^{page 149 note *} It may be enough to refer to a series of papers from 1860 to 1866 in the Journal of the Geological Society of London, by Mr Jamieson, of Ellon, forming one of the most valuable contributions made of late years to Scottish geology, and one frequently quoted and relied on by Sir C. Lyell. I give two quotations:—

“The land sank again until the sea in most places reached a height of from 30 to 40 feet above the present tide-mark. … . The clays and beds of silt forming the carses of the Forth, Tay, and other rivers were accumulated.”—1860. Vol. xvi. p. 371.

“A depression now took place. … . In the valley of the Tay and Forth this old coast-line was 25 or 30 feet above the present, but on the coast of Aberdeenshire, not more than 8 or 10. The old estuarine beds or carses of the Forth, Tay, and other rivers were formed, together with corresponding shingle beaches and caves along the coast.”—1865. Vol. xx. p. 195. In this paper the deposits of the Earn are specially described.

^{page 152 note *} For this series of sketches I am indebted to a young friend, Mr W. B. Murray, an art-student of our Edinburgh School, who has been very successful in his rendering of the scenes. Along the Earn it has, in three or four cases, been necessary to suppose the woods thinned, in order to show the real form of the ground, but this has been done as sparingly as possible. On the Teith there was less need for this except in Sketch 12, and even on the Earn all the finest examples of the terraces, such as those in Sketches 6 and 7, are given exactly as they appear in nature.

^{page 154 note *} Nowhere, perhaps, is this opinion more ingeniously stated and defended than in a series of papers by the late Mr Charles Maclaren. See his Select Writings recently published, vol. ii. pp. 186–201. It is from the valley of the Tay that he takes his examples.

^{page 155 note *} As this paper deals only with a question of local geology, I do not refer to any writers except those who have treated of the two rivers to which these researches are confined.

Mr Milne-Home applies this explanation of Lake Margins to the Terraces of the Teith, Trans. Roy. Soc. of Edin., vol. xvi. p. 416.

Dr Fleming supposes a lake to have occupied these river valleys, Lithol. of Edin. p. 76, 1859.

I may refer also to a discussion hefore the Geol. Society of Edin., 19th March 1867, the report of which appeared at the time. Only two theories, the Lacustrine and Marine, found support.

Mr Charles Nicolson, M.A., B.Sc, read a paper on the Surface Formations of the Tay at Perth, describing the Terraces, and advocating the view that they are of Lacustrine origin.

The President Dr Page, Mr Coyne, C.E., and others, gave “their opinion on these Terraces in opposition to Mr Nicolson's Lacustrine Theory maintaining their marine origin, Dr Page instancing the minute examination by Dr R. Chambers of the old sea margins, and Mr Coyne giving his opinion from minute measurements and personal observations.”

^{page 157 note *} There is a beautiful section showing that it consists of finely laminated sands with a little gravel and clay.

^{page 158 note *} Immediately beyond the farm-house the railway gains the summit of terrace c, and the view shown in sketch 1 is seen looking up the valley.

^{page 159 note *} The highest terrace c is made too prominent in the sketch.

^{page 161 note *} To illustrate this, two sections are given in Plate IV. Fig. 2 is from the terrace on the shores of Loch Lubnaig ; fig. 3 is from the railway cutting at Callander. In both the fine grey laminated sands, with their false beddings, are seen to have been denuded in a remarkable way, and are overlaid by coarse gravels.

^{page 166 note *} This does not apply when one follows the stream up among the mountains, where for the most part the terraces are absent. Is it that denudation has swept them wholly away? or is it that during the epoch of these old floods there were glaciers still lingering in the upper portions of the river-valleys ?

^{page 166 note †} More than twenty years ago, Mr Milne-Home described the terraces above Perth as haughs or river flats, but he seems to connect their formation with the bursting of lakes. See “Trans. Roy. Soe. Ed.,” vol. xvi. p. 418. I do not refer to other districts.

^{page 169 note *} See Fig 1, Pl. iv.

^{page 169 note †} There seems good ground for holding that the peat beds of the Earn belong to the time when the land stood comparatively high. But when Mr Jamieson makes the land again sink, and brings in the sea in order to deposit the estuarine mud of the carse, not only does the fossil evidence go against this, but there is the decisive fact already pointed out, that the peat and the carse clay are so associated as to form properly one deposit. At whatever time the one was formed, the other was also. The conclusion to which all the evidence seems to point is, that the whole system of these river terraces was formed at the time when the land was elevated above its present level.

^{page 171 note *} Seen in the valley of the Turrit, for example, above Crieff, and also in that of Monzie, in both of which it underlies the high-level gravels. My attention has been called to the fact that this view had been brought forward in the Memoirs of the Government Survey on Berwickshire, p. 50. 1863.

^{page 171 note †} This forms a continuation of terrace c shown in Sketch 4 on the left bank of the stream. It lies a short way further up the valley.

^{page 172 note *} See Plate iv. fig. 4, showing the arrangement of the gravels and sands forming the upper portion of the terrace.

^{page 173 note *} An attempt has been made to deny this on the ground that Scotland was submerged during the glacial epoch, while Picardy and Devonshire were not—the object being to show that the French valleys were excavated at a later period than the Scottish. But this argument can hardly be urged by those who hold that the formation of valleys is due mainly to subaerial forces and hardly at all to marine action. If the difference between the two countries be that France not having been submerged was continuously acted on by these eroding agencies, while Scotland was withdrawn from them by being buried beneath the sea, how will that prove the French valleys to be of later formation than the Scottish ? So far as that difference goes it should surely prove the reverse.

^{page 173 note †} “On the Amiens Gravel,” by Alford Tyler, Esq., F.G.S., “Journal Geol. Soc. Lond.” vol. xxiv. p. 103. In one respect Mr Tyler's reading is different. He regards the lower terrace, the loess, as the bank of the ancient river when in its ordinary state, and the higher terrace as its bank when in flood—referring both to the same period. This differs materially from the view which I have been led to take, namely, that each terrace is the highest flood-mark of its own time, just as the present banks and haughs are related to the floods of the present time.

^{page 175 note *} Page 101.

^{page 175 note †} It has been suggested by Mr Buchan of the Scottish Meteorological Society, that if the bed of the sea round our coast were elevated, and especially in the direction of Greenland the effect on the climate would be greatly to increase the river floods. Now already, on stratigraphical grounds, we have been led to the conclusion that it was precisely at that period of elevation that our high river terraces were formed. (See note, page 169.) The coincidence is remarkable. The whole stratigraphical evidence makes it probable that these high gravels were deposited just at the time when meteorology teaches us to expect that the river floods would be much beyond the present; and if even in the present state of things there could be such results as the Moray floods have to show, we may be prepared for the still more striking effects of that former age.

Note.—In this paper attention has been called to the absence of marine fossils from the terraces at Bridge of Earn and elsewhere. If such fossils should occur, it would be important to inquire whether they belonged to the time when the terraces were formed. Sometimes portions of antecedent deposits are overlaid or enclosed by the materials of the terraces—portions of rock in situ, for example, or of boulder clay. In the same way there might be found portions of those marine shell clays which belong to a previous period.