192 REPORT—1890. 
admit that if there are two electrolytic cells in series containing electrolytes 
AB, A’B’, Aand A’ being cations, B and B’ anions, the amounts of 
A and A’ or of Band B’ deposited by the same quantity of electricity 
are chemically equivalent ; but for a given electrolyte, the specification of 
the ions into which it will be decomposed by the current is not always 
known, or even ascertainable; moreover, there are cases in which the 
elements have more than one chemical equivalent, so that it is not prae- 
ticable to state this part of Faraday’s law in more definite terms than 
those given above. The recent work on the subject, will be considered 
in the answer to the question ‘ What are the ions ?’ in Part III. § b. 
But there is no question of doubt when the electrolytes are fused or 
dissolved compounds of monad elements only, and there are many other 
cases of dyad and triad compounds in which the chemical equivalence of 
the ions is well recognised, and in all these cases Faraday’s law in its 
complete form may be applied with confidence; and the final result is that 
with every monad atomic ion there is associated in electrolysis a certain 
definite quantity of electricity, positive or negative’; with every dyad 
atomic ion twice that amount, with every triad three times, and so on. 
And in all true electrolytes, the distribution of electricity is the distribution 
of these ions carrying their specific charges. 
(c.) The conduction of electric currents through electrolytes follows Ohm’s 
law.—It must be remembered that for metallic conductors it has been 
shown by Chrystal and Saunder that if the relation between electromotive 
force e and current i be represented by 
e=ir(1—hi?) 
then h is less than 107!*, showing that for these Ohm’s law is true 
with extreme accuracy. There are certain physical laws which, although 
originally discovered empirically, express as numerical relations necessary 
consequences of the nature of the physical quantities referred to in the 
laws. Thus Snell’s law of refraction (expressing, as is now known, the 
ratio of velocities of transmission in two media) is not a law in which 
one expects further experimental investigation to detect a deviation from — 
accuracy. Faraday’s law is another illustrative example. The inverse 
square laws, which perhaps merely express the property of transmission 
in straight lines, are also laws which seem to be strictly true, and not 
empirical approximations. There isa difference in character between 
these and such as the gaseous laws, in which more refined apparatus and 
methods detect divergences from the apparent simplicity. Now Ohm’s 
law for metals, being the most accurately verified of all laws, would seem to 
belong to the former class, and to be a necessary consequence of the nature 
of conduction itself. J. Hopkinson? suggests that the law asserts the 
principle of the superposition of the effects of electromotive forces in bodies 
in which the conduction is not complicated by residual charge, and it may 
therefore be regarded as a special case of the more general principle of 
superposition.? He divides the continuous effect of electromotive force on 
glass into four successive stages, and thinks that the same might hold if 
we could experiment fast enough for an electrolyte, the principle of super- 
position probably applying to all the continuously connected successive 
events. 
1 For the calculation of the amount of electricity on a monad atom see Lodge, 
B.A. Rep. 1885; Budde, Wied. Ann. 25, p. 562, 1885. 
2 Phil. Trans. 167, 1877, p. 614. 3 B.A. Rep. 1886, p. 309. 
