Mackintosh — The Sea v. Rivers. 159 



shaped, with an abrupt termination. The sides of the valleys 

 graduate into the slopes of the hills, and are uninterruptedly con- 

 tinued around detached hills (Eound Hill, for instance), in such 

 a manner as to show that the denudation which spared the hills also 

 excavated the valleys down nearly to their bottom. The level at 

 which the action of the streams commenced may, in some of the 

 larger valleys (Ash's valley for instance) be distinctly recognized, 

 being indicated by a sudden and abrupt increase in the steepness of 

 the declivity on both sides. In the lower parts of some of the 

 valleys, the stream cuts through drift similar to that of the open 

 country. In all instances where the streams debouche into the pass 

 of Church Stretton, which is a continuation of the plain of Shrop- 

 shire, they fail to cut a channel in comparatively incoherent gravel, 

 though their waters (the Oakham brook for example) flow as swiftly 

 as in some of the upper parts of their courses. While it is difficult, 

 if not impossible, to reconcile all these facts with the hypothesis of 

 subaerial aqueous agency, I think they can be at once accounted for 

 by the undermining and indenting action of the ocean. The gulleys 

 of the Longmynd are mainly winding and ramifying voes or creeks, 

 excavated while the mountain, as an island, was gradually, or at 

 intervals, suddenly rising above the sea. 



Absence of Correspondence between Biver-courses and Valleys. — In 

 tracing the courses of the larger streams and rivers of England and 

 Wales, we find, in many instances, a much more striking proof 

 of the pre-excavation of valleys than among the gullej^s of the 

 Longmynd. Generally speaking, there is a remarkable absence 

 of correspondence between the size, form, and direction of valleys, 

 and the size, velocity, and the course of the rivers by which they 

 are traversed. A river debouches on a plain through a valley which 

 is an arm or ramification of the plain — the line of escarpment 

 that bounds the plain running up each side of the valley. This 

 valley frequently extends far into the bosom of the hills, and is 

 covered by a continuation of the marine drift that is scattered over 

 the plain. After leaving the valley the river retains as much exca- 

 vating power as before ; but it fails, under equally favourable 

 circumstances, to cut a channel in the plain beyond a few feet 

 or yards in depth. Why, then, give the river credit for having worn 

 down the valley ? A river, after flowing through a plain without 

 making much impression in the shape of a channel, suddenly enters 

 a narrow gorge, through which it flows to the sea ; or the gorge is 

 merely a breach in a barrier, on the other side of which the river 

 re-enters a plain, where, with unim^^aired velocity, it flows only a 

 few feet below the general level. On the supposition of its having 

 formed the gorge, it must have commenced the work of excavation 

 when the level of the plain was at least as high as that of the 

 barrier, and the plain itself must have been worn down by the river. 

 Many considerations prove that the plain could never have been 

 worn down by the river, or its subaerial assistants but supposing 

 this to have been the case, why should the river have made such 

 a wide display of its denudating power on the area of the plain, 



