11)11] on P/i //steal Ghemixtrif and Ihr Dmirmr of hmaanitij. •_>2;5 



111 these examples the form of (Juldberg and Waat^e's law 

 indicates that one molecule of poison iind one molecule of a/ntitoxin 

 give two molecules of the compound toxin-antitoxin. In other cases, 

 especially with the so-called nerve-poisons, e.g. the ricin and the nerve- 

 poisons contained in the snake-venoms, the formula of equilibrium 

 indicates that three molecules of poison combine with three mole- 

 cules of antitoxin to form two molecules of the compound. This has 

 been regarded as indicating that the hemolysin of the snake-venoms 

 is different from the corresponding nerve-poison, but this conclusion 

 is not necessary ; it might be that the compound toxin-antitoxin 

 possesses a molecular weight 4/3 times as high when it enters in, i.e. 

 acts upon the nervous system, as when it enters in the red blood- 

 corpuscles. A similar remark may be made regarding the nerve- 

 poison and agglutinating poison in ricin. There is also another 

 possibility, namely, that the poison as well as the antitoxin, when 

 dissolved in the blood-corpuscles, possess a molecular weight wliich 

 is only t>vo-thirds of that characteristic of their solution in the 

 nervous substance, and that their compound has a double molecular 

 weight when dissolved in the nerves than when dissolved in the red 

 blood-corpuscles. Many experiments have been made to separate 

 the blood-poison from the nerve-poison, and through the admix- 

 ture of different substances it has been found possible to change 

 the toxicity in regard to blood in a different proportion from that in 

 regard to nervous substance. It must here be borne in mind that 

 these admixtures may act as sensibilisators in regard to tlie blood- 

 corpuscles, but not or only in a lesser degree in regard to the ner\'ous 

 substance, which may explain the effect without supposing that the 

 poison is different in the two cases. This case would then be similar 

 to that of mercuric chloride, which acts both as hemolysing and as 

 coagulating red blood-corpuscles, and through the admixture of 

 lecithin to the blood becomes less coagulating, and thereby more 

 strongly hemolysing. Of course, no one would conclude from this 

 fact that mercuric chloride contains two different poisons. The 

 circumstance that in the preparation of antibodies to a certain poison, 

 for instance, ricin, the strength of the antitoxin in regard to blood 

 is generally nearly proportional to its strength in regard to nervous 



