KAHLENBERQ — THEORY OF ELECTROLYTIC DISSOCIATION. 303 



t 



elektrolyte 4-7, was ganz aus dem Rahmen des Geforderten herausfallt 

 und nicht ohne Weiteres mit der iiblichen Auffassung vereinbar ist*'. 



Franklin and Kraus 7 have found that while NEL^C^, 

 !N"a!N"0 3 , and KI dissolved in liquid ammonia are excellent con- 

 ductors of electricity, the boiling-points of the solutions, like' 

 those observed by Walden in S0 2 , are not nearly as high as they 

 ought to be on the assumption that electrolytic dissociation takes 

 place. Again lately, Xicolo Castoro 1 found by means of the 

 freezing-point method that AgjNT0 3 , CdCl 2 , HgCl 2 , and ZnCl 2 

 have normal molecular weights in urethane. I have made a few 

 preliminary tests on the first three of these salts which show 

 that their solutions in urethane are conductors of electricity. 

 Verv recentlv Lines 2 found the molecular weights of succinic, 

 salicylic, and tartaric acids to be normal in pyridine, according 

 to the boiling-point method. These acids undoubtedly form 

 salts which re-dissolve in the excess of the solvent. Preliminary 

 tests have assured me that all three of these solutions are fairly 

 good conductors of electricity. The tartaric acid solution con- 

 ducts best, as might have been expected. 



While in many cases the molecular conductivity of non- 

 aqueous solutions increases with the dilution, this increase is 

 generally relatively slight. It has generally been impossible to 

 calculate the degree of dissociation of substances in non-aqueous 

 solutions from the conductivitv, because in these solutions the 

 molecular conductivity commonly either diminishes with the in- 

 crease of dilution, or it increases slightly with the dilution, ex- 

 hibiting no tendency to reach a maximum, or it remains prac- 

 tically constant, or soon reaches a maximum with so low a value 

 that completeness of dissociation can not consistently be as- 

 sumed. These facts will become evident to the reader by a 

 perusal of the figures contained in the original articles above 

 cited. I have already discussed at some length the difficulty of 

 calculating the degree of dissociation in non-aqueous solutions, 3 

 and shall therefore simply add here that the calculation of the 



^az. chim. Ital. 28 II, 317 (1898). 



*Jour. chem. Soc. London 79, 261 (1901). 



•Jour. Phys. Chem. 3, 395 (1899). 



