28 ON CHANGES IN THE DISPOSITION 



other similar evidences existing all over the rocky 

 coasts of Scotland. 



There is perhaps no more satisfactory proof of the 

 disjunction of islands from the main, than the ex- 

 istence, on them, of alluvial matters or fragments 

 derived from rocks which they do not contain. The 

 granite blocks on Lamlash, peculiar as that granite is, 

 demonstrate its former connexion with Arran. Staffa, 

 containing fragments of rocks not found even in Mull, 

 must once have been connected with the mainland 

 itself. But had these islands been disjoined from the 

 main by powerful oceanic or " diluvian" currents, the 

 alluvia must have been swept away; nor can this be 

 evaded by saying that they were then deposited, since 

 they often occupy places where it is impossible that 

 they could have rested under such circumstances. 

 Hence then, again, we must attribute these disjunc- 

 tions to that gradual action of the sea, of which the 

 evidences are every where, and no where more striking 

 than in those enormous walls and prolonged dykes of 

 trap, hereafter noticed under Denudation. 



But there is a rational view of the causes of currents, 

 or great movements of water, in the early condition of 

 the present earth ; while, if not actually proved by any 

 effects that can safely be assigned to them, they must 

 be believed in, because they must have been. This 

 cause was the elevation of the strata: but thus, instead 

 of any general current, or system of currents, from 

 fanciful causes, there must have been numerous partial 

 and irregular ones, differing also in time, as this work 

 proceeded, even down to those days which elevated 

 Italy and Ovvhyhee. We can imagine such events 

 producing almost any complications of simultaneous 

 and successive effects, with any intricacy of consequent 

 results that can be desired, and with powers competent 



