38 



inedy for the evils arising out of this ambiguity, 

 but to accompany the geological arrangement by 

 geological explanations; and thus the natural 

 history of a rock becomes an essential part of the 

 system of classification. 



For this reason it would have been desirable 

 to have added to each family of rocks, a full his- 

 tory of its geological relations. Thus, indeed the 

 present work was at first drawn up. It became 

 however but too obvious, when completed, that 

 it had assumed the appearance of a system of geo- 

 logy, deficient in many important particulars 

 which it was impossible to combine with the ob- 

 ject first in view, and the presence of which could 

 not be dispensed with in such a system. It seemed 

 preferable therefore to abandon this plan, and to 

 reserve for the contingent event of a future work, 

 that which could not with propriety be introduced 

 into the present. In truth, as the study of rocks 

 is essential to that of geology, so is the latter study 

 in a certain degree necessary to the former. Nei- 

 ther can well be understood without the other ; 

 and that an impossible object has not been at- 



