ORDER OF CRYSTALLIZATION IN IGNEOUS ROCKS 401 
formation of granite and of rhyolite. The rocks formed under 
these intermediate conditions confirm the evidence of the rhyolites 
as to the order of beginning of crystallization as Bowen has pointed 
out. There is not only no evidence that the difference of conditions 
brings about a reversal in order of crystallization but there is 
definite positive evidence that no such drastic change occurs. 
In view of the fact that the objections of Ziegler are almost 
entirely the outcome of theoretical considerations which it is 
dangerous to extend to such complicated systems, and that the 
evidence of the rocks themselves, on which Bowen based his con- 
clusions, is not questioned, it does not appear that the objections 
offered seriously impair the conclusions reached by Bowen. 
A knowledge of the order of beginning of crystallization is 
important for the elucidation of the differentiation of rock magmas. 
Differentiation is not uncommonly discussed in terms of the assump- 
tion that ferromagnesian minerals are minerals of early crystalliza- 
tion whether they are present to the extent of 70 per cent or of only 
5 per cent. This assumption is often quite obviously at variance 
’ with the evidence of the porphyritic facies of the rock types 
described, and in such cases leads, of course, to erroneous con- 
clusions. 
