224 Myr Chree, On the effect of an electric [May 21, 
in an originally saturated solution. Further the heating produced 
by an electric current is naturally more or less unequally distributed, 
and so must give rise to convection currents whose mechanical 
effect would naturally increase the rate of solution in an un- 
saturated solution. 
In a saturated solution the indirect influences of the current 
seem likely to be a minimum, and accordingly in the following 
experiments none but saturated—or very nearly saturated—solu- 
tions were employed. 
The salts examined were three in number, all chlorides, viz. of 
Sodium, Potassium and Calcium. The data as to their degrees of 
solubility and saturation are derived from Storer’s Dictionary of 
Solubilities, and the data as to their electrical conductivities 
from Wiedemann’s Llektricitdét, Erster Band. They were selected 
as representatives of various classes of salts. 
In sodium chloride the quantity of salt required for saturation 
increases with the temperature but at a very slow rate. According 
to Gay Lussac the percentage by weight of salt in the saturated 
solution at 0° is 35°52 and at 50° is only 36°98. In potassium 
chloride according to the same observer the percentage of salt in 
a saturated solution rises from 29°21 at 0° to 43°59 at 52°39. As 
to the quantity of salt in a saturated solution of calcium chloride 
different observers seem to vary considerably, but they all agree 
that it increases rapidly with the temperature. 
In Wiedemann’s Llektricitdét are curves whose abscissae give 
the amounts of salt in the solution and whose ordinates give 
the corresponding conductivities. Most, if not all, seem to agree 
in representing the conductivity as at first increasing with the 
amounts of salt but at a gradually diminishing rate. In some, e.g. 
calcium chloride, the conductivity attains a maximum and then 
diminishes before the point of saturation is reached. In others, 
e.g. potassium chloride, the form of the curve suggests that 
a maximum of conductivity would supervene but for its ap- 
pearance being forestalled by the point of saturation being 
reached. In sodium chloride the conductivity is still imereasing, 
but extremely slowly, as the point of saturation is approached. In 
calcium and in potassium chloride, especially the latter, the curve 
of conductivity is fairly steep near the point of saturation. 
The object of the experiments being to detect what was soon 
seen to be a very small effect, strong currents were used so as to 
get as large an effect as possible. ‘To prevent electrolysis a com- 
mutator giving a rapidly reversed current was employed. With 
comparatively weak currents there was in general no trace of 
electrolysis; but when, as sometimes happened, the current from 
6 or more large storage cells passed between electrodes of several 
square inches in surface, at a distance of two inches apart in the 
