CLASSIFICATION OF IGNFOUS ROCKS 560, 



monly asserted as the purpose of existing mineralogical systems 

 of rock classification, it requires but casual consideration to see 

 that a qualitative mineralogical system cannot express chemical 

 composition, and a thoroughly quantitative scheme has never 

 been formulated. 



Before stating the method of classification to be proposed, it 

 is important to point out some of the chemical and mineralogical 

 relationships obtaining in igneous rocks. And first it is to be 

 noted that while the chemical composition of a magma controls 

 in general the kinds of minerals that may crystallize from it, so 

 that quartz forms in the more siliceous rocks, and olivine in 

 those rich in magnesium and iron, still it does not fix absolutely 

 the kinds or the proportions of all of the rock-making minerals. 

 This is due to the fact that a number of these minerals consist 

 of similar elements in diverse proportions, so that two or more 

 different combinations of elements may be developed in chem- 

 ically similar magmas. Or, as is well known, some of the 

 minerals having a complex composition may be dissociated into 

 less complex ones. A familiar example is the experimental 

 melting of hornblende and the obtaining in its stead pyroxene 

 and magnetite. Another illustration of the same kind of rela- 

 tionship is the chemical identity of some hornblende-andesites 

 and some pyroxene-andesites. 



A striking illustration is furnished by the hornblendite and 

 camptonite of Gran, Norway, described by Brogger. 1 The two 

 rocks having almost identical chemical compositions are com- 

 posed in the first case of somewhat alkalic, aluminous hornblende, 

 and in the second of less aluminous hornblende and feldspar. 



The development of biotite in some rocks and its absence 

 from others having like chemical composition is well known; as 

 shown by its presence in some gabbros and its absence from 

 some chemically equivalent basalts ; its presence in certain 

 diorites and its absence from equivalent andesites; 2 its develop- 



1 Op. cit., pp. 60, 93. 



2 Iddings, J. P., "The Eruptive Rocks of Electric Peak and Sepulchre Moun- 

 tain, Yellowstone National Park,"' Twelfth Ann. Rept. U. S. Geol. Surv. (Washington, 

 1892), p. 653. 



