76 Studies of Diffusion through Rubber Membranes [Sept. 



the membrane, but that occasionally it did not. The reason for such 

 variations in the action of the ether could not be conveniently ascer- 

 tained at the time. 



The prompt Perforation of the collodion bags in our several 

 attempts, as stated above, to determine the diffusibihty through col- 

 lodion of the alcohol-ether soluble protein, recalled Dr. Gies' pre- 

 vious experiences and led him to suspect that the alcohol, in the Solu- 

 tions employed by us, was responsible for the observed destructive 

 effects on the collodion membrane in these experiments. He be- 

 lieved, also, that the previous variations in the action of ether on 

 collodion in dialysis experiments, as already related, were due to 

 differences in the degrees of purity of the ether employed. At Dr. 

 Gies' request, therefore, I made direct tests of the solvent action of 

 ether containing alcohol, and various other substances related in one 

 way or another to alcohol and ether. 



Collodion bags were made, in test tubes, from U. S. P. col- 

 lodion.^*^ It was found that such bags were not perforated by abso- 

 lute ether when it was poured into them 10 minutes after their re- 

 moval from the tubes, i. e., after fairly complete evaporation of the 

 residual alcohol. The time required for the evaporation of the 

 residual alcohol is dependent on the prevailing temperature. At low 

 temperatures the alcohol disappears from the collodion membrane 

 very slowly. Common ether (Merck's 0.720 sp. gr.), however, 

 when poured into such bags, passed through them almost immedi- 

 ately, with general Solution of the collodion, even after 2 hours of 

 preliminary exposure of the bag outside the mould. In the first 

 tests of the effects of alcohol it was found that absolute ether con- 

 taining 1.5 per Cent, or more of added absolute alcohol promptly 

 penetrated the bags. 



In a series of more careful tests of absolute ether containing 

 various percentages of added absolute alcohol, it was found that the 

 bags were penetrated promptly by ether containing more than 1,25 

 per Cent, of alcohol, but that the mixture containing 1.25 per cent. 

 of alcohol acted more slowly. Ether containing less than 1.25 ^r 

 cent. of alcohol exhibited no destructive action. 



Qualitative tests showed that acetone, acetaldehyde, ethyl acetate, 

 methyl alcohol and glacial acetic acid attack and penetrate collodion 



"An ether Solution containing alcohol. 



