CLASSIFICATION OF ROCKS. 67 



Now it is also found that certain rocks, or families, 

 possess a constant or a prevailing relative position in 

 nature ; some one, among many, being the lowest, and 

 some other the highest. If this arrangement were as 

 perfect as has been imagined, there would be a fixed 

 numerical order of succession ; or the algebraic nota- 

 tion which has been proposed, under the most unin- 

 telligible ignorance of the facts, would serve the pur- 

 pose of names. I have shown that there is no such 

 order ; though there is a prevailing one, of some use 

 in geological science. Further, however, there are 

 certain prevailing associations of rocks ; so as, among 

 local or casual differences, to permit of certain general 

 rules : while, lastly, some are always distinguished by 

 a stratified disposition, as others occur in irregular 

 masses. Such facts form the basis for a Geological 

 Classification of rocks ; which, comparatively disre- 

 garding their mineral characters, attempts to arrange 

 them as integrant parts of the structure of the earth. 



Now it is doubtless true, that certain mineral cha- 

 racters prevail, to the exclusion of others, in each of 

 the rocks or families thus distinguished by geological 

 relations : and if these distinctions were steady and 

 perfect, a mineralogical arrangement would suffice for 

 geological purposes : as, reversely, a geological one 

 would serve for those of a mineralogical classification. 

 But, to render such an arrangement unexceptionable, 

 the geological order of nature should itself be constant, 

 which it is not ; while the mineralogical classification 

 is not only imperfect, even in its own internal mecha- 

 nism, but at frequent variance with the geological one, 

 as I have fully shown. It is therefore useless for its 

 own declared objects, and pernicious when adopted 

 for geological purposes. But as a tolerably constant 

 peculiarity of mineral character attends the chief geo- 

 logical divisions of rocks, a geological arrangement, 



F 2 



