UNIVERSITY OF ILLINOIS LIBRARY AT SANA-CHAMPAIGN AGHICUt-TUP.r WON CIRCULATING CHECK FOR UNBOUND CIRCULATING COPY UNIVERSITY OF ILLINOIS Agricultural Experiment Station BULLETIN NO. 149 TUBERCULOSIS OF FARM ANIMALS BY CHAS. F. BRISCOE AND W. J. MACNEAL, \ URBANA, ILLINOIS, FEBRUARY, 1911 SUMMARY OF BULLETIN No. 149 1. Tuberculosis was known as early as 460 B. C. It is a disease due to a specific organism, B. tuberculosis. Four types of this organ- ism are recognized; the human, bovine, avian, and a type belonging to cold blooded animals. 2. The disease is characterized by the formation of tubercles as a reaction of the animal tissues. The tubercle bacilli may enter the body thru the act of breathing, eating and thru wounds ; and leave the body thru the mouth or nose, or in the milk, feces, urine, genital discharges, or discharges from wounds. The tuberculous animal rarely shows physical signs until the last stages of the disease. 3. In determining the presence or absence of tubercle bacilli in material supposed to be infectious, the method of animal inoculation is far superior to that of microscopical examination. This is true be- cause of the difficulty of distinguishing the tubercle bacilli from other acid-fast bacilli, and the difficulty of finding them when present in small numbers. Pages 317-327 4. Tubercle bacilli are widely distributed in market butter. Table i shows a total of 1233 samples of butter tested of which 163, or 13.2 percent, were positive; Table 2 shows 209 samples of oleomar- garine tested, with 9, or 4.3 percent, positive. 5. The percentage of samples containing tubercle bacilli shows no decrease when the work of the last few years is compared with the first work done twenty years ago. 6. Tubercle bacilli remain virulent in butter as ordinarily salted, for more than five months. 7. Two of six samples of butter tested at this laboratory contained tubercle bacilli virulent to guinea pigs. Pages 328-333 8. There is recorded in the literature the testing of 7845 samples of market milk for tubercle bacilli of which 537, or 6.8 percent, con- tained this organism. In the United States three authors found tu- bercle bacilli present in 78, or 17.5 percent, in a total of 447 samples of milk tested, while in Continental Europe four authors found this organism present in 205, or 4.8 percent, of 4229 samples tested. , 9. In 37 samples of market milk tested at this Station no tubercle bacilli were found. 10. Tubercle bacilli have been found in the milk from tuberculous cows with sound udders by many reliable investigators. These bac- teria probably gained access to the milk from other sources (manure, urine, dust) than, the interior of the udder, in most cases. 11. Tubercle bacilli are rarely excreted in the milk of tuberculous cows with sound udders, especially if the infection is localized; when the infection is generalized or when tubercle bacilli are free in the blood, these organisms may be excreted along with the milk. 12. Forty-seven samples of milk from ten tuberculous cows, tested at this laboratory, gave wholly negative results. None of the ten cows was found to be extensively tuberculous post mortem. Pages 334-362 13. Guinea pig inoculation tests are superior to microscopic tests in detecting the presence of tubercle bacilli in cow manure. By inocu- lation tests, tubercle bacilli in a suspension of .fresh cow manure can be detected when present in an amount as small as one hundred bil- lionth of a gram in the dose injected. 14. Ordinarily only a small percentage of tuberculous cattle are excreting virulent tubercle bacilli in their feces at any given time. 15. At this Station ninety-seven samples of feces from sixty-two cattle have been tested for tubercle bacilli by inoculation of guinea pigs. These organisms have been found four times in samples from three of the cows. Pages 362-390 16. At present the results of different investigators in detecting tubercle bacilli in the circulating blood are conflicting. This diffcr- ence may be due to the amount of sample injected into the test ani- mal ; the larger the amount of sample, the larger the percentage of positive findings.^ 17. The contention, of Rosenberger that tubercle bacilli are fre- quently found' in the blood by microscopic test has not been confirmed by other workers. 4. 18. As a diagnostic procedure the examination of the blood for tubercle bacilli is of little value. 19. Twelve samples of blood from eight cattle were tested for tu- bercle bacilli, both by the Rosenberger method and by guinea pig in- oculations. No test showed the presence of these organisms. Pages 390-402 20. There are about 200,000 deaths annually in the United States from tuberculosis, of which it is estimated (Park and Krumwiede) that 92 percent are caused by the human type of tubercle bacilli ; and 8 percent, or 16,000 cases, are caused by the bovine type. 21. In the cases caused by the bovine type of tubercle bacilli the disease is probably contracted very largely thru the food and is found almost exclusively among children. Pages 402-406 22. Though our knowledge concerning distribution of cattle tu- berculosis is yet incomplete, it is known to be extensive thruout the world ; ranging from 10 to 48 percent where most testing has been done. 23. The two things most favoring the distribution of this disease are the extensive trade in (tuberculous) cattle, and the difficulty of recognizing the disease by physical signs until it is far advanced. Pages 406-412 24. The recognition of tuberculosis in cattle depends almost en- tirely upon the tuberculin test. The positive test is accurate in about 98 percent of the cases as shown by slaughter. The negative test is not so reliable. 25. The Bang method of handling a tuberculous herd is recom- mended to owners of large herds. The modified Bang method by means of a "Stock Owners' Association," is recommended to owners of smaller herds. 26. The responsibility of ridding farm animals of tuberculosis must finally be placed upon the stock owner. The State may well formulate rules and regulations, provide for free tuberculin testing, and for popular and special education upon the subject of animal tu- berculosis. Pages 412-431 Note. The experimental and bibliographic work shown in this bulletin was done by Mr. Briscoe of the Department of Botany, which includes bacteriology. The manuscript in its original form was also prepared by him. Dr. MacNeal, Assistant Professor of Botany (bac- teriology) and a member of the Station Staff, directed the work and made a careful revision of the entire matter for publication. T. J. BURRILL, Head of Department. TUBERCULOSIS OF FARM ANIMALS BY CHAS. F. BRISCOE, INSTRUCTOR IN BOTANY, AND W. J. ASSISTANT CHIEF IN BACTERIOLOGY INTRODUCTION CAUSE Tuberculosis is a contagious disease widely prev- alent thruout the world. The infectious agent is a bacterium, Bacillus tuberculosis. This germ is conveyed from animal to animal, by direct contact, by means of food or drink, by inhalation, or it may gain entrance thru a wound. Bacillus tuberculosis is a microscopic plant. It has the shape of a very small, thin rod. It is a definite species. No tubercle bacillus can originate except it comes from a parent tubercle bacillus ; any more than a white oak tree can come from any other than a parent white oak tree. No animal, no person can take this disease unless this specific germ gains entrance to the body, and by its multipli- cation in the body produces its characteristic effects. It should be emphasized at the outset that Bacillus tuberculosis is not found everywhere. It cannot live long outside the animal body under ordinary conditions. It is easily killed by light, and is killed, when fully exposed, by drying, tho it may live for weeks or even months in apparently dry material. It is not found except in places where animals or human beings with the disease live or have lately lived. Tuberculosis is a purely contagious disease and in domestic animals it is contracted very largely thru close as- sociation with a diseased animal. The disease is not inherited, tho in some rare instances the tubercle bacillus may gain en- trance to the young before birth. A weak bodily resistance to the tubercle bacillus may be inherited. Tuberculosis was known to the Ancient Greeks as early as the fifth century before Christ/ Hip- ACCOUNT f . ' -B r* \ r 4.1. pocrates gave (460 B. C.) a description of this disease, tho he surely did not understand the tubercle as we know it today. But little more was written concerning the disease until the i/th century of our era; then came the period of anatom- ical investigation. About the middle of the iQth century what we may call the period of experimental investigation began, and among the investigations, that of Villemin, published in December, 1865, proving the infectiousness of the disease, is of fundamental importance. Villemin inoculated rabbits subcutaneously behind the ear with pus of yellow and gray tubercles from consumptive 317 318 BULLETIN No. 149 [February, men and also from tuberculous cattle, and produced in this way tubercles in the lungs and livers of the rabbits. Since then Vil- lemin's experiments have been repeated and confirmed by thous- ands of workers. In 1882, Robert Koch discovered the tubercle bacillus, stained, described and successfully cultivated it, and pro- duced tuberculosis by inoculating the germ, free from all other living material, into animals. In 1890, Koch prepared and de- scribed the substance known as tuberculin which has now become so useful as a diagnostic agent. Very early in the study of the tubercle bacilli, certain differ- ences were discovered between those present in mammals, those of birds, and those infecting fish, and these three distinct types have been recognized for many years. In 1898, Theobald Smith first recognized type differences between the tubercle bacilli pres- ent in human pulmonary tuberculosis and those present in tuber- culosis (pearl disease) of cattle. The former he called the human type, the latter, the bovine type. The organisms belonging to the human, bovine, and 'avian types are by far the most important economically and of these the human and the bovine, especially in America, concern us most. These latter were distinguished by Smith, 12 by differences in their morphology, in growth in artificial culture, in the reaction pro- duced in glycerin broth cultures, and in their virulence. The human type is less virulent for rabbits than the bovine type. The human type produces acid when grown for some weeks in five percent glycerin broth; the bovine type produces an alkalin reac- tion. The human type grows more readily and more abundantly than the bovine type on culture media in general. The human bacillus is longer, more slender, and curved; the bovine bacillus is shorter, plumper, and usually straighten VIRULENCE The possible practical importance of this differ- OF BOVINE ence in type is at once apparent. Koch perceived AND HUMAN this and followed up Smith's work with an ex- TYPES tensive investigation of the transmissibility of the human disease to cattle, and arrived at the start- ling conclusion which he first announced at the Tuberculosis Con- gress at London, 1901. He said that cattle can not be infected with the bacillus of the human type and that man need have little fear of infection with the bovine organism from animal products. This statement of Koch aroused considerable opposition at the time, and has stimulated the production, during the last nine years, of a great deal of work upon the subject. The opinion of the great majority of scientists, at the Sixth International Congress on Tuberculosis at Washington in 1908, was strongly against TUBERCULOSIS OF FARM ANIMALS 319 Koch's statement in regard to the danger of the bovine bacillus for man. There are on record numerous observations of the infection of cattle with tubercle bacilli derived from man. These have been reviewed by Ravenel 10 who concludes that there is no longer any doubt that such transmission can take place. Schweinitz and Dor- set" come to the same conclusion. That man can be infected by the bovine tubercle bacilli does not, from the nature of the question, readily admit of direct ex- perimental proof. But indirect evidence has been brought forward in abundance. Instances in which children and men have been infected by tubercle bacilli of the bovine type have been reported by a large number of investigators. Many of these instances are reviewed by Ravenel 10 and more recent reports of similar cases have been made by Hess," Lewis,' Park, 8 Duval," and Park and Krumwiede. 9 The case reported by Troje u may be given as an example in which the evidence was very clear. A young butcher in good health, and with no hereditary taint, wounded his right fore- arm while slaughtering a tuberculous cow. The wound at first healed under medical treatment but a month later proved to be tuberculous. Very many cases of tuberculosis in children have been reported where the infection had evidently come thru the in- gestion of milk and other dairy products. The isolation of the bovine type of tubercle bacilli from tubercles in children, especially from the lymphatic glands of the neck and intestines, is not at all uncommon. Primary intestinal tuberculosis and tuberculosis of the lymphatics in man are frequently caused by tubercle bacilli of the bovine type. These kinds of tuberculosis are thot by Koch" to be relatively very infrequent. On the other hand, Fibiger and Jensen 4 think that of all tuberculous cases in children 16 percent are of this type. Park and Krumwiede* found in an examination of a total of 435 consecutive cases of tuberculosis (that is all cases were examined regardless of type) 32, or 7.4 percent, caused by tubercle bacilli of the bovine type. Of adults over 16 years of age only one case of 297 examined was caused by bacilli of the bovine type, while in children between the ages of 5 and 1 6 years, 9 out of 54 or i6^i percent, and in children under 5 years of age, 22 out of a total of 84 cases examined, or 26.2 percent, were found to be caused by the bovine type of tubercle bacillus. Theobald Smith 13 has estimated that about one percent of tuber- culosis in man is of the bovine type. Charles F. Dawson says, "A summation of the views of various authors shows a prepon- derance of opinion against Koch's theory", in which he doubted that human and bovine tuberculosis are intertransmissible. 320 BULLETIN No. 149 [February, From the mass of apparently conflicting evidence in regard to the intertransmissibility of tuberculosis between the human and the bovine species, it is impossible at present to arrive at the whole truth. The following facts seem well established : 1. Healthy bovine animals do not naturally become infected with tubercle bacilli of the human type. 2. The human species can be infected with tubercle bacilli of the bovine type, and such infections appear to be rather common in children. 3. Bovine animals have been infected with tubercle bacilli derived from cases of human tuberculosis. It seems probable, in some cases at least, that these were instances of human infection with the bovine type of bacillus rather than instances of infection of cattle with the human type of the bacillus. 4. Evidence is still too slight for deciding whether the bovine type of tubercle bacillus may be, in nature, transformed into the human type. SUMMARIZED I- Tuberculosis is a disease due to a specific STATEMENT organism. It is contagious and transmitted from one animal to another. 2. It was known to the early Greeks, 460 B. C., but was little understood until the middle of this last century. 3. Several types of tubercle bacilli have been recognized; the human, bovine, avian, and a type belonging to the cold blooded animals. 4. The types are more or less intertransmissible from one species of animal to another. PATHOLOGY The most characteristic tissue change in the dis- TUBERCLE ease known as tuberculosis is the formation of tubercles. These are more especially found in the lungs, lymph glands, spleen and liver, but may be found in any organ and in almost all tissues of the body. They are not uncommon in the bones but rare in the muscles and nervous tissues. The tubercles vary in size from a millet seed, and less, to more than an inch in diameter. They are pearly white to a yellowish color. They are often hard, fibrous or calcareous but sometimes soft and full of pus. ENTRANCE ^ le tubercle bacilli may gain entrance to the body TO THE thru the lungs by particles breathed in, or they BODY may be absorbed directly thru the lining mem- branes of the alimentary tract alone with the TUBERCULOSIS OF FARM ANIMALS 321 food and may be carried to the lymph glands or lungs by the blood and lymph currents, and wherever arrested, multiply and irritate the tissue at that point. By the resulting increase of tis- sue cells a knot or tubercle is formed. From this primary center, which may be single or multiple, the tubercle bacilli may be car- ried by means of the lymph or blood vessels to other parts of the body. When many bacilli are carried thus by means of the blood vessels to all parts of the body, there is produced generalized, miliary tuberculosis. OPEN AND or g amsms ma y be restricted in their multi- c LOS ED plication and invasion, and a tough wall of tissue TUBERCULOSIS ma y f Qrm around the tubercle, shutting in the germs so they can not easily be carried to other parts of the body (Fig. i). This is called closed tuberculosis. In other cases, the tubercles are open and cavities are formed as is often the case in the lungs (Fig. 2). Since infectious pus is continuously poured out, this condition is known as open tubercu- losis. ESCAPE OF TU- This infectious pus may be coughed up and BERCLE BACILLI smeared over the manger, fodder, or drinking FROM THE BODY trough, or it may be swallowed and passed out with the dung. Tubercle bacilli may reach the intestinal contents also from tubercles of the intestines or they may be passed out in the bile secretions directly into the intestine. In such cases, the dung contains living tubercle bacilli, and thus the barn lot and pasture may become dangerous to other cattle. In a similar way the urine or secretions from the genital organs may carry out the germs. The milk from a tuberculosis udder is a very dangerous source of infection. APPEARANCE ^ n ^ ie ear ^y stages of the disease, there are no OF TUBER- outward manifestations; no fever, nor signs of CULOUS pain; the animal may be fat and sleek, eat heart- ANIMALS ily, give plenty of milk and apparently breed as well as an animal free from tuberculosis. If the udder or neck glands are affected, the change in these may be detected earlier, but not often even then until the disease is well advanced. In these cases knots can. be felt in the enlarged glands which are hard and painless. It is one of the characteristics of tuberculosis that squeezing upon the tuberculous nodule gives no pain, or much less pain than is usual for such knots found in other diseases. If the larynx, bronchi, or lungs are affected a cough is generally noticeable, especially after exercise such as running 322 BULLETIN No. 149 [February^ FIG. 1. SECTION THROUGH A PIECEOF LUNG FROM A cow, SHOWING CLOSED* TUBERCULOSIS. ENCAPSULATED TUBERCLES AT A. BRONCHIAL TUBE AT B. NATURAL SIZE. TUBERCULOSIS OF FARM ANIMALS 323 FIG. 2. SECTION THROUGH HUMAN LUNG SHOWING EXTENSIVE OPEN TUBERCULOSIS WITH CAVITIES.] {REDUCED. (AFTER G-. CORNET, DIE TUBERKULOSE, TAFEI, IV.) 324 BULLETIN No. 149 [February, around the lot for a short time. By placing the ear over the lungs an unusual sound may sometimes be heard a crackling, bubbling, or rasping sound. The breathing is more irregular. Tuberculosis of the uterus or ovaries may cause abortion or ster- ility, while an involvement of the alimentary tract often causes diarrhea. As the disease advances, even tho she may continue to eat a normal amount, the cow loses flesh. The hair becomes rough and loses its luster. The bones become more prominent, the eyes sunken, the ears droop, the head and neck are carried lower, there is a decrease in milk, and the milk becomes blue and thin. In the last stage, the animal becomes stiff; in many cases she can not lie down because of the difficulty in breathing. At this stage, a high temperature is often present; progressive emaciation and loss of strength continue until death ensues. Sometimes the disease in the earlier stages may be arrested, the tubercle encapsulated, and the cow may live for some time, but complete recovery is probably rare if it really does ever occur in the bovine species. i. The most characteristic thing about tubercu- SUMMARIZED , . . ,. , , , , P STATEMENT losis is the formation of tubercles as a reaction on the part of the animal tissues. 2. The tubercle bacilli may enter the body thru breathing, may be taken in with food and absorbed, and thru wounds in the skin. 3. Tuberculosis is called open when the lesions are discharg- ing tubercle bacilli, and closed when a confining membrane retains the organisms in the tubercle. 4. The tubercle bacilli leave the body thru the mouth or nose, or in the milk, feces, urine, or genital discharges. 5. Tuberculous animals in the early and moderately advanced stages appear well and active, and become emaciated only in the final stages of the disease. GENERAL METHODS EMPLOYED IX DETERMINING THE PRESENCE OF THE BACILLUS ' MICROSCOPICAL, EXAMINATION The microscopical method of examining milk, butter, feces, and urine for Bacillus tuberculosis, tho used to a considerable ex- tent by early investigators, is not at present considered very re- liable. PREPARING ^ examme tne material with the microscope a AND STAINING tm ' n smear ls made on a glass cover slip and fixed THE SMEAR by passing it thru the flame three times and then staining with a strong red dye, hot carbol fuchsin. TUBERCULOSIS OF FARM ANIMALS 325 This is washed in cold water and decolorized with ten percent nitric acid until only a slight pink color remains. The preparation is then dipped for ten seconds into eighty percent alcohol, and again washed in water. It is now counterstained with methy- lene blue ; dried and mounted in balsam. The tubercle bacillus has in its outer coating a waxy layer. This layer of wax prevents ordinary applications of stain from entering it rapidly, but when heat is applied the strong carbol fuchsin penetrates the organism readily. In the cold, ten percent, or even thirty percent nitric acid will not decolorize the tubercle bacillus, but will decolorize all other or- ganisms and tissues on the slide. After the acid is all washed out with water, the other material on the slide can be stained with the methylene blue solution. So in this preparation the tubercle bacil- lus appears as a thin, somewhat curved, rod, stained deep red, which contrasts sharply with the other material on the slide which takes the blue stain. RELIABILITY Other acid-fast organisms, however, take the same stain and can not be distinguished from the true tubercle bacillus by their appearance. Some of these may be eliminated by decolorizing with alcohol. Bang 1 recommends mi- croscopical examinations only for rapid work, as in the case of udder tuberculosis where hundreds of organisms may be found on a slide. It is not at all reliable where only a few are present in the material to be examined. By this method of examination, the true tubercle bacillus can not be accurately distinguished from other acid-fast organisms, such as the butter bacillus of Rabino- witsch or the Timothy grass bacillus of Moeller, and besides, one often fails to detect it when present in small numbers. A more reliable test for the presence of the bacillus is animal inoculation. From two to ten cubic centimeters of material may be examined by inoculation into a guinea pig; but if this material were spread out thin enough for microscopical examination it would make a surface so large as to take some days to examine it with a mi- croscope. ANIMAL INOCULATION STOCK ^ or our inoculations, guinea pigs alone have been ANIMALS used. They were bought with care, using only young, unused animals, weighing about 200 grams when purchased. When inoculated they weighed from 250 to 300 grams. In stock they were kept in pens, about 10 to 25 in a pen. Watering and feeding pans were scalded or steamed daily. The animals were fed clover hay, carrots, sugar-beets, and corn meal. 326 BULLETIN No. 149 [February, Those inoculated from each sample were, after inoculation, kept together in a closed galvanized INOCULATED . f ANIMALS wire ca e - They were bedded with sawdust or shavings which was changed frequently to keep the cage In good sanitary condition. Watering dishes were cleaned and steamed daily. The animals were fed twice and watered once a day. They were weighed weekly and notes on their condition made as to loss of weight, size of the lymphatic glands in the groin, axilla, any lesion at the point of inoculation, or any condition vary- ing from the normal. At the end of six (or more) weeks each guinea TUBERCULIN P^ was m j ecte ^ subcutaneously with two cubic centimeters of tuberculin, following the method of Anderson, who claims that this will detect not only the presence of tuberculosis but will distinguish between true and pseudo-tuber- culosis. In our work, we have found no pseudo-tuberculous or- ganisms. But the tuberculin has never failed either to kill or make very sick any guinea pig that had tuberculosis. It had none or but slight effect upon those free from the disease. We have so far injected 142 guinea pigs with the tuberculin ; thirteen of these had tuberculosis and all but one of these thirteen died in less than twenty-four hours after the injection. This one was made very sick and was killed by chloroform twenty-four hours after the injection and found to contain tuberculous lesions. The 129 having no tuberculosis as shown by autopsy were not visibly affected by the tuberculin. In the pigs free from tuberculosis autopsy often revealed a slight reddening at the point of injection of tuberculin, and in some cases a considerable swelling, but these guinea pigs apparently did not notice any inconvenience from the two cubic centimeters of tuberculin injected. They ate and be- haved otherwise as usual. POST MORTEM Twenty-four hours after injecting with tubercu- EXAM I NATION n ' n P ost mortem examination was performed on all the guinea pigs. Those not already dead from the tuberculin were chloroformed and all were carefully examined for tuberculosis. If any suspicious tissues were discovered, stained preparations were made and examined microscopically, tubes of egg medium were inoculated, tissues were fixed for sectioning, and two other guinea pigs were inoculated with an emulsion of the suspected tissue. If we obtained from all these tests results char- acteristic of the disease, we could be fairly sure that the organism in question was Bacillus tuberculosis. Such procedure is, however, necessary in order to distinguish from pseudo-tuberculous and other acid-fast organisms. TUBERCULOSIS OF FARM ANIMALS 327 REFERENCES 1. Bang, B., Measures against tuberculosis in Denmark. Sixth Internat. Cong, on Tuberculosis 4, part 2: 863. 1908. 2. Cornet, G., Historical account is largely from Cornet, Die Tu- berculose, 1: 11-19. 1902. 3. Duval, Charles W., Atypical tubercle bacilli in cervical aden- itis. Sixth Internat. Cong, on Tuberculosis 4, part 2 : 704-729. 1908. 4. Fibiger and Jensen, cited by Koch, R., Sixth Internat. Cong., on Tuberculosis 4, part 2 : 746. 1908. 5. Hess, A. F., The stability of type of the tubercle bacillus. Jour. Am. Med. Assoc. 52: 1011-1016. 1909. 6. Koch, Robert, The conference on the relations of tuberculosis of animals and of man. Continuation of the joint session of Sec- tions I and VII (in camera}. Sixth Internat. Cong, on Tuberculosis 4, part 2: 746. 1908. 7. Lewis, Paul A.j Tuberculous cervical adenitis. Sixth Internat. Cong, on Tuberculosis 4, part 2 : 692-696. 1908. 8. Park, William H., The types of tubercle bacilli present in eighty- four cases of human tuberculosis in New York City. Sixth Internat. Cong, on Tuberculosis 4, part 2 : 697-703. 1908. 9. Park, William H. and Krumwiede, C., The relative importance of the bovine and human types of tubercle bacilli in the different forms of human tuberculosis. Jour, of Med. Research. 23: 205-368. 1910. 10. Ravenel, M. P., Bovine tuberculosis a- factor in the causation of human tuberculosis. Report of the Tuberculosis Commission of the State of Maryland, 1902-1904, part 4: 32-42. Baltimore, 1904. n. Schweinitz and Dorset, The comparative virulence of human and bovine tubercle bacilli for some large animals. U. S. Dept. Agr. Bureau of Animal Industry, Bull. 52, part 2. 1905. 12. Smith, Th., A comparative study of bovine tubercle bacilli and human tubercle bacilli from sputum. Jour. Exp. Med. 3: 451. 1898. 13. Smith, Th., The channels of infection of tuberculosis. (Boston Med. and Surg. Jour.) Cited by Hess. Jour. Am. Med. Assoc. 52: ion. 1909. 14. Troje, Cited by Ravenel, above. EXAMINATION OF BUTTER FOR TUBERCLE BACILLI LITERATURE A detailed review of the literature on the exam- EXPLANATION . . r . ,. , .... , , f\c TADI co mation of butter for tubercle bacilli would be too \Jr IADL.C.O . -,.,,,. extensive for the purposes of this bulletin, so a tabulation of it is given instead. See Table i. The references are cited by numbers which refer to the authors arranged alpha- 328 BULLETIN No. 149 [February, TABLE 1. BACILLUS TUBERCULOSIS IN MARKET BUTTER 6 g Author 1> "rt Q 1890 1894 1895 1896 1897 1897 1897 1897 1897 1897 1897 1899 1899 1899 1899 1899 1899 1899 1899 1899 1899 1899 1899 1899 1899 1899 1900 1900 1900 1900 1901 1901 1901 1901 1902 1902 1904 1906 1908 Place Samples examined Samples positive Percent positive Remark 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 Brusaferro 7 Turin 9 20 13 42 14 17 ? 30 50 102 10 15 ? is 19 10 17 27 3 100 32 10 43 58 20 5 39 8 28 23 12 30 43 52 17 16 40 94 150 1 2 8 14 8 33 3 2 ? is 4 4 2 1 12 3 1 1 2 6 3 12 8 18 11.1 10.0 61.0 00.0 100.0 47.0 0.0 0.0 0.0 32.3 30 13.3 87.2 100.0 0.0 40.0 23.5 7.4 33.3 12.0 9.4 0.0 0.0 0.0 0.0 0.0 0.0 12.5 0.0 4.3 16.7 0.0 0.0 11.1 17.6 0.0 30.0 8.5 12.0 Microscopic method 16 tested, 2 lost Reported by Markl First series Second series Third series Fourth series Pseudo-tubercu- losis 5 percent Pseudo-tubercu- losis 8 percent Pseudo-tubercu- losis 4 percent 12 samples, 4 lost 39 samples, 11 lost Two are doubtful Butter from 88 dairies Roth 80 Zurich Berlin Obermuller 24 Schuchardt 33 Marburg Obermiiller 28 . Berlin Groning 11 Hamburg .... Wien Berlin Himesch 1 5 Rabinowitsch 28 Rabinowitsch* 8 . .. Petri 26 Philadelphia. Berlin Herman and Morganroth 1 8 Berlin Rabinowitsch 27 Rabinowitsch* 7 . . . . Berlin Berlin . .... Rabinowitsch 57 Berlin ... Rabinowitsch* 7 Berlin Obermuller 2 2 Korn 18 Berlin Freiburg. . . Konigsberg. . Konigsberg. . Milan Jaerer 1 7 . . CoTri 8 Weissenfeld 37 Bonn Grassberger 10 Wien Herbert 1 3 Tubingen. . . . Wiirttemberg Berlin ... Herbert 18 Herbert 18 Herbert 13 Abenhausen 1 Miinchen Marburg .... Helsingfors. . Marburg Hellstrom 18 Bonhoff * .... .... Pawlowsky 25 Kiew Tobler 36 Zurich Dorpat lyorenz 19 Markl 20 Wien Herr und Beninde 14 . Aujeszky 3 Budapest . . . Christiana . . . Rosen. . Stuttgart Leipsic Thu 36 Teichert 34 Reitz 29 Fjber 9 . . Totals . 1233 163 13.2 TUBERCULOSIS OF FARM ANIMALS 329 betically in the bibliography at the close of this section. In the table, the number of samples examined, the number in which tu- bercle bacilli were found, and the percent of positive findings are indicated. The examinations \vere always performed by the method of animal inoculation, except where otherwise noted in the column headed, "Remark". By the authors cited in this table (Table i), 1233 samples of market butter were examined, of which 163 were found to contain tubercle bacilli, or 13.2 percent of the total. GENERAL CON- There are great differences between the findings of SIDERATIONS different investigators, which may be due to the different sources of their material. In some cases different methods of work may have had a part also in determin- ing the results obtained. There is no evident progressive change in the percentage of infected samples during the period covered by these examinations. In this connection it may be of interest to note that tubercle bacilli have also been found in oleomargarine. In Table 2 are given the results obtained by seven investigators in the examina- tion of 209 samples, of which 9, or 4.3 percent, were found to contain tubercle bacilli. TABLE 2 BACILLUS TUBERCULOSIS IN OLEOMARGARINE Author V "5 Place Samples examined Samples positive Percent positive Remark Morganroth" 1899 1900 1900 1901 1901 1902 1908 Berlin Berlin 10 15 13 3 3 15 150 8 1 80.0 0.0 8.0 0.0 0.0 0.0 0.0 Nearly 8 percent 7 samples, 4 lost From 4 factories Annette 2 Annette 2 Bonhoff 6 Liverpool. .. Marburg .... Wien Markl 20 Thu 35 Christiana. . . L/eipsic Eber 9 Totals 209 9 4.3 There has been very little work in the United States on the examination of market butter for tubercle bacilli. Schroeder and Cotton 32 have made a study of the length of time that Bacillus tuberculosis will live in butter as ordinarily salted for the market, i. e. one ounce of salt to the pound of butter. They found by means of guinea pig inoculations that the tubercle bacilli in butter retained their virulence for 133 days at least. Schroeder 31 in a 330 BULLETIN No. 149 report upon further studies on the viability of tubercle bacillus in salted butter shows that it remains virulent to guinea pigs for 160 days or longer. He also reports the feeding of tuberculous butter to four hogs. The butter was three months old. Each hog re- ceived one ounce daily mixed with its feed. Three of the four hogs became tuberculous. The method usually employed in the examination METHOD OF , , J . ,f -,f. . , .. PROCEDURE butter for tubercle bacilli is as follows. Butter is bought in the market, taken to the laboratory and placed in a sterile dish. The piece is separated into two parts with a sterile knife, and from the central portion of the freshly cut surface a sufficient amount to use in the test is removed with a sterile spatula and is placed in a sterile, glass-stoppered bottle. This bottle is placed in a 38 incubator until the butter is melted, when it is injected subcutaneously into guinea pigs, a dose of from i to 5 cc. being given. In order to detect a small number of tu- bercle bacilli, the melted butter should be centrifuged and the sedi- ment injected. In this way the bacterial content of a larger amount of material can be placed under the skin of the test ani- mal. The sediment may also be examined microscopically. EXPERIMENTAL, WORK ON BUTTER Only eight samples of butter have been examined SAMPLES . ' ~ . EXAMINED m our laboratory. Ihe samples were all obtained from the University Experiment Station Cream- ery. The milk from which the butter was made came from the tuberculous herd of the University Experiment Station and from farmers' herds in the surrounding neighborhood. The samples of butter were furnished us directly from the moulding board and were put into 3-ounce sterile, glass-stoppered, wide-mouthed bot- tles. The butter was melted and three cubic centimeters inocu- lated intraperitoneally into each of three guinea pigs. The results are shown in Table 3. RESULTS samples two were lost by the guinea pigs dying one day after inoculation. Of the six samples completing the test, two, or 33^ percent, contained tu- bercle bacilli. It should be said that a number of samples of milk from the University herd have since been examined and none of these has proved to be tuberculous; it therefore seems probable that the tubercle bacilli came from some one of the farmers' herds. 79/7] TUBERCULOSIS OF FARM ANIMALS 331 TABLE 3 TUBERCLE BACILLI IN BUTTER FROM THE UNIVERSITY DAIRY 6 o fc d ^H.2 *H ^ a -2 2 o a L2 01 a .2 4> DC *s 5 "rt "S o re s ci U ri ID 0) ^H J 3 re o " " >" Autopsy ca C 's 3 -M 3 O o re .2 P 2 S ^ rt O ." rt o a CO '3 .r< H .5 Jg ^ 'i O 1 3 7-3- '09 Killed 67 Normal 4 7-3- '09 Killed 67 Normal 5 7-3- '09 Killed 67 Generalized tuberculosis in the lym- phatics, lungs, liver, spleen, etc. -L. -[_ -| 1_ 2 6 7-6- '09 Killed 65 Tuberculous only in the right super- ficial inguinal lymph gland _|_ -[_ -| L. 7 7-6- '09 Died 10 Acute infection * # 8 7-6- '09 Killed 65 Generalized tuberculosis in the liver, spleen, mesentery, diaphragm, lymph glands, etc. -|- -j- -| |- 3 24 7-28- ; 09 Died 1 Acute infection 25 7-28- '09 Died 1 Acute infection 26 7-28- '09 Died 1 Acute infection 4 27 7-31- '09 Died 1 Acute infection 28 7-31- '09 Died 1 Acute infection 29 7-31- '09 Died 1 Acute infection 5 33 8- 8- '09 Killed 59 Normal 34 8- 8- '09 Killed 59 Normal 35 8- 8- '09 Killed 59 Normal 6 63 8-14- '09 Killed 55 Normal 64 8-14-'09 Killed 55 Normal 65 8-14- '09 Killed 55 Normal 7 90 8-21- '09 Killed 65 Normal 91 8-21-'09 Died 1 Acute infection 92 8-21- '09 Killed 65 Normal 8 108 8-28- '09 Died 9 Acute infection 109 8-28- '09 Killed 62 I^arge tumor at the point of inocula- Nor- tion. No tubercles. Organs normal. t t mal 110 8-28- '09 Killed 62 Normal *Micrococci. fSeveral kinds of bacteria present. Summary: Bight samples, two lost; of the other six, two contained tubercle bacilli, 33% percent CONCLUSIONS 1. Bacillus tuberculosis is widely distributed in market butter. 2. The examinations in the last few years show no decrease in the percentage of samples containing tubercle bacilli as compared with the first work twenty years ago. 3. Bacillus tuberculosis remains alive in butter as ordinarily salted for a much longer time than butter is usually kept in storage. 4. Butter containing tubercle bacilli, when fed to hogs, may produce generalized tuberculosis. 5. Of the six samples of butter tested by us, two, or 33^3 per- cent, contained tubercle bacilli virulent to guinea pigs. 332 BULLETIN Xo. 149 [February, REFERENCES 1. Abenhausen, Einige Untersuchungen iiber das Vorkommen von Tuberkelbacillen in der Marburger Butter und Margarine. (Disserta- tion, Marburg, 1900.) Cited by Adolph Reitz. 2. Annette, Tubercle bacilli in Margarine. Lancet, 1900 (I) : 161. 3. Aujeszky, Ueber das Vorkommen der Tuberkelbacillen in der Budapester Marktbutter. Centralb. fur Bakt. Abt. I, Orig. 31: 132. 1902. 4. Ascher, Untersuchungen von Butter und Milch auf Tuberkel- bacillen. Ztschr. f. Hyg. 32: 329-344. 1899. 5. Bang, B., Measures against tuberculosis in Denmark. Sixth Internat. Cong, on Tuberculosis 4, part 2 : 863. 1908. 6. Bonhoff, H., Ueber das Vorkommen von Tuberkelbacillen in der Marburger Butter und Margarine. Hyg. Rundschau 10: 913-916. 1900. 7. Brusaferro, Alcune esperienze di inoculazione col burro del commercio. Gironale di med. veter. prat. Torino, 1890. Cited by Adolph Reitz. 8. Coggi, Sulla presenza di bacilli tubercolari nel burro di mer- catc di Milano. Centralb. f. Bakt. Abt. I, Ref. 27: 836. 1899. 9. Eber, A., Untersuchungen iiber den Tuberkelbazillengehalt der in Leipzig zum Verkauf kommenden Milch und Molkereiprodukte. Ztschr. f. Fl. und M. Hyg. 18: 309-319. 1908. 10. Grassberger, R., Ueber die nach intraperitonealer Injektion von Marktbutter bei Meerschweinschen entstehenden Veranderungen. Munchener med. Wchnschr. 46: 341-344. 1899. 11. Groning, Cited in Kolle und Wassermann, Handbuch der path- ogenen Mikroorganismen 2: 140. 12. Hellstrom, F. E., Ueber Tuberkelbacillennachweis in Butter und einigp. vergleichende Untersuchungen iiber pathogene Keime in Butter aus pasteurisiertem und nichtpasteurisiertem Rahm. Centralb. f. Bakt. Abt. I, Ref. 28: 542-555. 1900. 13. Herbert, Untersuchungen iiber das Vorkommen von Tuberkel- bacillen in der Marktbutter. Cited by Adolph Reitz. 14. Herr, F. und Beninde, M., Untersuchungen iiber das Vorkom- men von Tuberkelbacillen in der Butter. Ztschr. f. Hyg. 38: 152-181. 1901. 15. Himesch, cited by Adolph Reitz. 1 6. Herman und Morganroth, Ueber Bakterienbefunde in der But- ter. Hyg. Rundschau 8: 1081-1084. 1898. 17. Jager, H., Ueber die Moglichkeit tuberkuloser Infektion des Lymphsystems durch Milch und Milchprodukte. Hyg. Rundschau 9: 801-817. 1899. 18. Korn, Otto, Znr Kenntniss der saurefesten Bakterien. Cen- tralb. f. Bakt., Abt. I, 25: 532-537. 1899. 19. Lorenz, Chemisch-bakteriologische Untersuchungen der in Jur- /pi/] TUBERCULOSIS OF FARM ANIMALS 333 jew (Dorpat) vorkommenden Kuhbutter. (Dissertation, Dorpal 1901.) Cited by Adolph Reitz. 20. Markl, Zur Frage des Vorkommens von Tuberkelbacillen ID der Wiener Marktbutter und Margarine. Centralb. f. Bakt., Abt. I, Ref. 29: 954. 1901. 21. Morganroth, Ueber das Vorkommen von Tuberkelbacillen in der Margarine. Hyg. Rundschau 9: 1121. 1899. 22. Obermiiller, Kuno, Weitere Mitteilungen iiber Tuberkelbacillen- befunde in der Marktbutter. Hyg. Rundschau 9: 57-79. 1899. 23. Obermiiller, Kuno, Ueber Tuberkelbacillenbefunde in der Marktbutter. Hyg. Rundschau 7: 712-714. 1897. 24. Obermiiller, Kuno, Ueber Tuberkelbacillenbefunde in der Marktbutter. Hyg. Rundschau 5: 877. 1895. 25. Pawlowsky, Untersuchungen betreffend die Anwesenheit von Tuberkelbacillen in der Marktmilch und Butter. Cited by Adolph Reitz. 26. Petri, R. J., Zum Nachweis der Tuberkelbacillen in Butter und Milch. Arb. a. d. Ksrlchn. Gesundheitsamte 14: 1-35. 1898. 27. Rabinowitsch, Lydia, Weitere Untersuchungen zur Frage des Vorkommens von Tuberkelbacillen in der Marktbutter. Cited by Adolph Reitz. 28. Rabinowitsch, Lydia, Zur Frage des Vorkommens von Tuber- kelbacillen in der Marktbutter. Ztschr. f. Hyg. 26: 90-111. 1897. 29. Reitz, Adolph, Bakteriologische Untersuchungen mit der Stutt- garter Markt- und Handels-butter. Arch. f. Hyg. 57: 1-28. 1906. Centralb. f. Bakt., Abt. II, 16: 193-212. 1906. 30. Roth, Ueber die mikroscopische Untersuchung der Butter auf Bakterien, insbesondere auf Tuberkelbacillen. Cited by Adolph Reitz. 31. Schroeder, E. C., The occurrence and significance of tubercle bacilli in the feces of tuberculous cattle. Sixth Internat. Cong, on Tuberculosis 4, part 2: 599-606. 1908. 32. Schroeder and Cotton, Tubercle bacilli in butter; their occur- rence, vitality, and significance. U. S. Dept. Agr., Bureau of Animal Industry Circular 127. 33. Schuchardt, Einige Untersuchungen iiber das Vorkommen von Tuberkelbacillen in der Butter. (Dissertation, Marburg, 1896.) Cited by Adolph Reitz. 34. Teichert, Bakteriol.-chemische Studien iiber die Butter in der Provinz Posen mit besonderer Beriicksichtigung der Tuberkelbacillen. Cited by Adolph Reitz. 35. Thu, Hans, Untersuchungen auf Tuberkelbacillen in Milch, Butter und Margarine in Christiana. Centralb. f. Bakt., Abt. I, Ref. 36: 597. 1902. 36. Tobler, Maria, Beitrag zur Frage des Vorkommens von Tu- berkelbacillen und ancleren saurefesten Bacillen in der Marktbutter. Ztschr. f. Hyg. 36: 120-148. 1901. 37. Weissenfeld, Ueber Bakterien in der Butter und einigen an- deren Milchproclukten. Berl. klin. Wchnschr. 36: 1053-1055. 1899. 334 BULLETIN No. 149 [February, EXAMINATION OF MARKET MILK FOR TUBERCLE BACILLI LITERATURE The literature relating to the tubercle bacillus in milk is very extensive and we must content ourselves with giving it, in the main, in tabular form, as in the case of market butter. The cita- tions are made in these tables in the same way as they were in Table i. In these tables we have attempted to separate the literature concerning the presence of tubercle bacilli in market milk from that reporting the work done on milk of tuberculin-reacting cows with sound udders, a distinction not always made by the authors. For this reason, it may be that some items are misplaced in the tables, but in the main they are correct and it is worth while to make the distinction. There are two distinct questions that have called forth this great amount of work. On the one hand, the investigators have sought to determine the percentage of the sam- ples of milk found on the market containing tubercle bacilli, and on the other hand, other investigators have tried to determine whether a cow which evidently reacts to tuberculin, but which has sound udders, will give milk containing tubercle bacilli. Table 4 gives the literature on the bacilli in market milk. TABLE 4 BACILLUS TUBERCULOSIS IN MARKET MILK 1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 1 ( > Author a n Place Samples I examined | Samples 1 positive | Percent positive Remark Martin 47 1884 1890 1890 1890 1893 1893 1894 1894 1895 1895 1896 1897 1897 1897 1897 1897 1897 1898 1898 9 33 (13) 10 80 28 33 19 30 13 6 228 67 25 144 24 (?) (?) 57 3 (1) 3 4 4 1 8 2 12 9 7 4 7 (?) (?) 9 33.3 3.3 23.1 0.0 5.0 14.3 0.0 5.3 0.0 61.5 33.3 5.2 13.4 28 2.8 29 1 9.0 2.5 17.5 12 samples, 3 lost Microscopic Animal inociilation City milk Country 9 samples, 3 lost City milk Country City milk Country milk Corrected fierures Ernst 13 Boston Ernst 13 Boston. .... Munich .... St. Petersburg.. Copenhagen . Copenhagen . Washington.. Newark Berlin.. Gebhardt 1 " 6 Sacharbekow 37 Friis 15 Friis 15 Schroeder 38 Chester 10 Obermiiller 30 Bupre 8 . Halle Liverpool .... Liverpool .... Berlin Hope 18 Hope 18 Rabinowitsch and Kempner 36 Boyce 5 * 6 * Liverpool. . . Liverpool. . . - Genoa Boyce 5&e Massone 28 Kudinow** Dorpat Petri 33 .. Berlin . TUBERCULOSIS OF FARM ANIMALS 335 TABLE 4 Continued 6 Z Author 4) -4-> n) O Place Samples examined Samples 1 positive Percent positive Remark 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 Ott 31 1898 1898 1898 1898 1898 1898 1899 1899 1899 1899 1899 1899 1899 1900 1900 1900 1900 1900 1900 1n te sts 336 BULLETIN No. 149 [February, The determination of tubercle bacilli in milk is not a very dif- ficult problem. In this no attempt is made to find the source from which the bacilli entered the milk ; whether from tuberculous cows or from tuberculous persons handling the milk ; whether from the infected udder or from infected manure or dirt falling into the milk at milking time or otherwise. Most tests have been made by animal inoculation, and this to a large extent insures that the organisms found here have been true tubercle bacilli. As may be seen in Table 4, work was done upon this subject as early as 1884 by Martin. The first extensive work carried out in the United States was by Ernst in 1890. More extensive investigations have been made by the boards of health of Liver- pool and London, the work being reported by Hope, Delepine, Boyce, and others. More especial mention may be made of the results of Mueller and Beatty, who for several years carried out tests for the East Prussian Herdbook Society. They examined bacteriologically the mixed milk of dairy herds as a whole. Their purpose was to discover and remove from the herd, cows with tuberculous udders. They thot that in this way they would be able to prevent most of the infection by tubercle bacilli in dairy products. In this work Mueller has examined 1596 samples of milk and found 97, or 6.2 percent, containing tubercle bacilli. Beatty examined 272 samples and found 27, or 10 percent, con- taining these organisms. Further details of their paper will be found in the next section 1 of this bulletin. Anderson has examined milk from 102 dairies supplying the City of Washington. The herds supplying milk to this city con- tained considerable numbers of tuberculous cows as recent tuber- culin tests (1907) by Mohler had shown 214 reacting animals out of 1147 tested, or 18.6 percent. Other tests by the District Health Department upon 1095 cows in 51 herds in Virginia, Maryland, and the District of Columbia, also supplying milk to the City of Washington, detected 160 reacting animals, or 15.1 percent. An- derson considers these percentages considerably below the actual amount of tuberculosis in the dairy cows supplying that city, be- cause, he says, only the owners of herds supposed to be free from tuberculosis permitted the tests to be made. The author collected his samples in one-pint bottles with untampered stoppers. Fifty cubic centimeters of milk were added to one hundred cubic centi- meters of water (to decrease the specific gravity) and the mixture was centrifuged for one hour at 2000 revolutions per minute. One guinea pig was inoculated subcutaneously with 5 cc. of the sediment. The eight guinea pigs (usually only eight samples were tested per day) were placed together in the same cage, those that remained healthy being controls on their environment. Many animals died of acute infections with other bacteria in the milk but TUBERCULOSIS OF FARM ANIMALS 337 no attempt was made to identify any but the tubercle bacilli. In order to distinguish a tuberculous from a non-tuberculous guinea pig, 2 cc. of crude tuberculin was injected subcutaneously into each animal. This killed or made very sick any guinea pig that had tuberculosis. About 250 guinea pigs received the tuberculin in this way and no animal died that did not have tuberculosis. Two or three animals that had slight lesions did not die, but became very sick. All animals died whose lesions had caseated. The au- thor thinks the procedure a distinct help in eliminating infections with other acid-fast organisms. Samples from 102 dairies were tested. The milk from n, or 10.7 percent, of these 102 dairies contained tubercle bacilli. Of the 272 samples of milk tested, 49, or 1 8 percent, were lost by the test animal dying too early. Of the 223 samples remaining, 15, or 6.72 percent, were tuberculous. Hess studied the milk supply of New York City with three questions in mind : 1. What percentage of market milk in a large city like New York contains tubercle bacilli? 2. What percentage of the tubercle bacilli in milk is of bovine and what is of human origin? 3. What effect has this tuberculous milk upon children? The samples were taken from large 4O-quart cans, and not from bottle-milk. This was done because more than half the milk sold in New York City is not bottled but is the so-called "loose milk." The milk was collected in one-ounce, tight fitting, tin- capped bottles. It was taken indiscriminately from grocery stores, dairies, and large plants in various quarters of the city. Usually the milk was collected by the investigator himself. Each sample was tested by microscopic examination of stained prepara- tions and by the inoculation of two guinea pigs. The milk was centrifuged and the cream dipped off with a sterile copper spoon and placed in a sterile Petri dish. The "skim milk" was then poured off. Microscopic preparations were made from the sedi- ment and the cream. These were then mixed and injected into the animals. Of 112 samples tested, 5 were lost the guinea pigs dying before sufficient time had elapsed to detect tubercle bacilli and 2 were duplicates. The remaining 105 (the author says 107) gave 17 positive tests, or 16.2 percent (author gives 16 percent). There were also examined 8 samples of commercial pasteurized milk, and of these one was found to contain tubercle bacilli. The dealers stated that the milk had been pasteurized at 160 F. for forty seconds. In order to answer the second question, whether the tubercle bacilli found in the milk are of the human or the bovine type, eight cultures were isolated and worked with. In two other cases this 338 BULLETIN No. 149 [February, was attempted, but the cultures failed. Of these, seven conformed to the bovine type and one culture conformed to the human type. The author had collected samples especially from such dealers as had children and gave their children raw milk from the cans in the store. Of the 17 dealers selling milk which contained tubercle bacilli, 10 belonged to this category. These ten milk dealers had eighteen children. Sixteen of these children were tested with tuberculin; four reacted. One of these cases was of especial in- terest. This child consumed daily a pint of the poorest milk ex- amined. She was two years old, poorly nourished, and five months before had developed a cervical adenitis, which was incised at a dispensary. No other member of the family appeared to be tuber- culous. Recently the author was able to collect 44 human cases of undoubted bovine tuberculous infection of the mesenteric glands, 41 of which occurred in children. He thinks that the re- lation of bovine tuberculosis to man may be fairly stated as fol- lows: i. The bovine type of tubercle bacillus, altho less virulent for man than the human variety, is capable of infecting human be- ings. 2. Children are more susceptible than adults to bovine in- fection. EXPERIMENTAL, COLLECTION Champaign and Urbana are practically one city OF SAMPLES an d m st of the milkmen deliver milk in both towns. The samples were bought on the streets as the milkmen were delivering milk to their customers. They were all collected in pint or quart bottles with untampered stop- pers. After bringing the bottle to the laboratory, it was well shaken. TESTING OF Eighty cubic centimeters were placed into four SAMPLES sterile centrifuge tubes and these were centrifuged for thirty minutes at 1800 revolutions per min- ute. Then the milk was drawn from the middle of the tube with a Pasteur bulb pipette and the cream and sediment of the four tubes put into one tube ; this was again centrifuged thirty minutes at 2000 to 3000 revolutions per minute. The middle portion was drawn off from this tube and the remaining cream and sediment of the 80 cc. of milk was injected subcutaneously into three guinea pigs in graded doses, usually four, three, and one and one-half cubic centimeters respectively. In a number of cases one or two of the guinea pigs died too early to test the presence of tubercle bacilli ; but in only two cases did all the guinea pigs from any one sample die too early. Thirty-seven of the thirty-nine samples completed the test. Tubercle bacilli were not found in any sample. The results are recorded in Table 5. 1911] TUBERCULOSIS OF FARM ANIMALS 339 TABLE 5 TUBERCLE BACILLI IN MARKET MILK OF CHAMPAIGN AND URBANA Dairy number Sample number Guinea pig number Date of inoculation Mode of death Days after inoculation Autopsy Tubercu- losis 1 1 329 3-16-'09 Died 2 Acute infection 330 " Killed 63 Normal 331 " Killed 63 Normal 2 387 4-26- '09 Killed 37 Normal 388 < t Died 1 Acute infection 389 n Died 2 Acute infection 3 413 6-10-'09 Died 26 No tubercles 414 " Killed 43 Normal 415 " Died 26 No tubercles 4 443 6-18- '09 Died 18 No tubercles 444 Killed 42 Normal 445 " Killed 42 Normal 2 5 332 3-18- '09 Died 24 No tubercles 333 i 4 Killed 61 Normal 334 (i Died 16 Acute infection 6 398 6-4- '09 Died 3 Acute infection * . 399 a Died 2 Acute infection 400 ^t Died 1 Acute infection 7 410 6-10- '09 Killed 43 Normal "6" 411 " Died 27 No tubercles 412 " Died 27 No tubercles 3 8 341 3-29- '09 Killed 50 Normal 342 " Killed 50 Normal 343 ti Died 11 Acute infection 9 395 6-4-'09 Died 38 No tubercles "o 396 " Killed 49 Normal 397 11 Killed 49 Normal 10 428 6-1 7- '09 Died 26 No tubercles 429 " Killed 43 Normal 430 " Killed 43 Normal 4 11 350 3-30- '09 Died 46 No tubercles 351 '" Died 11 No tubercles 352 " Killed 49 Norm al 6 12 401 6-4- '09 Killed 49 Normal 402 " Died 32 No tubercles 403 Killed 49 Normal 13 533 8-31- '09 Killed 43 Normal 534 " Killed 43 Normal 535 " Died 4 Acute Infection 5 14 344 3-29- '09 Died 14 No tubercles 6 345 " Died 17 No tubercles 346 " Killed 50 Normal 15 437 6-17-'09 Killed 43 Normal 438 Killed 43 Normal 439 Killed 43 Normal 6 16 353 3-30-'09 Killed 49 Normal 354 Killed 49 Normal 355 Died 2 Acute infection 17 404 6-9- '09 Died 5 Acute infection 405 " Killed 44 Normal 406 11 Killed 46 Normal 7 18 416 6-ll-'09 Killed 42 Normal 417 " Died 23 No tubercles 418 " Killed 42 Normal 19 452 6-21-'09 Died 21 No tubercles 453 " Died 21 No tubercles 454 < i : Died 21 No tubercles 340 BULLETIN No. 149 [February, TABLE 5. Continued Dairy number Sample number be 1* P 9 S ^ c Date of inoculation Mode of death Days after inoculation Autopsy Tubercu- losis 8 20 419 6-11- '09 Died 24 No tubercles 420 !< Killed 42 Normal 421 " Killed 42 Normal 21 458 6-22- '09 Killed 44 Normal 459 " Died 2 Acute infection 460 " Killed 44 Normal 9 22 422 6-11- '09 Killed 42 Normal 423 " Killed 42 Normal 424 " Died 27 No tubercles 23 455 6-21- '09 Killed 45 Normal 456 " Killed 45 Normal 457 (i Died 15 Acute infection 10 24 425 6-11- '09 Died 6 Acute infection 426 " Killed 49 Normal '6 427 (i Killed 49 Normal 25 536 8-31-'09 Died 3 Acute infection 537 " Died 3 Acute infection 538 " Killed 43 Normal 11 26 434 6-17- '09 Died 23 No tubercles 435 Died 22 No tubercles 436 Killed 43 Normal 27 467 6-23- '09 Died 1 Acute infection 468 . i Died 13 Acute infection 469 ti Killed 43 Normal "6 12 28 446 6-18- '09 Died 18 No tubercles 447 " Killed 42 Normal . 448 M Killed 42 Normal 29 464 6-22- '09 Killed 44 Normal 465 Killed 44 Normal 466 a Killed 44 Normal 13 30 407 6-9- '09 Died 26 No tubercles 408 " Killed 44 Normal 409 " Died 26 No tubercles 31 470 6-23- '09 Died 13 Acute infection 471 " Died 13 Acute infection 472 4 4 Died 13 Acute infection 14 32 347 3^29- '09 Killed 50 Normal 348 " Killed 50 Normal 349 K Died 5 Acute infection 15 33 431 6-17- '09 Killed 43 Normal 432 " Died 23 No tubercles 433 " Killed 43 Normal 16 34 440 6-1 8- '09 Killed 42 Normal 441 " Killed 42 Normal 442 " Killed 42 Normal 17 35 449 6-21- '09 Died 15 Acute infection 450 " Died 21 Acute infection 451 1 1 Killed 39 Normal 18 36 461 6-22- '09 Died 2 Acute infection 462 " Died 2 Acute infection .... 463 " Killed 44 Normal 19 37 358 4-2- '03 Killed 48 Normal 359 n Died 8 Acute infection 360 " Killed 48 Normal 6 20 38 361 4-2- '09 Died 8 Acute infection ... 362 " Killed 46 Normal 363 " Died 9 Acute infection 21 39 384 4-26- '09 Killed 37 Normal 6 385 Died 1 Acute infection 386 K Died 2 Acute infection TUBERCULOSIS OF FARM ANIMALS 341 CONCLUSIONS 1. According to the literature reviewed, tubercle bacilli are very common in market milk, being found, in the 7845 samples tabulated in Table 4, 537 times, or 6.8 percent. 2. A composite of the three largest testings done in the United States in 1908 and 1909, those of Anderson (Washington), Fields (Louisville), and Hess (New York City), gives a total number of 447 samples, of which the number containing tubercle bacilli is 78 or 17.5 percent. The four largest series of tests in Conti- nental Europe, those of Mueller, Beatty, Smit and Eber, com- prise a total of 4229 samples, and of these only 205 or 4.8 percent contained tubercle bacilli. 3. The testing of thirty-seven samples of market milk from Champaign and Urbana, from twenty-one dairies, revealed no tu- bercle bacilli. REFERENCES 1. Anderson, J. F., The frequency of tubercle bacilli in the market milk of Washington, D. C. U. S. Hyg. Lab. Bull. 41 : 163-192. 1908. Also, Jour. Inf. Dis. 5: 107-115. 1908. 2. Ascher, Untersuchungen von Butter und Milch auf Tuberkel- bacillen. Ztschr. f. Hyg. 32: 329-344. 1899. 3. Beatty, Cited by Miiller. 4. Beck, M., Experimented Beitrage zur Untersuchung iiber die Marktmilch. Hyg. Rundschau 11: 490. 1901. 5. Boyce, published by Annette, Lancet, 1898 (I) : 159. 6. Boyce, Woodhead, Delepine and Hamilton, (British Med. Jour. 1897 (II) : 162.) Cited by Cornet und Meyer in Kolle und Wasser- mann, Hanclbuch der pathogenen Mikroorganismen 2: 140. 7. Brittlebank, J. W., Milk and tuberculosis. (Veterinary Record 19: '164. 1906) Ref. in Exp. Sta. Record 18: 581. 8. Buege, Ueber die Untersuchung der Milch auf Tuberkelbacillen. (Inaug. Dissertation, Halle, 1896.) Centralb. f. Bakt, Abt. I, Ref. 21: 70. 1896. 9. Bujwid, 0., Ergebnisse der Milchuntersuchung in Krakau be- ziiglich des Tuberkelbacillengehaltes. Centralb. f. Bakt., Abt. I, Ref. 30: 213. 1900. 10. Chester, F. D., Test of feeding the milk of a tuberculous cow to guinea pigs. Delaware Agr. Exp. Sta. Ann. Report 1895: p. no. 11. Delepine, S., (Brit. Mecl. Jour. 1898 (II) : 918.) Lancet 1898 (II) : 733-738. 12. Eber, A., Untersuchungen liber den Ttiberkelbacillengehalt der in Leipzig zum Verkauf kommenden Milch und Molkereiprodukte. Ztschr. f. Fleisch- und Milch-Hyg. 18: 309-319. 1908. T3_ Ernst, H. C., How far may a cow be tuberculous before her milk becomes dangerous as an article of food? Hatch Exp. Sta., Mass. Agr. Coll. Bull. 8: 1-38. 1890. 342 BULLETIN No. 149 [February, 14. Field, Cyrus W., Tubercule bacilli in market milk, Louisville, Ky. Louisville Times, July, 1909. 15. Friis, St., Beitrag zur Beleuchtung der Frage iiber die An- steckungsgefahr der Handelsmilch mit Bezug auf die Tuberculose. Cited by Mohler. 1 6. Gebhardt, Franz, Experimented Untersuchungen uber den Einfluss der Verdunnung auf die Wirksamkeit des tuberkulosen Giftes. Virch. Arch. 119: 127-148. 1890. 17. Hess, Alfred H., The incidence of tubercle bacilli in New York City milk. Jour. Am. Med. Assoc. 52: 1011-1016. 1909. 18. Hope, W. E., Report of the Medical Officer of Health, Liver- pool, 1897, on tuberculosis as affecting the milk supply of the city. Cited by Anderson. 19. Jaeger, H., Ueber die Moglichkeit tuberkuloser Infection des Lymphsystems durch Milch und Milchprodukte. Hyg. Rundschau 9: 801-817. 1899. 20. Kanthack and Sladen, Influence of the milk supply on the spread of tuberculosis. Lancet 1899 (I) : 74- 79. 21. Klein, E., Zur Kenntniss der Verbreitung des B. tuberculosis und pseudotuberculosis in der Milch sowie der Biologic des Bacillus tuberculosis. Centralb. f. Bakt. 28: 111-114. 1901. 22. Kudinow, N. P., Bakteriologische Untersuchungen der in Jur- jfew verkauflichen Milch. (Deutsche Ztschr. Thiermed. und ver- gleichende Path. 2: 147-151. 1898.) Cited by Mohler. 23. Macfadyen, Allen, Spread of tuberculosis by milk. A report to the Hackney Vestry. Lancet 1899 (I) : 849-850. 24. Magill, Discussion. Sixth Internat. Cong, on Tuberculosis 4, part 2: 531. 1908. 25. Marconi, II latte di vacca a Napoli in rapporto con 1'igiene e con la tuberculosi. (La Riforma Veterinaria 3: 435-452; 483-559. 1900.) Cited by Mohler. 26. Marshall, C. E., Mich. Agr. Coll. Exp. Sta. Bull. 184: 207-266. 1900. 27. Martin, Hippolyte, Recherches ayant pour but de demontrer la frequence de la tuberculose, consecutive a 1'inoculation de lait vendu a Paris, sous les portes cocheres. (Revue de Med. 4: 150- 161. 1884.) Cited by Mohler. 28. Massone," A., Sulla presenza del bacillo tuberculare nel latte del mercato di Geneva. (Annali d'igiene sperimentale 1897 p. 239.) Ref. Hyg. Rundcshau 7: 605. 1897. 29. Miiller, 0., Milk and dairy products as sources of infection in tuberculosis. Jour. Comp. Path, and Ther. 19: 19-33. 1906. 30. Obermiiller, Kuno, Ueber Tuberkelbacillenbefunde in der Marktmilch. Hyg. Rundschau 5: 877-883. 1895. 31. Ott, Ein weiterer Beitrag zur Milch-hygiene, Ztschr. f. Fleisch- und Milch-Hyg. 8: 69. if TUBERCULOSIS OF FARM ANIMALS 343 32. Pawlowsky, Untersuchungen betreffend die Anwesenheit von Tuberkelbacillen in der Marktmilch und Butter. (Bericht, 10 internat. Kong. f. Hyg. und Demographic.) Cited by Adolph Reitz. 33. Petri, Zum Nachweis der Tuberkelbacillen in Butter und Milch. Arb. a. d. Ksrlchn. Gesundheitsamte 14: 1-35. 1898. 34. Proskauer, Seligmann und Croner, Ueber die Beschaffenheit der in Berlin eingefuhrten danischen Milch. Ztschr. f. Hyg. 57: 173- 247. 1907. 35. Rabinowitsch, Lydia, (Deutsche med. Woch. 26: 416. 1900.) Cited by Anderson. 36. Rabinowitsch und Kempner, Beitrag zur Frage der Infectiositat der Milch tuberculoser Kiihe, sowie uber den Nutzen der Tubercu- linimpfung. Ztschr. f. Hyg. 31; 137-152. 1899. 37. Sacharbekow. Cited by Mohler. 38. Schroeder, E. C., Further experimental observations on the presence of tubercle bacilli in the milk of cows. U.S. Dept. Agr. Bur. Animal Industry Bull. 7: 75. 1894. 39. Smit, H. J., Ueber das Vorkommen von Tuberkelbacillen in der Milch und den Lymphdrusen des Rindes. Centralb. f. Bakt., Abt. I, Orig. 49: 36-71. 1905. 40. Thu, Hans, Untersuchungen auf Tuberkelbacillen in Milch, Butter und Margarine in Christiana. Centralb. f. Bakt. Abt. I, Ref. 36: 597. 1902. 41. Tonzig, C., Ueber den Anteil, den die Milch an der Verbrei- tung der Tuberkulose nimmt, mit besonderen Untersuchungen uber die Milch des Paduaner Markts. Arch. f. Hyg. 41: 46-67. 1902. 42. Woodhead and Wood, Tuberculosis and Milk supply. Lancet 1899 (I) : 395-396. LITERATURE Few of the earlier investigations concerning: tu- GENERALCON- . . SIDERATIONS bercle bacilli in milk of tuberculous cows with sound udders are free from objection. The ease with which milk is contaminated from outside sources was not then appreciated, as it is today. While in the case of market milk the problem is to determine only the presence or absence of tu- bercle bacilli in the milk regardless of the source of infection, here it is desired to know whether or not the sound udders of tu- berculous cows secrete tubercle bacilli along with the milk. So it is necessary to prevent any contamination from outside sources, such as the falling into the milk of manure infected with tubercle bacilli, infected urine, dust of the cow stable, or any contaminated 344 BULLETIN No. 149 [February, material whatsoever. The many possible sources of contamina- tion of the milk with tubercle bacilli make the problem a very dif- ficult one. The difficulty of reaching a conclusion is further indi- cated by the fact that the investigators reporting negative results and those reporting positive results are pretty equally divided. It will be seen by examining the literature, Ta- bles 6 and 7, that most of the results are positive TORS REPORT- .. .^. , i <-> /"v , ING NEGATIVE we CQme to Ostertag s work in 1899. Oster- RESULTS ta & took extra precautions to prevent contamina- tions, carefully grooming the cow, washing the udder with antiseptics, and drying with sterile wadding; also tak- ing care to have the milker clean and disinfect his hands. After rejecting the first milk, the sample to be examined was drawn into a sterile flask. He found on testing 50 cows which had reacted to tuberculin that 49 were giving milk entirely free from tubercle bacilli, according to his test. In the case of one cow the evidence was not so conclusive. Of four guinea pigs, inoculated with her milk, one became tuberculous. Ostertag concluded, however, that this infection was due to accidental contamination. Since Ostertag's work, McWeeney* 3 , Stenstrom" 3 , Ascher 2 , Miil- br 38 , and Smit 80 have done some very careful work with entirely negative results. The work of the last two authors was very ex- tensive and will be briefly discussed. Miiller examined the mixed milk from herds in order to detect the presence of cows with tu- berculous udders, as has been mentioned under market milk. Of 1596 samples tested, Miiller found 97, or 6.2 percent containing tubercle bacilli. When he found a tuberculous sample from a herd, an inspector was sent to examine the udders of each cow in this herd. In 59 of the 97 cases, tuberculous udders were found ; in 56 of these 59 cases, only one tuberculous udder was found in each herd ; in the other three cases, two tuberculous udders in each herd were found. Of the other 38 (author by mistakes says 32) samples, 19 were found to be from herds containing cows suffering from pulmonary tuberculosis ; twelve from herds containing one or several cases of uterine tuberculosis. In another positive sample the bacilli were derived from an animal affected with pulmonary tuber- culosis and several cases of intestinal tuberculosis in the herd. For the remaining five (two not accounted for) positive findings, there was no explanation found and other examinations of the milk of these five herds gave always negative results. But what are we to say of the 1499 herds where the milk showed no tubercle bacilli? Miiller assumed that 30 to 40 percent of the nearly 20,000 dairy cows in these herds were tuberculous, i. e., 6000 to 8000 cows. This is in accordance with the statistics given for that region. Now ac- cording to the views of some authors, the 6000 to 8000 tuberculous TUBERCULOSIS OF FARM ANIMALS 345 TABLE 6 BACILLUS TUBERCULOSIS IN THE MILK OF TUBERCULOUS Cows WITHOUT RECOGNIZABLE DISEASE OF THE UDDER 6 % Author 0) ri A Place id 2 o> ! o Cows positive Percent positive Remark i 2 3 4 5 6 7 8 9 1U 11 12 13 14 15 16 17 18 19 20 21 12 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 Zurn 66 1872 1873 1875 1880 1880 1883 1883 1884 1884 1884 1889 1889 1889 1889 1890 to 1895 1890 1891 1892 1893 1893 1894 1894 1894 1895 1895 1895 1897 1898 1898 1899 1899 1899 1899 1899 1899 1899 1899 1900 1900 Jena 1 1 (?) 1 1 91* 5 3 14 1 20 21 1 1 36 50 13 60 1 6 3 16 Herd 1 (?) 38 5 7 11 34 14 10 7 1 49 9 41 50 7 1 1 (?) 1 1 28* (?)t 4 1 11 2 1 15 2 6 1 2 3 1 Some (?) 6 Some 1 9 6 1 15 1 100.0 100.0 (?) 100.0 100.0 30.7* 0.0 (?)t 28.5 100.0 55.5 9.5 100.0 0,0 41.7 0.0 15.4 10.0 100.0 33.3 100.0 6.3 (?) 0.0 (?) 15.7 (?) 0.0 9.1 0.0 64.3 60.0 0.0 100.0 0.0 0.0 39.0 2.0 0.0 Fed to swine Fed to test animals 45.6 percent, of test animals positive Fed 3 hogs, 1 positive Fed 5 hogs, all positive 1 udder tuberculous post mortem Normal udders Fed to pigs and rabbits Normal udders Normal udders Normal udders Fed to calves Fed to swine Fed to guinea pigs 38 percent of test animals positive Guinea pig test Udders sound Udder tuberculous, post mortem 1 udder tuberculous, post mortem Microscopic Sound udders 50 cows examined; one doubtful Sound udders Sound udders Klebs 87 L/eipsic Semmer 59 Eeipsic Bellinger 8 Germany Bollinger 8 . . Germany Johne 26 (Compilation) . . . ,-r May 3 a Munich Imlach 85 Stein 62 . . Berlin Zschokke 05 . . . Germany Hirschberger 2 * . . . Bang 5 Munich Denmark . . : France Peuch 47 . . ... Galtier 21 France Ernst 18 Boston, Mass Schmidt- Muhlheim 36 Mac Fadyean and Woodhead 29 England Bang- 4 . Denmark . Philadelphia Smith and Schroeder 61 Washington, D. C. Ithaca, N. Y L,aw 88 Schroeder 58 Hills and Rich 23 .... Chester 10 Washington, D. C. Vermont Newark t Obermuller 4 1 Berlin Gehrmann 22 . ... Ravenel 51 Pennsylvania.. . . Liverpool Delepine 1 2 Delepine 1 2 England Bassett 7 Madison \Vis Rabinowitsch and Kempner 49 . . . Adami and Martin 1 . Ascher 2 Berlin Canada Konigsberg Rogers and Gamier 52 Italy Ostertag 4 2 Berlin Miiller 36 Bolle, Germany . . . Chicago Gehrmann and Evans 28 Marshall 30 Douerlas 1 7 . Enerland. . Compiled from the experimental work of various investigators up to 1883. f. Milk was fed to various test animals, among them two monkeys later found to be tuberculous; a result of doubtful significance. 316 BULLETIN No. 149 [February, TABLE 6. Continued 6 fc Author OJ -*-> a Q Place -rj * a> k * < rl v tj !-> 0> > te %'* O o o cL Percent positive Remark 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 Nonewitsch 40 Ostertag 44 1900 1901 1902 1902 1903 1904 1904 1904 1904 1906 1908 1908 1909 1909 1909 1909 Wilna, Russia. .... 6 18 15 50 56 3 19 57 13 20 1 10 33 2 2 4 3 12 1 7 5 4 3 2 2 50.0 0.0 0.0 0.0 21.4 0.0 5.3 12.2 38 4 20.0 0.0 30.0 0.0 0.0 100.0 50.0 Six women post mortem examination All udders normal Reacting cows Injected cultures into buffaloes Twenty-five examina- tions; six failed Milk taken with great care Natural tuberculosis, 33 tests Intravenous injection of bacilli Udders sound Berlin Mc\Veeney 33 . ....... Ireland Stenstrom* 3 Hamra Mohler 84 Prettner 48 Russell 55 Washington Germany Madison, Wis Moussu 35 Paris Mart el and Guerin 31 Martel and Guerin 31 Coquot and Cesari 11 De Jong 14 .... .. Germany Germany , . L/eiden Smit 60 Rotterdam Smit 60 Srnit 60 Rotterdam Rotterdam Royal Commission 53 L/ondon Total 748 131 17.9 TABLE 7 BACILLUS TUBERCULOSIS IN THE MILK 01? TUBERCULOUS Cows WITH VISIBLY DISEASED UDDERS 1 Author 0> ft Q Place & O * n v W *-* Cows positive Percent positive Remark 1 2 3 4 5 6 7 8 9 10 11 12 13 May 3a 1883 1883 1884 1889 1889 1891 1892 1892 1894 1896 1898 1899 1900 jvtunich 1 1 (?) 1 19 3 2 1 3 6 1 (8) 39 1 1 (?) 1 14 3 2 1 3 6 1 (?) 100.0 100.0 100.0 100.0 100.0 73.7 100.0 100.0 100.0 100.0 100.0 100.0 (?) 48 samples, 34 positive Bollinger 8 Bane" 3 . Germany Denmark Galtier 21 Peuch 47 Mac Fadyean 29 and Woodhead. . Bang 4 Fiorentini 19 Russell 54 Nocard and L/eclainche 39 . . . Delepine 18 . Rabinowitsch und Kempner 49 . . Douglas 1 ~ L/ondon Denmark Italy Madison, Wis. . . Paris Manchester Berlin ... J^jjcrlajKj Total 34 87.2 TUBERCULOSIS OF FARM ANIMALS 3+7 cows were all capable of secreting tubercle bacilli in their milk. But since the milk from these cows was entirely negative to guinea pigs subcutaneously injected, he concluded that the danger which is attached to the milk of cows with closed tuberculosis absolutely need not be considered, and the chief source of danger is the cow with tuberculosis of the udder. The investigations of Smit 60 were extensive and carefully done. He reviewed both the literature and methods of work at some length. His samples were taken with great care. The milk was drawn into a sterile flask thru a sterile rubber tube connected with a canula which was inserted into the opening in the teat. Sixty cubic centimeters of each sample were centrifuged and 5 cc. of an emulsion of the cream and sediment were injected into the muscle of the right hind leg of a guinea pig. The milk of 33 tu- berculous cows upon which the diagnosis was to be confirmed by post mortem examinations was tested, as above described, and in none of the samples were tubercle bacilli found. The milk of two cows, one with open and the other with closed tuberculosis, was tested for the period of a month by both microscopic examination and animal inoculation, and no tubercle bacilli were found in any sample. A four-year-old cow, in full flow of milk, reacting to tuberculin, was injected intravenously with an emulsion of 50 mg. of pure culture of tubercle bacilli of the human type in 10 cc. of normal salt solution. Her milk, feces, and urine were tested for tubercle bacilli on each pf three days before the injection. All these tests were negative. The same tests made after the injec- tion showed the presence of tubercle bacilli in the feces samples on the first and third day. The urine sample of the first day only contained tubercle bacilli. Eighteen samples of milk were taken during the month with which 23 guinea pigs were inoculated. The milk samples taken on the second day after the injection contained tubercle bacilli, all others were negative. In a similar manner a non-tuberculous cow was injected with 25 cc. of a thick emulsion of a tuberculous lung from another cow and her milk was tested for tubercle bacilli. The control tests before the injection were all negative. The tests of the urine following the injection of lung emulsion were negative; the tests of the feces samples were lost ; the tests of the milk on the first, second, and third days af- ter the injection showed the presence of tubercle bacilli in each sample. The tests of all other samples during the month were negative. INVESTIGA ^ ^ ie aut h rs reporting positive results on test- TORS REPORT- * n m ^^ ^ rom tuberculous cows with sound ud- ING POSITIVE d ers ma y b 6 mentioned Bang 4 ' 5 , Smith and Schroe- RESULTS der 61 , Ravenel 50 , Rabinowitsch and Kempner 49 , Mohler", Moussu" De Jong", The Royal Commis- 348 BULLETIN No. 149 [February, sion on Tuberculosis, England. 63 These investigators have made just as careful tests as those just reported who found negative re- sults. Special mention will be made here of only the last two. De Jong had special opportunity to investigate this question, as he was an officer in the official slaughtering house of Leiden, Holland. Cows that were condemned because of the tuberculin test became the property of the State. So De Jong was able to select just such cows as he chose and obtain samples immediately before slaughtering. This he did and selected cows with closed tuberculosis as determined by testing the saliva, trachial or bron- chial mucus and genital secretions. The samples were very care- fully taken. Before milking, the udder, the teats, and the whole hind parts of the cow were carefully washed with warm soap so- lution and especial attention paid to the opening in the teats. Af- ter the washing with the soapwater the hind parts of the cow were very carefully washed with 3 percent boracic acid, as were also the hands of the milker. The first part of the milk was discarded and only the last part drawn into a sterile flask. Forty cubic cen- timeters of this milk were centrifuged at 3200 revolutions per minute. The cream and sediment were mixed with a part of the skim-milk and 10 cc. of this were injected into the peritoneal cav- ity of one guinea pig and 5 cc. under the skin of the hind leg of another. He tested ten cows and found tubercle bacilli in the milk of three. The Royal Commission on Tuberculosis has made very thoro tests upon five cows, three of which, B, C, and F, showed obvious signs of tuberculosis, and two of which, D and E, showed no vis- ible signs. None of the five cows showed any sign of udder tu- berculosis during life. All the cows were subsequently slaugh- tered and the udders were carefully examined for tubercles and tubercle bacilli. There were none found except in one case, cow F, in which one quarter showed four small nodules which could not possibly have been detected during life. For the autopsy notes, see Table 8. Strictest precautions were taken in the collection of the samples. A pressure tube was connected to a metal catheter which was inserted into the milk sinus of the teat. The teat and the udder were thoroly cleaned. They were washed with a solu- tion of mercuric chloride and the opening of the teat carefully cleansed so that no tubercle bacilli could be pushed back into the sinus. The milk from each quarter was collected separately and was inoculated into guinea pigs and rabbits. Two guinea pigs were inoculated intraperitoneally from each quarter. Each animal received a dose of ten cubic centimeters of uncentrifuged milk plus the deposit of 20 cc. of centrifuged milk, or sometimes 6 cc. plus the deposit of 24 cc. of centrifuged milk. Three of the five cows TUBERCULOSIS OF FARM ANIMALS 349 TABLE 8 TUBERCLE BACILLI IN FECES AND IN MILK OP Six NATURALLY TUBERCULOUS Cows Lesions in the cow at autopsy Guinea pigs inoculated with feces G. pigs fed with feces Swine fed with feces Guinea pigs inoculated with milk & at (A tc $ 3 3 i 3 ^j ^o J3 O ^> O 0) t) ^H 13 c o ^^ Cow .2 u, 0} V i o> >% 3 O |4 3 O Ui 3 0) ^ A *3 O u he V 3 * i ^j 0) *""1 D ' 1 u ~" ~ -M V C A //] TUBERCULOSIS OF FARM ANIMALS 361 nicht zeigen, schadlich? Ztsch. f. Fleisch- und Milch-Hyg. 18: 205- 211. 1908. 44. Ostertag, R. Untersuchungen iiber den Tuberkelbacillengehalt der Milch von Kuhen, welche auf Tuberculin reagiert haben, klin- ische Erscheinungen der Tuberculose aber noch nicht zeigien. Ztschr. f. Hyg. 38: 415-457- 1901- 45. Ostertag, R., Cited by Smit, Centralb. f. Bakt., Abt: I, Orig. 49: 59. 1909. 46. Pearson, Cited by Mohler, U.S. Dept. Agr. Bur. Animal In- dustry Bull. 44: 1-93. 1903. 47. Peiich, F., Cited by Mohler, U.S. Dept. Agr. Bur. Animal In- dustry Bull. 44: 1-93. 1903. 48. Prettner, Beitrag zur Frage der Infectiositat der Milch von mit Tuberculose infizierten Tieren. Ztschr. f. Fleisch- und Milch- Hyg. 14: 222-224. 1904. 49. Rabinowitsch, L. and Kempner, W., Beitrag zur Frage der In- fectiositat der Milch tuberculoser Kuhe, sowie iiber den Nutzen der Tuber culinimpfung. Ztschr. f. Hyg. 31: 137-152. 1899. 50. Ravenel, M. P;, Tuberculosis and milk supply. Public Health 23: 289-296. 1897. 51. Ravenel, M. P., The passage of tubercle bacilli through the normal intestinal wall; a preliminary report. Public Health 29: 404- 405. 1903. 52. Rogers and Garnier, Cited by Mohler. 53. Royal Commission on Tuberculosis, Third Interim Report. Ref. Jour, for Comp. Path, and Ther. 22: 77. 1909. 54. Russell, H. L., The infectiousness of milk from tuberculous cows. Wisconsin Agr. Exp. Sta. Eleventh Ann. Rep. 1894, p. 196- 197. 55. Russell, H. L. and Hastings, E. G-., Infectiousness of milk from tubercular cows. Wisconsin Agr. Exp. Sta. Ann. Rep. 1904, p. 172-177. 56. Schmidt-Muhlheim, Cited by Smit, H. J., Centralb. f. Bakt. Abt. I, Orig. 49: 41. 1909. 57. Schreiber und Neumann, Gehen Rotlaufbazillen durch das nor- male Enter geimpter Rinder in die Milch iiber? Ztschr. f. Fleisch- und Milch-Hyg. 18: 57-58. 1908. 58. Schroeder, E. C., Further experimental observations on the presence of tubercle bacilli in the milk of cows. U.S. Dept. Agr. Bur. Animal Industry Bull. 7: 75-77. 1894. 59. Semmer, E., Versuche iiber die Uebertragbarkeit der Tubercu- lose (Perlsucht) der Rinder auf andere Thiere. (Deutsche Ztschr. f. Thiermed. und vergleichende Path. 2: 209-220. 1875.) Cited by Mohler. 60. Smit, H. J:, Ueber das Vorkommen von Tuberkelbacillen in der Milch und den Lymphdriisen des Rindes. Centralb. f. Bakt., Abt. I, Orig. 49: 36-71. 1909. 362 BULLETIN No. 149 [February, 61. Smith, Theobald and Schroeder, E. C., Some experimental observations on the presence of tubercle bacilli in the milk of tuber- culous cows when the udder is not visibly diseased. U.S. Dept. Agr. Bur. Animal Industry Bull. 3: 60. 1893. 62. Stein, Gottlieb, Experimented Beitrage zur Infectiositat der Milch perlsiichtiger Kiihe. (Inaugural Dissertation, Berlin, 1884.) Cited by Mohler. 63. Stenstrom, Beitrag zur Frage der Gegenwart von Tuberkel- bazillen in der Milch von Kiihen, welche nur auf Tuberkulin re- agieren. Ztschr. f. Fleisch- und Milch-Hyg. 14: 277. 1902. 64. Wesener, F., Kritische und experimentelle Beitrage zur Lehre von der Fiitterungtuberculose. Cited by Mohler. 65. Zschokke, Ein Weiteres zur Frage der Tuberkulosis. (Sch- weizer Archiv f. Thierheilkunde 26: 262-267. 1884.) Cited by Moh- ler. 66. Zurn, F. A., Cited by Mohler. EXAMINATION OF FECES FOR TUBERCLE BACILLI. INTRODUCTION. Schroeder and Cotton" first called attention to the frequency with which the feces of cattle contain living tubercle bacilli, and the possibility of their introduction into milk in ths way. "We have seen, also, from the work of the United States Public Health and Marine-Hospital Service that among 172 samples of city milk examined 121 or 70 percent contained a sediment after standing a fezv hours in the original containers, and that the sediment consisted in part of cow feces" . (Schroeder 7 pages 14-15.) Tru- man 12 examined 232 samples of milk collected in the city of Chi- cago. Of these, 158 samples or 68 percent contained visible sedi- ment. Of 212 samples of milk collected in other Illinois cities of over 10,000 inhabitants he found 88 percent containing visi- ble sediment. These facts indicate the danger to man from the use of milk contaminated with manure of tuberculous cows, and because of the frequency with which fecal material from cows is present in market milk this phase of the subject seems worthy of careful study. Furthermore, the recognition of the importance of tubercle bacilli in the manure is a rather recent development in the study of bovine tuberculosis. For these reasons it is con- sidered in some detail. LITERATURE Schroeder and Mohler 10 first proved that tubercle bacilli will pass thru the alimentary tract of cattle and pass out in their feces alive and virulent for hogs. They also performed several TUBERCULOSIS OF FARM ANIMALS 363 experiments by adding to milk, feces from cows known to be passing tubercle bacilli, and then inoculating guinea pigs with this soiled milk. In this way they were able to produce tuberculosis in the guinea pigs. They claim that the chief source of danger lies in the contamination of the milk with feces which may con- tain tubercle bacilli. Schroeder and Cotton* examined by micro- scopial tests seven natural cases of tuberculosis in cattle, and found tubercle bacilli in the feces in five or 71.43 percent of the cases. They consider the microscopical test a reliable means of determin- ing the bacilli in cow feces. They say "At first it was believed that the bacilli in the preparations belonged in whole or in part to the latter group (harmless acid-fast bacteria') but this view teas not tenable after it was discovered that similar bacilli could not be found in material from non-tuberculous cattle." (Schroeder and Cotton, 9 p. 9.) Inoculation experiments were made with the feces of the five cows, into guinea pigs, with the result that three of the five were shown to be passing virulent tubercle bacilli. They consider feces to be the most dangerous factor in the dis- semination of bovine tubercle bacilli. "In this respect feces must be regarded as having a place zmth cattle similar to that com- monly accorded to sputa unth tuberculous persons." (Schroeder and Cotton/ p. 22.) . They conclude that every cow known to be affected with tuberculosis must be regarded as positively dan- gerous, since we are unacquainted with any means by which it can be determined when tuberculous cattle or their feces become dangerous to the health of persons or animals. Schroeder 7 says that practically no tuberculous herd is very long free from a dangerously tuberculous animal, and since we cannot tell when a tuberculous cow becomes dangerous we are forced to assume for practical purposes that every tuberculous cow is dangerous from the moment she is affected with tuberculosis. She may be dangerous long before the disease is indicated by her physical appearance. He says that about forty percent of all cows reacting to tuberculin are passing virulent tubercle bacilli in the feces, this being the commonest mode of expelling these germs from a cow's body. In a later communication he further emphasizes the danger to man from this source. He collected twelve tuberculous cows from several dairy herds and kept them under observation for two years. They were in excellent general condition and had no visible symptoms of the disease. During the first two months microscopial examinations showed that five or 41 YJ, percent were expelling tubercle bacilli ; eighteen months later ten or 83^3 percent were expelling these germs in the feces. He says that ordinary market milk always contains sediment con- sisting of cow dung. The micro-organisms are very easily de- 364 BULLETIN No. 149 [February, tached and washed from the dung into the milk. Such milk inoculated into guinea pigs produces tuberculosis. Tubercle ba- cilli are found in the separated cream as well as in the sediment. Butter made from such cream also contains them in a condition virulent to guinea pigs. In ordinary salted butter they may re- main alive at least 1 60 days. In conclusion Schroeder 8 says that tuberculous cows frequently expel tubercle bacilli per rectum. These germs may come from the lungs and not necessarily from the intestinal mucosa. They are evenly distributed in the mass of the f eces and expelled in large numbers ; easily enter ,milk, butter and ice-cream and may thus be transmitted to human beings. He concludes further that since Mohler and Washburn have converted the bovine type of tubercle bacillus into the human type by pass- age thru cats (which work he accepts as trustworthy), this may occur in the human body as readily, and it is therefore impera- tively necessary for the protection of public health that all milk should be obtained from cows certainly free from tuberculosis, and from cows stabled, milked, and pastured in environments free from tuberculous infection, or else the dairy products should be pasteurized or sterilized before they are used as food in any form. Reynolds and Beebe 5 have also examined the feces of tubercu- lous cattle. They had for their material feces of 45 cattle which had reacted to tuberculin. These were, for the most part, pure bred cattle. The majority of them had reacted to tuberculin two years and a part several years before the experiment was made. The manure of sixteen of these was tested three times; that of one, five times; of two, four times; of fourteen, twice, and of seven, once. Of the entire herd, only one animal was excreting virulent tubercle bacilli in her feces. Her nasal secretions also contained Bacillus tuberculosis. The feces were collected with a rectal spoon The hind parts of the cow were previously disin- fected in most cases. Two grams of feces were rubbed up in sterile water. At first 5 cc. of this emulsion were injected sub- cutaneously; this killed all the guinea pigs. Later, 0.5 cc. of the emulsion was used and this worked fairly well. Microscopic test for the presence of tubercle bacilli in manure is not reliable according to these authors. In the single case in which the ani- mal inoculation proved the presence of the tubercle bacilli they were unable to detect these germs in the feces by the use of the microscope. They also consider it impossible to distinguish with certainty by microscopic examination alone, between tubercle ba- cilli and other acid-fast bacteria which may be present in the feces of the cow. The Royal Commission on Tuberculosis 8 tested the samples /p/z] TUBERCULOSIS OF FARM ANIMALS 365 of feces from six naturally tuberculous cattle, of which three showed physical signs of tuberculosis. For the purpose of the experiments, a considerable quantity of fecal matter was required. The action of the rectum was stimulated by injecting air into it thru a sterile glass tube and the feces were received directly into a sterilized pail applied to the margin of the anal orifice. This was necessary to 'prevent contamination from the vagina since three of the cows tested had extensive tuberculosis of the uterus and one, at least, was excreting purulent material from the va- ginal outlet. A portion of the feces was rubbed up in a mortar with sufficient normal salt solution to form an emulsion that could be pressed thru muslin. It was found that 0.05 cc. of this emul- sion injected intraperitoneally did not kill the test animals with acute peritonitis. Feces were fed also to guinea pigs and swine. The swine were young animals, about eleven weeks old, all be- longing to three litters, raised upon the same farm. They were all tested with tuberculin and found free from tuberculosis before they were used in the experiments. One pig from each of the two litters was kept as a control and when killed at the end of the experiment was quite normal. The feces were fed to the swine in previously sterilized milk, each animal being fed for about one week, receiving in this time several pounds of moist feces. Some guinea pigs were fed the tuberculous feces in sterilized milk, and others received a single dose administered by a pipette. The dose given was one cubic centimeter of the emulsion used for inocula- tion, a quantity fifty times that employed for intraperitoneal in- jection. Except those guinea pigs that died early and were thus thrown out of the test, every guinea pig save one inoculated with feces from the three obviously tuberculous cows become tubercu- lous. Every one of the five swine fed the feces from two of these cows became tuberculous. 'The feces of the other cow were not fed to swine. From cow D, one of the cows that showed no outward signs of tuberculosis, six of the fourteen guinea pigs inoculated became tuberculous. The 41 guinea pigs inoculated with feces from cow A, and 18 guinea pigs inoculated with feces of cow E remained free from tuberculosis. Four, swine were fed with the feces of cow A and one of them became tuberculous. Two swine were fed with the feces from cow D, and two with the feces of cow E. These animals remained free from tubercu- losis. The feeding of 36 guinea pigs with the feces of cows B and C produced only one tuberculous guinea pig. This one was' fed with the feces of cow C. The feces from these cows, B and C, were shown by inoculation into guinea pigs and by feeding swine to be very virulent. It thus appears that feeding suspected material to guinea pigs is not a reliable test for the presence of 366 BULLETIN No. 149 [February, tubercle bacilli. Their results show that five of the six naturally tuberculous cows were passing virulent tubercle bacilli in their feces as shown by inoculation and feeding tests. Two of these cows, A and D, were apparently in excellent health. Cow A, upon post mortem examination, was found to have tubercles in the lungs and in the small intestine; cow D had a bad lung, with pus in the bronchial tubes. For a tabulation of the results see Table 8, page 349, under literature of milk investigation; also the autopsy records of these cows, page 388. Peters and Emerson* consider the microscope alone of little value in the examination of feces for tubercle bacilli, because in feces similar bacilli are met with, which stain in the same way and cannot be distinguished morphologically from the tubercle bacilli. Moreover, the real tubercle bacilli may be present in such small numbers that a microscopic examination would be inadequate to disclose them, as was shown by Reynolds and Beebe. Guinea pig inoculations therefore must be resorted to; and not only that, but the characteristic lesions of tuberculosis must be produced in the animals before the presence of tubercle bacilli can be accepted as proven. In their experiments these authors used forty-one cows. The majority of the cows were in excellent physical con- dition. In all cases, the tuberculin test had been applied only a few weeks before making the examinations. The feces were taken with a rectal spoon. Microscopical examinations were made for acid-fast organisms, and when such were found, guinea pigs were inoculated. For these inoculations the manure was suspended in water and 0.5 to I cc. of the suspension injected. The guinea pigs died of septicemia in only a few instances. Feces from twenty-two cows were tested in this way and, of these, three were shown to be passing virulent tubercle bacilli. Two of these three were in good physical condition, and from their appearance the disease would not have been suspected. The third was thin but signs of the disease were not marked. Calmette and Guerin 2 injected pure cultures of tubercle bacilli into the ear vein of several rabbits. On each of seven days after these injections a rabbit was killed and immediately the bile was removed, centrifuged, and the sediment injected into four guinea pigs. Forty-five days later all the guinea pigs were killed and examined, with the following results : Two of the four guinea pigs inoculated with the bile sediment from the rabbit killed on the third day after the injection were tuberculous. One of the four guinea pigs inoculated with bile sediment from the rabbit killed on the fourth day after its injection was tuberculous. All of the eight guinea pigs inoculated with bile sediment from the fifth and TUBERCULOSIS OF FARM ANIMALS 367 sixth day rabbits were tuberculous, and three of the four guinea pigs inoculated with bile sediment from the seventh day rabbit were tuberculous. It is therefore evident that tubercle bacilli may be excreted by the liver and pass with the bile into the intestine. EXPERIMENTAL For the first series of examinations of the dairy THE SAMPLES COLLECTION OF , , , t tf ., herd the samples were all taken directly from the rectum of the cow with a sterilized bone spoon by one of us (Briscoe) and transferred to a sterile four ounce glass-stoppered bottle. This was usually done while the cow was in the act of defecating tmt at times the sample was taken by pushing the bone spoon several inches into the rectum, revolving and then withdrawing it. By repeating this operation once or twice eight to ten grams of feces were obtained. The spoon was washed in five percent carbolic acid and dried with a sterile cloth before taking each sample. The sample of feces for the second series, and all samples since then, have been taken by the method recommended by the Royal Commission of England, viz., by pumping air into the rectum and thus stimulating the cow to void. This method has worked exceedingly well. It was often possible in this way to cause a cow to void again immediately af- ter a natural defecation. Thick walled sterile glass tubes ten inches long with edges well rounded off in the flame were at- tached by rubber tubing to a rubber bulb air pump. The glass tubes were kept in a sterile tin case and one used for each cow. The tube was inserted into the rectum for five or six inches and air pumped in until the cow was stimulated to void. This re- quired only a few minutes and very rarely failed. The feces were caught in a one-half gallon sterile tin pail provided with a tight fitting cover. The pail was opened only while collecting the sample. It was held close up to the anus in order to avoid get- ting any material from the vagina. About five hundred grams of feces were collected for each sample. MAKING THE ^ ie sam pl e was mixed with a sterile glass rod SUSPENSION ar| d exactly one gram of it transferred to the inside of the neck of a sterile fifty cubic centi- meter, glass-stoppered, measuring flask.* This was moistened with sterile 0.8 percent salt solution, and rubbed to an emul- sion in "the neck of the flask with a sterile glass rod. The flask was then filled one-third full by the addition of salt solution *This method of preparing a suspension of feces has been described by MacNeal, Latzer, and Kerr. 3 368 BULLETIN No. 149 [February, and thoroly shaken, inverted, and the sediment allowed to fall back into the neck. By gradually tipping back the flask, the sedi- ment was left in the neck where it was again rubbed up with the glass rod. This was repeated until the bacteria of the feces were brought thoroly into suspension, and the pieces of straw ap- peared quite clean. The flask was now filled to the mark and the contents thoroly mixed. This gave a two percent suspension of feces. CENTRIFU Forty cubic centimeters of this suspension were CATION transferred by means of a sterile Pasteur bulb pipette to two sterile centrifuge tubes. These were centrifuged for five minutes at 1800 revolutions per minute, which treatment threw down only the coarser sediment. The supernatant liquid containing most of the bacteria was drawn off and put into other centrifuge tubes and again centrifuged thirty minutes at 2000 revolutions per minute. This produced a kind of differential sedimentation and is a process similar to that used by Strassburger 11 in his work on the occurrence of tubercle bacilli in the feces of men and also to that of MacNeal, Latzer, and Kerr 3 in their work on fecal bacteria of healthy men. The preparations for microscopical examinations STAINING . ,, % FECES were made by smearing the feces directly on SAMPLE FOR slides. This was done either by picking up shreds B. TUBERCU- of mucus from the surface of the fecal mass and LOSIS spreading them on the slides, or by taking some of the mass itself and smearing it over the slide and rubbing it down to a thin layer with another- slide. Besides the direct smears made in the two ways described, preparations of the sediment obtained by centrifugation were made and stained. All these preparations were stained in the manner described un- der general methods for staining Bacillus tuberculosis (page 324). INOCULATION - P 1 S S were inoculated with each sam- |NTO TEST pie. The dose used for inoculation was in most ANIMALS instances one cubic centimeter of the original two percent suspension of feces. In a few instances ice. of a concentrated suspension of the mixed bacteria obtained by the centrifuge according to the method of Strasburger given above was injected. In all cases the injection was made into guinea pigs under the skin of the right thigh in doses of one cubic centimeter each. It is rather surprising that only a few guinea pigs died of acute infection when inoculated with the cen- trifuged bacteria. It is evident that centrifuging concentrates the fecal bacteria and among them the tubercle bacilli when pres- ent. An attempt was made to compare centrifuged sediment with 1911} TUBERCULOSIS OF FARM ANIMALS 369 the original suspension by inoculating one guinea pig with the centrifuge sediment and another with one cubic centimenter of the two percent emulsion of the same sample. In no one of the twelve cases in which this was done have we found any tubercu- losis. It is therefore impossible to estimate from these tests the relative merit of the two procedures. Only two of the twelve guinea pigs inoculated with centrifuged material died of acute in- fection and apparently one of these was not due to the bacteria inoculated but largely to some improper food. The results indi- cate that guinea pigs may stand a much larger dose of bacteria if they are free from other undissolved material. DELICACY OF Three series of experiments to ascertain the deli- ANIMAL IN- Cac 7 f guinea pig inoculation as a means of test- OCULATION ing feces for tubercle bacilli have been performed. TESTS In these experiments definite amounts of a pure culture of bovine tubercle bacilli were added to suspensions of cow feces and then tested by injection into guinea pigs. The original of the cultures employed in the tests was obtained from Dr. Theobold Smith of Harvard Medical School, in October, 1908, and has since been propagated in our laboratory upon artificial media. The culture grows well on gly- cerin agar, glycerin potato, egg medium, and blood serum. It has been tested upon guinea pigs and shown to be virulent. A heavy growth upon an inclined glycerin agar was used. The mass of bacterial substance was very carefully removed from the agar surface by means of a strong platinum wire, avoiding as far as possible the admixture of moisture or of agar. The bac- terial substance was transferred to the inside of the neck of a weighed sterile 25 cc. measuring flask, and then accurately weighed upon an analytical balance. A little sterile 0.8 percent salt solu- tion was then added and the mass thoroly rubbed up in the neck of the flask with a sterile glass rod. The glass rod and inner wall of the neck of the flask were carefully washed with salt solution as it was added to the contents of the flask from a Pasteur bulb pipette. By vigorous shaking and by further trituration of mi- croscopic particles an apparently homogeneous suspension was ob- tained. It was then diluted to the mark on the flask (25 cc.). Microscopic examination showed most of the bacterial cells to be separate, but not all of them. Clumps of twenty or more cells and also some larger masses were still present. By allowing the suspension to settle a moment just before measuring out portions from it, these clumps were largely avoided in the material actually inoculated into the animals. 370 BULLETIN No. 149 [February, Experiment i For this experiment suspensions of a pure culture of bovine tubercle bacilli were desired in dilution of one to one hundred thousand and one to ten million. To make these suspensions 25 milligrams of the agar culture were weighed off and diluted accord- ingly.* The sample of feces was taken from heifer No. 73, which had given a negative tuberculin test. The results are recorded in Table 12. Every one of the guinea pigs in this experiment be- came tuberculous, even the controls. This was quite a disappoint- ment, as far as the controls were concerned, for the feces were supposed to be free from tubercle bacilli. Immediately upon find- ing that the controls were tuberculous, another sample of feces was obtained from heifer 73, and four healthy guinea pigs inocu- lated with suspensions of the same. At the end of 83 days autopsy TABLE 12 THE QUANTITY OF TUBERCLE BACILLI IN FRESH Cow MANURE NECESSARY TO PRODUCE TUBERCULOSIS WHEN INOCULATED INTO GUINEA PIGS Material sus- j pended, per cc. -2 c V Mode r. o> ' C ; c ID C Ti rfc Tubercle "-H ' '3 be of ctf fie Autopsy h a 9 _o o rt .2 JL 1 Cv^Co bacilli So death Q ,2 tn al "3 mg. mg-. tl 0. 0.0100 201 Killed 42 Generalized tuberculosis - - _l_ 20. 0.0100 -(- 195 Killed 52 Generalized tuberculosis - _l_ 20. 0.0100 -)- 196 Died 40 Generalized tuberculosis + -|- 20. 0.0100 193 Died 34 Generalized tuberculosis - - -j- 20. 0.0100 194 Killed 47 Generalized tuberculosis _j_ 0. 0.0001 202 Killed 67 Generalized tuberculosis _)_ 20. 0.0001 -j- 199 Killed 42 Generalized tuberculosis _l- 20. 0.0001 -[- 200 Killed 42 Generalized tuberculosis _j_ 20. 0.0001 197 Killed 42 Generalized tuberculosis - _l_ 20. 0.0001 198 Killed 54 Generalized tuberculosis - JL 20. 0.0000 191 Killed 42 Generalized tuberculosis - _(_ 20. 0.0000 192 Killed 42 Generalized tuberculosis + Table 12a. Supplementary test of Feces of Heifer 73 20. 0.0000 273 Killed 83 Normal 20. 0.0000 274 Killed 83 Normal 20. 0.0000 275 Killed 83 Normal 20. 0.0000 276 Killed 83 Normal showed them all to be healthy. It appears that by some mishap the control suspension contained some of the pure culture of tubercle bacilli. At the time there was available only one syringe and it is probable that the boiling which the syringe received be- tween inoculations was insufficient to kill all the tubercle bacilli ; and in this way some of the pure culture was carried over to the controls. *Tubercle bacilli Feces or salt solution. 0.025 -g. in 25 cc. 5 cc. Dil. A (.005 g.) in 100 cc. 10 cc. Dil. B (.000, 5 g.) in 50 cc. =Dil. A. i cc. contains =Dil. B. i cc. contains =Dil. C. i cc. contains 0.5 cc. Dil. C (.000,005 g.) in 50 cc. =Dil. D. I cc. contains Tubercle bacilli .001 g. .000,05 g. .000,01 g. .000,000,1 g. TUBERCULOSIS OF FARM ANIMALS 371 Experiment 2 Two dilutions of bovine tubercle bacilli were used in this ex- periment, the more concentrated being the same as the most di- lute in Experiment i, i.e., one ten millionth gram of bovine tu- bercle bacilli per cubic centimeter ; the highest dilution, one billionth gram per cubic centimeter. The fresh feces were obtained this time from heifer 74. She had just calved and was negative to the ophthalmic, cutaneous, and subcutaneous tuberculin tests. The sample was collected and suspensions made as described above. To make these suspensions 22 milligrams of germ substance from an agar culture were diluted as in Experiment I.* In injecting these doses into the animals, three syringes were employed. These were sterilized in the autoclave at fifteen pounds pressure for fif- teen minutes before using the first time. They were then steri- lized by boiling for fully fifteen minutes before being used again The suspensions free from tubercle bacilli were injected first, then TABLE 13 THE QUANTITY OF TUBERCLE BACILLI IN FRESH Cow MANURE NECESSARY TO PRODUCE TUBERCULOSIS WHEN INOCULATED INTO GUINEA PIGS. EXPERIMENT 2. Material sus- CO pended, per cc. 4 ri o *>5 Mode *d 01 5 10 cc. of Dil. D (0.000,000,05 g.) in 50 cc. =Dil. E, I cc. contains 0.000,000,001 g. i cc. of Dil. D (0.000,000,005 g-) in 50 cc. Dil. F, i cc. contains 0.000,000,000,1 g. - cc. of Dil. D (0.000,000,000,5 g.) in 50 cc. =Dil. G, i cc. contains 0.000,000,000,01 g. TUBERCULOSIS OF FARM ANIMALS 373 TABLE 14 THE QUANTITY OF TUBERCLE BACILLI IN FRESH Cow MANURE NECESSARY TO PRODUCE TUBERCULOSIS WHEN INOCULATED INTO GUINEA PIGS. EXPERIMENT 3. Material sus- pended, per cc. Centrifu- gation Guinea pig No. Mode of death a tC V >,a> ei a P *v Autopsy Micro- scopic exam. Cultures Feces mg. Tubercle bacilli mg. 20. 20. 20. 20. 0. 20. 20. 20. 20. 0. 20. 20. 20. 20. 0. 0. 20. 20. 0.000 001 0.000 001 0.000 001 0.000 001 0.000 001 0.000 000 10 0.000 000 10 0.000 000 10 0.000 000 10 0.000 COO 10 0.000 000 01 0.000 000 01 0.000 000 01 0.000 000 01 0.000 000 01 0.000 000 01 0.000 000 00 0.000 000 00 + + + 369 368 367 366 378 373 372 371 370 379 377 376 375 374 381 380 365 364 Died Died Died Died Killed Died Killed Died Killed Died Died Killed Died Killed Died Died Killed Died 10 1 10 13 54 1 53 6 53 6 1 53 5 53 3 55 11 No tubercles Died within a few hours No tubercles No tubercles Generalized tuberculosis No tubercles Generalized tuberculosis No tubercles Normal No tubercles No tubercles Generalized tuberculosis No tubercles Slight tuberculosis No tubercles Died when inoculated Normal No tubercles "o" + + + improper care. One of the animals selected for this experiment, guinea pig 380, died after it was brought into the room just before time for inoculation, and within 8 or 10 days one-half the guinea pigs in the experiment were dead. So this experiment was al- most a failure. However, we find even the greatest dilution caus- ing generalized tuberculosis (No. 376). A decided difference was observed between centrifuged and uncentrifuged samples. Note in this connection guinea pig 370, inoculated with a sample that was not centrifuged, having a di- lution of one-ten billionth gram of pure culture of bovine tubercle bacilli. This guinea pig remained healthy, while guinea pig 372, inoculated with the sample of the same dilution but centrifuged, contracted generalized tuberculosis. Also note that guinea pig 374, inoculated from the sample whose dilution was one-one-hun- dred-billionth gram per cubic centimeter, which was not centri- fuged, had only slight tuberculosis ; while guinea pig 376, inocu- lated from the sediment of the same sample after centrifugation had generalized tuberculosis. Conclusions i. A pure culture of bovine tubercle bacilli in quantities as small as one one-hundred-billionth gram suspended in sterile salt solution or in one percent feces emulsion produced tuberculosis in guinea pigs. 374 BULLETIN No. 149 [February, . 2. When the dilutions of pure culture of bovine tubercle ba- cilli were as high as one-hundred-billionth gram per cubic centi- meter of feces suspension the subcutaneous inoculation of the sed- iment of 20 cc. of the centrifuged suspension produced a more ex- tensive generalized tuberculosis than did a similar inoculation of one cubic centimeter of the same suspension not centrifuged. In dilutions containing a greater amount of germ substance this dis- tinction was not so clear, as very extensive generalized tuberculosis usually resulted from the injection of even the one cubic centi- meter of the uncentrifuged material. 3. No difference was noted in the effect upon guinea pigs whether the pure culture was inoculated in one cubic centimeter of sterile salt solution or in one cubic centimeter of one percent fresh feces suspension. Possibly higher dilutions may determine this point. METHOD OF Before obtaining the first shipment of guinea pigs PROCEDURE which were to be used in this work, a considerable number of miscroscopic tests were made. The samples were collected as previously described for the first series of tests. Two samples were taken from each cow, a mass sample of feces picked up from the floor after the cow had voided, and a sample scraped from the wall of the rectum by means of a sterilized bone spoon. From these two. samples, three microscopic prepara- tions were made. Shreds of mucus were first removed from the surface of the mass of feces and spread out on glass slides, fixed and stained. Then the mass was thoroly mixed and a smear of this mixed material made on a slide, fixed and stained. A third preparation was made by smearing the mucus fluid obtained from the rectum on a slide. All three of these smears were stained with carbol-fuchsin and methylene blue as described under General Methods (page 324). Each slide was examined 15 minutes and the number of acid-fast organisms recorded as shown in the Tables 15 and 16. If no acid-fast organisms were found in any of the slides made from the sample tested, it was recorded as a negative result. Usually only a few acid-fast organisms were found on one slide in the 15 minutes. Only five times from the 25 slides examined were as many as 25 acid-fast organisms found on a slide. At no time were shreds of tissue containing abundant acid-fast organisms found; the bacilli were generally separate. An acid-fast strepto- thrix was found in one sample of feces from cow No. 12. Often oval spore-like acid-fast bodies were found. These in no way re- TUBERCULOSIS OF FARM ANIMALS 375 sembled Bacillus tuberculosis and were never included as positive findings in the results tabulated. The only acid-fast organisms so included were short, thin rods which were often slightly curved, but sometimes quite straight. In appearance, they resembled in- dividual rods of the bovine tubercle bacillus, a slide of which was constantly kept at hand for comparison. The feces of both reacting and non-reacting cows were examined microscopically for tubercle bacilli. Table 15 gives the results from the reacting cows. Altogether 119 RESULTS TABI,E 15. MICROSCOPIC EXAMINATION FOR TUBERCLE BACII,I,I IN THE FECES OF REACTING Cows 6 X | o O 6 fe *cL a 3 cc Date of examination A. Mass of feces B. Scraping from the rectum Summary of -4-> tfl 0) *J Summary of test on cow 1. External mucus 2. Mixed material "8 % g ^c E J1 r<3 Pos. Neg. 10 12 13 15 16 19 28 29 33 35 36 1 2 1 2 3 1 2 3 4 5 6 7 8 1 2 1 2 1 2 1 2 3 4 5 6 7 8 1 2 3 4 5 1 2 1 2 3 4 1 2 3 6-29-08 8-20-08 6-18-08 7- 1-08 4-28-08 6-18-08 6-22-08 7- 1-08 7-11-08 8- 6-08 8-12-08 8-19-08 9- 2-08 6-19-08 8-28-08 6-24-08 8-13-08 6-28-08 8-18-08 6-18-08 6-22-08 7- 1-08 7-11-08 8- 6-08 8-12-03 8-18-08 9- 2-08 6-29-08 7- 1-08 9- 2-08 10-30-09 11-13-09 6-25-08 8-14-08 6-19-08 8- 6-08 9- 3-08 11-13-09 6-19-08 6-24-08 8-13-08 2 only one 4 < ( -1 - - J ) D 3 - f f 1 1 5 1 1 1 7 1 1 1 2 1 2 3 1 1 1 1 4 1 3 1 many * 24 several some 6 several several 4 several several 2 2 3 1 few 25 several some several 12 12 2 2 1 many 1 5 3 3 2 "6 many severa few severa 1 several . 2 *Acid-fast streptothrix. 376 BULLETIN No. 149 [February, TABLE 15. Continued r- A. I Vlass of : 'eces <+H O Kl 1 6 Z O . fc D 3. I MH O . rt 1* r> -w 5 '! External mucus Mixed laterial ~8 _ ~ c -2 '= i l~s be -_ f}l % oo % ^ O2 MH W, ummary test 13 ( si s^ 1^ J fe 35 ; o H 5 CC 0) iH c N ro H CC Pos. Neg. 37 1 6-19-08 o 2 8-28-08 o o o 2 39 1 6-23-08 few 2 3 4- 2 8-25-08 o 1 1 41 1 6-29-08 2 4 few 4- 2 9- 3-08 o o 1 1 42 1 6-19-08 o 2 8- 6-08 o o 3 9- 3-08 o o 4 10-30-09 o o o 4 47 1 6-29-08 6 2 J- 49 2 1 9- 3-08 6-24-08 12 "o 4- 1 1 2 8-26-08 o o 1 1 51 1 6-26-08 2 8-18-08 o o o 2 52 1 6-22-08 4 several several 2 7- 1-08 8 3 8-21-08 1 3 53 1 6-25-08 several several several 2 8-14-08 o o 3 8-18-08 o o 4 9- 2-08 _l_ 2 2 54 1 8-25-08 o 'o 1 55 1 6-29-08 3 many 2 4- 2 8--27-08 2 4. 3 11-13-09 1 3 56 1 6-25-08 6 several 3 4- ? 8-14-08 o o 3 8-30-09 o o 4 11-13-09 1 -1- 2 2 57 1 6-25-08 several several several ? 8-14-08 o o 1 1 58 1 6-26-08 few few 4. 2 8-18-08 o o 1 1 59 1 8-18-08 o o 1 60 1 6-24-08 26 24 several + ? 8-13-08 o o 1 1 61 1 8-11-08 o o 2 8-27-08 o o 2 62 1 6-29-08 3 25 4- 63 2 1 7- 1-08 9- 2-08 J. 1 1 1 64 1 8-20-08 o o 2 8-30-09 o o o 2 66 1 8-20-08 o 1 67 1 8-21-08 o o 1 70 1 8-11-08 o o 2 9- 2-08 + 1 1 72 1 8-30-09 1 4- 1 76 1 8-25-08 o 1 80 1 9- 2-08 2 4- 1 TUBERCULOSIS OF FARM ANIMALS 377 TABLE 15. Continued 6 X U 6 K 0) "S, 1 tc Date of examination A. Mass of feces B. Scrapin^- from the rectum Summary of test Summary of test on cow 1. External mucus 2. Mixed material 3. Cenlrilugcd sediment Pos. Neg. 83 86 87 97 98 109 S.H. Iv.H. L.B. W. S6 F. O.S. B. O.C. O. l 2 3 1 1 2 3 1 2 1 1 1 2 1 1 2 1 2 1 2 1 2 1 1 1 1 6-26-08 6-29-08 8-2C-08 6-23-08 6-19-08 6-23-08 8-21-08 6-22-08 8-21-08 6-23-08 9-30-09 9-30-09 11-13-09 9-30-09 11-13-09 11-28-09 11-13-09 11-28-09 11-13-09 11-26-09 11-15-09 11-26-09 11-15-09 11-15-09 11-15-09 11-15-09 3 1 2 13 2 few 4- + + + 0' 2 1 2 1 1 1 1 1 1 1 '0 2 2 2 2 2 1 1 1 1 several few several 2 2 few 4 12 + 2 1 few 1 5 - Total, 53 cows, 119 samples 53 66 samples were tested from 53 cows. Acid-fast organisms were found 53 times ; 66 times there were no such organisms found. Of the 53 reacting cows tested, acid-fast bacilli were at some time found in the feces of all but 19. Of these 19 in which such bac- teria were not found at all, 8 were tested only once, 9 twice, and i four times. Table 16 gives the results of examining the feces from 1 8 non-reacting cows. Forty samples were tested of which 23 were found to have acid-fast organisms and in 17 none were found. From 3 of the 18 cows no samples of feces examined con- tained acid-fast organisms. From two of these, two samples were tested, from the other, only one sample was used. Thus, 28 per- cent of the 53 reacting cows gave entirely negative results, while only 16^3 per cent of the non-reacting herd gave samples without acid-fast organisms. It would seem from this result that no sig- nificance can be attached to the microscopic finding of a few acid' fast bacteria in the feces of cattle. In this connection it is interesting to note the opinions of the following authors : Schroeder and Cotton 9 think that the micro- 378 BULLETIN No. 149 [February, TABLE 16 MICROSCOPIC EXAMINATION FOR TUBERCLE BACILLI IN THE FECES OF NON-REACTING Cows " G A. ] Mass of : eces b h aj . 6 * * W R rj * K Pos. Neg. A. 1 6-^9- '08 o several 4- 2 8-27- '08 14 3 8- 28- '08 6 3 B 1 6-29- '08 o several 9 7- 1-'08 3 8-12- '08 o 4 8-27- '08 o 1 3 c 1 6-29- '08 3 4 _L 9 7- 1-'08 o 3 8-'>7-'08 o 1 2 D 1 8- 7- '08 o 2 9- 1-'08 2 + 1 1 E 1 8-12- '08 o 2 9- 1-'08 o 2 F 1 8- 7- '08 1 4. 2 9- 1-'08 o o 1 1 G 1 8- 7- '08 o o 2 9- 1- ! 08 4 4- 1 1 H 1 8- 6- '08 3 -1- 2 9- 1-'08 2 2 I 1 6-22- '03 24 several several 2 8-25- '08 o 1 1 24 1 6-29- '08 5 2 4. 2 7- 1-'08 o 3 7-28- '08 o 1 2 26 1 6-29- '08 3 1 4- 2 7- 1-'08 o o 2 1 3 7-28- '08 3 4- 65 1 9- 2- '08 o 1 63 1 9- 2- '08 many 4. 1 69 1 9- 2- '08 2 8-11- '08 o o 2 85 1 7- 1-'08 4 4- 2 8-27- '08 1 4. 2 R 1 7-16- '08 3 . 3 4. 2 7-28- '08 4 4 6 4- 2 H 1 2 7-16-'08 7-28- '08 1 2 1 2 4. 2 o J 1 7-16- '08 '9 2 4- 2 7-28- '08 o 1 4 4. 7 Total 18 cows, 40 samples. 23 scope is adequate in determining- the presence of tubercle bacilli in feces of cattle. Reynolds and Beebe were not able to find the tubercle bacilli in the feces of cow No. 2 l /z with which they did produce tuberculosis in guinea pigs by subcutaneous inoculations. They think the microscopic method is not a reliable test for tubercle bacilli in cattle manure. Peters and Emerson 4 do not think that a rp/j] TUBERCULOSIS OF FARM ANIMALS 379 microscopical test is a sufficient means of testing for tubercle bacilli in the feces of a cow. Out of twenty-two cows in the feces of which they had found acid-fast organisms by microscopic test, guinea pig inoculations only showed three to be passing virulent tubercle bacilli. Bang 1 says the microscopic test is of no value ex- cept for rapid work when the test is positive and the tubercle bacilli are found in abundance ; as in the case of tuberculosis of the udder, where a stained preparation of a sample of milk may reveal hun- dreds of tubercle bacilli. ANIMAL INOCULATION TESTS FOR TUBERCLE BACILLI IN THE FECES OF REACTING CATTLE .,..-_,.. The method of collecting the samples and prepar- INOCULATION . >- i T TESTS m ' " ie feces suspension for inoculation has been given (see page 367.) The amount inoculated was one cubic centimeter of a one percent suspension of the feces. Only a few times was the suspension centrifuged and the sediment used for inoculation. Two guinea pigs were used for each sample, with few exceptions, when three were used. There have been made two complete series of tests including all the reacting dairy cattle present in the herd at the times these tests were made : The first series was made between the 28th of December, 1908, and the 4th of January, 1909. At this time samples from thirty-eight cows were tested and two were found to be passing virulent tubercle ba- cilli, cow No. 1004 and cow No. 83. The two guinea pigs inocu- lated from each were found to have generalized tuberculosis. The diagnosis of tuberculosis in the guinea pigs was confirmed by mi- croscopical examination, by cultures, and by inoculations of the tuberculous tissues of each of the four guinea pigs into two other guinea pigs. All of these tests were characteristic. The second series of tests was made between August 6 and Au- gust 9, 1909. Three weeks later it was noted that guinea pigs inocu- lated from samples of cows 72, 64, and 56 had enlarged superior inguinal lymphatic glands and for this reason another sample was collected from each of these three cows and inoculated into guinea pigs on August 31. Curiously enough the swelling of the lymph glands in the guinea pigs inoculated August 6-9 subsided, and at autopsy some weeks later, no evidence of tuberculosis was found in them. However, of the guinea pigs inoculated August 31 as a supplementary test, -two, Nos. 527 and 528, inoculated with feces of cow 56 became tuberculous. No other guinea pigs inoculated in this series of tests showed any evidence of tuberculosis at autopsy. Cow 56 was retested August 16, 1910, by inoculating two guinea pigs, 778 and 779, subcutaneously. One of these guinea pigs, 779, was wholly negative, the other, 778, became tuberculous and the 380 BULLETIN No. 149 [February, TABLE 17 ANIMAL INOCULATION TESTS FOR TUBERCLE BACILLI IN THE FECES OF REACTING Cows 6 fe o O JH 3. . i X! Guinea pig No. Date of inocula- tion Mode of death Days after 11OCU- Nation Autopsy Micro- scopic Cultures Reinocu- lation 10 1 252 2-4- '09 Killed 52 Normal 253 2-4-'09 Killed 52 Normal 2 512 8-9- '09 Killed 40 Normal 513 8-9- '09 Killed 40 Normal 3 640* 9-22- '09 Killed 14 Normal 641* 9-22- '09 Killed 14 Normal 12 1 254 2-4- '09 Killed 52 Normal 255 2-4- '09 Killed 52 Normal 13 1 78 8-19- '08 Killed 61 Normal 79 8-19- '08 Killed 61 Normal 80 8-19- '08 Killed 61 Normal 2 116 9-2- '08 Died 2 Acute infection 117 9-2- '08 Killed 62 Normal 118 9-2- '08 Killed 62 Normal 3 173 12-28- ; 08 Killed 54 Normal 174 12-28- '08 Killed 55 Normal 15 1 187 12-29- '08 Killed 54 Normal 188 12-29- '08 Killed 54 Normal 16 1 203 12-31- '08 Killed 53 Normal 204 12-31- '08 Killed 53 Normal 2 492 8-7- '09 Killed 39 Normal 493 8- 7- '09 .Killed 39 Normal 19 1 232 2-3- '09 Killed 51 Normal 233 2-3- '09 Killed 51 Normal 28 1 81 8-19- '08 Died 2 Acute infection 82 8-1 9- '08 Killed 59 Normal 83 8-19- '08 Killed 59 Normal 2 119 9-2- '08 Killed 62 Normal 120 9-2- '08 Killed 62 Normal 121 9-2- '08 Killed 62 Normal 29 1 663 10-30- '09 Died 1 Acute infection 664 10-30- '09 Died 1 Acute infection 2 680 11-13--09 Killed 51 Normal 681 11-13- '(J9 Died 6 Acute infection 33 1 205 12-31- '08 Killed 53 Normal 206 12-31- '08 Killed 53 Normal 2 498 8- 7- '09 Killed 41 Normal 499 8- 7- '09 Killed 41 Normal 35 1 682 11-13-'09 Died 12 Actite organisms organisms infection not acid fast not acid fast 681 11-13- '09 Killed 51 Normal 36 1 207 12-31- '08 Killed 53 Abscess in lung organisms organisms (no tubercles) not acid fast not acid fast 208 12-31- '08 Killed 53 Normal 2 502 8-7- '09 Killed 42 Normal 503 8-7- '09 Killed 42 Normal 37 1 177 12-29- '08 Killed 53 Normal 178 12-29- '08 Killed 53 Normal 2 642* 9-22- '09 Killed 14 Normal 643* 9-22- '09 Killed 14 Normal *Killed by a dog. TUBERCULOSIS OF FARM ANIMALS TABI,E 17 Continued 381 u V Guinea pig No. Date of inocula- tion Mode of death after inocu- lation Autopsj" Micro- scopic Cultures Reinocu- lation 39 1 181 12-29- '08 Killed 54 Normal 182 12-29- '08 Killed 54 Normal 41 1 209 12-31-'08 Killed 53 Normal 210 12-31- '08 Killed 53 Normal 2 506 8-9- '09 Killed 40 Normal 507 8-9- '09 Killed 40 Normal 42 1 665 10-30- '09 Died 1 Acute infection 666 10-30- '09 Killed 64 Normal 47 1 .37 9-15- '09 Died 50 Unknown Organisms cause not acid fast (no tubercles) 138 9-15-'08 Killed 107 Normal 139 9-15- '08 Died 57 Unknown cause (no tubercles) Organisms 49 1 211 12-31-'08 Killed 53 Normal not acid fast 212 12-31- '08 Killed 53 Normal 2 496 8-7- '09 Killed 49 Normal 497 8-7- '09 Killed 49 Normal 51 1 508 8-9- '09 Killed 40 Normal 509 8-9- '09 Killed 40 Normal 52 1 256 2-4- '09 Killed 52 Normal 257 2-4-'09 Killed 52 Normal 2 486 8-6- '09 Killed 39 Normal 487 8-6-'09 Killed 39 Normal 3 620 9-18- '09 Killed 60 Normal 53 1 84 8-19- '08 Killed 59 Normal 85 8,19- '08 Killed 59 Normal 2 86 8-19- '08 Died 3 Acute infection 122 9-2- '08 Killed 63 Normal 3 123 9-2- '08 Killed 63 Normal 124 9-2- '08 Killed 63 Normal 54 1 185 12-29- '08 Killed 54 Normal 186 12-29-'08 Killed 54 Normal 2 480 8-6- '09 Killed 39 Normal 481 8-6- '09 Killed 39 Normal 1 3 621 8-18- '09 Killed 60 Normal 622 8-18- '09 Killed 60 Normal 623 8-18- '09 Killed 60 Normal 55 1 684 11-13-'09 Died 5 Acute infection 685 11-13-'09 Killed 51 Normal 56 1 213 12-31-'08 Killed 53 Normal 12-31-'08 Killed 53 Normal 2 520 8-9- '09 Killed 44 Normal 521 8-9- 'Q9 Killed 44 Normal 3 527 8-31- '09 Killed 28 Tuberculosis -)- + -} j_ (superior inguinal) 528 8-31 -'09 Killed 28 Tuberculosis _j- ... (superior inguinal) 4 652 10-4- '09 Killed 53 Normal 653 10-4- '09 Killed 53 Normal BULLETIN No. 149 TABLE 17 Continued [February, * 6 s* V "a. . rt H CO rt 6 33 O & Date of inocula- tion Mode of death Days after inocu- lation Autopsy Micro- scopic Cultures Reinocu- lation 56 5 678 11-13-'09 Died 9 Acute Organisms Organisms infection not acid fast not acid fast 679 11-13--09 Killed 51 Normal 6 778 8-16- '10 Killed 80 Slight + + ++ tuberculosis 779 8-16- '10 Killed 143 Normal 57 1 258 2-4- '09 Killed 52 Normal 259 2-4- '09 Killed 52 Normal 2 484 8-6- '09 Killed 39 Normal 485 8-6- '09 Killed 39 Normal 3 644 9-22- '09 Killed 57 Normal 645 9-22- '09 Killed 57 Normal 58 1 260 2-4- '09 Killed 52 Normal 261 2-4- '09 Killed 52 Normal 2 518 8-9- '09 Killed 44 Normal 519 8-9- '09 Killed 44 Normal 59 1 262 2-4- '09 Killed 52 Normal 263 2-4- '09 Killed 52 Normal 2 524 8-9- 'C9 Killed 44 Normal 525 8-9- '09 Killed 44 Normal 60 1 234 2-3-'09 Killed 51 Normal 235 2-3- '09 Killed 51 Normal 2 500 8- 7- '09 Killed 41 Normal 501 8- 7- '09 Killed 41 Normal 63 1 514 8-9- '09 Killed 44 Normal 515 8-9- '09 Killed 44 Normal 64 1 215 12-31- '08 Killed 53 Normal 216 12-31- '08 Killed 53 Normal 2 478 8-6- '09 Killed 39 Normal 479 8-6- '09 Killed 39 Normal - 3 531 8-31- '09 Killed 28 Normal 532 8-31- '09 Killed 28 Normal 66 1 217 12-31- '08 Killed 53 Normal \ 218 12-31-'09 Killed 53 Normal 2 488 8-6- '09 Killed 39 Normal 489 8-6-'09 Killed 39 Ulcer at point Organisms Organisms of inoculation not acid fast not acid fast (no tubercles) 67 1 264 2-4- '09 Killed 52 Normal 265 2-4- '09 Killed 52 Normal 2 490 8-6- '09 Killed 38 Normal 491 8-6- '09 Killed 38 Normal 70 1 516 8-9- '09 Killed 44 Normal 517 8-9- '09 Killed 44 Normal 71 1 189 12-29- '08 Killed 54 Normal 12-29-'08 Killed 54 Normal 2 482 8-6- '09 Killed 39 Normal 483 8-6-'09 Killed 39 Normal 72 1 183 12-29 '08 Killed 54 Normal * 184 12-29- '08 Killed 54 Normal 2 476 8-6-'09 Killed 39 Normal 477 8-6- '09 Died 10 Acute Organisms infection net acid fast (no tubercles) 1911} TUBERCULOSIS OF FARM ANIMALS TABL.E 1". Continued 383 !* 0) "H. x I* 02 Guinea pig- No. Date of inocula- tion Mode of death Days after inocu- lation Autopsy Micro- scopic Cultures Reinocu lation 72 3 529 8-31--09 Killed 28 Normal 530 8-31--09 Died 2 Acute infection 4 547 9-4-'09 Killed 39 Normal 548 9-4- '09 Killed 39 Normal 73 1 522 8-9- '09 Killed 44 Normal. 523 8- 9- '09 Killed 44 Normal 76 1 219 12-31-'08 Killed 53 Normal 220 12-31- 'OS Killed 53 Normal 77 1 169 12-28- '08 Killed 54 Normal 170 12-28- '08 Killed 54 Normal 78 1 624 9- 18- '09 Killed 60 Normal 625 9-18- ; 09 Killed 60 Normal 79 1 171 12-28- '08 Killed 54 Normal 172 12-28- '08 Killed 54 Normal 80 1 266 2-4- '09 Killed 52 Normal 267 2-4- '09 Killed 52 Normal 83 1 268 2-4- '09 Killed 52 Gen. tuber. + + ++ 269 2-4- '09 Killed 52 Gen. tuber. + + 89 1 510 8-9-'09 Killed 40 Normal 511 8-9- '09 Killed 40 Normal 95 1 179 12-29- '08 Killed 54 Normal No acid- No acid- 180 12-29-'08 Killed 24 Tumor on liver fast organ- fast organ (No tubercles.) isms isms 104 1 236 2-3- '09 Killed 51 Normal 237 2-3- '09 Killed 51 Normal 109 1 494 8-7- '09 Killed 39 Normal 495 8-7-'09 Killed 39 Normal 2 667 10-30- '09 Killed 64 Normal 668 10-30- '09 Died 1 Acute infection 1002 1 221 12-31- '08 Killed 53 Normal 222 12-31- '08 Killed 53 Normal 1004 1 223 12-31--08 Killed 53 Gen. tuber. + + + 224 12-31-'08 Killed S3 Gen. tuber. + + + 1007 1 238 2-3- '09 Killed 51 Normal 239 2-3- '09 Killed 51 Normal 1008 1 240 2-3-'09 Killed 51 Normal 241 2-3- '09 Killed 51 Normal 1014 1 242 2-3- '09 Killed 50 Normal 243 2-3- '09 Killed 50 Normal 1017 1 244 2-3- '09 Killed 51 Normal 245 2-3- '09 Killed 51 Normal J.H. 1 246 2-3- '09 Killed 51 Normal 247 2-3- '09 Killed 51 Normal >. H. 1 671 10-30- '09 Died 1 Acute infection 672 10-30- '09 Died 2 Acute infection 2 676 11-13-'09 Died 9 Acute infection 677 11-13-'09 Killed 51 Normal 384 BULLETIN No. 149 TABLE 17 Continued [February, M 0^ c 2-0 s 3D Guinea pig No. Date of inocula- tion Mode of death Da3's after inocu- lation Autopsj* Micro- scopic Cultures Reinocu- lation L.H. 1 669 10-30- '09 Died 1 Acute infection 670 10-30- '09 Killed 64 Normal L. B. 1 686 11-15-'09 Died 13 Acute infection 687 11-15-'09 Died 13 Acute infection 2 710 12-1- '09 Killed 33 Normal 711 12-1-'09 Killed 33 Normal W.E 1 688 11-15-'09 Died 13 Acute infection 689 11-15-'09 Died 13 Acute infection 2 708 12-1- '09 Killed 33 Normal 709 12-1- '09 Killed 33 Normal S. 6. 1 690 11-15-'09 Died 3 Acute infection 691 11-15- '09 Died 3 Acute infection 2 702 11-25- '09 Killed 39 Normal 703 ll-25-'09 Killed 39 Normal F. S. 1 692 11-16-'09 Killed 49 Normal 693 11-15-'09 Killed 49 Normal O.S. 1 694 11-15-'09 Died 6 Acute infection 695 11-15- '09 Killed 49 Normal 0. C. 1 698 11-11-'09 Killed 49 Normal 699 11-15-'09 Died 4 Acute infection B. 1 696 11-15- '09 Killed 49 Normal 697 11-15-'09 Killed 49 Normal C.E. 1 700 11-15- '09 Died 3 Acute infection 701 11-15-'09 Killed 49 Normal lesions gave staining, culture and re inoculation tests characteristic for tubercle bacilli. There has been at irregular times a number of other tests for tubercle bacilli in the feces of these same and of other cattle; included in these were the tests of eight beef cattle. Tubercle bacilli were not found in any sample in these tests. The results of all the tests are shown in Table 17. Altogether, ninety- seven samples from sixty-two cattle have been tested. (For Au- topsy records of these cattle see Table 19.) In the feces of three of these, or (practically) five percent, virulent tubercle bacilli were present at the time, as shown by guinea pig inoculation. There is given in Table 18 a summary and com- parison of the microscopic examination and ani- mal inoculation test for tubercle bacilli in the feces of reacting cows. From this summary there seems to be no relation between the presence of acid-fast organisms found on a microscopic slide and the presence of viru- COMPARISON WITH THE MICROSCOPIC FINDINGS TUBERCULOSIS OF FARM ANIMALS 385 lent tubercle bacilli as proved by guinea pig- inoculation. Our re- sults are very similar to those of Reynolds and Beebe 5 and of Peters and Emerson 4 . They are unlike the results of Schroeder and Cot- ton* and of the Royal Commission on Tuberculosis of England.* The latter two found tubercle bacilli in the feces in a much higher percentage of tests. The percentage of positive cases that Schroe- der and Cotton found was based upon microscopic examination. However they demonstrated by guinea pig inoculation that cattle do pass virulent tubercle bacilli in their feces. The six cattle tested by the Royal Commission and possibly also, those used by Schroe- der and Cotton are not comparable in respect to the extent of in- fection to the large herds used in the experiments of Reynolds and Beebe, Peters and Emerson, and ourselves. We take it that the latter three herds are much more representative of the tuberculous cattle thruout this country. More work must be done before final conclusions are reached. Whether the percentage of cattle passing virulent tubercle ba- cilli in the feces is relatively large or relatively small, would ap- pear to depend upon the general condition of the herd and the stage of 'the disease in its individual members. Agricultural Experiment Station herds are probably above the average in the general con- dition of the animals and in freedom from emaciated and visibly diseased individuals. On the other hand, certain selected herds TABLE 18. SUMMARY AND COMPARISON OF THE MICROSCOPIC EXAMINATION AND ANIMAL INOCULATION TESTS FOR TUBERCLE BACILLI IN THE FECES OF REACTING Cows 6 2 o O Results with microscope Results with guinea pig Cow No. Results with microscope Results with guinea pig Posi- tive Nega- tive Posi- 1 tive Nega- tive Posi- tive Nega- tive Posi- tive Nega- tive 10 12 13 15 16 19 28 29 33 35 36 37 39 41 42 47 49 51 52 53 1 1 5 1 1 1 7 1 1 1 2 1 1 1 1 3 2 1 2 3 4 1 3 1 2 1 - 1 4 1 1 2 3 1 3 1 2 1 3 2 1 1 2 2 1 2 1 1 2 1 3 2 54 55 56 57 58 59 60 61 62 63 64 66 67 70 71 72 73 76 77 78 2 2 1 1 1 1 1 1 1 1 2 1 1 1 1 2 2 2 1 1 ^1 2 3 1 4 3 2 2 2 1 3 2 2 1 2 4 1 1 1 1 1 o 1 386 BULLETIN No. 149 [February, TABLE 18 Continued 6 Z O Results with microscope Results with animal test Cow No. Results with microscope Results with animal test Posi- tive Nega- tive Posi- tive Nega- tive Posi- tive Nega- tive Posi- tive Nega- tive 79 80 83 86 87 89 95 97 98 104 109 1002 1004 1007 1008 1 1 1 1014 1017 Grade heifer Small heifer Large heifer Lady Britton Wild Eyes Steer 650 Flower Strathmore Oxford 2d steer Oxford 2d cow Bull Columbia 11 (steer) Total 1 1 1 1 1 2 2 1 1 1 1 1 1 1 2 1 2 1 1 1 2 2 2 2 2 1 1 1 1 6 1 1 i i i' i 6 1 1 2 1 1 1 52 62 4 96 TABLE 19. POST-MORTEM INSPECTION OF REACTING CATTLE Cow No. Location of tuberculous lesions (a) Cattle slaughtered and inspected at Urbana* 10 28 36 37 42 47 49 52 53 54 56 57 58 77 86 87 97 Bronchial lymph glands, lungs, liver, ovary, mesenteric lymph glands : generalized tuberculosis : condemned. Bronchial and mediastinal lymph glands. Lungs, mesenteric lymph glands. Lungs, mediastinal lymph glands. Lungs, bronchial lymph glands. Heart, intestine, liver, and mesenteric lymph glands : generalized tuberculosis : condemned. Lungs, bronchial lymph glands. Lungs. Retro-pharyngeal and inguinal lymph glands. Liver, mediastinal and bronchial lymph glands. Lungs, liver, bronchial and mesenteric lymph glands : condemned. Liver, guttural, mediastinal, and bronchial lymph glands : con- demned. Large omentum, liver, guttural, mediastinal, and bronchial lymph glands : condemned. Bronchial lymph glands. Bronchial lymph glands. Lungs, pleura, diaphragm, liver, mediastinal and bronchial l3 r mph glands : generalized : condemned. Lungs, bronchial lymph glands : generalized : condemned. *We are indebted to the Dairy Department, University of Illinois, for the opportunity to make the post-mortem inspections upon the cattle slaugh- tered at Urbana, and for the records of inspection of the cattle slaughtered at Chicago, except for the last eight animals, the records of which have been furnished by the Animal Husbandry Department. TUBERCULOSIS OF FARM ANIMALS 387 TABLE 19. Continued Cow No. Location of Tuberculous lesions (b) Cattle slaughtered and inspected at Chicago.* 12 13 15 16 19 29. 33 39 41 51 59 60 61 64 66 67 71 72 76 79 80 83 95 104 109 1002 1004 1007 1008 1017 G.H. L.H. S.H. L.B. W.E. S.6. F. S. O. S. O. C. B. C. S. Liver, mediastinal lymph glands. Nothing found. Mesenteric lymph glands. Lungs, mediastinal, bronchial, and meseiiteric lymph glands. Mediastinal lymph glands. Nothing found. Nothing found. Bronchial lymph glands. Retro-pharyngeal and mesenteric lymph glands. Lungs, retro-pharyngeal and mesenteric lymph glands : con- demned. Parotid and mediastinal lymph glands. Bronchial lymph glands. Nothing found. Retro-pharyngeal lymph glands. Lungs, liver, bronchial and mediastinal lymph glands. Nothing found. Mediastinal lymph glands. Mesenteric lymph glands. Mediastinal lymph glands. Nothing found. Bronchial and mediastinal lymph glands. Lungs, bronchial and mediastinal lymph glands. Bronchial lymph glands. Retro-pharyngeal and mesenteric lymph glands. Non-tuberculous abdominal abscess. Lungs, bronchial and mediastinal lymph glands. Lungs, bronchial and mediastinal lymph glands. Bronchial lymph glands. Bronchial and mediastinal lymph glands. Lungs, bronchial and mediastinal lymph glands. Nothing found. Nothing found. Bronchial lymph glands. Nothing found. Nothing found. Nothing^found. Nothing found. Nothing found. Nothing found. Nothing found. Nothing found. *We are indebted to the Dairy Department, University of Illinois, for the opportunity to make the post-mortem inspections upon the cattle slaughtered at Urbana, and for the records of inspection of the cattle slaughtered at Chicago, except for the last eight animals, the records of which have been furnished by the Animal Husbandry Department. may be much worse than the average. It is important to point out that even in the well-fed and apparently healthy herds, which have reacted to tuberculin, some animals do pass tubercle bacilli in the feces. That the general percentage is as high as indicated by the 388 BULLETIN No. 149 [February, work of some authors does not seem warranted when \ve consider the list of all the naturally tuberculous cows that have been proved by guinea pig inoculation to be passing virulent tubercle bacilli in their feces. This list comprises sixteen animals as follows : SCHROEDER AND COTTON, 1. Cow No. 113 Bad advanced tuberculosis. Still alive. 2. Cow No. 373 Generalized tuberculosis at autopsy. 3. Cow No. i Generalized tuberculosis at autopsy 4. Cow No. 552 Still alive in good condition. REYNOLDS AND BEEBE, 5. Cow No. 2 l /2 Generalized tuberculosis at autopsy. PETERS AND EMERSON, 6. Cow No. 102 No autopsy. The cow was poor but the symptoms of the disease were not marked. The number of acid-fast bacilli in the feces as revealed by the microscope was exceedingly large. In every field a number of them could be located, and in one place an area of degenerating tissue with many bacilli clinging to it was found. 7. Cow No. 20 Showed no sign of tuberculosis. Her feces contained fairly large numbers of acid-fast organisms. No autopsy. 8. Bull No. 204 Is of special interest because of his exceptionally fine appearance. This was a large Aberdeen Angus bull, ^remarkably fat and sleek. However, he had a persistent but moderate cough. No autopsy. ROYAL COMMISSION ON TUBERCULOSIS, ENGLAND, 9. Cow B Autopsy showed bad generalized tuberculosis. 10. Cow C Autopsy showed bad generalized tuberculosis. 11. Cow A Autopsy showed tuberculosis of intestine and lungs. 12. Cow D When alive, fat and in good condition. Tuberculosis could be detected only by tuberculin test. Autopsy showed large tubercles in the lungs and bronchi. 13. Cow F Autopsy showed severe generalized tuberculosis, apparently originating in the alimentary tract. NEW CASES IN THIS REPORT. 14. Cow 83 Autopsy showed generalized tuberculosis. 14. Cow 1004 Autopsy showed tubercles in the lungs and bronchial lymph glands. 16. Cow 56 Autopsy showed tubercles in the lungs, liver, and mesenteric, and bronchial lymph glands. It should be noted that of these 16 cows, that have been shown by guinea pig inoculation to be passing virulent tubercle bacilli in their feces, the eleven slaughtered were very badly tuberculous. Every one of the eleven showed abscesses of the lungs; seven, ulcers of the intestines ; and six, tuberculosis of the liver. Every one of these surely should have been condemned at slaughter. Of the animals not examined post mortem, two were in advanced stage of tuberculosis and three in apparently good condition. The most important fact to be noted here is that a cow may be in an apparently good condition and yet be passing tubercle bacilli in her feces which will produce tuberculosis in guinea pigs. It appears TUBERCULOSIS OF FARM ANIMALS 389 also that under natural conditions a cow does not pass virulent tubercle bacilli in the feces unless her lungs, intestines, or liver are affected with open tuberculosis. But this can - not be ascertained from the outward appearance of the animal. CONCLUSIONS 1. Microscopic tests show acid-fast micro-organisms in the feces of non-reacting as well as of cattle reacting to the tuberculin test. 2. Microscopic tests can not, in all cases, be depended upon to reveal virulent tubercle bacilli when they are present. 3. By guinea pig inoculations, tubercle bacilli in a suspension of fresh cow feces can be detected when present in an amount as small as one hundred billionth of a gram in the dose inoculated. 4. Reacting cows in an apparently good state of nutrition may be passing virulent tubercle bacilli in such quantities that one cubic centimeter of a two percent emulsion of their feces will produce generalized tuberculosis in guinea pigs. 5. Under ordinary conditions only a small percentage of tuber- culous cattle are excreting tubercle bacilli in their feces at any given time. 6. At this Station, ninety-seven samples of feces from sixty- two tuberculous cattle have been tested for tubercle bacilli by in- oculation of guinea pigs. Tubercle bacilli were found four times in samples from three animals. REFERENCES 1. Bang 1 , B., Measures against tuberculosis in Denmark. Sixth Internat. Cong, on Tuberculosis. 4. part 2 : 863. 1908. 2. Calmette et Guerin, Sur 1'evacuation de bacilles tuberculeux par la bile dans 1'intestin chez les animaux porteurs de lesions latentes ou occnltes. Comp. rend, de 1'Acad. des Sci. 148: 601. 1909. 3. MacNeal, Latzer and Kerr, The fecal bacteria of healthy men Jour, of Infect. Dis. 6: 123-169. 1909. 4. Peters, A. T. and Emerson, C., Dissemination of tuberculosis by the manure of infected cattle. Nebraska Agr. Exp. Sta. 22d. Ann. Rep.: 136-142. 1909. 5. Reynolds, M. H., and Beebe, W. L., Dissemination of tubercu- losis by the manure of infected cattle. Univ. of Minn. Agr. Exp. Sta. Bull. 103. 1907. 6. Royal Commission on Tuberculosis of England, 3d. Interim Re- port. Ref. Jour, for Comp. Path, and Ther. 22: 77. 1909. 7. Schroeder, E. C., The unsuspected but dangerously tuberculous cow. U. S. Dept. Agr., Bureau of Animal Industry, Cir. 118, 1907. 390 BULLETIN No. 149 [February, 8. Schroeder, E. C., The occurrence and significance of tubercle bacilli in the feces of tuberculous cattle. Sixth. Internat. Cong, on Tuber. 4, part 2 : 599-606. 1908. 9. Schroeder, E. C. and Cotton, W. E., The danger from tubercle bacilli in the environment of tuberculous cattle. U. S. Dept. Agr.. Bureau of Animal Industry Bull. 99. 1907. 10. Schroeder, E. C. and Mohler, J. R., The tuberculin test of hogs and some methods of their infection with tuberculosis. U. S. Dept. Agr., Bureau of Animal Industry Bull. 88. 1906. 11. Strasburger, J., (I.) Ein Verandertes Sedimentirungsverfaliren zum mikroscopischen Nachweis von Bacterien. (II.) Ueber den Nachweis von Tuberkelbacillen in den Feces. Munchener. med. Wchnschr. 47: 533-535- 1900. 12. Trueman, John M., Milk supply of Chicago and twenty-six other cities. University of Illinois Agr. Exp. Sta. Bull. 120. 1907. TUBERCLE BACILLI IN BLOOD OF TUBERCULOUS ANIMALS LITERATURE In the Tables 20, 21 and 22 giving the literature upon this sub- ject two methods of determining the presence or absence of tu- bercle bacilli in the blood are contrasted, that of bacterioscopy and that of experimental inoculation. Calmette 11 has briefly summarized the bacterioscopic methods as follows: "Bacterioscopy is the direct search for the tubercle bacillus in the juices of the organism or in the blood. It must be regarded as an exceptional method, because, even in the case of sero-pleural effusions, the bacilli are discovered at the most in two percent of the cases (Netter). It is so persistently negative in the case of blood that there is nothing gained by using it for the study of that fluid. The ingenious method of indirect bacterioscopy, known, since the work of Jousset, as inoscopy, in which the blood is cen- trifugated after the fibrin has been artificially digested or rendered incoagulable by means of various procedures (washing \vith soda, after Bezangon, Griffon and Philibert; the leech method of Le- sieur), is so delicate in its application, and fraught \vith so many errors on account of the presence of other acid-fast, non-tubercu- lous bacilli, that its results cannot be interpreted with sufficient ac- curacy for clinical purposes. The same is true of other methods of bacterioscopy. based on a previous hemolysis of the red blood-cells (alcohol-hemolysis of Loeper and Louste, hydrohemolysis of Nat- tan-Larrier and Bergeron). At the beginning of a tuberculous in- fection the bacilli are so few in number in the blood and in serous exudates that there is no possibility of finding them in these fluids." TUBERCULOSIS OF FARM ANIMALS 391 Rosenberger has since devised a modification of the direct bac- terioscopic method, the essential features of which are sedimenta- tion of citrated blood, smearing the sediment in thick layers on slides and laking the. hemoglobin with distilled water before stain- ing for tubercle bacilli. Uhlenhuth and Xylander 47 have employed "antiformin", a mixture of chlorinated soda and caustic soda, to dissolve the blood clot, and have examined the sediment of the resulting solution for tubercle bacilli, a method very similar in principle to that of Jousset. Schnitter has first laked the freshly drawn blood by mixing it with three percent acetic acid or with three percent citric acid. The washed sediment of this mixture was then digested with fifteen percent antiformin solution. The remaining sediment, very minute- in quantity, was washed with water and then stained and examined for tubercle bacilli. None of these methods of bacterioscopy is so delicate as that of animal in- oculation. By this method some of the suspected material either without special preparation or after treatment by some of the various methods given above is injected into test animals. Table 21 ''Tubercle bacilli in blood of animals injected with pure culture" shows that some of the results of each author quoted are positive by the animal inoculation test. Of the three authors giving results by both methods, Anderson has wholly negative re- sults by the microscopic test, while his results by the inoculation test are positive in sixty-six and two-thirds percent of the cases. Anderson, also, obtained positive culture results by inoculating glycerine potato medium with the blood of these same animals. TABLE 20. TUBERCLE BACILLI IN THE BLOOD OP MAN Microscopic test Animal inoculation test Author ri t o> M oo u > a A O Method m rf '5 O 05 3 o Remark O V O 0) * a. PH 1 1884 Direct smear 4 4 100 Acute cases -> 1884 Direct smear Chronic cases from dead body 3 1884 Direct smear 8 8 100 Acute cases f roni dead body \ 1884 Direct smear 1 1 100 Blood taken during life e. 1884 Direct smear 9 4- 9 After long search 6 Rutime3-er 1 ". . . . 1885 Direct smear 1 1 100 Patient just dead Sticker 44 188=; Direct smear 3 3 100 8 Sticker 44 188S Direct smear 1 Made 80 prep- arations 392 BULLETIN No. 149 [February, TABLE 20. Continued 6 a Author V "rt O Microscopic test Animal inoculation test Method te V to 3 Q Positive Percent o r. a O Positive Percent Remark 1C 11 12 13 14 IS 1(> 17 is 19 20 21 22 23 24 25 26 27 2s 2-> 30 31 32 33 34 35 36 37 38 39 Bezancon 5 Griffon and Philibert 1903 1904 1904 1904 1904 1905 1906 1908 1909 1909 1909 1909 1909 1909 19C9 1909 1909 1909 1909 1909 1909 1909 1909 1909 1909 1909 1909 1910 1910 1910 1910 Homogeni- sation Inoscopy Leech method Alcohol- hemolysis Rosenberger Rosenberger Rosenberger Rosenberger Rosenberger 7 Rosenberger Rosenberger Rosenberger 7 Rosenberger Rosenberger Staubli technic and Antiformiii Rosenberger Aiitiformiii 7 Rosenberger Rosenberger Rosenberger Rosenberger 2 2 100 35 26 35 30 11 1 3 5 31 4 14.3 16.7 Animal experi- ment Guineapig test Guinea pig test Guineapig test Guineapig test Suinea pig test Two cases pos- itive by guin- ea pig test [n the positive one, guinea pig inocula- tion failed Preliminary report Guineapig test Guineapig test Guineapig test Animal tests Guineapig test Guineapig test Bergeron 8 26 Garv 17 . . T V S2 L/esieur 30 3 300 12 20 7 8 7 10 48 18 10 17 10 7 38 22 25 7 18 56 7 51 10 33.3 Loeper and 25 L,udke 26 3 1 1 3 1 6 100 100 7 Huguenin 19 .. .. Rosenberger 38 . . . Forsyth 1 4 300 10 1 100 83 5 Rosenberg 37 Webb 49 Stoll 45 Miller 30 Brown 8 13 10 48 50 20 40 Bernstein 4 and Fried 1 9 12 11 14 5* 2 90 21 44 78 9 Anderson 1 Ljebermeister 23 .. Ravenel and 86 Smith Burnham and 9 L/yons 17 Dailey 13 Petty and 35 Mendenhall Brem 7 Schnitter 42 Hewatt and 1 8 Sutherland .... L/ippmann 24 Tausig 46 ? Foster 15 Buryille- 1 Holmes 37 Clifford 12 White and 51 Averj* 47 *The author considers at least four of these five positive results to have been due to acid-fast bacilli in the distilled water used. I9H} TUBERCULOSIS OF FARM ANIMALS 393 TABLE; 21. TUBERCLE BACILLI IN BLOOD OF ANIMALS INJECTED WITH PURE CULTURE OK B. TUBERCULOSIS 6 fc 1 2 3 4 5 (> 7 8 9 10 11 12 13 14 Author 1) 4-1 cd Q Animal Test posi- tive after injection Microscopic test Animal inoculation test Method to -*j rt Q Microscopic test Animal inoculation test Method t ~ ^ M-l O Jj jj O fc/j r^ . f-4 ^^ u O d) ~*~* MH "^* ^f cc ^ ft o <4-t if D <^ 4-> jj J cd f^ tf) 3 Autopsy c.i a^l o 6 C 3 /5 o S H >%0 ~TH "^ d Hastings" reported the testing in Massa- chusetts of 24,685 cattle of which 12,443 or 50 percent reacted to tuberculin. In 1905, Russell' 3 reported work done in 1903 and 1904, in which 21 herds in Wisconsin having wide spread infection were tested. These 21 herds aggregated 933 cows of which 518 or 55 percent were tuberculous. In 1901, Pearson, and Ravenel 11 tested twelve herds having a total of 599 cattle, 484 or 80 percent reacting. In 1908, Ward and Raring 1 ' reported the testing of 22 herds in California near San Francisco totaling 1022 cows of which 326 or 31.9 percent reacted. In 1908, -Moore 8 compiled figures obtained from work done under the direction of the Commissioners of Agriculture of Xew York State and of private veterinarians, and found 32.26 percent of the cows reacting to the tuberculin test. DISTRIBUTION Denmark. Bang" reported the tuberculin testing IN FOREIGN from 1898 to 1904 of 40,624 head of cattle. Of COUNTRIES the calves under six months 12.1 percent were tuberculous; of the yearlings (6 to 18 months) 27.5 percent; of cows i^ to 2^/2 years, 38.6 percent; of those 2 /^ to 5 years, 44.9 percent ; and of those over 5 years old 48.0 percent were tuberculous. Belgium. Heymans" said that there are about 2,000,000 cattle in Belgium of which about 1,000,000 are milk cows. About 40 percent of the cows and 10 percent of the other cattle or an average of 25 percent of all cattle have tuberculosis. Germany. Ostertag 10 in 1907, recorded the testing of 215,837 dairy cattle in Saxony of which 115,100 or 43.6 per- sent reacted to tuberculin. Xocard 9 estimated that 25 percent /pi/] TUBERCULOSIS OF FARM ANIMALS 409 of all dairy cattle in the regions of Brie and Beance in France were tuberculous, while some other regions were free from this disease. The Royal Veterinary College of England 12 in 1900, re- ported the testing of 20,930 animals of which 5441 or 26 percent reacted. There are some parts of the world, as for example, Japan, the northern parts of Norway and Sweden, the Steppes of Russia, the Island of Jersey, and parts of Africa and South America where bovine tuberculosis is entirely absent or very rare. 15 These are places where the importation of cattle has been small in amount and this disease has not yet been disseminated. FACTORS INFLUENCING DISTRIBUTION The factors that are continuing to extend tuber- QF CATTLE culosis among farm animals are extensive trading in cattle on the one hand and the subtleness of the disease on the other hand. In order to improve the herds thorough- bred stock has been imported and incidentally tuberculosis has been introduced with it. Salmon 10 says, "Unfortunately, the breeders of Great Britain were not as skillful in avoiding tuberculosis as they were in increasing the size, perfecting the form, and hastening the maturity of animals, and the result has been not only that they un- wittingly propagated the disease, but that they distributed it in the most e.vtensk'c manner". The improvement and increase in transportation facilities, together with the wider recognition of the value of a few pure bred animals in improving the. quality of a herd, have led to extensive buying and selling of cattle for breeding purposes. This trade in cattle is very essential to the wel- fare of the live stock industries, but it should be clearly recognized that precisely this is the chief mode of introducing tuberculosis into a previously healthy herd. The owner of a healthy herd can- not be too careful in selecting the animals which he expects to add to it. It is the nature of this disease to disguise its presence. The cow with the disease looks well, may breed well, eat well, and continue to give milk. In fact, it may be only in the last stage of the disease, long after she has become dangerous by distributing the germs in her feces, urine', or milk, that physical signs of tu- berculosis appear. Thus it often happens that a whole herd may become infected before the owner recognizes the presence of the disease. STOCK SHOWS Stock shows, state and county fairs are other AND FAIRS means of disseminating the disease. At such a fair where the stock drink from the same watering trough, where in the show ring they may nose each other, a bad tu- 410 BULLETIN No. 149 [February, berculous cow may infect the stock of a number of other farms and the fact may not be discovered for months or years afterward. Stock owners ought to exercise great caution in taking their stock to such shows. No stock should be permitted to be shown un- less the owner has a certificate of a negative tuberculin test for the stock he is entering. Laws were passed in Denmark seventeen years ago to prevent tuberculous cattle and swine from entering stock shows and fairs and these laws have materially lessened tu- berculosis in Denmark. HOGS FOLLOW- ^ ne practice of having hogs follow corn-fed cat- ING CORN- tie is another great means of distributing tuber- FED CATTLE culosis. It has been noted for some years that hogs shipped from the dairy regions, both in this and in foreign countries, have upon slaughtering far more tubercu- losis than hogs from other parts of the country. This is due to the fact that hogs readily take tuberculosis by eating corn out of manure or by drinking the milk of tuberculous cattle. Hogs should never be permitted to follow tuberculous cattle. Calves and hogs should not be fed separator milk from the creameries unless the milk has been pasteurized or boiled. VENTILATION ^ * s we ^ known that living out of doors and get- AND OUT- ting an abundance of pure air tends to check and DOOR LIFE even to arrest tuberculosis in human beings. This may also be true among cattle. But it does not follow from this that if cattle live out of doors entirely, they will be entirely free from tuberculosis. Ward and Haring 18 reported the testing of 1022 cattle in California of which 326 or 31.9 per- cent were tuberculous, while tests in New York State compiled by Moore showed a percentage of 32.26 having this disease. In New York the cattle are more or less housed in barns thruout the year while in California the cattle tested by Ward and Haring were kept almost exclusively out doors in large pastures or ranges and only put in open sheds at milking and feeding time. The herds were otherwise handled much like the Eastern herds, the cattle were associated together at feeding times, and the herds were built up by "buying in" fresh cows. This indicates that the climate is not a very important factor in preventing the spread of tuberculosis in cattle. Whatever the climate, or the breed, or the stock, or the vigor, when tuberculosis is introduced by purchase or otherwise, unless means are taken to arrest its spread, it will finally infect and destroy the whole herd, /pi/] TUBERCULOSIS OF FARM ANIMALS 411 SUMMARY 1. Tuberculosis among farm animals is found extensively thruout the whole world. 2. Our knowledge concerning distribution is as yet very in- complete. 3. The three important means of ascertaining the distribution of tuberculosis among farm animals are thru the reports of lab- oratory tests of dairy products, tuberculin testing of cattle, and government meat inspection. 4. In particular instances the percentage of tuberculous cat- tle are very high, in some states ranging from 32 to 80 percent. 5. The percentage of tuberculous cattle in foreign countries where much more testing has been done than in the United States ranges from 25 to 48 percent. 6. The two things most favoring the distribution of this dis- ease are the extensive trade in (tuberculous) cattle; and the dif- ficulty of recognizing the disease by physical signs until it is far advanced. 7. The tuberculin test should always be applied before adding a newly purchased animal to the herd. REFERENCES 1. Anderson, J. F., The frequency of tubercle bacilli in the market milk of Washington, D. C. U. S. Hyg. Lab. Bull. 41 : 163-192. 1908. Also Jour. Inf. Dis. 5: 107-115. 1908. 2. Bang, B., Measures against animal tuberculosis in Denmark. Sixth Internat. Cong, on Tuberculosis. 4, part 2: 855-867. 1908. 3. Field, C. W., Tubercle bacilli in market milk of Louisville, Ky. Louisville Times, July, 1909. 4. Hess, A. H., The incidence of tubercle bacilli in New York City milk. Jour. Am. Med. Assoc. 52. 1011-1016. 1909. 5. Hughes, D. A., Precautionary sanitary legislation against tu- berculosis of the domesticated animals in the United States. Sixth Internat. Cong, on Tuberculosis, 4, part 2: 971. 1908. 6. Heymans, Discussion. Results accomplished in Belgium to- ward the controlling of tuberculosis of animals. Sixth Internat. Cong, on Tuberculosis 4, part 2 : 522. 1908. 7. Melvin, A. D., The economic importance of tuberculosis of food producing animals. Sixth Internat. Cong, on Tuberculosis 4, part 2 : 501-511. 1908. 8. Moore, V. A., Bovine tuberculosis in New York State. Cited by Ward and Haring. 1908. 9. Nocard, Cited by Salmon. 412 BULLETIN No. 149 [Ffbnutry, 10. Osterag, R., Die Milchwirtschaft mid die Bekampfung der Rindertuberkulose. Zeit. f. Fleisch- und Milch-Hyg. 18: 41-50. 1907. 11. Pearson, L. and Ravenel, M. P., Tuberculosis of cattle and the Pennsylvania plan for its repression. Perm. Dept. of Agr. Bull. 75, 1901. 12. Royal Vet. Col. England, After Salmon, U. S. Dept. of Agr., Bureau of Animal Industry Bull. 38:12. 1906. 13. Russell, H. L., Two ways of treating tuberculosis in herds. Wis. Agr. Exp. Sta. Bull. 126. 1901. 14. Russell and Hastings. Bovine tuberculosis in Wisconsin. Wis. Agr. Exp. Sta. Bull. 84. 1901. 15. Salmon, D. E., Tuberculosis of the food producing animals. U.S. Dept. of Agr., Bureau of Animal Industry Bull. 38: 1906. 1 6. Ward and Haring. The prevalence of tuberculosis among dairy cattle in the vicinity of San Francisco, California. Sixth In- ternat. Cong, on Tuberculosis 4, part 2 : 564-569. 1908. METHODS OF COMBATING ANIMAL TUBERCULOSIS RECOGNITION BY PHYSICAL, MEANS The first essential in combating tuberculosis of farm animals is the recognition of the disease. In general this cannot be done by a physical examination alone. No expert can recognize tuber- culosis, except in the advanced stages, by making a physical ex- amination ever so carefully. In advanced cases or in moderately advanced cases a cough is often present. In tuberculosis of the udder an emaciated quarter or knots in the udder may be noted. In the last stages of the disease, generally only shortly before death, the animal becomes much emaciated and then the condition may be recognized from the external appearance. RECOGNITION BY MEANS OF THE TUBERCULIN .TEST The tuberculin test is by far the best means of recognizing tuberculosis in cattle in its early stages. It is made by first as- certaining the normal temperature of the cow, then injecting a dose of tuberculin under the skin of the neck and then observing the effect upon the temperature produced by the tuberculin in- jected. TUBERCULIN Tuberculin is the product of an extensive growth of B. tuberculosis in an artificial medium. The medium used is six percent glycerine meat broth. This medium is made and sterilized in broad bottomed, cotton stoppered flasks. These flasks are inoculated with a pure culture and are kept at body temperature (37 Centigrade) for some weeks until' a thick /9//] TUBERCULOSIS OF FARM ANIMALS 413 crumpled membranous growth covers the surface of the glycerine broth. The bacteria are killed by heat and then removed by fil- tering thru a fine clay filter. The filtrate is then placed in an evaporating dish and concentrated to one-tenth its original vol- ume. This concentrated filtrate is tuberculin. Tuberculin as just described is prepared by the Bureau of Animal Industry of the United States Department of Agriculture and is diluted with weak (one percent) carbolic acid to eight times its volume before use. In this form, it is more convenient for use than the original thick syrupy liquid. It is sent out free of charge in either the concen- trated or in the diluted form read for use. OBSERVATION i cattle to be tested should be kept in the ^rr -ri_irr -i-c.-i stable, resting quietly for some hours at least be- OF THE TEM- 7 J . PERATURE * ore tne temperature is taken. Disturbing fac- tors such as excessive feeding, drinking large amounts of cold water, or excitement of the animal should be avoided. The temperature is taken with a self -registering physi- cian's clinical thermometer, placed in the rectum of the animal. It is necessary to throw down the column of mercury in such an instrument each time it is used, and the position of the column should be carefully examined each time just before insertion of the thermometer, to be certain that the mercury is down below 99 Fahrenheit. Neglect of this precaution may lead to very seri- ous error. The thermometer is tied with a string around the con- striction and then fastened to the cow with a hair clamp or tied around the tail with a string so that if the thermometer falls it will not strike the floor and break. Before inserting the ther- mometer it should be washed in 5 percent carbolic acid, wiped with a sterile cloth and smeared with vaseline. It is then carefully inserted into the rectum for its whole length, where it is allowed to remain for three to five minutes. Two persons working to- gether, one reading, the other recording the temperatures, can keep in use five or six thermometers at one time. A cow's normal temperature is 101 to 102.5 Fahrenheit; if because of any fever, being in heat, calving, or something else, the temperature is abnormal the tuberculin test should not be made at this time. Such animals should be excluded from the test and should be tested at a subsequent time when the cause of the abnormal temperature is no longer present. A good method of procedure is, on the first day, to determine the normal temperature of the cow by beginning at eight o'clock in the morning and tak- ing her temperature every two hours until six o'clock in the even- ing. The average of these temperatures may be regarded as her normal temperature. Excluding all cows with abnormal tern- 414 BULLETIN No. 149 [February, peratures, inject the tuberculin into the others at 8 p. m., and be- gin again the next morning at six o'clock and observe the tempera- ture at two hour intervals or more frequently thruout the day. By comparison of the temperature on the day following the injec- tion of the tuberculin with the temperature on the previous day the result of the test is determined. For the correct interpretation of the test a man of training, experience and judgment is neces- sary. INJECTING THE ^ or tne m j ect i n of the tuberculin a good syringe TUBERCULIN w ^ tn several sharp stout needles is required. The needles may be sharpened from time to time on an oil stone. The syringe should hold from five to ten cubic cen- timeters and should be graduated in cubic centimeters. It should be sterilized by boiling before use. The injection is made sub- cutaneously in the neck. Here the skin is thin and is easily pulled into a fold. The needle, if sharp, can be pushed in with but little trouble. It should be held almost parallel to the surface of the neck, so as not to enter the muscles. Have the set screw placed so that the exact amount can be injected. For an animal weigh- ing a thousand pounds two cubic centimeters is the usual dose; for animals weighing more or less, a proportionally greater or smaller amount of tuberculin should be used. . s POSITIVE AND Usually in the positive reactions the maximum NEGATIVE temperature on the day following the injection REACTION is from 3 to 5 Fahrenheit above the normal temperature; if the rise is less than 2F. it is to be considered doubtful. Some consider 1.5 F. above the normal individual temperature as determined on the previous day as suf- ficient to indicate a positive reaction. The elevation of tempera- ture should come on gradually, remaining practically at its maxi- mum for a few hours, and gradually subside. A sudden elevation which is not continued is not to be considered a reaction. All doubtful cases should be retested after six to eight weeks. On the opposite page are some examples of temperature observations in typical positive and in typical negative cases. RELIABILITY ^^ positive reactions of the tuberculin test are OF POSITIVE very reliable. The Bureau of Animal Industry REACTIONS reports tests from 1893 to 1908, upon 400,000 cattle; 24,784 of the reacting cattle were slaugh- tered and of these, 24,387 or 98.39 percent were found to be tuberculous. In Pennsylvania 4000 animals that had given char- acteristic reactions were slaughtered and examined, and the pres- ence of the disease was demonstrated in all but eight of these animals, 99.8 percent. Fifty-eight of the cattle used in the ex- TUBERCULOSIS OF FARM ANIMALS 415 Reaction Reaction t> iO 00 J> 00 ON M . if) 10 o rH & to rH rl rH rH cN O S3 $.^ S3 : >-> 0, *-> CM >O kH IO rH 00 CM O 4> O O 00 he 0) <-> M O iO Cl rf Tf Tf rH rH rH O O O O O c M-l cd s^ IO O <-H rH 0000 H l- fsi \O iO O O O o r-( rH O O r- O O O O O O rH rH VH rH rH S5 5 M JJ p-J cd 0) ^ SS88 rH rH rH rH 0) 'O O to ON \O gc 0) iO O O >O c 0) O ^D^ rH f^ rH O rH O O O O O 'C w^ a> o yl S88S 3 o M-1-OrHON P ^ il iO O O I> 's g ^ fNl .-. rH rH O 00 O rH rH rH " rH '. "rt iO *- rH ' DH rH C^ N rH 0000 rH rH rH r- O fcrH CO 00 * M VO CN! t/j H tog O rH rH rH r* O O O O O O w go rH << 3| -M .^ O O Q O O rH ON ON ON 00 rH A <*H O Weight OOO O O ON rH rH rH he VO rH O tO \O rH rH 0) 0) ho O 00 1> O Breed -, 1_ >> W he i> he w he CO * _, VH >^ VH >^ ^ 0) 0, 0> H Q C. C a o c h Breed rH M rH r-l O O a DH B i^* rsi \G f^J -*-> o * og S" ^ >O M M O OO O rH rH rH r-. 5 8^ -*' VO rH rH O O O O * U Cn 4) M 00 I> * CO 3 Hl^ o o So "rt 0) o 'd CO to vo VO B toS ^O IO rH rH O O O O V 00 * rt t^ iO 00 S 800 1 416 BULLETIN No. 149 [February, H W PH O ffi w 3 O O H w 55 w o TO O I9ii\ TUBERCULOSIS OF FARM ANIMALS 417 "n o "I O o o 6 4 W 2 * ^ O W K x S ' n : CU Q r 1 W H O M {H S o Sow 191 1] TUBERCULOSIS OF FARM ANIMALS 419 perimental work of this bulletin that had given positive tuberculin reactions have been slaughtered. In seventeen of them, slaughtered and inspected by us in Urbana, typical lesions of tuberculosis were found in every case. Of the forty-one slaughtered and inspected at Chicago, lesions of tuberculosis were found in only twenty-four. The work of the government inspectors is necessarily done'rapidly, and it is not impossible that slight lesions, unimportant from the standpoint of food value of the carcass, might be overlooked. The tuberculin test is not an absolutely reliable test for the presence or absence of tuberculosis. It is, however, very reliable when positive; not so reliable when negative. It is by far the most reliable means known to determine the presence of tuber- culosis. NATURE OF ^ consideration of the nature of the reaction as THE REACTION explained by Theobald Smith 9 may do much to clear up the confusion regarding the failure to react which sometimes occurs in tuberculous animals. The tubercle bacilli growing in the body induce certain tissue changes and arouse certain new functions of the tissues which are in the nature of a partial immunization. It is the action of the specific product, resulting from these changes, upon the tuberculin that causes the latter to split off a poisonous substance, which poison acting upon the tissues causes the elevation of temperature. Now, when the lesion is healing there is formed about it a fibrous wall which tends to isolate it from the surrounding tissues. This wall prevents the specific product of the tissue changes, induced by the tubercle bacilli, from getting outside it, and thus, not being able to act upon the tuberculin, it fails to split off the poisonous product, and no elevation of temperature occurs. Therefore in a cow with only a few healing lesions a reaction may not occur. But such a lesion may afterward become active and so give rise to a positive reaction at a later time. ' Harding and Smith 4 have described a cow (Millie D.) which had reacted and then failed to react for three years. At her death, they found a few tubercles, no larger than a pea and completely encapsulated. These tubercles, crushed in sterile water and injected into guinea pigs, produced generalized tuberculosis. The period between infection and the time when a tuberculin reaction will first take place is known as the period of incubation. There must be produced enough of the "specific product" formed by the reaction of the tissues which are stimulated by the pres- ence of the tubercle bacilli, in order to react with the tuberculin to produce the fever. Just how long a period of incubation is required is not known. It probably varies considerably in indi- vidual cases from a few weeks to several months. 420 BULLETIN No. 149 [February, An important series of tests described by Moore 8 may serve to illustrate the failure of the tuberculin test to detect all cases of tuberculosis in a herd at a single test. In January, 1904, a herd of 491 cattle was tested with tuberculin. Of these only 96 cows failed to react. The stables were disinfected with five per- cent carbolic acid and the reacting animals all immediately re- moved. The apparently healthy animals were tested every six months afterward and all subsequently reacting animals promptly removed. The non-reacting animals were guarded carefully from reinfection. The result was as follows : 1904 Jan. 491 tested 395 reacted 96 did not react 80.4 percent reacted 1904 July 96 tested 3i reacted 65 did not react 32.2 percent reacted 1905 Jan. 65 tested 8 reacted 57 did not react 12.3 percent reacted 1905 July 57 tested 15 reacted 42 did not react 26.3 percent reacted 1906 Jan. 42 tested 15 reacted 27 did not react 35-7 percent reacted 1906 July 27 tested 3 reacted 24 did not react ii. i percent reacted 1907 Jan. 24 tested 2 reacted 22 did not react 8-3 percent reacted 1907 July 22 tested I reacted 21 did not react 4-5 percent reacted 1908 Jan. 21 tested I reacted 20 did not react 4-7 percent reacted 1908 July 20 tested I reacted 19 did not react 5-0 percent reacted Several of these animals which failed to react at the first test but did react later were examined post mortem. The lesions were slight in extent but of long standing and surrounded by a wall of fibrous tissue. Moore thinks these cows were infected and the lesions healed or partially healed before the first test January, 1904. Afterward the arrested lesion became active. So the os- cillation of the morbid process between an active and an arrested condition is an important factor in the result of the tuberculin test. The cow may fail to react during the early or incubation pe- riod, or at a time when the lesions are healing or partially healed but also the reaction may fail in very advanced cases. While this last is rare and in such cases the disease may usually be rec- ognized by a physical examination, yet it may escape recognition, and such an occurrence is liable to cause very serious trouble. For such an animal is usually passing out enormous numbers of the tubercle bacilli from her body and may quickly infect a large part of the other animals in the herd. Special care should be exercised to detect these very important cases. RECOGNITION BY MICROSCOPIC AND ANIMAL TESTS Special laboratory procedures are useful in detecting the dan- gerously tuberculous cow; to detect the cow that is constantly passing virulent tubercle bacilli in the milk, feces, or other se- cretions in order that she may be removed from the herd. Micro- scopic examination may be useful where a very bad udder is sus- TUBERCULOSIS OF FARM ANIMALS 421 pected or something of this nature; but this test can not be relied upon unless the organisms are found in great numbers. Negative results are practically worthless. The testing of feces by micro- scopic examination and by animal inoculations has already been discussed. Uterine and nasal discharges may be tested by similar methods. In many European countries the testing of milk by guinea pig inoculations in order to discover the cow with tuber- culous udder is a common laboratory practice. Mention of this subject has been made in connection with the report of Mueller's work. (See page 344.) HARPOONING Another means of testing is by harpooning. This is more frequently used to determine the nature of knots or supposed tubercules in the udder, especially, or upon any part of the body. A harpoon, which is a needle with a beard at the tip, is thrust into the suspected swelling, revolved once, and quickly withdrawn. It brings with it, on the hook, a small bit of tissue which can be examined directly with the microscope or inoculated into animals. None of these methods can take the place of the tuberculin test but they are useful as supplements to this test. DISPOSAL, OF THE TUBERCULOUS Cow What shall be done with the tuberculous cow when she has been recognized? If the cow is wasting in body or physical signs of tuberculosis are evident, she should be killed at once and buried, or burned, or put in the rendering tank. When the cow is in ap- parently good or fair condition she may be valuable for beef, or she may be useful for the production of milk and the raising of calves for some years. Slaughter under inspection is usually the most OlAllf^LJTpp ' UNDER IN satisfactory means of disposal unless the cow is SPECTION especially valuable for the production of milk and butter and the raising of calves. To the owners of small herds slaughter under inspection is to be recommended. The preservation in quarantine of the tuberculous portion of the herd according to the Bang method can sometimes be recommended to owners of large herds. . THE BANG METHOD. What is known as the Bang method of handling a tuberculous herd in quarantine, and the building up of a healthy herd from a tuberculous herd was first carried out by Prof. Bernard Bang of the Royal Veterinary College of Copenhagen, Denmark. Bang's experience with the tuberculin test had shown him, that tho ac- 422 BULLETIN No. 149 [February, curately revealing the presence of tuberculosis it gave little knowl- edge of the extent of the disease in the cow, which might often indeed have only a slight infection. His special desire was to devise a system eventually to get rid of the tuberculous cow; to prevent at once further spread of the disease, and to interfere as little as possible with the breeding of cattle. An outline of the method is as follows: 1. Apply the tuberculin test. 2. Separate the herd into reacting and non-reacting herds. 3. Provide for the two herds entirely separate stables, pas- tures, and attendants, if possible separate farms. 4. Thoroly disinfect the stables, watering troughs, and all utensils coming into contact with the healthy herd. 5. Test the well herd every six months, and remove the re- actors and again thoroly disinfect the stables. 6. Take the calves of the tuberculous herd immediately from their mothers and feed them milk from non-reacting cows or milk pasteurized at 85 C. 7. Add no calves to the non-reacting herd except those show- ing a negative reaction to tuberculin, subcutaneously administered, at the age of six months. 8. No cows "bought in" must be added to the well herd un- til thoroly tested with tuberculin. 9. All milk from the tuberculous herd must be pasteurized or sterilized before used as food for man or animals. 10. As soon as an animal in the tuberculous herd arrives at an advanced stage of the disease indicated by physical signs, she should be immediately disposed of. The application of the tuberculin test should be ~T H P T I I R P R P 1 ) TEST IN can "i e d out as previously described. This test THE BANG should be made subcutaneously and should be in- METHOD terpreted with care. As has been said the positive results are quite reliable ; but it is the negative results that give the trouble in applying the Bang method. Three things at least must be guarded as carefully as possible. I. A cow that is very badly infected may fail to react. Most generally however her physical condition would at least show that the cow was not in a good state of health and should be discarded. One such cow may soon infect the whole herd and should be watched for with utmost care. 2. If the test was made during the period of incubation of a part of the cattle, which is almost sure to be the case in some of the animals, especially of a large herd in .which this disease is just getting a start, some of the infected animals would not react ; but later the disease would develop and TUBERCULOSIS OF FARM ANIMALS 423 become evident in these. 3. In cases where the tubercles are suf- ficiently encysted and the disease becoming arrested, a reaction may not occur, as was shown by Moore (already cited). These cows may react at a subsequent Jest. Because of these last two facts, the failure of the tuberculin to show the presence of virulent tubercle bacilli in the body of the cow during the period of incubation and also when the tubercle is sufficiently encapsulated to prevent the reaction, it is probably advisable when the larger part of the herd reacts, say seventy percent or more, not to separate it into two parts but to consider the whole herd tuberculous, and begin to build up a healthy herd from the non-reacting calves and thoroly tested cattle that are "bought in". After finding a smaller part of the herd tubercu- 1OUS by the tuberculm test > the tw herds should LING OF THE ^ i mme diately separated and provided with sep- TWO HERDS arate stables, feed lots and pastures. It is better when practicable to have separate farms. The two herds should be cared for by different persons. If this is not convenient, the hands should be provided with different sets of over-suits, and shoes for each barn so that no disease germs may be carried on their clothing from the tuberculous herd to the well herd. The healthy herd is tended first and then the diseased herd. One should never go directly from the tuberculous herd to the healthy herd. One of the serious difficulties is providing water WATER ^ or bot ^ 1 nerds - I* i g expensive to provide two sets of wells or cisterns, especially when in a few years one expects to be rid of the tuberculous herd and then will not need the extra well for water supply. The living germs may easily be carried in the water. Therefore in case there is only one source of water, it must be so arranged that the water will flow from the healthy herd to the tuberculous herd. The stables where the non-tuberculous herd is to DISINFECTION , , . _ '.i 1 * r i r+' OF THE be kept must be thoroly disinfected. First, they STABLES should be thoroly cleaned, and then disinfected with mercuric chloride solution in proportion of one part of mercuric chloride to one thousand parts of water. The best way to apply this solution is with a spray pump. The wood work and floors of the whole stable, around the doors, the posts, fencing, and any other place on the premises where the tuberculous cows may have left tubercle bacilli, should be disinfected. After the disinfecting solution has dried, all the wood work should be painted, oiled, or white-washed. 424 BULLETIN No. 149 [February, It is necessary from time to time to retest the THE WEL we ^ kerd to see *kat tnere rema i n no tubercu- HERD lous animals. It is better to test every six months, and at least as often as once per year. The re- actors must be immediately removed and the stables again thoroly disinfected each time. No cow should be tested for three weeks before or after calving, nor if she has an abnormal temperature from any cause; but if omitted from the regular test she should be tested as soon afterward as is convenient. HANDLING OF Calves Q f tuberculous cows very rarely contract THE CALVES tuberculosis from their mothers before birth. It is estimated that not more than four percent are tuberculous or become tuberculous at the age of six months if they are immediately removed from the mother at birth and fed wholesome milk. Even a calf from a cow with tuberculous geni- .tal organs may be born free from the germs. Every calf should be tested at the end of six months before it is permitted to enter the healthy herd. Cows that are "bought in" must not be added to BOUGHT IN ^ e we ^ k ei "d unt il they are thoroly tested. If they come from a region where there is considerable tuberculosis they should be kept from the non-tuberculous herd for 4 to 6 months until at least two tests have been made before putting them with the well herd. It is by far better to buy cows only from a herd known to be entirely free from tuberculosis. The non-tuberculous herd is the great desideratum when you are buy- ing much more to be sought than the non-tuberculous individual ; because it is not unlikely that the non-tuberculous individual, proved so only by the tuberculin test, especially if she comes from a tuberculous herd or a region where tuberculosis is plentiful, may on the second or third test prove to be tuberculous ; and the buyer may find it out only too late, to save his well herd from becoming infected. It is necessary to pasteurize all milk from the tu- berculous herd before use as food for man or for TION OF MILK . mi i j- f f R __ nR _ PPFn animals. The best disposition of milk from tu- DtrUKtrttU- . .. ING TO HOGS berctilous cattle is to use it in the manufacture of AND CALVES some of the dairy products, as butter, or evapo- rated cream. In these the thoro pasteurization does not harm the finished products. It is the lack of attention to pasteurization that has caused so much trouble in many of the dairy regions. A cow with a tuberculous udder can infect all the milk from a number of herds when this is mixed together in the large containers at a creamery, and the skim milk from this cream- TUBERCULOSIS OF FARM ANIMALS 425 ery when fed to calves and pigs will produce tuberculosis. This is one of the most common causes of infection among these animals. Bang 1 applied his method first on the Thurebylille ' farm in 1892. A special government grant gave EXPERIMENT ^ m ^ e ^ un ^ s - When this herd was first tested 131 animals reacted and only 77, mostly young animals, were found to be healthy. The reactors and the non- reactors were at once separated but only with a board partition be- tween them in the stables, the shed occupied by the calves forming part of the healthy division, which should have been separate. They were attended by two sets of hands. The cattle were kept entirely separate at all times both in the stable and pasture. At times the project was discouraging. The reactors at one time from the well herd were as many as 9 percent. Things necessary to be done were not clearly understood by the farmer and at times a highly tuberculous cow would be allowed to remain on the farm for some time, dischaiging immense numbers of virulent bacilli. In spite of all these drawbacks, year by year the healthy herd in- creased and in 1907, five years after the experiment started, the herd consisted of 211 animals without one reacting to the tubercu- lin test. This herd is now supplying "milk for infants" in the city of Copenhagen, a very superior milk. This experiment has been repeated so many times in Denmark that it has now come to be known as the Bang system. It is especially Successful on the .larger farms and estates, more particularly where the herds can be placed on separate farms. On such farms the herds have been cleared of tuberculosis in two or three years. But not only on the large farms but on the small farms this method of Bang has been successful. Bang 1 records in 1905 that on 66 small farms, with an average of 29 head of cattle per farm, there were 1045 reacting animals and 780 healthy animals. This method was put into practice and a few years later there were 1896 healthy animals and none reacting. Harding and Smith 4 in New York State were equally successful in converting a tuberculous herd into a healthy herd. In Wiscon- sin 8 and in Pennsylvania 5 the Bang method has been applied with success. However both in the United States and in Canada the Bang method has not proved satisfactory to most of the stock own- ers who have tried it. Klein 5 , State Veterinarian of Pennsylvania, says that the Bang method is not popular and has not been adopted in his state except in a few instances. The reasons appear to be the restricted market for the heated milk and the extra labor in maintaining the two separate herds. Pope 7 of East Orange, New Jersey, quotes from .a letter received from Hon. W. C. Edwards, Rockland, Ontario. 426 BULLETIN No. 149 [February, "I most certainly think that the Bang 1 system is a most excellent one for any one to practice. I may say, however, that I do not think that many farmers or even breeders, will go to the expense and trouble involved in the practice. Personally I may say that the riddance of tuberculosis thru the Bang system is very effective \ but it is very expensive to practice it, as we practice it here. If it was to do over again I would have separate farms ........ " "ASSOCIATE ""* seems pl aus iW e that an "associate farm" where FARM" *ke ne i"hbors could combine their forces in stamp- ing out tuberculosis from among their cattle would be a good thing. There the owners of small herds could econom- ically use the Bang method in ridding their herds of this disease. It would only be necessary to have a small tract of land suf- ficient to pasture the tuberculous cows kept by the association. The tract of land, with sufficient buildings could be owned, rented, or leased as each association should choose for itself. The "associ- ate farm" would take care of the tuberculous cow or cows for each member of the association for which would be paid a stipulated price agreed upon by the members. Only cows valuable either for butter making or more especially for their calves would be expected to be kept on the farm. Each member would get back his own calves to add to his sound herd upon his own farm, and have also his share of the proceeds of the dairy products that were marketed. The advantages of the "associate farm" would be several: i. The disease would be much more completely isolated. The tu- berculous cows of the whole neighborhood could easily be kept from the non-tuberculous cows. The method would thus avoid in a large measure the difficulties and expenses of the Bang system as practiced by the individual farmer, and moreover would insure greater protection for the healthy herds. 2. There would be an advantage in marketing the dairy products. The public interest in the matter of pure dairy products is becoming more and more pronounced, and when any dairyman could show that he had no tuberculous cattle upon his farm, his dairy products would be at a premium. The cream from the tuberculous herd at the associa- tion farm could be made into the highest grade butter after proper pasteurization. The method thus combines the best principles of the Bang system and does away with the expense of two sets of stables and pastures and two sets of hands at each farm. It pro- vides the small stock owner the opportunity of successfully using the Bang method to rid his herd of tuberculosis, and at the same time gives him an advantage in marketing his dairy products. TUBERCULOSIS OF FARM ANIMALS 427 THE DUTY OF THE STOCK OWNER There is expense involved, and every farmer will have to de- termine for himself whether the expense in freeing his herd from tuberculosis by the Bang method will be greater than that of slaughtering and purchasing new stock. There will be expense, trouble and aggravation in any way he rids his farm of this dreaded disease. But the expense will not be so great as keeping the disease in his herd, which sooner or later is sure entirely to de- stroy it. The problem is here and it must be faced, and the sooner the fight is begun, for a fight it must be, the less will be the strug- gle and expense. Moreover, in the end, this will be the task of the individual farmer in whose herd the disease exists. Indeed the Nation and the State will have and should have their share of responsibility. But tuberculosis like a noxious weed must be kept out of each individual herd, and the cattle owner is the only one that can constantly keep guard over his own flock. To this end the stock owner should educate himself concerning tuberculosis, concerning its slow development in the 'animal body, the ease with which it is overlooked, its contagiousness, its manner of spread from herd to herd and from animal to animal within the herd, its detection by the tuberculin test and by other means ; and he should learn to understand especially the failure that he may experience by relying too implicitly upon the negative tuberculin test. More than this, not only must he inform himself upon this subject, but he must help teach his neighbor; for his ignorant neighbor may maintain a tuberculous herd in such condition as to menace the stock of all the surrounding farms. DUTY OP THE STATE It is widely recognized that it is an obligation of the state to take an active part in the control and prevention of tuberculosis among farm animals. For this purpose Germany and England have appropriated large sums of money to carry on scientific investiga- tions which they are doing thru their Commissions on Tubercu- losis. A large amount of investigation has been done in this coun- try by the United States Government thru its Bureau of Animal Industry, as well as by various state governments. As has been said before, in the end it will fall to the lot of the stock owner- to eradicate and to keep out of his own herd this dreaded disease. But it will be the part of the state to organize the work, to combine the efforts of the various live stock interests, health organizations and other societies that they may all work together in harmony to one end, the complete eradication, finally, of tuberculosis from among the farm animals. This is possible. An example of ef- fective prevention may be seen in the Island of Jersey where by vigilant care the cattle are kept quite free from tuberculosis. 428 BULLETIN No. 149 [February, ORGANIZATION tuberculosis work of the state is usually or- OF THE WORK ganized and carried out under a Board of Live Stock Commissioners. This is better than the creation of a special board for this single disease. Extra funds will be necessary and should be provided so this board may secure the proper help to carry on its work efficiently. It would be folly for this board to attempt a radical elimination of all tuberculous cattle. This has been tried in other states and found impracticable. 'It will be necessary to proceed along- conservative lines. It is the combined effort of all concerned that is the most effective in elimi- nating 1 this disease. To preserve the harmony and to keep at work all the organizations concerned in eradicating tuberculosis from farm animals will be one of the most difficult as well as the most important of the tasks that the Board of Live Stock Commission- ers will have to perform. Another! important work of the Board should be TUBERCULIN ^ providing 1 of free tuberculin testing of cattle TESTING when requested by stock owners. This testing should be done under proper restrictions and con- trol. The stock owner should submit to such a control of the react- ing animals that they may not become a menace to other stock and to the health of the people. Each reacting cow should be branded by cutting the letter T in the right ear, so that the few unscrupu- lous stock ow r ners could not impose upon their neighbors by selling the fat, sleek, reacting cows for healthy animals. This testing should be free, for then it will be done more usually by authorized experts, which is necessary for the best results. Furthermore the report of the testing which is so essential for the further and ef- fective control of the reacting cattle will thus be in the hands of the state officials. All reacting animals that are not especially desired CONTROL OF r i r ,, THE REACTING breeding purposes or for the production of ANIMALS ' dairy products should be slaughtered under State or Federal inspection. When it is desired to retain such reacting animals as mentioned above, proper quarantine regu- lations should be enforced under the control of the Board of Live Stock Commissioners. Printed regulations and advice should be provided by this board when requested by the owners of such stock. An inspector should visit the quarantined herd not only to see that the regulations are carried out but to help the stock owner by advising him of the best and most scientific methods of ridding his herd of tuberculosis. The sale of tuberculous animals except as such should be prohibited by law. An indemnity of the real value of a tuberculous cow should be 1911} TUBERCULOSIS OF FARM ANIMALS 429 paid the owner when it is the opinion of a state official that a cow should be destroyed in order to protect the stock of the surround- ing neighborhood. This procedure should be used sparingly and only when absolutely necessary. It should be remembered that very few states, if any, have ever succeeded in ridding themselves of tuberculosis by buying up the tuberculous animals. Maine may possibly prove to be the one exception, and Maine can succeed, if she does finally, only because there is very little tuberculosis in the state. Illinois can not wisely undertake such a procedure at pres- ent. It is especially unwise to pay a fictitious price for such in- demnity. The value of a tuberculous cow can never equal the price of a non-tuberculous cow of similar grade however well she may look from outward appearances. The indemnity should be measured by the actual market value of the tuberculous cow rec- ognized as tuberculous. To encourage stock owners to keep their herds PUBLIC LIST- , , & , , . ., j ui i free from tuberculosis it seems advisable to keep ING OF NON- , ,. ,. f . i i j j. TUBERCULOUS anc * P u b nsn nsts * owners of such herds, mdicat- HERDS ing the size of each herd, and the time it has been kept free from this disease as shown by official testings. Such publications may be undertaken by the various live stock associations but the testing and reporting should be done under the direction and control of the Board of Live Stock Com- missioners. The records of the Board should be available to any purchaser and not only would it help the purchaser to select stock from herds free from tuberculosis as shown by official tests, but it would be an incentive to stock owners to keep their herds free from this disease as it would put a premium upon such herds. EDUCATION Another important duty of the state is to carry on POPULAR AND PP ll l ar education along these lines. This should SPECIAL k e P rom ted and carried out by all organizations concerned and at all available opportunities. There is not any one thing that will do more to rid farm animals of tuberculosis than the education concerning this disease of every person who has to deal with such animals. Public demonstrations showing the value of the tuberculin test by slaughtering some of the reacting animals, especially some of those whose physical ap- pearance is that of healthy cattle, with the demonstration of the lesions in such cattle, will convince the most skeptical of the worth of this valuable test. Popular lectures and exhibitions of the tu- berculous tissues from cattle and hogs will be very profitable. It might also prove economical and wise for the state to pro- vide facilities for the training of men expert and skillful in the handling of this disease. At least good general instruction in the subject for all agricultural college students would seem advisable. 430 BULLETIN No. 149 [February, SUMMARY 1. Recognition of tuberculosis in dairy cattle depends almost entirely upon the tuberculin test. 2. The positive tuberculin test is accurate in about 98 percent of the cases as shown by slaughter; the negative test is not so reliable. 3. For handling the tuberculous cow one of the following methods is advised : a. In advanced cases where there are physical signs of the disease or when the cow is not very desirable for breeding or dairy purposes, slaughter under inspection is best. b. When it is desirable to keep the reacting cow for the purposes mentioned aboye, the Bang method is recommended es- pecially to large stock owners. c. For owners of smaller herds the modified Bang method by means of a stock owners' association, so the tuberculous ani- mals may be farther removed from the healthy herd, is to be recom- mended. 4. The responsibility of ridding farm animals of tuberculosis must be placed upon the stock owner before there can be any hope of permanent success; for only the stock owner can keep close watch over his herd. 5. The State will have her share of responsibility in formulating rules and regulations, providing for free tuberculin testing, and carrying out a system of popular and special education upon the subject of animal tuberculosis. REFERENCES 1. Bang, B., Measures against animal tuberculosis in Denmark. Sixth Internat. Cong, on Tuberculosis 4, part 2: 850-868. 1908. 2. Dammann and Stedefeder, (Dfeuts. Tierarztl. Wschr., 1909: 345.) Ref. Jour, of Comp. Path. 22: 258. 1909. 3. Fisher, I., Report on national vitality, its wastes and conserva- tion. Bulletin of the Committee of One Hundred on National Health. U.S. Senate Document No. 419. 1910. 4. Harding and Smith, The Bang method of controlling tubercu- losis, with an illustration of its application. New York Agr. Exp. Sta. Bull. 277. 1906. 5. Klein, L. A., Control of tuberculosis in domestic animals in Pennsylvania. Sixth Internat. Cong, on Tuberculosis 4, part 2: 547- 556. 1908. 6. Moore, V. A., The value of tuberculin in the control of tuber- culous herds. Sixth Internat. Cong, on Tuberculosis 4, part 2 : 918- 926. 1908. TUBERCULOSIS OF FARM ANIMALS 431 7. Pope, J. E., Tuberculosis of cattle from the point of view of the farmer. Sixth Internat. Cong, on Tuberculosis 4, part 2: 571- 583. 1908. 8. Russell, H. I., The history of a tuberculous herd of cows. Wis. Agr. Exp, Sta. Bull. 78: 3-16. 1899. 9. Smith, Th., After Moore. ; NM!