18 



sition must be designated. It is easy to compre- 

 hend that, in the first of these events, it would 

 be impossible to describe the most common geo- 

 logical fact without a tedious repetition ; and that, 

 in the latter, there would be generated a degree 

 of confusion in geological descriptions, which 

 would be productive of the most injurious con- 

 sequences. 



The disadvantages therefore that may be in- 

 dividually opposed to the advantages already 

 stated to arise from a mineralogical classification 

 of rocks, are the following. 



It is unnatural, as far as relates to the most 

 important purpose for which the knowledge of 

 rocks deserves our attention. 



In separating the simple from the compound 

 rocks, it disjoins, and with considerable incon- 

 venience, that which nature has united ; while T 

 in many cases, it associates others which are es- 

 sentially separated from each other in nature. 



In describing a rock but once, and indepen- 

 dently of its different connexions, it abandons one 

 of the most important circumstances in the his 



