Classifications of Rocks. 35 



Thus/ the different varieties of granite, of gneiss, of micaceous 

 schist, or of other rocks, have been approximated under cer- 

 tain leading mineral characteristics, so as to found mineralo- 

 gical families ; each of which, however various its members, is 

 distinguished by one general title. The term genus has, by 

 some, been applied to these associations $ and that of species, 

 by others ; but with no great propriety, as they are not 

 amenable to the rules which regulate the forms and differences 

 of organized beings, to which alone it is possible to apply 

 these distinctions usefully. On such a basis may be formed a 

 mineralogical arrangement, or classification of rocks 5 and on 

 it, arrangements of this nature have actually been attempted. 



Now, it is further found that, in nature, there are certain 

 rocks, or families, which possess either a constant or a pre- 

 vailing position ; or that, where many occur together, some 

 one is always the lowest, and some other the highest. If this 

 arrangement were as perfect as it has been imagined, there 

 would be a fixed numerical order of succession ; or every rock 

 might be indicated as well by a number as by a name. It is 

 now well known that this is not the fact ; though there is a 

 prevailing order, which may be rendered of use in geological 

 science. It is also found that there are certain prevailing 

 associations among rocks, or families, in nature 5 and that, in 

 each of these, whatever differences may occur in different 

 places, there are some general rules to which we can always 

 safely refer. Lastly, it is also observed, that, of these different 

 rocks, there are some which are always distinguished by their 

 stratified disposition, while others are as invariably found in 

 shapeless and irregular masses. Such facts present a basis 

 for a geological classification of rocks, or for one which, com- 

 paratively disregarding their mineral characters, attempts to 

 arrange them as integrant parts of the structure of the earth, 

 or of the great system of nature. 



It must now be observed, that certain mineral characters 

 prevail, to the exclusion, more or less complete, of others, in 

 each of those families of rock which are thus distinguished by 

 their geological positions and relations. If these characters, in 

 rocks, were steady and perfect, a mineralogical arrangement 

 might itself be rendered perfect for geological purposes 5 if the 



D 2 



