222 MICROBES AND TOXINS 



reverse of the chemical theory of Ehrlich ; the one simplifies 

 where the other complicates. 



In the chemical theory the same quantity of antitoxin ought 

 to combine with the same quantity of toxin, so, to account for 

 the irregularities in the actual facts, there have been introduced 

 the hypotheses of the haptophore group separate from the 

 toxophore, of the toxones, toxoids, &c. On the contrary, 

 Bordet supposes that the antitoxin really unites with the toxin 

 in varying proportions : the toxin can fix, can, so to speak, 

 dye itself with toxin in greater or less amount just as starch 

 can take up varying quantities of iodine and become thereby 

 stained a more or less dark blue. In the same way, in 

 haemolysis the corpuscles can absorb variable amounts of the 

 active substance according to the concentration of the solutions 

 and the duration of the contact. The distance of this idea 

 from the theory of chemical equivalents is apparent. When a 

 given quantity of toxin is mixed with a quantity of antitoxin 

 insufficient for complete neutralization, what occurs is not a 

 monopolization of the antitoxin by a portion of the toxin mole- 

 cules, forming a complete combination with it while the rest of 

 the toxin remains free (i.e., the chemical conception). On the 

 contrary, the antitoxin is equally distributed over all the toxin 

 present, so that the latter is attenuated throughout and presents 

 a diminished activity. To return to the same analogy, it is 

 faintly dyed. It can produce toxic effects qualitatively different 

 from those due to an intact toxin or a toxin completely 

 neutralized without necessitating the hypothesis of a special 

 chemical condition (toxones).^ 



The phenomenon of Danysz-Dungern (the antitoxin has a 

 different action on toxin when the mixture of the same 

 quantities is made at one instead of several additions) does not 

 compel the hypothesis that in toxin there are several chemically 



^ The fact first observed by Ehrlich, namely, the difiSculty of preparing 

 exactly neutral mixtures of toxin and antitoxin, can thus be easily explained. 

 The effects produced by a haemolysin more or less neutralized and the 

 experiments of Grossberger and Schattenfroth on the toxin and antitoxin of 

 the bacillus of quarter-evil have confirmed Bordet 's views of the nature of 

 the toxin + antitoxin reaction. 



