570 JAMES ASTON 
of physical chemistry, more especially the theory of solutions and 
the phase rule of Gibbs. The application of these laws is of very 
recent date. In fact, it is hardly necessary to go back more than ten 
years to cover the period of their development. Because of this fact 
the drawing of well-worked-out analogies is very difficult; a consid- 
erable amount of investigation has been done with the simpler alloys, 
but in geology the field is practically barren. It will be necessary, 
therefore, to confine the discussion to the fundamental types of solidi- 
fication from the viewpoint of their solubility relations, and to point 
out the application of these laws to certain alloys, and analogous 
minerals and rocks, wherever possible; leaving to the judgment and 
imagination of the reader the possibility of their further application 
to geological problems. 
Proceeding from this viewpoint, it is at first essential to grasp a 
few fundamental facts. Alloys are solutions, and obey the same 
laws in freezing as aqueous solutions of salts do in crystallizing. It 
is merely a question of fluidity at a different range of temperature. 
Not only is this true, but it is likewise true that solubility is not limited 
to any particular state of the interacting substances. Thus we may 
have solubility in the solid state, or solid solutions. ‘This is shown 
in the tendency for mutual diffusion of solid gold and solid platinum; 
or more well known, perhaps, in the harveyizing of armor plate, 
where carbon is absorbed by iron at temperatures well below the 
point of fusion. These solid solutions you are more familiar with 
as isomorphous mixtures or mixed crystals. 
COOLING-CURVES 
In obtaining the solubility curves for salts in solution, the usual 
procedure is to make analyses of the saturated solution at various 
temperatures, and in this way to obtain a series of points which may 
be plotted into a temperature-composition curve. In the case of 
alloys, this method is not so well suited, since we are working at 
higher temperatures; also we often have to deal with changes in the 
solid, as well as in the liquid state. For alloys we obtain the necessary 
data for individual mixtures by means of the cooling-curve, a curve 
which is a function of time and temperature, and shows the change 
in the rate of cooling as the temperature is lowered. Breaks in the 
