600 WILLIAM LAWTON GOODWIN ON THE 
sulphates of alumina have no definite composition. An essential difference 
between solution and combination seems at first sight to lie in the fact that 
solubility is a function of temperature. Thus, the quantity of a salt which will 
dissolve in water generally increases with the temperature regularly and appar- 
ently without a limit. It is possible that at low temperatures the molecules of 
water are associated together in groups, and that a molecule of salt attaches itself 
to a whole group. When the temperature is raised these groups are broken 
up, and molecules of water are broken off and ready to unite with farther 
molecules of the salt. Or, on the supposition that the molecules of salt occupy 
the interspaces, it is easy to see how arise of temperature would increase the 
solubility by increasing the number of interspaces. This supposition is 
strengthened by the observed decrease of the coefficient of viscosity of water 
with rise of temperature, and the increased coefficient of solutions of salts. 
Gases certainly mix irrespective of anything like chemical attraction, but when 
liquids and solids are considered, a different set of phenomena is observed. 
Some liquids “mix” in all proportions, others only within a limit. If solution 
is a mere interpenetration of the particles of the two bodies, how can it be 
explained that oil does not dissolve in water, but does dissolve in ether? It 
may be that the particles of oil are too large to find their way into the inter- 
stices of water, otherwise the kinetic theory is as favourable to interpenetra- 
tion between oil and water as between alcohol and water. The case of water 
and ether may be cited as an example of a limit to the solution of each liquid 
by the other. The molecules of ether must be either very large or very far 
apart, for with a molecular weight of 74 it has a specific gravity of only ‘723. 
Its great mobility would seem to point to the latter as probable. If this be 
the case the molecules of water should have no difficulty in finding room among 
the molecules of ether. It can be generally stated that bodies which are like 
each other in chemical constitution dissolve in each other, and it seems as if 
something like chemical action between bodies must be assumed to explain 
this fact. There is no definite boundary between solutions and chemical com- 
pounds.* Alloys occupy a sort of middle-ground, but many of these can be 
obtained in definite crystals ; but this is no more a proof of chemical combina- 
tion than the production of definite crystals of isomorphous salts. It is pro- 
bable that there is a gradual progression from simple mixture to the most 
stable chemical combination, and for any particular case by a change of tem- 
perature we may pass from a state of chemical combination to one of simple 
mixture. Substances resembling each other chemically tend to mix merely, 
* Dr Ramsay has shown (Jour. Chem. Soc., Oct. 1877) that hydrates of alumina and ferric oxide 
give off their water at a regular rate depending on the temperature, while salts containing water of 
crystallisation show different rates corresponding to definite hydrates. Something very like this is seen 
in the phenomena of solution. 
