THE ACTION OF ETHER OX PROTOPLASM. 463 



But there is also another important fact to be considered. 

 Separating the cells of Laminaria is a network of cell walls. 

 This is composed of cellulose and cellulose-like material which 

 offers little resistance to dissolved substances and is obviously 

 one of the main paths for an electric current. The adsorptive 

 powers of cellulose are well known, although it apparently 

 adsorbs electrolytes much more readily than non-polar com- 

 pounds. In the finely divided condition in which the cellulose 

 occurs in the cell walls it is not improbable that substances of 

 low surface tension like ether, alcohol, and chloroform would be 

 selectively adsorbed on it and more or less concentrated there. 

 We should at least be led to expect this from Gibbs' adsorption 

 equation. Moreover the surfaces of the cells would also tend to 

 adsorb the anesthetic. If such adsorptive processes occur, then 

 the conductivity of the cellulose framework as well as the con- 

 ductivity of the cell surfaces would doubtless be decreased, and 

 the decrease in conductivity would be greater than that which 

 occurred in the mass of the solution. If this is true then it 

 might in itself account for the variations in conductivity found 

 by Osterhout. 



It is interesting to note that the experiments of McClendon 

 (8) on the diffusion of electrolytes from anesthetized pike eggs 

 also fail to take cognizance of the direct effect of the anesthetic 

 on the diffusion rate. It might be thought that the difference 

 in diffusion rate would also explain McClendon's experiments. 

 This is not the case. The presence of 2 per cent, alcohol would 

 lower the diffusion rate less than 8 per cent. McClendon 

 apparently found a lowering of as much as 50 per cent, in the 

 total amount of chlorides diffused from the eggs. But his results 

 are not as trustworthy as it might at first sight be thought. 

 McClendon determined the diffused chlorides nephelometrically 

 as silver chloride, precipitating them with silver nitrate. Now 

 the precipitation of silver chloride varies under diverse conditions; 

 there is a well known tendency for it to go into colloidal solution. 

 The presence of alcohol either alone, or in conjunction with the 

 albuminous substances also present, might act as a peptizing 

 agent and hinder the precipitation of the silver chloride. Alcohol 

 occasionally acts as a peptizing agent, and the precipitation of 



