JACQUES LOEB 481 



is not ionized at its isoelectric point and does not combine with salt), 

 and that there is no additional source of p.d. which might be ascribed 

 to adsorption. Figs. 4 and 5 give the transport curves for cane-sugar, 

 NaCl, KCl, LiCl, MgClz, MgS04, BaCla, Ce(N03)3, and Na4Fe(CN)6 

 at pH 4.7; i.e., when gelatin is non-ionized. There cannot be the 

 least doubt that the curves for all these salts (with the exception of 

 Ce(N03)3 and Na4Fe(CN)6) are of the nature of the cane-sugar curve; 

 i.e., they show only that part of the curve which corresponds to the 

 second rise of the transport curves in Figs. 1 and 3, and which must be 

 ascribed chiefly, if not exclusively, to the osmotic forces. It might be 

 added that the curve for Naa oxlate does not commence to rise until 

 the concentration of the salt is m/16. The initial rise and drop of 

 the transport curves in Figs. 1 and 3, which is the expression of the 

 electrical forces, is entirely lacking in all the curves at the isoelectric 

 point of gelatin, Figs. 4 and 5, with the exception of the curves for 

 Ce(N03)3 and Na4Fe(CN)6, to which we shall now give our attention. 



In previous publications the writer has already called attention to 

 the fact that on the alkaline side of the isoelectric point the presence 

 of salts with a trivalent cation has the effect of reversing the sign of 

 the P.D. between gelatin and water. When the pH is > 4.7, i.e. 

 when gelatin exists in the form of Na gelatinate, the Donnan equili- 

 brium causes the expulsion of NaOH from the gelatin into the bound- 

 ing liquid with the result that gelatin assumes a negative and the 

 bounding liquid a positive charge. When, however, some CeCls 

 or LaCls is added the liquid assumes a negative and the gelatin a 

 positive charge.^ This reversal of the sign of charge by trivalent 

 cations had been discovered by Perrin in his experiments on electrical 

 endosmose.^" The reversal may either be due to a reaction between 

 Ce(N03)3 and isoelectric gelatin, in which a compound is formed which 

 dissociates into a complex positively charged gelatin-Ce cation and a 

 negative ion, presumably NO3, or the addition of the salt brings 

 the pH to a value below 4.7. In either case the behavior of the 

 curve for Ce(N03)3 in Fig. 4 becomes clear. When Ce(N03)3 solu- 

 tions of pH 4.7 are separated by a collodion-gelatin membrane (of 



9 Loeb, J., /. Gen. Physiol, 1919-20, ii, 659. 



^"Perrin, J., /. chim. physique, 1904, ii, 601 ; 1905, iii, 50. Notice sur les litres 

 et travaux scicntifiques de M. Jean Perrin, Paris, 1918. 



