Experimental Studies on the Preparation and Effects of Antitoxins for Tubercu/osis. BY E. L. TRUDEAU, M.D., AND E. R. BALDWIN, M.D. (Saranac Laboratory for the Study of Tuberculosis, Saranac Lake, JV. Y.) FROM THE AMERICAN JOURNAL OF THE MEDICAL SCIENCES, December, 1898, and January, 1899. Extracted from The American Journal of the Medical Sciences, De) intravenously with virulent tuberculous mate- rial, and recovered; (c) treated with tuberculin subcutaneously in in- creasing doses. VII. Ass injected : (a) Subcutaneously with dead cultures of non- virulent tubercle bacilli on thymus bouillon; (6) with precipitated tuberculin from cultures of the same non-virulent tubercle bacilli on proteid-free media ; (c) alkaline extracts of the bacilli with dead bacilli, subcutaneously ; (d) living non-virulent tubercle bacilli. VIII. Rabbits: (a) Inoculated intravenously with non-virulent tubercle bacilli and recovered; (6) inoculated intraperitoneally with virulent tubercle bacilli and recovered. With serums from the foregoing we tried to carry out the following plan of tests : 1. Effect of serum on healthy animals. 2. Treatment of tuberculous animals with serums to show influence on course of disease and temperature. 3. Test of germicidal influence in vivo and in vitro. 4. Test of power to neutralize effect of tuberculin in small and fatal doses. * The tuberculin used in this work was made from full-grown bouillon cultures of non- virulent tubercle bacilli from human source, evaporated over a water-bath to one-tenth volume and filtered through clay. 0.100 c.c. usually sufficed to kill guinea-pigs six weeks tuberculous. t The cultures denoted “ non-virulent ” were from tubercle bacilli of human origin, grown four years in the incubator, and which only occasionally killed guinea-pigs in six months to one year. TRUDEAU: ANTITOXINS FOR TUBERCULOSIS. 5 5. Test of effect on local reaction and temperature produced by tuberculin in tuberculous animals. We must acknowledge at the outset that, for various reasons, we were unable to carry out all of these tests with all of our serums, and the work is incomplete to that degree. Methods used in some tests were changed in others because they were thought to be fallacious, particularly in testing for anti-tuberculin. Consequently there is no true basis for comparison of all the serums tried. In addition to the serums prepared b}r ourselves, we have tried five or six from other sources. In the present uncertain state of serum-therapy for tuber- culosis it is undesirable to mention the names of their originators. Part I. I. Sheep injected intravenously with filtrate of cultures of tubercle bacilli on thymus bouillon. June 26, 1894. No. 1, treated, weight 64 lbs.; No. 2, control, weight 64 lbs. Both animals are “ wethers” and in good condition. Began injections of filtrate of cultures of (non-virulent) tubercle bacilli on calf thymus bouillon into sheep No. 1. Four doses: June 26th, 10 c.c.; June 30th, 20 c.c.; July 3d, 25 c.c. ; July 11th, 25 c.c. Tem- perature three hours after first dose 103.8.° July 15th. Weights: No. 1, 52 lbs. (loss, 12 lbs.; no other effect noted); No. 2, 64 lbs. August 4th. Weights: No. 1, 57 lbs.; No. 2, 67 lbs. 10th. Weights: No. 1, 561 lbs.; No. 2, 67 lbs. No. 1 did not re- cover weight, and on this account bleeding was postponed. October 21th. No further improvement or change noted. No. 1 bled 500 c.c. by canula. 175 c.c. serum collected. This animal was so much weakened by bleeding that it was killed ten days later. No lesions found at autopsy. Effect of No. 1 Sheep-serum on the Course of the Disease and on the Temperature.* October 21th. Took 6 guinea-pigs, average weight 523 grammes ; (a) treated with serum, 4 guinea-pigs, average weight of 504 grammes; (6) controls. Before inoculation (a) received five doses of 4-5 c.c. of the serum subcutaneously; total, 24 c.c. each. No effect to be noted on temperature. November 5th. Weights: (a) Treated, average 517; (6) controls, average, 511. All were inoculated in groin with virulent material from the lung of a rabbit in which tubercle bacilli were scanty. Each of (a) was injected with 4 c.c. of serum. 12th, 18th and 23d. (a) Injected with 5, 5, and 2 c.c.; total, 12 c.c. Serum pigs lost more weight than the controls. They showed no local irritation. * Weight and temperature tables are omitted to economize space. 6 TRUDEAU: ANTITOXINS FOR TUBERCULOSIS. Result. Average time of death : (a) 44 days ; (b) 62 days. Au- topsies revealed generalized tuberculosis of all organs. The serum apparently hastened death and caused loss of weight in healthy animals before inoculation. A preliminary test of the antitoxic power of this serum on pigs ninety- three days tuberculous showed that it did not prevent a temperature reaction after the use of tuberculin, and the outlook was so unpromising that further tests were abandoned. The bleeding of the sheep being postponed so long, the method may, be said to have had (hardly? a fair trial. II. Fowls inoculated intraperitoneally with mammalian tubercle bacilli of increasing virulence. September 10, 1894. We took 12 fowls (chickens and cocks), two- thirds grown, having an average weight of 1026 gm. The first inocu- lation was of 1.5 c.c. of an emulsion of non-virulent tubercle bacilli from bouillon culture which had been grown four years in an incubator oven. October 27th. Average weight 1671. Second inoculation: 0.7 c.c. of an emulsion of non-virulent tubercle bacilli from a bouillon culture which had been cultivated two years. December 20th. Average weight 1717. Third inoculation : 1 c.c. of an emulsion of a virulent culture (cultivated four months on serum). January 21, 1895. Average weight (10 fowls) 1686. Fourth inocu- lation : 2 c.c. of an emulsion of a virulent culture (growing five months on serum). February Vbth. Average weight (10 fowls) 2006. Fifth inoculation : 1.5 c.c. of the juice of crushed lymph nodes from a rabbit; third passage of virulent tuberculosis. April 27th. Average weight (10 fowls) 1894. Sixth inoculation: 1.5 c.c. of the juice of the lungs and omentum; sixth passage of vir- ulent tuberculosis. May 22d. (Not weighed). Seventh inoculation: 1.5 c.c. of the juice of the spleen and omentum ; eighth passage of virulent tuberculosis. July 19th. (Not weighed.) Eighth inoculation: 1.5 c.c. of the juice of the omentum; seventh passage of virulent tuberculosis. August 10th. Average weight 1684. Ninth inoculation: 1.5 c.c. of the juice of the omentum; sixth passage of virulent tuberculosis. September 4th. Tenth inoculation: 1.5 c.c. lung; eighth passage of virulent tuberculosis. During the course of these inoculations only two of the fowls died, and those from injuries received in fighting. The rest were bled to death at various times after the seventh inoculation. Most of them were bled twelve and twenty-six days following the last inoculation. Only small amounts of serum were obtained, and part was preserved by trikresol. In no case was there found the slightest evidence of a past or present tuberculosis on post-mortem examination. June 15, 1895. We took an emulsion in water of the first culture of tubercle bacilli on serum (virulent), (a) Mixed one-half with 6 c.c. serum ; no antiseptic added. (6) Mixed one-half with 6 c.c. of 0.6 per cent. NaCI solution. Both were allowed to stand six hours at room Test of the Germicidal Power of Fowl-serum. TRUDEAU: ANTITOXINS FOR TUBERCULOSIS. 7 temperature in a dark closet. We then inoculated 3 guinea-pigs, average weight 358 gm., with (a) 2 c.c. each ; and 3 guinea-pigs, average weight 345 gin., with (6) 2 c.c. each. All were inoculated sub- cutaneously in the right groin. Result: All became tuberculous in the usual way : (a) lived 77 days ; (b) lived 80 days. There was no germicidal influence in vitro on the tubercle bacilli. The effect of this serum on the temperature of seven tuberculous guinea- pigs and rabbits was tried, with the result that doses of 1.5 to 2 c.c. seemed to cause some elevation in six hours. The amount of serum was insufficient for more of such experiments. Influence of Fowl-serum on the Course of the Disease in Guinea-pigs. September 20, 1895. We took 5 pigs, average weight 514.2 gm., for treatment ; anrl 5 pigs, average weight 487.6 gm., for controls. All inoculated with virulent tuberculous material from lung of guinea- pig; each receiving 0.25 c.c. in the right groin. We began treatment the same day, using subcutaneous doses of from 1 to 2.5 c.c. every three to ten days, and giving a total quantity of 10 c.c. each. Ab- scesses formed in a few instances, probably from skin infection. Result: The treated animals lived 57 days; the control animals 581 days. Tests of the temperature of these animals showed no perceptible influ- ence of the serum. The limited quantity of serum available precluded tests with tuberculin. The results accord with those published recently by Auclair,3 and were especially disappointing because fowls have such high natural immunity to mammalian tuberculosis and are able to dis- pose of large quantities of human tuberculous material. III. Sheep injected subcutaneously with increasing amounts of tuber- culin. IV. Sheep inoculated intravenously with living non-virulent tubercle bacilli cultures and injected with tuberculin in increasing doses. April 22, 1895. (III.) Sheep (wether), weight 70 lbs. (IV.) Sheep (wether), weight 65 lbs. Sheep (IV.) received 10 c.c. tubercle bacilli from a non-virulent bouillon culture in the saphenous vein. No harm- ful effects followed. May 16th, 10 c.c. same. Weights: (III.) 105 lbs.; (IV.) 80 lbs. After this dose (IV.) lost weight and strength steadily during the summer, so that nothing more was attempted until December of the same year. December 15, 1895. Weights: (III.) 105 lbs.; (IV.) 80 lbs. Tuber- culin Test. Injected both with 0.200 tuberculin. The temperature of (III.) rose from 102° to 103.5°; (IV.) reacted from 102° to 105°. Tuberculin Injections. During the next three months both sheep were injected with gradually increasing doses of tuberculin; at first every second day, then less frequently, according to the loss of weight. (HI.) received 19 doses; the largest, 50 c.c.; the total quantity, 184 c.c. (IV.) received 23 doses ; the largest, 20 c.c.; the total quantity, 64.5 c.c. Both animals retained their weight until the maximum dose was reached, which for sheep (IV.) was evidently overwhelming, as it lost 15 lbs. during the following three weeks, finally becoming so weak and 8 TRUDEAU: ANTITOXINS FOR TUBERCULOSIS. cachectic that it was killed. No tuberculous lesions were revealed by autopsy, but the liver and spleen were found much atrophied; there was a calculus in the pelvis of one kidney. The absence of tuberculous lesions and the apparent good health of this sheep up to the time of the tuberculin injections remind one of the effects mentioned later by Maf- fucci and Vestea57 as the result of intravenous inoculations of living tubercle bacilli in sheep. The cachexia in one sheep was probably in- duced by the tuberculin injections. No signs of hsemoglobinuria were noted in these animals; this is mentioned by Niemann66 as occurring after massive doses of tuberculin in goats, owing to the large amount of glycerin contained in it. Sheep (III.) withstood the dose of 50 c.c. of tuberculin fairly well, and was bled March 21,1896, five days later. 1000 c.c. were taken by canula from the external jugular vein. The weight decreased 20 lbs. during the following three weeks. Part of the serum mis kept aseptic and the rest was preserved with camphor. _ Test of the Germicidal Power of the Serum of Sheep (JZT.) Tuherculinized. March 23, 1896. We took mixtures of: (a) 6 c.c. of emulsion in 0.6 per cent. NaCl sol. of a washed sputum rich in tubercle bacilli, adding 12 c.c. of a serum without antiseptic. (6) 6 c c. of emulsion of sputum as above; 12 c.c. of 0.6 per cent. NaCl solution. A stained drop showed one or two tubercle bacilli in a field. The mixture was placed in a cool, dark closet for six hours. We then took 3 guinea-pigs, average weight 692, each of which received 6 c.c. of (a) subcutaneously, and 3 guinea-pigs, average weight 702, each of which received 6 c.c. of (b) subcutaneously. Result: All became infected in the usual time, no difference&in the course of the disease being observed. (They were therefore used for testing tuberculin later.) Autopsies revealed rather chronic tubercu- losis in all. No germicidal influence was manifested by the serum. Effect of the Serum of (JIT.) on the Course of the Disease in Guinea-pigs. March 23, 1896. We took 6 pigs, average weight 721 gm., and treated them with serum; 4 pigs, average weight 766, were taken as controls. All were inoculated in the right groin with one oese of washed sputum emulsion (same as in above experiment).* Injections of serum preserved with camphor were begun on the following day. It was warmed and administered intraperitoneally to facilitate absorption. Doses of 2 to 5 c.c. were continued every second to fourth day for fifty days. The treated animals received total amounts varying from 50 to 65 c.c. Result: By the twentieth day all the treated pigs were much more emaciated than the controls. One died from peritonitis from puncture of the stomach in injecting of serum. Three more treated pigs died in fifty days from tuberculosis, while the controls were still vigorous. The controls were therefore killed on the fifty-first day for comparison, and the lesions were found practically the same as in the treated. The * Sputum was used for these inoculations, thinking more nearly to approach the infective power of tubercle bacilli for human beings. TRUDEAU: ANTITOXINS FOR TUBERCULOSIS. 9 serum appeared to act harmfully; at least when given in the perito- neum, though the two remaining treated animals survived 110 to 132 days, showing lesions which were chronic, but notsunusual in character. The effect on temperature was not noted in above experiment. Test with Tuberculin of Serum (III.') Antitoxic Power. Eight pigs inoculated with sputum were tested on the nineteenth, thirtieth, and thirty-third days of disease with serum and tuberculin mixed, given subcutaneously. The conditions of the experiments were so unsatisfactory that they deserve only brief mention. So far as could be judged, no favorable influence was observable. In some pigs the serum seemed to cause fever, and since the bleeding of the sheep was undertaken only five days after a large dose, it is conceivable that some of the tuberculin may have still been contained in the serum. V. Ass J.; inoculated intravenously with living, non-virulent tubercle bacilli. December 13, 1894. Male ass(J ); weight, 445 lbs. Appeared old, but sound. We injected 7 c.c. of an emulsion of tubercle bacilli in 0.6 per cent, sterile NaCl solution into an ear vein. (About one-third went into the subcutaneous tissue, producing induration and a cold abscess.) No effect could be noted on the health of the ass. January 15, 1895. Weight 460 lbs. We gave successfully 15 c.c. of an emulsion of tubercle bacilli in 0.6 per cent. NaCl solution ; the tubercle bacilli were taken from three bouillon tubes. No injurious effect was seen. March 1th. Weight 480 lbs. We injected 15 c.c. of a similar emul- sion of tubercle bacilli taken from 6 tubes, with partial success. April 22d. Weight, 490 lbs. We attempted to inject 10 c.c. of a strong emulsion of tubercle bacilli in 0.6 per cent, sterile NaCl solu- tion into the external jugular; most of it went into the subcutaneous tissues. May 16th. Weight, 475 lbs. We gave 12 c.c. of a thick emulsion of tubercle bacilli in 0.6 per cent, sterile NaCl solution into an ear vein. Result: Died in twelve hours from an embolus in the pulmonary artery. The emulsion was probably too thick and produced a clot. No evidence of tuberculosis of the lungs, liver, or spleen could be found at autopsy, nor on microscopical examination of sections. Enough serum was saved from the heart cavities to test its bactericidal influence. Test of the Bactericidal Effect of Serum. Ass J. ( V.). May 18, 1895. (a) 4 c.c. of the above serum, unfiltered, was mixed with 1 c.c. of the crushed and strained juice from the spleen of a tuber- culous guinea-pig that had died in five weeks; (b) 2 c.c. 0.6 per cent. NaCl solution was mixed with 0.50 c.c. of the same juice. Both stood five and a half hours in a dark closet. 4 pigs, average weight 315 gm., received 1.25 c.c. each of (a). 2 pigs, average weight 293 gm., received 1.25 c.c. each of (b) in the right thigh. Result: All became uniformly tuberculous as usual, indicating that the serum had no effect on the bacillus, though the experiment is obvi- ously inconclusive. 10 TRUDEAU: ANTITOXINS FOR TUBERCULOSIS. VI. An ass was inoculated subcutaneously and intravenously with virulent cultures of tubercle bacilli and virulent tuberculous material from animals; it was then injected with tuberculin in increasing doses. February 25, 1895. Subcutaneous inoculations. Female ass (R.); weight, 450 lbs. ; full-grown, but young ; in fine condition. The normal rectal temperature varies from 98.4° to 99.5°. The first inoculation was of 12 c.c. of an emulsion of the liver and omentum of a rabbit that had died in fifty-one days of acute tuberculosis (fourth passage through rabbits). There were few tubercle bacilli in the emulsion; a cheesy abscess resulted, which was opened and healed. April 6th. Weight, 455 lbs. Second inoculation: 12 c.c. of an emulsion of the liver and omentum of a rabbit that had died forty days (fifth passage); tubercle bacilli were fairly numerous. An abscess followed. May 22d. Weight 430 lbs. Third inoculation : 4 c.c. of an emulsion of the liver and omentum of a rabbit that had died in twenty-two days (seventh passage); tubercle bacilli were few. Injections were made in three or four places; no abscesses followed, but there was permanent induration. June 9th. Fourth inoculation : 4 c.c. of an emulsion of the liver and omentum of a rabbit that had died in forty-nine days (seventh passage) ; tubercle bacilli fairly numerous. An abscess resulted; otherwise its health remained good. It was turned out to pasture for the summer. August 9th. Intravenous inoculations. First inoculation : 5 c.c. of an emulsion of the lymph nodes and omentum of a rabbit three months tuberculous (seventh passage) ; it was injected into an ear vein ; tubercle bacilli were scanty in the emulsion. No disturbance of health that was perceptible followed. 2Uh. Second inoculation: 2 c.c. of an emulsion from the lymph nodes and omentum of a rabbit three months tuberculous (eighth pas- sage) : tubercle bacilli were numerous. Following this the ass soon became daily thinner and weaker, and breathed somewhat quickly. September 1st. Losing weight. Evening temperature 100.2°. 13th. Injected with 0.030 tuberculin; reaction to 104°. November 19th. Gaining weight; evening temperature 100°. 20th. Tuberculin injections. Begun with 0.030, and gradually in- creased every two to ten days for five months. Thirty doses were given, aggregating 173.5 c.c. The last dose was 33 c.c. Her weight gradually increased from 350 to 420 lbs., and the reactions became less. The last dose caused a temperature of 101°, but no depression. Little disturbance was noted from the tuberculin after the first four or five doses. May 4, 1896. First bleeding : Ten days after the largest dose, 1500 c.c. of blood was taken from the jugular vein by canula. Some of the serum obtained was preserved aseptically in tubes, while the rest was kept with lumps of camphor added. No depression followed the bleed- ing, so the animal was turned out to pasture for the summer. During the summer of 1896 ass R. was quite thin, weighing only 345 lbs., but she had regained her weight in the autumn—to 400 lbs. on October 15 th. November 2d. Tuberculin injections were resumed, beginning with a dose of 0.500 c.c., which was followed by a rise of temperature to 102°. TRUDEAU: ANTITOXINS FOR TUBERCULOSIS. 11 The doses were rapidly increased, causing about the same reaction, until December 29th, nearly two months. The last dose was 10 c.c. ; the aggregate 46.5 c.c. An interval of thirty days was then given before bleeding, during which time the ass gained 20 lbs. in weight. January 27, 1897. Second bleeding: 1800 c.c. was taken from the left jugular. Test of the Germicidal Power of the Serum of Ass Ii. (FT). First serum: The same method was used as in the fowl and sheep serums except that virulent cultures were employed for infective material. May 6, 1897. (a) 12 c.c. of serum (without antiseptic) was mixed with 6 c.c. of an emulsion of tubercle bacilli from a first culture on serum. (6) 12 c.c. of a 0.6 per cent, solution of NaCl was mixed with 6 c.c. of tubercle bacilli emulsion as above. Both were allowed to stand twelve and one-half hours in a dark closet. We took 3 pigs, average weight 517 gm., and inoculated them with 3 c.c. each of (a); and 3 pigs, average weight 512 gm., which we in- oculated with 3 c.c. each of (6). Result: After twenty-six days the evidence of uniform disease was so palpable in all that they were killed by tuberculin. All were found to have lesions of generalized tuberculosis. Second serum: This serum was not tried on animals, but was added to bouillon cultures without heat to test its inhibitive power on the growth of tubercle bacilli in vitro. January 31, 1897. We took 4 flasks containing 50 c.c. each, of bouil- lon -f 10 c.c. or 5 c.c. of serum (R. second), 5 flasks containing 25 c.c. each of bouillon + 2.5, 1.25, 0.63, 0.50, and 0.25 c.c. of (R. second) 4 flasks containing 25 c.c. each,of control bouillon -f- 10, 1.25, 0.50, and 0.25 c.c. of hydrocele fluid, and 1 flask containing 25 c.c. of con- trol bouillon only. All were planted with non-virulent tubercle bacilli, surface growth. Only one flask became contaminated. No retardation of growth was seen unless the serum was present in large proportion. In the latter case the alkalinity was so increased by the serum that from comparison with the hydrocele fluid cultures and from former observation we may attribute the lessened growth to the wide differences in this respect. Result: There was no reason to suspect any antibacillary effect of a specific nature from the above experiment. Effect of the First Seram of Ass JR. ( VI.) on the Course of the Disease in Guinea-pigs. May 4, 1896. 10 pigs were inoculated in the right groin with a viru- lent culture of tubercle bacilli, the first culture on serum. 6 pigs, aver- age weight 500 gm., were treated with serum. 4 pigs, average weight 520 gm., as controls (a). 2 pigs, average weight 515 gm., as controls (fb); which were not inoculated, but were treated. 1 pig, weight 541 gm., as control (c); which was neither inoculated nor treated, but was kept in the same cage for comparison. Injections of serum into the peritoneal cavity were begun on the fourth day, and were continued every second or third day for fourteen doses, which ranged from 1 c.c. to 6 c.c. The average total amount for each pig was 45.7 c.c. 12 TRUDEAU : ANTITOXINS FOR TUBERCULOSIS. Result: The treated died on an average of 45.5 days after inocula- tion. Controls (a) died on an average of 42.5 days after inoculation ; controls (6) died on an average 50 days after inoculation ; control (c) gained 33 grammes weight. No effect could be observed either on the appearance of the treated animals or on their temperature and lesions as compared with the con- trols. The controls (6), which were to show the effect of the serum on healthy animals, were inoculated by mistake May 20th. They serve, however, to show that the serum failed to prevent or modify the disease when given prior to inoculation. All died with generalized tuber- culosis. The prolongation of life was not sufficient to be significant. Possibly a more favorable result would have been obtained with this serum had it been used subcutaneously in smaller doses. The quantity at our disposal was too small to carry out a second test and to try its effect on tuberculous eyes of rabbits. The second serum from ass R. was thus tried, as will be seen hereafter. (The experiments on the antituberculin or antitoxic power of first serum, ass R., will be given in Part II.) Effect of the Second Serum of Ass R. ( FA) on the Course of the Disease. February 12, 1897. 15 pigs were inoculated in the right groin with a culture of tubercle bacilli of moderate virulence. 5 pigs, average weight 634 gm., were treated with the second serum. 5 pigs, average weight 602 gm., controls.* (a) 1 pig, weight 580 gm., control (6), was not inoculated, but was treated. 1 pig, weight 455 gm. control (c), neither inoculated nor treated. Injections of serum were begun next day under the skin of the ab- domen ; doses of from 0.050 to 1 c.c. at intervals of one to three days were given until May 4th—nearly three months. No marked indura- tion of skin nor abscesses resulted. The total for each pig was 17.4 c.c. Result: 4 treated pigs died after an average of 95.2 days; 1 treated pig survived four months, and was then killed. 2 controls (a) died after an average of 75.5 days; 3 controls (a) survived four months and were then killed 1 control (6) lost 10 grammes in four months. 1 con- trol (c) gained 15 grammes in four months. No influence was manifest on the animals, nor on their temperatures taken before and after doses of 1 c.c., and compared with controls. Nothing unusual was noted in the lesions. All had chronic generalized tuberculosis. This experiment was in direct contrast to that with the first serum in the less virulent inoculation and with smaller doses of serum given under the skin instead of into the peritoneal cavity. The serums seemed to have no favorable influence either way. VII. Ass injected subcutaneously : (1) with dead non-virulent cultures of tubercle bacilli on thymus bouillon; (2) with precipitated tuberculin made from cultures of non-virulent tubercle bacilli on proteid-free media; (3) with alkaline extract of tubercle bacilli mixed with dead tubercle bacilli; (4) with living non-virulent tubercle bacilli. January 4,1895. Female ass (H.), weight 475 lbs., full-grown, rather “phlegmatic;” in good condition and apparently sound; normal rectal temperature—97° to 98°. * The other five pigs were treated with Serum VII. See page 1$7 TRUDEAU: ANTITOXINS FOR TUBERCULOSIS. 13 1. Injections of non-virulent cultures of tubercle bacilli on thymus bouil- lon which had been hilled by trikresol. Began injections of emulsion of tubercle bacilli in thymus bouillon culture with 0.2 per cent, trikresol, on alternate sides of neck. January 4, 1895, 1 c.c.; 11th, 1.5 c.c.; February 8th, 5 c.c.; March 4th, 8 c.c.; April 8th, 10 c.c. No rise of temperature followed these injections, and only indurated spots remained; no abscesses resulted. The animal was not visibly affected, except that she lost 15 pounds, which was regained by April 17th, when she was turned out to pasture for the summer. Nothing further was done until December 11th, when a dose of 0.500 tuberculin was injected without an ensuing reaction. December 12th. 2. Injections of precipitated tuberculin. We began injections of toxin obtained from cultures of non-virulent tubercle bacilli on liquid media containing asparagin, mannit, or ammonium carbonate in mixtures with salts as in the formulas of de Schweinitz84 and Proskauer and Beck.70 The toxin was prepared as follows : To the clay-filtered culture fluid 2 per cent, acetic acid was added, followed by ammonium sulphate to saturation. Nearly all the proteid matter arising from the fully grown cultures was thus precipitated. It was then collected, redissolved in water, reprecipitated by alcohol, filtered, washed with alcohol and ether, dried, and weighed. Solutions of the solid substance, 0.005 in 1 c.c., were made in weak sodium carbonate. 0.5 c.c. of this prepara- tion was not fatal to tuberculous guinea-pigs, though producing high temperature and local irritation at the site of injection. This pure tuberculin or toxin was injected every second day in in- creasing doses up to 7 c.c. until January 30, 1896. The total quantity was 26.85 c.c. Abscesses then formed at site of injection, and occa- sionally thereafter from the following treatment: 3. Injection of alkaline extract and dead tubercle bacilli. The toxin was next prepared by adding 1 per cent. NaOH to the cultures before filtration; warming to 40° C.; filtering through cotton; precipitating by 2 per cent, acetic acid, which gave an abundant flocculent sub- stance ; filtering again ; washing out the acid with water by decantation ; re-solution in weak alkali. Naturally this fluid contained many dead and disintegrated tubercle bacilli, as the filtration was carried through cotton and paper. We found it impracticable to filter through clay, since most of the dissolved poison was left on the filter. From February until June, 1896, fifteen injections were given sub- cutaneously, in increasing doses, of these solutions well diluted and freshly made. The largest amount represented 3500 c.c. of cultures. In the aggregate, 17,500 c.c. of cultures grown in 100 to 150 c.c. flasks were used. Considerable induration persisted after these injections, and on three occasions they were followed by septic abscesses which necessitated an interruption of the treatment. The solutions not being sterilized, and the dissolved poison and dead bacilli both being such strong irritants, we despaired of further increasing the doses. Through- out the time of treatment the ass had but little disturbance of health except at the time of the abscesses, when she did not eat well. The abscesses were opened as soon as formed, and healed promptly. By June 1st she had lost fifteen pounds, but regained it by the 15th inst. June 24, 1896. First bleeding : Nine days after the last dose of toxin and tubercle bacilli, 1000 c.c. of blood was drawn from the right 14 TRUDEAU: ANTITOXINS FOR TUBERCULOSIS. external jugular vein ; 400 c.c. of serum was obtained aseptically. All the abscesses had healed at this time except two or three small pustules. The animal was then put into the pasture. Although this had been an unpromising experiment, owing to the abscesses, we decided to try the serum, omitting the germicidal test on animals. Effect of First Serum, Ass H. ( VII.), on Course of Disease in Guinea-pigs. June 26, 1896. 10 pigs inoculated in right groin with tuberculous material from a guinea-pig ; virulent infection. 6 pigs, average weight 710.8, treated with serum; 4 pigs, average weight 713.7, (a) controls; 2 pigs, average weight 780, (6) controls, not inoculated, but treated with serum. The serum (with camphor added to insure preservation) injections into peritoneum were begun on next day and given every second day for a month with doses of 1 to 7 c.c.; total quantity, 41.5 c.c. each. Result: Average time of death : treated 47 days; (a) controls 49.7 days. Controls (b) lost weight and were inoculated later, but developed the disease in usual way. The temperatures of all were taken on the twenty-second day of disease and after the dose of 7 c.c. serum, with- out showing any influence from serum. The antituberculin tests will be found with the other serums in Part II. 4. Ass inoculated subcutaneously with non-virulent living cultures. November 7, 1896. Took ass H. (same as was used in VII.) after being in pasture all summer. Weight 490 lbs. Indurations from former treatment remained, but were smaller. The first inoculation was of 25 c.c. of culture fluid with tubercle bacilli rubbed up in mortar, weighing in moist condition 0.0221—from an actively-growing culture of non-virulent tubercle bacilli on acid bouillon. The injection was into the right shoulder. The temperature six hours later was 99°. November 20th. Second inoculation : 25 c.c. of a culture containing 0.1353 tubercle bacilli into left shoulder. Several small aseptic ab- scesses from the first inoculation were to be seen. December 7th. Thix-d inoculation : 25 c.c. (the tubercle bacilli were not weighed, but more wTere used); the injection was made into the right groin; there was no abscess from last dose. 19th. Fourth inoculation: 40 c.c. into the left groin; no abscess re- sulted. January 6, 1897. Fifth inoculation : 40 c.c. into the right shoulder; temperature 98.5 ; no abscess followed. Much induration of the skin remained after all the inoculations, but little disturbance of health was noticeable. Weight 500 lbs. February 6, 1897. Second bleeding: Thirty days after the last dose, weight 510 lbs.; 2000 c.c. blood was taken from the right jugular vein ; 1000 c.c. serum was collected aseptically. February 13, 1897. Second serum: We each, containing 50 c.c. bouillon -f- 10 c.c., 5 c.c., or 2.5 c.c. serum;; flasks each, con- taining 50 c.c. bouillon + 2 or 1.25 c.c. Taf normal serum, and; control flasks each, containing 50 c.c. bouillon -j- 2.5 c.c. or 1 c.c. of NaOH solution. Germicidal Power of the Serum, Ass H., in Culture. TRUDEAU: ANTITOXINS FOR TUBERCULOSIS. 15 The serum was added aseptically, no heat nor antiseptic being used. The NaOH was added to the controls because the serum increased the alkalinity of the bouillon. All were planted with non-virulent tubercle bacilli from a bouillon-serum culture. Result: March 20th, thirty-five days had grown luxuriantly* there were no contaminations; no inhibitive influence of the serum was to be seen. Effect of the Second Serum of Ass H. on the Course of the Disease in Guinea-pigs. February 12, 1897. 15 pigs were inoculated in the right groin with the culture of tubercle bacilli of moderate virulence. 5* pigs, average weight 620 gm., were treated with serum subcutaneously. 5f pigs, average weight 602 gm., (a) controls, were not treated. 1 pig, weight 575 gm., (6) control, was not inoculated, but was treated. 1 pig, weight 455 gm., (e) control, was neither inoculated nor treated. The doses and intervals were the same as in the experiment with the serum of ass R., the two serums being compared in this way. (See page Result: 4 treated pigs lived on an average 106.7 days; 1 survived 4 months; 2 controls lived on an average 75.5 days; 3 survived 4 months. No influence could be ascribed to the serum so far as the lesions or the course of the disease were concerned. Effect of the Second Serum of Asses R. and H. on the Course of Eye- tuberculosis in Rabbits. January 30,1897. 12 rabbits were inoculated in the anterior chamber of the left eye with one drop of weak emulsion of a pure culture of tuber- cle bacilli of moderate virulence. (R.) 4 rabbits, average weight 2028 gm., were treated with serum from Ass R. subcutaneously. (H.) 4 rabbits, average weight 1741 gm., were treated with serum from Ass H. subcutaneously. (C.) 4 rabbits, average weight 1705 gm., were used as controls. February 12th. Twelfth day. We began treatment with doses of from 2 to 6.5 c.c. of the serums every second or third day, under the skin of the abdomen; treated until March 23d—forty days; the total amount given each was 66 c.c. Result: The serum produced induration, but no abscesses nor dis- turbance of health. There was no change in the appearance of the eyes in the treated which was not seen in the control rabbits; no differ- ence in temperature was noted when taken six hours after the first dose of serum; the disease in the eyes ran the usual course. VIII. Rabbits: 1. Inoculated intravenously with non-virulent cultures of tubercle bacilli; recovered. 2. Inoculated in peritoneal cavity with virulent tuberculous material; disease arrested or recovered. March 31, 1896. We took 3 rabbits, weights 2210, 2210, 2085 gm. Each received 1 c.c. of an emulsion of a pure culture of non-virulent tubercle bacilli in water into the aural vein. After temporarily losing weight they appeared completely well six months later. * Same lot as “ second serum, ass R.page 1JL t Same controls as in “ second serum, ass R.page 1JL 16 TRUDEAU: ANTITOXINS FOR TUBERCULOSIS. October 3d. All were inoculated intraperitoneally with an emulsion from a caseous lung of a monkey. The three control rabbits all died in two months. All the above animals survived and were apparently recovered by January 2, 1897, when they vTere bled to obtain serum- Some chronic lesions were found in one animal, but were not pro. gressing. Effect of the Serum of Rabbit (VIII.) onthe Course of the Disease. January 5,1897. We took 5 pigs, average weight 555.2 gm., for treat- ment ; and 5 pigs, average weight 553.4 gm., for controls. All were inoculated in the right groin with a pure culture of tubercle bacilli of weak virulence. 7th. We gave subcutaneous doses of 0.050 to 0.850 c.c. of the rabbit serum every second to third day, until February 21st—forty-five days. Total quantity, 9.65 c.c. each. Induration of skin was produced, but there were no abscesses; no other effect was perceived. Result: 4 pigs treated died after an average of 92.5 days; 1 survived 4 months and was then killed. 3 control pigs died after an average of 77.3 days; 2 survived 4 months and were then killed. No difference in the lesions was to be seen on post-mortem examination, but the serum pigs that died outlived the controls that died. In a single tuber- culous pig no effect on the temperature was produced by a dose of 1.70 c.c. The limited quantity of this serum precluded further investigation. In addition to the serums prepared by ourselves, we made a prelimi- nary test of some serum said to be from a horse treated with toxins obtained from virulent bacilli. We injected three tuberculous pigs every other day for two months without apparent effect in prolonging life, and hence were unable to confirm the author’s statements regard- ing the serum. We must state with reference to all the foregoing experiments in treating pigs with serums that they were only preliminary, and the results with so few animals are not viewed by us as conclusive. Part II. Tests of the Antitoxic Power of Serums in Tuberculin Poisoning. In these experiments we have considered : 1. Fatal doses of tuberculin* in sound animals. 2. Fatal doses in tuberculous animals. 3. Small doses in tuberculous animals to show the effect: (a) on the temperature; (6) on the local reaction. Many difficulties present themselves in any attempt to demonstrate a specific antitoxin in tuberculosis by methods analogous to those used so satisfactorily in diphtheria. The principal of these relate to the * Under the term “ tuberculin ” we here include the various extracts of tubercle bacilli, but usually mean'tbe original Koch fluid. TRUDEAU: ANTITOXINS FOR TUBERCULOSIS. 17 difficulty of obtaining a highly poisonous product in small bulk unless tuberculous instead of sound animals are employed for the tests. Objections can easily be made to the use of tuberculin and diseased animals for testing serums. In the first place, tuberculin when pre- pared by boiling heat may not represent the unaltered toxins of tubercu- losis, though containing more toxic extractive substances from the bacilli than when made by evaporation of cultures at low temperatures Yet we cannot safely say that tuberculin is a wholly altered product until we have further knowledge of its chemical nature. We know, indeed, that boiling heat not too prolonged does not destroy its most prominent physiologic action on tuberculous animals. The propriety of using tuberculin in serum tests principally depends upon the view one may take of the character of the reaction. That question may need further elucidation, but it seems to us that the best explanation we have from observations up to date supposes tuber- culin to be a partly specific irritant, both to tuberculous foci and to the sus- ceptible organism in general. The local and general reaction is caused partly by the poison contained in the tuberculin, which irritates the sensi- tized cells composing the tubercle, and partly by those toxins set free by the hypersemia or the enzyme (?) action directly or indirectly produced by the dose of tuberculin (see Babes and Proca7). Whether the toxins liberated are the same, chemically, as those obtained from artificial cultures of tubercle bacilli is not easy to determine ; but that some are set free seems obvious from the profound general disturbance brought about by a minute dose of tuberculin. According to this theory, the poisons are stored up in the tubercles, and in part at least derived from the dead or weakened bacilli, as shown by Babes and Proca7 in ex- periments with dead tubercle bacilli. Crookshank and Herroun23 and Matthes63 61 have extracted albuminous poisons out of caseous material from tuberculous lesions which had the properties of tuberculin. Kahler43 and Lenoir51 found albumoses in the urine of patients being treated with tuberculin, which produced tuberculin reactions and were present in the urine in excess of the amount injected as tuberculin. Thus we have some evidence of the discharge of accumulated pyrogenic substances ; and that the tuberculin injected, in the minute dose necessary to cause the reaction, may not itself play the prominent part as a poison, but rather as the excitant to the discharge of other specific poisons. Therefore, but for the uncontrollable amounts of hoarded toxin ex- ploded, as it were, coupled with the local congestions, there would be no objection to proving the activity of serums in this way. Given the serum of a tuberculous animal which had been treated with tuberculin and had tolerated the reactions well, we might hope, if an anti-body had been in the serum, that it would be effective against the specific poisons developed in the animal body by tuberculosis, though not necessarily 18 TKUDEAU : ANTITOXINS FOR TUBERCULOSIS. against tuberculin as obtained from cultures. By analogy, such an antitoxin ought also to manifest its activity on other tuberculous ani- mals injected with a dose of tuberculin just large enough barely to pro- duce a reaction (and at the same time with a dose of such serum suffi- cient surely to neutralize it). Should it be weak in activity, it would still seem possible to obtain some effect in that way. Furthermore, it might reasonably be expected that the fatal effects of doses of tuberculin just sufficient to cause death in tuberculous animals with uniform lesions might be inhibited by the administration of an efficient serum ; or, at least, that the lives of the animals might be appreciably pro- longed. Such tests ought to give qualitative results, though, from the unknown quantities of toxin to be dealt with, they manifestly would not serve as quantitative standards for serums. On the other hand, healthy animals cannot be suitable for tests until we can obtain sufficiently toxic unaltered products of tubercle bacilli; while if the serums proved to be wholly or chiefly bactericidal, and not antitoxic, they would not reveal that fact in non-tuberculous animals. As is well known from Koch’s original experiments, healthy guinea- pigs tolerate very large doses of tuberculin. This fact renders it nec- essary to employ such a quantity of ordinary tuberculin to produce death that the effects of the unaltered ingredients of the culture fluid, such as glycerin, albumoses, or peptone, should be remembered, as well as the shock produced by the traumatism of large injections. We occupied ourselves for some time in attempts to produce a con- centrated tuberculin or toxin from cultures; but while it is compara- tively easy to produce tuberculin, even from non-virulent cultures, which will usually be fatal in doses of 1 c.c. to 100 grammes of pig (the strength of toxin mentioned by Maragliano as used to test his serum), yet this ratio did not suffice in our experiments to kill pigs weighing above 250 grammes. In addition to the objections mentioned, we found its absorption from under the skin very uncertain. We then sought by precipitation from cultures on proteid-free media to obtain a pure toxin which would be fatal in small doses. We also attempted to procure a concentrated poison from the bacilli by using weak alkaline extracts, but none of our preparations were quickly fatal to healthy animals in small doses, though producing cachexia and death after some time. After testing various preparations subcutaneously and intraperitoneally, without encouragement, we abandoned the idea of utilizing healthy animals for antitoxin or antituberculin tests. In the course of this work we used fifty guinea-pigs, weighing from 150 to 450 grammes. More recently, fresh extracts from crushed, living tubercle bacilli gave no more uniform fatalities. Clinical experience with human tuberculosis confirms the fact that TRUDEAU: ANTITOXINS FOR TUBERCULOSIS. 19 the poisons are of relatively low toxicity—where their immediate effects are considered—as compared with those of diphtheria, tetanus, etc. The same fact suggests itself as a reason for the indifferent results of experiments in sero-therapy with tuberculosis. The effect of tuberculin administered to an animal after the disease becomes established is, nevertheless, so powerful, and the dose so small, that the objections mentioned do not enter. A surely fatal dose of tuberculin for a tuberculous guinea-pig, though not uniform, is small enough for testing serums; yet there are other factors which are not to be disregarded in such experiments, and which are very difficult to control. The local reaction varies considerably in intensity in the same series of animals. Differences in virulence of the inoculation material used, variations in lesions and in the extent of caseation, the age, and the frequent pregnant state of the animals, suffice to require large numbers of animals in any experimental work on tuberculosis. All these factors we found especially important in testing for anti-tuberculin. For the reasons mentioned, the fatal dose of tuberculin necessarily varies. Differences in the stage of the disease, the injection of serums mixed with tuberculin, or given separately may produce diverse results owing to different rate of absorption* or other reasons. We also found that non-tuberculous serums, and even physiological salt solution when used under like conditions, gave results that dampen over-enthusiasm in making deductions. Yet in the experiments that follow, in the course of which two hundred and fifty pigs were used, there can be seen indi- cations of a favorable influence or possibly an antitoxic effect from some of the serums on tuberculous guinea-pigs tested with fatal doses of tuber- culin. Details of the Methods Used to Test the Antitoxic Power of Serums. Guinea-pigs were inoculated in lots of ten to twenty, at different times, with cultures or other tuberculous material of varying virulence. The point of inoculation was usually below the right groin in the sub- cutaneous tissue. The amount of tuberculin used to kill them varied from 0.100 to 0.500 grammes, according to the stage and type of disease in the animals; it was made from non-virulent bouillon cultures. The same bottle was used in nearly all the tests. To test the effect on temperatures doses of 0.001 to 0.002 were employed. The doses of serum ranged from 1 to 12 c.c. Some lots were tested with doses of serum and tuberculin proportioned to the weight of the pigs. * The slow absorption of the infiltration produced by injections under the skin of guinea- pigs of sheep, cow, ass, and rabbit serums was a point specially noted by us in all this work, and has been of late particularly contrasted with the better absorption of horse serum by Uhlenhuth.90 20 TRUDEAU: ANTITOXINS FOR TUBERCULOSIS. Loss of weight and strength were taken as the most practical criteria of the condition of the diseased animals at the time of testing. The tests were made from twenty-one to fifty-six days after infection. To avoid false deductions from lack of absorption of tuberculin when mixed with serum, it was given separately in most of the experiments. Serums were injected at the same time or before the tuberculin, sub- cutaneously, or, in some instances, intraperitoneally; the tuberculin always under the skin. Temperatures were taken in the rectum at the time of injection and six hours later. Some of the surviving animals were usually killed for comparison after twenty-four hours. Normal horse serum, diphtheria antitoxin, normal sheep serum, 0.6 per cent. NaCl solution, and hydrocele fluid were used to control anti- tuberculosis serums in comparative tests. Serum of Ass R, First Bleeding. March 23, 1896. 16 guinea-pigs of an average weight of 692 grammes were inoculated in the groin with a non-virulent sputum, which kills in from 60 to 100 days. May 7th. We tested 12 pigs on the forty-fifth day. Each received serum subcutaneously in doses of from 1 to 8 c.c. mixed in 7 instances with 0.300 c.c. of tuberculin. (I.) Of the 7 pigs that received serum and tuberculin 5 survived, while 2 died in from 13 to 15 hours. (II.) The 2 pigs that were injected with serum only survived and showed no effect. (III.) 3 pigs, controls, received tuberculin only and died in from 9 to 10 hours. Three days later (II.) received 0.300 c.c. tuberculin and died in from 10 to 12 hours. Result. The infiltrate was not well absorbed. The lesions were not very uniform, but the result seems favorable to the serum, except that there was an insufficient number of controls. No protection was seen when the serum was given three days before the tuberculin. May 13, 1896. 4 pigs were tested on the fifty-first day. The serum was given intraperitoneally in doses of from 8 to 10 c.c.; the tuber- culin subcutaneously in simultaneous doses of 0.300 c.c. 2 pigs that received serum and tuberculin died in 10 to 12 hours; 2 pigs used as controls and receiving tuberculin only, died in 10 hours. Result. Inconclusive; there was no better absorption of the serum infiltrate. April 28th. 12 pigs of an average weight of 529 grammes were inocu- lated in the groin with sputum that was of moderate virulence (though two died in from 28 to 29 days). May 28th. 10 pigs were tested on the thirtieth day. The serum was given subcutaneously in simultaneous doses of from 3 to 12 c.c. Tuber- culin in a separate place in divided doses: 12 m., 0.300 c.c.; 6 p.m., 0.200 c.c. Of 7 pigs that received serum and tuberculin, 3 survived, while 4 died in from 9 to 10 hours; control, of 3 pigs receiving tuber- culin*. only, 2 survived ; 1 died in 15 hours. Result. The serum infiltrate was not well absorbed, though the de- pression from the tuberculin was not marked until after the second dose. The controls had a higher temperature following the first dose. The TRUDEAU: ANTITOXINS FOR TUBERCULOSIS. 21 lesions were not uniform and the controls were too few. No conclusion was warranted. July 11,1896. 20 pigs of an average weight of 496.7 grammes were inoculated intraperitoneally with a small drop of the juice of very viru- lent tuberculous tissue. August 1st. We tested 10 pigs on the twenty-first day. All were very ill with acute tuberculosis. Serum was given intraperitoneally ; tuber- culin subcutaneously at the same time to 7 pigs; the dose of the tuber- culin was 0.400, of the serum 2 to 8 c.c. 1 survived, while 6 died in from 6 to 22 hours; 3 control pigs that received tuberculin only died in 9 hours. Result. Autopsies revealed acute miliary tuberculosis very much alike in all. The amount of serum absorption was doubtful. The result seemed slightly favorable to the serum, but the condition of the animals was bad for the test. Serum of Ass H, First Bleeding. May 27, 1896. 10 pigs of an average weight of 534 grammes were in- oculated in the groin with a non-virulent sputum, which kills in 60 days. July 11th. They were tested (fifty-first day). Serum was given intra- peritoneally in doses of from 2 to 10 c.c.; tuberculin subcutaneously in simultaneous doses of 0.400 c.c. Of 5 pigs that received serum and tuberculin 2 survived and 3 died in from 6 to 9 hours; of 5 control pigs that received tuberculin only, 3 survived, while 2 died in from 6 to 11 hours. Result. The serum evidently was not absorbed to any extent. The lesions were not uniform, though the controls were of sufficient number to neutralize that objection. Evidently no conclusion was warranted, as too many controls survived. July 11, 1896. 10 pigs from a lot of 20 were inoculated intraperi- toneally with very virulent material (see previous experiment, July 11th) ; 2 pigs that died were not used ; all became very ill. August 1st. 8 pigs were tested (twenty-first day). Serum was given intra- peritoneally in doses of from 2 to 8 c.c. ; tuberculin subcutaneously in simultaneous doses of 0.400 c.c. Of 4 pigs that received both serum and tuberculin 1 survived and 3 died in from 5 to 10 hours; the 4 controls that received tuberculin only, died in from 6 to 10 hours. Result. The lesi ons were very uniform and were those of acute miliary tuberculosis. The serum was probably unabsorbed, and the condition of the animals was bad for the test. No conclusions were warranted. Comparisons Between Various Serums, etc. The following experiments were made to compare the effects of vari- ous serums when given with and without tuberculin. In explanation of the serums tried which were other than our own products, it may be stated that: “Cow D.” Was obtained from an animal that had recovered from pulmonary tuberculosis and had been injected with large quantities of tuberculin. 22 TRUDEAU: ANTITOXINS FOR TUBERCULOSIS. “ Horse I.” Was said to have been treated with tuberculous “ toxins ” from virulent cultures without bacilli. “ Horse II.” Had been inoculated subcutaneously with non-virulent cultures. “ Horse III.” Was said to have been treated with “ toxins” from virulent bacilli freed from the culture fluid. “Ass M.” Was a tuberculinized healthy ass. In addition to the anti-tuberculous serums we used normal horse serum, diphtheria antitoxin from the New York City Board of Health, sheep serum from a healthy sheep, and a 0.6 per cent. NaCl solution. July 28, 1896. To test the serum from cow D, 10 pigs of an average weight of 714 grammes were inoculated in the groin with virulent tissue juice from a tuberculous pig ; 5 died before the test in from 40 to 50 days. September 28th. We tested 5 pigs on the sixty-second day. Serum was given subcutaneously, mixed with tuberculin, to 3 pigs, the serum in doses of 5, 5, and 8 c.e., the tuberculin in a dose of + 0.400 c.c.; all died in from 4 to 7 hours; 2 control pigs received 0.400 c.c. of tuber- culin and both died in from 3 to 5 hours. Result. All the serum was absorbed. The lesions were uniformly advanced caseous tuberculosis. No conclusion could be drawn, as the dose of tuberculin was probably excessive. August 24,1896. To test the serums from cotv D and horse I, 10 pigs of an average weight of 714 grammes were inoculated in the groin with virulent tissue juice from a tuberculous pig; 2 died on the 43d day. October 7th. We tested 8 pigs on the forty-fourth day. Serum wras given subcutaneously, mixed with tuberculin; 3 pigs receiving the serum from cow D in doses of 2.8, 2.6, and 2.7 c.c. + 0.300 tuberculin died in from 6 to 8 hours; 2 pigs that received the serum from horse I in doses of 2.36 to 2.64 c.c. -j- 0.300 tuberculin died in 8 hours. Of 3 controls that received 0.300 c.c. tuberculin only,l survived and 2 died in 8 hours. Result. The serum infiltrate was not well absorbed, and the dose of tuberculin was too large. No effects favorable to the serums could be seen. October 12, 1896. We tested the serums of cow D and ass H, first bleeding. 10 pigs of an average weight of 657.4 grammes were inocu- lated in the groin with a pure culture of virulent tubercle bacilli that killed in from 40 to 50 days. They were tested on the thirty-second day. Serum was given subcu- taneously to guinea-pigs in dose of 0.005 to 1 gramme of pig, diluted with equal quantities of a 0.6 per cent. NaCl solution. Tuberculin in a dose of 0.0006 to 1 gramme of pig, diluted, was given subcutaneously 30 minutes later, on the opposite side of the abdomen ; 5 pigs received the serum of cow D in doses of from 2.5 to 2.75 c.c. and tuberculin in amounts of from 0.250 to 0.450; 1 survived, 4 died in from 10 to 14 hours; 2 pigs received the serum of ass H in doses of 2.5 and 2.75 c.c. and tuberculin in doses of 0.250 and 0.450; both died in from 9 to 10 hours; 3 controls received tuberculin only in doses of from 0.250 to 0.450 and died in from 9 to 13 hours. Result. The serum was incompletely absorbed, except in the pig that survived. The lesions were very uniform and showed generalized tuber- TRUDEAU: ANTITOXINS FOR TUBERCULOSIS. 23 culosis with beginning caseation. Very little, if any, protective influ- ence of the serum could be seen. August 23,1897. Serum, of horse II. 5 pigs of an average weight of 508 grammes were inoculated in the right groin with a virulent first culture. September 25th. They were tested on the thirty-third day. 3 pigs re- ceived serum subcutaneously in a dose of 0.005 to 1 gramme of pig and tuberculin in a dose of 0.200; 1 survived, 2 died in 12 hours; 2 pigs received tuberculin only in the same dose and died in 12 hours. Result. Slightly favorable to serum protection in one pig. The following experiment is so instructive that we give the full de- tails in tabular form. The apparently marked protective influence exerted by both the serums and the physiological salt solution illustrates the fallacies in deductions of supposed specific effects of the serum as determined by such tests. November 23, 1896. We tested the serums of horse II, rabbit VIII, and ass R in a 0.6 per cent. NaCl solution. 19 pigs of an average weight of 517 grammes were inoculated with a pure culture of tubercle bacilli of moderate virulence that killed in from 50 to 60 days. We tested 13 pigs on the forty-second day with serum and salt solu- tion, in a dose of 0.006 to 1 gramme of pig, diluted with an equal quan- tity of water; it was given subcutaneously in the right side, and tuber- culin in a dose of 0.0002 to 1 gramme of pig, diluted with water, Avas given subcutaneously, 45 minutes later, in the left side. 5 pigs re- ceived the serum of horse II in doses of 2.58 c.c. to 3.90 c.c.; 1 died in 15 hours, 4 survived; 2 pigs received the serum of rabbit VIII in doses of 3.48 to 3.51 c.c. and both survived; 6 controls received tuber- culin only, and all died in from 7 to 11 hours. We tested 5 pigs on the forty-fifth day. 1 pig that received 2.70 c.c. of the serum of ass R survived ; 2 pigs that received 0.6 per cent. NaCl, 3.00 to 3.19 c.c., survived; 2 controls died in 10 hours. See Table I. (A) and (B). Result. The lesions in this lot of animals were very uniform, and the favorable showing for the serums seemed more than a coincidence. The tuberculin reaction was typical in the organs, the spleens being enormously enlarged and ecchymotic; there was free serum in the cavities; other local congestions were marked. The serum infiltrate seemed to be absorbed but slightly in most animals, even after two days. The results seen in Tables I. (A) and (B) led us to make comparative tests of various non-specific serums and of salt solution under the same conditions. The number of animals in some cases was too small to make the results of any value, but they are suggestive. Some experiments are conveniently tabulated. November 13, 1897. The serum of horse II and normal horse serum “ jyy» 27 pigs of au average weight of 537 grammes were inoculated in the right groin with a culture of fair virulence ; 5 died. The serums and tuberculin were given subcutaneously and in different places 45 minutes apart; the serum in doses of 0.006 to 1 gramme of pig, the tuberculin in doses of 0.200 to 0.500 in pigs Nos. 1 to 15, and in doses 24 TRUDEAU: ANTITOXINS FOR TUBERCULOSIS Weights. Tuberc. Temp. Died. Days Serum, Serum, 12.45 Condition. survived Autopsies. No. Gain or loss. kind. 12 M P.M. 7 P.M. or XI/23,’C6 1/4, ’97 Dose. Dose. 12 M. 7 P.M. Hour. Hours. killed. 1 430gm. 430gm Ogm. Horse II 2.58 c.c. 0.086 102.5° 103° Lively Survived General caseous tuberculosis. 17 days 2 680 645 —35 3.87 0.129 103.5 101 Fair Survived General caseous tuberculosis (less). 16 days 3 460 465 + 5 “ 2.79 0.093 103 101 “ Killed Acute tuberculosis , beginning caseation ; some 2d day infiltrate at site of injection. 4 660 650 —10 3.90 0.130 103 102 Survived 14 days General caseous tuberculosis. 5 465 460 — 5 “ 2.76 0.092 103.5 101 U 3 A.M. 15 Acute tuberculosis; ruptured spleen ; serum partly unabsorbed. 6 610 580 —30 Rabbit I 3.48 0.116 102.5 101 Weak Survived Acute tuberculosis ; serum all absorbed. 1 day 7 565 585 +20 3.51 0.117 102.8 103 Depressed Survived 1>2 day “ miliary tuberculosis ; beginning caseation. 8 685 620 -65 Control 0 0.124 103. 3 95 Dying 9.30 P.M. 9 9 535 675 +140 (pregn’t) 0 0.135 102 97 Weak 9 10 562 550 —12 il 0 0.115 102 98.5 Depressed 10 30 11 u u u a “ 11 540 540 0 0 0.108 101.5 105 Weak 11 12 490 465 —25 ii o 0.093 103.5 Dying Weak 7 30 7 << (< a u a 13 505 515 +10 “ 0 0 092 103.5 101 9.30 9V2 ■< « « << a 14 590 r>pf*. 98 1896 35th day of disease Table I (B) —Test of 5 Pigs on Forty fifth Day. No. Weights. 12.30 1 P.M. 12.30 7.30 1/7,’97 P.M. P.M. P.M. 15 425 gm. 450 gm +25 g n. Ass R.I. 2.70 C.C. 0.090 101° 104.5° Lively Killed General miliary tuberculosis; beginning case- 2d day ation ; serum infiltrate still under skin. 16 425 500 +75 NaCl 3.30 0.100 101.6 102 Weak “ NaCl solution absorbed. 17 390 365 —25 “ 3.19 0.073 102.5 106 Lively U << u a a u »< it 18 485 435 —50 Control 0 0.087 101.5 99.5 Weak 11 P.M. 10 tt n n a a n 19 499 525 +26 “ 0 0.108 102 101.8 “ “ 10 “ “ “ “ “ “ Table I (A).—Test of 13 Pigs on Forty second Day. TRUDEAU: ANTITOXINS FOR TUBERCULOSIS. 25 Weight. Serum, Serum, Tuberc. 12.45 Temp. Condition. Died. Survived Autopsies. No Xl/13,’97 XIl/25,’97 Gain or Iras. kind. 12 M. Dose. P.M. Dose. 12 M. 6 P.M. 12 M. 6 P.M. Hours. or killed. 1 662gm. 580gm. —82gm. Horse II 3.48 c.c. 0.232c. C. 102° 96° Good Fair 15 General tuberculosis; beginning caseation of liver and spleen ; some infiltrate from the serum. 2 600 480 —120 2.88 0.192 103.5 97 U 9 General tuberculosis; beginning caseation, liver, etc.; serum infiltrate unabsorbed. 3 490 400 —90 “ 2.40 0.160 100 97.5 Poor Weak 10 General tuberculosis; more caseation of liver. 4 470 360 —110 N 2.16 0.144 100.6 98.5 Fair “ 10 General tuberculosis; beginning caseation; serum infiltrate unabsorbed. 5 620 565 -55 0 0 0.226 101.5 102 ‘ Fair 28 hours General tuberculosis; beginning caseation; spleen not large. 6 560 500 —60 0 0 0.200 101.8 102 “ “ Surv. 12 dys Gen. tuberc.; slight caseation ; spleen not large. 7 490 415 —75 N 2.49 0.166 102.6 96.5 “ io General tuberculosis; more caseation ; serum in- filtrate unabsorbed. 8 425 380 —40 II 2.28 0.152 103 98.5 Poor 4< 15 General tuberculosis; slight caseation; serum in- filtrate fairly absorbed. 9 405 360 —45 0 0 0.144 101.5 96 Good Weak 10 General tuberculosis; beginning caseation. 10 365 340 —25 II 2.04 0.136 102.8 102 “ Fair Surv. 22dys General tuberculosis; slight caseation ■ marked caseation of the lungs. 11 355 340 —15 0 0 0.136 103.8 104.5 “ “ “ 45 “ General tuberculosis; marked caseation. 12 385 340 —45 N 2.04 0.136 103.8 103.5 Fair “ “ 18 “ Gen. tuberc.; marked caseation except spleen. 13 465 430 —35 II 2.58 0.172 102.2 96.5 Good Weak 10 General tuberculosis; beginning caseation ; serum infiltrate unabsorbed. 14 435 360 —75 0 0 0.144 104.4 105 Poor Fair Surv. 22 dys Gen. tuberc.; more caseation ; marked in lungs. 15 350 370 +20 0 0 0.074 102.5 101.8 Good 15 General tuberculosis; beginning caseation. 16 555 560 + 5 N 3.36 0.112 102.2 100 Strong Surv. 37 dys General tuberculosis; massive caseation in lungs; not in liver or spleen. 17 500 510 +10 N 0 0.102 102.2 102 “ 15 General tuberculosis ; beginning caseation. 18 705 560 —145 0 0 0 97 Weak Weak 12 Not treat’d; too weak General tuberculosis ; advanced caseation of liver and spleen. 19 495 430 -65 0 0 0.086 103.6 95 Fair Fair 10 General beginning caseous tuberculosis. 20 470 430 —40 II 2.58 0.086 102 100 “ ‘ 15 General beginning caseous tuberculosis; serum fairly absorbed. 21 560 525 —35 0 0 0.105 103 100.5 Good “ 18 Ruptured liver; general tuberculosis. 22 395 400 + 5 II 2.40 0.080 102.5 103 li Surv. 5 dys General tuberculosis; beginning caseation; spleen small. 23 400 340 —60 N 2.04 0.068 102.5 99.8 Weak 18 General tuberculosis; slight beginning caseation ; serum unabsorbed. Table II.—Tested 23 on Fokty-second Day. 26 TRUDEAU: ANTITOXINS FOR TUBERCULOSIS. Weights. Serum. Temp. Died or killed. Tuberc. 1’ondi- No. Xl/2,’96. 1/7,’97 Gain or loss. Kind. 12.30 P.M. Dose. 12.45 p.m. Dose. 12.30 P.M. 7.30 P.M. tion. 7.30 Hours. Days. Autopsies. 1 705gm. 610gm. — 95gm. 3.66 C.C. 0.122 c.c. 102.4° 100° Fair 24 1 General caseous tuberculosis; serum infiltrate not absorbed from skin. 2 710 510 —200 Control 0 0.102 95 Dying 8 General caseous tuberculosis. 3 710 630 — 70 0 0.126 102 103.2 Fair Vi/2,’97 Chronic caseous tuberculosis (Albino). Table IV. Weights. Seram. Temp. Died or killed. Condi- Tuberc. Condi- No. Gain or loss. tion. Dose. tion Autopsies. XII/30,’96 ix/8,’97 Kind. Dose. 12 M. 6 P.M. 6 P.M Hours. Days 1 655gm. 625gm. —30gm. Fair Ass R 3.75 c.c. 0.125c.c. 103° 100.4° Fair 15 General caseous tuberculosis ; serum infiltrate not (doe) all absorbed. 2 545 585 —60 U il 2.90 0.097 102.2 102.6 %t K. 10 General caseous tuberculosis. 3 610 545 —65 Poor Ass II 3 27 0.109 102.4 103.4 “ 2% General caseous tuberculosis ; serum well absorb’d 4 640 585 —55 Good “ “ 3.50 0.117 102.6 99.2 Weak io General caseous tuberculosis; serum infiltrate not (doe) well absorbed ; more caseation. 5 670 620 -50 Horse I 3 72 0.124 102.4 103.2 Fair K. 9 General caseous tuberculosis; moderate caseation 6 445 385 -60 Poor “ «< 2.30 0.077 102.6 99.6 Weak D. 9 General caseous tuberculosis; slight beginning caseation. 7 525 410 —115 111 NaCl 2.46 0.082 101.4 103.2 Fair 15 a n a '< 8 490 425 —65 Fair (doe) Poor “ 2.55 0.095 103 104 “ K. 10 it a “ “ 9 525 440 -85 Control 0 0.088 101.2 95 Weak 15 « 10 525 470 —55 Good “ 0 0.094 103.2 101. [ 9 “ “ “ “ 11 620 610 —10 “ 0 0.122 102.2 104.5 Fair K. 9 a u u a 12 655 Died l/l 0, ’97 Emaciated; acute peritonitis. 13 480 I/ll,’97 Emaciated; nearest lymph nodes tuberculous. 14 450 “ 1/27 ,’97 Emaciated; nearest lymph nodes and spleen tuberculous. 15 590 n/6,’97 General caseous tuberculosis. Table III. TRUDEAU: ANTITOXINS FOR TUBERCULOSIS. 27 of from 0.100 to 0.500 in pigs Nos. 15 to 23. Of 8 pigs that received the serum of horse II, 2 survived and 6 died in from 9 to 15 hours. 6 pigs received the serum of horse “ N,” 2 survived, while 4 died in from 10 to 18 hours. (See Table II.) Result. The lesions were fairly uniform, but the test was too equivo- cal for any value, and the serum infiltration was not so well absorbed as in Table I. November 2, 1896. Hydrocele fluid. 3 pigs were inoculated with a culture of tubercle bacilli of moderate virulence. They were tested on the sixty-sixth day. Hydrocele fluid (0.006 to 1 gramme of pig) was injected subcutaneously after diluting it one- half ; tuberculin, in a dose of 0.0002 gm. to 1 gramme of pig, was given subcutaneously 15 minutes later. (See Table III.) Result. No conclusion could fairly be made. December 30, 1896. We tested the serums of horse I, ass R 2d, ass H 2d, and a 0.6 per cent. NaCl solution. If? pig3 of an average weight of 558.5 grammes were inoculated in the right groin with a virulent pure culture, that killed in from 30 to 50 days. They were tested on the fortieth day. The serum of horse I was given in a dose of 0.006 to 1 gramme of pig; the serum of ass R 2d, the serum of ass H 2d, and 0-2 per cent, of NaCl, 07006, were given in the same dose; the tuberculin was given in a dose of 0.0002 to 1 gramme of pig, 40 minutes after the serum. The serum was given subcuta- neously on the left side; the tuberculin on the right side; both were diluted one-half with water. (See Table IV.) Result. These animals revealed fairly uniform lesions much like those in Tables I. (A) and (B), and at nearly the same stage in the dis- ease. All had caseous retroperitoneal lymph nodes ; enlarged, and begin- ning caseous, liver and spleen, and miliary tuberculosis in the lungs. The amount of caseation varied, being somewhat less in Nos. 6, 7, 8, and 11 (Table IV.) than in the rest. This latter factor seemed to influence the severity of the poisoning by tuberculin, which is usually more marked the greater the caseation and the age of the disease. Therefore the experiments in Tables IIP. and IV. have little use for safe deductions. January 5, 1897. We tested the serums of cow D, ass R 2d, a normal sheep, and 0.6 per cent. NaCl solution. 10 pigs of an average weight of 483.4 grammes were inoculated with a culture of tubercle bacilli of moderate virulence from the same source as in the experiment of Novem- ber 23, 1896. February 16th. We tested them on the forty-second day. The serums, etc., were given subcutaneously, diluted with an equal volume of 0.6 per cent. NaCl, giving 0.004 to 1 gramme of pig; tuberculin, diluted, was given subcutaneously, 45 minutes later, in a dose of 0.0002 to 1 gramme of pig; 2 pigs that received the serum of cow B died in from 12 to 20 hours; of 2 pigs that received the serum of ass R 2d, 1 survived and 1 died in 12 hours; of 2 pigs that received the serum of a sheep (nor- mal), 1 survived and 1 died in 9 hours; 2 pigs received NaCl in 0.6 per cent, solution, and died in from 9 to 10 hours; 2 controls died in from 9 to 12 hours. Result. The infiltrations from the ass and cow serums were not well absorbed; the others were entirely so. The NaCl seemed completely absorbed when the tuberculin was given. The lesions were very uni- form. Possibly some protection was shown by the serums. 28 TRUDEAU: ANTITOXINS FOR TUBERCULOSIS. Days in- Weight. Tuberc. 0.6 per cent. NaCl solution. Temperature. Survived No. oculated. II/18,’97 Loss or gain. 12 M. Dose. 12 m. Dose. 3 P.M. Dose. 6 P.M. 12 M. 5.30 P.M. or died. Autopsies. 1 50 350gm. —125gm. 0.350 Control 0 0 0 100° Dying 6 hours General tuberculosis; slight caseation; re- action. 2 50 475 — 75 0 “ 0 0 0 102 104.4° K. after 24 hours General tuberculosis; slight caseation; tuber- culin not given. 3 50 555 —115 0.555 NaCl 3.33 c.c. 3.30 c.c. 3.50 C.C. 104.2 100.2 15 hours General tuberculosis; NaCl nearly all ab- sorbed ; reaction apparent. 4 50 565 — 55 0.566 “ 3.39 3.29 3 50 103.2 101.2 11 hours 5 44 430 — 40 0.430 “ 2.50 2.50 2.50 103.2 105 Survived “ 6 44 505 + 70 0.505 Control 0 0 0 102.2 99.4 10% hrs. “ “ No. Inocula- tion. Test day. Weight. Serum, etc. January 29th. Temperature. January 28th. Temperature. January 29th. Temp. Condition. xi/5,’96 1/28,’97 Kind. Dose. 11.30 A.M. 5.30 P.M. 11.30 A.M. 5.30 P.M. variation. 1 xx/5,’96 85th 395 gm. 510 gm. Control 0 102.5° 101.6° 101° 101° 0 Fat 2 “ “ 285 340 Rabbit I 1.70 c.c. 102 102 99.2 102 +2.8° Thin 3 ll “ 475 525 Ass R 2 2.62 102.5 101.2 102 101.4 +1.6 Fat 4 a “ 445 540 NaCl 2.70 102 101.6 100.2 101.6 +1.4 “ 5 X/ 5,’96 116th 480 420 Ass R 2 2.10 102 101.4 101.2 102 +0.8 Poor Died in 18 hours; was quite ill. 6 (< “ 380 500 “ 1.25 101.5 100.4 101.2 100.8 —0.4 Good 7 << “ 545 “ 2.77 101 100.3 101 100.4 —0.6 “ 8 Not inoc ulated. 470 2.00 101 101.2 +0.2 “ Table VI. (A). Table V. TRUDEAU: ANTITOXINS FOR TUBERCULOSIS. 29 January 20.1897. We tested the serums of horse II and diphtheria antitoxin (300 units'). 10 pigs of an average weight of 529.2 grammes were inoculated in the groin with a culture of tubercle bacilli of moder- ate virulence (2 died before testing). February 26th. We tested 8 pigs on the thirty-seventh day. The serums were given subcutaneously, diluted with an equal volume of water in a dose of 0.005 to 1 gramme of pig (2 to 4 c.c.); the tuberculin, diluted, was given subcutaneously, 40 minutes later, in a dose of 0.0006 to 1 gramme of pig (0.234 to 0.489 c.c.); 4 pigs that received the serum of horse II died in from 7 to 11 hours; of 2 pigs that received diphtheria antitoxin, 1 survived, while 1 died in 6 hours; 2 controls died in from 8 to 11 hours. Result. The serums were not perfectly absorbed and the dose of tuber- culin was so large that no deduction is warranted. Note.—These animals had been used on the twenty-second day of the disease with the serum of ass R 1st and small doses of tuberculin (see experiment of January 20, 1897). February 19, 1897. We compared the serum of ass R, second bleeding, and a 0.6 per cent. NaClsolution. 16 pigs of an average weight of 437.5 grammes were inoculated in both groins with a culture of tubercle bacilli of weak virulence (all but 3 pigs had gained weight at the end of two months). April 23d. We tested all on the sixty-third day. The serum and the NaCl were given, undiluted, subcutaneously in a dose of 1 to 2 c.c.; tuberculin, diluted, was given 1£ hours later, subcutaneously, in a dose of 0.300 c.c. ; 6 pigs that received the serum of ass R, 2d, all died in from 7 to 11 hours; 3 pigs that received NaCl solution, all died in from 7 to 9 hours; of 7 controls, 2 survived and 5 died in from 9 to 12 hours. Result. The serum infiltration was not completely absorbed ; the NaCl solution was entirely so. The lesions were very uniform and showed chronic fibro-caseous lymph nodes ; very large, slightly caseous spleen ; beginning caseation in the liver ; miliary tuberculosis of the lungs. The dose of tuberculin was rather larger than was necessary for most of the animals ; no influence could be seen. Serum of ass M. 3 pigs were taken 55 days after infection (a); 2 pigs, 6\ months after infection (6); tuberculin was given 40 minutes later than the serum, (a) 2 pigs received the serum of ass M in a dose of 1 and 2 c.c. subcutaneously, and tuberculin, 0.100 c.c.; 1 survived and 1 died in 10 hours; 1 control-pig injected subcutaneously with tu- berculin, 0.100 c.c., died in 12 hours; (b) 1 pig received of the serum of ass M 2.5 c.c., and of tuberculin, 0.100 c.c.; it died in 10 hours; 1 control-pig received tuberculin, 0.100 c.c., and died in 12 hours. (See Tabled-) Result. There was no certain protection. A 0.6 per cent. NaCl solution in repeated doses, with large doses of tuberculin. 6 pigs of an average weight of 480 grammes r-eeeived 4 inoculatfolis on December 30, 1896, and 2 on January 5, 1897,‘4e?a pure culture of tubercle bacilli of moderate virulence; NaCl solution was given in a dose of 0.006 to 1 gramme of pig, and tuber- culin in a dose of 0.001 to 1 gramme of pig. c Table V) Result. The excessive doses of tuberculin and the prolongation of life in the animals that received the NaCl solution suggest that the 30 TRUDEAU: ANTITOXINS FOR TUBERCULOSIS. influence of the latter may be due to a supporting effect on the failing hearts of the poisoned animals by adding bulk to the blood or diluting and promoting the excretion of the poison. May 27, 1897. 0.6 per cent. NaCI solution in a single dose. 15 pigs of an average weight of 676.7 grammes were inoculated in the groin with a first culture of virulent tubercle bacilli; 4 died in from 5 to 34 days. July 2, 1897. We tested 11 on the thirty-sixth day. The NaCI solution was given subcutaneously to six in a dose of 0.010 to 1 gramme of pig; the tuberculin, subcutaneously, in a dose of 0.150 to all, 15 minutes later; of the 6 pigs that received NaCl solution in doses of 4.75 to 7.30 c.c., 2 survived and 4 died in from 6 to 12 hours ; of 5 controls that received tuberculin only, 3 survived and 2 died in from 6 to 12 hours. Result. No protective influence of the NaCl can be seen in this ex- periment, though it seems apparent in some others. The lesions showed virulent infection, but not uniform caseation. Effect of the serums on the temperature of tuberculous guinea-pigs (with- out tuberculin). Serum of ass R, second; serum of rabbit VIII; and solution of the NaCI, 0.6 per cent. Miscellaneous pigs, inoculated with non-virulent tubercle bacilli and having chronic disease, were used. (See Table VI.) CA> Result. There was apparently no influence on the temperature. Test 3 days later of the first four pigs (Table VI., B) with tuberculin. Ass R, second serum, was given in separate places at the same time, subcutaneously. No. Serum. Tuberc. Dose. Temperature. Kind. Dose. A. M. P. M. Variation. 1 . Ass R 2 2.15 c.e. 0.015 c.c. 101° 105° +4° 2 . 0 0 0.010 101 104 +3 3 . 0 0 0.015 102.4 104.6 +2.2 4 . Ass R 2 2.70 0.0155 101.6 102.4 +1.8 Table VI. (B). Result. There was no effect in preventing a rise of temperature. January 20,1897. Effects of serum on the temperature when given with small doses of tuberculin. 10 pigs were inoculated with bacilli of mod- erate virulence. (1 died of enteritis.) C Table vMj) They were tested on the twenty-second day. The serum of ass R, second, and NaCI, 0.6 per cent, solution, were given in doses of 0.004 to 1 gramme of pig, diluted one-half; tuberculin in a dose of 0.0005 to 100 grammes of pig ; they were given in separate places at the same time, subcutaneously. Result. The controls seemed to have a greater rise of temperature, but the NaCI pigs seemed to have less rise than the serum animals, thus making a specific effect of the serum less likely. January 20, 1897. Serum of ass R, second bleeding, and of horse II. 6 pigs of an average weight of 492.5 grammes were inoculated with a pure culture of tubercle bacilli. TRUDEAU: ANTITOXINS FOR TUBERCULOSIS. 31 Table YIT. Weight. Serum, etc. Tuberc. Temperature. No. i/20,’97 11/11/97 Gain or loss. Kind. Dose. Dose. 12 M. 6 P.M. Variation i 665 gm. 605 gm. —60gm. Ass R 2 2.40 c.c. 0.003 c.c. 102. 2° 102.2° 0 2 400 395 — 5 “ 1.58 0.0019 103.6 105 +1.4° 3 390 400 + 5 “ 1.60 0.002 108.8 104.4 +0.6 4 420 415 — 5 “ 1.66 0 102 102.6 +0.6 5 375 345 —30 NaCl 1.48 0 0017 102 101.6 —0.4 6 390 440 +50 “ 1.76 • 0.022 103 103.6 +0.6 7 445 450 + 5 Control 0 . 0.022 102.2 104.6 +2.4 8 610 610 0 0 0 0.003 102 103.6 +1.6 9 10 920 875 —45 0 0 0.0043 103 104.8 +1.8 The test was on the twenty-fewrt-h day. Serum was given in doses of 0.006 to 1 gramme of pig; tuberculin in doses of 0.0005 to 100 grammes of pig; they were given simultaneously, in separate places under the skin. CToble vm a) No. Weight. Serum. Ass R 2 Dose. Tuberc. Dose. Temperature. 1/20,’97 11/18/97 Gain or loss. 12.30 p.m. 5.45 P.M. Variation. 1 475 gm. 450 gm. —25gm. 2.70 c.c. 0.0022 c e. 102.8° 104.6° 41-8° 2 555 545 —10 3.27 0.0032 102.2 Died soon after dose 3 430 415 —15 2.49 0.0024 102.8 104.4 41-6 4 480 400 -80 Control 0.0024 103.8 105.2 41.4 5 500 535 435 “ 0.0020 103.2 104.4 41.2 6 515 505 —10 “ 0.003 102.6 102.8 40.2 Table VIII. (A). Result. The serum did not prevent the temperature rise due to tuber- culin. The serum of horse II was tested on the forty-second day. Serum was given in a dose of 0.004 to 1 gramme of pig; tuberculin in equal vol- ume was given subcutaneously 10 minutes after the serum. Liable vm b; Result. There was no influence upon the effect of non-fatal, large doses of tuberculin. January 5, 1897. Serums of horse I and horse II. 10 pigs were inocu- lated with a pure culture of tubercle bacilli of moderate virulence. (See Table IX.) They were tested on the forty-ninth day. The serum and the tuberculin were given in separate places under the skin of the abdomen, simul- taneously. 32 TRUDEAU: ANTITOXINS FOR TUBERCULOSIS. Table VIII. (B). Weight. Serum. Tuberc. Temperature. No. 1/20,’97 III/4,’97 Gain or loss. Kind. Dose. Dose. 10.30 A.M. 5.45 P.M. Variation 4 480 gra. 380 gm. —lOOgm. Horse II 1.10 c.c. 0.076 e.c. 102° 104.4° +2.4° 3 430 420 —10 “ 1.68 0.084 102. 2 105 +2.8 6 515 520 +5 “ 2.08 0.104 102.4 99 —3.4 1 475 435 —40 Control 0 0.097 101.8 103 +1.2 5 500 555 +55 “ 0 0.111 102 105 +3 Result. There was evidently no preventive effect on the temperature reaction. March 27, 1897. Effect of the serums of R and II on the local reaction after tuberculin injections into the eyes of rabbits (see Part I.). Three weeks after discontinuance of treatment and 56 days after the inocu- lation of the eyes, the congestion of the acute stage in the disease was decreasing; some eyes had become caseous to a large extent. A small dose (0.010) of tuberculin was given to 5 of the animals (not mixed with serum); serum was given at the same time. 2 received 3 c.c. each of serum E,; 2 received 3 c.c. each of serum II; 1 control received tuberculin only. Table IX. Weight. Serum. Tuberc. Temperature. No. 1/5,’97 11/23,’97 Gain or loss. Kind. Dose. Dose. 12.30 PM 5.30 p.m. Variation 1 440 gm. 280 gm. —60gm. Horse I 2.0 c.c. 0.002 c.c. 101.2° 105° +3.8° 2 455 465 +10 “ 1.5 U 102.4 105.4 +3 3 377 405 +28 “ 1.5 ft 103.2 104.4 +1.2 4 777 760 —17 Horse II 2.0 101.2 105.2 +4.0 5 360 410 +50 “ 1.0 103 105.8 +2.8 6 760 630 —130 “ 1.5 103.2 104.6 +1.4 7 407 465 +58 Control 0 <« 102.2 104.8 +2.6 8 792 785 — 7 “ 0 u 102.2 104.2 +2 9 490 470 —20 “ 0 103 105.6 +2.6 10 685 690 + 5 0 u 103 104.2 +1.2 Result. There was slight additional redness in all, with a tempera- ture of 104.5°, an increase of 2° to 2.5° after six hours. This was not less with the serum rabbits. April 1,1897. The test was repeated with three controls; the dose of tuberculin was 0.020. Result. There was slight increase of congestion and temperature in all, without perceptible difference between the controls and the serum animals. There was no influence in preventing local reaction. TRUDEAU: ANTITOXINS FOR TUBERCULOSIS. 33 Resume. The results of our four years’ work in experiments upon four sheep, three asses, twelve fowls, eighteen rabbits, and four hundred and fifty guinea-pigs are to be found in the following summary : 1. A sheep was injected intravenously with killed thymus cultures. The result was so unsatisfactory that the serum-tests were not conclusive. 2. Chickens were inoculated intraperitoneally with mammalian tuber- culosis. The serum revealed no germicidal or inhibitive action on the tubercle bacilli, nor favorable influence on the course of the disease in guinea-pigs. 3. A sheep was injected with tuberculin. The serum was wanting in germicidal, antitoxic, or curative effect so far as tested. 4. A sheep was inoculated intravenously with non-virulent cultures. Cachexia followed, and the serum was therefore not used. 5. An ass was inoculated as in (4.); it died from pulmonary embolus. The serum was not bactericidal to tubercle bacilli. 6. An ass was inoculated with virulent tubercle bacilli and treated with tuberculin. The serum showed no germicidal nor curative, but possibly some antitoxic effect. 7. An ass was inoculated with non-virulent tubercle bacilli and treated with various extracts of tubercle bacilli and dead bacilli. The serum showed no activity. 8. Rabbits were inoculated with non-virulent and virulent tubercle bacilli, and recovered. Their serum possibly conferred some protection in tuberculin poisoning, and possibly prolonged the lives of treated guinea-pigs. With a full appreciation of the uncertainty of correct conclusions from tests of the serums other than our own product which were tried with tuberculin, we may state that only one indicated antitoxic power. This was obtained from a horse inoculated with non-virulent cultures. That the apparent protection against fatal tuberculin poisoning occa- sionally seen was not necessarily due to the specific antitoxic power of the serums is made probable by the similar effects of physiological salt solution seen at times. None of the serums appeared to prevent local or general reaction from small doses of tuberculin, nor to influence the temperature of tuberculous animals. Disappointing as these results may seem, the writers feel that, in the light of recent contributions made by Ehrlich, Wasserman and Behring10 to our knowledge of the mechanism of immunity and antitoxin pro- duction in the body, the outlook for an efficient tuberculosis antitoxin is by no means a hopeless one. In conclusion, we would express our thanks to Drs. S. W. Hewetson, W. S. Nelson, and J. A. Wilder, all of whom have aided in the many arduous details of these experiments. Ay»i- >. • 34 TRUDEAU: ANTITOXINS FOR TUBERCULOSIS. Literature. 1. Abel. (Injections of Dead Tubercle Bacilli.) Deutscb. med. Wochens., 1892, p. 482. (Ref. Masur u. Kockel.) 2. Arloing, Rodet et Courmont. Etudes experimentale sur la propri