THE CLASSIFICATION OF THE IGNEOUS ROCKS 7 
the same significance. . The terms, rhyolite-porphyry, trachyte- 
porphyry, andesite-porphyry, etc., by their substitution for the 
objectionable names, quartz-porphyry, quartzless porphyry, etc., 
would carry with them the idea of varietal rather than specific 
variation from the family type, and would, moreover, obviate 
the danger of their being interpreted in terms of the age classi- 
fication. 
Combination of chemical and mineralogical composition as a basis 
for rock classification Probably the majority of those petrolo- 
gists who define rocks as objects would agree that chemical and 
mineralogical composition with texture should occupy the fore- 
most places in rock classification.’ It would probably be more 
satisfactory, were it practicable, to adopt chemical composition 
divorced from mineral composition as the primary basis in 
classification, but we are, per force, compelled to look first to 
the mineral composition, and work backward from this to the 
chemical composition —the chief factor in determining mineral 
composition. In the past the mineralogical examination of 
rocks has been largely qualitative, resulting, in some cases, in 
the classing together of rocks strikingly different as regards 
their ultimate chemical composition, but a stage has now been 
reached where such a method is no longer adequate. Pirsson 
has called attention to the necessity of paying greater regard to 
the relative quantities of the several essential constituents of a 
rock, thus making a rough estimation of its ultimate chemical 
composition.” 
Specific, generic, and family rock names are applied to arbitrary rock 
types separated from one another by no sharp lines.— It follows, from 
the gradations generally observed to connect the families of the 
igneous rocks, that the names which we adopt to designate any 
individual rock, or class of rocks, is applied as aZype name in the 
sense that it applies to a particular rock or collection of related 
1Cf. TEALL: British Petrography, p. 69; WHITMAN Cross: loc. Gilig, DotKOS Io 1ee 
Ipprincs: On Rock Classification, JouR. GEOL., 1898, VI, p.93; F. ZIRKEL: Lehrbuch 
der Petrographie, I, p. 829, 1893; W. C. BrOGGER: Die Gesteine der Grorudit- 
Tinguait Serie, Christiania, 1894, p. 92. 
2Tgneous Rocks of Yogo Peak, Montana, Am. Jour Sci. (3) L, p. 478. 
