48 A NATURAL CLASSIFICATION OF ROCKS. 



subordination to his general knowledo-e of animal structure. The subor- 

 dinate relation of the mineral to the rock is more obscure than the same 

 relation of the bones to tlie animal as a whole, since it is true that the 

 mineral makes up the whole of the majority of rocks. The subordination of 

 the crj'stalline minerals to the rock iniit has been above thus strongly 

 insisted upon since the opposite view seems to be taken by many litho- 

 logists, who appear to study as microscopical mineralogists, instead of work- 

 ing as lithologists proper. 



If it is possible to find the principles of the natural classification of rocks 

 they ought to be applicable to any rock, whatever may be its age and condi- 

 tion. By the natural classification of rocks is meant that system which will 

 place together those forms nearest allied in their general cliaracters, composi- 

 tion, structure, and origin, when the rock as a whole is considered, and not 

 certain of its characters only. The present artificial classifications of rocli^s 

 pick out certain mineralogical characters, and render the rock characters sub- 

 ordinate to them. These classifications are, to a certain extent, natural, and 

 afford a convenient method of arrangement, requiring on the part of the 

 lithologist who follows them simply skill in the determination of minerals. 

 The method works well in some places, in others it masses together a most 

 heterogeneous collection of rocks in a single species — causing some rock 

 names, like diorite for instance, to remind one of the old use of the term 

 schorl in mineralogy. The employment of the minerals alone to determine 

 the rock species is like Linnceus's use of the stamens and pistils in botani- 

 cal classification — a convenient but artificial system. One in objecting to 

 the sexual system in botany does not reject the use of the stamens and pistils 

 in classification, but he does object to their over-riding all other characters. 

 They are to have their just and proportionate w^eight, but no more. So, too, 

 in lithology the minerals in rocks hold a similar relation to them that the 

 sexual organs do in plants — they may comprise all or but little of the rock 

 or plant. Minerals are entitled to their just and joroportionate weight in 

 rock classification but no more; they are not to be allowed, in my judg- 

 ment, to become superior to the rock itself No single character should Ije 

 allowed to over-ride all the others, that is : the presence or absence of a 

 single mineral ought not to remove a rock from the species to which iill its 

 other characters assign it. In the current classifications it frequently' hap- 

 pens that the name given to the rock depends upon the particular portion 

 of the hand-specimen from which the microscopic section was taken. 



