200 PUBLIC HEALTH BACTERIOLOGY 



reaction. The standardization of diphtheria toxin and 

 antitoxin was the next step. Von Behring called a toxin 

 containing ioo minimum lethal doses (for a 250 grm. guinea- 

 pig) in 1 c.c, a "normal toxin solution" (D.T.N^M 250 ), 

 and a serum capable of neutralizing it c.c. for c.c, a 

 " normal antitoxin " or an " antitoxin unit." 



Ehrlich, in working at the subject, more exactly 

 measured the toxin unit by introducing a time-limit, 

 namely, that one unit must kill the guinea-pig in 4 to 5 

 days. He also varied von Behring's method of testing 

 the antitoxin, by first mixing the toxin and antitoxin 

 outside the body, and thereafter injecting ; whereas von 

 Behring injected them separately and at different parts. 

 He prepared in this way an antitoxin, which he kept in a 

 stable condition by drying in a vacuum and preserving 

 in the dark in a dry atmosphere and at a low temperature. 

 With this antitoxin he is able to standardize new toxins, 

 and from them new antitoxins. In the course of this 

 work he made some discoveries. In the first place, while 

 the death of a guinea-pig in 4 to 5 days gave a fair measure 

 of 1 toxin unit, when 100 such units were mixed with the 

 amount of antitoxin necessary to neutralize, namely, 

 1 antitoxin unit (which was determined from previous 

 measurement), and injected into a guinea-pig, it was not 

 easy to estimate whether there was exact neutralization, 

 or less, or more. If the antitoxin were markedly insufficient 

 to neutralize the toxin, then symptoms such as paralysis, 

 etc., would arise which would proclaim this. But the 

 conditions of the experiment were such that no marked 

 signs could be expected. The further test, therefore, 

 was devised, namely to find the amount of toxin which, 

 plus 1 unit of antitoxin would still be able to kill a guinea- 

 pig, on injection, in 4 to 5 days. (When a new serum is to 

 be standardized, the amount of serum which, mixed with 

 this last-mentioned amount of toxin, just suffices to prevent 

 the death of the guinea-pig before 4 days, is taken as one 

 unit of antitoxin.) Theoretically one might have expected 

 that, if 1 toxin unit killed a 250 grm. guinea-pig in 4 to 5 

 days, and 1 antitoxin unit exactly neutralized 100 toxin 

 units, the injection of 1 antitoxin unit mixed with 101 

 toxin units would have left 1 toxin unit free to have killed 



