104 A. P. MATHEWS 



words, ions like that in Fig. 2. As a matter of fact, if we try to 

 apply this formula to the results on toxicity, a discrepancy between 

 the response of protoplasm and colloidal albumin to salts is at once 

 noticed. I have already called attention to this discrepancy. The 

 discrepancy is this: While in the colloidal solutions the opposite 

 action of the ions, dissolving and precipitating, is clearly apparent, 



and that opposite action is propor- 

 tional to the ionic potentials of the 



*PROTciD\ _ anion and cation respectively, for 

 protoplasm in general and for the 



COLLOID I ferment studied by McGuigan a dif- 



\^ ./ ferent relationship is seen in that, 



instead of the positive ion counter- 



actin b y its ener y content the 



leaving the colloid both negative and positive negative ion, aS it OUght to do On 

 at different places. 



the theory developed, a summation 



of effects is noticed. The iodides of the metals, instead of being 

 less poisonous than the chlorides, as they should be, are more 

 poisonous. The explanation of these facts is to be sought on the 

 basis of the differently charged colloids present. 



If these amphoter colloidal particles exist in protoplasm, or if we 

 are dealing in protoplasm with a mixture of both negative and positive 

 colloids, each ion will tend to precipitate. In the twin ions, if potas- 

 sium is replaced by an ion of greater potential energy, the proteid will 

 tend to be precipitated, since the negative part will be partially neu- 

 tralized; and similarly if chlorine is replaced by an anion of greater 

 potential. Both ions, therefore, will exert an action in the same direc- 

 tion, and there will be a summation of action in this case instead of a 

 difference. The formula for toxicity would become : 



Total action = (E ca tion salt" -^cation colloid) + (^anion salt" ^anion colloid) 

 = Vacation salt T -^anion salt) (-^cation colloid T -^anion colloid) 5 



or, writing E' c and E' a for the ionic potentials of the anion and cation 

 of the salt introduced, and E c and E a for the ionic potentials of the 

 ions bound to the colloid, 



log FWU4+4) - (E c +E a )]+C 



