A REVIEW OF IGNEOUS ROCK CLASSIFICATION 67 



together. It is pointed out that even the subrangs contain a 

 great diversity of rocks, exhibiting" wide variations in almost 

 every constituent, and that the qualitative nomenclature is 

 correspondingly varied. The latter is in part due to " the 

 indefiniteness, confusion, and redundancy of modern nomencla- 

 ture," as shown by C.I.P.W., and admitted by Evans; partly to 

 the inclusion of all rocks chemically belonging to the subrang 

 whatever their texture, which is an important factor in qualita- 

 tive nomenclature. The wide variations in chemical composition, 

 however, are considered by Evans to be due to defects in the 

 classification brought about by the mode of calculation and the 

 subdivision of the norm. Such variations are inherent and 

 inevitable in any classification on a quantitative basis, even in 

 the smaller compartments. The compartments of similar size 

 in the qualitative classifications (such as Hatch's) probably 

 contain an even wider range of rocks. It is to be remarked, 

 however, that the method of the Quantitative Classification does 

 not admit of overlapping between the various compartments, 

 and that each compartment contains rocks differing from those 

 of any other. 



In the investigation of an abnormal manganiferous series of 

 igneous rocks from India, Dr. L. L. Fermor found the Quantita- 

 tive Classification to lack the elasticity necessary to accommodate 

 the series, whilst he found it quite easy to fit it into Hatch's 

 classification. This, however, is very natural, since Hatch's 

 classes are based on silica percentage and are extremely com- 

 prehensive. The wider the compartments of a system, the easier 

 it is to place a given rock. 



The type of Fermor's kodurite series is a phaneric rock 

 consisting of orthoclase, a manganese garnet (spandite), and 

 apatite, and is the basic member of a series of differentiated 

 igneous rocks ranging in acidity from quartz-orthoclase rocks, 

 through intermediate quartz-kodurites and basic kodurites, to 

 manganese-pyroxenites and garnet-rocks. The manganese 

 content in the analysed rocks of this series ranges from 10*50 

 to o'98, and in the typical kodurite is 9*08. The calculation of 

 the norms of these rocks presented some difficulty, which was 

 got over by introducing tephroite (manganese-olivine) as one 

 of the normative molecules. Fermor then shows that the 

 manganese, although fourth in importance in the chemical 

 analysis, and third in the norm, does not affect the classification 



