SSSSJSSSS5SS \ ^^ M^ \>.jp-^«i:r 1 CORNELL UNIVERSITY. THE pi. SPTotocr Itibrarg THE GIFT OF ROSWELL P. FLOWER FOR THE USE OF THE N. Y. STATE VETERINARY COLLEGE 1897 8394-1 Digitized by Microsoft® Cornell University Library R 108.N53 V.2 Researches of the Loomis laboratory. 3 1924 000 339 659 »•' s Digitized by Microsoft® This book was digitized by Microsoft Corporation in cooperation witli Cornell University Libraries, 2007. You may use and print this copy in limited quantity for your personal purposes, but may not distribute or provide access to it (or modified or partial versions of it) for revenue-generating or other commercial purposes. Digitized by Microsoft® Digitized by Microsoft® Digitized by Microsoft® RESEARCHES OF THE LooMis Laboratory Digitized by Microsoft® Digitized by Microsoft® RESEARCHES OF THE LooMis Laboratory Medical Department OF THE University of the City of New York. VOLUME 2. NEW YORK: Press of STETTINER, LAMBERT & CO., 22, 24 & 26 Reade Street. 1892. Digitized by Microsoft® A/S3 Digitized by Microsoft® CONTBKTS. PAGE Some Toxicological Points in a Case of Homicide by Morphin. R. A. WiTTHAUS, M.D I Tiie Use of Automatic and other Models in Teaching Physio- logy. W. Oilman Thompson, M.D iS Primary Tuberculosis of the Bronchial Glands. Henry P. LooMis, M.D 52 Six Cases of Supposed Hydrophobia. J. M. Byron, M.D 76 Pure Cultures of Lepra Bacillus. J. M. Byron, M.D. ....... 87 On the Detection of Quinin in Chemico-Legal Cases. R. A. WiTTHAUS, M. D 91 The Functions of the Cerebellum as indicated by recent Ex- periment. Edward D. Fisher, M.D 102 The Arrangement of the Muscular Fibres of the OEsophagus. C. G. Coakley, A.m., M.D 113 A Method of Making Photomicrographs of Low and Medium Powers. Henry S. Stearns, M.D iiS A Method of Obtaining Zirconium Oxid. H. C. Vanden- x bergh, M.D 117 Arsenic Trioxid. Ivin Sickels, M.S., M.D 121 Pure Cultures of Tubercle Bacilli obtained from Sputum. Frank Abbott, Jr., M.D 125 A Method of Double Staining Bacteria in Cover-Glass Prepara- tions and Tissues. J. M. Byron, M.D 126 Double Staining Gonococci in Pus, etc. Frank Abbott, Jr., M.D 127 A New Flagellated Bacillus, the Spirillum Flagellatum. Frank Abbott, Jr., M.D 128 A Method of Fixing Sections on the Cover Glass for Staining Micro-organisms in Tissue. J. M. Byron, M.D 129 Micro-organisms in Typhus Fever. J. M. Byron, M.D., and Egbert Le Fevre, M.D 130 A Method of Preserving Urinary Casts. Warren Coleman, M.D 134 An Improved Pressure Filter. J. M. Byron, M.D 135 Digitized by Microsoft® Digitized by Microsoft® SOME TOXICOLOGICAL POINTS IN A CASE OF HOMICIDE BY MORPHIN. R. A. WITTHAUS, M.D., Professor of Chemistry, Physics, and Hygiene. Some account of the analysis in the case of the People vs. Carlyle W. Harris is, we believe, worthy of perma- nent record, as it is apparently one of but two cases of homicide by morphin, or by an opiate in which the presence of morphin in the cadaver was demonstrated by modern methods, and one of a comparatively small number of cases in which the alkaloid has been separated from the cadaver. The criminal courts of the county of New York have had but little to do with the crime of poisoning. During the past thirty-five years they have had to consider only four cases of murder by poison. In 1859 Alexander Stevens was convicted of the murder of his wife by arsenic, and was sub- sequently executed. The analysis in this case was made by Prof. R. O. Doremus. In 1872-73 Frederick Heggi was twice tried, and acquitted upon the last trial. In 1888 Wil- helmina Lebkuchner poisoned her three children, but was acquitted on the ground of insanity. In 1892 Carlyle W. Harris was convicted of the murder of his unacknowledged wife by morphin. During the period between 1879 and 1892 two other cases of murder by poison occurred, in which no trial was had, because the person who administered the poison committed suicide at the same time. These were the cases of Albert Deffarge in 1879 and of Christine Fritz in 1885. The former poisoned his son and himself with mor- phin, according to the verdict of the coroner's jury, al- though no analysis was made. The latter poisoned herself and two children with potassium cyanid. During the ^ Digitized by Microsoft® 2 Some Toxicologtcal Points in a same period two unsuccessful attempts at poisoning were the subjects of trials : in 1879 Joseph and Mary Volckmer were convicted of an attempt to murder by morphin ; and in 1888 Mary Dygutis attempted to murder her employer and his family with Paris green. Although cases of accidental and suicidal poisoning by the opiates are of extremely frequent occurrence, cases of supposed or proved murder by these agents are but rarely recorded. An exhaustive search of the literature and an extensive correspondence with prosecuting officers have led to the records of only nine cases of supposed or proved murder by morphin since the discovery of that alkaloid : 1. The earliest case (and also the first recorded case of murder by an alkaloid) was the affaire Castaing in 1823. A French physician was convicted and executed for the mur- der of his friend by morphium acetate. The conviction was had entirely upon moral evidence, the post-mortem and analysis having afforded no positive indications.' 2. In 1871 Dr. Medlicott was convicted of murder by morphin and atropin' in Michigan. The analysis, by the Stas method, showed the presence of morphin by the nitric acid, ferric chlorid, Frohde, iodic acid, platinic chlorid, and sulphuric acid and potassium dichromate tests ; and the presence of atropin by the sulphuric acid and physiological tests. The deceased was found dead in bed, and the autopsy was made the same day. 3. The Sonzogno case in Italy, in 1874, was one of sup- posed morphin poisoning. The analysts who first ex- amined the viscera reported the presence of morphin, al- though they had failed to obtain the ferric chlorid reaction. It was during the examinations of this case, and of another at about the same period, that Selmi demonstrated the ex- istence of ptomains and pointed out the possibility of their being mistaken for the vegetable alkaloids.^ 4. In November, 1875, the case of the People vs. Eliza A. Stone was tried in Columbia County, New York, and the ' Tardieu, " Empoisonnement," p. 910. " W. H. Saunders, " The Ruth-Medlicott Poisoning Case," Mich. Univ. Med. Jour., Ann Arbor, 1871-72, 641-652. " F. Selmi, " Sulle Ptomaine," etc., Bologna, 1878. Digitized by Microsoft® Case of Homicide by Morphin. 3 defendant was acquitted, the evidence being considered in- sufficient as connecting her with the cause of death. Dr. John T. Wheeler, who made the post-mortem, informs me that the " autopsy showed absence of disease and presence of venous congestion. But chemical examination showed no morphin in the stomach." I can find no published ac- count of this case. 5. Dr. A. Krauss, of Tiibingen, reports' a case of homicidal morphin poisoning which occurred in 1878 in a German village, in which the chemical evidence is not conclusive, if it was no more extensive than is indicated in the report. The contents of the stomach, " treated by the method of Julius Otto, yielded needle-like crystals, visible with a mag- nifier, and a granule of iodic acid was added to their solu- tion. As the solution was thus colored red to red-violet the presence of morphin was demonstrated." The iodic acid test is incorrectly described. Taken alone or in conjunction with the fact that the substance is crystalline, it is by no means proof of the presence of morphin." 6. Dr. Krauss also reports at length " the singular case of Anna Thormahlen, who was tried in 1876 and acquitted, although nine weeks after the death of her husband she accused herself of having poisoned him with morphin,^ and gave a circumstantial account of the manner of adminis- tration. The body was exhumed seventy-one days after death. " The autopsy, however, gave in every respect such negative results that we unreservedly concur in the expres- sion of the chemical expert, that we can consider it as not having been made" (p. 167). 7. The Deffarge case, mentioned above, in which no analysis was made. 8. The Volckmer case, mentioned above, which was an unsuccessful attempt. 9. The Harris case. ' Friedreich's Blatter, 1883, xxxiv., 370-378. " Since the above was written, I have received, through the kindness of Dr. Krauss, a MS. copy of the report of the chemical analysis in this case. From this it appears that the iodic acid test was properly applied, and that no attempt was made to obtain any of the other morphin reactions, although apparently the material at the disposal of the analysts was sufficient. " Friedreich's Blatter, ig87, xxxviii.,.153-185. „^ Digitized by Microsoft® 4 Some Toxicological Points in a Lacassagne ' states that from 1825 to 1880 there were in France 2,123 cases of criminal poisoning (excluding suicides) which were the subject of judicial investigation. Of these twenty-one were by " opium, laudanum, morphin, etc." No further particulars are given. In none of the above eight cases previous to the Harris •case, except the Ruth-Medlicott case, was the presence of morphin satisfactorily shown by chemical evidence. But few cases of suicidal or accidental death from mor- phin have been recorded in which an analysis has demon- strated the presence of the poison. The principal points in these cases are as follows : 1. Dr. C. U. Shephard, Jr. :' Man. Died in about one hour from seven and a half grains of morphium sulphate. Dr. H. D. Geddings obtained from two fluidounces of stomach con- tents one and a half grains of morphium acetate by the Erd- mann-Uslar process, and identified it by the iodic acid, nitric acid, and ferric chlorid reactions. 2. Langsdorff :^ Man, get. 25. Died, in five hours from 0.4 gramme (6.17 grains) morphium chlorid, taken in mistake for quinin. The stomach was not pumped out before death. Autopsy next day. M. Scheid separated morphin (amount not stated) from the viscera, and iden- tified it by the nitric acid, Frohde, iodic acid, ferric chlorid, and Pellagri reactions. 3. Maschka :* Woman, ast. 24. Died in nineteen hours from an unknown dose of a salt of morphin. Stomach not evacuated. Prof. Lerch obtained a crystalline residue from the contents of the stomach, which gave the ferric chlo- rid and nitric acid reactions for morphin. The residue weighed one-half grain, but was not pure morphin. 4. Reyher:' Man, aet. 30. Died in about eight hours. Stomach not evacuated. Kauzmann ' separated 0.1052 gramme (1.62 grains) of crystallized morphin from two- 1 Arch. del'Anthrop. Crim., Par., 1886, i., 260. ' Tr. S. Car. Med. Assoc, 1880, xxx., 25-27. ' Vrtljschr. f. ger. Med., 1888, N. F. xliii., 212-221. * Vrtljschr. f. pr. Hlk., Prag, 1867, xcvi., 23-25. » Deut. Arch. f. kl. Med., 1868, iv., 602-609. ' Diss. Dorpat, 1868, pp. 80-86. Digitized by Microsoft® Case of Homicide by Morphin. 5 thirds of the stomach and contents, and obtained affirmative results with the silver nitrate, iodic acid, Husemann, Frohde, and ferric chlorid reactions. The morphin reactions were also obtained with the extracts from the intestines, bile, blood, liver, and liquid which escaped from the mouth and nose. 5. Dr. Edwards ' extracted crystallized morphin from the contents of the stomach of a child, and identified it by the nitric acid, iodic acid, and ferric chlorid reactions. 6. Kreyssig" separated twenty-two grains (?) of crystallized morphin from the material vomited by a young woman six hours after she had taken half a teaspoonful of solid mor- phium acetate. The residue responded to the ferric chlorid reaction. No reaction was obtained with the residues from the faeces and urine. 7. In an insurance case in 1890 the author obtained re- actions for morphin with the ferric chlorid, Husemann, Erdmann, Pellagri, iodic acid, and nitric acid tests from the extract from the contents of the stomach. No quantitative determination was attempted. There are also a number of recorded cases in which the statement is made that morphin was detected, without detail of the reactions employed ; and several cases of death from laudanum and other preparations of opium, in which morphin, or meconic acid, or both, were detected, with more or less certainty. In the Harris case morphin was detected in the stomach and intestines only. The parts analyzed were the contents of the bladder, the stomach and its contents, the intestines and their contents, the liver, the brain, and a pint of the bloody liquid which accumulated in the abdominal cavity during the post-mortem. The method of analysis was practically the Erdmann- Uslar modification of the Stas process. The material was extracted three times with water faintly acidulated with hydrochloric acid, and the solutions separated by filtration and expression. The acid aqueous liquid was evaporated in small portions at a temperature not exceeding 60° C. to the ' Pharm. Journ, and Trans., 1857, xvii., 42. ' ""'''lii^m'cfhfMicrosom 6 Some Toxicological Points m a consistence of a thin syrup. After cooling, strong alcohol was gradually added, during constant stirring, until nothing further was precipitated. After standing twenty-four hours a further quantity of alcohol was added, and the liquid fil- tered after two or three hours. The alcohol was distilled off from the filtrate after the addition of water. The acid aqueous liquid was then shaken successively with several portions of petroleum ether, of amylic alcohol, and of pe- troleum ether again, each solvent being separated and evap- orated apart. The aqueous liquid was then rendered alka- line with ammonium hydroxid, and immediately shaken with petroleum ether, which, after separation, was filtered into a glass capsule and evaporated, leaving a residue in which quinin would be found if present. This extraction with petroleum ether was repeated three to five times, the aqueous liquid being each time acidulated and again ren- dered alkaline immediately before the agitation with the ether. The acidulated aqueous liquid was then again ren- dered alkaline with ammonium hydroxid and immediately agitated with amylic alcohol, the extraction with amylic alcohol being repeated also three to five times under the same conditions, and the separated alcoholic layers evapo- rated. During this treatment, although the addition of am- monia did not produce an immediate precipitate in the aqueous liquid, when it was allowed to stand over night the walls of the vessel were found to be coated with an abun- dant crystalline deposit, some of which also floated in the liquid. These crystals somewhat resembled those of mor- phin in shape, but were found to contain magnesium, phos- phoric acid, and arsenic, probably produced by the action of the arsenical embalming liquid upon the normal phos- phates. It may be mentioned that a considerable propor- tion of arsenic passed into the amyl alcohol from the acid aqueous solution, and appeared in those residues in the oc- tahedral crystals of the trioxid.' The residues of evapora- tion of the amylic extracts were partly gummy and partly imperfectly crystalline, and all yellow in color. Repeated attempts at purification were made by dissolving the resi- ' These crystalline residues will be further considered in another paper. Digitized by Microsoft® Case of Homicide by Morphin. 7 due in dilute hydrochloric or sulphuric acid, agitating the acid aqueous liquid with amylic alcohol so often as this took up any color, then alkalinizing with ammonium hydroxid and extracting with amylic alcohol ; and by solution in ethy- lic alcohol. But as it was found that the accompanying im- purity behaved toward these solvents in the same way as morphin, and as each operation was probably attended by some loss of material, the attempt at further purification was abandoned. Slight modifications in the earlier stages of the process were adopted with the blood, with the contents of stomach and of bladder, and with the liver. With the liver the acid aqueous liquid was extracted with a mixture of equal vol- umes of chloroform and ether before treatment with petro- leum ether. The contents of the stomach were first treated with alcohol, acidulated with hydrochloric acid, filtered after standing, the residue washed with alcohol, the alcohol expelled, and the remaining acid aqueous liquid treated directly with immiscible solvents. The blood and contents of the bladder were first freed of albuminoids by heating with a cold saturated solution of sodium sulphate and acetic acid, filtration, and removal of excess of sodium sulphate. The tissue of the stomach and its contents were treated separately during the first stages of the process, including the examination for quinin, but the residues were sub- sequently united and examined together for morphin. The residue left by the evaporation of the amyl alcohol extract was certainly not pure morphin ; a yellow, oily material, probably a ptomain, persistently accompanied it, and was still present in the final residue, to which the iden- tification tests were applied. It is certain that during these attempts at purification an appreciable quantity of morphin was lost, and that, under the conditions, an accurate quanti- tative estimation was not practicable, although an attempt at one, a mere rough guess, was made by a comparison of the weight of the free alkaloid with its weight after conversion into the chlorid. The tests applied gave results which were certainly con- clusive proof of the presence of morphin. Certain of them were simply group reactions, indicating the presence of Digitized by Microsoft® 8 Some Toxicological Points in a some alkaloid, without individualizing it in any way. Such were the reactions with phosphomolybdic acid, phospho- tungstic acid, potassium and cadmium iodid, and auric chlorid. Affirmative results were obtained with nine char- acterizing reactions, no one of which taken alone is peculiar to morphin, but which taken together so positively char- acterize that alkaloid that in a defence in which every possible and improbable theory of cross-examination which suggested itself to exceptionally well-qualified legal counsel, aided by eminent chemical counsel, was utilized, no attempt was made to question the presence of morphin. The re- actions of morphin were obtained with the ferric chlorid, the iodic acid, the Frohde, the Husemann, the Erdmann, the sulphuric acid and potassium chlorate, the ferric chlo- rid and ferricyanid, the nitric acid, and the Pellagri re- actions. The remainder of the liquid was used in an endeavor to obtain a physiological reaction, which, however, owing to the small amount of the material remaining, and to the lack of sensitiveness on the part of animals to the action of morphin, gave results which at the best could only be considered as subsidiary. When injected into the lymph pouch of a small frog, this solution produced a distinct diminution in reflex sensibility. The respiration was not markedly affected, and the animal recovered completely in about two hours. The intestines and their contents were treated separately from the stomach and apart from each other during the ear- lier stages of the process, but the residues from tissue and contents of the intestines were united for testing both for quinin and for morphin. The amylic residues were contaminated with the same cadaveric material that existed in the stomach residue, and attempts to isolate the vegetable alkaloid in a condition of absolute purity were equally fruitless. Nevertheless, very distinct reactions for morphin with all the tests above described were obtained. The weight of impure residue from the tissue and con- tents of the stomach was 0.0021 gram, weighed as chlorid; that from the intestines weighed .0086 gram under like Digitized by Microsoft® Case of Homicide by Morphin. 9 conditions, making a total of 0.0107 gram, equal to seven- teen-hundredths grain, of impure morphium chlorid. It has been shown by Bergeron and L'Hote' that water which has been agitated with amylic alcohol produces, when injected into frogs, guinea pigs, or rabbits, symptoms closely resembling those of narcotism, even when the amy- lic alcohol has been purified; and that while a frog was killed in eight minutes by the injection of one cubic centi- metre of water which had been agitated with amylic alcohol and filtered, another frog into which 0.0125 morphium chlorid was injected did not suffer. As these experiments raise a serious question concerning any value which may attach to the physiological test with the stomach, the re- mainder of the residue from the intestine, not used in applying the chemical tests, which had been reserved, was dissolved in water and used for physiological testing and for comparison of the effects of the amylic residue con- taining morphin with those obtained from another amylic residue but free from morphin, and with those caused by a solution of morphin of known strength. Three small frogs received by injection into the abdominal cavity : I., 0.009 gram of the residue left by the evaporation of the amyl alcohol which had been agitated with the acid aqueous extract from the contents of the small intestine, a quantity which would represent a much greater amount of amyl alcohol than that which was injected in Bergeron and L'Hote's experiment; II., 0.002 gram of morphium sul- phate; III., the remainder of the morphin residue from the intestines, weight not determined. The bulk of liquid in- jected was one cubic centimetre in each. A fourth frog which had been kept under similar conditions was ob- served for comparison. The results were as follows : ' C. R., Ac. d. Sc, Par., 1880, xci., 390-393. Digitized by Microsoft® lO I. Some Toxicological Points in a II. 2 P.M. : Respiration 66 ; injection 2.15 p.m. No apparent excitement fol- lowing injection. 2.35: Respiration 48. III. 2P.M.: Respiration 62 ; in- 2 P.M. : Respiration 61; in- jection 2.25 P.M. jection 2.20 p.m. Animals both very uneasy after injection ; crawl about ; stiffen up against side of bell glass when disturbed — a condition somewhat resembling that produced by a very small dose of strychnin, but much less intense and of much shorter duration. 2.35 : Respiration 69. I2.35 : Respiration 40. 2.40: All about equally sensitive to irritation, but do not respond to mechanical stimulus as readily as IV. The tetanoid movements in II. and III. have ceased. 2.45: Respiration 48. 3.00 : Movements natural jumps when struck. 3.10 : Respiration 56. 3.20 : Pupils somewhat con tracted. Point approached to eye ; eye closed before contact and immediately re- opened. Animal moves away. Withdraws leg with kick- ing motion and jumps away when toe is pinched. Animal sits in normal po- sition, part of weight on front legs. Pointed glass rod applied to coccyx ; 2.45 : Respiration 60. 3.00 : Movements slug- gish; crawls when struck, without jumping ; hind legs dragged. Lies on belly with front legs doubled under. 3.10: Respiration 63. 3.20 : Pupils not modified. Suffers contact of point with cornea ; closes eye slowly, does not open for several seconds. After second contact of point with cornea keeps eye closed nine seconds. Ani- mal does not move. Withdraw legs sluggishly not jump. 2.45 : Respiration 40. 3.00 : Movements slug- gish; crawls when struck, does not jump; hind legs somewhat better than II. 3.10: Respiration 37. 3.20 : Pupils not modified. Suffers contact with point ; closes eye slowly, does not open for several seconds. After second contact keeps eye closed two min- utes and twenty seconds. Animal does not move. when toe is pinched. Do Animals lie on belly with front legs doubled under. away, does not crawls jump. Conditions remain about the same until 4.30. At 6 p.m. animals all in normal condition. moves limbs slightly, neither crawls nor jumps! can be lifted by rod without attempting to move. These results are in accordance with those of Bergeron and L'Hote, of Trombetta,' and of others, as to the lack of sensitiveness of the physiological test; .002 gram (equal to about one-thirty-third of a grain), a quantity more than sufficient to give all the characteristic chemical reactions did not cause the death of the frog, nor any very marked symptoms beyond a very transitory period of excitement Case of Homicide by Morphin. 1 1 followed by diminution in reflex sensibility. So far as the appearances went, however, those manifested by the frog which received the suspected solution corresponded closely with those of the one which received morphin, and differed from those of No. I. except as to the respirations, which were somewhat diminished in frequency in Nos. I. and III., and not materially affected in No. II. The evidence which can be expected from the physiological test can, therefore, only be confirmatory, at the best, until some available animal is found at least as sensitive to the action of morphin as the human infant. The absence of marked symptoms in No. I. seems to indi- cate that the poisonous substance observed by Bergeron and L'Hote is volatile, and consequently not liable to interfere with the physiological test applied with the residue of evapd- ration, unless indeed the amylic alcohol used by them was imperfectly purified. The conditions as to the time which had elapsed between the death and the analysis in this case are also of interest. Toxicological authors are unanimous in the opinion that although the detection of morphin has frequently failed in cases of known poisoning by it under apparently favorable conditions,' it is by no means as prone to decomposition in the presence of putrid material as it was at one time sup- posed to be. Several instances are recorded in which it has been recovered months after contact with cadaveric and putrid substances. Orfila' obtained the nitric acid reaction for morphin from a mixture with human intestines which had been allowed to putrefy during three months. Laroque and Thibierge ' obtained the nitric acid and ferric chlorid reac- tions of morphin from extracts of mixtures with putrescible material which had been exposed for several months, and even for a year. Flandin * separated morphin " in very ponderable quantity " from a mixture with muscular tissue which had been allowed to putrefy two months. Huse- mann " is of the opinion that " even when months or a year ' See Wormley, University Med. Mag., Philadelphia, May, 1890. ' "Tox.," sime ed., ii., 295. » Galtier, "Toxic," ii., 86. < " Traite des i3igit^edit)^ilVlicrosoft® ' " Toxic," p. 620. 12 Some Toxicological Points in a have elapsed since burial, the possibility of a positive result, of the analysis is not excluded," and states that Stas detected morphin in two cases of exhumed bodies — in one case after thirteen months. Tidy ' detected morphin in considerable quantity, and also a trace of meconic acid, in the dried-up stomach of a man who had died about four months before^ Taylor Mound both morphin and meconic acid in organic liquids which had been allowed to undergo decomposition for fourteen months. Pellacani ' obtained the reactions of morphin from a mixture with calf's blood, in the proportion of one to five hundred, which had been exposed twenty- seven months. Russo-Giliberti and Dotto* detected mor- phin by the nitric acid, Frohde, and Erdmann reactions, in a mixture with Jaksch's fluid which had putrefied during seventy days. Dotto " obtained evidence of the presence of morphin by the Frohde, Erdmann, and nitric acid reactions from mixtures with putrid materials after thirty-three months. In the case under consideration the conditions for the de- tection of the poison were, so far as known, as favorable as possible, in view of the fact that the analysis was begun fifty-three days after death. The stomach had not been pumped out during life. The condition of the cadaver was by no means such as would be expected nearly two months after death. The casket and its contents were perfectly dry. The cadaver was extremely well preserved, the only exter- nal evidences of post-mortem change being that the face was very slightly darkened in color, and the eyes, mouth, and lower part of the face were coated with a white mould, which also appeared in a small patch on the left forearm. With these exceptions the cadaver resembled that of a per- son very recently dead. There was not the faintest odor of putrefaction, even when the body was opened; and the surface of the body was perfectly natural in color, even the early post-mortem discoloration of the abdomen being ab- sent. The internal organs were firm in texture, even the ' Woodman and Tidy, For. Med., Philadelphia, 1877. p. 340. " " Poisons,'' 3d Am. ed., 556. 5 Riv. sper. difreniat,, Reggio-Emilia, 1887-88, xiii., Pt. 2, 237-248. " Sicilia Med., i88q, i., 53. ' Id., 18^0, ii., 442Digitized by Microsoft® Case of Homicide by Morphin. 13 brain was sufficiently firm to permit of being removed entire and examined as to its gross appearances, and was no more softened wfien examined chemically than is usual with brains taken from recent subjects. In short, putrefaction had been practically arrested — an efiect which is. referable to the co-operation of several causes. The body had been embalmed with a solution of sodium arsenite in dilute alco- hol, by injection into the femoral artery and into the ab- dominal cavity with a trocar. It is probable, however, that the embalming liquid did not penetrate the smaller vessels, as the vascular system had not been previously washed out, and the organs, with the exception of the lungs, maintained their normal color and contained blood. The body was in a subterranean vault built of brick, cement, and bluestone, near the top of a hill, and in a soil consisting of rather coarse gravel, under conditions which insured its protection fron* moisture, and probably also the maintenance of a suf- ficiently low temperature to greatly retard, if not to arrest, putrefaction. A point of considerable interest in the case was with regard to the presence of quinin. As it was in the history of the case prior to the exhumation that morphin and qui- nin had been prescribed in medicinal proportion by the de- fendant, the analysis was directed primarily to the detection of these two alkaloids, and, if possible, to a determination of their relative proportion should both be found to be pres- ent. With every organ examined, the residue in which qui- nin would be found, if present, was extremely small in amount, only weighable in one case, and in some barely vis- ible. To avoid the possibility of loss great economy of material and the avoidance of repeated manipulations were imperative. These residues were therefore dissolved in from four to six drops of very dilute hydrochloric acid and the solution examined by two tests, one the most delicate of the chemical reactions, that with bromin water and ammo- nium hydroxid (the thalleiochin reaction); the other not so delicate, but more within the lay comprehension — the sense of taste. In the application of the thalleiochin reaction a compar- ative testing was simultaneously made with a solution of Digitized by Microsoft® 14 Some Toxicological Points in a Homicide by Morphin. quinin i : 8,000 as chlorid, using the same reagents in the same amounts. In every instance the characteristic green color appeared with the comparative solution and remained absent with an equal bulk of the solution of the residue from the organs. To apply the taste test a small strip of filter paper was impregnated with the liquid, dried at a slightly elevated temperature, and placed upon the tongue. In no case was any impression of bitterness perceived. Under similar con- ditions two drops of a solution of quinin i : 2,000 as chlorid, similarly dried upon paper, produces a distinctly bitter taste which persists for several minutes. Incidentally the liquid was examined, before the applica- tion of the above tests, for fluorescence.' It has been suggested that the quinin reaction may have failed either by reason of the conversion of quinin into one of its isomers — quinidin or quinicin — or by reason of its ab- sorption and elimination. The former supposition is not ten- able, because such conversion, if it occur in the system, does not take place in the stomach, and consequently the residue from the contents of the stomach should have responded to the tests had quinin been present; and for the more cogent reason that quinidin and quinicin respond to the thalleio- chin reaction in the same manner as quinin. To meet the second supposition — that when quinin and morphin are taken in the proportion of one-sixth of a grain of the latter to nearly five grains of the former, the quinin may be entirely eliminated and a portion of the morphin re- main — I should have wished to offer experimental evidence. This was not done, however, for the reason that it is impossi- ble to experimentally fulfil the conditions actually existing or supposed to exist in this case, or to approach sufficiently near to the conditions of ante-mortem administration in a human subject, death twelve hours later, embalming and excep- tional preservation of the body for fifty-three days, to ren- der the results of such experiment " relevant and material." In the course of this analysis several points of toxicologi- cal interest were incidentally investigated, some of which will form the subject of later papers. ' See paper on quinin reactions, tost Digitized by Microsoft^ THE USE OF AUTOMATIC AND OTHER MODELS IN TEACHING PHYSIOLOGY. W. OILMAN THOMPSON, M.D. The secret of success in teaching a scientific subject lies in the abundant illustration of established facts, and the fundamental principles of physiology are based upon many well-known chemical and physical laws which admit of liberal illustration. No doubt the most thorough method for students to learn physiology is by experimental labo- ratory work, but in large medical colleges, and certainly in public schools and academies, this is not altogether feasible. In medical colleges it is still customary to teach physio- logy by didactic lectures to large audiences. It is not practicable, nor is it desirable, to take several hundred men into a laboratory and let them experiment upon the lower animals. Very few medical students expect or desire to become expert physiologists, and their time is already fully occupied in the laboratory training in the more practical work of urine analysis, histology, pathology, etc. The so- called courses in " practical physiology " offered in many medical schools are really laboratory courses in histology or physiological chemistry, and not of operation upon living animals at all. For the latter is required special surgical technique and the use of elaborate and delicate recording apparatus, which comparatively few students have sufficient leisure or ability to master. There are, however, many facts in physiological science which can as well be demon- strated to several hundred students simultaneously as to one. The number of experiments which can be effectively seen more than ten feet awey is limited, hence many very in- structive demonstrations upon living animals — such as stim- ulation of peristaltic movements, the effect of stimulation Digitized by Microsoft® i6 The Use of Automatic and Other of the chorda tympani, etc. — are more than useless in the presence of a class of several hundred. It is with the pur- pose of illustrating upon a large scale some of the more important facts and observations in physiology that I have devised a number of automatic or mechanical models, and the object of describing them is to show that with very limited expenditure, and even with limited mechanical skill, it is possible to make many extremely serviceable and striking illustrations for the lecture room. I have practically tested the value of such demonstrations for several years, and my experience with classes in physiology has convinced me that such models are a most valuable aid. 1 frequently receive letters from teachers of physiology in academies or common schools, asking where they can obtain charts, dia- grams, or apparatus to illustrate their work. The models which are herein described are for the most part so simple in construction that they can be made by any teacher who is willing to expend a little time and trouble, and the interest which they awaken in students is an ample reward. Most of them are entirely novel, but a few are cheaper adaptations of more complicated and expensive apparatus. The proper place of models in teaching should be con- stantly borne in mind. It is easy to make them so large as to be merely grotesque, unless their relation to the whole organ, of which they may represent but a part, is clearly ex- plained by description and reference to diagrams, and, when possible, to the life-sized natural object. Bearing this in mind, it is my design always to reproduce an object in three dimensions of space, rather than by the flat surface of a diagram alone. Take, for example, the diagram of the large nuclei near the base of the brain as they appear in sections made at various levels in any half-dozen different text books. It is always bewildering to a student to seek to acquire a correct idea of their shape and exact position, and reconcile one diagram with the next; but simple models enlarged, and representing the nuclei entire, can be used to explain the level at which the various sections are made, and their appearance in any position; then the nuclei models may be placed within an enlarged brain model and their relation to the adjoining parts is readily explained. Students are Digitized by Microsoft® Models in Teaching Physiology. \y apt to tire of diagrams. They either think that they have them all in their text books at home, or they fail to under- stand them because the relation to other regions of the body is obscure, or the diagrams are not made upon a uni- form scale of enlargement with one another, or they repro- duce only one surface of an object. The special use of automatic or working models, moreover, is to illustrate the mechanism of certain physiological processes which cannot be as readily shown by other means. THE MATERIALS USED. The first requisite for large models is lightness combined with durability. After experimenting with various kinds of papier-mach6, modeller's clay, wax, etc., I have devised a very simple material which is moulded with great ease and which quickly becomes exceedingly hard and firm, while its light weight and great cheapness are special advantages. Wire cloth or " wire gauze," such as is sold for fly screens at any hardware shop, may be had in large sheets or rolls at two cents a square foot. It is easily bent into any form, which it retains, and may be readily cut and trimmed with a stout shears and tacked down by double-pointed tacks to a supporting board. It is then covered on both sides with strips of newspaper, one or two inches in length, which are dipped in very thick carpenter's glue. The glue pene- trates the meshes of the wire cloth and holds the papers together on both sides. The thickness and strength of the model may be indefinitely increased by adding successive layers of paper and glue. Time is saved by painting the glue over the surface in thick layers with a brush and rap- idly laying the strips of paper upon it. The last layer only need be laid on with any special care to secure smooth- ness of surface. After the glue is thoroughly dried, one or two thick coats of paint, of any desired color, are applied and the whole is shellacked. When finished the model be- comes as hard as board, and it possesses great advantages over papier-mach6, which is more expensive, and usually brittle unless subjected to great pressure. This new compo- sition is smooth and very hard, water-tight (for cold water), it never warps, breaks, nor cracks, and when painted it is 2 Digitized by Microsoft® 1 8 The Use of Automatic and Other difficult to believe that it has been made of such cheap ma- terials. A great variety of models illustrative of physio- logical anatomy may be formed from it. For brevity of description hereafter I shall refer to this material simply as "wire paper." Clay is useful for modeUing the brain, show- ing its development and structure, but it soon cracks as it dries ; and plaster reproductions of clay models are rela- tively difficult and expensive to make, besides being heavy. Where great delicacy in moulding a surface is required, clay, putty, or modeller's wax may be used and then covered by layers of paper and glue without the wire gauze. When dry, the clay may be scooped out from the interior, leaving a hard paper shell, very light in weight, which is strength- ened by a dozen or two additional layers of newspaper and glue. MODELS OF THE NERVOUS SYSTEM. Reflex Action. — A simple working model to illustrate reflex nei .'c action can be operated by compressed air. It repre- sents motor, sensory, and inhibitory nerves, a ganglion cell, and a sensory cell. Take a board one and a half by three feet. At each of two diagonal corners fasten a small upright block bearing on its face a cardboard or feather arrow pivoted on a pin. Arrange at the shaft of each arrow a piece of rubber tub- ing so that a puff of air forced through the tubing will de- flect the arrow. Join the two rubber tubes in the centre of the board by a glass T-tube. To the long arm of the T-tube attach a third rubber tube to represent the afferent nerve. The T-tube is concealed beneath a polygonal ganglion cell moulded out of the "wire paper." A common bellows sends a current of air through the afferent nerve (tube). This cur- rent, on reaching the T-tube, divides and deflects the arrow S (sensation) and the arrow M (motion). A fourth rubber tube, I, serves for inhibition. It conveys two strings, one of which is tied around the motor tube, the other around the sensory tube beneath the ganglion cell. The free ends of the strings are brought out behind the block S. At the mo- ment of expected deflection of the arrow M pull the string passing around the motor tube ; it is thereby compressed and motion is ^^hibked. j^l^sen|a^^n alone results. Or, Models in Teaching Physiology. 19 at the moment of expected deflection of the arrow S, pull the string passing around the sensory nerve (tube), and mo- tion alone results. By manipulating the free ends of the strings, retarded motion, retarded sensation, or complete in- hibition may be illustrated. By varying the force of the compressed-air current slow or quick motor and sensory re- sponses are shown. The Brain. — A hollow model of a hemisphere of the hu- man brain, three or four feet long, is made from the " wire Fig. I.— Model to illustrate Reflex Action. I, path of inhibitory impulse. M, motor dial; S, sensory dial; paper." In the photograph, Fig. 2, an actual human hemi- sphere is shown, for comparison, at I. In this photograph unavoidable foreshortening, due to the large size of the model and peculiar reflection from its sur- face, distorts the fissures and convolutions, which are care- fully outlined in paint upon the model. A large number of layers of paper and glue make the model extremely hard and firm. The lobes, various landmarks, and motor areas are drawn in chalk on the external surface before the class, and Digitized by Microsoft® 20 The Use of Automatic and Other can be subsequently rubbed out. The mesal surface is made from a single piece of "trunk-board," i.e., the most rigid kind of pasteboard, on which the typical diagram is painted, representing the mesal fissures, convolutions, and corpus cal- losum with its rostrum and genu. This is not shown in the photographs. After this is de- monstrated the painted " trunk-board " is removed, like the side of a box, and within are seen the lenticular and caudate nuclei (Fig. 3, I), the thalamus (Fig. 3, II), and the corpora quadrigemina (Fig. 3, III). The convolutions about the fissure of Sylvius (Island of Reil) stand out clearly at Fig. 3, V. The mottled appearance in the photograph of the con- cave surface of this hemisphere is due to the fact that the Fig. 2. — Model of a Cerebral Hemisphere. (Outer surface of Fig. 3.) I, human hemisphere, life size, for comparison. motor, sensory, and special sense areas are painted within it in various colors. Ortho-chromic negative plates were not used, hence the reds and yellows appear very dark. The nerve fibres from the medulla, etc., are represented by strings passing up from beneath, between the several gan- glia or nuclei, and radiating out to the cortex. The strings are colored window cord, arranged to match the painted cortical areas to which they severally pass. The great un- mapped frontal area is clearly shown. The nuclei are modelled first in clay and then covered by paper and glue. The clay is next scooped out through a perforation which is subsequently closed ; the paper and glue are thickened by successive layers until the nuclei are as hard as wood but very light. Holes are bored through and between them with a gimlet for %^a|^^;^^^^ds. The motor and Models in Teaching Physiology. 21 other areas being localized upon the outer surface, the model is turned about, the mesal surface is removed, and the inner surface of the cortex is shown to the class, with its radiating fibres. Association fibres are omitted, to avoid indistinctness from having too many cords. They may be shown during the lecture by looping cords from one corti- cal area to another. The operculum and speech centre are particularly well understood from this model, It is used also to show the effect of localized lesions at the cortex which affect limited centres, in distinction from lesions near the ganglia, which include so many and diverse fibres. It Fig. 3. — Model of the Interior of a Cerebral Hemisphere, showing the nuclei, corona radiata. motor, sensory, and special-sense "centres" in the cortex. (Inner surface of Fig. 2.) I, caudate nucleus; II, thalamus; III, two of the corpora quadrigemina; V, operculum. is well to make a duplicate set of nuclei and thalami, so that they can be turned over in various positions to show the appearance of sections cut at different levels. The strings may be fastened to small wire rings pasted by tape to the cortex (such can be bought at a stationer's ready for past- ing), or they may be passed through small perforations to the external surface and their ends covered by final layers of paper and glue. They are readily kept taut by pulling upon the bundle of cords beneath the ganglia. If desired, the cortex mav be covered by a close-fitting Digitized by Microsoft® 22 The Use of Automatic and Other jacket of canvas or sacking painted with fissures and repre- senting the cranial bones. The landmarks for trephining being pointed out to the class, a circular flap may be cut out of the canvas with scissors, representing a trephine opening in the skull. The convolutions appear below, and the hemisphere may be examined inside to ascertain if the trephining was performed over the correct area. This brain model possesses decided advantages over the imported open-wire models. It is larger and it does not at- tempt to show too much at once. Tne nerve fibres are few and typical, the centres are distinct, the cortex is better represented, and the materials cost but a few dollars. Fig. 4. — Model of Median Perpendicular and Horizontal Sections of the Brain, showing nuclei, corona radiata, and face, arm, and leg "centres." I, sections through the tempero-splienoidal lobes, arranged to fold under III by hinges; II, median perpendicular section through a natural human brain for comparison of size; III, nuclei. Another very instructive brain model, designed to show the relations of the nuclei to median perpendicular and hori zontal sections, is made as follows : the outlines of two such sections are copied on tracing paper from Dalton's or Luys' photographs of brain sections. The tracing paper is then put in any projection lantern and magnified upon a screen to a diameter of four feet. This accurately enlarged outline is next copied on a piece of " trunk-board," and the periphery is cut out with a saw. The oarts 'of the perpen- *^ ^ Digitized by MicrosonW ' ^ *^ Models in Teaching Physiology. 23 dicular section which project below the horizontal section (Fig. 4, I) are hinged with strips of canvas, glued on, so that when not in use they are folded against the under surface of the horizontal section, and the model lies flat. The nuclei (Fig. 4, III) are turned out of wood on a carpenter's lathe, and are glued firmly to the sections. Long strips of "trunk- board," an inch wide, are bent in the form of hoops to sug- gest the outlines of sections through the cortex at various levels, and to these hoops colored strings are tied to repre- sent radiating fibres which converge toward the nuclei and internal capsule. The thalami — also turned out of wood — are placed in position. Additional strings may be adapted during the demonstration of this model to represent con- FiG. 5. — Model to illustrate the Distribution of the Columns of the Cord in the Medulla. I, section through the gray matter of the cord; II, lateral columns of the cord; III, peduncles of the cerebrum; V, V, peduncles of the cerebellum. necting fibres. The two " trunk-board " sections are painted to imitate the gray and white matter, and various colors are added to mark off the centres for movement of muscles of the -face, arm, and leg, etc. In the photograph (Fig. 4) a median perpendicular section of a normal human brain is represented at II, to give an idea of relative size. The Medulla. — The decussation and other distribution of the fibres in the medulla is always a difficult subject for stu- dents to understand. To facihtate this study a model is constructed as follows : two pieces of board, each a foot square, are cut with a saw to represent the outlines of a cross-section through the gray matter of the upper cervical Digitized by Microsoft® 24 The Use of Autotnatu and Other cord. The boards are held about fifteen or eighteen inches apart, with their flat surfaces parallel, and a strip of wire cloth is tacked to their margins. In this way a wire box is made, having a wooden top and bottom, and, when covered by paper and glue and painted, it represents a section through the gray matter of the cord (Fig. 5, I), showing the anterior and posterior horns and commissure. The columns of the cord are made of parallel bundles of twine laid in the grooves formed in the sides of the gray box (Fig. 5, II). The deeper bundles are glued to the gray matter (box) to keep them in position. The bundles of twine are continued to form the peduncles of the cerebrum (Fig. 5, III) and the peduncles Fig. 6.— V, Model of the Floor of the Fourth Ventricle. VI, IV, Models of Transverse Sections through the Medulla, showing deep nuclei of certain cranial nerves. I, respiratory centre; II, posterior columns from the spinal cord. of the cerebellum (Fig. 5, V, V). They are made of the white cotton twine that is sold in hardware shops for mops, and they can be easily dyed of different colors to represent the anterior, lateral, and posterior columns respectively. These fibres can be readily arranged to show the decussa- tion in the motor tract in the medulla. On examining one of the cerebral or cerebellar peduncles, it is easy to see the colors of the bundles of twine of which it is composed, trace them back, and find whether they decussate or not, and along which columns of the cord they run. Another model of the medulla (Fig. 6, V) may be used in Digitized by Microsoft® Models in Teaching Physiology. 25 connection with the preceding one. It represents the floor _jf the fourth ventricle. It is formed of a board on which the " wire cloth" is moulded to represent the median groove (Fig. 6, I), the eminence in the floor of the fourth ventricle, and also the corpora quadrigemina. The fibres forming the lateral and antero-lateral boundaries of the ventricle are made of unravelled clothes-line or of more expensive win- dow cord (Fig. 6, II, III). The respiratory centre and deep origins of the cranial nerves arising from beneath the floor of the fourth ventricle are indicated by colored paints upon the surface. It is desirable also to have several cross- sections of the medulla (Fig. 6, IV, VI), made of soft board and painted to represent the deep nuclei of the various Fig. 7. — Model showing the path of voluntary motor impulses from the cortex to muscles at I. nerves. These sections can be placed perpendicular to the model of the floor of the ventricle, to show where the dif- ferent cross-sections are made, which in the text books give the appearance of the medulla so much variety, dependent upon the level at which they are cut. The valve of Vieus- sens may be cut from paper and shown in position over the model of the floor. This model is especially useful for show- ing the positions of the numerous nerve centres which are crowded together beneath the floor of the fourth ventricle. Motor Nerve Model. — The path traversed by a motor im- pulse from the cortex to the muscle is demonstrated by the following model : a board eighteen inches in diameter is sawed and painted to represent the outline of a typical cross- section of the spinal cord (Fig. 7). Two stiff wire rods are Digitized by Microsoft® 26 The Use of Automatic and Other screwed into the board in the motor path in the lateral col- umns. They are crossed above to represent the decussation at the medulla, and are capped with nerve cells (made of " v^ire paper " or wax) to represent their termination in the cortical gray matter. Towards the periphery the motor path is con- tinued by strings, which pass from the rods through large Fig. 8. — Working Model of the Spinal Cord, showing the origin of the spinal nerves, typical spinal cord segments, and the distribution of the white fibres in the cord. I, a normal human spinal cord for comparison of size. nerve cells (made of " wire paper" or wax) in the gray mat- ter painted on the board, outward to terminate in ganglion cells of "wire paper" resting upon small blocks of wood to- represent, diagrammatically, striped muscle fibres (Fig. 7, 1).. Digitized by Microsoft® Models in Teaching Physiology. 27 On this model may be pointed out the effect of lesions in va- rious situations above or below the decussation. It serves merely to isolate the motor path and impress its course upon the mind. A similar model may be made to illustrate the sensory tracts. Working Model of the Spinal Cord. — This model illustrates many of the laws of reflex action, and may be used to show the anatomy of the cord as well, and the result of lesions at any part. It is about seven feet high, and a life-sized model of the human cord is placed beside it in the photograph (Fig, 8, 1) for comparison. The idea of this model occurred to me on seeing a familiar diagram of the cord in Landois and Stirling's Physiology, and I have found it far more instructive than many diagrams of the cord. The model represents typical segments of the cord in the lumbar, lower dorsal, upper dorsal, and cervical regions, with typical groups of fibres running between. Of course the width of the model is necessarily out of proportion to its length, but this in no way affects the clearness of the de- monstration. The model is thus made : a piece of a half-inch gas pipe is screwed perpendicularly through the centre of a thick board, two and a half by one and a half feet, that serves as a support for the pipe, which is held by a large nut on the under side of the board. This pipe represents the median canal of the cord. Four flat boxes are made to represent the four typical segments of the cord. Each box has a wooden top and bottom, sawed to represent the outline of a segment of the cord, with anterior and posterior fissures. The boards have a central perforation to allow them to slip over the gas pipe, to which they are fastened at appropriate heights by wires, or such clamps as are sold by dealers in chemical wares to be attached to iron rods. They also have numer- ous smaller perforations, through which colored strings pass to represent bundles of perpendicular nerve fibres. Th^ sides of each are formed by " wire cloth " tacked on to the wooden top and bottom of the box. The sides are perforated to allow of the outward passage of groups of strings repre- senting the anterior and posterior nerve roots. The ganglion for each posterior root is made of cotton. On the surface of each box are painted, diagrammatically, the gray matter and Digitized by Microsoft® 28 The Use of Automatic and Other the areas of the different columns of Turck, Goll, Burdach, €tc., through which areas pass perpendicular strings of cor- responding color. Other strings laid horizontally on the surface of the segments represent decussations, continua- tions of the root fibres, etc. Above the cervical segment the strings are arranged to exhibit decussation of the motor tract in the medulla. At the level of the lumbar region the anterior and posterior roots are not joined, but the strings of the posterior root pass inside of the lumbar box, run over small pulleys, and join the strings of (i) the motor root on the same side, (2) the motor root of the opposite side, (3) the motor roots of the cervical segment, and, finally (4), a string passes straight upward and through the sensory tract of the opposite side and terminates in a movable arrow fastened to a wire at the top of the model, indicating the further trans- mission of sensory impulses onward towards the brain. The motor roots are stiffened with wire, so that pulling upon any sensory string of the lumbar segment causes the corresponding motor root to move. With this model the following facts may be clearly de- monstrated : I. Take the sensory bundle of fibres in the hand and irritate one of them [i.e., pull the string) : a reflex of the leg on the same side occurs (movement of the motors root). 2. Irritate more strongly (pull another string) : the reflex passes to the opposite leg. 3. Irritate more strongly still, and motion occurs by irradiation to a motor root of the segment above. 4. Finally irritate all the fibres (pull all the strings), and general convulsion occurs. 5. Pull the string running up to the arrow, and sensation results. 6. Sensa- tion may be made to occur before motion or after it, and motion or sensation may be produced with any desired ra- pidity or mutual relation. 7. Take a large piece of card- board and bend it around any group of nerve fibres, to show - ^hat takes place when they are injured or excised. The effect of lateral, antero-lateral, and other incisions of the cord may be thus distinctly described. 8. The effect of cut- ting the anterior or posterior roots on the proximal or dis- tal side of the ganglion may also be explained on this model. 9. The model gives a very fair idea of the manner in which the fibres run through the cord, join the nerve roots, decus- sate, etc. Digitized by Microsoft® Models in Teaching Physiology. 29 CLAY MODELS. A useful demonstration of the motor and sensory tracts in the spinal cord is made as follows : a large flat cake of modeller's clay, one inch thick and two feet wide, is cut out before the class to the shape of a typical transverse section of the cord. The insoluble dry powdered pigments used by painters may be applied to the surface to mark off distinctly the different columns of the cord ; the clay itself furnishes- the natural gray. Colored straws are stuck into the clay to represent the fibres of the longitudinal columns (Goll, Bur- dach, direct cerebellar, etc.) ; colored strings represent the root fibres and their distribution within the spinal cord. The same materials are very useful for modelling before the class the development of the brain and the growth of its em- bryological forms. A section of the cord may also be made by sawing out a thick block of wood, and the several col- umns may be cut out and painted, and then put together again like a dissected map. AUTOMATIC GLAND MODELS. Certain facts in regard to a typical gland and its mechan- ism are illustrated as follows : the typical gland is tubular and consists of epithelial cells, basement membrane, vessels, and a regulating nerve supply. The relations of blood pres- sure and obstruction to the rate of secretion are shown, and the fact that the secretion differs from the blood in com- position and reaction. Upon a board three feet by two feet is hollowed out with a mallet and gouge the outline of (Fig. 9, V) the gland with fundus and a neck opening on the free surface at Fig. 9, 11. The depression thus made is lined by cotton lampwick unravelled, to represent connec- tive-tissue fibres of basement membrane, upon which rest small columnar sponges tacked to the board to keep them in place. The sponges represent columnar gland cells.. Beneath the sponges a piece of lead gas pipe is concealed,, bent like a horse-shoe. With an awl a small hole is bored in the pipe beneath each sponge. The vein and artery (Fig. 9, IV) are made of rubber tubing and terminate in a capillary plexus of rubber tubing joined by glass T, Y> and Digitized by Microsoft® 30 The Use of Automatic and Other U tubes. One of the rubber capillaries joins the lead pipe in a concealed position beneath a cell (sponge). From a pressure bottle arterial blood (water colored with a solution of phenophthalin rendered slightly alkaline) flows to the capillaries (Fig. 9, V). It is seen coursing through the glass tubing. Part of this red fluid returns by the rubber vein (Fig. 9, IV) to the bottom of a cylindrical glass jar (Fig. 9, 1). In the bottom of this jar are a few drops of acetic acid and two or three drops of a strong aqueous solution of indigo Fig. 9. — Automatic Model to illustrate the Mechanism of Gland Secretion. Ill, pressure bottle for arterial supply; IV, vein and artery; V, capillary plexus surrounding the fundus of the gland; I [, receptacle for the secretion; I, recep- tacle for the returned venous blood; VI, ganglion cell with afferent and efferent nerves. blue. The red fluid is at once decolorized by the acid and promptly re-colored by the indigo blue ; thus blue venous " blood " is seen to rise in the jar (Fig. 9, I). The rest of the red arterial fluid flows into the concealed lead pipe and spurts out in a tiny jet beneath each sponge. Under each sponge a few grains of powdered tartaric acid are sprinkled. As the Digitized by Microsoft® Models in Teaching Physiology. 31 alkaline fluid is absorbed by the sponges it is at once decolor- ized and made acid by the tartaric acid, and the sponges swell and fill the lumen of the gland, and soon a colorless secretion begins to trickle out of the neck into a glass dish {Fig. 9, II). The nerve mechanism is diagrammatically repre- sented by a large ganglion cell (Fig. 9, VI) of " wire paper." An afferent impulse comes to the nerve cell from irritation of the free surface at Fig. 9, II ; another may be brought in from distant centres. An efferent impulse travels to the vessels through vaso-motor nerves, and another to the gland cells directly, as indicated by the arrows. The reaction of the red arterial fluid is shown to the class with litmus and found alkaline, while the clear, colorless secretion of the gland is found to be acid. It is thus impressed on the student that chemical changes have actually occurred in the gland cells and that secretion is not exclusively filtration. It is easy to see that by raising and lowering the stand upon which the pressure bottle (Fig. 9, III) rests, variations in the rapidity of the secretion will occur. The effect of pressure upon the vein, artery, or capillaries may also be demonstrated. This model I have found very serviceable to illustrate many fun- damental principles in regard to a typical gland secretion which apply equally to nearly all glands. The changes in the red fluid to blue and to white make a very striking experi- ment. There are many aniline and other pigments which are of great service in demonstrations of this kind. Congo-red solution, which changes to deep blue by altering the reac- tion, is one of the most useful. To show dialysis I fre- quently employ a deep blue aqueous solution of litmus on one side of the membrane with acid on the other. The gradual change of color of the litmus to bright red is visi- ble at a considerable distance if an appropriate background is furnished. The materials of this gland model are so cheap — a sponge, a few feet of tubing, a piece of lead pipe, a glass bottle and jar, a bit of wire cloth, and cords — and its construction is so easy, that it is within the reach of any teacher and will be found extremely useful. Automatic Kidney Model. — The mechanism of urine secre- tion is illustrated on the same plan with the preceding model of a typical gland, only the urine, instead of appearing as a Digitized by Microsoft® 32 The Use of Automatic and Other colorless secretion after leaving the gland, is re-colored of an amber hue by a few drops of fustic placed in the bottom of the receiving jar (Fig lo, U). The model presents a Malpighian corpuscle with afferent and efierent vessels, a venous plexus, convoluted tubule, loop of Henle, and collecting tube. The Malpighian cor- FiG. 10. — Automatic Model for the Mechanism of Urine Secretion. A, tube bringing arterial fluid from the pressure bottle to the Malpighian corpuscle; V, receptacle for venous fluid returned from the plexus about the convoluted tubule; U, receptacle for the urine. puscle is made from a cocoanut split in half (Fig. lO, A) and painted. I have tried papier-mach6 and other materials, but nothing is so simple, inexpensive, and durable as this. The cocoanut conceals a glass T-tube. The red arterial fluid Digitized by Microsoft® Models in Teaching Physiology. 33 from the pressure bottle (Fig. 10, A) enters this T-tube and the stream divides. A part passes through the convoluted tubule, loop of Henle, etc. (made of painted rubber tubing of various sizes, with glass tube joints), is received in the jar (Fig. 10, U), decolorized by acid, and then colored yellow by fustic. The remaining part passes out of the efferent vessel and is seen flowing through the bent glass tubing which forms the capillary plexus, returning as venous blood (indigo solution) to the jar (Fig. 10, V). The change in reaction from alka- line blood to acid urine is shown by litmus. As the quantity of urine increases it becomes less strongly acid and paler (since only a limited quantity of acid and fustic were added to the jar). The jar (Fig. 10, V), having been allowed to fill with blue blood, is emptied and replaced, when it immedi- ately refills with red arterial blood, thus emphasizing the fact that in coursing through the kidney the arterial blood does not become especially venous, but changes in other ways by parting with materials that form the urine. The effect of increased or diminished arterial pressure is shown by raising or lowering the pressure bottle. Venous or arte- rial obstruction, or obstruction in the tubule (casts), all give striking results, as seen by the quantity of fluid discharged. Large Kidney Model. — In order that the exact relations in the kidney of the preceding automatic secretion model may be understood a large model of* the entire kidney (Fig. 11) can be made. It is constructed of the "wire paper," and it represents a kidney with successive layers dissected away, finally show- ing the individual systems of tubules, which are formed of rubber tubing. Th.e Malpighian corpuscles are formed of English walnuts, smoothed and painted. The renal vein (Fig. II, II), artery (Fig. 11, III), and ureter (Fig. 11, V) are formed of "wire paper." The hollow pelvis and calices are plainly shown, with the orifices of the (rubber) tubules open- ing at the papillae. The smaller blood vessels are made of graduated rubber tubing with glass joints, ending in cords at the glomeruli, all being painted appropriately to repre- sent the course of venous and arterial systems. In the pho- tograph a human kidney is shown for comparison at Fig. II, I. 3 Digitized by Microsoft® 34 The Use of Automatic and Other Automatic Micturition Mode/.— The principle of this model resembles that of the two preceding. The model (Fig. 12) represents the two kidneys with their renal veins, arteries, and ureters, together with the bladder, lower abdominal wall, and spinal cord centre, with nerves for control of micturition. The kidneys are made of "wire paper." Within each kidney is a glass T-tube. The ar- terial blood (red phenophthalin solution) flows from the pressure bottle (Fig. 12, A) to both kidneys. It returns in part through the renal veins to the glass vessel (Fig. 12, V). Fig. II. — Large Kidney Model. Ill, artery; V, ureter. I, human kidney, natural size; II,^vein; The rest of the fluid flows through the remaining arm of the T-tube into the ureter and thence to the bladder. The bladder is a rubber bag (like a toy balloon, but made of better rubber). It becomes distended with urine and is gradually seen to rise above the rim of the pelvis and push forward the abdominal wall, which is made of several super- imposed layers of sheet rubber tacked on to the supporting board. In the neck of the bladder a rubber tube is inserted, representing the urethra. From behind the supporting Digitized by Microsoft® \ Models in Teaching Physiology. 35 board, through a small aperture, this urethra is kept closed by a clamp. On releasing the clamp, the bladder being full, the urine flows in a vigorous stream from the urethra; the bladder is seen to contract and recede below the abdominal wall, and the latter also contracts and presses upon the blad- der. The clamp is then re-applied and the bladder refills. The fluids used are identical with those mentioned in the Fig. 12. — Model to illustrate the Mechanism of Micturition. A, tube for arterial fluid flowing from pressure bottle to kidneys ; V, receptacle for venous blood returning from kidneys ; U, receptacle for urine, with urethra, above which are seen the bladder and abdominal wall. The arrows indicate the direc- tion of nerve current to and from the micturition " centre" in the spinal cord. two preceding models, and pressure and obstruction effects are demonstrated in like manner. It can be shown that the urine is a continuous sec^tLon^^,j(,bj^ton^^^^ variations in quantity, which are in part depenaent upon the absolute or 36 The Use of Automatic and Other relative pressure of the blood vessels. Incontinence from retention may be demonstrated by using a weak clamp, which yields when a certain intra-vesical pressure is at- tained. Suppression may also be demonstrated. The nerve mechanism is shown by white strings extending from the micturition centre in the lumbar region of the spinal cord to the fundus and neck of the bladder, the abdominal walls, and upward towards the brain, etc. Hasmoglobinuria,- with alkaline urine, is shown by emptying the first vessel of urine and allowing the red arterial fluid to run directly through the model without being decolorized. In models of this kind so little acetic acid is required to change the reaction, and so little fustic or other concentrated pigment is required lo change the color of the fluids used, that these substances, when added to the bottles or jars before entering the lec- ture room, are not visible a few feet away, and the illusion produced by rapid changes in color and reaction is complete. A LIVER LOBULE. Fig. 13. — Large Model of a Lobule of the Liver. Interior shows transverse section of central vein and its capillaries. Exterior shows branches of intra- lobular vein, hepatic artery, and branches of bile ducts. It is extremely difficult to give a correct conception of a lobule of ^ the liver by ordinary charts or diagrams. The majority of physiological text books fail in the attempt. I iA simple liver lobule model is easily made (Fig. 13) at a small cost. A u J 4. t Digitized JiyMicrosoft® . , , A board two feer m aiatneter is cut m the form of a Models in Teaching Physiology. 37 hexagon or an octagon, and a wall of " wire paper" eighteen inches high is built up around it and nailed to the periphery of the board. " Wire paper" tubes of various sizes are then made. One is fastened perpendicular to the centre of the board (the central vein of the lobule), from which capil- laries radiate out towards the periphery, where they anasto- mose with tubes leading in from the branches of the portal vein, seen ramifying over the surface of the lobule. The bile duct branches (painted another color) and the small hepatic arterial branches are also seen on the surface of the lobule. A number of polygonal wooden blocks are painted to represent the liver cells, and they may be packed within the lobule between the capillary tubes. They are removed at will to show the grooves in their sides and angles where the first bile channels and the capillaries respectively lie. By turning this model in various directions toward the au- dience the relations of the text book flat surface drawings are made much mofe clear. MODELS FOR RESPIRATION. So many illustrations of the phenomena of respiration can be shown upon the living etherized animal at a distance, that models are less helpful for this branch of the subject. With the aid of tambours, pneumographs, etc., applied to the outside of the chest wall, and of manometer tubes placed within it, the movements of respiration under varying con- ditions may be magnified to any desired extent. Yet some of the physics 'Of respiration may be shown by the use of large bottles, in which the glass bottoms are replaced by a sheet of rubber cloth tied tightly to the bottle. To the centre of the rubber a cork is fastened, which serves for making traction and drawing down the rubber diaphragm. The lungs are represented by two rubber toy balloons tied to a Y-tube which runs through a cork in the neck of the bottle (Fig. 14, II). Traction on the diaphragm causes air to rush in at the Y- tube and expand the lungs. A perforation may be made in the side of the bottle, which is fitted by a cork that admits a glass tube to which a third rubber balloon filled with colored water is a.ttached. .This balloon represents the DigifizeaDy Microsoft® 38 The Use of Automatic and Other heart. The Y-tube joins a manometer. Traction on the rubber diaphragm may be shown to affect the pressure within the heart as well as the lungs (Fig, 14, I). The effect of opening the chest wall is shown by a similar bottle (Fig. 14, III), with a lateral perforation closed by a cork through which a glass tube passes. This glass tube is closed by a rubber tube slipped over it and compressed by a clamp. Only one lung (rubber balloon) is required. Trac- tion on the diaphragm inflates the lung when the lateral opening in the chest wall (bottle) is closed by the clamp. If the cork be removed air enters between the lung and the glass chest wall, and the lung remains collapsed. If the cork be replaced and the clamp be opened, air rushes in Fig. 14. — Models for Respiration. I, to show relations of pulmonary pressure to the heart; II, to show the inflation of the lungs by a rubber diaphragm; III, the same as II, but with a perforation in the bottle for the admission of air or fluid. through an opening smaller than the calibre of the tube at- tached to the lung (representing a bronchus), and the lung partially expands (Fig. 14, III). There is nothing very new in this kind of model, there- fore but few suggestions are offered here. Any dealer in chemical apparatus supplies ready-made bottles of the shape and size required. The fallacy of all such models must be made clear to the student, namely, that the chest wall is represented by a rigid glass. This, however, does not materially affect certain of the physical problems involved, so long as the '^^"^^t^'^hfmf&rCm^ sufficiently. Models in Teaching Physiology. AN AUTOMATIC HEART. 39 This model is designed to show the relations which the opening and closure of the heart valves bear to the move- ments of the ventricular walls under conditions of varying rhythm, as well as normally. The apparatus is far more difficult to make than any of the other models, yet with a little patience, and assistance Fig. 15. — Automatic Heart Model. I, auricle; II, heart wall; III, median ventricular septum; IV, aorta with sinuses of Valsalva, and aortic cusps closed; V, mitral valve open. from a mechanic or clock-maker, it can be duplicated. A stout board, about four and one-half feet square, separates the heart in front from the machinery behind which moves it. The from I constructed myself. The machinery was made under my direction by a very ingenious mechanic, Mr. R. D. Gray, who also constructed for me in 1885 a Digitized by Microsoft® 40 The Use of Automatic and Other new form of camera with which 1 was enabled to take in- stantaneous views of the beating heart, at the rate of six in one second, upon one plate. I explained to him the move- ments of the valves and ventricular walls as I desired to re- produce them, and he worked out the mechanical details with remarkable accuracy. ' Fig. i6. — Automatic Heart Model. (Rear view of Fig. 15, showing clockwork.) V, perpendicular shaft worked by crank attached to clock; I, toggle joint; II, slit in board, perforated by the wire which supports and moves the heart wall in front (see Fig. 15); III, weights balancing the cusps in front, to which they are connected by a wire working on a pivot. Description of the Model: The Frojit. — The auricles are rep- resented as closed, and they are stationary. The auricles (Fig. 15, I), median ventricular septum (Fig. 15, III), papil- lary muscles, puligp^^r^^^^^^gg^l^orta (Fig. 15, IV) are Models in Teaching Physiology. 41 made entirely from " wire paper." The ventricles are repre- sented as cut open in a vertical plane, so as to expose their two cavities and also show all the valves cut in sections through their cusps (Fig. 15, V, etc.). The cusps of the valves are composed of small steel springs bent nearly straight, and made wider and stifier by being reinforced by common tape dipped in shellac. The springs enable the cusps to "give " a little when the valves are slammed together. All the valves open and shut in proper rhythm. The ventricu- lar walls are also movable. They are made of two stout brass wires bent to the proper curvature, to which are fastened wide-jointed pieces of pasteboard. This gives both strength and lightness to the large walls, three feet high. The Back (Fig. 16). — The motive power is a Waterbury clock, reinforced by the spring of a music box to give a little more power. (Two clocks would do as well.) The clock is wound up, and in revolving it causes a central straight metal rod (Fig. 16, V) to work up and down like a shaft. To this shaft are fixed toggle-jointed metal rods (Fig. 16, I), which are fastened through a narrow slit in the board (Fig. 16, II) to the two wires supporting the walls of the ventricles in front. When the shaft (V) ascends the toggle-joints (I) are pushed outward and carry with them the wires (supporting the ventricular walls) in front. When the shaft descends the toggle-joints (1) are drawn in and they approximate the wires in front. In this way a to-and-fro movement of the ventricular walls is secured. As these walls move to and fro the appearance at a distance simulates very well the rhythmical contraction of the heart. Each valve cusp is pivoted upon a wire which runs through to the back of the supporting board, where it is balanced by a weight to which a string is tied (Fig. 16, III). The string runs to the central shaft (V). As the shaft V ascends it pulls the string with it, which raises the weight and thereby causes the cusp, fastened to the same wire with the weight, to rise also. When the shaft descends the string is slackened, the little weight falls and with it the corresponding cusp. By adapting the weights and the attachment of the strings to the shaft, it is easy to have all the cusps move in proper Digitized by Microsoft® 42 The Use of Automatic and Other rhythm. One or more cusps may be rendered immovable or partly movable at any time by permanently hooking up their corresponding weights, thus showing mitral or aortic leakage, etc. The following facts are demonstrated with great facility upon this model : 1. The rhythm of the ventricular walls. 2. The rhythm of the valves, showing how the auriculo- ventricular valves open as the semilunar valves close, and conversely. 3. The effect of a cusp made rigid and thereby causing a leak (regurgitation). 4. The sinuses of Valsalva, showing the pockets behind the aortic and pulmonary cusps. 5. There is little or no change in the longitudinal dia- meter of the heart in systole, but a great shortening of the transverse diameter occurs. 6. The movement of the model produces sounds which, though differing of course in pitch, resemble those of the heart in rhythm and in cause. One element of these sounds is distinctly heard from the clapping together of the valves, the other from the friction of the ventricular walls. By hooking up all the cusps and immobilizing them from behind, the muscular element alone remains. This difference can be heard distinctly many feet distant. 7. It is seen clearly that the cusps are slammed suddenly together like trap-doors, with great "'force and emphasis, and that they are not " floated up," as is frequently erroneously stated in physiological text books. If they were, where would they " float " when a man is in a recumbent posture or standing on his head ? In the former case one cusp would " float up " in position, while the other floated back against the ventricular wall ! In the latter case they could assuredly not " float up." Moreover, the normal position of the heart, when the body is erect, does not favor such " floating." 8. It is also demonstrated that the cusps of the mitral and tricuspid valves do not and cannot overlap 2,% the valves close, but that their ajl,^uja^ ^s^u^acej^^^ approximated and Models in Teaching Physiology. 43 pressed the more firmly together the firmer the intra-ven- tricular pressure. 9. I have fastened a cord to a little brake which slows down the clock or stops it completely at will. This cord represents a vagus nerve. The clock is wound and the heart starts its pulsations very rapidly (palpitation). Standing several feet distant and pulling upon the vagus cord, the heart may be slowed to any desired rate of beat ; relaxing the cord allows the rapidity of action to be resumed. 10. The vagus cord makes it easy to stop the heart sud- denly in full diastole, in full systole, or in any midway con- dition, in order to compare the position of the valves, open or shut, with the position of the ventricular walls, relaxed or contracted. This model is naturally not to be criticised from a strict anatomical standpoint — no model can supply everything — but a great deal of good physiological demonstration can be made with it, and, once regulated, it works with unfailing accuracy. I have used it in the lecture room for several years, and many students have testified to the aid it has given them in getting a clearer understanding of the cardiac mechanism. It is so large that the movements of the valves, which are five inches in diameter, can be seen at any dis- tance. Although this model serves to demonstrate so much, it is simple enough in structure — a cheap clock, with some brass wires and rods, a few little weights and strings, are all the machinery required. There are no chorda3 tendineas. I tried to attach them, but it is difficult to represent them un- less by using elastic bands, and the real chordae are not elastic. They] may be better demonstrated by a separate model (see Fig. 18). An Automatic Model for Rhythm. — This model is intended to illustrate but a single physical fact, namely, that there is nothing remarkable in a continuous inflow being converted into an intermittent outflow. In the heart the inflow of the venous blood is not absolutely constant, being momentarily checked by the auricular systole, but it is nevertheless true that the blood occupies a much longer time in flowing into the heart than in leaving it. Moreover, when the intra-car- Digitized by Microsoft® 44 The Use of Automatic and Other diac pressure has reached a certain maximum the heart emp- ties itself, just as this model empties itself when a maximum pressure is attained. The model (Fig. 17) consists of an oblong, water-tight tin box, six by two and a half by two and a half inches. The box is closed except for a wide opening at one end (Fig. 17, I), and a small perforation in the top, near the opposite end, which admits a rubber tube. The box is suspended by a central wire axis placed a little behind the middle, i.e., a little nearer the end in which the rubber tube is fixed. Fig. 17. — Model to show the conversion of a continuous stream from the pressure bottle at II into an intermittent outflow from the pivoted tin box at I. Within, a transverse partition divides the box into two un- equal compartments, and the larger one opens through the large mouth. The partition does not quite reach from the floor to the roof of the box, so that water can flow over the top of it from the smaller into the larger partition. When empty the longer, heavier end of the box swings down and makes the long axis of the box perpendicular. Colored water is then allowed to flow in from a pressure bottle (Fig. 17, II) through the rubber tube. This fills the smaller compartment, makes it heavier, and therefore tips Digitized by Microsoft® Models in Teaching Physiology. 45 the box back to a horizontal position, as shown in Fig. 17. The water rises to the top of the partition and flows over into the larger compartment, which again disturbs the equi- librium and tips the box up perpendicularly, so that all the water is suddenly discharged. The flow from the rubber tube meanwhile continues and soon fills the small compart- ment over again, righting the box as before. This works with perfect rhythm, and by raising or lowering the pressure bottle (Fig. 17, II) the frequency of discharges from the tin heart is varied, but always with exact rhythm. THE HEART VALVES. A model of the mitral valve closed, with its chordae ten- FiG. 18.— Model of the Mitral Valve, closed, papillary muscles. I, auriculo-ventricular ring; II, dineae and the papillary muscles, is made as follows (Fig. 18) : The mitral valve of an ox heart is carefully dissected out, attached to the auriculo-ventricular ring. The ring is then cut across and opened out flat, laid on a paper, and the out- lines of the flattened cusps are drawn. The drawing is then enlarged to three feet in length, and from this pat- tern a piece of wire cloth is cut. The wire cloth, being Digitized by Microsoft® 46 The Use of Automatic a7id Other bent into a ring, gives a perfect model of the magnified mitral valve, one foot in diameter. One side of the valve is nailed to an upright board, and the vsrire cloth is cov- ered by a double layer of paper and glue. The cusps of the valve are pressed together until the orifice of the ring (mitral orifice) is closed tightly (Fig. i8, I). Strings are glued to the ventricular surface of the cusps and to their margins, to represent the chordse tendinea^. The strings are collected into larger and firmer bundles, and the latter are drawn down and fastened to the apices of the papillary muscles, represented by cones of "wire paper" tacked to a horizontal band for support. This model makes clear the following details : 1. The fact that chords tendineae often proceed from a single papillary muscle to more than one cusp. 2. The branches and anastomoses of the chordae as they approach the cusps. 3. The manner in which the cusps are approximated in order to close the orifice, and the appearance of the auricu- lar and ventricular surfaces of the closed mitral valve. 4. The strong " purchase " that the branching chordee have upon the cusps from their numerous points of attach- ment to them. 5. The impossibility of regurgitation under normal con- ditions. 6. The fact that some of the papillary muscles stand well out toward the lumen of the ventricle, thus enabling the chordas to be attached to better advantage. It is useful to make a similar model with the valve open instead of closed. SCHEMA OF THE CIRCULATION. The conditions of pressure in the heart, arteries, capilla- ries, and veins respectively are shown by the following ap- paratus (Fig. 19) : The heart is a Davidson or other syringe. The aorta, capillaries, and vena cava are represented by rubber tubing of different calibres with glass T-tubes inserted ; for anas- tomoses straight glass tubes of various calibres are inserted as manometers in the arteries, veins, and capillaries. The Digitized by Microsoft® Models in Teaching Physiology. 47 tubes are held upright by wooden sticks or trunk hoops bent at a right angle and screwed to the supporting board. This schema is based upon an old idea, but the form of it pictured in Fig. 19 is inexpensive and it works admirably. The tubes in the aorta and vena cava should be at least half an inch in calibre, and the others are much smaller. Colored water is used to fill the tubes. On this schema are shown the great pressure of the fluid in the aorta, which may be made to rise three feet, and the negative venous pressure near the heart (rubber bulb). The local pressure may be made to rise abnormally in any tube of the series by com- pressing the distal tubing. Fig. 19. — Circulation Schema. Showing the varying height of the fluid in the different manometer tubes. The effects upon the pressure of slow and rapid heart beats and of strong and weak pulsations are clearly seen at a long distance. If the amount of fluid in the schema be re- duced, air will be sucked in at the vena cava, and the effect of air in the circulation, churned up with the blood, is seen as the bubbles flow through the T-tubes. The schema also shows that the collective area of the arterioles is much greater than the aorta, and demonstrates the resistance which their walls offer through friction. THE PULSE WAVE MODEL. The velocity of the pulse wave, its gradual diminution in amplitude as it recedes from the heart, retardation of the Digitized by Microsoft® 48 The Use of Automatic and Other pulse wave, and the effect of slow and quick, feeble and forcible heart beats upon the arterial pulse, are shown by the model represented in Fig. 20, A. A rubber tube ten feet long, with very elastic walls, is bent and fastened to an upright board. Levers -(light wood or straw) are pivoted so as to rest upon the tubing at inter- vals. When water is forced through the tubing by a Da- vidson syringe (representing the heart) the movements of the pulse wave are transmitted to the levers. The lever next the heart rises promptly and extensively. The next lever rises perceptibly later and a little less extensively, and so on. Fig. 20.— Models for the Circulation. A, model to show retardation and diminution of the pulse wave by means of levers resting upon a curved rubber tube; B, model to show the difference between the blood pressure, recorded by a manometer, and the pulse wave, recorded by a lever. The difference between the blood pressure and the pulse wave, and the fact that a sphygmograph does not give an accurate record of blood pressure, are illustrated by the model shown in Fig. 20, B. A stream of fluid is forced from a Davidson syringe through four feet of rubber tubing. A lever rests upon the tubing and moves with the pulse wave. A manometer (mercurial) is inserted in the tube. If the tube is kept overfilled with fluid there will be very little movement of the lever (sphygmograph) with each pulse wave, but the pressure in the manometer will be seen to be very high. The same slight movement of the lever is of Digitized by Microsoft® Models in Teaching Physiology, 49 course obtained if the tubing is relatively empty, or if the heart beat (rubber bulb) is very feeble, while the mano- metric pressure is seen to be very low. MODEL OF ARTERY AND CAPILLARIES. This model is made as follows (Fig. 21) : A sheet of " wire cloth " three feet by four or five feet is laid on a table, and with chalk the outlines of an artery, arteri- oles, and terminal capillaries are drawn upon it. With shears this outline is cut out, and the walls are bent to- gether through a half-circle, as shown in Fig. 21. A couple of dozen red blood discs are modelled out of wax and painted. They should be adapted to the width of the narrowest capil- lary. Some white blood corpuscles and blood plaques of Fig. 21, — Model of an Artery, Arterioles, and Capillaries, showing red blood discs rolling along the capillaries. proportionate size are made fresh from modeller's clay or putty. The wall of the large artery is thickened by ad- ditional layers of paper and glue. On the back the layers may be represented as peeled off in part, to show the several coats of the artery. On the rest of the back of the artery a small nutrient artery with its vein and vaso-motor nerve are represented by red, blue, and white cords respec- tively. The model is used for these demonstrations : 1. The several arterial coats (painted diagramraatically). 2. The nutrient vessels. 3. The collective area of the branches of the artery com- pared with the parent stem. 4. The capillaries and their anastomoses. 5. The appearance of the blood stream as seen by the microscope in the capi^^|^^ ^^ /j^.^^„^„^@ 50 The Use of A utomatic and Other The red discs are rolled along in the middle of the stream, with their long axes parallel to the long axes of the vessel. 6. At the bifurcation of a vessel they are made to loiter and balance until some corpuscle from behind rolls against them. 7. When they reach a very narrow capillary (being wax) they are doubled in two and manage to squeeze through. 8. The white corpuscles are more pliable (clay), and they are shown to loiter near the walls and to elongate in passing through narrow capillaries. 9. At intervals they are worked through small perfor- ations in the capillary walls, until they come out on the opposite side and are free in the inter-capillary spaces. 10. The occasional jamming of corpuscles in a capillary and the back-flow through an anastomotic branch, and the obstructive effect upon the circulation of such jamming. 11. The formation of an intra-arterial clot by means of blood plaques and fibrin, and the shape of such a clot when a small artery is tied off and the clot reaches gradually back to the next branch. MODEL OF VEIN VALVES. A simple model of the valves in the veins and lymphatic vessels, or of the semilunar valves of the heart, is thus made: A wide lamp chimney is purchased which bulges outward on two sides (to accommodate a broad, flat flame). The cusps of the valves are cut out of rubber cloth in the shape of watch pockets. The curved margins of the rounded pockets are fastened within the chimney by water-tight ce- ment (" Stratena " is the best). When dry, colored water may be poured in at the chimney top from a pitcher. It runs into the bulging parts of the chimney (corresponding to the sinuses behind the cusps) and slams the rubber cusps to- gether. If they are carefully fitted the chimney may be filled to the top without leaking. When the water is poured in at the bottom of the chimney it of course runs directly through and presses the rubber cusps back into the bulging sides of the chimney. I desire to expr6^«^?rf9feWfa?^^% Dr. M. Allen Starr Models in Teaching Physiology. 51 for valuable assistance in modelling the brain nuclei and ar- ranging the segments and tracts in the large cord model, and to Dr. J. West Roosevelt for suggestions in regard to certain ingenious details in the mechanism of the heart models. I do not wish to imply that the models here described are to replace other auxiliaries for teaching; they are only intended to make clearer certain demonstrations and add interest to some subjects which are apt to prove stumbling blocks to students. It is easy to give wrong impressions with auto- matic models, unless one is careful to emphasize the points of difference as well as of resemblance to the real objects ; but this very difference may be made of value by fixing a fact more clearly. There are so many new and long words employed in physiological text books that students are apt to think they understand a description when they oftentimes fail to grasp the fundamental point. I have frequently been surprised to find how some simple demonstration of an ob- ject familiar by name to an intelligent student, but never before regarded in three dimensions of space, would appeal to him and awaken an entirely new conception. Digitized by Microsoft® PRIMARY TUBERCULOSIS OF THE BRONCHIAL GLANDS. AN EXPLANATION OF SOME FACTS IN THE ETIOLOGY OF PHTHISIS. HENRY P. LOOMIS, M.D., Professor of Pathology; Visiting Physician and Curator to Bellevue Hospital. In the last number of the Researches, I published the result of the examination of the bronchial glands for tubercle bacilli of fifteen persons who died suddenly, either from accident or acute disease, and in whom no other tubercular lesions were found at the autopsy. The results obtained induced me to continue the inves- tigations, until I am now able to offer the conclusions ob- tained from thirty autopsies and experiments upon animals — a sufficiently large number to furnish, I believe, some posi- tive facts as to the frequency of primary tuberculosis of the bronchial glands, and an explanation of the development of pulmonary tuberculosis in persons With no hereditary phthisical tendencies — persons, in fact, who at the time of the development of phthisis are under the best hygienic conditions, and in whom the tuberculosis seemed to be ex- cited by an acute simple bronchial catarrh or by some other acute non-tubercular disease. May not the presence of the tubercle bacillus in the bronchial glands explain the development of tubercular meningitis and tubercular joint disease in those cases where no primary focus can be found ? These are the cases which have always puzzled the pathologist ; and here, I believe, we have an explanation of many of them. It is a well-known fact that tubercles are deposited in the bronchial glands of adults, subsidiary to deposits in the lungs, in about og^J:fx^|^ o^^^^^^ oi phthisis. Some Primary Tuberculosis of the Bronchial Glands. 53 cases which show only a single small patch of tubercular change in the lungs, present bronchial glands extensively diseased and often entirely caseous. It may also be added, in accordance with the report of cases as given below, that well-marked tuberculosis of the bronchial glands may ex- ist where the lungs are perfectly normal and no evidence of tuberculosis can be found elsewhere in the body. It was with the object of ascertaining the frequency of this latter class of cases that I commenced, eighteen months ago, a se- ries of examinations of non-tubercular cases, and, to render the conclusions as free from error as possible, associated control experiments upon animals. In cases of phthisis in infancy and childhood it is an equally well-established fact that an abundant deposit of tubercles in the bronchial glands is of great frequency, the disease in the glands being ofttimes as important as that in the lungs, and much more considerable. These evidences led Bollinger to carefully examine the bronchial glands of all children who died dur- ing the great epidemic of measles in Munich in 1887. He found and proved that tuberculosis may be latent in a child in apparently perfect health, and also demonstrated the presence of tubercle bacilli in the swollen lymphatic glands at the root of the lungs and in the mediastinum in children who gave no evidence whatever of tuberculosis in the lungs or other parts of the body. It has further been demon- strated that in most cases of phthisis in children, the disease has been preceded by a long-existent, but latent, glandular tuberculosis, the lymphatic glands in children being in a state of functional activity, and therefore by the slightest irritation becoming the seat of acute congestion. Papavoine mentions forty-nine children, of whom eleven were affected with tubercles in the bronchial glands without any in the lungs. From the foregoing it seems reasonable to question whether a similar glandular condition may not exist in adults, only becoming appreciable when a lowering of the vitality, or some unknown cause favoring the escape of the bacilli from the glands, sets up a general tuberculosis, then easily recognized. I have been unable to find any report or statemei^^-^^^yp^^^yj^^^loWfeich have as yet been 54 Primary Tuberculosis of made to settle this point. The only printed statistics which give an idea of the frequency of primary tuberculosis of the bronchial glands are made by Dr. Chambers in a report (" Deceminum Pathologicum ") of post-mortem examinations on five hundred arid sixty-six tubercular patients at St. George's Hospital during a period of ten years. He gives nine cases from among these as having glands and not the lungs involved; all were cases, however,in which the diagnosis was only made by gross appearance, and it is easily seen how this is liable to error. Professor "Von Zierassen, in his work on " Pulmonary Tuberculosis," when writing of latent tuberculosis in the glands of children, simply suggests the possible occurrence in adults. Ziegler, in his last work on " Pathological Anatomy," under the subject of " Phthisis," states that bronchial glands may be tuberculous throughout their entire extent and yet no changes at all be discovered in the lungs. How he reaches this conclusion he does not say. Cornil introduced into the trachea of guinea-pigs a few drops of tuberculous culture and observed submucous tubercular lesions with rapid generalization, and afterward demonstrated tubercle bacilli in the bronchial glands, thus showing that the bacilli may reach the bronchial glands directly through the tracheal mucous membrane. Before my attention was called especially to the bronchial glands I had made many autopsies on cases of acute pulmonary or acute general tuberculosis, and although it seemed un- doubted that lesions must exist somewhere to show where the tubercle bacilli entered the body, multiplied, and gained headway before being able to produce a general in- fection, the spot was not ascertainable. For the last year I have examined carefully the bronchial glands in this class of cases, and have been surprised to find in how many instances the glands were extensively involved and the lesion far ad- vanced and apparently antedating the general eruption. It has been demonstrated that tubercle bacilli may be inhaled in respired air. The position of the bronchial glands and the arrangement of the lymphatics in the re- spiratory tracts render it easier for the inhaled tubercle bacilli to enter them than the lung parenchyma, for it is to be rememberefl;gtfh'2*c/a!tytfWs;riwaff€hing of the trachea the Bronchial Glands. 55 between the right and left bronchi is a group of glands, from ten to fifteen in number, varying in size from a small pea to an olive, and that just beneath the mucous membrane of the lower portion of the trachea, and the mucous surface of all the bronchi, is an immense network of lymphatic ves- sels running directly into the bronchial glands ; also differ- ent lymphatics reach these glands from the lungs, the pleura, and adjoining parts. The blood supply of the gland is from the bronchial arteries, and the blood is returned to the bronchial veins and so to the right side of the heart. On the mucous surface of the trachea and larger bronchi the tubercle bacilli would be more liable to lodge than to be carried down to the pulmonary alveoli. If the integrity of these mucous surfaces is weakened, as occurs during mild attacks of bronchial catarrh, the direct passage of the tubercle bacilli into the bronchial glands would be favored. The tubercle bacilli, having once gained entrance to the bronchial glands, would there find fit soil for development and a temperature most favorable to their growth. In the cases which I have examined the tubercular infection did not appear to have begun simultaneously in all the glands, nor to have advanced with the same rapidity in each gland. One or two abnormally large glands are probably found at the bifurcation of the trachea, which form an irregular, rounded, lobulated mass, measuring two or even more inches in diameter. The other bronchial glands vary in size from a natural bulk to that of an &g^ ; a few only or several may be affected. Many of the glands affected will present a general tu- bercular lesion, being dense, firm, and scattered throughout with tubercular masses. These glands are often very much pigmented, and are generally surrounded by a dense capsule of connective tissue firmly binding the gland to the sur- rounding parts. The natural thin involucrum of the gland always thickens as tuberculosis advances. This condition was especially noticed in the bronchial glands removed from the cases which I have collected, and which were proven by inoculation experiments to have contained ba- cilli. Adhesions to the bronchial tubes are formed through the medium of an areolar tissue, which is at first loose in its Digitized by Microsoft® 56 Primary Tuberculosis of texture, but becomes more dense. It would seem to be a conservative process, limiting the tuberculosis and account- ing for the fact that bacilli were present within the organ- ism without producing any evil effects. Sometimes the glands become so infiltrated as to constitute a yellowish- white, friable substance in which no vestige of the pristine texture remains discernible. Frequently in the glands are to be found encysted, cheesy, or calcareous masses, some- times quite large, the cretaceous material being, as a rule, near the centre of the gland. Fig. I is an accurate reproduction of a collection of en- .A B FiG- I- — Enlarged Tubercular Bronchial Gland. A, trachea; B, right bron- chus; C, enormously enlarged bronchial gland, proven by examination to be tu- bercular; D, collection of bronchial glands between right and left bronchus, slightly enlarged. larged bronchial glands removed from Case XXX. Six glands were removed, one of which was nearly as large as the one shown in the drawing. These were injected in the usual way into a healthy rabbit, which at the autopsy showed the characteristic lesions of a general tuberculosis, proving that the enlarged glands contained tubercle ba- cilli. The posterior part of gland C was cut into and pre- sented the gross appearance of a chronic tubercular process with caseation ^^l]mi0hBmrit® the Bronchial Glands. 57 It is possible, although hardly probable, that in the cases where the lungs were examined very carefully for the slight- est trace of a recent or old tubercular process, the tuber- cle bacilli might have reached the glands after first having lodged in the pulmonary parenchymata, setting up so slight or so localized changes as to be overlooked, then to be car- ried by the lymphatics to the glands at the root of a lung. It seems much more probable that the bacilli reach the glands through the lymphatics of the mucous membrane, of the trachea, of the large bronchi, or even of the smaller one. Since the bronchial veins return the blood to the right heart and so back to the lungs, may not the development of an acute pulmonary tuberculosis from infected bronchial glands be explained by the venous blood system taking up some elements from a softened, cheesy lymphatic gland and car- rying this substance to the lungs? Dr. T. M. Prudden has lately shown me a perfectly normal lung that was laid open by a vertical incision from apex to base, and included the structures at the root of the lung. At the root of the lung, just above a large bronchus, was seen an enlarged bron- chial gland, four centimetres in length and three centimetres in width, pigmented, dense, and firm. In its upper part were seen two large tubercles, one the size of a small cherry, which on microscopical examination presented the character- istic appearances of a tubercular mass. The autopsy and microscopical examination of the case were conducted by Dr. T. M. Prudden, and no tubercular lesions were found, after careful examination, in any part of the body. Here was unquestionably primary tuberculosis of the bronchial glands. How long the tubercle bacilli were in this gland it is impossible to say, as well as how long they would have remained there comparatively harmless. Anal- ogous cases are found in encysted cheesy masses at the apex of the lungs, which remain long after physical signs and ra- tional symptoms point to any change. How great the danger must be while the bacilli are present in the organ- ism, no matter how apparently circumscribed they appear to be, any one can see. The demonstration of the tubercle bacilli confined to the bronchial glands would explain the manner in which they might gain entrance into the lymph Digitized by Microsoft® 58 Primary Tuberculosis of or blood current, and so be carried to a point furnishing a fit soil for their development. It was with the hope of obtaining some facts which might throw light upon these questions that a series of autopsies was undertaken on persons who had died suddenly, either from accident or from acute non-tubercular disease. In all such cases a careful examination was made of each organ, especially the lungs, for evidences of old tubercular processes ; if none were found, then the bronchial glands were removed and placed in melted paraffin, so that the hot paraffin would destroy all putrefactive bacteria that might contaminate the surface of the glands, and would render them sterile until an animal could be inoculated with their substance. Within half an hour after removal the glands were taken from the hardened paraffin, washed with a bichloride solution, and reduced to a pulp in a ster- ilized crucible, to which a small amount of distilled water was added to render the fluid obtained from the glands thin enough to pass through the needle of a sterilized Koch syringe. About two cubic centimetres of the fluid which had been obtained from the pulverized bronchial glands were then injected into the right pleural cavity of a healthy rabbit. The rabbit was kept under the best hygienic sur- roundings, and, if it did not die before the end of six weeks, was killed by being bled to death. A careful record was kept of the autopsy on the rabbit, and of the microscopical examination of the organs. If the first rabbit at the au- topsy showed evidence of tuberculosis, a second rabbit was inoculated, under antiseptic precautions, with material ob- tained from the first, to determine positively the character of the lesions found in the rabbits which were inoculated with the bronchial gland substance. Eleven rabbits died within twelve hours to six days, either from a faulty inoc- ulation, too large a quantity of fluid being used, the animal being injured by the needle of the syringe, or possibly be- cause the glands had undergone decomposition, the autopsy on the body from which they were removed having been made too late. From these cases no conclusions could be reached, and they are not included in the list of experi- ments. Digitized by Microsoft® the Bronchial Glands. 59 The question might arise, in determining the presence of the tubercle bacilli in the bronchial gland, instead of inocu- lating animals with the material of the glands, why not cut sections of these glands and stain them for the tubercle ba- cilli ? The reasons, to my mind, are two, viz.: in the first place, the bacilli m the glands may be so few and scattered that it might necessitate a hundred sections being made to obtain, under the microscope, that portion of the gland containing the bacilli ; and, secondly, in the appreciation of a small number of bacilli a personal factor comes in — one observer affirming their presence, while to another they are obscure. In answer to the objection that the material may have become contaminated before its inoculation into the animal, it may be stated that all the steps of the opera- tions were conducted under the strictest antiseptic precau- tions, even to the removal of the glands from the dead body by sterilized instruments. Eight out of the thirty persons on whom autopsies were made had in their bronchial glands active tubercle bacilli. None of the persons presented any evidence of old or recent tubercular process, yet the bacilli were within the bronchial glands, ready to enter the general lymphatic system of the lungs and set up a localized pulmonary tuberculosis, or, by entering the general lymph or blood current, give rise to a general tuberculosis. The following is a report of the cases upon whom autop- sies were made, with the result of the inoculation experi- ments upon rabbits. A detailed report of the first fifteen cases was given in the previous number of the Researches. Here I will simply mention these cases by number, and state whether or not tubercle bacilli were found in the bronchial glands. A complete description will be given of the last fifteen cases. Case I. — An intoxicated woman was taken during the night to a police station and locked up in a cell. On the following morning she was found lying on the floor uncon- scious. Before medical aid could be summoned she died. The autopsy showed that she had small cirrhotic kidneys and had died from uraemia. Digitized by Microsoft® 6o Primary Tuberculosis of The single enlarged bronchial gland removed from this woman, which measured one-half inch by one inch, had all the appearance of an old tubercular process. As the rest of the glands appeared normal, it seems probable that the infection of the rabbit came from this enlarged gland. This rabbit at the autopsy presented similar lesions to those found in rabbits inoculated with a pure culture of the tu- bercle bacilli. Case II. — G. J., aged 32, a prize-fighter, who had come from California to give exhibitions in New York, was seized, upon his arrival here, with a chill, developed lobar pneu- monia, and died on the seventh day of the disease. This man, in perfect health and of extraordinary strength and physical vigor, is suddenly seized with an acute disease and dies on the seventh day. He is found to have had tu- bercle bacilli in his bronchial glands, the presence of which was demonstrated not only in stained sections, but by the results of the inoculations of rabbits. It would seem that the amount of virus in the second rabbit was so small that the animal readily overcame the infection. Case III. — M. F., aged 30, was perfectly well until her confinement. After delivery she developed symptoms of puerperal septicaemia, and died on the ninth day. While the nodules found in the liver of the rabbit inocu- lated from the bronchial glands removed from this case, presented an appearance resembling tubercular nodules, the control experiment showed conclusively that they were non-tubercular. Case IV. — J. R., aged 40, a street vender, while following his business pushed a cart before him about the streets. As he sat on the curbstone with his cart in front of him, on the afternoon of May 17th last, a truck was driven against the cart, the handle of which struck him in the chest. He fell over and died almost instantly. The case proved, firstly, that the bronchial glands con- tained no tubercle bacilli, and, secondly, that the glands re- mained perfectly sterile for twenty-four hours in the paraf- fin. Case V. — J. E.,aged 30, was admitted to Bellevue Hos- pital on May 27th, in the eighth month of pregnancy. She had been in perfect health until two weeks previous. At the time of admission she presented the symptoms of puer- peral uraemia. Four days later labor was induced and a dead child delivered. On the following morning urasmic coma was complete, and she died in the evening. The apparently normal bronchial glands ot this young woman, who had been in perfect health until a short time before death, fj[}^f}^^^§y^%g,^ hygienic surroundings, the Bronchial Glands. 6i furnished a solution which, when injected into a healthy rabbit, produced general tuberculosis. At the autopsy this rabbit presented appearances similar to those which are found in rabbits inoculated with a pure culture of the tu- bercle bacilli, and in its turn furnished, on sterilized pota- toes, an equally pure culture of the bacilli. Case VI. — G. W., aged 38, who had been drinking heav- ily for three weeks, was found in an unconscious condition on the street, and taken to the alcoholic ward of Bellevue Hospital. lie died in four hours. At the autopsy there was the ordinary "wet brain" of the pancreas, which was twice the normal size, showed exten- sive areas of hasmorrhagic necrosis. Inoculation experiments showed that the bronchial glands contained no tubercle bacilli. Case VII. — Frederick W., aged 35, of robust physique, who had been in the habit of going on periodical sprees for the past six years, was brought into Bellevue Hospital with the symptoms of acute alcoholism, and died twenty hours after admission. Cirrhotic kidneys, dilated heart, pericarditis with effu- sion, and cerebral oedema, were the lesions found at the au- topsy. The lesions found in the inoculated rabbit were so char- acteristic of tuberculosis that a control inoculation seemed unnecessary, and it was not performed. Case VIII. — J. D., aged 60, was in perfect health until six days before his death, when he suddenly became un- conscious and presented the usual symptoms of cerebral apoplexy. There is no evidence that the bronchial glands removed from this case contained tubercle bacilli ; the animal inocu- lated evidently died of non-tubercular peritonitis. Case IX. — George E., aged 38, stevedore, while engaged in unloading a vessel, was instantly killed by the falling of a beam. Inoculation experiments made with the bronchial glands removed from this case showed that they were perfectly normal. Case X. — D. M., aged 66, who had suffered over a year with cardiac symptoms, died suddenly. At the end of six weeks the rabbit inoculated with the bronchial glands from this case was killed. The left lung was normal, but the right was completely consolidated and bound to the chest wall. Single and collected nodules pro- jected from the surface and studded the pulmonary tissue. Three small cavities were noted at the base. A nodule as large as an oUve, containing grumous material and creamy- Digitized by Microsoft® 62 Primary Tuberculosis of looking pus, completely filled the right lobe of the liver. Two nodules, the size of millet seeds, were found in the right kidney. The mesenteric glands were enlarged and pigmented. A second rabbit was inoculated in the right pleural cavity with material obtained from the first, and died at the end of seven weeks with the characteristic lesions of miliary tuberculosis. Case XI. — James D., aged 46, of strong, vigorous phy- sique, was admitted to the hospital with the symptoms of lobar pneumonia, and died on the fourth day of the disease. There is no evidence that in this case the bronchial glands contained tubercle bacilli, although one gland was very much enlarged and contained a calcareous mass, showing that it must have been at some time the seat of cheesy metamorphosis. Case XII. — Fanny W.,aged 17, when in excellent health, was attacked with acute pericarditis and died on the sixth day. This case not only showed that no tubercle bacilli were demonstrated in the bronchial glands removed from the young woman, but that the pericarditis was non-tubercular. Case XIII.— Robert H., aged 20, was in perfect health until attacked with typhoid fever. In the third week of the disease he died. * Immediately after the removal the bronchial and inguinal glands were reduced to pulp in separate crucibles, and four cubic centimetres of the fluid obtained from each were injected respectively into the right pleural cavities of two healthy rabbits. Six weeks later the rabbits were killed, and the organs in both were found normal. Case XIV. — A. D., aged 34, farmer, was admitted to Bellevue Hospital June 22df after an illness of sixteen days. The diagnosis of trichina poisoning was made. The patient died on the fourth day after admission. The bronchial glands of this case contained no tubercle bacilli. Case XV.— F. F., aged 45, was found dead by the poHce on the sidewalk of one of the down-town streets. The kidneys were typical examples of the " small cirrho- tic kidney." The left heart was hypertrophied and showed no valvular lesion. The lungs were normal, and no evi- dence of any old or recent tubercular process was found in the body. The brain was not examined. J^-i The bronchial glands removed from this man were proven to contain tubercle bacilli. Case XVI.— Unknown man, aged about 38, was brought Digitized by Microsoft® ^ ^ the Bronchial Glands. 63 into Bellevue Hospital by ambulance. He died within eight hours and no history was obtainable. Autopsy, twelve hours after death. — The lungs were nor- mal and no evidence of any tubercular process was found anywhere in the body. The man died from a rupture of an aneurism of the transverse portion of the arch of the aorta. Seven bronchial glands were removed, one of which was very much enlarged (two inches in diameter). One-third of this gland was preserved to examine microscopically for tu- bercle bacilli, if the inoculation experiments should show that the rabbit died of tuberculosis. Inoculation Experiments and Results. — Five cubic centime- tres of the fluid obtained from the glands were inoculated in the usual way into a healthy rabbit. Three weeks after in- oculation the rabbit had a litter of five healthy young. The rabbit was killed at the end of eight weeks and all the or- gans were found in a healthy condition. Case XVII. — Bridget N., aged 48, for a number of years a helper at the almshouse, while under the influence of liquor took by mistake an overdose of morphine and died within six hours. Autopsy, twenty-four hours after death. — All the organs preseAted by their gross appearance a normal condition, with the exception of an old patch of softening in the frontal lobe of the brain. Six bronchial glands were removed ; one was enlarged and fibrinous-looking, another contained a black, soft mass which looked like coagulated blood. Inoculation Experiments and Results. — A healthy rabbit was inoculated with three cubic centimetres of the fluid ob- tained from the glands. The rabbit was killed by an acci- dent at the beginning of the fifth week of inoculation. An autopsy was at once made, and, with the exception of local- ized adhesions over the right lung, all the organs were nor- mal. Case XVIII.— Patrick M., aged 32, fell by accident from one of the wharves into the East River. He was rescued within a few minutes, but in spite of prompt treatment died at the end of two hours. Autopsy, made ten hours after death, at the request of the coroner. — The man was of fine ^physique and in perfect Digitized by Microsoft® 64 Primary Tuberculosis of health, all his organs being in normal condition. The lungs presented the characteristic appearance of death by drown- ing. Inoculation Experiments and Results. — Two and one-half cubic centimetres of fluid which was obtained from ten nor- mal-looking bronchial glands removed from the case were injected into the right pleural cavity of a healthy rabbit. Seven weeks later the rabbit was killed and all the organs were found in a normal condition. There were no adhesions or pigment deposits to show where the needle was intro- duced. Case XIX. — Rosa S., aged 40, was found drowned at the foot of Morton street. Nothing of her previous history is known. Autopsy, made twenty -four hours after the body was found. — The woman was robust-looking and apparently had not been in the water many hours. All the organs were in a normal condition, with the exception of the heart, which showed excessive left ventricle hypertrophy with no valvular lesion. Many of the arteries, and especially the aorta, were atheromatous. Five large gall stones were found encysted in the gall bladder. Ten bronchial glands were removed, one of which was very large and contained a cheesy mass the size of a cherry in its centre. Inoculation Experiments and Results. — Three cubic centi- metres of the fluid obtained from the glands, together with the cheesy centre of the enlarged gland, were injected into a healthy rabbit. Nine weeks later the rabbit was killed. It was found that the fluid had been injected by mistake into the upper portion of the peritoneal cavity behind the right lobe of the liver. At the point of inoculation there was no evidence of any inflammatory change beyond a small patch of prithysabitis. The lymphatics of the posterior abdominal wall were beautifully mapped out by the pigment derived from the glands. AH the organs of the rabbit were in a per- fectly healthy condition. This case shows two interesting points: first, that the enlarged cheesy gland contained no tubercle bacilli, for the whole mass was reduced to a fluid and injected ; second, that the injection was perfectly harm- less, producing no inflammation or adhesions. Digitized by Microsoft® the Bronchial Glands. 65 Case XX. — Michael F., aged 45, died of lobar pneumonia on the seventh day of the disease. Before his last illness he had always enjoyed the best of health. Autopsy, twenty-eight hours after death. — The lower lobes of both lungs were in the second stage of lobar pneumonia. Old adhesions bound the left lung in places to the chest wall. With these exceptions all the organs were in a normal condition. Eight fibrinous-looking, but only slightly en- larged, bronchial glands were removed. Inoculation Experiments and Results. — Five cubic centi- metres of the fluid, obtained by preparing these glands in the usual way, were inoculated into a healthy rabbit. Seven weeks afterward the rabbit was killed, and at the autopsy the following condition was found : The right lung was bound to the chest wall throughout its whole extent. Four or five masses projected from the surface and extended into the substance of the lower lobe of the right lung. These nodules were surrounded by consolidated areas and had all the appearance of being tubercular. There was a mass the size of a large cherry behind the lungs, which contained a whitish, cheesy material. This I believe to have been an enlarged tubercular bronchial gland, although stained cover-glass preparations of the cheesy material failed to reveal any tubercle bacilli. The left lung was normal, as were also the abdominal organs. A rabbit which had been inoculated from the first rabbit died by an accident before any conclusion could be reached, so we have no positive way to prove that the lesions of the first rabbit were tubercular. From a careful examination of a number of sections I feel almost certain that the changes found were those of a localized tuberculosis, but I have not included this case with those which had tubercle bacilli in their bronchial glands. Case XXI. — S. C.'s child, aged 2^ years, was always perfectly well to within ten days of its death. The diag- nosis of general capillary bronchitis was made, but, the ques- tion of tuberculosis being raised, an autopsy was requested. Autopsy, made twelve hours after death. — All the organs were found in a normal condition, with the exception of the lungs, which showed the lesions of catarrhal pneumonia 5 Digitized by Microsoft® S6 Primary Tuberculosis of secondary to a general bronchitis. No tuberculosis was found. Seven normal-looking bronchial glands were re- moved and kept for six hours wrapped in a cloth soaked in a solution of carbolic acid. Inoculation Experiments and Results. — Three cubic centi- metres of the fluid obtained from the glands were injected in the usual way into a healthy rabbit. Eight weeks later the rabbit was killed and all its organs were found in a nor- mal condition. There was localized adhesion of the lung at the point where the needle pierced the chest cavity. Case XXII.— Mary M., aged 72, died of senility at the Workhouse hospital. She was well nourished and hid been confined to her bed only one week. Autopsy, made twenty-four hours after death. — About a pint of clear serum was found in the right pleural cavity. The lungs were normal, with the exception of oedema and congestion of the lower lobes. The liver and kidneys both presented the lesions of an old interstitial change (cirrhotic). Five enlarged bronchial glands were removed. Inoculation Experiments and Results. — Three cubic centi- metres of the fluid obtained from the glands were injected into the right pleural cavity of a rabbit. The rabbit, which had remained perfectly well, was killed at the end of eight weeks. The thoracic organs were normal. The abdominal cavity contained a small amount of clear serum. Through- out the liver were a number of very small whitish-gray points, having very much the appearance of miliary tu- bercles. However, a microscopical examination of these masses showed that they were not tubercles, but small ca- pillary infarctions (mechanical). The enlarged glands removed from this case contained no tubercle bacilli, although the gross appearance of the lesions found in the rabbit's liver was very similar to that of tuber- culosis. The microscopical examination, however, showed these not to be tubercular. Case XXIII. — Anna H., aged 22, was admitted to Belle- vue Hospital suffering from cardiac disease. Her friends said she had been perfectly well to within two weeks of her admission, when she was obliged to take to her bed on ac- count of shortness of breath and severe pain and palpitation Digitized by Microsoft® the Bronchial Glands. 6y over the heart. Cardiac asystoliutn was pronounced, and it was impossible to accurately diagnose the loud blowing murmurs heard over the prsecordial region. She died at the end of ten days. Autopsy, twenty-four hours after death. — There was exten- sive ulceration and reversion of the aortic valve, with fibrin- ous deposits on the free surfaces. The lesion was appa- rently an acute ulcerative process engrafted on an old lesion of the aortic valve. All the rest of the organs were nor- mal, with the exception of the lower lobes of the lungs, which showed the lesions of hypostatic congestion. Inoculation Experiments and Results. — Eight bronchial glands of normal size and free from pigment were prepared in the usual way, and five cubic centimetres of the fluid were injected into a healthy rabbit. The rabbit died on the for- tieth day after inoculation. At the autopsy it was found that the fluid had been introduced into the upper portion of the peritoneal cavity. An adhesive peritonitis had been setup, and a large abscess formed to the right and behind the liver, filled with grumous pus. The thoracic organs were normal, although the abscess had not the appearance of a tubercular abscess. Still, to settle the question, a sec- ond rabbit was inoculated with some of its contents. All the organs of this rabbit, which was killed at the end of six weeks, were found normal. The bronchial glands removed from this case contained no tubercle bacilli, and the peri- tonitis and abscess found in the inoculated rabbit were not tubercular, as was proven by the control experiment. Case XXIV. — Patrick B., aged 38, had complained of heart trouble for a long period, but was able to be about and attend to his business to within eight days of his death. During the last week the symptoms of acute endocarditis developed, accompanied by high fever and a rapid pulse. Autopsy, \.yitx\.\.^ hours after death. — All the flaps of the aortic valve presented the lesions of an acute exudative pro- cess engrafted on a chronic lesion. In the lower lobe of the right lung, close to its root, was found a small circumscribed abscess, the result of an infarction. With this exception the lungs were perfectly normal. There were no evidences of any tubercular changes. Two large infarctions were found Digitized by Microsoft® 68 Primary Tuberculosis of in the spleen, together with a few small ones in the kid- neys. All the bronchial glands were very much enlarged and softened. The case was one of acute ulcerative endo- carditis. Micrococci were found in stained sections cut from the aortic valve. Inoculation Experiments and Results. — Fourteen bronchial glands were removed and three cubic centimetres of the fluid obtained from them were injected into a healthy rabbit. Af- ter seven weeks the rabbit was killed. The lymphatics of the right pleural cavity were beautifully mapped out by the injected gland pigment. With the exception of localized adhesions at the point where the needle had pierced the chest cavity, the lungs were perfectly normal. A few small nodules the size of a pin's head were found in the liver, but a microscopical examination showed them to be not tubercles, but emboli in the vessels. This case showed by the inocula- tion experiments that the bronchial glands were free from tubercle bacilli, and also that the enlarged bronchial gland close to the embolic abscess in the lungs contained no sep- tic micro-organism, or not enough to cause the rabbit's death from septicaemia. Case XXV. — Michael M., aged 32, in perfect health, fell from a third-story window upon the pavement below and died immediately from fracture of the skull. Autopsy, made twenty-eight hours after death. — All the organs were found in a normal condition. Five bronchial glands were removed, one of which was about twice the normal size. Inoculation Experiments and Results. — Five cubic centi- metres of the fluid obtained from the glands were injected into the right pleural cavity of a healthy rabbit. Eight weeks afterward the rabbit, which had become emaciated, was killed. A large, completely encysted abscess was found filling the right pleural cavity, producing collapse of the lung. The cavity of the abscess contained a thick, gru- mous, whitish-gray pus, and had somewhat the appearance of a tubercular abscess. The left lung was normal, as also were the rest of the organs, with the exception of two or three suspicious little nodules in the liver. One of these, together with some of the fluid from the abscess, was in- Digitized by Microsoft® the Bronchial Glands. 69 jected into the right pleural cavity of a second rabbit. This rabbit was killed at the end of two months, and all its or- gans were found in a perfectly normal condition. The pleural cavity showed no evidence of the injection. The result of the inoculation of the second rabbit proved conclusively that the nodules and abscess found in the first rabbit were not tubercular, and also that even the enlarged bronchial gland removed from the man was free of tubercle bacilli. Case XXVI. — G. W., aged 35, went to bed in one of the uptown hotels in apparently perfect health. He blew out the gas and was found dead in the morning. * Autopsy, made thirty-two hours after death. — Body well nourished, blood fluid, oedema and congestion of the lungs. There was thickening, but no adhesions, of pleura over the right apex; with this exception the lungs were normal, as were all the rest of the organs. Six bronchial glands were removed, all of them being very much enlarged, and one measured four inches in length and one and one-half inches in diameter and contained a cheesy, calcareous, tubercular-looking mass. Inoculation Experiments and Results. — Four cubic centi- metres of the fluid obtained from the glands, including the cheesy mass found in the large gland, were injected into the right pleural cavity of a large, healthy rabbit. Seven weeks later the rabbit was killed. It was well nourished and apparently healthy. Organized adhesions were found obliterating the right pleural cavity ; the lung was collapsed. The appearance was of an old pleurisy with absorption of the fluid. The lung was normal, as were the rest of the organs. No evidence of tuberculosis was found anywhere. In this case all the bronchial glands were greatly enlarged, and one had the appearance of being tubercular. Still, the inoculation experiment proved it not to be so. This case shows how one might be led into error by relying upon gross appearance alone. Case XXVII. — Female, aged 50, was found in the street in a stupid condition, due, as supposed, to the effects of liquor. She was brought to the hospital by ambulance. Digitized by Microsoft® 70 Primary Tuberculosis of where she died at the end of twelve hours without regain- ing consciousness. Autopsy, thirty-six hours after death. — In the brain were found two small clots ; one involving the anterior and under portion of the occipital lobe on the right side, the other the island of Reil and the overlapping parietal lobe. All the cerebral arteries were atheromatous. The left ventricle was markedly dilated ; there were no valvular lesions. Kidneys cirrhotic; liver slightly " fatty." There were a few old adhesions over both lungs, but no old or recent tubercular changes were found in their substance. Ten bronchial glands were removed, about normal in size and very much pigmented. Two of the glands were about three times the normal size and fibrinous-looking. Inoculation Experiments and Results. — The glands were prepared in the usual way. Two cubic centimetres of the fluid were injected into a healthy rabbit. Seven weeks after inoculation the rabbit was killed. The right pleural cavity presented a normal appearance, except where the needle was introduced. There was connective-tissue in- crease, and on the costal pleura were ten or fifteen tubercles the size of a pin's head. In the right lung a number of tubercular nodules, varying in size from a pin's head to a pea, were found ; one of these nodules was quite large and contained a cheesy centre. In the left lung only a single small tubercular mass was found. A few tubercles were found in the substance and on the inferior surface of the left lobe of the liver. The rest of the organs were normal. The patient was stricken down in the street with an apo- plectic stroke, and was, as far as could be learned from her friends, in good health at the time of her death. The lesions found in the rabbit inoculated with the bronchial glands were so characteristic and so evidently started at the point of injection that an inoculation of a second rabbit was not deemed necessary. Case XXVIII. — George M., aged 35, was admitted to Bellevue Hospital on May 2d. He had been sick, as he expressed it, " off and on " for nearly a year. For the past month he had be^n^-^miligey^^^^l^^^ouse and unable to the Bronchial Glands. 71 work. There were well-marked symptoms of cardiac dis- ease and an excessive enlargement of the liver and spleen. Two weeks after admission the patient died, becoming in- tensely jaundiced towards the last. Autopsy, made twelve hours after death. — The lungs were perfectly normal, with the exception of some old adhesions over the right side, especially at the apex, and an encysted, fibrinous-looking nodule the size of an olive at the left apex. The heart showed the lesions of an acute endocarditis en- grafted on a chronic lesion. The liver weighed eleven and one-half pounds and was amyloid, as was also the spleen. No tubercular lesions were found anywhere in the body, with the exception of the nodule at the apex, which pre- sented the appearance of a healed phthisis. Ten bronchial glands of normal size and, with the exception of two, of normal appearance, were removed. On opening these two glands, small, cheesy masses were seen scattered throughout their substance ; and, by squeezing, they were made to pro- trude as " little white worms." Eight of the enlarged mesen- teric glands were also removed. Inoculation Experiments and Results. — Four cubic centi- metres of the fluid obtained from the bronchial glands were injected in the usual manner into a healthy rabbit, and an- other rabbit was also inoculated with equal amount of fluid obtained from the mesenteric glands. Six weeks later the first rabbit was killed and all the organs were found to be healthy. A small mass was found attached to the ribs at the point of puncture. There were no adhesions, and an exam- ination showed it to be composed entirely of new connec- tive tissue. The rabbit inoculated from the mesenteric glands was also killed and all its organs were found in a normal con- dition. A microscopical examination of the nodule found in the lungs showed the ordinary appearances of a tubercular pro- cess which had been recovered from by new connective- tissue formation. The experiments proved, first, that if the bronchial glands had ever contained tubercle bacilli, they were free from them at the time of death, and that the white, cheesy masses found in two or three of the glands were non- Digitized by Microsoft® 72 Primary Tuberculosis of tubercular in character; second, that the mesenteric glands contained no tubercle bacilli. Case XXIX.— Henry F., aged 48, died after a pro- longed sickness of advanced cardiac and Bright's disease. Autopsy, made twenty hours after death. — There was car- diac hypertrophy and dilatation, with extensive lesions of the mitral valves. The lungs were normal, with the excep- tion of an infarction, the size of a small apple, in the lower right lobe, which had gone to the formation of an abscess. The liver and kidneys were cirrhotic ; all the bronchial glands were very much enlarged, soft, and extensively pigmented. Inoculation Experiments and Results. — Three cubic centi- metres of the black fluid obtained from ten of the glands were injected into a healthy rabbit. At the end of six weeks the rabbit was killed. A circumscribed, dense, pig- mented mass was found on the surface and extending for a slight distance into the substance of the lower lobe of the right lung. There were no adhesions, but most of the lym- phatics of the pleura on the right side were filled with pig- ment. The mass in the lungs was due to the needle piercing the lung substance and setting up an inflammation which ended in the formation of new connective tissue. All the organs were normal and no evidence of tuberculosis was found. I think the excessive enlargement of the bronchial glands in this case was due to the absorption from the encysted abscess cavity. They certainly contained no tubercle ba- cilli, nor did the rabbit show any of the lesions of septic infection from the glands. Case XXX. — Nora S., aged 70, who had been a helper in the wards of Bellevue Hospital for over a year, suddenly became unconscious and died from what was supposed to be apoplexy. Autopsy, made fourteen hours after death. — The cause of death was found to be due to a hemorrhage into an old, de- generated area in the corpus striatum. There was also ex- cessive cardiac dilatation. The lungs were perfectly normal, with the exception of excessive pigmentation. Eight bron- Digitized by Microsoft® the Bronchial Glands. 73 chial glands were removed, one of which was very large and fibrinous-looking ; all were dense and pigmented. Inoculation Experiments and Results. — Two cubic centime- tres of the fluid obtained in the usual way from the glands were injected into a full-grown, healthy guinea-pig. One week after inoculation the animal began to lose flesh and appetite, and failed steadily in health for two months after inoculation, at which time it was killed. The animal was very much emaciated. On opening the chest both lungs were found studded with tubercles, and tubercles the size of a pin's head were seen all over the costal and diaphragmatic pleura and both surfaces of the pericardium. The liver and spleen were completely de- stroyed by cheesy, tubercular masses. This is one of the most interesting cases collected. The woman was under observation for a long time and always seemed to be perfectly well, although somewhat enfeebled b)' her age. She spent her time in a ward where there were many phthisical cases. She died suddenly. Her lungs were normal, being even entirely free from adhesions. One bronchial gland was very much enlarged and was the seat of a chronic inflammation. This gland unquestionably, and possibly also some of the normal-sized ones removed, con- tained tubercle bacilli, as the animal inoculated died of well-marked general tuberculosis. General Review of Cases. — The period of experimentation extended over eighteen months. During this time the sub- stance from the bronchial glands removed from forty -eight persons was injected into healthy rabbits, which were kept under the best hygienic conditions. The majority of these forty-eight persons died suddenly, and none of them after a prolonged illness. A careful examination of all the organs in such persons failed to reveal any old or recent tubercular process. Eighteen of the forty-eight experimental inocula- tions were failures, for the rabbits died before any tuber- cular infection, if present, could have shown itself. Conse- quently these have not been included in the thirty from which my conclusions are drawn. The bronchial glands re- moved from eight of the thirty cases contained tubercle ba- cilli, as demonstrated by the development of tuberculosis in Digitized by Microsoft® 74 Primary Tuberculosis of rabbits inoculated with the substance of the bronchial glands. In five of these eight cases death occurred sud- denly from accident, while they were in apparent health, and none of the thirty, with two exceptions, were sick more than ten days. Three of the twenty-one negative cases caused lesions in the inoculated rabbits, which resembled so closely tubercular nodules that the inoculation of a second rabbit was necessary to determine their non-tubercular char- acter. The results of the inoculation of the gland substance in Case II. illustrates in a striking manner the possible pre- sence of tubercle bacilli in the bronchial glands in persons of strong and vigorous physique and in the fullest physical activities of adult life. Cases I. and V. furnished, in the fluid obtained from their bronchial glands, enough tubercle bacilli to set up in the rabbits inoculated a miliary tubercu- losis, from which pure cultures of the tubercle bacilli were made on sterilized potatoes. Cases VII. and X. both produced in the rabbits inoculated a profuse pulmonary tuberculosis. While the gland substance in Case XV. contained tubercle bacilli, the lesions found in the first rabbit could only be rendered positive by the inoculation of a second rabbit. In almost all the cases inoculated the pigment contained in the glands was seen in the rabbits at the point of injection, and it spread out along the lymphatic channels, which were beau- tifully mapped out by it. There is one point we must not lose sight of in the analysis of these cases : they were principally taken from the lower walks of life, most of them hospital cases. Such persons must necessarily be surrounded by impure air, and are more exposed to the possible inhalation of tubercular bacilli in large numbers than are those who, in large and well-ven- tilated apartments, live under better hygienic surroundings. Case XXVII. furnished bronchial glands from a person suddenly stricken with apoplexy, in apparently perfect health. These glands contained tubercle bacilli, which would have caused the death of the inoculated animal from tuberculosis if it had been permitted to live long enough. Case XXX. is exceedingly interesting from two facts: First, the patient had spent the last year of her life in Digitized by Microsoft® the Bronchial Glands. 75 apparently good health in a hospital ward, surrounded by phthisical patients. She died suddenly. Second, her bronchial glands furnished such an intense tubercular virus as quickly to overpower the inoculated animal by a general tuberculosis. BIBLIOGRAPHY. Quain, Richard : British Medical Journal, 1878, p. 863. Barely : L'Adenopathie Tracheo-bronchique. West : Diseases of Infancy and Childhood. Ancell, Henry : Treatise on Tuberculosis, p. 232. Babes : Gazette des Hopitaux. Paris, 1888. Bollinger : Deutsche Medicinische Wochenschrift, 1888, xxx. Williams: Pulmonary Consumption. Walsh : Diseases of the Lungs, 4th edition. Huguenin ; Ziemssen's Cyclopaedia. Fuller, H. W. : Diseases of the Chest, London, p. 395. Jenner : Times Gazette, vi., p. 426. Furth : Internat. klin. Rundschau, ii., 973. Wien, 1888. Edwards, Blanche : La France Medicale, ii., 1888. Krumbholz, H.: Cor. Blader Allg. arzte. Ver., v. Tiiringen. Loeb : Jahrbuch fiir Kinderh., 86, xxiv., 353. Digitized by Microsoft® SIX CASES OF SUPPOSED HYDROPHOBIA. J. M. BYRON, M.D., Director of the Bacteriological Department of the Loomis Laboratory. Since 1886, when the first antirabic inoculations began in Paris, the alleged cases of hydrophobia have increased to an alarming extent wherever an antirabic institute has been established. This fact has attracted the attention both of the medical fraternity and the public. Whether this increase is due to a better knowledge of the disease, or to the effect upon the public mind created by the ever-increasing num- ber of patients who upon the slightest pretext apply for treatment at the Pasteurian institutions, or to the neglect of cauterization of the wound in the hope that the antirabic inoculation would prove sufficient to counteract the effect of the virus, or to the inoculations themselves, can only be decided by a close clinical observation of each case, together with the subsequent experimentation upon animals. Hy- drophobia is a very rare disease. In some countries, where prophylactic measures are taken by the authorities, such as muzzling of dogs, etc., hydrophobia has disappeared from the list of causes of death. Germany may be cited as an example of this kind, where the mortality from rabies has been zero from 1885 to date. In other countries. West Indies, South America, etc., where the laws are not so strict regarding muzzling of dogs and where these animals are very nume- rous, hydrophobia is very seldom talked about. It is ad- mitted by most authorities to be an infectious disease, which starts from unknown causes in dogs, and is propagated by their bite, the virus being contained in the saliva. Dubuo6' was the first to demonstrate the localization of the virus of rabies in the central nervous system, principally in the bulb. Of this fact the Pasteurian school has profited to produce ' " De la Physiologic pathologique et du Traitement rationnel de la Rage," Paris, 1879. Digitized by Microsoft® Six Cases of Supposed Hydrophobia. j'j artificially the disease by inoculating portions of the central nervous system of infected animals under the dura mater of healthy rabbits or other animals, thereby increasing its virulence to such an extent as to reduce its period of incuba- tion from three or four vi^eeks to six or eight days. The object of this paper is not to treat of the Pasteurian methods, well known to the medical profession, but incidentally to mention the fact claimed by this school regarding the ab- solute certainty of artificially producing hydrophobia in animals inoculated subdurally with parts of the central nervous system of animals whose death has been due to rabies, so that any suspicious case of death that has pre- sented symptoms resembling those of hydrophobia can be at once proven to be or not to be a genuine case of rabies. And as far as experimentation goes, it is the only means that we possess to-day of distinguishing the cases of lysso- phobia from those of genuine rabies. So that if an animal inoculated under the dura with portions of the medulla of the suspicious case does not die with the characteristic symptoms of laboratory rabies, it should be considered a case of lyssophobia — a disease whose symptoms greatly resemble those of the so-called genuine hydrophobia, but which is not caused by the inoculation with the virus, but by a hyperexcitable nervous system and an unbal- anced mind, or some other disease, with perhaps the casual bite of some animal supposed to be rabid. Occasionally, though, rabbits may die with ascending paralysis resembling rabies, when inoculated under the dura with dead animal matter or other substances, as has been observed by some experimenters ; but in such cases the central nervous system of the animal does not reproduce the disease in others. The six cases which I have experimented with, excepting one, belong to the category of false hydrophobia. These include the most important cases that have presented themselves to the medical profession of this city and its vicinity during the past two years. I have not been able to collect a com- plete clinical history of each case, but am in the possession of the reports of the post-mortem examinations, made, in the majority of instances, by able pathologists in this city. The experiments have been performed by me in this Labo- Digitized by Microsoft® 78 Six Cases of Supposed Hydrophobia. ratory under the most favorable circumstances, and follow- ing strictly the principles established by Pasteur in this branch of medical science. A portion of the central ner- vous system has either been inoculated immediately after the autopsy, or kept in neutral glycerin for a fevir hours, which it is known does not affect the virulence of the cord. The emulsions were prepared with sterilized water, and a few drops injected under the dura mater of at least two healthy rabbits at a time, and the rabbits kept under obser- vation for at least a period of ninety days. Case I. — A lad 15 years of age was bitten, six weeks before his death, on the wrist, by a dog supposed to be mad. During the period that elapsed from the time of his being bitten until four days before his death the boy seemed to enjoy perfect health. The first manifestation of the disease was violent delirium, which continued to grow worse when he entered Bellevue Hospital in the last days of December, 1889, and was transferred to the insane pavilion. Here he became rapidly worse, throwing himself from the bed to the floor, and showing signs of the delirium of persecution. His eyes had a wild look, and his conjunctivas were con- gested. At times he would seem for a moment to be ra- tional ; his face would become alternately flushed and pale. Knowing nothing of the previous history of the case, Drs. Fitch and Douglass, the physicians in charge, made a diag- nosis of acute mania. The patient, during the last day of his life, vomited profusely, and the stomach was washed out. While sitting in a chair he suddenly died, presenting some symptoms of asphyxia. Difficulty in swallowing fluids had not been obserted, as none had been administered. The autopsy was performed twelve hours after death by Dr. Henry P. Loomis, and, with the exception of an engorgement of the cerebral vessels and slight pulmonary oedema, no lesions of any kind in the organs of the body could be found. Microscopical examination of the lower portion of the me- dulla showed a congestion of the capillaries, with no struc- tural change. Portions of the cord were removed and kept in neutral glycerin for about twelve hours, and two healthy rabbits inoculated by the usual method under the dura mater with an engd^^on^ac^.^^y^^c^^the cord, after thor- Six Cases of Supposed Hydrophobia. "jg oughly washing it with sterilized water and using only the central portion, which had not been acted upon by the gly- cerin. Both rabbits enjoyed apparently good health until January 13th, 1890, when one became morose, refused food, and died after presenting the symptoms of laboratory rabies, i.e., progressive paralysis beginning in the hindquarters and advancing rapidly until the muscles of respiration were affected and the animal died from asphyxia. The other rabbit continued to live in perfect health for over ninety days. The cord of the rabbit which died with symptoms of hydrophobia was inoculated into a second series of rabbits. One of these died after the twenty-third day, but did not present the characteristic symptoms of laboratory rabies. A third series was inoculated from the cord of this rabbit, and both lived over the allotted period of ninety days with- out presenting any pathological manifestations. I regret to say that, by a hasty judgment on my part, I told Dr. H. P. Loomis that I thought this a case of hydrophobia, basing my opinion on the symptoms presented by the first rabbit that died after inoculation. But subsequent experi- ments have proven to me that it was not a case of genu- ine rabies, for, if so, all the re^inoculations should have produced an uninterrupted series of rabid animals. Such cases, although rare, are not uncommon. Case II. — On Sunday, November 30th, 1890, George Kendall, of Arlington, N. J., took up a strange dog which he found on the road. The dog had broken a leg, and, on being examined, bit Kendall on the lower lip, taking away a piece of the skin. The wound bled profusely, and was skil- fully cauterized and dressed antiseptically. The wound healed nicely, and the man apparently enjoyed good health until a week before he was taken to the Chambers Street Hospital. At this time, while in the company of other men, he was about to drink a glass of ginger ale when he was seized with spasms and was unable to swallow the beverage. He presented no other symptoms until the day he was taken to the hospital, a week after, when, while per- forming his morning ablutions, he was seized with a series of spasms, one after the other in quick succession. Help was called by his frightened employer, and he was bound Digitized by Microsoft® 8o Six Cases of Supposed Hydrophobia. and carried to the Chambers Street Hospital of this city. There he remained either unconscious or in convulsions for twenty-four hours, when he died. The autopsy was per- formed by the late Dr. Stevens, assisted by Dr. Ira Van Giesen, on January 25d, 1891. Body well nourished ; rigor mortis well marked. Thoracic Cavity. — A few old adhesions in the left side ; right side normal. Lungs natural color ; substance of lung congested and red. Mucous membrane of bronchial tubes congested, and the tubes filled with mucus. Heart. — Valves normal. Sub-endocardial haemorrhages in the left ventricle. Abdomen. — Peritoneum over the intestines congested. No fluid in the abdominal cavity. Liver. — A little nutmegged and congested. Spleen. — Normal in size ; slightly congested. Intestines. — Large intestine congested ; small intestine con- gested. Head. — Dura mater congested ; cerebral substance diminished in consistence. Sections of the brain tissue show no change in white or gray matter. Cord normal ; slightly congested. Portions of the medulla oblongata were removed and kept in neutral glycerin for about fourteen hours, emulsion- ated, and inoculated subdurally in two healthy rabbits. Another rabbit was inoculated in the ej'e by Ferran's method, and another hypodermically. All of these enjoyed perfect health over the allotted period of ninety days. Case IIL — On February loth, 1891, Dr. Stevens, not feel- ing well, called in Dr. Wright. He said that in the latter part of December, 1890, he had scratched his finger while performing an autopsy. This was followed by some lymph- angitis, fever, chills, etc., from all of which he soon recovered. His general health had been good, but he thought he had been working too hard, and therefore felt tired. On January 23d, 1891, he had performed an autopsy on a subject at the Chambers Street Hospital, who, it was thought, had died from hydrophobia. The scratch on his finger had healed except for a space that was covered by a small scab, and where there was still a little pus. He did not remove the spinal cord, but his hands were immersed for some time in the bronchial and pulmonary seeretions. On January 26th he underwent a series of antirabic injections. These ended Digitized by microsofm Six Cases of Supposed Hydrophobia. 8i February gth. Daily they were given about the loins sub- cutaneously. On February 5th he had a headache and felt languid. The next day he found he was very tired on go- ing upstairs. On February 7th he noticed there was a tingling, numb sensation in his feet, and some heavings. On the 8th the loss of power was more marked. On the 9th he had the tingling sensation in his hands, and he began to have a sense of discomfort in the muscles of the back and limbs. When seen by Dr. Wright on the loth he had no febrile disturbance and was pretty comfortable. His gait was ataxic, and it was with some difficulty that he could walk across the room. The loss of power in his hands was not appreciable ; sensation was intact. Sphincters, pupils, etc., were normal. That day he succeeded in going down- stairs. Pulse was about 56 to the minute, which he said was normal. The case was regarded as one of anterior poliomyelitis. Dr. Amidon saw him in consultation later on that day, and he coincided in the diagnosis ; but neither physician felt en- tirely satisfied with this diagnosis, and looked on the case with great anxiety. The loss of power in the legs slowly and gradually became more marked up to the end, the exten- sors being more affected than the flexors, until finally the paralysis became almost complete. Paralysis of the arms set in later, following the same course, but did not become com- plete. Reaction of degeneration was marked in a few days, as was also the atrophy. On the nth the feeling of discom- fort increased to actual pain, which in a few days became extreme. It seemed to be at first fairly well localized along the great nerve trunks, as the sciatic, etc. There were also along the back and the shoulder blades areas of very marked muscular hyperaesthesia. These pains at first seemed par- oxysmal, coming on at 8 or 9 A.M. and disappearing in the afternoon, and for two or three days antedated themselves by an hour. The patient was given fluid extract of ergot, drachm j., t. i. d., with eight grains of calomel. On the loth he was, also, cupped over the spine. On account of the periodicity of the pain, the patient was given large doses of quinine by injection and by the mouth, with beneficial effect at first. He was also given some phenacetin. All ^ Digitized by Microsoft® 82 Six Cases of Supposed Hydrophobia. drugs, however, except morphine hypodermically, failed to have any marked effect on the pain. The case was then regarded more as one of peripheral neuritis than a polio- myelitis. He never had spasms of any kind. On the 12th or 13th of February he complained of a numb sensation in his mouth and lips ; a few days later his voice became some- what nasal. About the i6th he began to have some diffi- culty in deglutition, due to muscular paresis ; nevertheless, until the last few days he was able to take enough nourish- ment by the mouth. About the 17th there began to be some acceleration of the pulse, this becoming more marked on exertion. A slight rise of temperature occurred on the 19th. Respiratory difSculty began on the 20th, and gradually increased, until it was necessary to begin artificial respiration at about 9 P.M. on the 21st ult. His heart stopped beating about 11:30 on the 22d of February, after fifteen hours of continuous artificial respiration and oxygen in- halations. During the last week he had retention of urine, compelling the use of the catheter. There was no inconti- nence of either urine or faeces. His intelligence was perfect until the last, excepting for the stupor due to the large amount of morphine given hypodermically. The duration of his illness from the headache on the 5th was seventeen days. His temperature was elevated during the last two or three days. Nothing abnormal was found in his urine. A microscopical examination by Dr. Van Giesen showed no lesions of any kind in the spinal cord. The autopsy was performed by Dr. Farquhar Ferguson,, and, on his authority, I can state it was entirely negative. Unfortunately the body had been embalmed four hours be- fore the post-mortem was performed. Portions of the central nervous system were handed to me by Prof. W. G. Thompson, and were kept in neutral glycerin for thirty hours, and then inoculated by the usual method in four healthy rabbits, two subdurally, one in the eye, and one under the skin. All these animals lived in perfect health over ninety days. I beg to remark that, according to the authority of Dr. Ferguson, all the organs were permeated with the embalming fluid, having been injected through the carotid, although I was unable to detect the existence of any Digitized by Microsoft® Six Cases of Supposed Hydrophobia. 83 of the usual metallic substances used in the process of em- balming. Case IV. — To Dr. Ruhl, who attended the patient during his illness, I owe the clinical notes of this case : A young man, 27 years of age, healthy, temperate, was bitten in the hand by a dog, four weeks previous to May 20th, 1891, when the first symptoms began. He was very excited when he called at Dr. Ruhl's ofHce, who prescribed for him. The following day he grew worse, and, according to his family, began snarling, barking, etc. It is to be noticed that the friends and family had been speaking about the danger of a dog's bite and the agonies of hydrophobia two weeks previous to the outburst of the first symptoms, and that the patient not until then gave any thought to the bite. The man died the 21st of May, nineteen hours after his call at Dr. Ruhl's office. On May 21st last, through the kindness of Deputy Coroner Jenkins, I was invited to perform the autopsy on the case, the report of which is as follows : Rigor mortis was well marked ; hypostasis rather pro- nounced. The right hand presented a few abrasions, not very deep, which I was told were the marks of the dog's bite. Thorax. — A few old pleuritic adhesions were found at the apices of both lungs. Left lung normal ; right lung, lower lobe presented a large spot of lobar pneumonia, possibly produced by introduction of food. Larynx. — Rather con- gested, and filled with bloody froth. Heart normal. Blood dark and fluid. Abdomen. — Liver slightly congested, with a very slight cirrhosis. Spleen normal. Stomach contained a small amount of thick mucus, and presented in the large cul-de-sac a few heemorrhagic patches. Kidneys congested ; size normal ; capsule adherent, indicating slight interstitial changes. Other organs normal. Head. — The dura and pia mater slightly congested ; sub- stance of the brain in a healthy condition. Portions of the medulla were inoculated, after the usual method, in two rabbits on May 22d, 1891. One of these rabbits died on the nth of June with the symptoms of laboratory rabies. The other rabbit died on the i8th of June with the same symptoms. The inoculations from these Digitized by Microsoft® 84 Six Cases of Supposed Hydrophobia. animals to other healthy rabbits were followed by no re- sults. Case V.— Nine months before his death Lawyer R. S. Bar- tine, of Asbury Park, N. J., was bitten in the hand by a stray cat, who was persecuted by a dog. The wound excited no interest, and healed completely and rapidly. Friends of the man chaffed him about the matter, until he consulted his physician, who gave him some medicine and assured him that he was in no danger at all. He, being a man of a very nervous temperament, allowed the fear of hydrophobia to prey upon his mind to such an extent that he finally was obliged to take to his bed. Shortly after this he was seized with convulsions, which were repeated at intervals, becoming very frequent, until on the evening of June Sth he died. His body was placed in a vault, and, the consent of the family being obtained, an autopsy was performed on June 9th, 1891, by Dr. Frank Abbott, Jr., Assistant in the Microbiological Department of this Laboratory. Some of the relatives of the dead man objected to the opening of the body, but con- sented to allow the brain to be removed. The vessels of the dura mater were found to be congested, and so were those in the pia mater and the substance of the brain. All the arteries were in an advanced stage of athero- matous degeneration. At the base of the brain an extensive haemorrhage was found— about five ounces of blood — possi- bly due to rupture of the basilar artery, but the decomposi- tion was so far advanced that minute examination could not be well made. The vessels were so far degenerated that it was impossible to determine with accuracy the extent of the rupture. Portions of the brain and cord were preserved in neutral glycerin, and subdural inoculations made in four healthy rabbits, all of which are living at the present date. Case VL — The sixth case is that of a small boy who was very badly bitten by a dog in Elizabeth, N. J. The child was taken to the hospital and his scalp wounds dressed, needing about one hundred stitches. The wounds healed nicely, and he was discharged as cured. After a few days he was brought again, presenting severe nervous symptoms, and a diagnosis of hydrophobia made. The child died and the autopsy was Digitized by Microsoft® Six Cases of Supposed Hydrophobia. 85 performed by myself October nth, 1891, a few hours after death. All the inoculations made with the cord of this boy inva- riably produced paralytical rabies in rabbits. This is the only case out of six Which can be considered as hydrophobia. The first case presented symptoms of general nervous ex- citation and died suddenly. Then, some way or other, it was discovered that, a few weeks before the first symptoms ap- peared, a supposed mad dogbit the patient and the suspicion of rabies arose. Nobody knew whether or not the dog died of rabies. The experiments in the laboratory, where one of the inoculated rabbits died with the ascending pa- ralysis which is characteristic of laboratory rabies, accord- ing to the Pasteurian school, lent color to the idea that it was a case of genuine hydrophobia ; but had it been so, an uninterrupted series of rabid rabbits should have been the result. Cases of rabbits inoculated under the dura, which die presenting symptoms of ascending paralysis after some time of apparent health, are not uncommon, but if other rab- bits are inoculated with the cords of the former there is not the continued series that characterizes Pasteur's laboratory rabies. Kendall's case was experimentally absolutely neg- ative. In the village where he was bitten no other case of rabies, either in man or beast, occurred, and certainly it would be a singular coincidence for a mad dog to bite this man alone. Clinically there is no doubt that it was a case of true lyssophobia. Then Dr. Stevens, without any absorbent surface through which the alleged virus of a man who died from the fear of hydrophobia could enter,' has himself inoculated with laboratory virus and dies of ascending paralysis ; while an- other pathologist. Dr. Ira Van Giesen, who assisted him to perform the autopsy, wounded himself with the knife with which he was cutting the portion which in the body of a hy- drophobic is considered most virulent, i.e., the spinal cord, and does not suffer any consequences. Dr. Stevens' body was embalmed, and consequently the experiments lose most of their value, but the clinical history points clearly to the cause of his death. Digitized By microsoft® 86 Six Cases of Supposed Hydrophobia. Galrauth was bitten by a dog which was afterwards kept under observation in this Laboratory for over six weeks without presenting symptoms of hydrophobia or any other disease. Two rabbits inoculated with the man's cord died with the paralysis resembling laboratory rabies, but re-inoc- ulations in a second series of rabbits proved it to be, not a case of hydrophobia, but of lyssophobia. Mr. Bartine's case is the most striking of all. Nine months elapsed between the time he was bitten and the ap- pearance of the first symptoms of hydrophobia — an extraor- dinary period of incubation. The autopsy proved him to have died of haemorrhage of the brain, due, perhaps, to the increased vasal tension induced by his mental condition. Four of these cases followed one another in rapid succes- sion during a time when the public press occupied its valu- able space with endless columns relating the agonies and deaths by hydrophobia. These experiments were undertaken to investigate such cases as were more prominent, with the results stated. Had the diagnosis of hydrophobia remained undisputed a few more cases would probably have been added to the statis- tics of death by rabies. Digitized by Microsoft® Digitized by Microsoft® •snupijq rjda'l }0 a-inni" 3"id •siuiuioq siuinSuBS Buciij , 1) ^i'35»«5fcI-T3;0'V5\^S<'v 'O _«■ s ^0 '5_-'ji®-=,'^A^f tumi^iP^^B ^""11 ^fecf by Microsoft® PURE CULTURES OF LEPRA BACILLUS. J. M. BYRON, M.D., Director of the Bacteriological Department of the Loomis Laboratory. Having had the opportunity, in the month of June lastyear, to obtain a Tew nodules of leprosy tissue from a patient at Charity Hospital, I made a series of inoculations in animals and artificial cultures. The fresh tissue was made into a fine emulsion in sterilized water, and inoculated subcutane- ously into four animals, two guinea-pigs and two rabbits. One of the rabbits and one of the guinea-pigs were also inoculated in the anterior chamber of the eye. Neither of the animals contracted the disease, and in two the au- topsy did not present any abnormality. One of the guinea- pigs, which is still living, presented after a few days a very large abscess at the point of inoculation, but the pus and the shreds of necrotic tissue discharged did not show the bacilli of leprosy. Out of the same patient twelve tubes were inoculated with lymph obtained from the wound made, and kept in an incubator at 37° C. After three days ten out of the twelve tubes presented on the surface of the agar many small colo- nies, which proved to be of a bacillus resembling the tuber- cle bacillus in both form and microchemical reactions. Pure cultures of the bacillus of leprosy, or at least claimed to be such, were obtained for the first time by Bordoni-Ufire- duzzi. Other experimenters have also claimed this honor, although in not a single case has the disease been reproduced in animals by inoculation of the pure cultures. Such a fact, however, would not very seriously impair the pathogenic character of the germ towards man, as other well-estab- lished micro-organisms which are pathogenic in man do not transmit the disease to animals. The inoculations which I made in animals "^j^ti^^^ ^^SmM^ ^ ^''P^'' '^""''^ ^^"^ 88 Pure Cultures of Lepra Bacillus. to prove that the bacillus of leprosy is not transmissible, at least to rabbits and guinea-pigs. The bacilli of leprosy, as is well known, are abundantly found in the granulomatous new formations of leprosy, in the pus and detritus of leprosy ulcerations, etc. They have not been found in blood, saliva, urine, etc., of lepers, confin- ing themselves only to the tissues which they invade. The bacilli appear as small, curved rods, about three /< in length and 0.2 yM wide, presenting the appearance of short chains of beads, strikingly resembling tubercle bacilli, from which it would be almost impossible to differentiate them, except by the characteristic reactions. In pure cultures the rods are rather thicker and many have their extremities club- shaped ; probably these are only early involution forms of the rods when artificially cultivated. To obtain a pure culture of leprosy bacillus the skin in- volved is washed first with soap and water, then with i : i,ooo solution of bichloride of mercury, next rinsed with alco- hol, and dried with ether. A deep scarification is made, enough to reach the diseased tissue. The little wound is squeezed, and with a sterilized platinum loop a small drop of the lymph which oozes is transferred to the surface of an agar tube. The best formula that I know for preparing agar suited for this kind of culture is the following : Ordinary bouillon, loo parts; glycerin, 6 parts; rock candy, 0.5 part; agar, 1.5 parts. Boil until the agar is com- pletely melted, neutralize, and filter. Separately dissolve four parts of gelatin in twenty parts of bouillon, filter, neutral- ize, and add to the filtered agar. The adding of the solu- tion of gelatin after the preparation of the agar is to avoid its being transformed into paragelatin or non-coagulable gelatin, which happens if it be boiled for a very long time, as is necessary when preparing agar. This culture medium answers admirably not only for lepra bacillus, but also for tubercle bacillus, gpnococci, and other germs which scarcely grow or do not grow at all on ordinary media, and, best of all, discards from laboratory use blood serum, which in- volves so many tedious manipulations to obtain it in a pure condition. The infected tubes are kept in an incuba- tor at a temperature of 37° C. .After thirty-six or forty-eight Digitized by Microsoft® ■' ■^ ^ Pure Cultures of Lepra Bacillus. 89 hours, minute, star-like, grayish-white little dots appear on the surface of the agar. These little colonies spread slowly for two or three days, when the growth seems to stop. Then they are about the size of a small pin's head. From this point the growth is extremely slow, and it takes months before they are as large as in the tube of the photograph. In plate or dish cultures (Petri's) kept at 37° C. the colonies appear as round, irregular, whitish dots, which under a low power look like little, irregular heaps, darker in the centre, of a yellowish color on the outside, limited by wavy edges. The colonies grow both upon the surface and in the depth of the agar. Linear cultures obtained by passing an infected needle upon the surface of an agar tube are, 1 think, quite characteristic ; but care should be taken that no condensa- tion water collects in the tube, otherwise the true appear- ance of the culture is impaired. At the end of thirty-six or forty-eight hours a slight line of grayish-white color begins to appear along the track of the needle, and from here the growth continues to extend outward, producing after many weeks a beautiful fern-like appearance, almost transparent, of a grayish-white color, and about one-sixteenth of an inch wide. The culture is slightly prominent in the centre, from which point radiate little rib-like heaps that run to the edges. Old potatoes dipped in a solution of five per cent glycerin and one per cent of candy sugar are also a good, thriving ground for this germ, the only difficulty being in keeping the potatoes from drying and to select those having a neutral reaction. The growth is absolutely colorless, and cannot be detected except by preparing cover glasses from the vicinity of the needle track. Cultures of lepra bacilli in distilled sterilized water do not thrive, but if the water con- tains a small amount of organic matter the germs will multi- ply with relative rapidity. Those experiments I have not yet finished. Cultures in bouillon with the addition of gly- cerin and sugar grow luxuriantly during the first two weeks as small, whitish, glutinous flakes, which float in the clear medium and which afterwards subside to the bottom. The best temperature for the growth of the lepra bacillus seems to be that of the human body. The growth, though, even under the most favorable conditions, is exceedingly Digitized Dy Microsoft® 90 Pure Cultures of Lepra Bacillus. slow. The resisting power of lepra bacilli to the action of the exterior surroundings is as great as, if not superior to, the tubercle bacillus. I have kept dry cultures of this germ at the temperature of 98° F. for six months, and when re-inoc- ulated in fresh media a new and vigorous growth sprang up from them. The bacilli, when stained, present under the microscope some oval spaces which do not take the dye, giving the ap- pearance of a little chain of beads. Such spaces are con- sidered by most bacteriologists as spores. Experiments on animals in order to produce this disease artificially have been performed by several observers, among them Damsch and Vossius, who state that they have been able to produce local nodules of leprosy, but not a general infection. The experiments which I have undertaken are yet far from com- pletion. The exceedingly long period of incubation and the slow evolution of the disease forbid me from giving to-day any conclusions, which I am sure would be premature and without any practical value. Later on perhaps I may be able to publish further details as to the results of experimen- tation on animals. Digitized by Microsoft® ON THE DETECTION OF QUININ IN CHEMICO- LEGAL CASES. R. A. WITTHAUS, M.D., Professor of Chemistry^ Physics, and Hygiene. The literature of toxicology contains but little with re- gard to the cinchona alkaloids, as, owing to their compara- tively non-toxic character, they are but rarely the objects of chemico-legal investigation. A recent case, however, has shown that the question of the presence or absence of quinin in a cadaver may become one of great importance. It was the claim of the defence that a capsule taken by the deceased, shortly before the manifestation of symptoms of morphin poisoning, contained four and one-half grains of quinium sulphate and one-sixth grain of morphium sulphate. But on an analysis of the stomach and intestines being made, fifty-three days after death, the reactions for morphin responded distinctly, while only negative results were obtained with the three most important quinin reactions. If the tests for quinin are reliable, their failure in a case such as this, associated with an affirmative result with the morphin tests, would go far toward supporting the theory on the part of the prosecution that the contents of the capsule were not as claimed, but that morphin had been substituted. I therefore sought to determine the limits of delicacy and reliability of the three principal tests for quinin : that by taste, that by fluorescence, and the thalleiochin reaction,, under different conditions. Affirmative results with these three tests are not suffi- cient to demonstrate the presence of quinin, to the exclusion of quinidin, as the two alkaloids behave toward them in much the same way. ^^hftl^^l^bYl^m^m^^ them, we believe. 92 On the Detection of Quinin amount to a demonstration of the absence of both quinin and quinidin in quantity beyond the limit of delicacy. The Test by r^j/"^.— According to Fliickiger,' the bitter taste of quinin is no longer detectable in a solution of the dilution of 1 : 100,000. Presumably in this matter the personal equa- tion of the delicacy of taste, which differs materially in different individuals, constitutes an important element. I have been unable to detect bitterness in a solution of a qui- nin salt of I : 50,000, but could detect it faintly in one of i:20,ooo, and distinctly and persistently in one of 1:5,000 whether the solution was tasted directly, or dried upon paper. The Fluorescence Test.— In 1862 Fluckiger' pointed out that the blue fluorescence exhibited by acid solutions of quinium sulphate, when examined in direct sunlight, against a black surface, afforded a very delicate test for the presence of that alkaloid, being still visible, under favorable condi- tions, in a solution of i : 500,000. Subsequently Kerner ' claimed that, when examined by the light from a Geissler tube, the fluorescence is still visible in a solution of the sulphate of the dilution of 1 : 8,000,000. Following Kerner, who states that " the hydracids of the halogens prevent the reaction entirely," most writers as.sert that solutions of the chlorid are destitute of fluorescence ; in some cases qualifying the statement by the supposition of the presence of free hydrochloric acid. Thus Prescott* says: "The hydracids of chlorine, bromine, etc., the cyano- gen hydracids, and thiosulphuric acid in union with quinine, do not give fluorescence." Husemann and Hilger" do not include the hydracids in the list of substances which prevent fluorescence : " The solutions of its salts (quinin) . . . show strong blue fluor- escence, which, however, is prevented by sulphocyanic, ' Ztsch. f. an. Chem., 1872, xi., 318. ' Schweiz. Ztsch. f. Pharm., vii., 22 ; ex Ztsch. f. an. chem., 1862, i., 373. • Arch. f. Physiol , ii., 200 ; ex Ztsch. f. an. Chem., 1870, ix., 134, * " Organic Analysis," N. Y., 1887, p. 130. e "Pflanzenstoffe," -^f^^^^^^y MiCrOSOft® in Chemico-Legal Cases. 93 ferro-, palladio-, and platinocyanic acids, and by hyposul- phites." In the Harris trial, Prof. Theo. Wormley, being asked a long hypothetical question, answered : "As I understand the question, the residue is dissolved in a small quantity of dilute hydrochloric acid, then examined for fluorescence. Under these conditions the failure to produce fluorescence was not conclusive evidence of the absence of quinine. This conclusion of mine is the general statement of the authors, that oxy salts of quinine are the fluorescent salts. The state- ment in regard to chloride is specifically made in the United States Dispensatory," in which it is stated that a pure saturated solution of quinine chloride is destitute oi fluorescent prop- erties, but a dilute — certain dilute solutions — may present fluorescence, provided there is no free acid present. Ques. — What can you say of your own experience and obser- vation in regard to that? Ans.—T\\2X solution of quinine — any form of chloride in the presence of free hydrochloric acid — is not fluorescent." Kerner's original statement, quoted above, is undoubtedly true, but it is not the whole truth, for while a large- excess of hydrochloric acid does destroy the fluorescence of solu- tions of the chlorid, a slight excess not only does not de- stroy it, but renders it more brilliant. To determine the value of the fluorescence test the follow- ing experiments were made. Except where otherwise stated, the solution used was 10 cc. in bulk, contained in test tubes one-half inch in external diameter, carefully selected of clear, white glass, and supported in a frame open on one side only, and blacked on the others, and the color observed by looking down into the mouth of the tube. Series A. — A solution made by dissolving o.i gram of alka- loidal quinin in i cc. HCl, sp. gr. 1.0503. and 99 cc. H,0. The solution is strongly acid, and contains nearly ten times as much HCl as is necessary to neutralize the quinin. This solution (A) was used undiluted (= i : 1,000), and diluted in the proportions of 1 : 2,000, i : 5,000, r : 8,000, i : 10,000, I : 15,000, I : 20,000, I : 50,000, i : 100,000, and i : 500,000. ' See U. S. Disp., 15th ed., p. I2iq. Digitized by Microsoft® 94 On the Detection of Quimn When examined by diffuse daylight, at a north window, on a rainy afternoon, fluorescence was distinct in the solu- tions of I : 1,000, I : 2,000, i : 5,000, and i : 50,000 ; intense in I : 8,000, I : 10,000, I : 15,000, and i : 20,000, the maximum of intensity being in the solution of i : 20,000, which was quite as brilliantly fluorescent as a solution of the sulphate with excess of sulphuric acid (series B) of the same degree of dilution. The solution i : 100,000 presented a pale but distinctly recognizable fluorescence; while none could be de- tected in I : 500,000. When examined with a converging^ beam from the zircon light, in a dark room, the fluorescence in this and in the other series was found to be more brilliant, but could not be detected in those solutions which failed to show it in diffuse daylight. The use of this strong artificial light is therefore not an advantage. No fluorescence could be detected in the solution i : 500,000 when placed in flat cell, illuminated with a cone of direct sunlight produced by a glass lens, and examined against a black background. Large drops (0.05 cc.) of the solutions i : 1,000 and i : 5,000 showed very distinct fluorescence when examined in a 3-inch, round- bottom glass evaporating dish in direct sunlight, against a black surface, without a lens. In the solution i : 10,000 fluor- escence was not detected under these conditions. Series D. — A solution made by dissolving 0.105 gram of gummy quinium chlorid, obtained by dissolving the alka- loid in a slight excess of dilute HCl, concentrating over the water bath, and evaporating to dryness in vacuo over HjSO„ dissolved in 100 cc. H,0. This solution (i : 1,000) was neutral in reaction, and was used in the same degrees of concentration as in the A se- ries. In diffuse daylight, under the same conditions as A, pale fluorescence was visible in the concentrations between 1:1,000 and I : 20,000 ; the maximum of intensity being at i : 2,000, which was about equal to that of a solution of the sulphate, with excess of acid of i : 100,000. In the solutions i : 50,000 and I : 100,000 it was not detectable with certainty, and in 1 : 500,000 it was absent. Viewed in direct sunlight, the so- lution I : 50,000 showed a pale fluorescence, and in that of I : 100,000 the blue color was barely visible. When illu- Digitized by Microsoft® in Cliemico-Legal Cases. 95 minated in a dark room with tlie light from a Geissler tube, the concentrations i : 50,000 and i : 100,000 were fluores- cent. Series G. — A solution of 2.1 grams Rosengarten's crystal- lized " hydrochlorate of quinine " dissolved in 100 cc. H^O, constituting i : 50 of the alkaloid. The neutral solution, which was nearly saturated, was ex- amined undiluted, and diluted to i : 100, i : 250, i : 500, I : 1,000, I : 2,000, I : 5,000, I : 8,000, I : 10,000, I : 20,000, I : 50,000, and i : 100,000. In diffuse daylight, under the same conditions as A, i : 50, which was yellow in color, showed no fluorescence ; but in the dilutions i : 100 to I : 10,000 the blue color was distinct, the maximum of in- tensity being at i : 250. In i : 20,000 it could not be detected with certainty, and in i : 50,000 and i : 100,000 it was absent. In direct sunlight fluorescence was very faintly visible in I : 50 and in i : 50,000. Examined with the light from a Geissler tube, it was distinct with i : 50 and pale with I : 50,000. Series H. — The same solution as in G, with the addition of 0.5 cc. HCl, sp. gr. 1.0503, to 50 cc. of the solution. Ex- amined in the same concentrations asG. With diffuse daylight, on the same day as A, D, and G, i : 50 did not show fluorescence, in i : 100 it was discernible, though pale, while in i : 250 to i : 8,000 it was very distinct, the maximum of intensity being at i : 2,000, which was about equal to that of the solution of the sulphate with ex- cess of acid (series B) of i : 10,000. In i : 10,000 the blue was visible, while in the higher dilutions it was not de- tected. In direct sunlight the yellow i : 50 did not flu- oresce, but the blue color was detectable up to i : 50,000, and with the Geissler tube illumination up to i : 100,000. Whether viewed by diffuse daylight, by direct sunlight, or by Geissler tube illumination, in test tube or in capsule, the solutions of series H, containing an excess of hydrochloric acid, were much more brilliantly fluorescent than those of series G of like degrees of concentration, which were neu- tral. From these experiments it appears that neutral solutions of quinium chlorid fluoresce distinctly between the dilu- Digitized by Microsoft® 96 On the Detection of Quinin tions of I : 100 and i : 10,000 when examined by diffuse day- light without prism, lens, flat cell, or other paraphernalia; that the fluorescence is increased in intensity by the pres- ence of a moderate excess of free hydrochloric acid ; and that it is distinctly discernible in direct sunlight in a large drop (0.05 cc.)of solutions between i : 1,000 and i : 5,000 in concen- tration. While Kerner's"fluoroscope," which consists of a Geissler tube within an outer tube, into which the solution to be ex- amined is introduced, may permit of the detection of quinin in solutions of higher dilution than i : 500,000, and may be of service for certain uses, it is not serviceable for medico-legal examinations. Kerner states that by its use the fluorescence, of "very small quantities of liquid, 15 to 20 cc," can be de- tected. Such a bulk of liquid may be a puddle clinically, but toxicologically it is an ocean. Moreover, the apparatus could be used but once with safety, for, owing to its con- struction and its alleged delicacy of i : 8,000,000, there would always be doubt that it had been completely cleansed after having been once filled with a solution of quinin. In a toxicological analysis the conditions would be (and were in the Harris case) those of series D ; the residue of evaporation of petroleum ether from the alkalized watery solution having been evaporated to dryness, dissolved in dilute hydrochloric acid, again evaporated to dryness, and the residue dissolved in about 0.05 cc. of water. Under these circumstances no free acid can be present. Or, if the hydrochloric acid solution be examined without evaporation and resolution, the conditions are those of series A. In either event fluorescence would be detectable in 0.05 cc. of liquid containing i : 1,000 to i : 5,000 of alkaloidal quinin — which represent -j-sVir to TuiTnr of a grain. It would also be visible in more concentrated solutions up to i : 100. Saturated solutions are not frequently met with in toxicological analy- ses. Should they be, the obstacle is easily remediable by dilution. Solutions of quinium sulphate without excess of acid pre- sent only famt fluorescence, not as distinct as that with solu- tions of the chlorid under like conditions. Series C. — A solution ofEimer & Amend's " sulphate of Digitized by Microsoft® ^ in Chemico-Legal Cases. 97 quinine " containing 0.1153 of the salt in 100 cc. H^O was ex- amined undiluted (= i : 1,000 of the alkaloid), and diluted as in series A. A pale blue was observed by diffuse daylight in solutions 1 : 1,000, i : 2,000, i : 5,000, i : 8,000, i : 10,000, i :i 5,000, and I : 20,000, having its maximum of intensity at i : 2,000, which was about equal to that of a solution of the sulphate, with excess of acid (series B) containing i : 100,000. In 1 : 50,000 no fluorescence was observed by diffuse daylight, but it became visible, although very faint, in i : 50,000 and in I : 100,000 in direct sunlight, and with Geissler tube illumi- nation. Series B. — For purposes of comparison a solution was made by dissolving o.i gram of alkaloidal quinin in i cc. of dilute HjSOj, sp. gr. 1.1926, and 99 cc. H^O. The strongly acid solution (i : 1,000) was examined undiluted, and in the same dilutions as A. With the exception that in the dilution ot i : 20,000 the intensity of color in the chlorid solution with excess of acid (series A) fully equals that of the sulphate under like con- ditions, the solutions of this series were more brilliantly fluorescent than those of the others. In the dilutions from 1 : 1,000 to 1 : 20,000 the color was very brilliant, in i : 50,000 it was distinct, in i : 100,000 pale, and in 1 : 500,000 detect- able by comparison with a tube filled with water, in diffuse daylight on a rainy day. The fluorescence of solutions of quinidin sulphate and chlorid with excess of acid is more brilliant than that of the quinin salts under like conditions. Series E. — A solution of o.i gram quinidin dissolved in I cc. dilute HCl, sp. gr. 1.0503, and 99 cc. H^O. Examined in the same dilutions as A-D. In diffuse daylight, under the same conditions as A, the solution i : 1,000 exhibits a pale fluorescence, about equal to that of B i : 10,000 ; i : 2,000 a distinct color, equal to that of B i : 50,000; i : 5,000 to i : 20,000, intense fluorescence, more brilliant than the corresponding dilu- tions of the B series ; 1 : 50,000, a distinct color, equal to those of E 1 : 2,000 and B i : 50,000 ; i : 100,000, a pale blue, equal to those of E i : 1,000 and B i : 100,000. In ^ Digitized by Microsoft® 124 Arsenic Trioxid. of well-defined spaces, in the middle of each of which is a very small crystal, and from this crystal as a centre minute crystals are arranged in close radiating masses, as shown in Fig. 9. In this figure the enlargement is about twenty- five hundred diameters. In the change from vitreous to porcellaneous the mass in- creases in bulk, becomes filled with fine cracks, turns white, and absorbs water. The absorption of water causes partial solution, followed by crystallization, in which a portion of the water is set free to pass into the next deeper layer of the oxid, repeating the process until the entire mass has become crystallized. Digitized by Microsoft® PURE CULTURES OF TUBERCLE BACILLI OBTAINED FROM SPUTUM. FRANK ABBOTT, JR., M.D. The complicated technique of Koch's classic method to obtain pure cultures of tubercle bacilli from tubercular tissues has induced several experimenters to try and obtain them directly from tubercular sputum. In this Laboratory we have attempted several times, with very satisfactory results, the following method, which recom- mends itself by its simplicity. The sputum to be used should contain a very large num- ber of tubercle bacilli and be as fresh as possible. The patient's mouth and throat are rinsed with a i: 10,000 solution of bichloride of mercury, and the expectoration collected in a sterilized shallow dish (Petri's). It is then diluted with about an equal volume of bouillon-glycerin, and kept in an incubator at 37" C. for twelve to twenty-four hours. At the end of this time the temperature is raised to 55° C. and the sputum kept in it for one hour. With a very fine pipette some of the sediment is taken and spread upon several agar- glycerin tubes and kept in the ordinary way at 37° in the incubator. Almost invariably at the end of fourteen to sixteen days the tiny colonies of tubercle bacilli appear to the naked eye as yellowish, round, creamy little dots, which afterward ac- quire the dry appearance of ordinary cultures. Under low power, by using refracted light, the single col- onies are round, fibrillar, of a yellowish-brown hue, with a central nucleus of dark color. From the periphery of the colonies, which is not well defined, radial filaments project outward. By reflected light the colonies have a pinkish color and appear as brilliant, porcelain-like heaps. Digitized by Microsoft® 98 On the Detection of Quinin I : 500,000 no fluorescence was visible, either in diffuse day- light or in direct sunlight. Series F.— A solution of o.i gram quinidin dissolved in I cc. dilute HS,0„ sp. gr. 1.1926, and 99 cc. virater. Ex- amined under the same conditions as E. In the dilutions i : 1,000 to i : 20,000 the fluorescence was most intense, exceeding that of the corresponding terms of the B series in brilliancy. It was still distinct in i : 50,000 and in i : 100,000, but not detectable in i : 500,000. The color with quinidin is more violet than that with quinin. Between the concentrations of i : S,ooo and i : 20,000 solu- tions of quinidin are intensely fluorescent in presence of ex- cess of HCl or of H^SO,. With the oxacid the range of intense fluorescence is extended to i : 1,000. In both cases the color is distinctly perceptible between i : 1,000 and 1 : 100,000. The Thalleiochin Reaction. — This test, as at present applied, consists in the addition, first, of an aqueous solution of bromin, which produces neither color nor precipitate with quinin or with quinidin, and subsequently of ammonium hy- droxid, which produces with those alkaloids a brilliant green, varying in shade from pea green to emerald, and, in highly concentrated solutions, a white precipitate. In some cases a violet color also makes its appearance, either before the appearance of the green, when, after the addition of the bromin water, the drop is exposed to the vapor from a neighboring drop of ammonium hydroxid solution ; or ac- companying and within the green. The extreme limit of delicacy of this test, under the condi- tions in which I have experimented with it, was with a solu- tion of I : 20,000, and the limits of a clear and certain reac- tion at I : 10,000. Dragendorff' gives the limit as i : 5,000; Fliickiger'' as i : 20,000; and Kerner,' using chlorin water, also as I : 20,000. In toxicological investigations, where the question of pre- sence or absence of quinin in a barely weighable residue is to be determined, large bulks of liquid are not at command. ' Ermittl. d. Gifte, 2d ed., 189. " L. c, p. 319. Dibith^ W Microsoft® in Chemico-Legal Cases. 99 Test tubes can be rarely used, and the use of the tubes of 20 to 30 centimetres in length, mentioned by Kerner and Fliickiger, would imply extreme dilution. Moreover, the appearance or non-appearance of a given color can be- judged of with perfect certainty in one drop of 0.05 cc. In the application of the test a drop of the size mentioned should be placed upon a porcelain (or white glass) surface ; a much smaller drop of the bromin solution placed near to it with another stirring rod, and the two mixed with a third. A small drop of the diluted ammonium hydroxid solution is then placed near by, and, after waiting a moment to see if the color appears at the edge of the drop, as it does in solu- tions of medium concentration, mixed with the main drop. The relative proportions of bromin and ammonia used, and the amount of quinin present, exert an influence upon the re- action. To determine the extent of this the following ex- periments were made with the solutions of series A, B, E, and G, mentioned above, using bromin solutions and ammo- nium hydroxid solutions of varying degrees of concentra- tion, the amounts of liquid experimented upon, and of re- agents used, being as nearly as possible the same in all cases. The bromin waters used were : A. A freshly prepared, saturated solution, containing about 3 pts. Br. in 100 cc. B. A, diluted with an equal volume IIjO (= ^ saturated). C. B, diluted with an equal volume HjO (= 3^ saturated). D. C, diluted with an equal volume HaO (= Ys saturated). The ammonium hydroxid solutions used were : A. Sp. gr. 0.9471 = 13.3^ NHs (= % aqua ammonise ff.) B. Sp. gr. 0.9725 = 6.6% NHs (= yi A.) C. Sp. gr. 0.9860 = 3.3^ NH, (= X A.) D. Sp. gr. 0.9927 = ^.^% NHs (= % A.) The results of these experiments can be most concisely expressed in tabular form : Digitized by Microsoft® lOO as o en ^0- On the Detection of Quinin a . 5*2 'S'k o o a& Series A. — Quinium Chlorid with Excess of Acid. A. A. I.D. S.D. S.F. A. B. T D. S.D. S.F. A. C. I.D. S.D. S.t. A. D. I.D. S.D. S.F. o B. A. I.D. S.D. S.F. B. B. I.I. SD. S.F. B. C. I.I. S.D. S.D. S.F. B. D. I.I. I.I. S.D. S.F. €. A. I.D. S.D. S.F. S.F. S.F. C. B. I.I. I.D. S.D. S.D. S.D. S.F. C. C. I.D. I.D. S.D. S.F. S.F. S.F. C. D. I.D. I.D. S.D. S.F. S.F. S.F. D. A. I D. I.D. S.D. S.F. S.F. S.F. D. B. I.D I.D. S.F. S.F. S.F. D. C. I.D I.D. S.F. S.F. S.F. D. D. I.D. I.D. S.F. o Series G.— Quinium Chlorid in Neutral Solution. A. A. W.I. W.I. W.I. I.I. I.I. I.I. S.D. S.D. S.F. A. D. W.I. W.I. W.I. I.D. I.D. I.D. S.F. B. A. W.I. W.I. W.I. I.D. S.D. S.D. S.F. B. T). W.I. W.I. W.I. I.D. I.D. S.D. S.F. S.F. C. A. W.I. W.I. W.I. I.D. I.D. S.D. S.D. C. B. W.I. W.I. W.I. I.D. S.D. S.D. C. D. W.I. W.I. W.I. I.D. I.D. S.D. S.F. D. D. W.I. W.I. W.I. I.D. I.D. S.D. S.D. S.F. o Series B. — Quinium Sulphate with Excess of Acid. A. A. I.I. I.I. I.D. S.D. S.D. S.F. B. A. .... . . . I.F. I.F. S.F. S.F. D. A. .... I.I. I.I. I.D. S.F. o D. D. .... S.D. S.D. S.F. I.I. =: an immediate and intense reaction ; I.D. = an immediate and distinct reaction; S.D. = a distinct reaction, appearing slowly; S. F. = a faint reaction, appearing slowly; O = no color after or during five minutes; W.I. =a green color, and iridescent film before the ammonia is mixed with the drop, and a white precipitate with pea-green flocks when the ammonia is brought in contact. Digitized by Microsoft® in Chemico-Lezal Cases. loi ^-^« ■:5-*>'i '^' s'-^flf f!!!?,;-^ c^"^ ^ :i.i*v' -^ f.^. ^*^ .^^' 9* ^^-.'^ m-'^ ■*^:- '^»<^.," t\^ Digitized b)/ picfosbtt®^