OF THE SEA AND LAND. 35 



On Denudations. 



Though I have enquired of such destructions of 

 land as are obviously dependent on assignable cur- 

 rents, there are other appearances of this nature, for 

 which it is difficult to discover a cause. These have 

 been called denudations. But as this term has been 

 applied to every marked loss of land, under whatever 

 cause, I must here limit myself to those of unknown 

 or doubtful origin; merely suggesting one or other 

 of the known causes, where such explanations seem 

 admissible. It is easy to suppose that every ridge or 

 peak has been thus produced, or shaped: but the 

 former remarks on the elevations of strata, show that 

 such a principle must be cautiously adopted and care- 

 fully limited. In countries of trap, however, where 

 the masses are horizontal and the subjacent strata at 

 low angles, we find summits of that rock separated 

 from similar ones, or from larger tracts, in such a 

 manner as to indicate the removal of the intermediate 

 land. In Scotland, the instances of this nature are 

 so abundant that it is unnecessary to specify them : 

 though I shall have occasion to notice one of the most 

 remarkable immediately, in Morven. Where conical 

 detached mountains are formed of strata at low angles, 

 the same explanation must be given ; and the loss of 

 land is inferred in a similar manner, by the corre- 

 spondence of the distant portions. The most remark- 

 able instances of this nature, in our own island, occur 

 on the western coast of Rossshire. Here, a compli- 

 cated group of mountains, consisting of nearly hori- 

 zontal sandstone, is based on gneiss and shaped into 

 independent masses and cones, with intervening chasms 

 and valleys, or extensive open tracts, of mere gneiss, 

 often of many miles in extent. And while, through- 



D 2 



