40 MORRIS LOEB 



those readers who will compare the mass of experimental 

 material and will convince themselves of the simple relations 

 which the various phenomena appear to bear toward each 

 other. As far as this test is concerned, the hypothesis will be 

 found to fulfill its purpose. Shall it therefore be accepted? 



The objections which have been raised are twofold, phys- 

 ical and chemical. Oliver J. Lodge concedes the inge- 

 nuity and physical "orthodoxy" of the treatment, and it 

 might appear that his criticisms to the earlier papers have 

 been largely obviated by the further development of the 

 theory; indeed they apply chiefly to points which I have 

 omitted as no longer seeming essential to the theory, such as 

 relations between the viscosity and the friction which a 

 solvent opposes to the motion of the ions. But he appears 

 mainly to object to the neglect of conduction by the solvent, 

 which he believes would explain the rate of transference 

 of the ions, without the assumption of unequal velocities of 

 negative and positive ions. It is doubtful whether such an 

 explanation would also elucidate the troublesome diffu- 

 sion phenomena as well as does "dissociation" in the hands 

 of Nernst. E. Wiedemann thinks that hydrates are the cause 

 of better conduction, and explains osmotic anomalies by the 

 assumption of a polymerization of the solvent. Aside from 

 Planck's proof, on thermodynamic grounds, that such poly- 

 merization would not affect the vapor tension phenomenon 

 in this way, in very dilute solutions, Ostwald pertinently 

 asks why the electrolyte need cause a polymerization which 

 no non-electrolyte does. 



The objections from a chemical standpoint have been 

 chiefly raised by Henry E. Armstrong, and are not all co- 

 gent. The following appear to be the most important at 

 the present time: 



1. Anhydrous hydrochloric acid and pure water do not 



