KAHLENBERO — THEORY - OF ELECTROLYTIC DI880CIATION. 351 



eminently one for the chemist. Each solution will have to be 

 examined separately and then it will appear that chemically 

 analogous solutes in the same solvent will have a similar be- 

 havior, the closeness of the agreement being determined by the 

 degree of the analogy; and finally from such a study of solu- 

 tions, which can and should begin with the most concentrated, 

 the behavior of the most dilute solutions will appear as a limit- 

 ing case, — and then we shall see the present theory of solutions 

 in its true relation to the facts. And finally, as far as the an- 

 swer to the question, — What must be the relation between solvent 

 and dissolved substance in order that the resulting solutions 

 may conduct electricity? 1 — is concerned, we are unfortunately 

 as yet in the dark ; just as we do not know why certain solids 

 conduct electricity and others do not. The essence of electrical 

 conduction in electrolvtes and that in metals is, after all, not so 

 radically different as is frequently supposed. The further ex- 

 perimental investigation of the general problem of electrical 

 conduction will, let us hope, ere long give us the true key to the 

 situation. 



Laboratory of Physical Chemistry, 



University of Wisconsin, 



Madison, May, 1901. 



^The investigation of D. Konowalow [Wied. Ann. 49, 733 (1893)] are of special 

 interest in this connection. 



