128 Mr. W. W. J. Nicol on the Molecular 
KC1 8-4, NaCl 10-99, 
KN0 3 5-6, K 2 S0 4 1-0 + , 
Na 2 S0 4 ... 2-0 + . 
All at 20° C. 
We have therefore arrived at the following : — 
In the case of salts when compared in pairs, the increase 
of molecular volume by concentration is greatest with the less- 
soluble salt. Why is this ? When viewed in the light of the 
theory of solution which I recently stated in a paper* read 
before the Royal Society of Edinburgh, the reason is evident. 
In that paper I suggested that the solubility of a salt 
in water was due to " the attraction of the molecules of water 
for a molecule of salt exceeding the attraction of the mole- 
cules of salt for one another ;" and that " as the number of 
dissolved salt-molecules increases, the attraction of the dis- 
similar molecules is more and more balanced by the attraction 
of the similar molecules ;" and that this last increases until 
the two forces balance, when saturation takes place. I also 
showed that the rate of increase in the density of a salt-solution 
is less than the rate at which it becomes more concentrated, 
but that when a solution is sufficiently dilute no further 
dilution affects its specific gravity ; as a consequence, the 
molecular volume of the salt dissolved is constant. It is 
evident from the above that the effect of concentration in 
increasing the molecular volume of a solution is due to the 
molecules of salt coming more and more within the sphere of 
one another's attraction, and being thus brought nearer and 
nearer to the point at which crystallization takes place. Now 
what is true of the same salt in solutions of different strengths 
is equally true of solutions of the same strength of salts whose 
solubilities are different ; and it follows that the less soluble 
salt will have its molecular volume increased by concentration 
faster than the more soluble salt. Such, I believe, is the 
explanation of the variations in the differences above observed; 
and, as a consequence, if the salt-solution be sufficiently dilute, 
the differences between their molecular volumes are constant. 
The next point to be considered is the influence of rise 
of temperature on the molecular volumes of salt-solutions. 
Table VIII. contains the requisite data. These are the results 
of single experiments, and are therefore not of equal accuracy 
with the data in Table I., nor, in consequence, do the values 
at 20° correspond exactly with those given in that table ; but 
they are accurate in general to + '00005 in the specific gravity. 
The column headed Difference contains the increase of 
molecular volume due to heating from 20° to 40°. This refers 
to the salt alone ; for water = 1 , both at 20° and 40°. 
* Phil. Mag. 1883, xv. p. 91. 
