126 PHYSICAL CHEMISTRY IN MEDICINE. 



product of time when the toxin is kept for a long time, such 

 as the toxoids. The differences observed in the toxicities 

 of a toxin when only partially saturated with antitoxin, 

 and those observed in the behavior of different animals 

 toward the neutralized toxins, all furnish important 

 support for these assumptions. But we must look upon 

 it as a fact well established through investigations on 

 colloids, that in their changes in state a part of their 

 reactions may be influenced, while another part may 

 remain untouched; that through the mixture of colloids 

 which neutralize each other manifold new, in no sense 

 preformed, aggregates can be produced, and that such 

 aggregates may at one time allow the effect of the one 

 colloid, in this case the toxin, at another time the other, 

 the antitoxin, to become apparent. Great differences 

 must in consequence result in the intensity and in the 

 picture of the intoxication in different animals. Against 

 EHRLICH'S theory ARRHENIUS and MADSEN, as well as 

 GRUBER and PIRQUET, have set up the idea that in 

 toxin- antitoxin mixtures we always have to do with a 

 dissociation of compounds having only a weak affinity 

 for each other. This conception, which originated from 

 a too far-reaching generalization of a special case, allows 

 only of the existence of completely neutralized aggregates, 

 beside traces of their components, and is, therefore, in- 

 capable of explaining the multitude of experimental facts 

 which have been obtained in the study of different toxins, 

 more especially diphtheria toxin. Against this inade- 

 quate assumption arose the strength of the EHRLICH 

 theory, which has rendered the great service of having 

 been the first to fix in the minds of investigators the 

 newly discovered and scarcely calculable varieties of 



