TOXINS AND ANTITOXINS 207 



he designates as the "haptophore" or " anchoring" group. The other, 

 the one by which the toxin molecule exerts its deleterious action, must 

 be more easily changed or destroyed; this he calls the "toxophore" 

 or " poison " group. In the altered toxin-bouillon in which a part of 

 the poisonous action has been lost while the antitoxin-neutralizing power 

 is intact, the toxophore group of some of the toxin must have been 

 changed or destroyed. Such altered toxin he speaks of as "toxoid." 



In support of this hypothesis and for the purpose of perfecting the 

 methods of standardization, Ehrlich was led to determine, for a large 

 variety of specimens of diphtheria toxin, the precise quantity, in 

 cubic centimeters, which was necessary to neutralize exactly one unit 

 of his standard antitoxin. This he accomplished by making a series of 

 toxin-antitoxin mixtures, in each of which the quantity of antitoxin 

 was exactly one unit, while the amount of toxin was gradually increased. 

 These mixtures were injected into guinea-pigs of 250 grams weight. 

 It is self-evident that in such an experiment the mixtures containing 

 the smaller quantities of toxin would have no effect upon the guinea- 

 pigs. Soon, however, a mixture would be reached in which toxin would 

 be sufficiently in excess of antitoxin to produce the symptoms of slight 

 poisoning, as evidenced in local edema, rise of temperature, etc. The 

 largest quantity of toxin which could be added without producing such 

 symptoms was then regarded as exactly neutralizing one antitoxin unit. 

 This quantity of toxin Ehrlich speaks of as "Limes zero" (Limes = 

 threshold) or, briefly, " L ." 



For instance : 



One antitoxin unit + 0.6 c.c. toxin No symptoms of poisoning. 



0.8 c.c " 



" 0.9 c.c " " 



" 1. c.c. " " " " 



" 1.1 c.c Local edema. Paralysis in 30 days. 



" 1.2 c.c Death in 10 days. 



In this example, L , therefore, equals 1 c.c. 



It is obvious, however, that because of the great difficulty in esti- 

 mating the very slightest evidences of toxic action in guinea-pigs, a 

 more exact method of standardizing the poisons against antitoxin 

 would be to determine how much toxin would be required to neu- 

 tralize one antitoxin unit and still be sufficiently in excess to cause 

 the death of a guinea-pig of 250 grams in four to five days . This would 

 then correspond to the action of one toxin unit, unmixed with antitoxin. 

 A priori it would seem that this value (expressed by Ehrlich as " Limes 



