262 INFECTION AND IMMUNITY 



able to immunize against toxin with such mixtures, thereby proving 

 the identity of the haptophore groups of the two substances. The 

 importance of this observation will become more evident in con- 

 nection with the section on the "side-chain theory." 



METHOD OF PARTIAL ABSORPTION OF TOXIN. Ehrlich 17 has gathered 

 more exact data in support of his views from what he terms the 

 "Method of Partial Absorption" of toxin by antitoxin. 



In order to understand this method clearly, it is necessary to 

 remember that Ehrlich 18 believes the union of toxin with antitoxin 

 to take place according to the chemical laws of valency. Just as in 

 H 2 oxygen has an atomic valency of 2 for hydrogen, so, in the 

 fully neutralized toxin-antitoxin compound, he believes antitoxin to 

 have a valency of 200 for toxin. It would require, according to 

 this, 200 T or toxin molecules to satisfy the affinities of one antitoxin 

 molecule. 19 



This belief is based upon the following consideration: In de- 

 termining the LQ dose, or fully neutralized toxin-antitoxin union, 

 Ehrlich, as well as Madsen, found that the number of T units con- 

 tained in such a dose was almost regularly a factor of one hundred, 

 recurring again and again as 25, 33, 50, 75, etc. This pointed to 

 more or less regularity in the deterioration of toxin into toxoid, and 

 to a more or less regular relation of toxin to toxon. Now, as we 

 have seen before, if we could procure a perfectly pure toxin, the 

 L dose plus one toxin unit would give us the L + dose; that is, 

 one toxin unit in excess of full neutralization would suffice to kill 

 a guinea-pig of 250 grams in four to five days. Since a perfectly 

 pure toxin, however, has not been obtainable up to the present 

 time, it is clear that the number of pure toxin bonds contained in 

 L + must be less than the actual number of neutralizing units in the 

 combination, a part of the antitoxin being bound by toxon and 

 toxoid. The actual values obtained for the number of T units in 

 L + has never exceeded 200, and has usually been more than 100, 

 the highest value ascertained by Madsen being 160. Given, there- 

 fore, a combining value which, being a multiple of one hundred, 

 is often more than one hundred, but in an obviously impure state 

 has never reached 200, it is most likely that 200 represents the 

 actual value sought for. 



-''Ehrlich, ' ' Gesammelte Arbeiten zur Immunitatsf orsch., " Berlin, 1904. 



18 Ehrlich, Deut. med. Woch., 1898. 



19 Ehrlich, ' ' Schlussbetrachtungen, ; ' Nothnagel's System. 



