224 The American Geologist. April, 1900 
lection and relative order of governing' principles, therefore, 
the classification is like those of Teall and Kemp, above given. 
In the selection of mineralogy as the prime factor in classifi- 
cation the writer follows, besides those above mentioned, 
Brogger, Zirkel, and practically Michel-Levy (according to 
his writings, although in his table he bases his chief division 
upon structure). 
For precedent in dropping out mode of occurrence, as an 
important factor in rock classification, one may quote Fouque 
and Michel-Levy, who substitute structure for mode of oc- 
currence in their grand divisions of granitoid, porphyric, and 
trachytoid. Zirkel and Teall have also omitted the mode of 
occurrence as a factor of the first degree of importance. 
As concerns the omission of age as an important principle 
in rock classification Teall* observes that "geologic age 
should not receive direct expression in petrographic nomen- 
clature. The petrographer should define his rock types with- 
out direct reference to age and then describe the distribution 
of his types in time. He should follow in short the method of 
the paleontologist. A similar course should be adopted so far 
as matters relating to origin and geographic distribution are 
concerned. The classification of igneous rocks should there- 
fore be based primarily on facts of composition and texture." 
Also Broggerf remarks that "to encumber entirely similar 
rocks with two different names only because they are of dif- 
ferent age, — for example, one Devonian, another Tertiary or 
recent, — seems now to be looked upon more and more as un- 
justified." 
Concerning the application of mineralogy as a prime factor 
in rock classification Michel-LevyJ observes : 
The study of structures, in fine, only serves to constitute some large 
groups and to subdivide rocks of the same mineralogic and chemical 
composition — such, for example, as the acid rocks. In the last analy- 
sis it is always indispensable to have recourse to the mineralogic com- 
position to create families which have come from an analogous original 
magma. But it is necessary to take exact account of the principal com- 
ponent elements and of their relative importance from a quantitative 
*0p. cit. p. 69. 
tDie Gesteine der Grorudit-Tinguait Serie ; Kristiania, 1894 p. 87. 
JPrincipes a suivre pour une Classification universelle des Roches, 
Compte Rendu. Lausanne, p. 155. 
