500 JOHN JOHNSTON AND PAUL NIGGLI 
those systems in which solid solutions can form, as indeed one 
might expect a priori. The diffusion is enormously accelerated by 
rise of temperature, and may be influenced by pressure, though it 
would be premature to regard this as established yet, far less to 
generalize from it. In any case the rate of diffusion under geologic 
conditions, wherever diffusion is possible at all, is probably sufficient 
to reduce in the course of geologic time the degree of heterogeneity 
of systems which originally were not too coarse grained. Never- 
theless old rocks frequently show well-defined layers of isomorphous 
crystals which, so far as we know, can form homogeneous mix- 
crystals; this is true especially for plagioclases, hornblendes, 
augites, and garnets, and indicates that in these cases the rate of 
diffusion is vanishingly small. 
EFFECTS OF TEMPERATURE AND PRESSURE ON SYSTEMS SOLID-FLUID 
General considerations.—In discussions of the change of volume 
(to take a specific instance) which accompanies the process of melt- 
ing, one often encounters the argument that this change is given by 
the difference between the densities of the crystals and of the glass, 
as measured at ordinary temperature. This argument may be alto- 
gether misleading, because it involves the tacit assumption that 
the expansion coefficients of both crystals and glass are identical— 
an assumption which certainly can fit the facts only in a few excep- 
tional cases.‘ Of course until the necessary quantities have been 
actually determined, one can only use the densities observed under 
ordinary conditions; but if one does so, the limitations of any 
results calculated in this. way must be recognized, and the conclu- 
™ The truth of this statement is perhaps more obvious from the following mathe- 
matical formulation. If V; and V, are the specific volumes of glass and crystals, 
respectively, at ordinary temperature, the change of volume 
AV=V,—-V,. 
The change of volume at the temperature ¢ is 
AV,=(V,+AV,)-(V,+4V,) 
=AV+(AV,—AYV,) 
where AV, and AV, are the expansion produced by the increased temperature in glass 
and crystals, respectively. The term (AV,—AV,) may be positive or negative, 
and might even be of such magnitude as to cause the sign of AV, to be opposite from 
that of AV, if the latter were itself small. 
