42 
Journal of Agricultural Research 
Vol. VIII, No. 2 
the same protective power whether present alone or in the serum. But 
the results obtained at the same time with serum 96 and globulin 96 
do not substantiate this view. It is to be noticed that three guinea 
pigs that received 1 c. c. each of globulin 96, containing 0.147 gm. of 
pseudoglobulin, died before the close of the test, as well as three that 
received 1 c. c. each of the heated globulin 96, containing 0.150 gm. of 
pseudoglobulin. On the other hand, two guinea pigs out of three, that 
had received 0.075 gm. of globulin (mostly pseudoglobulin, with a small 
amount of euglobulin) in the form of 1 c. c. of serum 96 survived. Like¬ 
wise, two out of three survived after receiving 2 c. c. of serum 96. From 
these latter results alone the inference might be drawn that a protective 
action existed in the serum which was absent in the globulin prepara¬ 
tions. The early death of practically all the guinea pigs that received 
globulin 96 (Table II) led to the suspicion that the virus C was too 
strong. 1 Accordingly test 5 was repeated, using the weaker virus D and 
increasing the quantities of globulin 96 injected (test 7, Table III); 
otherwise the two tests were the same.' (For a description of the virus, 
technic of the injections, etc., see page 41.) 
In test 7 (Table III) the results obtained with globulin 48 are prac¬ 
tically the same as those obtained in test 5 (Table II). A larger number of 
guinea pigs did not survive, although a weaker virus was used. Plainly 
serum 48 protected a larger number of guinea pigs against virus D than 
against virus C. In test 7 the large number of survivals with serum 48 
as compared with globulin 48 would indicate that, the serum contained 
protective bodies which were absent in the globulin preparations. This is 
just the reverse of the result in test 5, in which the stronger virus was 
used. For serum 96 and globulin 96, the results in test 7 are essentially 
similar to those obtained in test 5, except that a few more of the globu¬ 
lin guinea pigs survived. It is apparent from the results that no definite 
statements can be made as to whether the survivals were due to the weaker 
virus or the increased amounts of globulin. 
In the tests that were made the protective power of globulin 48 
heated was as great as that of globulin 48 unheated, although the former 
was prepared from serum heated for six hours in a 6o° C. air oven in the 
presence of one-third saturation ammonium sulphate. In so far as the 
flasks containing the mixtures to be heated were at room temperature 
when placed in the oven, it is practically certain that the temperature of 
6o° C. was actually reached inside the flasks only toward the end of the 
heating period. This method of heating was first used because it was 
desired to heat the mixture up to 6o° C. without exposing any part of it 
to a temperature much higher than this. In general it was similar to the 
method of heating described by Banzhaf (2, p. 115). On the other hand, 
globulin 96, heated, was not as potent in protective power as the corre- 
1 When the tests were in progress the time of death of the animals was noted almost every hour, day 
and night. Little would be gained by inserting these figures into the tables. 
