MAGMATIC DIFFERENTIATION OF IGNEOUS ROCKS 3109 
IL 
REVIEW OF THE PHYSICAL CHEMISTRY OF THE CRYSTAL- 
LIZATION OF IGNEOUS MAGMAS 
In the examination of these laws two different methods can be 
used: (a) the synthetic, in which there is an opportunity for 
precision-determinations, especially of temperature. Previous 
investigations on the crystallization of silicates from melts have 
been, nearly without exception, conducted at the pressure of one 
atmosphere; (0) the analytic, mainly based on the study of the 
structure of the rocks. In this manner we may examine the 
sequence of crystallization, and so also the “‘individualization- 
fields” of the minerals, further the mix-crystal systems, the chemical 
composition of eutectic intergrowths, etc.—all under the physical 
conditions, especially with regard to pressure and time, present 
during the solidification of the different igneous rocks. 
The synthetic method forms the important base. The analytic 
method gives us, in particular, information as to the extent to 
which the results of investigations conducted chiefly at atmospheric 
_ pressure and during short periods of time, can be transferred to 
apply to the physical conditions under which the crystallization 
of magmas took place. 
The two methods, therefore, go hand in hand and complete 
each other. 
REMARKS ON THE STRUCTURAL CRITERIA FOR THE SEQUENCE OF 
CRYSTALLIZATION 
The sequence of crystallization in igneous rocks may usually 
be determined by the complete, partial, or wanting idiomorphism 
of the minerals, by the inclusions, by deposition on a solid body 
(Fixkorper-Absaiz), by ‘together-swimming”’ structure (synneusis- 
strukiur, see below), by law-governed intergrowths, etc. 
The complete zdiomorphism of a primary mineral A against all 
the other minerals shows that its crystallization was finished before 
the commencement of the solidification of the others. From the 
partial idiomorphism of the primary mineral A against the primary 
mineral B whose idiomorphism is wanting, we can infer that the 
crystallization of A had commenced at an earlier stage than the 
crystallization of B. But we must not draw the more extensive 
