THE ORIGIN OP IGNEOUS ROCKS. 
185 
From the wide distribution of the rocks of the alkali groups 
it would appear that they are of more frequent occurrence 
than those of sub-alkali groups; but this arises from the fact 
that it is easy to recognize the former by the presence of 
certain characteristic minerals, whereas the latter are not 
so easily characterized, and it requires more complete infor¬ 
mation concerning the chemical composition of the whole 
group or a knowledge of the mineralogical character of the 
various rocks of the whole series to establish its identity. 
Upon investigating the data already accumulated, which 
indicate the character and distribution of the various igneous 
rocks over the earth, it is evident how essential it is that 
ample, and if possible complete, collections should be made 
of all regions where such rocks occur, and that, besides a 
mineralogical examination, there should be a complete and 
accurate chemical analysis of the material. 
In the absence of such complete information it is not pos¬ 
sible to establish definite laws regarding the geographical 
distribution of chemically different magmas, but sufficient 
is known of them to warrant the conclusion that the igneous 
rocks in various parts of the globe form natural groups of 
varieties in each region, and that they have resulted from 
the chemical differentiation of different kinds of general 
magmas—that is, they do not originate immediately from a 
universal magma of homogeneous composition, or from a 
homogeneous liquid couche beneath the crust of the earth. 
However, the recurrence in different parts of the earth’s 
surface of the same varieties of general magma leads to the 
conclusion that they may be the results of a differentiation 
of a still more uniform magma, which may have been homo¬ 
geneous throughout its whole extent and may represent an 
original liquid couche. There is at present nothing to indi¬ 
cate at what period in the earth’s history this supposed 
primitive differentiation took place. Its extreme remote¬ 
ness would be proven if it could be shown that the chemical 
character of general magmas which furnished groups of 
igneous rocks in various periods from pre-Cambrian to the 
27 —Bull. Phil. Soc., Wash., Vol. 12. 
