766 REPORT—1885. 
Optical observations, combined with electrical conductivity deter- 
minations, like those of Reinold and Riicker with soap-films, may be ex- 
pected to lead to further knowledge of their thickness, nature, &c. 
A hygrometer founded on the resistance of such films would perhaps 
indicate not what is ordinarily called the ‘ humidity ’ of the air, i.e. its 
nearness to saturation, but its actual vapour pressure. 
Possible methods of answering Question 1. 
T find it difficult to make useful suggestions with regard to an experi- 
mental attack on the question of Ohm’s law in electrolytes. 
It is to be remembered that experiment is useless except on a very 
good and well-considered plan. Something like 1 in 1,000 accuracy must 
‘be obtained, and higher may be aimed at. 
Although the best experiments hitherto are probably those of Kohl- 
yausch and Nippoldt with alternating currents, it does not seem to me 
that such currents, employed with electrodes, are suitable for a really 
accurate determination. The capacity of polarisable electrodes is enor- 
mous, and apparently instantanecus, and the use of alternating currents 
by no means avoids effects due to this polarisation ; although it is possible 
very nearly to eliminate these effects by calculation, as Kohlrausch did. 
Methods of avoiding all polarisation, by working in closed circuits 
wholly electrolytic, are tempting. Such, for instance, as the damping 
effect, when a magnet swings over a liquid, or when a vessel of liquid 
is spun between the poles of a magnet (e.g. Guthrie and Boys).' The 
electrolyte may be made to partake of the motion of the vessel by using 
a jelly. But the calculation of such experiments would (to me, at any 
rate) be difficult, and it is not likely that the experiments themselves 
are susceptible of minute accuracy. 
Probably the best plan of research would be some modification of the 
ingenious method planned by Maxwell for the British Association Com- 
mittee on Ohm’s law in metals. (See Brit. Assoc. Report, Glasgow. ) 
If perfectly similar tapping electrodes could be got, it would be 
sufficient to compare the E.M.F.s at the ends of two troughs or tubes, 
one wide, the other narrow, placed in series, so that the same current 
might flow through both. A null method of comparing E.M.F.s is 
simple enough. For instance, a liquid Wheatstone bridge might be made, 
one branch consisting of the very wide and very narrow tubes, the other 
of a uniform tube or trough of moderate width (see figure). A and B 
are battery electrodes, while c and b, which are carefully chosen, make 
connection with an electrometer; c being movable. A galvanometer seems 
dangerous, as allowing slight currents which might disturb the equality 
of the tapping electrodes. 
The liquid employed might be chosen for convenience, and a natural 
liquid to use would be sulphate of zinc, with amalgamated zinc electrodes. 
It is not to be assumed that the polarisation of these is really nil, 
but it is certainly feeble, and will thus diminish the uncertainty and 
risk of error attending the experiment. 
To avoid difficulties due to the inevitably unequal rise of temperature, 
an intermittently strong and weak current can be employed as in the 
Maxwell-Chrystal method? (cf. Schuster, Brit. Assoc. Report, Belfast). 
1 Proc. Phys. Soc. or Phil. Mag., 1879 and 1880. 
2 Brit. Assoc. Report, 1876, p. 36, Glasgow. 
