ON THE DESTRUCTION OF ROCKS. 245 



of the whole, and that even all the rocks of this cha- 

 racter which occur in Arran and Sky, may once have 

 been clinkstones or compact blue claystones. The very 

 same phenomena occur among volcanic rocks; and, 

 in both, many of the substances to which the unmean- 

 ing term trachyte has been applied, are only the 

 produce of this change. 



It is not safe to carry this speculation too far; but 

 it is plain that it opens a way towards the explanation 

 of many circumstances in the history of the trap 

 family, which have been hitherto matters of difficulty. 

 It is not easy to imagine, for example, that rocks so 

 earthy and loose as some of the claystones are, were 

 the produce of igneous fusion ; but since such masses 

 of the harder rocks can undergo that change of cha- 

 racter which has been described, this difficulty ceases. 

 In the same manner, it has often been objected to the 

 igneous theory of trap, that it is inconsistent with 

 the presence of clay or unconsolidated rocks in the 

 subjacent positions in which they sometimes occur. 

 The case described as existing in Bute, solves this 

 difficulty; as it will explain many others where similar 

 anomalies have been found to exist. 



Of the mechanical Agents which tend to destroy 



Rocks. 



The mechanical causes which operate in effecting 

 the destruction of rocks may act, either on masses 

 already decomposed, or on rocks which have under- 

 gone no change; but it is unnecessary to distinguish 

 the two cases. The most universal of these is fric- 

 tion, which is here enumerated in the first place, al- 

 though not always called into action till the larger 

 parts have been separated by other causes. But it i 



