2 1 1 1897 2 1 7 7 5 1 Average 1.9 1.5 1.5 1.6 1.4 4.1 3.8 2.6 1.5 2.2 1.1 1.2 Ratio of 100 7.8 6.1 6.1 6.6 5.7 16.7 15.6 10.7 6.1 9.0 4.5 4.9 Mean Monthly Temperatdke. From " Beobaclitungen der meteorologischen Stationen im Eonigreich Bayem." Year. J. F. M. A. M. J. J. A. s. 0. N. D. 1889 —4 -3 -1 7 15 18 17 16 11 8 1 -4 1890 1 -5 3 7 14 14 16 17 12 6 2 -7 1891 -6 -3 3 5 13 16 17 15 13 9 1 -7 1892 -2 1 1 7 13 16 16 19 14 7 3 -3 1893 -9 2 4 9 12 16 18 17 13 9 1 -3 1894 -5 1 4 10 11 14 18 16 11 8 3 —1 1895 -5 -8 1 8 11 15 18 17 16 7 5 Average -4 -2 2 8 13 16 17 •17 13 8 2 -3 Fahrenheit 25 28 36 46 55 61 63 63 55 46 36 27 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVER. 557 VIENNA, Monthly Typhoid Deaths. From " Veroffentlichungen des Kaiserlichen Gesundheitsamtes." Yteir. J. F. u A. M. J. J. A. s. 0. N. D. 1888 7 7 12 8 7 9 4 4 5 9 5 26 1889 18 14 9 9 12 5 5 5 5 9 2 8 1890 6 7 7 7 6 6 4 6 11 7 3 7 1894 7 5 8 5 8 10 3 12 2 5 4 5 1895 5 3 2 2 5 6 13 12 6 11 14 7 Average 8.6 7.2 7.6 6.2 7.6 7.2 5.8 7.8 5.8 8.2 5.6 10.6 Ratio of 100 9.8 8.2 8.6 7.0 8.6 8.2 6.6 8.8 6.6 9.3 6.3 12.0 Mean Monthly Tempekatuee. From " Jahrbiicher der k. k. Central-Anstalt fiir Meteorologie und Erdmagnetismus." Tear. J. F. M. A. u. 3. J. ■ A. s. 0. N. D. 1888 1889 1890 1891 1892 1893 1894 -3 -2 1 -6 —1 -8 -3 -3 -1 -2 —2 1 2 2 4 1 6 4 2 6 8 8 9 9 7 10 10 15 15 18 16 16 14 14 17 18 20 16 17 17 17 18 18 19 19 18 19 19 23 18 18 21 17 21 19 20 15 12 14 16 16 15 16 8 11 9 12 9 11 12 2 3 4 3 2 3 5 e -4 -5 1 -2 1 1 Average Fahrenheit -3 27 32 4 39 10 50 16 61 18 64 19 66 19 66 15 59 10 50 3 37 -1 30 558 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVEE. CHICAGO. Monthly Typhoid Deaths. From Reports, Local Department of Health. Tear. J. F. M, A. M. J. J A. s. 0. N. D. 1889 30 21 15 12 16 18 29 64 77 68 68 35 1890 53 136 103 45 82 107 86 115 95 72 67 47 1891 67 61 71 136 408 167 200 182 198 171 150 186 1892 311 187 76 56 70 55 211 179 138 92 67 47 1893 41 30 41 58 56 CO 55 76 86 81 43 43 1894 46 26 27 30 31 31 37 62 71 68 38 34 1895 30 21 26 30 30 18 36 59 76 90 60 42 1896 87 89 65 33 31 44 58 64 87 89 60" 44 1897 38 46 41 19 13 23 27 42 48 61 44 35 1898 29 32 41 94 67 35 55 45 65 62 56 55 Average 75 59 51 51 80 56 79 88 94 85 65 57 Ratio of 100 8.8 7.0 6.0 6.0 9.5 6.7 9.4 10.5 11.2 10.1 7.7 6.8 Mean Monthly Tempekatuke. From " Monthly Weather Review," U. S. Weather Bureau. Year. J. F. H. A. M. J. J. A. s. 0. s. D. 1888 15 23 30 45 53 67 72 69 60 48 41 31 1889 29 20 38 47 57 62 70 71 63 49 39 41 1890 31 32 29 46 53 70 72 68 60 51 42 31 1891 30 29 31 47 53 66 67 69 69 53 34 36 1892 19 30 31 44 52 64 72 71 64 54 35 23 1893 12 21 33 44 52 68 74 70 64 53 36 25 1894 27 23 41 47 56 71 73 71 66 62 34 32 1895 18 17 32 46 59 70 70 72 69 46 36 30 1896 27 27 31 53 65 67 72 73 61 50 38 33 1897 22 29 35 46 55 65 74 69 69 58 39 25 Average 23 25 33 46 55 67 72 70 64 51 37 32 SEDGWICK AND WINSLOW. BACILLUS OF TYPHOID FEVER. 559 PHILADELPHIA. Monthly Typhoid Deaths. From Reports, Local Department of Health. Tear. J. F. H. A. H. J. J. A. s. - 0. N. D. 1888 63 46 40 37 84 49 62 169 100 67 36 32 1889 62 79 61 41 64 50 68 83 70 63 33 66 1890 126 54 52 52 51 36 56 62 57 47 39 34 1891 50 44 102 141 76 42 49 42 53 35 23 26 1892 51 68 51 37 30 24 20 40 44 37 11 27 1893 43 34 38 35 61 37 26 47 47 29 25 35 1894 43 18 20 25 36 24 29 50 34 31 29 31 1895 36 64 48 40 39 38 33 36 32 43 30 30 1896 34 23 21 40 46 27 31 38 34 17 28 63 1897 36 18 27 41 50 32 25 49 24 20 31 48 Average 54 45 46 49 54 36 40 62 49 39 28 39 Ratio of 100 10.0 8.2 8.4 9.0 10.0 6.7 7.4 11.5 9.0 7.2 5.2 7.2 Mean Monthly Tempebatukb. From " Monthly Weather KeTiew," U. S. Weather Bureau. Year. J. F. mI A. H. i. J. A. b: 0. N. D. 1888 28 34 35 51 61 73 72 74 ; 64 50 46 36 1889 39 29 42 53 65 71 75 73 66 53 47 44 1890 42 41 39 52 63 74 75 74 67 55 46 32 1891 36 40 38 54 61 71 72 74 72 55 44 43 1892 31 35 36 51 62 74 77 76 67 56 44 33 1893 24 32 39 51 61 72 77 76 66 58 44 36 1894 37 32 47 51 64 73 78 73 70 57 42 37 1895 31 25 38 52 62 74 73 77 72 53 47 39 1896 31 34 36 55 67 70 78 77 68 54 50 35 1897 31 36 43 53 63 69 76 74 68 58 46 38 Average 33 34 39 52 63 72 75 75 68 55 46 37 560 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVER. NEWARk. Monthly Typhoid Cases. From Report of Local Department of Health for 1899. Year. J. F.' M. A. M. J. J. A. s. 0. N. D. 1890 93 23 21 17 16 7 20 10 22 27 34 57 1891 88 42 43 18 18 11 15 167 207 137 92 38 1892 36 27 19 11 4 4 16 32 30 17 16 17 1893 5 3 9 6 8 10 11 . 26 12 21 7 7 1894 2 4 6 9 6 3 3 10 13 21 6 5 1895 2 3 2 1 6 4 4 31 38 21 21 15 1896 10 5 3 2 3 6 4 14 25 29 7 8 1897 5 5 11 7 5 2 8 7 14 ■ 11 13 15 1898 5 3 2 3 3 7 6 38 59 29 16 8 1899 2 2 301 . 67 27 9 19 28 30 12 10 8 Average 24.8 11.7 41.7 14.1 9.6 6.3 10.6 36.3 45.0 32.5 22.2 17.8 Ratio of 100 9,2 4.3 15.4 5.2 3.6 2.3 3.9 13.4 16.7 12.0 8.2 6.6 Mean Monthly Temperatdee. From " Monthly Weather Review," U. S. Weather Bureau. _. . Tear. J. ' T. Ml A. M. J. J. A. s. a N. D. 1890 39 38 36 49 60 71 73 72 65 54 43 30 1891: 33 36 36 51 59 69 70 72 69 53 43 41 : 1892 30 34 34 49 59 72 74 73 64 54 41 30 1893 22 28 35 47 59 68 74 73 . 62 55 41 34 1894 33 28 43 49 60 70 76 71 67 54 40 35 .1895 29 25 36 48_ 61 71 71 74 70 50 45 37 1896 29 31 33 53 66 69 76 75 66 53 49 32 1897 30 33 40 50 62 67 75 72 66 55 44 35 1898 33 33 45 48 58 71 76 76 70 56 43 32 1899 29 25 36 49 61 72 74, 72 64 56 43 34 Avepge I 31 31 37 49 60 70 74 73 66 54 43 34 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID KKVi-K. 561 PARIS, M Monthly Typhoid Deaths. From " Annuaire statistique de la ville de Paris." Year. J. p. M. A. M. J. J. A. s. 0. N. D. 1888 146 78 52 58 54 52 81 51 70 65 69 71 1889 69 62 57 43 53 71 102 153 120 92 " 84 208 1890 74 39 45 47 51 57 44 54 76 92 71 73 1891 65 59 53 47 36 30 37 43 40 39 54 46 1892 50 36 48 37 48 78 90 89 97 105 62 69 1893 48 49 50 47 29 29 63 73 72 48 33 29 1894 25 53 289 84 34 46 33 37 21 22 29 24 1895 11 9 13 21 13 25 22 30 43 34 24 26 1896 35 17 21 10 25 9 30 35 26 17 28 9 Average 52 40 63 39 34 40 50 56 56 51 45 64 Katio of 100 9.0 6.9 10.9 6.7 5.9 6.9 8.6 9.7 9.7 8.8 7.7 9.3 Mean Monthly Temperatchb. From " Annuaire statistique de la Tille de Paris." Year. J. F. M. A. M. J. J. A. s. 0. N. D. 1888 1 4 7 13 16 16 16 15 8 8 3 1889 1 2 4 9 15 19 18 17 14 10 6 1890 6 2 6 9 14 15 16 17 15 9 6 -.3 1891 —1 3 6 8 12 16 17 16 15 12 ; 5 5 1892 2 4 . 4 10 15 17 18 19 15 9 ; 8 1 1893 —1 6 9 14 14 18 19 20 1.0 11 i 5 3 1894 3 5 8 12 12 16 18 17 14 10 ' 7 4 1895 4 5 11 14 16 18 18 19 9 r 9 5 1896 2 3 9 9 13 17 19 16 15 9 3 4 Average 1 2 5 9 13 16 17 17 15 9 6 2 Fahrenheit 34 36 41 48 55 61 63 63 59 48 43 36 86 562 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVER. NEW ORLEANS. Monthly Typhoid Deaths. From Reports, Local Department of Health. Year. J. F. M. ^ M. J. J. A. s. 0. N. D. 1886 5 3 2 3 i 3 3 3 1 3 1887 2 4 1 8 4 1 4 2 4 2 7 1890 7 6 3 2 3 6 7 4 2 3 1 6 1891 5 1 1 1 3 6 7 6 10 2 4 13 1892 4 1 1 2 2 6 3 10 10 2 5 5 1893 2 2 5 1 1 6 4 1 4 4 4 5 1896 7 2 7 8 4 12 9 14 8 4 4 11 1897 10 4 3 7 6 16 21 18 10 11 19 16 Average 5.2 2.9 2.5 2.7 3.0 7.4 7.0 , -7.5 6.1 4.1 5.0 8.2 Ratio of 100 8.5 4.7 4.0 4.5 4.9 11.9 11.3 12.2 9.9 6.7 8.1 13.4 Mean Monthly Temperattjke. From " Monthly Weather Review," XJ. S. Weather Bureau. Year. J. p. M. A. -M. J. J. A. s. 0. N. D. 1888 56 59 60 70 73 77 81 78 75 68 59 51 1889 53 53 61 70 74 78 83 81 79 70 59 64 1890 65 64 62 70 74 81 82 81 78 69 64 56 1891 53 63 61 68 74 81 81 81 78 68 60 56 1892 49 61 59 69 74 79 80 82 77 71 62 56 1893 50 61 61 72 76 80 83 82 80 69 60 58 1894 58 55 63 71 75 78 79 80 80 71 60 58 1895 52 45 62 68 74 80 82 82 82 69 60 54 1896 52 56 61 71 78 80 83 83 79 70 65 55 1897 51 58 69 68 74 82 84 82 79 74 64 57 Average 54 57 62 70 75 80 82 81 79 70 61 , 56 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVKK. 563 ATLANTA. Monthly Typhoid Deaths. Obtained, in correspondence, by courtesy of Local Board of Health. Tear. J. p. M. A. M. J. J. A. s. 0. N. D 1893 1 1 3 3 4 5 11 13 7 9 5 1 1894 1 1 3 6 11 12 7 6 2 1 • 1895 3 1 3 4 12 14 20 6 5 1896 3 2 4 2 3 7 13 8 10 8 5 3 1897 1 1 10 10 11 9 6 4 3 1898 4 3 1 4 4 6 5 8 8 7 5 2 Average 1.5 1.0 2.0 1.8 2.5 6.0 9.0 10.7 9.2 9.3 4.5 2.r> Ratio of 100 2.5 1.7 3.3 3.0 4.2 10.0 15.0 17.8 15.3 15.5 7.5 4.2 Monthly Temperatcee. From " Monthly Weather Review," U. S. Weather Bureau. Tear. J. F. M. A. M. J. J. A. s. 0. N. D. 1893 36 46 51 64 67 74 81 77 73 62 51 47 1894 47 45 57 62 69 76 76 76 73 62 49 46 1895 40 34 51 60 67 77 77 77 76 60 52 44 1896 42 45 49 66 75 75 78 80 75 61 56 44 1897 39 48 55 60 68 79 78 76 74 66 53 45 1898 47 43 57 56 73 79 78 77 74 60 49 44 Average 42 43 53 61 70 77 78 77 74 62 52 45 564 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVEK. CHARLESTON. Monthly Typhoid Deaths. From Reports, Local Department of Health. Year. J. F. M. A. M. J. J. A. B. 0. N. D. 1888 3 3 1 1 2 2 7 4 5 4 4 1889 3 2 2 4 1 4 3 5 3 5 3 5 1890 4 6 3 2 2 6 6 8 4 9 2 4 1891 5 2 1 1 6 3 3 5 2 1892 5 1 2 1 4 3 3 3 3 1 1 1898 1 4 2 2 1 4 2 4 3 1 1894 1 2 2 2 1 4 1 2 4 4 2 1895 1 2 1 2 2 10 3 2 5 3 2 1896 3 5 3 3 2 6 4 5 4 3 1 5 1897 1 2 2 4 3 5 5 7 1 3 7 Avei'age 2.7 2.7 2.0 1.7 1.5 2.9 4.4 4.3 3.8 4.3 2.2 2.8 Ratio of 100 7.6 7.6 5.7 4.8 4.2 8.2 12.5 12.2 10.8 12.2 6.2 7.9 Mean Monthly Temperature. From " Montlily Weather Review," U. S. Weather Bntean. Tear. J F. M. A. M. J. J. a. 8. 0. N. D. 1888 51 54 55 66 72 78 . 78 80 74 64 56 47 1889 52 47 55 63 74 77 81 78 76 65 60 60 1890 59 61 56 65 73 82 80 80 76 68 62 51 1891 50 58 55 65 70 80 80 81 76 64 56 55 1892 48 53 55 64 72 78 80 81 75 66 57 52 1893 43 56 56 68 72 78 83 79 78 68 58 54 1894 53 53 61 65 72 77 79 80 78 68 57 52 1895 49 41 56 64 70 79 81 82 78 66 58 51 1896 48 52 55 66 77 79 82 81 77 67 63 49 1897 47 55 61 66 72 80 82 81 75 70 62 54 Average 50 53 57 65 72 79 81 80 76 67 59 53 SEDGWICK AND WINSLOW. BACILLUS OP TTPHOID FEVKB. 565 EMPIRE OF INDIA. Monthly Typhoid Admissions, British Troops in India. From Report on Sanitary Measures in India in 1896-97. Vol. XXX. Period. J V. M. A. M. J. J. A. s. 0. N. D. 1886-95 1896 518 65 418 75 689 202 1427 214 1795 160 1365 152 1441 175 1718 214 1400 179 923 90 745 92 879 177 Total 583 493 891 1641 1955 1517 1616 1932 1579 1013 837 1056 Average Ratio of 100 53 3.9 45 3.3 81 5.9 149 10.9 178 13.0 138 10.1 147 10.7 175 12.8 144 10.5 92 6.7 76 5.5 96 7.0 Monthly Range of Temperature. From " Handbuch der Klimatologie," J. Hann. Zweite Auflage. Stuttgart, 1897. Difference between the monthly mean and the yearly mean. Central India, Deccan, 20.8° N., 78.0° E., 390 M. J. F. M. A. M. J. J. A. s. 0. N. D. -6 -3 2 6 8 3 -4 -7 Punjab , 31.1° N , 72.3° E , 200 M. J. F. M. A. . M. J. J. A. 8. 0. N. D. -12 -10 -3 3 7 10 9 7 6 —7 -11 566 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVER. SANTIAGO DE CHILE. Typhoid cases received at Hospital S. Francisco de Borja and Hospital S. Juan de Dios, 1886-1895. Figures from essay, " La Fiebre Tifoidea en Santiago," by Pedro V. Garcia, P., " Revista Cliilena de Hijiene." Torao III, NiSni. 11. J. F. M. A. M. J. J. A. s. 0. N. D. Total Ratio of 100 121 13.0 121 13.0 102 11.0 87 9.4 65 7.0 49 5.3 52 5.6 47 5.1 49 5.3 60 6.5 '■ 73 ' 7.8 107 11.5 Mean Monthly Temperature. From " Observaciones meteoroldjicas lieclias en el Observatorio Astrondmico de Santiago." Tear. J. p. M. A. M. J. J. A. s. 0. N. D. 1882 20.7 18.9 16.4 12.6 10.3 8.0 7.4 9.5 12.4 15.2 16.6 18.4 1883 19.1 18.9 15.3 12.8 9.9 7.5 6.8 8.9 10.8 13.3 16.2 18.9 1884 21.7 18.2 15.3 13.3 9.0 7.0 6.4 10.3 10.9 13.2 16.4 19.0 1885 18.7 18.3 16.4 10.3 8.8 7.5 6.4 9.4 12.6 13.5 18.0 17.4 1886 19.9 18.1 16.5 13.4 10.2 6.2 8.1 8.7 11.5 14.4 16.5 19.4 1887 19.8 18.4 16.4 13.1 9.7 8.5 8.6 10.5 11.7 13.4 16.0 18.1 Average 20 18 16 13 10 7 7 9 12 14 16 18 Fahrenheit 68 64 61 55 50 45 45 48 54 57 61 64 BUENOS AYRES. Monthly Typhoid Deaths, 1876-1897. J. F. M. a. M. J. J. A. 8. 0. N. D. Total Ratio of 100 573 10.4 534 9.8 632 11.6 728 13.4 642 11.8 487 9.0 317 6.8 284 5.2 233 4.3 262 4.8 317 5.8 432 7.9 Mean Monthly Temperature, 1876-1897. Figures from essay, " La Fiebre Tifoidea en Buenos Aires," by Dr. Diego T. R. Davison, " Anales del Departamento Nacional de Hijiene." Ano VIII. Niim. 13. J. r. M. A. M. J. J. A. s. 0. N. D. Centigrade Fahrenheit 23.5 73 22.8 73 21.2 70 16.7 63 13.2 55 10.3 50 10.4 50 11.5 53 13.3 55 16.1 61 19.8 68 22.4 72 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVER. 567 III. INTERPEETATION OF THE STATISTICAL RESULTS. An examination of the curves plotted as above described shows that a very striking parallelism exists between the monthly variations in temperature and typhoid prevalence. Of the thirty communities considered, eighteen show this parallelism to be almost perfect; these are the Empire of Japan, the States of New York and Massa- chusetts, the District of Columbia, and the cities of Atlanta, Baltimore, Berlin, Boston, Buenos Ayres, Denver, Leipsic, London, Mobile, Montreal, New York, St. Paul, San Francisco, and Santiago. Three other typhoid curves — those for India, for Charleston, and for New Orleans — rise with the- temperature in spring, and fall with it in autumn, but show a temporary decrease in the disease during the time of great- est heat. In all these twenty-one cases the connection between the two factors seems too close not to indicate a vital relation. In the northern cities — Montreal, Boston, Denver, and St. Paul — the curve of typhoid is acute ; in cities with a more and more equable temperature the curve of the disease is progressively flattened, the limit being reached in the case of San Francisco. In the northerly localities the maximum occurs in September and October; in the southern cities, with a milder winter, it comes in August (Atlanta) or July (Charleston and Mobile). In the two cities of the southern hemisphere the curves of both typhoid fever and temperature are exactly reversed. In the case of the tropical and sub-tropical regions — India, Charleston, and New Orleans — it appears that the rise with the temperature, after beginning in the usual fashion, is checked by some other factor, perhaps strong sun- light or extreme dryness. (See Plates I.-VIII.) It remains now to consider the nine cities which show more or less irregular curves, and to see if their abnormalities are capable of explanation. These nine cities are Chicago, Cincinnati, Dresden, Munich, Newark, Oakland, Paris, Philadelphia, and Vienna. The first thing to notice in this connection, and the one all previous students of seasonal variation have neglected is the necessity of discriminating between sharp epidemic outbreaks of the disease and the slow succession of isolated cases which characterize that condition known to the older sanitarians as "endemic." The term endemic has been so misused and has become so associated with the idea of a mysterious miasm inherent in a geographical region, that it cannot be safely used in a more scientific sense. At the same time a distinction, vital to the epidemiologist, must be drawn between the infection which reaches a number of persons at once through a single medium as water or milk, and the slower, more complex process by which a disease passes from person to person through a population, the path of the 568 SEDGWICK AND WINSLOW. BACILLUS OF TYPHOID FEVEK. contagious material being different in each individual instance. For this sort of infection which spreads gradually in a community instead of striking a large number of persons at a single blow, the term "prosodemic," meaning "through" or "among" the people, has been suggested. In the examination of data bearing on the question of the seasonal prevalence of typhoid fever it is obviously the prosodemic disease which should be mainly considered. Cases of thif^ sort furnish a large number of independent facts which may be averaged together fairly; while an epidemic must always be a perturbing element. Thus, for example, a public water supply furnishes exceptional facilities for the distribution of infection from its watershed to a large number of individuals. Twelve hundred cases of typhoid fever at Plymouth, Pa., derived from a single house on the banks of a reservoir have, for a study of normal seasonal variations, far less significance than fifty cases, in which the paths of infection are separate and independent. Curves of seasonal variation which are based on a small number of cases will always be liable to show irregularities due to single epidemics ; and if our tables of typhoid deaths be inspected, it will at once be seen that four of the nine exceptions to a regular seasonal distribution are due to this cause. Thus the form of the Oak- land curve is distorted by the epidemic of twenty-two deaths in July, 1893, which we are informed by the local authorities was due to an infection of the milk supply. The largest number of deaths in any other month in the ten years was seven, so that this irregularity could not be compensated. Similarly, the Munich curve owes its peculiarity to the epidemic of thirty-five deaths in June and July of 1893, the largest number in any other month being nine. The curve for Vienna is controlled, in a similar way, by an epidemic in Decembei', 1888, and January and February, 1889. In all these cases the curve would follow the temperature more or less normally if these perturbations were eliminated. Again for Dresden the total number of deaths is so small that eight cases in April, 1894, cause a notable distortion. That the typhoid in this city did follow the temperature when there was enough of it to give average results is shown by Fiedler's figures for 1850-60, quoted above. We may thus consider that the irregularities of the Oakland, Munich, Vienna, and Dresden curves are explained by the fact that the number of cases considered is too small to eliminate the haphazard effect of epidemics. There remain to be explained the exceptions offered by Cliicago, Cincinnati, Newark, Paris, and Philadelphia, in all of which cities the amount of material is amply sufficient to prevent mere chance irregularities. If the curves for these five cities be compared, it will at once be noted that they exhibit a remarkable resemblance. Besides the summer rise, each curve SEDGWICK AND WINSLOW. — BACILLUS OP TYPHOID FEVER. 569 exhibits two secondary maxima, one in December or January, the other between March and May. If our general theory be correct, there must in these localities be some special condition tending to produce typhoid epidemics in the early winter and the early spring, which modifies the normal influence of the season. Fortunately, we know exactly what this influence is. These five cities — and of the thirty communities we have considered, these five only — draw their water supply from surface sources liable to gross pollution. The epidemics of March, 1899, at Newark ; of May, 1891, at Chicago; of January, 1888, and December, 1889, at Paris, as well as the lesser winter and spring outbreaks in other years, were unquestionably due to the public water supplies of those cities. We have here then a special condition influencing the occurrence of epidemics in cities having surface water supplies and therefore de- ranging the normal course of prosodemic typhoid. The heavy autumn rains and the spring floods consequent on the melting of the winter's snow, carry into surface water supplies a larger amount of pollution than reaches them at any other time, — as is well shown by a comparison of the bacterial content of surface water at various sea- sons. We may venture to generalize by saying that winter and spring epidemics are characteristic of those cities whose water-supply is most subject to pollution ; they are absent from communities which use filtered water or water obtained from adequately protected watersheds. Finally, then, it appears that of the thirty commimities we have studied, all but four, in which the number of cases is too small to furnish average results, give typhoid curves corresponding to one of three types, — the normal temperature distribution, the subtropical modification, and the modification due to winter and spring water- epidemics. These latter types of distribution are explicable as the resultant of a combination of the temperature factor with another. We may therefore conclude that wherever a sufficient number of cases has been considered a direct relation between typhoid fever and temperature appears to be general and invariable. IV. CONCLUSION" OE THE AUTHOES THAT THE SEASONAL PREVALENCE OF TYPHOID FEVEE DEPENDS MAINLY UPON SEASONAL TEMPEEATUEE. The increase of typhoid fever with a gradual rise in the mean air temperature of a given locality appears to be a phenomenon so widespread and significant as to indicate beyond reasonable doubt some relation between the two factors. Whether this rela- tion be direct or indirect must be determined by considerations as to the aetiology of the disease and as to the relation of temperature to the various vehicles mainly con- cerned in its transmission. 570 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVER. The methods by which prosodemic typhoid may spread are ahnost innumerable. The last link in the chain is, in most cases, some article of food or drink, and the food becomes infected, in many instances, from the fingers of a typhoid patient or of his unprofessional attendants. The transmission of typhoid fever on a large scale by water and milk has led sanitarians to minimize unduly this direct personal element in its aetiology. In a well-organized, thoroughly sanitary city dwelling the distinction between contagion and infection is an important one ; but in dirty surroundings typhoid becomes, for all practical purposes, a contagious disease. This fact, in itself, throws some little light on its seasonal prevalence. A large number of persons who live ordinarily in cities, surrounded by many sanitary safeguards, in vacation time are exposed in camps and summer resorts to abundant opportunities for filth infection. The autumn fever, in small part at least, occurs among those who are attacked on such summer vacations or immediately after their return home. Again, several special sources of food contamination have a more potent influence at this season of the year. Those observers are perhaps correct who consider that ground waters are most dangerous when the wells are at their lowest and liable to receive impurities from a wide area. Professor Gualdi would explain the facts by attaching great significance to raw vegetables as vehicles for the transmission of t3-phoid fever ; and he has traced out a more or less close connection between the consumption of these articles and the amount of typhoid in Rome. Most original of all is the sugges- tion of Bonne, who seeks to explain the autumnal maximum at Hamburg by the in- creased amount of bathing in the Elbe beginning with the July heat Of the three great intermediaries of typhoid transmission, fingers, food, and flies, the last is even more significant than the others in relation to seasonal variation. Since the emphasis laid on this vehicle of infection by the surgeons who studied the conditions of the late Spanish War, our conception of its importance has grown more and more considerable. There can be little doubt that many of the so-called " sporadic " cases of typhoid fever which are so difficult for the sanitarian to explain are condi- tioned by the passage of a fly from an infected vault to an unprotected table or an open larder. The relation of this factor to the season is of course close and complete ; and a certain amount of the autumnal excess of fever is undoubtedly traceable to the presence of large numbers of flies and to the opportunities for their pernicious activity. None of the factors noted, however, nor the whole of them taken together, seem to us to account satisfactorily for the observed phenomena. Neither the agency of insects, nor the exposure of urban subjects to rural unsanitary conditions, though both are undoubtedly important, can be held to account for a phenomenon so con- SEDGWICK AND WINSLOW. BACILLUS OF TYPHOID FEVER. 571 stant, so striking, and so universal. The parallelism between the curves of typhoid and of temperature is too close not to suggest in the strongest manner some direct relation such as was postulated by Murchison, Liebermeister, and Davidson. No one doubts a direct correlation between the growth in a wheat-field and the changes of temperature during the changing seasons. The fundamental properties of protoplasm are so constant that there seems no reason to doubt a similar favorable effect of the warmth of summer, not on the crop of typhoid plants growing in human bodies, but on the survival seed which passes from one body to another through the environment. This is theoretical ; but the experiments reported in the first section of this paper furnish practical evidence to confirm the a priori hypothesis that it must be more difficult for an organism habituated to a temperature of 98° F. to persist in Nature when the thermometer is at 30° than when it is in the neighborhood of 80°. We do not wish to assert that the typhoid bacillus multiplies in the environment during the summer months of a temperate climate. It is the absence of the destruc- tive influence of cold, rather than any stimulating influence of heat, which permits the rise culminating in the autumnal maximum. In fine, the probable mechanism of the seasonal changes according to our conception is as follows : — The bacteriology and the aetiology of typhoid fever both indicate that its causal agents cannot be abundant in the environment during the colder season of the year. The germs of the disease are carried over the winter in the bodies of a few patients and perhaps in vaults or other deposits of organic matter where they are protected from the severity of the season. The number of persons who receive infection from the discharge of these winter cases will depend, other things being equal, upon the length of time for which the bacteria cast in these discharges into the environment, remain alive and virulent. The length of the period during which the microbes live will depend largely upon the general temperature; as the season grows milder, more and more of each crop of germs sent at random into the outer world will survive long enough to gain entry to a human being and bear fruit. The process will be cumu- lative. Each case will cause more secondary cases ; and each of the latter will have a still more extensive opportunity for widespread damage. In our opinion the most reasonable explanation of the seasonal variations of typhoid fever is a direct effect of temperature upon the persistence in Nature of germs which proceed from previous victims of the disease. PART III. BIBLIOGRAPHY. A. ON DISEASE ATTRIBUTED TO POLLUTED ICE AND ICE-CREAM. 1. Nichols, A. H. Report on an Outbreak of Intestinal Disorder attributable to the Contamination of Drinking Water by Means of Impure Ice. Seventh Ann. Report, S. B. H., Mass., 1876, p. 4G7. 2. Smart, Charles. On Mountain Fever and Malarious Waters. Am. Journ. Med. Sci., Jan. 1878, p. 17. 3. Sickness from Impure Ice. Second Ann. Report, S. B. H., Conn., 1879, p. 90. 4. Chamberlain, C. W. Impure Ice. Fifth Ann. Report, S. B. H., Conn., 1882, p. 295. 5. Duclaux. Les Inipuretes de la glace. Ann. d'Hyg., 1884, 3rd series, XII, p. 97. 6. Riche, A. Emploi de la glace dans I'alimeutation. Ann. d'Hyg., 1893, 3rd series, XXX, p. 47. 7. Derange. Epidemie de fifevre typhoide due a I'ingestion de glace impure. Rev. d'Hyg. at de Police Sanitaire, XX, 4, 1898, p. 295. 8. Turner, G. Report on an Epidemic of Enteric Fever due to the Consumption of Ice-Cream. Practitioner, London, 1892, XLIX, p. 141. 9. Munro, A. C. Epidemic of Enteric Fever, traceable to Infected Ice-Creams and Water-Supply and attacking over Eight Hundred Persons. Public Health, London, 1894-95, VII, p. 30. 10. Vaughan, V. C, and Perkins, G. D. Ein in Eiscrgme und Kase gefundener giftproducirender Bacillus. Arch. f. Hyg., 1896, XXVII, p. 308. 11. Hope. Ice-Cream as a Vehicle of Infection in Typhoid Fever. Liverpool M.-Chir. J., 1898, XVIII, p. 185. 12. Taylor, L. H. Report upon the Epidemicof Typhoid Fever at Plymouth, Pennsylvania. First Ann. Report State Board of Health, Penn., 1885, p. 176. B. ON THE BACTERIOLOGY OF NATURAL ICE, SNOW AND HAIL, AND OF ICE-CREAM. 13. Sanderson, J. B. The Origin and Distribution of Microzymes (Bacteria), in Water, etc. Quar. Jour. Mic. Sci., XI, 1871, p. 323. 14. Cohn, F. Untersuchungen ixber Bakterien. Beitrage zur Biologic der Pflanzen, Band I, 1875, II, p. 220. 15. Leidy, J. Organisms in Ice. Proc. Acad. Nat. Sci., Phila., 1884, p. 260. 16. Pohl. (Cheiaical and bacteriological investigations in relation to the water-supply of St. Peters- burg.) Wratsch, 1884, No. 9. 17. Gardiner, J. T. Report on the Purity of Ice from Onondaga Lake, the Erie Canal at Syracuse, and from Cazenovia Lake. Report to the State Board of Health of New York, July, 1886. 18. Breunig, J. Bakteriologische Untersuchung des Trinkwassers der Stadt Kiel. (Inaug. Disser- tation.) Kiel, 1888. (Ref. Baumgarten's Jahresbericht, IV, p. 480.) 19. Kowalski. Ueber bakteriologische Wasseruntersuchungen. Wiener klinische Wochenschrift, 1888, Nos. 10, 11, 14, 15, and 16. (Ref. Centr. f. Bakt., IV, p. 467.) 20. Heyroth. Ueber den Reiulichkeitszustand des kunstliches Eises. Arb. aus dem K. Gesundheit- samte, IV, 1888, p. 1. 574 SEDGWiCK AND WINSLOW. — BACILLUS OF TTPIIOID FEVER. 21. Report upon the Pollution of Ice Supplies. Twenty-first Ami. Eeport, S. B. H., Mass., 1889, p. 143. 22. Character and Quality of the Ice-Supply of London. Lancet, Vol. LXXI, 1893, II, p. 269. 23. Girard, C. et Bordas, F. Analyses chimiques at bacteiiologiques des glaces consommees k Paris. Ann. d'Hyg., 1893, 3rd series, XXX, p. 78. 24. Christomonas, A. C. Ueber einige Eigenschaften des kiinstlichen Eises. Oesterr. chem. ztg., 1898, p. 486. 25. Janowsky. Ueber den Bakteriengehalt des Schnees. Wratsch, 1888, IX, p. 727. Also Ceutr. f. Bakt., IV, 1888, p. 647. 26. Schmelck. Eine Gletscherbakterie. Centr. f. Bakt., IV, 1888, p. 545. 27. Schmelck. Bakteriologische Untersuchungen des Trinkwassers in Christiania. Centr. f. Bakt., VIII, p. 102. 28. Bujwid, 0. Die Bakterien im Hagelkornern. Centr. f. Bakt., Ill, 1888, p. 1. Also Ann. de I'lnst. Past., 1887, I, p. 592. 29. Foutin, W. M. Bakteriologische Untersuchungen von Hagel. Wratsch, 1889, Nos. 49, 50. (Ref. Centr. f. Bakt., VII, 1890, p. 372.) 30. Giacosa, P. Studii sui germi di microrganismi nella neve delle alte montagne. Giornale della R. Accademia di medicina, 1890, Nos. 11-12. 31. Abbott, A. C. Bacteriological Study of Hail. Johns Hopkins Hospital Bulletin, 1890, Vol. I, No. 5, p. 56. 32. Dominguez, S. e Calumbide. Bacterias del hielo. Ann. asist. piib. Buenos Aires, 1891-2, II, pp. 213, 263. 33. Scofone, L. Esame batteriologico delle acque di neve di torrente e di lago. Arch, per le sc. med. de Torino, XX, 1895, n. 12. 34. Scofone, L. Esame batteriologico di neve e ghiaccio. Arch, per le sc. med. de Torino, 1897, XXI, p. 437. 35. Harris, A. E. Report to Chairman and Members of the Public Health Committee, Islington, 1894. (Ref. Lancet, LXXII, 1894, p. 862.) 36. Nield-Cook, J. The Bacteriology of Ice-Cream. Public Health, London, 1895-6, VIII, p. 252. 37. Stevenson. Bacteria in Ice-Cream. Food and Sanitation, April 16, 1898. 38. Wilkinson, J. Ice-Creams : their Manufacture and Bacteriology. Public Health, London, 1899, XI, p. 255. 39. Nystrom, C. Om fasnings och forruttnelseprocesserna pa Spetsbergen, Upsala, Lakareforeuings forhandlingar, t. IV. 40. Couteaud. Bacteriologie de la zone glaciale. Arch, de med. nav., 1893, LIX, p. 119. 41. Levin. Les microbes dans les regions arctiques. Ann. de I'Inst. Pasteur, XIII, 1899, p. 558. C. ON THE EFFECT OF FREEZING AND OTHER LOW TEMPERATURES UPON BACTERIA. 42. von Frisch, A. Sitzungsber. der Wiener Akad. d. Wissenschaften, LXXV, p. 257. 43. Pictet et Young. De Paction du froid sur les microbes. Compt. rend. Acad, des Sc, XCVIII, 1884, p. 747. 44. Finkler und Prior. Forschungen ilber die Cholerabakterien. Bonn, 1885. 45. McKendrick. The Action of Cold on Microphytes. (Communication to Biological Section, British Association for the Advancement of Science.) Natiu-o, XXXII, 1885, p. 561. 46. Forster, J. Ueber einige Eigenschaften leuchtender Bakterien. Centr. f. Bakt., 1887, II, p. 337. 47. Fischer. Bakterienwachsthum bei 0° C. Centr. f. Bakt., 1888, IV, p. 89. 48. d'Arsonval, A. Emploi de I'acide carbouique liqu^fi^ pour la filtration et la sterilization rapides des liquides organiques. Compt. rend. Acad, des Sc, 1891, CXII, p. C67. SEDGWICK AND WINSLOW. BACILLUS OF TYPHOID FEVER. 575 49. Forster, J. Ueber die Entwickelung von Bakterien bei niederen Temperaturen. Centr. f. Bakt., 1892, XII, p. 431. 50. Fischer. Deutsche medizinische Woehenschrift, 1893, No. 23. 51. Fictet, R. De I'emploi methodique des basses temperatures en biologie. Archives des sciences physiques et naturelles de Geneve, 3r(l period, XXX, No. 10, p. 293. 52. d'Arsonval et Charrin. Influence des agents atmospheriques, en particulier de la lumiere, du froid, sur le bacille pyocyanogfene. Compt. rend Acad, des Sc, 1894, CXVIII, p. 151. 53. Do. Influence des agents cosmiques (electricite, pression, lumiere, froid, ozone, etc.) sur revolu- tion de la cellule bacterienne. Arch, de physiol. norm, et path, 1894, 6 s., VI, p. 335. 54. Weber, A. Zur Aetiologie der Krebspest. Arb. aus dem K. Gesundheitsamte, XV, 1899, Heft 2. 55. Mason, W. P. Examination of Water. New York, 1899, p. 117. 56. Ravenel, M. P. The Eesistance of Bacteria to Cold. Medical News, Phila., June 10, 1899. 57. Cadeac et Malet. Becherches experimentales sur la virulence des matieres tuberculeuses dess^- chees, putriiies ou cong^lees. Lyon Med., 1888, p. 229. 58. Frankland, Grace and Percy. Micro-Organisms in Water. London, 1894. p. 253. 59. Kasansky, M. W. Die Einwirkung der Winterkalte auf die I'est- und Diphtheriebacillen. Centr. f. Bakt'., 1899, XXV, p. 122. 60. Abel, R. Versuche tiber das Verhalteu der Diphtheriebacillen gegen die Einwirkung der Winter- kalte. Centr. f. Bakt., 1895, XVII, p. 545. 61. Seitz, C. Bakteriologische Studien zur Typhus- Aetiologie. Munchen, 1886. 62. Billings, J. S. Sanitary Engineer, January 29, 1887, p. 211. 63. Chantemesse et Widal. Eecherches sur le bacille typhique et I'etiologie de la fievre typhoide. Arch, de Phys. norm, et path., 1887, p. 217. 64. Bazhenoff. (Effect of different agents on the typhoid bacilli of Ebert-Koch.) Ejened. klin. gaz. St. Petersburg, 1885, V, pp. 81, 95. 65. Janowski. Zur Biologie der Typhusbacillen, II. Centr. f. Bakt., VIII, 1890, p. 449. 66. Remlinger, P. Sur la sensibilite du bacille d'Eberth aux variations de temperature. Compt. rend, de la Societe de Biologie, 1897, 9 juillet. 67. Koch, R. Deutsche medizinische Woehenschrift, 1884, No. 32. 68. Raptschewski. Wratsch, 1886, No. 5. 69. Cornil et Babes. Les bacteries. Paris, 1890. Vol. II, p. 183. 70. Schrnff. Zum Sanitatsbericht tiber den Kreis Neuss filr 1893. (Cit. by Weiss.) 71. Finkelnburg. Zur Frage der Variabilitat der Cholerabacillen. Centr. f. Bakt., 1893, XIII, p. 114. 72. Karschinski, W. Archiv des Laboratoriums fur AUg. Pathologic an der Universitat zu War- schau, 1893. 73. Renk. Ueber das Verhalten der Cholerabacillen im Eise. Portschritte der Med., 1893, p. 396. 74. Uffelmann, J. Weitere Beitrage zur Biologie der Cholerabacillen. Einfluss der Kalte auf ihre Lebensfahigkeit. Berliner klin. Wochenschr., 1893, p. 158. 75. Wnukow, N. Ueber den Einfluss langerer Einwirkung niedriger Temperaturen auf die Kochs- chen Komma-bacillen. Wratsch, 1893, No. 18. 76. Montefnsco, A. Azione della basse temperature sulla virulenza degli spirilli del colera. Annali dell' Instituto d' igiene dell' Universita di Eoma, 1893, III, p. 31. 77. Abel. (Ref.) Centr. f. Bakt., XIV, 1893, p. 184. 78. Kasansky, M. W. Ueber den Einfluss der Kalte auf die Cholerabakterien von Koch und ahnliche Vibrionen von Pinkler-Prior, Miller, Deneke und die Vibrionen Metschnikoff. Centr. f. Bakt., 1896, XVTI, p. 184. 79. Weiss. Ueber das Verhalten der Cholera-Erreger bei niedrigen Temperaturen. Zeit. f. Hyg., XVIII, 1894, p. 492. 576 SEDGWICK AND WINSLOW. — BACILLUS OF TYPHOID FEVEK. D. on' the quantitative studies of the destruction of bacteria by freezing. 80. Pengra, C. P. Purification of Water by Freezing. Twelfth Ann. Report, S. B. H., Michigan, 1884. p. 79. : 81. Frankel, C. Ueber den Bakteriengehalt des Eises. Zeit. f. Hyg., I, 1886, p. 302. 82. Wolffhuegel und Riedel. Die Vennehrung der Bakterien im Wasser. Arb. aus dein K. Gesund- heitsaiute, Berlin, 1886, p. 455. 83. Prudden, T. M. , On Bacteria in Ice and their Relations to Disease, with Special Referenoe to the Ice-Supply of New York City. The Medical Record, 1887, March 26 and April 2. 81. Bordoni-UflFreduzzi. Die Biologische Untersuchung des Eises in sein Beziehung zur offentlichen Gesundheitspflege. Centr. f. Bakt., II, 1887, p. 489. 85. Russell, H. L. Preliminary Observations on the Bacteria in Ice from Lake Mendota, Madison, Wis. The Medical News, August 17, 1889. 86. Mills, H. F. Typhoid Fever in its Relation to Water Supplies. Twenty-second Ann. Report, S. B. H., Mass., 1890, p. 542. , E. ON THE SEASONAL PREVALENCE OF TYPHOID FEVER AND ITS RELATION TO SEASONAL TEMPERATURE. 87. Bartlett, Elisha. The History, Diagnosis, and Treatment of Typhoid and of Typhus Fever, 1842. 88. Flint, A. Clinical Reports on Continued Fever, Philadelphia, 1855. 89. Griesinger. Infektionskrankheiten. Vol. IIj Abth. 2 of Virchow's Handbuch der speciellen Pathologic und Therapie. Erlangen, 1857, p. 122. 90. Tweedie, A. Lectures on Fevers. Lancet, 1860, Vol. I, p. 1. 91. Haller. Die Volkskrankheiteh in ihrer Abhangigkeit von den Witterungsverhaltnissen. Wien, 1860. 92. Hirsch, A. Handbuch der historisch-geographischen Pathologie. Erlangen, 1860. p. 177. 93. Murchison, Charles. A Treatise on the Continued Fevers of Great Britain. 1862. 94. Fiedler. Statistische Mittheilungen tiber Typhus abdominalis. Archiv der Heilkunde, III, 1862, p. 155. 95. Buhl, L. Ei'n Beitrag zur Aetiologie des Typhus. Zeit. f . Biol., I, 1865, p. 1. 96. Seidel, L. Ueber den numerischen Zusammenhang, welcher zwischen der Haiitigkeit der Typhus- Erkrankungeu und dem Stande des Grundwassers wahrend der letzten 9 Jahre in Mtinchen hervorgetreten ist. Zeit. f. Biol., I, 1865, p. 221. 97. Seidel, L. Vergleichung der Schwankungen der Regenmengen mit den Schwankungen in der Haufigkett des Typhus in Milnchen. Zeit. f. Biol., II, 1866, p. 145. 98< Liebermeister. Zur Aetiologie des Abdominaltyphns. Deutsche Klinik., 1866, p. 89. 99. V. Pettenkofer, M. Ueber die Schwankungen der Typhussterblichkeit in Munchen von 1850 bis 1867. Zeit. f. Biol., IV, 1868, p. 1. 100. Socin, B. Typhus, Regenmenge und Grundwasser in Basel. Basel, 1871. 101. Virchow. Ueber die Sterblichkeitsverhaltnisse Berlins. Berliner klinische Wochenschrift, IX, 1872, p. 597. 102. Murchison, C. A Treatise on the Continued Fevers of Great Britain. Second Edition, London, 1873. 103. Liebermeister, K. Typhoid Fever. Ziemssen's Cyclopedia of the Practice of Medicine. (Ameri- can Edition, edited by A. H. Buck, 1874). 104. Cousot. llltude sur la nature, I'^tiologie, et le traitement de la fifevre typho'ide. 1874. p. 117. 105. Oesterlen, F. Handbuch der medicinischen Statistik. Tubingen, 1874. SEDGWICK AND WINSLOW. — BACILLUS OP TYPHOID FEVEE. 577 106. Buchan, A., and Mitchell, A. The Influence of Weather on Mortality from different Diseases and at different Ages. Journ. Scot. Meteor. Soc., 1874-5, N. S., IV, p. 187. 107. Sander, F. Handbuch der ofifentlichen Gesundheitspflege. Leipzig, 1877. 108. Sixth Ann. Report, Secretary, State Board of Health of Michigan. 1878. p. 293. 109. Oldendorff, A. Article, Morbiditats- und Mortalitats-Statistik, in Eulenburg's Real-Encyclopadie der gesammten Heilkunde, Vol. IX, 1881, p. 304. 110. Fodor, J. Deutech. Viert. f. off. Gas., VII, p. 206. Hygienische Untersuchungen tlber Luft, Boden, und Wasser. Braunschweig, 1881-2. 111. Welch, F. H. Enteric Fever, Philadelphia, 1883. 112. Hirsch, A. Handbook of Geographical and Historical Pathology. Second Edition (translated by Creighton, London, 1883). 113. Baker, H. B. The Relation of the Depth of Water in Wells to the Causation of Typhoid Fever. Twelfth Annual Report S. B. H., Michigan, 1884, p. 89. (Am. P. H. Assoc, October 16, 1884.) 114. Virchow und Guttstadt. Die Anstalten der Stadt Berlin fur die offentliche Gesundheitspflege und den naturwissenschaftlichen Untersicht. Berlin, 1886. p. 71. 115. V. Pettenkofer, M. Zum gegenwartigen Stand der Cliolerafrage. Miinchen und Leipzig, 1887. 116. Pifitor. Das offentliche Gesundheitswesen in Berlin, 1882-5. 1887. p. 57. 117. Soyka. Zur Aetiologie des Abdominaltyphus. Arch, f . Hyg., VI, 1887, p. 260. 118. V. Pettenkofer, M. Der epidemiologische Theil des Berichtes iiber die Thatigkeit der zu Erfor- schung der Cholera iin Jahre 1883 nach Aegypten und Indien entsandten deutschen Commission. Miinchen und Leipzig, 1888. 119. Almqnist, E. Ueber EinfltUss von Jahreszeit und Witterung auf das Auftreten von Infektions- krankheiten mit besonderer Berlicksiohtigung der localen Epidemieen. Zeit. f. Hyg., V, 1888, p.l. 120. Goldberg, B. Der Einfluss des Witterungsganges auf vorherrschende Krankheiten und Todesur- sachen. Erganzungshefte zum Centr. f. allgem. Ges. II, 1889, p. 317. 121. Magelsen. Ueber die Abhangigkeit der Krankheiten von der Witterung. Leipzig, 1890. 122. Davidson, A. Geographical Pathology. N. Y., 1892. 123. Woodhead, G. S. Minutes of Evidence, Royal Commission on Metropolitan Water Supply, Lon- don, 1893, ]). 506. 124. Korosi, J. Statistik der infektiosen Erkrankungen in den Jahren 1881-91 und Untersuchung des Einflusses der Witterung. Berlin, 1894. 125. Fodor, J. Hygiene des Bodens. Weyl's Handbuch der Hygiene, I, 1896, p. 194. 126. Jessen, F. Witterung mid Krankheit. Zeit. f. Hyg., XXI, 1896, p. 287. 127. Knoevenagel, 0. StudieniiberKrankheitsdispositionen und iiber Genius epidemicus. Deuts. Vier- telj. f. offentl. Gesundheitspflege, XXVIII, 1896, p. 298. 128. Baker, H. B. The Etiology and Pathology of Typhoid Fever. Twenty-fourth Annual Report of the Secretary of the S. B. H. of Michigan, 1897, p. cxlii. 129. Berger, H. Die Bedeutung des Wetters fiir die ansteckenden Krankheiten. Therapeutische Monatsbefte, XII, 1898, pp. 139, 201. 130. Ruhemann, J. Meteorologie und Infektionskrankheiten. Zeitschrift fiir Diatetische und Physi- kalische Therapie, I, 1898, p. 312. 131. Weichselbaum, A. Epidemiologie. Weyl's Handbuch der Hygiene, IX, 1899, p. 441. 132. Curschmann, H. Typhoid Fever and Typhus Fever. Nothnagel's Encyclopedia of Practical Medicine. English Translation, edited by Osier. Phil., London, 1901. EXPLANATION OF THE PLATES. Plates I.- VIII. are based upon the statistics given on pp. 540-566, as is stated on p. 539. Abscissae indicate months; ordinates indicate temperatures (shown by broken lines), and also percentages of yearly typhoid-fever mortality (solid lines) except in the curves for Newark, N. J. (Plate VI.), the Empire of India (Plate VII.), and Santiago de Chile (Plate V.), in which deaths, not cases, are indicated. It is important to remember that the curve of typhoid deaths in each case has been moved back exactly two months from its true position, and that for typhoid cases one month, as is explained on p. 539. PLATE I. PLATE II. /\ Ji 7/7. Mi y 5^ ut. -Jc n.—Tet 'Tia—Jan. ftqy j 4^ Jan. f.^y '^ \ ^7 o" / '■-_ f^'/ ' / * \ / / ^ t \ , 1 \ U ' 1 * / / / / f j 1 \ \ /ox~ 1 t \ n° / / ■ 1 t 1 1 1 \ o U / / t 1 / * \ ^ y 1 1 1 1 \ \ \ \ A / t I \ s \ \. \ \ \ \ \ -s% P /0 \ / 1 >, u U \ ^ Mc 1 /. /y Afc t/. ^r.-l}p hold— Man / 1 / July l^i V \ /^ar. / / / O^ 1—1 1 1 V-i r\ or. roui Typhoid deaths Temperature Denver 1 Jc n Mi 7 \ ^^ vf. Ml 'y ^f \ft Ja n. Ja 7. — Tei ip.-Jci n. f 'iV / \ •7 n° f/lvA 1 1 / ' \ I 1 — <5 7 1 t 1 f \ \ \ i I \ lO 7, /nv t U A 1 f / /' \ » \\ \\ \ \ V\ \ rf / 1 1 1 1 \ \ \ JTV o /o V \ \ V O u X \ O/' Mi / 1 V w *^ V r-7^p hoi(^—M ar Ju /y Ale K Ma r. 4 Montrea, 1 f \ Baltimore 1 PLATE III. PLATE IV. Jc n. M jK -54 pt Jc w.—Ten •ip.—J'an. May .Se, of. Jan. wiy n' IK% lOX- • ,' '' \ \ \ u n' t t 1 1 / / \ \' \ \ /\ Y \ / \ \ lO" ~v ■J T U 1 / \ ' V \ \ \ \ \ \ \ C»/ M nr. Ju ly A/c w. M u hoid-Mar. 1 July fjov. 1 I Mar. San Francisco 1 1 1 1 1 1 Cincinnati 1 "^pfioid deaths Temperature J IS'A 7/f. M 1 1 ay- Se '' / \ ■> — ' T \— ^ / / \ ' 1 \ 1 1 \ / r f 4 — '\ \ \ \ \ \ \ \ \ \ Jc m. (St \ \ \ \ Ji vn-Ten — 7 tp.-Ji 7rT. fO%- 1 1 t 1 1 1 1 1 1 1 1 1 1 1 1 1 1 / / / / / / \ \ \ \ \ \ \ \ \ \ \ \ \ \ ^ \ \ \ \ \ \ 5 / / / • \ \ \ \ -10% IT'. M 7r