20 
diphtheria antitoxic serum, say 0.01 c. c., was injected subcutaneously 
into a guinea pig. Twelve hours later, 0.5 c. c. of a fresh virulent 
culture of the diphtheria bacillus was inoculated into the guinea pig, 
weighing, say, 300 grains. If the guinea pig lived, 0.01 c. c. of 
the serum was considered sufficienth^ active to protect 300 grams of 
guinea pig, and 1 c. c. of the serum w'ould consequent!}" protect 30,000 
grams of guinea pig, and the strength of the serum was expressed as 
1: 30,000. The strength of a serum according to this method was 
expressed in such figures as 1: 100,000 or 1: 500,000. 
Control animals were inoculated to insure the virulence of the cul- 
ture used. The culture used was considered sutficiently virulent if 
the control guinea pig died within thirty hours. 
A serum with an immunizing strength of 1: 100,000 meant that 1 c. c. 
would protect 100, OOO grams of guinea pig; and a serum whose 
strength was 1: 500,000 meant that 1 c. c. of the serum would protect 
500,000 grams of guinea pig against a certainly lethal dose of diph- 
theria culture or later the toxine when injected in accordance with 
the above procedure. 
Madsen® showed as a result of comparative tests that the German 
method of expressing the strength of antitoxic sera in antitoxin units 
was quicker, cheaper, easier, and more accurate than the French method 
devised by Roux. He showed that with the French method it is often 
impossible to determine differences as great as 1: 100,000 and 1: 200,000. 
This method was finally abandoned b}" Roux himself in favor of the 
German procedure. 
In the earlier methods of testing the strength of diphtheria anti- 
toxin, according to both the methods of Behring and Roux, the two 
substances were always inoculated separatel}" into a guinea pig, usually 
in different places. Sometimes the toxine and antitoxin were inocu- 
lated at different times in order to make a distinction between the 
immunizing power of a serum and its curative power. 
The method devised by Ehrlich in collaboration with Kossel and 
Wassermann in 1891^ was based upon an entirely new principle. The 
.scientific researches of Behring and Kitasato had previously established 
the fact that tetanus poison and its antitoxin neutralize each other in 
a test tube outside of the bod}-. Ehrlich, Kossel, and Wassermann satis- 
fied themselves that this neutralization between the diphtheria poison 
and its antibod}" takes place at once when the two substances are mixed 
in a test tube, and they believed at that time (1894) that it took place 
in accordance with the law of simple proportions. 
^Madsen, Thorvald; Ueber messung der starke des antidiphtherischen serums. 
Zeitschr. f. hyg., Leipzig, v. 24, pp. 425-442. 
^L'eber Gewinnung und Verwendung des Diphtherieheilserums. Deut. med. 
Woch., 1894, no. 16. 
