THE TUBERCLE BACILLUS 489 



From this table it is evident that out of a total of 1,042 cases, 101 

 only were bovine in origin and over 50 per crent of these occurred in 

 children under five years of age. Fifty-one out of the 59 cases occurring 

 in the 161 infants were directly or indirectly traced to alimentary 

 infection. 



It seems reasonably accurate, therefore, to state the case as follows : 

 Human adults are relatively insusceptible to bovine infection. Such 

 infection can take place, but is unusual. Below 16 years of age the 

 human race is relatively more susceptible and up to this age the danger 

 of milk infection is unquestionably great, this source accoxmting for 

 about one-third of the cases. Below 5 years the danger is greatest. 

 This table alone should form sufficient evidence to silence absolutely any 

 doubts as to the dangers of milk infection and remove any objections 

 to the most rigid sanitary control of milk supplies. 



On the other hand, it also shows that Behring's original claims were 

 far too sweeping and can not be upheld. 



Rosenberger^ has recently reported finding tubercle bacilli in the 

 circulating blood of all cases of human tuberculosis which he examined. 

 This announcement aroused much interest and has led to many investiga- 

 tions by other workers. Rosenberger's results were obtained by morpho- 

 logical examination of smears of citrated blood taken from the patients, 

 dried upon slides and laked with distilled water. Many other observers 

 have failed to confirm Rosenberger's results. Anderson ^ examined 

 47 cases in which tubercle bacilli were found in the sputum and one case 

 of joint. tuberculosis. In none of these 48 cases was he able to obtain 

 tubercle bacilli, neither by morphological examination nor by guinea-pig 

 inoculation. Brem ' subsequently found that laboratory distilled water 

 may frequently contain acid-fast saprophytes — a fact which may 

 account in many cases for errors when morphological examination alone 

 is relied upon and blood examined by the technique of Rosenberger. 

 This, too, is suggested by the finding of acid-fast bacilli in the blood of 

 perfectly healthy individuals. Therefore, although the bacilli may be 

 present in the blood in a certain number of cases it does not seem likely 

 that they are so distributed in anything hke the high percentages found 

 by Rosenberger.* 



Bacillus tuberculosis (typus humanus) is pathogenic for guinea- 



1 Rosenberger, Am. Jour, of Med. Sc, cxxxvii, 1909. 



« Anderscm, U. S. P. H. Service, Hygienic Lab., BuU. 57, 1909. 



» Brem, Jour. A. M. A., liii., 1909. 



' Suzuki and Takaki, Centralbl. f . Bakt., Ixi, 191 1. 



