614 
Journal of Agricultural Research 
Vol. XXVIII, No. 7 
by combining 1 cc. of a l-to-100 solution of barium chlorid with Q9 cc. of a 
l-to-100 solution of sulphuric acid. A practice was made of reading the tests 
after they had been incubated at 37° C. for a period of from 36 to 48 hours. 
The degree of agglutination has been indicated (in Table V) by the characters 
+, P, S, and —. The + sign has been used to indicate complete clumping of 
the bacteria, P a partial clumping, S a slight trace of agglutination, and the 
minus sign to indicate no evidence of clumping. 
In addition to the bimonthly agglutination and complement-fixation results 
indicated in the charts of the individual animals, the results of cultural and 
inoculation tests of milk, placentas, fetuses, uterine fluids, etc., are also recorded. 
OBSERVATIONS RELATIVE TO IMMUNITY IN ANIMALS WHICH HAD 
PREVIOUSLY ABORTED 
At the time this experimental work was undertaken there appeared to be a 
difference of opinion as to the theory of living organism vaccination. The use 
of massive doses of Bacterium abortus created the impression that saturation of 
the subject with abortion bacteria was the object desired. Under these con¬ 
ditions the prevention of abortion losses was attributed more to the develop¬ 
ment of a tolerance to abortion bacteria than to resistance to their invasion. 
This conception appeared to be justified to some degree in view of the fact 
that little distinction was made between infected and noninfected animals as 
proper subjects for immunization. In other words, the tendency was to assume 
that the noninfected animal could be rendered resistant to the infection by 
the subcutaneous administration of massive doses of the organisms, and that 
the Bacterium abortus udder-infected animal likewise derived resistance of a 
similar character by contributions to the infection borne at the time of treat¬ 
ment and possibly for months or years previously. Whether or not injections 
of biological products have a tendency to overcome infection of the udder with 
abortion bacteria is a problem that has received very little attention. In the 
absence of evidence to support such a contention it is difficult to conceive logically 
how the same beneficial results can be reasonably anticipated from the treat¬ 
ment of both types of animals. 
Studies that were made of 12 cows, all of which had previously aborted, and 
from the fetuses, placentas, or uterine fluids of which Bacterium abortus had 
been isolated, furnished information that has some bearing on this matter. 
All of these animals were eventually successfully bred. Their milk was then 
tested for the presence or absence of Bacterium abortus by guinea-pig inocula¬ 
tions. It was determined that the milk produced by one-third of the number 
contained abortion bacteria. The milk from eight of the group failed to produce 
abortion disease in any of the pigs that were inoculated. 
Subsequently to collecting the milk samples, 11 of the 12 cows received by 
the mouth variable amounts of fetal-stomach contents containing Bacterium 
abortus. One cow received one administration, 4 received the material on two 
occasions, and 6 on three different dates. Cultural tests which were made of 
the exposure material in all cases indicated the presence of viable abortion 
bacteria in an uncontaminated condition. Ten living calves were produced 
by this group. Failure was experienced at time of parturition in isolating 
Bacterium abortus from the placentas or uterine fluids of eight of the number 
by cultural means or guinea-pig inoculations. The animals having uteri free 
from Bacterium abortus at time of parturition, as indicated by the above-men¬ 
tioned tests, were those that from the earlier milk tests gave evidence of having 
overcome the infection in their bodies. 
Bacterium abortus was isolated from the fetuses, placentas, or uterine fluids 
of the remaining four cows, despite the fact that animal No. 413 was not sub¬ 
jected to ingestion exposure during the present pregnant period and two of the 
number (413 and 460) aborted. All four were shown to have been carriers of 
Bacterium abortus infection in their udders during the early months of preg¬ 
nancy. (See Table II.) 
