ROCK CLASSIFICATION 93 



of similarity, are, nevertheless, so diverse in others, and the laws 

 of their relationships are so unlike, that it is advisable to con- 

 sider them separately. The present discussion will be confined 

 to rocks of the first category — that is, eruptive or igneous 

 rocks. 



The most distinctive features of any igneous rock are those 

 inherent in the mass, namely, the chemical composition, the 

 mineral composition, and the manner in which the constituents 

 are grouped together — the texture and physical aspect of the 

 rock. 1 Other characters of quite as essential natures are, how- 

 ever, less distinctive ; such are the form or dimensions of the 

 body of the rock and its formal relations to adjacent rock 

 masses ; that is, its body as a geological unit and its occurrence 

 or habitat, and finally its connection or relation to other igneous 

 rocks — its association — and hence its origin. 



These two groups of characters are of different orders. The 

 first are clearly material, having to do with the chemical ele- 

 ments and the minerals making the substance of a rock. The 

 second are modal, having to do with forms and relationships 

 among rocks. Failure to recognize this difference has led to 

 confusion of ideas and of methods in classification and in didac- 

 tic treatment. The rapid development of speculation along the 

 lines of rock genesis in conjunction with the accumulation of 

 many facts regarding the composition, occurrence, and associa- 

 tion of igneous rocks, has prevented the proper consideration of 

 these phases of the science apart from one another. 



It seems necessary to call attention to a difference which in 

 this science, at least, must always exist between a classification 

 of the material under investigation- — the rocks — and a treat- 

 ment for purposes of instruction or of discussion of the whole 

 subject of petrology, or of the rocks of any petrographical prov- 

 ince. 



It may be thought at this point that the writer's conception 

 of a classification is too narrow and restricted, and that a system 



*Cf. Michel-Levy, A., Structures et Classification des Roches Eruptives, Paris, 



1889, p. 34. 



