494 
Vital statistics show that 14 out of every 100 people that die 
succumb to tuberculosis, while of the remaining 86 more than one- 
half show lesions of tuberculosis on post-mortem, although dying 
from some other cause. The statement of Von Behring, above men- 
tioned, is practically pertinent in regard to the relation of human 
tuberculosis to the milk supply, especially in connection with the 
results of those investigators who have studied market milk and 
found from 5.2 to 55 per cent of the samples examined to contain 
tubercle bacilli. Le Blanc, Ripper, Jemma, and de Michele consider 
the milk of tuberculous cows dangerous even when bacilli are not 
present, on account of the toxin it contains. Michellazzi has injected 
such milk into tuberculous animals and obtained a reaction. 
To eliminate all tuberculous cattle from the herd, or to pasteurize 
all milk coming from untested cattle, should therefore be the object 
of all producers of milk, and sanitarians will be remiss in their 
whole duty should they neglect to guard against the products of 
tuberculous animals in their attempts to eradicate tuberculosis 
from man. 
TUBERCLE BACILLI IN OTHER DAIRY PRODUCTS. 
Since milk is so often infected with tubercle bacilli, it is very evi- 
dent that food products made from milk without submitting it to 
lethal temperatures during the process of their manufacture must 
frequently harbor virulent tubercle bacilli in undesirable numbers. 
The investigations of Rabinowitsch, Klein, Laser, Bang, Petri, 
Dawson, Markl, Moller, and many others have conclusively shown 
that tubercle bacilli may be present in butter, buttermilk, margarin, 
and cheese, when these products are offered for sale. Butter made 
in the customary manner and stored under the ordinary market con- 
ditions until time of sale, if dangerous through the presence of 
tubercle bacilli at the time of its manufacture, may retain its viru- 
lence through several months. This statement has been adequately 
proved by a series of experiments recently performed by Mohler, 
Washburn, and L. Rogers, of the Bureau of Animal Industry. In 
this work three samples of butter were tested. The first sample was 
made from milk to which bovine tubercle bacilli had been added just 
before churning. They were obtained from a luxuriantly growing 
culture upon glycerin bouillon. Ten centigrams were removed from 
the surface growth of the flask, carefully mixed in a sterilized solution 
and added to 10 gallons of milk. The second sample was made from 
milk obtained from a cow affected with tuberculosis of the udder. 
In this milk tubercle bacilli of extreme virulence were present in 
great numbers. Both the first and second samples of butter were 
salted in the usual proportions of 1 ounce of salt to 16 ounces of 
