NATURE OF ANTITOXIC ACTION 565 



with equal affinity. The presence of epitoxoids would manifestly 

 explain the above phenomenon. The L dose would represent 

 toxin + epitoxoid molecules all united to antitoxin molecules, and 

 the addition of another M.L.D. of toxin would not result in 

 there being a free fatal dose, but in the added toxin taking the 

 place of epitoxoid. Several lethal doses would need to be added 

 before the mixture was sufficient to produce a fatal result that is, 

 L t - L would equal several M.L.D's. 



Ehrlich observed another fact strongly in favour of the 

 existence of toxoids, namely, that in the course of time the 

 toxin might become much weakened, so that in one case observed 

 the M.L.D. came to be three times the original fatal dose, and 

 still the amount of antitoxin necessary to neutralise it completely 

 was the same as before. Ehrlich also investigated the effects of 

 partial neutralisation of the L amount of toxin that is, he added 

 to this amount different fractions of an immunity unit and 

 estimated the toxicity of the mixture. He found by this 

 method that the neutralisation of the toxin did not take place 

 gradually, but as if there were distinct bodies present with 

 different combining affinities the graphic representation of the 

 effects of the mixture not being a curve but a step-stair line. 

 Thus he distinguished proto-, deutero-, and trito-toxins (with 

 corresponding toxoids). It will thus be seen that Ehrlich re- 

 gards the combination toxin-antitoxin to be a firm one, and that 

 the neutralisation phenomena are to be explained by the 

 complicated constitution of the crude toxin. 



The chief criticism of Ehrlich's views has come from the 

 important work of Madsen and Arrhenius. Their main con- 

 tention is that the toxin-antitoxin combination is not a firm one 

 but a reversible one, and is governed by the laws of physical 

 chemistry. For example, in the case of a mixture of ammonia 

 and boracic acid (i.e., of a weak base and a weak acid) in 

 solution, there is a constant relation between the amounts of 

 each of the substances in the free condition and the amounts 

 in combination, the combination is reversible, so that if some 

 of the free ammonia were removed a certain amount of the com- 

 bined ammonia would become dissociated to take its place; 

 further, if to the mixture, in a state of equilibrium, more 

 ammonia or more boracic acid were added, part would re- 

 main free while part would combine. Accordingly, if toxin 

 and antitoxin behaved in a similar manner, an explanation of the 

 Ehrlich phenomenon would be afforded. Madsen and Arrhenius 

 have worked out Jthe question in the case of a great many toxins, 

 and find that the graphic representation of neutralisation is in 



