180 Proceedings of the Royal Society 
and a standard coil of 10 ohms. A rocking commutator permitted 
the electrodes of the Thomson quadrant electrometer to he connected 
either to the ends of the standard coil, or to the platinum wires that 
came from the electrolyte. In this way, a simple comparison of 
deflections while a steady current was flowing gave the ratio of the 
resistances of the standard coil and the liquid column between the 
platinum points. 
IVhen the current was kept steadily in one direction, the platinum 
points became gradually polarised to different potentials ; hut yet 
the difference between the reading when the current was flowing, 
and the reading when the current was stopped, was very constant, 
whatever this latter reading (the approximate zero) might be. By 
reversing the current from time to time, this polarisation at the 
points could be kept down ; and in no case was the approximate 
zero ever so great as to show any tendency to fall off appreciably 
when left for several minutes to itself. 
A few preliminary experiments convinced me of the possibility of 
obtaining results by this method; but it is only within the last 
few months that a complete series of experiments have been carried 
out, for which I have to thank Messrs J. W. Macdonald and G. 
Gregory Smith, students in the Physics Library of the University. 
These gentlemen investigated very carefully the variation of resist- 
ance with density of solutions of sulphuric acid and water at the 
ordinary temperature (about 10° C.). The following table gives the 
final reductions : — 
Kesi stance. 
Density. 
57*87 ohms. 
1*647 
35 
1*540 
26*74 
1*470 
20*57 
1*410 
15*55 
1*340 
14*18 
1*282 
13*9 
1*204 
14*44 
1*164 
16*47 
1*130 
19*36 
1*098 
25*93 
, 1*069 
52*19 
1*033 
