586 REVERSAL OF SIGN OF CHARGE 



ions, as was the case in regard to the influence of ions on the density 

 of charge of the membrane.^ If this were the case, SO4 should shift 

 the steep rise of the curve more to the acid side and Ca more in the 

 opposite direction. The curves given in this paper (and many- 

 other curves not published) show that this is not the case. This sup- 

 ports the view that the mechanism for the influence of ions on the 

 density of charge and for the influence of the hydrogen ions on 

 the reversal of the sign of charge is not the same. 



The curves in Fig. 1 show that there are three relative minima of the 

 charge of the membrane. One lies at about pH = 4.0 or shghtly 

 above, when the membrane is neither positively nor negatively 

 charged. The drop leading to this minimum is not an additive func- 

 tion of the oppositely charged ions and differs in this respect from the 

 drop leading to the other two minima, the latter being due to the dim- 

 inution of the density of charge on the membrane caused by that 

 ion which has the opposite sign of charge as the membrane. We 

 have shown in the preceding experiments^ that this depression occurs 

 when the concentration of the electrolyte exceeds a certain value. One 

 of these minima lies on the acid side, namely at a pH less than 2, 

 owing to the fact that the concentration of the HNO3 added is m/100 

 or more. This minimum is due to the depressing effect of the anions 

 (NO3 and SO4) of the solution upon the density of the positive charge 

 of the membrane. The third minimum lies at a pH of about 12 or 

 above and is caused by the depressing effect of the cation of the 

 solution, Na, K, etc., upon the density of the negative charge of the 

 membrane. If we had continued to increase the concentration of acid 

 there would have been another rise of the curve due to the gas pressure 

 effect of the acid solution; and the same might have happened if we 

 had been able to increase the concentrations of the KOH added; 

 strong alkali solutions, however, dissolve the membrane. 



III. Membranes Treated with Casein and Egg Albumin. 



The isoelectric point of casein is, according to MichaeHs,^ identical 

 with that for gelatin, since it lies at a pH of about 4.7. When we treat 

 coUodion membranes over night with a 1 per cent casein solution in- 



^ Michaelis, L., Die Wasserstoffionenkonzentration, Berlin, 1914. 



