116 A . Holmes — Classification of Igneous Rocks. 



the conclusions arrived at, and while the writer must be held entirely- 

 responsible for all expression of opinion, he wishes to acknowledge 

 his gratitude and indebtedness to Professor Watts and Dr. Evans, 

 both of whom, in numerous discussions, have freely offered suggestions 

 and criticism which have proved to be of the greatest value. 



In recent years the chief criteria on which systematic classifications 

 of igneous rocks have been based are {a) mineral composition, 

 (b) chemical composition, and (c) texture, or (V) mode of occurrence 

 (Cross, 1910, p. 473). Although the two latter factors have often been 

 considered interdependent, it is now becoming generally recognized 

 that they are by no means wholly so, and modern custom tends more 

 and more to relegate each of them to a subsidiary position in classifica- 

 tion. For the working petrologist, the mineral composition of an 

 igneous rock is, in a great majority of cases, its most important 

 characteristic. Unfortunately, many minerals, such as the pyroxenes 

 and amphiboles, are capable of a wide range in composition. More- 

 over, a small percentage of igneous rocks are incompletely crystallized. 

 It has therefore to be recognized, and accepted as at present an 

 unavoidable limitation, that the whole field of igneous rocks cannot be 

 reliably classified on a mineralogical basis that will also faithfully 

 reflect the chemical composition. On the other hand, a chemical 

 classification, whether founded on normative minerals, or otherwise, 

 demands far more analyses than can in practice be obtained. In 

 addition to the requirements of field-work, it is therefore necessary 

 to have at least two systems of classification, one mineral, the other 

 chemical. Clearly, for purposes of comparison, the two systems 

 should be arranged as closely as possible along parallel lines. 



In the Quantitative Classification of Cross, Iddings, Pirsson, and 

 Washington, the chemical composition is expressed, not in oxides, 

 but by a series of standard minerals known as the norm, as opposed to 

 the mode, which is the actual mineral composition. The first division, 

 into Classes, is based on the relative proportions of the salic and femic 

 groups of minerals in the norm (corresponding terms for the mode are 

 felsic and mafic). As Mr. Gr. W. Tyrrell (p. 63) has pointed out, the 

 five classes that are adopted correspond in principle, though in 

 greater detail, to Brogger's division of igneous rocks into leucocratic 

 and melanocratic types. 



The second division, into Orders, is based in the first three classes 

 on the ratio of quartz, or if quartz be absent of felspathoid, to 

 felspar. The orders thus correspond partially to Professor Shand's 

 division of igneous rocks into oversaturated, saturated, and under- 

 saturated types (1913, p. 513, and 1915, p. 340). One of the most 

 serious defects of the CLP. W. Classification, to which attention was 

 drawn by Tyrrell in 1914 (p. 68), lies in the fact that the method of 

 subdivision into orders in Classes IV and Y is based on ratios of the 

 femic minerals, thereby introducing a most confusing break and lack of 

 parallelism between the subdivisions of Classes I to III, and those of 

 IV and V. Tyrrell urges that the salic divisions should be carried on 

 through all the classes, and for the same reason the femic divisions 

 of the later classes might, if desired, be carried back through the 

 early classes, thus providing each rock analysed with a double symbol. 



