io Journal of Agricultural Research v 0 i. ix, no. t 
Another fact related to the expulsion of abortion bacilli with milk from 
the udders of cows is that in the numerous tests made with milk from 
many different cows the abortion bacillus was never found in the milk 
of a cow unless both her milk and heir blood serum possessed agglutinating 
properties for suspensions of abortion bacilli. This fact is interesting 
and important not only on its own account but also because it serves as 
strong circumstantial evidence to prove that the work of the writers on 
the occurrence of abortion bacilli in the milk of cows is trustworthy. 
It does not mean, however, that the milk of all cows which react with 
the agglutination test for abortion disease is infected, as the writers have 
repeatedly tested milk from reacting cows without detecting abortion 
bacilli. 
Regarding reacting cows with uninfected udders, it appears that their 
blood serum gradually loses its power to agglutinate suspensions of 
abortion bacilli. The writers wish, however, to have this statement 
taken cautiously, as the evidence behind it is not yet sufficient to give 
it the rank of a proved fact. If this statement, on further study, should 
prove true, it, together with other facts, will justify the conclusion that 
the persistence of agglutinating and complement-fixing substances in 
the blood of cows, relative to abortion disease, is intimately associated 
with the abortion bacilli that enter the body through the lymphatics 
from infected udders, as abortion bacilli do not maintain themselves in 
the bodies of cows elsewhere than their udders and gravid uteri. 
That abortion bacilli do not maintain themselves in the bodies of cows 
elsewhere than the regions named is a fact of which the writers have 
obtained fairly convincing proof. It was found that abortion bacilli 
injected into the veins of normal, nonpregnant cows disappeared from 
their circulating blood in the course of a few hours; and when such cows 
were killed sometime afterwards, though their blood had become positive 
with agglutination tests, the germs could not be found in their bodies 
unless it was in their udders and associated lymph glands. One case in 
the records of experiments is remarkably impressive as an illustration of 
the tendency of abortion bacilli to lodge in the udder. The case is that 
of an adult, virgin, female animal, a heifer, approximately 4 years old, 
which was given an injection of abortion bacilli into one of her jugular 
veins. Later it was found that the infection 1 had established itself in 
her virgin udder, which was not functioning and never had functioned. 
Another series of tests, probably even more convincing than the 
foregoing, was a careful search for abortion bacilli in the bodies of natur¬ 
ally infected as distinct from artificially infected cows. The cows were 
killed and their blood, spleens, livers, kidneys, brains, ovaries, uteri, 
udders, milk, synovial fluid from various joints, nerve tissue, lymph 
glands from all portions of the body, etc. tested for abortion bacilli 
1 The term “infection” is used here and elsewhere in this paper as signifying the discoverable presence of 
abortion bacilli, and not as implying the development of observable lesions of disease. 
