15 



served, in one class, the simple is placed in ano- 

 ther ; or rather it is not at all to be found in the 

 catalogue of rocks, being described in the mine- 

 ralogical arrangement. Or, the occasional pre- 

 sence of calcareous carbonate in a specimen of 

 micaceous schist, renders it necessary to erect 

 a new species, when perhaps a whole continent 

 may not produce a few yards in extent of such a 

 variety. Thus also we may be led to confound in 

 description, rocks so widely different in geolo- 

 gical connexion as granite and the members of 

 the overlying (or trap) family ; because their dif- 

 ferences in mineral composition and structure are 

 often such as not to be defined by words. 



In concluding these general objections, it 

 may therefore be remarked, that to attempt a 

 perfect classification of rocks, in the present state 

 of our knowledge, appears an useless as well as 

 an injurious sacrifice to a logical order, to which, 

 as far as we can yet perceive, nature refuses 

 to conform. Mineralogists appear, in this in- 

 stance, to have been misled by the example of 



