376 The American Geologist. June, L896 
Iddings* states practically the same thing in his monograph 
on the rocks of Electric peak and Sepulchre mountain, and 
other prominent authors may be quoted to the same end. It 
seems to the writer that nothing is more needed to advance 
petrographic science than the definite use of rock terms, and, 
if it is necessary to know the conditions under which a rock 
consolidated and its geological age, it is evident that in many 
cases we cannot name a rock at all. No mineralogist would 
consent for a moment to base a classification of minerals on 
other than contingent facts, and although, since rocks are in 
most cases mixtures of several minerals, a rock classification 
must be a classification of mixtures, which vary in each 
magma from point to point, it is nevertheless true that certain 
mixtures, of very nearly the same composition and structure, 
are found in many parts of the world, and admit of being 
characterized in such a w T ay, that the term applied to them 
may be made to have a definite meaning. On the other hand, 
if the fact that rocks are mixtures be not borne in mind the 
tendency to make rock species may be carried to excess. If 
the minerals that make up the larger part of most igneous 
rocks were great in number, a rock classification would require 
so man} r terms as to have little practical value, but happily, 
as all geologists know, this is not the case, and with the 
enormous advance in late years of our knowledge of the com- 
position of rocks it is perhaps not visionary to hope that a 
set of terms, which will endure, will soon be agreed upon for 
the most universal mineral mixtures. 
With many of the coarse granular rocks, in which the 
mineral components are easily recognized, there is already a 
set of terms in use nearly the world over, but with many rocks 
occurring as dikes, and with the micro- and cryptocrystallme 
rocks, there is still much confusion. What appears to be a 
step in the right direction is the division by Fouque and Levy 
of the plagioclase rocks into an oligoclase-andesine series and 
a labradorite-anorthite series. The former will contain most 
andesites and diorites and may be called the diorite family ; 
the latter will contain most basalts, diabases, and gabbros and 
may be called the gabbro family. 
*13th Annual Report of the U. S. Geological Survey, p. G63. 
