ON ELECTROLYSIS. 771 
It may also be possible to examine the speed with which an ion, start- 
ing from a vessel at one end of along tube, makes its appearance in 
another vessel at the other end of it, there being in this second vessel 
some sensitive chemical test for faint traces of it; such, for instance, as 
iron in sulpho- or ferro-cyanide, silver in chloride, copper in ferrocyanide. 
Or again, by putting a detecting substance in the tube, the journey 
of an ion may be continuously watched. 
Other things more or less vague might be tried; such as seeing whether 
there is any difference between applying a high H.M F. to given electrodes 
in a long narrow vessel, and a small E.M.F. to similar electrodes in a 
short wide vessel, so that the same current goes through both. One must 
not expect any variation in the primary products of electrolysis probably, 
but there may be some difference in the secondary effects. 
8. How much of the Current is conveyed by the water and how much 
by the dissolved salt in any given case ? 
To answer this question it seems to me sufficient to determine the 
amount of free acid generated at either pole, and to compare it with the 
amount of metal deposited in the same time by the same current. 
For instance, with copper sulphate solution and platinum electrodes. 
If the salt conducted the whole current, one equivalent of free acid would 
be produced at the anode; if the water conducted the whole, one equiva- 
lent of free acid would appear at the cathode; it is easy to see this by 
drawing the section of a cell and considering its action; and go, by de- 
termining the proportion of free acid which actually does appear at either 
pole for given strength of solution and intensity of current, it is easy to 
reckon how much of the current is by each substance conducted. Tt is 
obviously necessary to diminish diffasion between the cathode and anode 
vessels, and to avoid. electric endosmose and the use of porous dia- 
phragms. It is also probably advisable to stir or scrub the surface of 
both electrodes, so as to avoid a layer of some other liquid forming there, 
and so confusing the whole reaction. 
Migration data, if explained in the way I suggest, will also give a 
determination of the proportion of current conveyed by salt and b 
solvent ; and it will be interesting to compare the two methods of deter- 
mination, and see if they give the same result. 
If there is a discrepancy, as is very probable, it will be necessary to 
examine its cause, and to see whether it wholly upsets the suggested 
migration hypothesis, or only entails some slight modification. 
9. Is any quasi-electrolysis possible across an air-space ? 
If there are in acid-water any free dissociated atoms electrically 
charged, it is just possible, though very unlikely, that they might be 
pulled out of the liquid by electrostatic attraction. 
Thus, if hot acid liquid be one plate of a condenser, the other plate 
being a hot metal slab half a millimetre above it, and kept at a very 
different potential, it is just conceivable that the evaporating steam might 
contain a trace of free hydrogen or free oxygen, according to the sign of 
the potential of the plate. 
3D2 
