23 
smears were examined by at least two observers and a part of them by 
three, and, moreover, cultures and animal inoculations were negative. 
The third point brought out by the experiments is that tubercle 
bacilli were demonstrated in the blood of 7 of the 8 rabbits experi- 
mentally infected. In 3 of the 7 they were found both by cultures 
and by animal inoculations. From this it would seem that 
when rabbits are infected by subcutaneous inoculation of tubercle 
bacilli the latter are found in the circulating blood in a large propor- 
tion of the cases, but not in sufficient numbers to be detected in 
smears. It is of especial interest to note that the blood of a rabbit 
(No. 1), which at the autopsy did not present any naked-eye evidences 
of tuberculosis, was infective for a guinea pig. 
Tubercle bacilli were found in the blood of only 1 guinea pig 
experimentally infected. It seems, therefore, that whereas tubercle 
bacilli are frequently found in the blood of tubercular rabbits, it is 
unusual tofind them in the blood of tubercular guinea pigs and humans 
in numbers or virulence sufficient to infect fresh animals. 
The culture of tubercle bacilli from the blood of the experimentally 
infected rabbits is one of the few, if not the first, recorded instances in 
which tubercle bacilli have been grown from the blood of tubercular 
animals. 
