^{page 390 note 1} Professor Ramsay says of the estuary of the Humber (op. cit. p. 147), ‘The sea effected a breach in the rocks. Then suppose these lands to have been heaved up, …. the river then ran through it.’ Prof. Eamsay says much more in proof that he does not deny the efficacy of marine denudation, but we have not space to quote it here.—EDIT.Google Scholar

^{page 390 note 2} Many of the gorges are so narrow that a river could not have found room to bend round and return during the progress of the supposed excavation.

^{page 391 note 1} This I have endeavoured to prove as regards millstone grit and other rocks (see article on Brimham Rocks—Geol. Mag. 04, 1865 Google Scholar). But I think it can be shown that limestone is more or less exempt from any process of dissolution caused by the atmosphere apart from the grinding action of streams. Many limestone cliffs would appear to be both rain-proof and air-proof, as they still retain the smooth grooves and undercuts imprinted on them by the sea. These marks may be distinctly traced on those parts of the Eglwyseg rocks, in Denbighshire, which are not undergoing fragmentary dilapidation. The most striking proofs, however, of the resistance offered by Caboniferous Limestone to mere rain may be seen on a table land to the northwest of Minera, and about eight miles from Llangollen. A number of nearly square flags of limestone, separated by a very regular system of joints, lie flat on the surface. From a distance they look like a vast assemblage of grave-stones. Here and there whole flags or ranges of flags have been carried bodily away, without leaving the slightest trace of their existence, and that, most assuredly, by no kind of atmospheric action. Nearly every remaining stone presents a series of peculiar marks, consisting of smooth, semicircular grooves, from an inch to eight or nine inches in depth. These grooves are generally straight, but sometimes winding, and generally, though not always, roughly parallel; they often turn round at nearly right angles, and run into each other or vanish at the border of the flag. They sometimes terminate in circular perforations. No one would say that glaciers could have formed such a number of deep and complicated grooves. Bain is out of the question, its effect being evidently, as one may sec on the spot, to roughen, and that to no very great extent, the smooth curvilinear outlines of the grooves. There is no appearance of any former riverchannel, and rivers, had they been here, could not have given rise to such indentations. There is only one explanation left, and it may be seen exemplified on the sea-coast of Wales—(I have seen this at Aberystwyth)—namely, the backward and forward motion of pebbles driven by the advancing and receding waves of the sea. Similar wave-worn flags may be seen farther to the south, and I have no doubt in other localities. Their position must be at least 1000 feet above the present level of the sea.

^{page 392 note 1} It is a remarkable fact, that narrow gorges are generally very free from accumulations of drift.

^{page 392 note 2} The best example of this may be seen about two miles to the north of Rhayader, on the left hand side, near to where the railway crosses the Wye.

^{page 392 note 3} MrRandall, , F.G.S., tells me there are similar marks on the rocks under the marine drift near Coalbrookdale.Google Scholar

^{page 392 note 4} Sometimes erroneously called Llyn Gwyn.

^{page 393 note 1} Why is not the sea cutting across ridges and valleys on the west coast of Ireland? Because in Ireland it is now forming ridges and valleys as it anciently did in Wales. On any given coast the sea must be either destroying or producing inequalities.

^{page 393 note 2} Any given area of the earth's surface must, during a given lapse of time, be longer below than above the sea-level, otherwise the proportions of land and sea could not be maintained.

^{page 394 note 1} The shores of the intricate channels and inlets on the Paciffic coast of British North America, if elevated from the sea, would present but slight difference from sides of the narrow valleys in the Rocky Mountains at an altitude of 3,500 feet.” — DrHector, , Quart. Journ. Geol. Soc., Vol. xvii, part 1.Google Scholar

^{page 395 note 1} To the west of Aber, the channel excavated by the Ogwen may be traced from the sea to Llyn Ogwen as a depression distinct from the outline of the wide area over which it flows in the lower part of its course, and equally distinct from the great gorge of Nant Francon, which, with the precipice of Ben Glog, must have existed before the supplementary action of the river commenced. Fig. 10, Plate XV., is a view (from a photograph) of all that a considerable-sized tributary of the Ogwen has been able to effect in modifying the outline of Nant Francon.

^{page 396 note 1} Between Llangollen.and Corwen, on the left hand side of the road, streams with cascades may be seen descending from the Berwyn Mountains, with little or no channel to mark their course. Behind Corwen a powerful stream has, in some places, excavated no channel at all, in other places a miniature gorge. The excavating power of this stream is assisted by a steeply-inclined course, and numerous cascades; and the gorge may, with the greatest safety, be regarded as the measure of its denuding capabilities, as the ground on each side rather falls away from it than towards it.

^{page 396 note 2} Fig. 12, Plate XV., is a distant view of this pass. In most of the wet passes of Wales and the Lake District, with which I am acquainted, the streams which flow away from the cols could have had no share in the formation of the cols themselves, which indicate an agency cutting straight through a ridge, and cannot be referred to an improbable linear coincidence in the original sources of the streams.

^{page 396 note 3} See Mr.Kinahan's able article on raised beaches in Ireland in Geol. Mag., 08., 1866.Google Scholar

^{page 397 note 1} Excepting at intermediate levels where streams have had space to acquire considerable volume, without descending so far as to lose sufficient inclination of channel. This may be called the zone of maximum atmospheric denudation—the higher and lower lands the minimum zones.

^{page 398 note 1} Answers to other parts of Mr. Maw's article will be found forestalled in the preceding pages.