ON ELECTROLYSIS. 725 
Tests for electrolytic conduction as distinct from metallic. 
So far as experiment has yet gone, electrolytic conduction is found to 
obey Ohm’s law—a remarkable and important fact, if fact it accurately 
be—and we will return to it later. 
The obedience to Ohm’s law shown by electrolytes prevents our 
drawing any easy and sharp line of demarcation between the two classes 
of bodies. To distinguish between them we have to study what happens 
at a boundary or junction of the two classes of conductors. 
At such a place Ohm’s law utterly breaks down; a finite E.M.F. is 
needed to drive any permanent current, however small. And the reason 
is that the atoms which have conveyed the electricity through the electro- 
lyte can accompany it no longer, and have either to give it up and let it 
go on without them, or cling to it and stop the current. In either case we 
have what is called polarisation. If the clinging power of the atoms is 
greater than the applied E.M.F., the current may wholly cease, and as 
soon as the E.M.F. is removed, will spring back again, the opposition to 
it being no longer like friction, but like a strained spring. 
If the applied E.M.F. overcome the atomic force, the current flows 
on, leaving the discarded atoms to do what they like. They may com- 
bine with each other and separate from the liquid, in which case we have 
visible chemical decomposition; or they may combine with each other 
and dissolve in the liquid; or they may combine with the liquid, forming 
secondary products ; or they may combine with the electrodes. 
They usually cling tightly to the electrode, even if unable actually to 
combine with it, and by thus altering its surface they may give rise to a 
permanent opposite current. The four possibilities are— 
Combination with, or solution in, the liquid. 
Combination with, or deposition upon, electrodes. 
Combination with something already dissolved in liquid. 
Combination with each other and freedom. 
Wherever visible decomposition occurs, there is no doubt of elec- 
trolysis, so that is the most simple and obvious test of electrolytic con- 
duction. But the conductivity may be so bad that no visible product 
forms inacentury. In that case polarisation, if present, isa test. As 
Helmholtz says: ‘ Galvanometers are sensitive enough to shew a current 
which could only decompose a milligramme of water in a century.’ ! 
But this test also may fail by reason of secondary actions. It is the 
merest commonplace that an ordinary voltameter behaves as a very 
leaky condenser. A continual drain of electricity goes through it, 
however small the E.M.F. applied; and when charged, and left, the charge 
is found very rapidly to diminish. It is facts like these of course which 
have so often suggested slight metallic conductivity, and which at one 
time tempted Faraday to postulate this. , 
With certain fused salts the decay of polarisation is so rapid that 
polarisation hardly appears to exist; and quite a strong current can be 
sent through them without any visible decomposition. Clark ? has shown 
that in these cases it is the solution of the liberated ions and their rapid 
diffusion which causes this apparent metallic conduction. There is pro- 
? One may notice that even such a current as this is still 10 electrostatic units per 
second, 
2 J. W. Clark, Phil. Mag., October 1885. 
