580 REVERSAL OF SIGN OF CHARGE 



On the basis of this theory it seems natural that membranes should 

 as a rule assume a negative charge when in contact with water. The 

 fact which requires a further explanation is the possibility of a reversal 

 of this sign of charge. 



Perrin has shown that in certain cases acids are able to cause a 

 membrane in contact with water to be positively charged. The 

 fact that addition of alkali to an acid solution restores the original 

 negative charge to such a membrane was explained by Perrin on 

 the assumption that in an alkaline solution the OH ions are adsorbed. 

 This latter assumption seems unnecessary since the addition of 

 alkali to an acid may merely serve to lower the hydrogen 

 ion concentration below the level required to make the membrane 

 positive. It is, therefore, only necessary to explain why hydrogen 

 ions in sufficient concentration impress a positive charge on so many 

 membranes and the following is a contribution to the solution of this 

 problem. 



We have seen in the preceding papers of this series that a collodion 

 membrane not treated with a protein always assumes a negative 

 charge when in contact with a watery solution, even if this solution 

 is acid. This was demonstrated in two ways, first by electrical endos- 

 mose,^ and, second, by common osmosis.^ The experiments with 

 electrical endosmose do not lend themselves so well to the exact 

 determination of that hydrogen ion concentration at which the mem- 

 brane becomes positive on account of electrolysis and the secondary 

 chemical reactions at the electrodes. The method of common osmosis 

 is free from this source of error. When we separate a watery solution 

 from pure water by a collodion membrane the pure water diffuses 

 into the solution at a rate which is not only a function of the gas pres- 

 sure of the solute but also of the electrostatic forces of the ions in 

 solution. The rate increases with increasing valency of the anion 

 and diminishes with increasing valency of the cation. When, how- 

 ever, the collodion membrane has been treated for some time with a 

 protein it assumes a positive charge when the hydrogen ion concen- 

 tration exceeds a certain value; and in this case the watery phase of 

 the double layer assumes a negative charge. This is proved by the 

 fact that when we separate a watery solution with a sufficiently high 

 hydrogen ion concentration from pure water by a collodion membrane 



