VETERINARY MEDICINE. 583 



evaporation, poor growth, or contamination tooli place, making some elimination 

 of flasks necessary. 



" Broadly speaking, the reaction cnrve in glycerin broth divides tubercle 

 bacilli into 2 types. The bacilli which possess a low degree of virulence for rab- 

 bits and the power to grow well on glycerin media in the early generations 

 produce one type of reaction curve, while those which are virulent for rabbits 

 and which in the early generations grow slowly and with difliculty on glycerin 

 media form the other type of curve in glycerin broth. These 2 types of glycerin 

 reaction curve are again divisible into groups accox-ding to their final reactions. 



" The curves of adjacent groups show much the same general direction and 

 there is a gradation from one group to the next ; but the reaction curves of the 

 groups at both extremes are widely divergent. When any large number of 

 viruses is examined there will be found a small percentage of cases, which, by 

 cultural characteristics and virulence, belong to one type of tubercle bacilli 

 while they would be classed with the opposite type of bacilli if judged by their 

 glycerin reaction curve alone. On repeated tests this reversed glycerin reaction 

 curve may, or may not, be a constant feature of these particular viruses, 

 although the conditions under which they have been cultivated are apparently 

 the same in the several tests. Undetected variations of the culture medium 

 must be taken into consideration ; it is not advisable to depend on the reaction 

 curve obtained from one lot of broth only, but several examinations of a virus 

 are desirable. In from 30 to 40 per cent of the viruses retested, the reaction 

 curves belong to different groups, that is, the end reaction may be high in one 

 test, and low or medium in the next. In only 3 instances was the variation so 

 great as to justify the classification of the reaction curves into different types. 

 In about half the cases the degree of acidity produced has been in direct ratio to 

 the amount of growth. 



" There is also no constant relation between irregularities of culture and viru- 

 lence on the one hand, and irregularity of the glycerin reaction curve on the 

 other. Some viruses which culturally and in virulence showed nothing unusual 

 have given very atypical curves, while perfectly normal reaction curves were 

 produced by viruses which from cultm-al and virulence tests could not be called 

 quite typical. 



" The glycerin reaction curve is undoubtedly a valuable corroborative evi- 

 dence of a division of tubercle bacilli into 2 types. Its value is lessened, how- 

 ever, by the number of irregular and atypical reactions encountered, while as a 

 practical aid in determining the type of an individual virus, it is also much 

 handicapped by the length of time required to carx'y it out." 



Report in regard to avian tuberculosis in mammals, D. A. De Jong (Ann. 

 Inst. Pasteur, 2.'f (1910), No. 11, pp. 895-906). — According to this author avian 

 tubercle bacilli can spontaneously infect, in addition to man, the ape, pig, bovine, 

 rabbit, rat. and white mouse. He assumes that such cases will be observed 

 much oftener as soon as it can be proved that the mammalian tubercle bacillus 

 can be converted by simple mutation into the avian type of bacillus. 



The tuberculin test and its limitations, E. G. Hastings {HoanVs Dairy- 

 man, 42 (1912), No. 50, pp. 1525, 1537-15-'t0).— This article points out the fact 

 that the tuberculin test has cei'tain limitations which must be taken into con- 

 sideration by those using and interpreting the test. 



The intracutaneous test for detecting' tuberculosis in bovines, M. Cheis- 

 TiANSEN (Maancdskr. Dyrlwger, 22 (1910), No. 16, pp. 337-352, figs. J,; ahs. in 

 Berlin. Tlcrdrztl. Wchnschr., 27 (1911), No. 28, pp. 509, 510).— Out of 75 

 animals, .30 gave a positive reaction with this test, and on slaughter 27 of the 

 30 animals were found to be tuberculous. The author recommends a further 

 critical study of the test. 



