158 THEORY OF COLLOIDAL BEHAVIOR 



by the colloidal particles, and that the salt ion with the opposite 

 sign of charge to that of the colloidal particle diminishes the 

 charge of the latter, while the salt ion with the same sign of charge 

 as the colloidal particle increases the charge of the latter, and the 

 more the higher the valency of the ion of the salt. The idea that 

 such an adsorption occurs is definitely refuted by the experiments 

 discussed in Chap. II (see Figs. 1 and 2). It might, however, 

 be possible that the ion with the same sign of charge as the 

 colloidal particle might increase the charge of the colloidal 

 particle in some other way than through adsorption, and it was 

 necessary to test this possibility, which has found acceptance on 

 the part of some chemists. The writer's experiments on anomal- 

 ous osmosis have shown that when a dilute solution of a salt is 

 separated from pure water by a collodion membrane coated 

 with gelatin, if the salt solution and the water are brought to the 

 same pH by adding an acid, e.g., HNO 3 , the potential difference 

 on the two opposite sides of the membrane increases with the 

 valency of the cation of the salt used, i.e., in the order 



Ce>Ca>Na 



This influence was found to be due to a diffusion potential. 1 

 Nevertheless it seemed necessary to determine whether or not 

 these cations influenced the charge of suspended particles of 

 gelatin chloride in the same way. If this were true, the depress- 

 ing effect of CeCl 3 on the charge of the micellae should be less 

 than the depressing effect of CaCl 2 , and the depressing effect of 

 CaCl 2 should be less than the depressing effect of Nad, provided 

 the pH is on the acid side of the isoelectric point. 



If, on the other hand, the Donnan effect alone determines the 

 depressing effect of the salt on the charge of the suspended 

 particles of gelatin, this depressing effect should be exclusively 

 due to the anion of the salt on the acid side of the isoelectric 

 point of the gelatin, while the cation of the salt should have no 

 effect. This follows from the discussion in the preceding chapter 

 according to which the P.D. between gelatin chloride solution 



and water is determined by the value of log (1 + -), where z 



y 

 and y are the anions. The cations do not enter into the term on 



1 LOEB, J., J. Gen. Physiol., vol. 4, pp. 213, 463, 1921-22. 



