ON THE THEORY OF SOLUTION. on7 
our renowned visitors and their collaborateurs, and the discovery is one 
of the most valuable contributions to chemical physics that has been 
made of recent years. The visitors call this quality the ‘ratio of disso- 
ciation.’ Professor Armstrong would rather call it ‘ measure of affinity.’ 
I would be inclined to point out that the term ‘dissociation’ is not 
happily chosen, and that ‘affinity’ really explains very little, and that it 
would be better to call it by a new name whose full meaning will require 
further investigation, and would call it ‘ measure of ionisation.’ 
In the first place as to the term ‘ dissociation.’ In all other cases of 
dissociation, e.g. in an electric arc, the elements are so far free from one 
another that they diffuse independently of one another. The term ‘ dis- 
sociation’ is no doubt vague, but it is time we had a more definite notion 
of it. I would certainly confine the use of the term to such cases that 
there was no link connecting the elements that would prevent their 
diffusing independently of one another. As long as there is any link 
connecting the elements of molecules together which essentially prevented 
one of them getting away without the others following, I would not agree 
to say that the elements were dissociated. Hence I object to the term 
dissociation as applied to the ions in an electrolyte. All agree that one 
cannot escape or diffuse without the other following; it may be due to 
electrical forces between them, it may be for other causes; but in either 
case I would refuse to call them dissociated. The possibility of indepen- 
dent diffusion I look upon as a test of dissociation. I would therefore 
appeal to both sides to adopt some neutral term such as ‘ionisation’ to 
express the state of ions in electrolytes. Now as to the proofs that the 
ions are absolutely independently mobile in the liquid, and the assumption 
from this that they are free like the molecules of a gas, being kept apart 
by the molecules of the solvent. This seems a very misleading way to 
speak of the condition. In the first place itis acknowledged that different 
solvents have different powers of ionising a given substance, thereby 
conclusively proving that the function of the solvent cannot be properly 
described as merely giving the ions space to resolve themselves. And 
those who speak so acknowledge that it is only an analogy, or a facon de 
parler. But it seems a very misleading analogy, which leaves out the 
really active part that the solvent plays, and attributes to it a purely 
passive part. The argument of van ’t Hoff that the osmotic pressure in 
very dilute solutions depends only on the kinetic pressure, and not on the 
forces between the molecules, seems to cut against the conclusion that 
these forces must necessarily be small; it seems to show that, whatever 
forces there are between the ions, they will produce the right amount of 
osmotic pressure if only they are so far independent that each ion can 
carry cr its bombardment independently of the other. As this only 
requires the space within which they are bombarding about to be small 
compared with the space rate of variation of the force between the ele- 
ments, and as this is quite consistent with there being plenty of connec- 
tion between the elements, it follows that the laws of osmotic pressure so 
explained do not in the least militate against there being bonds between 
the elements. The whole argument is, however, I think, fallacious, in 
that it assumes a particular theory as to the action between the semi- 
permeable membrane and the liquid. It would follow from this theory 
that one molecule of a salt could never produce osmotic pressure in its 
own neighbourhood by any forces of attraction between it and the sol- 
vent. Now if we apply this on a large scale to the case of an ocean 
