21 
amount of an emulsion of tubercle bacilli was added; tubercle bacilli 
were demonstrated by smears, culture, and guinea-pig inoculations, 
showing that the bactericidal action, if any, was very slight. 
The smears in every instance were carefully examined by two and 
a portion of them by three different observers. I am satisfied that 
if acid-fast bacilli had been present they would not have escaped 
detection. 
The blood from 8 rabbits experimentally infected was examined 
for tubercle bacilli. The blood from 7 of the 8 was examined by 
glycerin-potato cultures and guinea-pig inoculations of the whole 
heart’s blood and of the sediment from the citrated blood in the ice 
chest for twenty-four hours. The blood from one rabbit (No. 1) 
was examined only by injection of 5 c.c. into a guinea pig. 
Smears were made from the sediment of the citrated blood in the 
ice chest of 7 of the rabbits. As the blood of a number of the rab- 
bits was proved by cultures and guinea-pig inoculations to contain 
tubercle bacilli, an unusually thorough search was made for acid- 
fast bacilli in the sediment smears of these cases. Two smears were 
made from the blood of each rabbit. The smears were examined b}^ 
three trained laboratory men, but not in a single instance were 
tubercle bacilli found, though the blood contained tubercle bacilli 
in sufficient numbers to grow on potato and to infect guinea pigs. 
Six of the 8 rabbits showed naked-eye appearances of tuberculosis 
at the autopsy made after the blood was taken. The blood of 1 of 
the 2 remaining was infective, even though there were no naked-eye 
lesions of tuberculosis. 
Tubercle bacilli grew on the glycerin potato in 3 of the 6 cases, 
both from the whole heart’s blood and from the citrated sediment. 
The blood was infective for guinea pigs by both methods in 4 of the 
6, yet tubercle bacilli could not be found in the smears. 
These results with the blood of rabbits are strongly against the 
idea that the blood serum of individuals with tuberculosis has a bac- 
teriolytic power sufficient to lessen or destroy the virulence of tubercle 
bacilli in the blood stream so that, while they may be found in 
smears, they will not grow on cultures or infect animals. 
The blood from only one guinea pig (No. 105) experimentally 
infected was found to contain tubercle bacilli by inoculation into 
fresh animals. The blood from all the others was negative, both 
culturally and by inoculation. No tubercle bacilli were found in the 
smears made from the citrated blood of No. 105 in the ice chest over 
night, though it was infective for the fresh animal. 
It is of interest to note that the blood of only 1 guinea pig out of 
13 contained tubercle bacilli, while that of 7 out of 8 rabbits experi- 
mentally infected contained them in sufficient numbers to be demon- 
strated either culturally or bv inoculation. 
