JACQUES LOEB 469 



flowing junction of Lamb and Larson^ was used in a simplified and 

 also less accurate form, which, however, gave results of sufficient 

 accuracy for our purpose. Table III gives some of the results showing 

 that the order of efficiency of the various salts and the influence of 

 concentration are the same in diffusion potentials as in the p.d. across 

 the membrane. The diffusion potentials are, however, considerably 

 lower than the p.d. across the membrane (Tables I and II). The 

 diffusion potentials depend on the difference in the relative velocity 

 of the oppositely charged ions of a salt. A comparison of the values 

 in Table I and Table III suggests that the source of the p.d. across 

 the membrane is the same as that of the diffusion potential if we 

 assume that at pH 3.0 the cations experience a greater retardation 

 in the diffusion through protein films than anions. 



The figures in Table I, II, or III do, however, not explain the drop 

 in the curves of Fig. 1 which occurs when the concentration of the 

 salt reaches m/32. The cause for this depression lies probably in 

 the influence of the concentration of the salt on the value of e; i.e., 

 P.D. between gelatin chloride and water inside the pores of the mem- 

 brane. It has been shown in previous experiments that salts depress 

 the P.D. between gelatin particles and surrounding liquid and that 

 the reason for this depression is furnished by the Donnan theory 

 of membrane equilibria.^ The method of these experiments was 

 briefly as follows : 



1 gm. of fine particles of powdered gelatin rendered first isoelectric 

 and of an equal size of grain was put into 200 cc. of various concen- 

 trations of a salt (NaCl, CaCl2, etc.) in water and containing 8 cc. 

 of 0.1 N HCl per 100 cc. The gelatin remained in this solution for 

 2 hours at 20°C. under frequent stirring. The suspension was then 

 put on a filter and the gelatin freed from the supernatant liquid. 

 The gelatin was then melted by heating to 45° and cooled rapidly 

 to cause solidification and the p.d. between the gelatin and the super- 

 natant liquid was then measured with the aid of a Compton elec- 

 trometer. Table IV gives the result. The reader will notice that 

 in no case do any of the salts cause a rise in the p.d. between gelatin 

 and liquid. The observed p.d. could, however, be calculated with a 

 fair degree of accuracy from Donnan's equilibrium equation. 



'Lamb, A. B., and Larson, A. T., /. Am. Chan. Soc, 1920, xlii, 229. 



