SO 
Journal of Agricultural Research 
Vol. VTII, No. a 
In general there was from three to four times as much globulin in the 
globulin preparations as there was in the same volume of serum. Thus, 
io c. c. of serum 48, January 6, 1916, contained 0.411 gm. of total globu¬ 
lin; about three-fourths of this was pseudoglobulin. The globulin 
concentrate prepared from this serum contained five times that amount 
of globulin, 2.110 gm., all of which was pseudoglobulin. This was 
determined by direct coagulation rather than by precipitation with ammo¬ 
nium sulphate, because the results are more accurate by the former 
method. Whether the globulin preparations were three or four times 
as potent in protective power as the same volume of serum can not be 
stated definitely, for reasons already given. The extent to which other 
proteins were removed during the globulin concentration is indicated in 
the last column of Table I. 
CHANGES IN THE SERUM PROTEINS DURING THE COURSE OF 
HYPERIMMUNIZATION 
The changes in the serum proteins which take place during the course 
of hyperimmunization against diphtheria were studied by Hiss and 
Atkinson (6) and by Ledingham (7). Similar studies were made by 
Banzhaf and Gibson (3) on the plasma of horses immunized simultane¬ 
ously against diphtheria and tetanus. They state (3, p. 203): 
The observations of Atkinson and Ledingham, so far as we are aware, are the only 
determinations of the quantitative relation of the serumglobulin content and antitoxic 
potency throughout the course of immunization. The subject is of extreme impor¬ 
tance because of the constant association of the antisubstance with the serumglobulin. 
Hartley’s work on this subject (5) was published more recently (see 
p. 56 ). 
The following quotation from the work of Banzhaf and Gibson (3, p. 
206) will serve the purpose of briefly summarizing some of the more 
important results obtained by the above investigators: 
While the greatest rise in the serumglobulin was usually coincident with maximum 
antitoxic potency, as already pointed out, the extent of this increase in the serum¬ 
globulin was practically independent of the antitoxic potency when the results on 
more than one horse were contrasted. There may be, then, no relation between the 
absolute or percentage increase of the serumglobulin and the antitoxic potency in 
the plasma of different horses. The increase in the serumglobulin of refractory horses 
may surpass that in the plasma of some of those yielding a high antitoxin. 
Obviously a study of anthrax serum during immunization was desir¬ 
able for at least two reasons: (1) To ascertain whether the increase in 
potency was accompanied by an increase in globulin, and (2) in so far 
as the Bacillus anthracis does not form a soluble toxin while the bacilli 
of diphtheria and tetanus do, the serum changes would undoubtedly 
throw light on both the practical and theoretical knowledge of many 
related problems in immunity. 
In figure 1 the analyses of the serums of horses 48 and 96 and mules 
148 and 149 are plotted against the time of bleeding. The immuniza- 
