1889.] Viscosity and Conductivity of Electrolytes. 27 



lytic frictional resistance of about equal amount allows the finding 

 of an absolute size of molecules which approaches the sizes found 

 by other methods." 



But is it necessary to suppose that the individual atoms are 

 moved through the liquid ? May the ions not be parts of complex 

 molecules of salt and solvent which have to be dragged through 

 the liquid? Wiedemann* mentions that such a suggestion has 

 already been made but that it is very risky, yet the doctrine of 

 electrolysis of molecular aggregates as opposed to that of disso- 

 ciated atomic ions has some adherents, and expressions might be 

 quoted from the writings of well-known supporters of the dis- 

 sociation-theory to shew that they acknowledge the necessity for 

 regarding the electrolytic molecule as complex in some special 

 cases. I think it may be well to recall that certain phenomena 

 may naturally lead to the same view, although I am well aware 

 that these phenomena have been otherwise explained in a manner 

 that is accepted as satisfactory. 



The view that electrolysis consists in the convection of elec- 

 tricity by single atoms only, or their chemical representatives, is 

 based upon the resolution of the processes taking place in an 

 electrolytic cell into independent phenomena. Electric endos- 

 mose, the unequal dilution of solution at the electrodes and the 

 deposition of ions are all treated separately. Electric endosmose 

 is regarded as the result of the electrification by contact of the 

 boundary layer of the solution in the porous partitions which 

 divide the cell, and the explanation has been regarded as com- 

 plete since von Helmholtz shewed that the difference of potential 

 of the boundary layer necessary to explain the effect was not 

 more than a few volts. The unequal dilution at the electrodes is 

 explained by the theory of migration of the atomic ions with 

 unequal velocities, and this explanation has received strong con- 

 firmation from Kohlrausch's calculation of resistance, based on 

 these atomic ionic velocities and Lodge's experimental verification 

 of the calculation in the case of hydrogen. But, apart from 

 these reasons, we have no direct evidence that the ions are simply 

 atoms, or their chemical representatives. One of the greatest 

 desiderata in electrolysis is the determination of the actual ions 

 in any case of electrolysis, but it does not seem practicable to 

 identify them. Hitherto the tendency has been to assume atomic 

 ions if possible, and yet complex ions cannot be excluded. I have 

 taken the following cases of electrolysis from Wiedemann f, who 

 is himself a strong opponent of the idea of complex molecular 

 decomposition in general. 



* See Electricitat, vol. it., pp. 953, 962. 

 t lb., vol. ii. 



