ON ELECTROLYSIS AND ELECTRO-CHEMISTRY. 187 
Part I. 
General Electrolytic Phenomena. 
Tn order to analyse the actions taking place in electrolysis, we may 
imagine the electrolyte in the cell divided into three portions by two 
parallel partitions of porous non-conducting substance; the two end 
portions, the anode and the cathode vessels, contain the positive and 
negative electrodes respectively ; the middle portion, while it allows the 
transmission of electricity through it, may be imagined protected from 
any change of composition which, in the absence of partitions, might be 
effected by diffusion, or mechanical transfusion, or convection currents of 
liquid. How far such an ideal partition is realisable in practice will 
appear later. The electrodes may be any electrical conductors, solid or 
fluid, alike or different. For a typical specimen we cannot regard an elec- 
trolytic liquid otherwise than as a mixture of solutions of chemical com- 
pounds, though the amount of all but one of the constituents of the 
mixture may be so small as to be regarded merely as impurities, which it 
would not even be possible to detect by ordinary chemical means. The 
remarkable sensitiveness of electrolytic properties to change, in conse- 
quence of the admixture of very minute portions of impurity, renders this 
necessary. 
Thus Von Helmholtz has already said in his Faraday Lecture that he 
has detected the polarisation corresponding to the decomposition of a 
quantity of water of the order 1x10-" gramme. And Gore! has 
shown that the effect of chlorine upon the E.M.F. of a Pt-Mg voltaic 
couple in distilled water is such that the presence of one part of chlorine 
in seventeen thousand million parts of water could be detected thereby. 
The neglect of considerations of this kind finds very remarkable illustra- 
tion in the history of electrolysis. It is now generally known that the 
experiments upon very pure water, especially those of Kohlrausch,? have 
so far changed the views upon the matter that, whereas at one time water 
was regarded as the conducting part of a solution, pure water is now looked 
upon as probably not conducting at all. Kohlrausch obtained water the 
ratio of whose conductivity to that of mercury was 0°71 x10 -" at 215° C., 
and its sensitiveness for small quantities of impurity approximated to that 
_ of the sense of smell, since when exposed in a room containing tobacco- 
smoke its conductivity doubled in three hours. The simplification that 
would be introduced by regarding the typical electrolytic cell as contain- 
ing a perfectly pure chemical compound liquid cannot therefore be 
realised in practice, and any part of a theory which depends for its sup- 
port on such an assumption must, for the present at any rate, be held in 
suspense. 
When an electromotive force is made to act between the electrodes of 
such a cell as that described above, so that a current is shown in a gal. 
vanometer included in the circuit, the following actions take place :— 
(a.) A part of the electrolyte is decomposed, the products of the de- 
composition are deposited at the electrodes, and these either (i.) are visibly 
set free, (ii.) unite with the electrodes, or (ii1.) unite chemically with the 
solution in the anode or cathode vessel as the case may be, and in the 
' Proc. Roy. Soc. Tune 14, 1888, vol. 44, p. 301 
* Pogg. Ann. Ergz. B. 8, 1876, p. 1; Wied. Elec. 
