550 
of the District of Columbia, because in both places the number tested 
is only a small portion of the total number in use, and the percent- 
ages of tuberculosis obtained respectively may have been influenced 
largely by the motive that prompted the application of the test. 
When tests are made at the request of those who own dairy herds, it 
ma}^ be assumed that the owners of exceptionally good herds will be 
in the majority, and the percentage of tuberculosis discovered will 
be low. If, on the other hand, the tests are largely forced for the 
protection of public health because tuberculosis is suspected among 
the tested animals, the percentage of tuberculosis found will rise to 
the maximum figure. 
The writer has personally tested a large number of dairy herds 
in widely separated localities, and in all his tests did not have the 
good fortune to find a single herd entirely free from tuberculosis. 
Most of these herds, however, were tested at the request of owners 
who had some reason to suspect tuberculosis among their cattle, and 
hence this discouraging experience can not be used as a reason for 
assuming that few perfectly healthy dairy herds exist. 
From the figures and estimates that are available it seems fair 
to conclude that not less than 20 per cent of our dairy cows are 
tuberculous, and that tuberculosis occurs to some extent in about 
30 per cent of our dairy herds. These are believed to be conservative 
figures, but they must be taken strictly as having purely and simply 
the value of an estimate. In some European countries, where better 
statistics are available than in the United States, it is safe to con- 
clude that not less than 40 per cent of all dairy cows are tuberculous, 
and this high percentage will be reached among our dairy cattle 
before long unless vigorous means are used to prevent the further 
spread of tuberculosis among them. 
THE FREQUENCY WITH WHICH DAIRY PRODUCTS HAVE BEEN 
PROVEN TO CONTAIN TUBERCLE BACILLI. 
The truest test of the measure in which the- public is exposed 
to tubercle bacilli from bovine sources is the frequency with which 
tubercle bacilli occur in dairy products. 
Without reviewing investigations of older date or those made 
in foreign countries, four comparatively recent investigations made 
in America show how common the occurrence of virulent tubercle 
bacilli in milk is. The largest of the four investigations showed 
that 15, or 6.7 per cent, of 223 samples of milk contained tubercle 
bacilli. The milk was obtained from 102 dairies, among which 11, 
or 10.7 per cent, were distributing infected milk. a The second 
investigation showed that 2, or 2.7 per cent, of 73 samples of milk 
a J. F. Anderson, United States Public Health and Marine-Hospital Service, 
Hygienic Laboratory Bulletin No. 41, pp. 163-192. 
