VAPOUR PRESSURES AND REFRACTIVE INDICES OF SALT SOLUTIONS. 285 
and citric acids. The previous results were completely confirmed. 
The law is as follows :— 
The molecular volume of a salt in dilute solution is a quantity com- 
posed of two constants, one for the metal and another for the salt radical. 
It follows that the replacement of one metal, or salt radical, by another 
metal, or salt radical, is always attended by the same volume charge, no 
matter how they may be combined together. 
The presence or absence of water of crystallization in one or both of 
the salts has no effect on the above law; it therefore follows that it has 
the same volume as the solvent water. Water of constitution, however, 
shows itself in solution by possessing a volume markedly different from 
that of the rest of the water. 
These results point to the presence in solution of what may be 
termed the anhydrous salt, in contradistinction to the view that a 
hydrate, definite or indefinite, results from solution; or, in other words, 
no part of the water in solution is in a position, relative to the salt, 
different from the remainder. 
II, Saturation of Salt Solutions. Part II. 
It is found that the molecular volumes of a series of solutions of 
different strengths of the same salt may be satisfactorily expressed by 
the formula :— 
M, V. = 1800 + na + 026 — ny, 
Where x = number of molecules of salt per 100 H,0, and a, 8, and y 
constants depending on the salt, 
r=na+n7B — ny; 
and 
‘= — n? 
ae + nB — ny. 
This last is the mean molecular volume of the salt in solution. The- 
curve is a parabola, and is such that ee twice the solubility of the salt 
Y 
in question .*, £ = solubility ; but this is also the apex of the parabola ;. 
saturation is therefore reached when the further addition of salt would 
produce diminution of the mean molecular volume of the molecules. 
already present. The last molecule before saturation, enters into solution 
with a volume sensibly equal to the mean, as is shown thus :— 
(na + 076 — n¥y)—((n—1)a + (n—1)28 =@ —1)*y) =a 4+ 7p — ny, 
When n= Bay 
2y 
III, Supersaturation of Salt Solutions.’ 
In these papers experiments are described which lead to the con- 
clusion that the only truly supersaturated solutions are those which 
result from the fact that, when a hot solution is cooled, a finite time 
ig required for the excess of salt to crystallize out—what is usually 
* Published (1) Phil. Mag., June, 1885 ; (2) Phil. Mag., September, 1885, 
