664 
Journal of Agricultural Research 
Vol. XXXI, No. 7 
EXPERIMENTAL PROCEDURE 
The experimental work was carried out in a herd of approximately 
200 head of dairy cattle in which infectious abortion had prevailed 
for about five years. An agglutination test of the herd, made March 
17, 1922, showed that the blood serum of 22 cows caused clumping of 
a Bad. abortus suspension in amounts as small as 0.001 c. c., and that 
the end point in the case of 8 others was 0.002 c. c. Results of this 
test on these 30 cows are given in Table I. 
Table I.— Blood-serum, reactions obtained in Bacterium abortus agglutination tests 
on SO cows 
+=Pronounced agglutination. S=Slight agglutination. 
P=Partial agglutination. — * No agglutination. 
In making these agglutination tests six different quantities of 
blood or milk serum were used. There is placed in six different test 
tubes these respective amounts: 0.04, 0.02, 0.01, 0.005, 0.002, and 
0.001 c. e. " There is then added to each tube 1 c. c. of a 0.5 per cent 
carbolized suspension of Bad . abortus similar in density to a barium- 
sulphate suspension resulting from the addition of 1 c. c. of a 1: 100 
solution of barium chloride to 99 c. c. of a 1:100 solution of sul¬ 
phuric acid. Results are indicated by these signs: +, which signifies 
pronounced agglutination of the bacteria; P, clumping incomplete; 
S, a trace of agglutination; and —, no agglutination. Readings of 
the tests were made after incubation for 36 hours. 
After identifying the marked reactors to the abortion test, com¬ 
posite samples each containing 40 to 50 c. c. of milk or udder secre¬ 
tions, were taken from these 30 animals, as were blood specimens 
also, with the idea of detecting those cows which carried the infection 
in their udders. 
The milk was kept in the refrigerator over night, after which the 
cream layer was pipetted off. The specimens were then thoroughly 
shaken, warmed to body temperature, and injected intra-abdominally 
into guinea pigs in 5 c. c. amounts. Two guinea pigs were used in 
most cases for making this preliminary test; in two or three cases 
four were inoculated. In obtaining the agglutinin titer of the milk, 
a small quantity of rennet was used to curdle the specimens. Milk 
