INFLUENCE OF A SLIGHT MODIFICATION OF THE 



COLLODION MEMBRANE ON THE SIGN OF 



THE ELECTRIFICATION OF WATER. 



By JACQUES LOEB. 

 (From the Laboratories of The Rockefeller Institute for Medical Research.) 



(Received for publication, November 25, 1919.) 



I. 



When we separate a solution of an electrolyte from pure water by 

 a collodion bag the water will as a rule diffuse into the solution. In 

 a preceding paper^ it was shown that the forces determining this 

 diffusion are different for concentrations of electrolytes below or above 

 a certain value (which is for certain sodium salts about m/16). The 

 forces causing the diffusion of water into the solution below this 

 critical concentration are predominantly electrical, while the forces 

 causing the diffusion above the critical concentration are predomi- 

 nantly (or perhaps exclusively) molecular. 



The electrical forces of diffusion depend upon the sign, the valency, 

 a third property of the ions (which we arbitrarily designated as their 

 radius), and in addition upon the concentration of the ions in solution. 

 There is still another variable to be considered; namely, the nature of 

 the membrane. We have already called attention to the fact that a 

 collodion membrane which has once been treated with a gelatin solu- 

 tion shows a different osmotic behavior from a membrane not treated 

 with gelatin. The gelatin treatment consisted in this that the col- 

 lodion bags with which the experiments were made were filled over 

 night with a 1 per cent gelatin solution (isoelectric or nearly isoelectric). 

 The next day the gelatin solution was poured off and the interior of 

 the bags washed out about six times or more with warm water to 

 remove as much gelatin as possible. They were then put for days into 

 water to dissolve still more and remove the last traces of gelatin. 



^ Loeb, J., /. Gen. Physiol., 1919-20, ii, 173. 



255 



