400 ION SERIES AND PROTEINS. Ill 



true for the LaCls curve. From this it follows that the depressing 

 effect of these three salts on gelatin chloride is practically exclusively 

 a function of the concentration of the CI ion, while no effect of the 

 cation seems to be noticeable. In all these cases the pH of the gelatin 

 solution was not altered by the addition of the salt. 



When we prepare 1 per cent solutions of gelatin chloride of pH 3.0 

 in solutions of Na salts with the anion of a weaker acid, e.g. Na2 oxa- 

 late, Na4Fe(CN)6, the pH is increased and we are in danger of attrib- 

 uting erroneously a depressing effect to the anion which in reality 

 is caused by the increase in pH. In Fig. 6 the effects of the addition 

 of equal concentrations of NaCl, Na2S04, and Na4Fe(CN)6 on gelatin 

 chloride of pH = 3.0 are plotted. In the case of Na4Fe(CN)6 only 

 the lowest concentrations, from m/8,192 to m/1,024, could be used, 

 since in these only did the pH of the protein solution remain = 3.0. 

 Fig. 6 shows that the depressing effect of these salts increases rapidly 

 with the valency of the anion. When the concentration of the salt 

 was only m/ 1,024 a drop in the viscosity was already noticeable. 

 This drop was small in the case of NaCl (from 2.8 to 2.6), was greater 

 in the case of Na2S04 (from 2.8 to 2.35), and considerably greater in 

 the case of Na4Fe(CN)6 (from 2.8 to 1.5). The objection might be 

 raised that since Na2S04 has twice as many cations as NaCl of the same 

 concentration and Na4Fe(CN)6 has four times as many cations, it was 

 the difference in the concentration of the cations which caused the 

 difference in the drop. This is refuted by the fact that Na2S04 

 causes a drop to 1.8 at a concentration of m/256 while NaCl causes the 

 same drop at a concentration of above m/64 which is about four times 

 as high. If the concentration of the cation were responsible for the 

 drop the two concentrations should be as 1 : 2. Na4Fe(CN)6 causes the 

 same drop of the viscosity to 1.8 at a concentration less than m/1,024. 

 Hence the concentration of Na4Fe(CN)6 required to cause the same 

 diminution of the specific viscosity as that caused by m/64 NaCl is 

 less than yV of the latter, while it should be at the least only ^ if the 

 cation Vv^ere responsible for the drop. The depressing effect of the 

 anion seems to increase almost in proportion to the square of its 

 valency, as the Hardy- Whetham rule demands. 



We have selected viscosity experiments, but experiments on os- 

 motic pressure and on swelling lead to the same formulation of the 

 difference in the effect of acids and salts. 



