140 
MESSES. W. R. BOUSFIELD AND C. ELSPETH BOUSFIELD 
16. Volume Changes in the Solvent. —When a solute is dissolved to form a 
concentrated solution, if the solute is a solid it assumes the liquid form with an 
accompanying change of volume owing to its liquefaction. We shall find good 
reason to suppose that dilution of the concentrated solution involves no further 
volume changes of the liquefied solute, and that the contraction which takes place 
on dilution is attributable to volume changes in the solvent. It is true that, owing 
to the compressibility of the atoms and molecules, the volume of the solute may be 
slightly diminished by dilution, owing to the molecular compression produced by 
further combination with the solvent. But this possible progressive volume change 
in the solute is too small to be traceable in the experimental results, and it will be 
seen later that both the specific heat and the specific volume of the solute, when 
once it has been reduced by solution to the liquid state, can be treated as constant 
for all isothermal dilutions. Thus for isothermal dilutions the contraction which 
takes place on dilution may be ascribed wholly to changes in the water, within the 
limits of experimental accuracy. These changes are due to 
(1) The disappearance of some of the bulkier molecules in the free water; 
(2) The increased density of the combined water. 
Assuming that water is a mixture chiefiy of trihydrol or ice molecules (H 2 0) 3 , 
dihydrol (H 2 0) 2 , and hydrol or steam molecules (H 2 0), it is probable that the 
addition of the solute reduces the proportion both of ice molecules and of steam 
molecules, both of which are comparatively bulky molecules (see Bousfield and 
Lowry, ‘Phil. Trans.,’ A, vol. 204, 283, 1905). At the same time both the freezing- 
point and the vapour pressure of the solution are depressed. These depressions give 
us data from which, in conjunction with data as to conductivities and viscosities, the 
number of combined water molecules can be ascertained with some approach to 
accuracy. In the case of the most concentrated NaCl solutions all the necessary 
data, at temperatures of from 18° C. to 20° C., exist, and in a former paper 
(Bousfield, ‘Trans. Chem. Soc.,’ vol. 106, 1821, 1914) the water combination for 
three concentrated solutions of NaCl, which are nearly the same as solutions I., II., 
and III., was evaluated. By the help of these data we shall be able to arrive at the 
specific heat and the specific volume of the liquefied solute and thence at the volume 
changes in the solvent itself. 
17. The Specific Contraction of the Water of the Solution. —It has been shown 
in a former paper (Bousfield, ‘Trans. Chem. Soc.,’ vol. 107, 1409, 1915) in the case 
of a liquid solute that the contraction which takes place on the solution of a gramme 
molecular weight E of the liquid solute in H grammes of water is ‘ 
AXu = X-Xo, 
where 
Xo = E(w-Uo) 
