Another series of observations made by Captain Graah, during a survey of Greenland between 1823 and 1829, and by Dr. Pingel in 1830-32, adds not a little to the geological interest of the "outskirts," in their bearing on glacial phenomena of ancient date. Those Danish investigators, with one of whom, Dr. Pingel, I conversed at Copenhagen in 1834, ascertained that the whole coast from latitude 60 to about 70 degrees north has been subsiding for the last four centuries, so that some ancient piles driven into the beach to support the boats of the settlers have been gradually submerged, and wooden buildings have had to be repeatedly shifted farther inland.* (* "Principles of Geology" chapter 30.)

In Norway and Sweden, instead of such a subsiding movement, the land is slowly rising; but we have only to suppose that formerly, when it was covered like Greenland with continental ice, it sank at the rate of several feet in a century, and we shall be able to explain why marine deposits are found above the level of the sea, and why these generally overlie polished and striated surfaces of rock.

We know that Greenland was not always covered with snow and ice, for when we examine the Tertiary strata of Disco Island (of the Upper Miocene period) we discover there a multitude of fossil plants, which demonstrate that, like many other parts of the arctic regions, it formerly enjoyed a mild and genial climate. Among the fossils brought from that island, latitude 70 degrees north, Professor Heer has recognised Sequoia Langsdorfii, a coniferous species which flourished throughout a great part of Europe in the Miocene period, and is very closely allied to the living Sequoia sempervirens of California. The same plant has been found fossil by Sir John Richardson within the arctic circle, far to the west on the Mackenzie River, near the entrance of Bear River, also by some Danish naturalists in Iceland to the east. The Icelandic surturbrand, or lignite, of this age has also yielded a rich harvest of plants, more than thirty-one of them, according to Steenstrup and Heer, in a good state of preservation, and no less than fifteen specifically identical with Miocene plants of Europe. Thirteen of the number are arborescent; and amongst others is a tulip-tree (Liriodendron), with its fruit and characteristic leaves, a plane (Platanus), a walnut, and a vine, affording unmistakable evidence of a climate in the parallel of the arctic circle which precludes the supposition of glaciers then existing in the neighbourhood, still less any general crust of continental ice, like that of Greenland.* (* Heer, "Recherches sur la Vegetation du Pays tertiaire" etc. 1861 page 178.)

As the older Pliocene flora of the Tertiary strata of Italy, like the shells of the Coralline Crag, before adverted to, Chapter 12, indicate a temperature milder than that now prevailing in Europe, though not so warm as that of the Upper Miocene period, it is probable that the accumulation of snow and glaciers on the mountains and valleys of Greenland did not begin till after the commencement of the Pliocene period, and may not have reached its maximum until the close of that period.

Norway and Sweden appear to have passed through all the successive phases of glaciation which Greenland has experienced, and others which that country will one day undergo, if the climate which it formerly enjoyed should ever be restored to it. There must have been first a period of separate glaciers in Scandinavia, then a Greenlandic state of continental ice, and thirdly, when that diminished, a second period of enormous separate glaciers filling many a valley now wooded with fir and birch. Lastly, under the influence of the Gulf Stream, and various changes in the height and extent of land in the arctic circle, a melting of nearly all the permanent ice between latitudes 60 and 70 north, corresponding to the parallels of the continental ice of Greenland, has occurred, so that we have now to go farther north than latitude 70 degrees before we encounter any glacier coming down to the sea coast. Among other signs of the last retreat of the extinct glaciers, Kjerulf and other authors describe large transverse moraines left in many of the Norwegian and Swedish glens.

CHRONOLOGICAL RELATIONS OF THE HUMAN AND GLACIAL PERIODS IN SWEDEN.

We may now consider whether any, and what part, of these changes in Scandinavia may have been witnessed by Man. In Sweden, in the immediate neighbourhood of Upsala, I observed, in 1834, a ridge of stratified sand and gravel, in the midst of which occurs a layer of marl, evidently formed originally at the bottom of the Baltic, by the slow growth of the mussel, cockle, and other marine shells of living species intermixed with some proper to fresh water. The marine shells are all of dwarfish size, like those now inhabiting the brackish waters of the Baltic; and the marl, in which myriads of them are embedded, is now raised more than 100 feet above the level of the Gulf of Bothnia. Upon the top of this ridge (one of those called osars in Sweden) repose several huge erratics consisting of gneiss, for the most part unrounded, from 9 to 16 feet in diameter, and which must have been brought into their present position since the time when the neighbouring gulf was already characterised by its peculiar fauna. Here, therefore, we have proof that the transport of erratics continued to take place, not merely when the sea was inhabited by the existing Testacea, but when the north of Europe had already assumed that remarkable feature of its physical geography, which separates the Baltic from the North Sea, and causes the Gulf of Bothnia to have only one-fourth of the saltness belonging to the ocean.

I cannot doubt that these large erratics of Upsala were brought into their present position during the Recent period, not only because of their moderate elevation above the sea-level in a country where the land is now rising every century, but because I observed signs of a great oscillation of level which had taken place at Sodertelje, south of Stockholm (about 45 miles distant from Upsala), after the country had been inhabited by Man. I described, in the "Philosophical Transactions" for 1835, the section there laid open in digging a level in 1819, which showed that a subsidence followed by a re-elevation of land, each movement amounting to more than 60 feet, had occurred since the time when a rude hut had been built on the ancient shore. The wooden frame of the hut, with a ring of hearthstones on the floor, and much charcoal, were found, and over them marine strata, more than 60 feet thick, containing the dwarf variety of Mytilus edulis, and other brackish-water shells of the Bothnian Gulf. Some vessels put together with wooden pegs, of anterior date to the use of metals, were also embedded in parts of the same marine formation, which has since been raised, so that the upper beds are more than 60 feet above the sea-level, the hut being thus restored to about its original position relatively to the sea.

We have seen in the account of the Danish kitchen-middens of the Recent period that even at the comparatively late period of their origin the waters of the Baltic had been rendered more salt than they are now. The Upsala erratics may belong to nearly the same era as these. But were we to go back to a long antecedent epoch, or to that of the Belgian and British caves with their extinct animals, and the signs they afford of a state of physical geography departing widely from the present, or to the era of the implement-bearing alluvium of St. Acheul, we might expect to find Scandinavia overwhelmed with glaciers, and the country uninhabitable by Man. At a much remoter period the same country was in the state in which Greenland now is, overspread with one uninterrupted coating of continental ice, which has left its peculiar markings on the highest mountains. This period, probably anterior to the earliest traces yet brought to light of the human race, may have coincided with the submergence of England, and the accumulation of the boulder-clay of Norfolk, Suffolk, and Bedfordshire, before mentioned. It has already been stated that the syenite and some other rocks of the Norfolk till seem to have come from Scandinavia, and there is no era when icebergs are so likely to have floated them so far south as when the whole of Sweden and Norway were enveloped in a massive crust of ice; a state of things the existence of which is deduced from the direction of the glacial furrows, and their frequent unconformity to the shape of the minor valleys.

GLACIAL PERIOD IN SCOTLAND [NOTE 25].

Professor Agassiz, after his tour in Scotland in 1840, announced the opinion that erratic blocks had been dispersed from the Scottish mountains as from an independent centre, and that the capping of ice had been of extraordinary thickness.* (* Agassiz, "Proceedings of the Geological Society" 1840 and "Edinburgh Philosophical Journal" 49 page 79.)

Mr. Robert Chambers, after visiting Norway and Sweden, and comparing the signs of glacial action observed there with similar appearances in the Grampians, came to the conclusion that the Highlands both of Scandinavia and Scotland had once been "moulded in ice," and that the outward and downward movement and pressure of the frozen mass had not only smoothed, polished, and scratched the rocks, but had, in the course of ages, deepened and widened the valleys, and produced much of that denudation which has commonly been ascribed exclusively to aqueous action. The glaciation of the Scotch mountains was traced by him to the height of at least 3000 feet.* (* "Ancient Sea Margins" Edinburgh 1848. Glacial Phenomena "Edinburgh New Philosophical Journal" April 1853 and January 1855.)

Mr. T.F. Jamieson, of Ellon, in Aberdeenshire, has recently brought forward an additional body of facts in support of this theory. According to him the Grampians were at the period of extreme cold enveloped "in one great winding sheet of snow and ice," which reached everywhere to the coast-line, the land being then more elevated than it is now. He describes the glacial furrows sculptured on the solid rocks as pointing in Aberdeenshire to the south-east, those of the valley of the Forth at Edinburgh, from west to east, and higher up the same valley at Stirling, from north-west to south-east, as they should do if the ice had followed the lines of what is now the principal drainage. The observations of Sir James Hall, Mr. Maclaren, Mr. Chambers, and Dr. Fleming, are cited by him in confirmation of this arrangement of the glacial markings, while in Sutherland and Ross-shire he shows that the glacial furrows along the north coast point northwards, and in Argyleshire westwards, always in accordance with the direction of the principal glens and fjords.

Another argument is also adduced by him in proof of the ice having exerted its mechanical force in a direction from the higher and more inland country to the lower region and sea-coast. Isolated hills and minor prominences of rock are often polished and striated on the land side, while they remain rough and jagged on the side fronting the sea. This may be seen both on the east and west coast. Mention is also made of blocks of granite which have travelled from south to north in Aberdeenshire, of which there would have been no examples had the erratics been all brought by floating ice from the arctic regions when Scotland was submerged. It is also urged against the doctrine of attributing the general glaciation to submergence, that the glacial grooves, instead of radiating as they do from a centre, would, if they had been due to ice coming from the north, have been parallel to the coast-line, to which they are now often almost at right angles. The argument, moreover, which formerly had most weight in favour of floating ice, namely, that it explained why so many of the stones did not conform to the contour and direction of the minor hills and valleys, is now brought forward, and with no small effect, in favour of the doctrine of continental ice on the Greenlandic scale, which, after levelling up the lesser inequalities, would occasionally flow in mighty ice-currents, in directions often at a high angle to the smaller ridges and glens.

The application to Scandinavia and Scotland of this theory makes it necessary to reconsider the validity of the proofs formerly relied on as establishing the submergence of a great part of Scotland beneath the sea, at some period subsequent to the commencement of the glacial period. In all cases where marine shells overlie till, or rest on polished and striated surfaces of rock, the evidence of the land having been under water, and having been since upheaved, remains unshaken; but this special proof rarely extends to heights exceeding 500 feet. In the basin of the Clyde we have already seen that Recent strata occur 25 feet above the sea-level, with existing species of marine testacea, and with buried canoes, and other works of art. At the higher level of 50 feet occurs the well-known raised beach of the western coast, which, according to Mr. Jamieson, contains, near Fort William and on Loch Fyne and elsewhere, an assemblage of shells implying a colder climate than that of the 25-foot terrace, or that of the present sea; just as, in the valley of the Somme, the higher-level gravels are supposed to belong to a colder period than the lower ones, and still more decidedly than that of the present era. At still greater elevations, older beds containing a still more arctic group of shells have been observed at Airdrie, 14 miles south-east of Glasgow, 524 feet above the level of the sea. They were embedded in stratified clays, with the unstratified boulder till both above and below them, and in the overlying unstratified drift were some boulders of granite which must have come from distances of 60 miles at the least.* (* Smith of Jordanhill, "Quarterly Journal of the Geological Society" volume 6 1850 page 387.) The presence of Tellina calcarea, and several other northern shells, implies a climate colder than that of the present Scottish seas. In the north of Scotland, marine shells have been found in deposits of the same age in Caithness and in Aberdeenshire at heights of 250 feet, and on the shores of the Moray Firth, as at Gamrie in Banff, at an elevation of 350 feet; and the stratified sands and beds of pebbles which belong to the same formation ascend still higher—to heights of 500 feet at least.* (* Prestwich, "Proceedings of the Geological Society" volume 2 page 545; Jamieson, "Quarterly Journal of the Geological Society" volume 16 1860.)

At much greater heights, stratified masses of drift occur in which hitherto no organic remains, whether of marine or freshwater animals, have ever been found. It is still an undecided question whether the origin of all such deposits in the Grampians can be explained without the intervention of the sea. One of the most conspicuous examples has been described by Mr. Jamieson as resting on the flank of a hill called Meal Uaine, in Perthshire, on the east side of the valley of the Tummel, just below Killiecrankie. It consists of perfectly horizontal strata, the lowest portion of them 300 feet above the river and 600 feet above the sea. From this elevation to an altitude of nearly 1200 feet the same series of strata is traceable, continuously, up the slope of the mountain, and some patches are seen here and there even as high as 1550 feet above the sea. They are made up in great part of finely laminated silt, alternating with coarser materials, through which stones from 4 to 5 feet in length are scattered. These large boulders, and some smaller ones, are polished on one or more sides, and marked with glacial striae. The subjacent rocks, also, of gneiss, mica slate, and quartz, are everywhere grooved and polished as if by the passage of a glacier.* (* Jamieson, "Quarterly Journal of the Geological Society" volume 16 1860 page 360.)

At one spot a vertical thickness of 130 feet of this series of strata is exposed to view by a mountain torrent, and in all more than 2000 layers of clay, sand, and gravel were counted, the whole evidently accumulated under water. Some beds consist of an impalpable mud, like putty, apparently derived from the grinding down of felspar, and resembling the mud produced by the grinding action of modern glaciers.

Mr. Jamieson, when he first gave an account of this drift, inferred, in spite of the absence of marine shells, that it implied the submergence of Scotland beneath the ocean after the commencement of the glacial period, or after the era of continental ice indicated by the subjacent floor of polished and grooved rock. This conclusion would require a submergence of the land as far up as 1550 feet above the present sea-level, after which a great re-upheaval must have occurred. But the same author, having lately revisited the valley of the Tummel, suggests another possible, and I think probable, explanation of the same phenomena. The stratified drift in question is situated in a deep depression between two buttresses of rock, and if an enormous glacier be supposed to have once filled the valley of the Tummel to the height of the stratified drift, it may have dammed up the mouth of a mountain torrent by a transverse barrier, giving rise to a deep pond, in which beds of clay and sand brought down by the waters of the torrent were deposited. Charpentier in his work on the Swiss glaciers has described many such receptacles of stratified matter now in progress, and due to such blockages, and he has pointed out the remnants of ancient and similar formations left by extinct glaciers of an earlier epoch. He specially notices that angular stones of various dimensions, often polished and striated, which rest on the glacier and are let fall when the torrent undermines the side of the moving ice, descend into the small lake and become interstratified with the gravel and fine sediment brought down by the torrent into the same.* (* Charpentier, "Essai sur les Glaciers" page 63 1841.)

The evidence of the former sojourn of the sea upon the land after the commencement of the glacial period was formerly inferred from the height to which erratic blocks derived from distant regions could be traced, besides the want of conformity in the glacial furrows to the present contours of many of the valleys. Some of these phenomena may now, as we have seen, be accounted for by assuming that there was once a crust of ice resembling that now covering Greenland.

The Grampians in Forfarshire and in Perthshire are from 3000 to 4000 feet high. To the southward lies the broad and deep valley of Strathmore, and to the south of this again rise the Sidlaw Hills to the height of 1500 feet and upwards. On the highest summits of this chain, formed of sandstone and shale, and at various elevations, I have observed huge angular fragments of mica-schist, some 3 and others 15 feet in diameter, which have been conveyed for a distance of at least 15 miles from the nearest Grampian rocks from which they could have been detached. Others have been left strewed over the bottom of the large intervening vale of Strathmore.* (* "Proceedings of the Geological Society" volume 3 page 344.)

It may be argued that the transportation of such blocks may have been due not to floating ice, but to a period when Strathmore was filled up with land ice, a current of which extended from the Perthshire Highlands to the summit of the Sidlaw Hills, and the total absence of marine or freshwater shells from all deposits, stratified or unstratified, which have any connection with these erratics in Forfarshire and Perthshire may be thought to favour such a theory.

But the same mode of transport can scarcely be imagined for those fragments of mica-schist, one of them weighing from 8 to 10 tons, which were observed much farther south by Mr. Maclaren on the Pentland Hills, near Edinburgh, at the height of 1100 feet above the sea, the nearest mountain composed of this formation being 50 miles distant.* (* Maclaren, "Geology of Fife" etc. page 220.) On the same hills, also, at all elevations, stratified gravels occur which, although devoid of shells, it seems hardly possible to refer to any but a marine origin.

Although I am willing, therefore, to concede that the glaciation of the Scotch mountains, at elevations exceeding 2000 feet, may be explained by land ice, it seems difficult not to embrace the conclusion that a subsidence took place not merely of 500 or 600 feet, as demonstrated by the marine shells, but to a much greater amount, as shown by the present position of erratics and some patches of stratified drift. The absence of marine shells at greater heights than 525 feet above the sea, will be treated of in a future chapter. It may in part, perhaps, be ascribed to the action of glaciers, which swept out marine strata from all the higher valleys, after the re-emergence of the land.

LATEST CHANGES PRODUCED BY GLACIERS IN SCOTLAND.

We may next consider the state of Scotland after its emergence from the glacial sea, when we cannot fail to be approaching the time when Man co-existed with the mammoth and other mammalia now extinct. In a paper which I published in 1840, on the ancient glaciers of Forfarshire, I endeavoured to show that some of these existed after the mountains and glens had acquired precisely their present shape,* (* "Proceedings of the Geological Society" volume 3 page 337.) and had left moraines even in the minor valleys, just where they would now leave them were the snow and ice again to gain ground. I described also one remarkable transverse mound, evidently the terminal moraine of a retreating glacier, which crosses the valley of the South Esk, a few miles above the point where it issues from the Grampians, and about 6 miles below the Kirktown of Clova. Its central part, at a place called Glenarm, is 800 feet above the level of the sea. The valley is about half a mile broad, and is bounded by steep and lofty mountains, but immediately above the transverse barrier it expands into a wide alluvial plain, several miles broad, which has evidently once been a lake. The barrier itself, about 150 feet high, consists in its lower part of till with boulders, 50 feet thick, precisely resembling the moraine of a Swiss glacier, above which there is a mass of stratified sand, varying in thickness from 50 to 100 feet, which has the appearance of consisting of the materials of the moraine rearranged in a stratified form, possibly by the waters of a glacier lake. The structure of the barrier has been laid open by the Esk, which has cut through it a deep passage about 400 yards wide.

I have also given an account of another striking feature in the physical geography of Perthshire and Forfarshire, which I consider to belong to the same period; namely, a continuous zone of boulder clay, forming ridges and mounds from 50 to 70 feet high (the upper part of the mounds usually stratified), enclosing numerous lakes, some of them several miles long, and many ponds and swamps filled with shell-marl and peat. This band of till, with Grampian boulders and associated river-gravel, may be traced continuously for a distance of 34 miles, with a width of 3 1/2 miles, from near Dunkeld, by Coupar, to the south of Blairgowrie, then through the lowest part of Strathmore, and afterwards in a straight line through the greatest depression in the Sidlaw Hills, from Forfar to Lunan Bay.

Although no great river now takes its course through this line of ancient lakes, moraines, and river gravel, yet it evidently marks an ancient line by which, first, a great glacier descended from the mountains to the sea, and by which, secondly, at a later period, the principal water drainage of this country was effected. The subsequent modification in geography is comparable in amount to that which has taken place since the higher level gravels of the valley of the Somme were formed, or since the Belgian caves were filled with mud and bone-breccia.

(FIGURE 35. OVAL AND FLATTISH PEBBLES IN DESERTED CHANNELS.)

Mr. Jamieson has remarked, in reference to this and some other extinct river-channels of corresponding date, that we have the means of ascertaining the direction in which the waters flowed by observing the arrangement of the oval and flattish pebbles in their deserted channels; for in the bed of a fast-flowing river such pebbles are seen to dip towards the current, as represented in Figure 35, such being the position of greatest resistance to the stream.* (* Jamieson, "Quarterly Journal of the Geological Society" volume 16 1860 page 349.) If this be admitted, it follows that the higher or mountainous country bore the same relation to the lower lands, at the time when a great river passed through this chain of lakes, as it does at present.

We also seem to have a test of the comparatively modern origin of the mounds of till which surround the above-mentioned chain of lakes (of which that of Forfar is one), in the species of organic remains contained in the shell-marl deposited at their bottom. All the mammalia as well as shells are of recent species. Unfortunately, we have no information as to the fauna which inhabited the country at the time when the till itself was formed. There seem to be only three or four instances as yet known in all Scotland of mammalia having been discovered in boulder clay.

Mr. R. Bald has recorded the circumstances under which a single elephant's tusk was found in the unstratified drift of the valley of the Forth, with the minuteness which such a discovery from its rarity well deserved. He distinguishes the boulder clay, under the name of "the old alluvial cover," from that more modern alluvium, in which the whales of Airthrie, described in Chapter 3, were found. This cover he says is sometimes 160 feet thick. Having never observed any organic remains in it, he watched with curiosity and care the digging of the Union Canal between Edinburgh and Falkirk, which passed for no less than 28 miles almost continuously through it. Mr. Baird, the engineer who superintended the works, assisted in the inquiry, and at one place only in this long section did they meet with a fossil, namely, at Cliftonhall, in the valley of the Almond. It lay at a depth of between 15 and 20 feet from the surface, in very stiff clay, and consisted of an elephant's tusk, 39 inches long and 13 in circumference, in so fresh a state that an ivory turner purchased it and turned part of it into chessmen before it was rescued from destruction. The remainder is still preserved in the museum at Edinburgh, but by exposure to the air it has shrunk considerably.* (* "Memoirs of the Wernerian Society" Edinburgh volume 4 page 58.) In 1817, two other tusks and some bones of the elephant, as we learn from the same authority (Mr. Bald), were met with, 3 1/2 feet long and 13 inches in circumference, lying in an horizontal position, 17 feet deep in clay, with marine shells, at Kilmaurs, in Ayrshire. The species of shells are not given.* (* Ibid. volume 4 page 63.)

In another excavation through the Scotch boulder clay, made in digging the Clyde and Forth Junction Railway, the antlers of a reindeer were found at Croftamie, in Dumbartonshire, in the basin of the river Endrick, which flows into Loch Lomond. They had cut through 12 feet of till with angular and rounded stones, some of large size, and then through 6 feet of underlying clay, when they came upon the deer's horns, 18 feet from the surface, and within a foot of the sandstone on which the till rested. At the distance of a few yards, and in the same position, but a foot or two deeper, were observed marine shells, Cyprina islandica, Astarte elliptica, A. compressa, Fusus antiquus, Littorina littorea, and a Balanus. The height above the level of the sea was between 100 and 103 feet. The reindeer's horn was seen by Professor Owen, who considered it to be that of a young female of the large variety, called by the Hudson's Bay trappers the caribou.

The remains of elephants, now in the museums of Glasgow and Edinburgh, purporting to come from the superficial deposits of Scotland have been referred to Elephas primigenius. In cases where tusks alone have been found unaccompanied by molar teeth, such specific determinations may be uncertain; but if any one specimen be correctly named, the occurrence of the mammoth and reindeer in the Scotch boulder-clay, as both these quadrupeds are known to have been contemporary with Man, favours the idea which I have already expressed, that the close of the glacial period in the Grampians may have coincided in time with the existence of Man in those parts of Europe where the climate was less severe, as, for example, in the basins of the Thames, Somme, and Seine, in which the bones of many extinct mammalia are associated with flint implements of the antique type.

PARALLEL ROADS OF GLEN ROY IN SCOTLAND.

(PLATE 2. VIEW OF THE MOUTHS OF GLEN ROY AND GLEN SPEAN, BY SIR T. DICK LAUDER. VV. Hill of Bohuntine. VVV. Glen Roy. V(inverted)V. Mealderry. V. Entrance of Glen Spean VV(superscript)V. Point of division between Glens Roy and Spean.)

Perhaps no portion of the superficial drift of Scotland can lay claim to so modern an origin on the score of the freshness of its aspect, as that which forms what are called the Parallel Roads of Glen Roy. If they do not belong to the Recent epoch, they are at least posterior in date to the present outline of mountain and glen, and to the time when every one of the smaller burns ran in their present channels, though some of them have since been slightly deepened. The almost perfect horizontality, moreover, of the roads, one of which is continuous for about 20 miles from east to west, and 12 miles from north to south, shows that since the era of their formation no change has taken place in the relative levels of different parts of the district.

(FIGURE 36. MAP OF THE PARALLEL ROADS OF GLEN ROY OR LOCHABER.

A. five miles distant south-west from this point is Fort William, where the Lochy joins an arm of the sea, called Loch Eil. Vertical lines. Cols or watersheds at the heads of the glens—once the westward outlet of the lakes. Dots. Conspicuous delta deposits as laid down by Mr. T.F. Jamieson.)

Glen Roy is situated in the Western Highlands, about 10 miles east-north-east of Fort William, near the western end of the great glen of Scotland, or Caledonian Canal, and near the foot of the highest of the Grampians, Ben Nevis. (See map, Figure 36.) Throughout nearly its whole length, a distance of more than 10 miles, three parallel roads or shelves are traced along the steep sides of the mountains, as represented in the annexed view, Plate 2, by the late Sir T. Dick Lauder, each maintaining a perfect horizontality, and continuing at exactly the same level on the opposite sides of the glen. Seen at a distance, they appear like ledges, or roads, cut artificially out of the sides of the hills; but when we are upon them, we can scarcely recognise their existence, so uneven is their surface, and so covered with boulders. They are from 10 to 60 feet broad, and merely differ from the side of the mountain by being somewhat less steep.

On closer inspection, we find that these terraces are stratified in the ordinary manner of alluvial or littoral deposits, as may be seen at those points where ravines have been excavated by torrents. The parallel shelves, therefore, have not been caused by denudation, but by the deposition of detritus, precisely similar to that which is dispersed in smaller quantities over the declivities of the hills above. These hills consist of clay-slate, mica schist, and granite, which rocks have been worn away and laid bare at a few points immediately above the parallel roads. The lowest of these roads is about 850 feet above the level of the sea, the next about 212 feet higher, and the third 82 feet above the second. There is a fourth shelf, which occurs only in a contiguous valley called Glen Gluoy, which is 12 feet above the highest of all the Glen Roy roads, and consequently about 1156 feet above the level of the sea. * (* Another detached shelf also occurs at Kilfinnan. (See Map, Figure 36.)) One only, the lowest of the three roads of Glen Roy, is continued throughout Glen Spean, a large valley with which Glen Roy unites. (See Plate 2 and map, Figure 36.) As the shelves, having no slope towards the sea like ordinary river terraces, are always at the same absolute height, they become continually more elevated above the river in proportion as we descend each valley; and they at length terminate very abruptly, without any obvious cause, or any change either in the shape of the ground or in the composition or hardness of the rocks.

I should exceed the limits of this work, were I to attempt to give a full description of all the geographical circumstances attending these singular terraces, or to discuss the ingenious theories which have been severally proposed to account for them by Dr. Macculloch, Sir T. Lauder, and Messrs. Darwin, Agassiz, Milne, and Chambers. There is one point, however, on which all are agreed, namely, that these shelves are ancient beaches, or littoral formations, accumulated round the edges of one or more sheets of water which once stood for a long time successively at the level of the several shelves.

(FIGURE 37. SECTION THROUGH SIDE OF LOCH.

AB. Supposed original surface of rock. CD. Roads or shelves in the outer alluvial covering of the hill.)

It is well known, that wherever a lake or marine fjord exists surrounded by steep mountains subject to disintegration by frost or the action of torrents, some loose matter is washed down annually, especially during the melting of snow, and a check is given to the descent of this detritus at the point where it reaches the waters of the lake. The waves then spread out the materials along the shore, and throw some of them upon the beach; their dispersing power being aided by the ice, which often adheres to pebbles during the winter months, and gives buoyancy to them. The annexed diagram (Figure 37) illustrates the manner in which Dr. MacCulloch and Mr. Darwin suppose "the roads" to constitute mere excrescences of the superficial alluvial coating which rests upon the hillside, and consists chiefly of clay and sharp unrounded stones.

Among other proofs that the parallel roads have really been formed along the margin of a sheet of water, it may be mentioned, that wherever an isolated hill rises in the middle of the glen above the level of any particular shelf, as in Mealderry, Plate 2, a corresponding shelf is seen at the same level passing round the hill, as would have happened if it had once formed an island in a lake or fjord. Another very remarkable peculiarity in these terraces is this; each of them comes in some portion of its course to a col, or parting ridge, between the heads of glens, the explanation of which will be considered in the sequel.

Those writers who first advocated the doctrine that the roads were the ancient beaches of freshwater lakes, were unable to offer any probable hypothesis respecting the formation and subsequent removal of barriers of sufficient height and solidity to dam up the water. To introduce any violent convulsion for their removal was inconsistent with the uninterrupted horizontality of the roads, and with the undisturbed aspect of those parts of the glens where the shelves come suddenly to an end.

Mr. Agassiz and Dr. Buckland, desirous, like the defenders of the lake theory, to account for the limitation of the shelves to certain glens, and their absence in contiguous glens, where the rocks are of the same composition, and the slope and inclination of the ground very similar, first started the theory that these valleys were once blocked up by enormous glaciers descending from Ben Nevis, giving rise to what are called, in Switzerland and in the Tyrol, glacier-lakes. In corroboration of this view, they contended that the alluvium of Glen Roy, as well as of other parts of Scotland, agrees in character with the moraines of glaciers seen in the Alpine valleys of Switzerland. It will readily be conceded that this hypothesis was preferable to any previous lacustrine theory, by accounting more easily for the temporary existence and entire disappearance of lofty transverse barriers, although the height required for the supposed dams of ice appeared very enormous.

Before the idea of glacier-lakes had been suggested by Agassiz, Mr. Darwin examined Glen Roy, and came to the opinion that the shelves were formed when the glens were still arms of the sea, and, consequently, that there never were any seaward barriers. According to him, the land emerged during a slow and uniform upward movement, like that now experienced throughout a large part of Sweden and Finland; but there were certain pauses in the upheaving process, at which times the waters of the sea remained stationary for so many centuries as to allow of the accumulation of an extraordinary quantity of detrital matter, and the excavation, at many points immediately above the sea-level, of deep notches and bare cliffs in the hard and solid rock.

This theory I adopted in 1841 ("Elements," 2nd edition), as appearing to me less objectionable than any other then proposed. The phenomena most difficult to reconcile with it are, first, the abrupt cessation of the roads at certain points in the different glens; secondly, their unequal number in different valleys connecting with each other, there being three, for example, in Glen Roy, and only one in Glen Spean; thirdly, the precise horizontality of level maintained by the same shelf over a space many leagues in length, requiring us to assume, that during a rise of 1156 feet no one portion of the land was raised even a few yards above another; fourthly, the coincidence of level already alluded to of each shelf with a col, or the point forming the head of two glens, from which the rain-waters flow in opposite directions. This last-mentioned feature in the physical geography of Lochaber Mr. Darwin endeavoured to explain in the following manner. He called these cols "land-straits," and regarding them as having been anciently sounds or channels between islands, he pointed out that there is a tendency in such sounds to be silted up, and always the more so in proportion to their narrowness. In a chart of the Falkland Islands, by Captain Sulivan, R.N., it appears that there are several examples there of straits where the soundings diminish regularly towards the narrowest part. One is so nearly dry that it can be walked over at low water, and another, no longer covered by the sea, is supposed to have recently dried up in consequence of a small alteration in the relative level of sea and land. "Similar straits," observes Mr. Chambers, "hovering, in character, between sea and land, and which may be called fords, are met with in the Hebrides. Such, for example, is the passage dividing the islands of Lewis and Harris, and that between North Uist and Benbecula, both of which would undoubtedly appear as cols, coinciding with a terrace or raised beach, all round the islands if the sea were to subside."* (* R. Chambers, "Ancient Sea Margins" page 114.)

The first of the difficulties above alluded to, namely, the non-extension of the shelves over certain parts of the glens, might be explained, said Mr. Darwin, by supposing in certain places a quick growth of green turf on a good soil, which prevented the rain from washing away any loose materials lying on the surface. But wherever the soil was barren, and where green sward took long to form, there may have been time for the removal of the gravel. In one case an intermediate shelf appears for a short distance (three quarters of a mile) on the face of the mountain called Tombhran, between the two upper shelves, and is seen nowhere else. It occurs where there was the longest space of open water, and where the waves may have acquired a more than ordinary power to heap up detritus.

The unequal number of the shelves in valleys communicating with each other, and in which the boundary rocks are similar in composition, and the general absence of any shelves at corresponding altitudes in glens on the opposite watershed, like that of the Spey, and in valleys where the waters flow eastward, are difficulties attending the marine theory which have never yet been got over. Mr. T.F. Jamieson, before cited, has, during a late visit to Lochaber, in 1861, observed many facts highly confirmatory of the hypothesis of glacier-lakes which, as I have already stated, was originally advanced by Mr. Agassiz. In the first place, he found much superficial scoring and polishing of rocks, and accumulation of boulders at those points where signs of glacial action ought to appear, if ice had once dammed up the waters of the glens in which the "roads" occur. Ben Nevis may have sent down its glaciers from the south, and Glen Arkaig from the north, for the mountains at the head of the last-mentioned glen are 3000 feet high, and may, together with other tributary glens, have helped to choke up the great Caledonian valley with ice, so as to block up for a time the mouths of the Spean, Roy, and Gluoy. The temporary conversion of these glens into glacier-lakes is the more conceivable, because the hills at their upper ends not being lofty nor of great extent, they may not have been filled with ice at a time when great glaciers were generated in other adjoining and much higher regions.

Secondly. The shelves, says Mr. Jamieson, are more precisely defined and unbroken than any of the raised beaches or acknowledged ancient coast-lines visible on the west of Scotland, as in Argyllshire, for example.

Thirdly. At the level of the lower shelf in Glen Roy, at points where torrents now cut channels through the shelf as they descend the hill-side, there are small delta-like extensions of the shelf, perfectly preserved, as if the materials, whether fine or coarse, had originally settled there in a placid lake, and had not been acted upon by tidal currents, mingling them with the sediment of other streams. These deltas are too entire to allow us to suppose that they have at any time since their origin been exposed to the waves of the sea.

Fourthly. The alluvium on the cols or watersheds, before alluded to, is such as would have been formed if the waters of the rivers had been made to flow east, or out of the upper ends of the supposed glacier-lakes, instead of escaping at the lower ends, in a westerly direction, where the great blockages of ice are assumed to have occurred.

In addition to these arguments of Mr. Jamieson, I may mention that in Switzerland, at present, no testacea live in the cold waters of glacier-lakes; so that the entire absence of fossil shells, whether marine or freshwater, in the stratified materials of each shelf, would be accounted for if the theory above mentioned be embraced.

When I examined "the parallel roads" in 1825, in company with Dr. Buckland, neither this glacier theory nor Mr. Darwin's suggestion of ancient sea-margins had been proposed, and I have never since revisited Lochaber. But I retain in my memory a vivid recollection of the scenery and physical features of the district, and I now consider the glacier-lake theory as affording by far the most satisfactory solution of this difficult problem. The objection to it, which until lately appeared to be the most formidable, and which led Mr. Robert Chambers in his "Sea Margins," to reject it entirely, was the difficulty of conceiving how the waters could be made to stand so high in Glen Roy as to allow the uppermost shelf to be formed. Grant a barrier of ice in the lower part of the glen of sufficient altitude to stop the waters from flowing westward, still, what prevented them from escaping over the col at the head of Glen Glaster? This col coincides exactly in level, as Mr. Milne Home first ascertained, with the second or middle shelf of Glen Roy. The difficulty here stated appears now to be removed by supposing that the higher lines or roads were formed before the lower ones, and when the quantity of ice was most in excess. We must imagine that at the time when the uppermost shelf of Glen Roy was forming in a shallow lake, the lower part of that glen was filled up with ice, and, according to Mr. Jamieson, a glacier from Loch Treig then protruded itself across Glen Spean and rested on the flank of the hill on the opposite side in such a manner as effectually to prevent any water from escaping over the Glen Glaster col. The proofs of such a glacier having actually existed at the point in question consist, he says, in numerous cross striae observable in the bottom of Glen Spean, and in the presence of moraine matter in considerable abundance on the flanks of the hill extending to heights above the Glen Glaster col. When the ice shrank into less dimensions the second shelf would be formed, having its level determined by the col last mentioned, Glen Spean in the meantime being filled with a glacier. Finally, the ice blockage common to glens Roy, Spean, and Laggan, which consisted probably of a glacier from Ben Nevis, gave rise to the lowest and most extensive lake, the waters of which escaped over the pass of Muckul or the col at the head of Loch Laggan, which, as Mr. Jamieson has now ascertained: agrees precisely in level with the lowest of all the shelves, and where there are unequivocal signs of a river having flowed out for a considerable period.

Dr. Hooker has described some parallel terraces, very analogous in their aspect to those of Glen Roy, as existing in the higher valleys of the Himalaya, of which his pencil has given us several graphic illustrations. He believes these Indian shelves to have originated on the borders of glacier-lakes, the barriers of which were usually formed by the ice and moraines of lateral or tributary glaciers, which descended into and crossed the main valley, as we have supposed in the case of Glen Roy; but others he ascribes to the terminal moraine of the principal glacier itself, which had retreated during a series of milder seasons, so as to leave an interval between the ice and the terminal moraine. This interspace caused by the melting of ice becomes filled with water and forms a lake, the drainage of which usually takes place by percolation through the porous parts of the moraine, and not by a stream overflowing that barrier. Such a glacier-lake Dr. Hooker actually found in existence near the head of the Yangma valley in the Himalaya. It was moreover partially bounded by recently formed marginal terraces or parallel roads, implying changes of level in the barrier of ice and moraine matter.* (* Hooker, "Himalayan Journal" volume 1 page 242; 2 pages 119, 121, 166. I have also profited by the author's personal explanations.)

It has been sometimes objected to the hypothesis of glacier-lakes, as applied to the case of Glen Roy, that the shelves must have taken a very long period for their formation. Such a lapse of time, it is said, might be consistent with the theory of pauses or stationary periods in the rise of the land during an intermittent upward movement, but it is hardly compatible with the idea of so precarious and fluctuating a barrier as a mass of ice. But the reader will have seen that the permanency of level in such glacier-lakes has no necessary connection with minor changes in the height of the supposed dam of ice. If a glacier descending from higher mountains through a tributary glen enters the main valley in which there happens to be no glacier, the river is arrested in its course and a lake is formed. The dam may be constantly repaired and may vary in height several hundreds of feet without affecting the level of the lake, so long as the surplus waters escape over a col or parting ridge of rock. The height at which the waters remain stationary is determined solely by the elevation of the col, and not by the barrier of ice, provided the barrier is higher than the col.

But if we embrace the theory of glacier-lakes, we must be prepared to assume not only that the sea had nothing to do with the original formation of the "parallel roads," but that it has never, since the disappearance of the lakes, risen in any one of the glens up to the level of the lowest shelf, which is about 850 feet high; for in that case the remarkable persistency and integrity of the roads and deltas, before described, must have been impaired.

We have seen that 50 miles to the south of Lochaber, the glacier formations of Lanarkshire with marine shells of arctic character have been traced to the height of 524 feet. About 50 miles to the south-east in Perthshire are those stratified clays and sands, near Killiecrankie, which were once supposed to be of submarine origin, and which in that case would imply the former submergence of what is now dry land to the extent of 1550 feet, or several hundred feet beyond the highest of the parallel roads. Even granting that these laminated drifts may have had a different origin, as above suggested, there are still many facts connected with the distribution of erratics and the striation of rocks in Scotland which are not easily accounted for without supposing the country to have sunk, since the era of continental ice, to a greater depth than 525 feet, the highest point to which marine shells have yet been traced.

After what was said of the pressure and abrading power of a general crust of ice, like that now covering Greenland, it is almost superfluous to say that the parallel roads must have been of later date than such a state of things, for every trace of them must have been obliterated by the movement of such a mass of ice. It is no less clear that as no glacier-lakes can now exist in Greenland [Note 26], so there could have been none in Scotland, when the mountains were covered with one great crust of ice. It may, however, be contended that the parallel roads were produced when the general crust of ice first gave place to a period of separate glaciers, and that no period of deep submergence ever intervened in Lochaber after the time of the lakes. Even in that case, however, it is difficult not to suppose that the Glen Roy country participated in the downward movement which sank part of Lanarkshire 525 feet beneath the sea, subsequently to the first great glaciation of Scotland. Yet that amount of subsidence might have occurred, and even a more considerable one, without causing the sea to rise to the level of the lowest shelf, or to a height of 850 feet above the present sea-level.

This is a question on which I am not prepared at present to offer a decided opinion.

Whether the horizontality of the shelves or terrace-lines is really as perfect as has been generally assumed is a point which will require to be tested by a more accurate trigonometrical survey than has yet been made. The preservation of precisely the same level in the lowest line throughout the glens of Roy, Spean, and Laggan, for a distance of 20 miles east and west, and 10 or 12 miles north and south, would be very wonderful if ascertained with mathematical precision. Mr. Jamieson, after making in 1862 several measurements with a spirit-level, has been led to suspect a rise in the lowest shelf of one foot in a mile in a direction from west to east, or from the mouth of Glen Roy to a point 6 miles east of it in Glen Spean. To confirm such observations, and to determine whether a similar rate of rise continues eastward, as far as the pass of Muckul, would be most important.

On the whole, I conclude that the Glen Roy terrace-lines and those of some neighbouring valleys, were formed on the borders of glacier-lakes, in times long subsequent to the principal glaciation of Scotland. They may perhaps have been nearly as late, especially the lowest of the shelves, as that portion of the Pleistocene period in which Man co-existed in Europe with the mammoth.

CHAPTER 14.

CHRONOLOGICAL RELATIONS OF THE GLACIAL PERIOD AND THE EARLIEST SIGNS OF MAN'S APPEARANCE IN EUROPE—CONTINUED.

Signs of extinct Glaciers in Wales. Great Submergence of Wales during the Glacial Period proved by Marine Shells. Still greater Depression inferred from Stratified Drift. Scarcity of Organic Remains in Glacial Formations. Signs of extinct Glaciers in England. Ice Action in Ireland. Maps illustrating successive Revolutions in Physical Geography during the Pleistocene Period. Southernmost Extent of Erratics in England. Successive Periods of Junction and Separation of England, Ireland, and the Continent. Time required for these Changes. Probable Causes of the Upheaval and Subsidence of the Earth's Crust. Antiquity of Man considered in relation to the Age of the existing Fauna and Flora.

EXTINCT GLACIERS IN WALES.

The considerable amount of vertical movement in opposite directions, which was suggested in the last chapter, as affording the most probable explanation of the position of some of the stratified and fossiliferous drifts of Scotland, formed since the commencement of the glacial period, will appear less startling if it can be shown that independent observations lead us to infer that a geographical revolution of still greater magnitude accompanied the successive phases of glaciation through which the Welsh mountains have passed.

That Wales was once an independent centre of the dispersion of erratic blocks has long been acknowledged. Dr. Buckland published in 1842 his reasons for believing that the Snowdonian mountains in Caernarvonshire were formerly covered with glaciers, which radiated from the central heights through the seven principal valleys of that chain, where striae and flutings are seen on the polished rocks directed towards as many different points of the compass. He also described the "moraines" of the ancient glaciers, and the rounded masses of polished rock, called in Switzerland "roches moutonnees." His views respecting the old extinct glaciers of North Wales were subsequently confirmed by Mr. Darwin, who attributed the transport of many of the larger erratic blocks to floating ice. Much of the Welsh glacial drift had already been shown by Mr. Trimmer to have had a submarine origin, and Mr. Darwin maintained that when the land rose again to nearly its present height, glaciers filled the valleys, and "swept them clean of all the rubbish left by the sea."* (* "Philosophical Magazine" series 3 volume 21 page 180.)

Professor Ramsay, in a paper read to the Geological Society in 1851, and in a later work on the glaciation of North Wales, described three successive glacial periods, during the first of which the land was much higher than it now is, and the quantity of ice excessive; secondly, a period of submergence when the land was 2300 feet lower than at present, and when the higher mountain tops only stood out of the sea as a cluster of low islands, which nevertheless were covered with snow; and lastly, a third period when the marine boulder drift formed in the middle period was ploughed out of the larger valleys by a second set of glaciers, smaller than those of the first period. This last stage of glaciation may have coincided with that of the parallel roads of Glen Roy, spoken of in the last chapter. In Wales it was certainly preceded by submergence, and the rocks had been exposed to glacial polishing and friction before they sank.

Fortunately the evidence of the sojourn of the Welsh mountains beneath the waters of the sea is not deficient, as in Scotland, in that complete demonstration which the presence of marine shells affords. The late Mr. Trimmer discovered such shells on Moel Tryfan, in North Wales, in drift elevated more than 1300 feet above the level of the sea. It appears from his observations, and those of the late Edward Forbes, corroborated by others of Professor Ramsay and Mr. Prestwich, that about twelve species of shells, including Fusus bamfius, F. antiquus, Venus striatula (Forbes and Hanley), have been met with at heights of between 1000 and 1400 feet, in drift, reposing on a surface of rock which had been previously exposed to glacial friction and striation.* (* Ramsay, "Quarterly Journal of the Geological Society" volume 8 1852 page 372.) The shells, as a whole, are those of the glacial period, and not of the Norwich Crag. Two localities of these shells in Wales, in addition to that first pointed out by Mr. Trimmer, have since been observed by Professor Ramsay, who, however, is of opinion that the amount of submergence can by no means be limited to the extreme height to which the shells happen to have been traced; for drift of the same character as that of Moel Tryfan extends continuously to the height of 2300 feet. [Note 27.]

RARITY OF ORGANIC REMAINS IN GLACIAL FORMATIONS.

The general dearth of shells in such formations, below as well as above the level at which Mr. Trimmer first found them, deserves notice. Whether we can explain it or not, it is a negative character which seems to belong very generally to deposits formed in glacial seas. The porous nature of the strata, and the length of time during which they have been permeated by rain-water, may partly account, as we hinted in a former chapter, for the destruction of organic remains. But it is also possible that they were originally scarce, for we read of the waters of the sea being so freshened and chilled by the melting of ice-bergs in some Norwegian and Icelandic fjords, that the fish are driven away, and all the mollusca killed. The moraines of glaciers are always from the first devoid of shells, and if transported by ice-bergs to a distance, and deposited where the ice melts, may continue as barren of every indication of life as they were when they originated.

Nevertheless, it may be said, on the other hand, that herds of seals and walruses crowd the floating ice of Spitzbergen in latitude 80 degrees north, of which Mr. Lamont has recently given us a lively picture,* (* "Seasons with the Sea-Horses" 1861.) and huge whales fatten on myriads of pteropods in polar regions. It had been suggested that the bottom of the sea, at the era of extreme submergence in Scotland and Wales, was so deep as to reach the zero of animal life, which, in part of the Mediterranean (the Aegean, for example), the late Edward Forbes fixed, after a long series of dredgings, at 300 fathoms. But the shells of the glacial drift of Scotland and Wales, when they do occur, are not always those of deep seas; and, moreover, our faith in the uninhabitable state of the ocean at great depths has been rudely shaken, by the recent discovery of Captain McClintock and Dr. Wallich, of starfish in water more than a thousand fathoms deep (7560 feet!), midway between Greenland and Iceland. That these radiata were really dredged up from the bottom, and that they had been living and feeding there, appeared from the fact that their stomachs were full of Globigerina, of which foraminiferous creatures, both living and dead, the oozy bed of the ocean at that vast depth was found to be exclusively composed. [Note 28.]

Whatever may be the cause, the fact is certain, that over large areas in Scotland, Ireland, and Wales, I might add throughout the northern hemisphere on both sides of the Atlantic, the stratified drift of the glacial period is very commonly devoid of fossils, in spite of the occurrence here and there, at the height of 500, 700, and even 1400 feet, of marine shells. These, when met with, belong, with few exceptions, to known living species. I am therefore unable to agree with Mr. Kjerulf that the amount of former submergence can be measured by the extreme height at which shells happen to have been found.

GLACIAL FORMATIONS IN ENGLAND.

(FIGURE 38. DOME-SHAPED ROCKS, OR "ROCHES MOUTONEES," IN THE VALLEY OF THE ROTHAY, NEAR AMBLESIDE, FROM A DRAWING BY E. HULL, F.G.S.* (* "Edinburgh New Philosophical Journal" volume 11 Plate 1 page 31 1860.))

The mountains of Cumberland and Westmorland, and the English lake district, afford equally unequivocal vestiges of ice-action not only in the form of polished and grooved surfaces, but also of those rounded bosses before mentioned as being so abundant in the Alpine valleys of Switzerland, where glaciers exist, or have existed. Mr. Hall has lately published a faithful account of these phenomena, and has given a representation of some of the English "roches moutonnees," which precisely resemble hundreds of dome-shaped protuberances in North Wales, Sweden, and North America.* (* Hull, "Edinburgh New Philosophical Journal" July 1860. )

The marks of glaciation on the rocks, and the transportation of erratics from Cumberland to the eastward, have been traced by Professor Phillips over a large part of Yorkshire, extending to a height of 1500 feet above the sea; and similar northern drift has been observed in Lancashire, Cheshire, Derbyshire, Shropshire, Staffordshire, and Worcestershire. It is rare to find marine shells, except at heights of 200 or 300 feet; but a few instances of their occurrence have been noticed, especially of Turritella communis (a gregarious shell), far in the interior, at elevations of 500 feet, and even of 700 in Derbyshire, and some adjacent counties, as I learn from Mr. Binney and Mr. Prestwich.

Such instances are of no small theoretical interest, as enabling us to account for the scattering of large erratic blocks at equal or much greater elevations, over a large part of the northern and midland counties, such as could only have been conveyed to their present sites by floating ice. Of this nature, among others, is a remarkable angular block of syenitic greenstone, 4 1/2 feet by 4 feet square, and 2 feet thick, which Mr. Darwin describes as lying on the summit of Ashley Heath, in Staffordshire, 803 feet above the sea, resting on New Red Sandstone.* (* Ancient Glaciers of Caernarvonshire, "Philosophical Magazine" series 3, 21 page 180.)

SIGNS OF ICE-ACTION AND SUBMERGENCE IN IRELAND DURING THE GLACIAL PERIOD.

In Ireland we encounter the same difficulty as in Scotland in determining how much of the glaciation of the higher mountains should be referred to land glaciers, and how much to floating ice, during submergence. The signs of glacial action have been traced by Professor Jukes to elevations of 2500 feet in the Killarney district, and to great heights in other mountainous regions; but marine shells have rarely been met with higher than 600 feet above the sea, and that chiefly in gravel, clay, and sand in Wicklow and Wexford. They are so rare in the drift east of the Wicklow mountains, that an exception to the rule, lately observed at Ballymore Eustace, by Professor Jukes, is considered as a fact of no small geological interest. The wide extent of drift of the same character, spread over large areas in Ireland, shows that the whole island was, in some part of the glacial period, an archipelago, as represented in the maps, Figures 39 and 40.

Speaking of the Wexford drift, the late Professor E. Forbes states that Sir H. James found in it, together with many of the usual glacial shells, several species which are characteristic of the Crag; among others the reversed variety of Fusus antiquus, called F. contrarius, and the extinct species Nucula Cobboldiae, and Turritella incrassata. Perhaps a portion of this drift of the south of Ireland may belong to the close of the Pliocene period, and may be of a somewhat older date than the shells of the Clyde, alluded to in Chapter 13. They may also correspond still more nearly in age with the fauna of the uppermost strata of the Norwich Crag, occurring at Chillesford. [Note 29.]

The scarcity of mammalian remains in the Irish drift favours the theory of its marine origin. In the superficial deposits of the whole island, I have only met with three recorded examples of the mammoth, one in the south near Dungarvan, where the bones of Elephas primigenius, two species of bear (Ursus arctos and Ursus spelaeus ?), the reindeer, horse, etc., were found in a cave;* (* E. Brenan and Dr. Carte, Dublin 1859.) another in the centre of the island near Belturbet, in the county of Cavan.

Perhaps the conversion into land of the bed of the glacial sea, and the immigration into the newly upheaved region of the elephant, rhinoceros, and hippopotamus, which co-existed with the fabricators of the St. Acheul flint hatchets, were events which preceded in time the elevation of the Irish drift, and the union of that island with England. Ireland may have continued for a longer time in the state of an archipelago, and was therefore for a much shorter time inhabited by the large extinct Pleistocene pachyderms.

In one of the reports of the Geological Survey of Ireland, published in 1859, Professor Jukes, in explanation of sheet 184 of the maps, alludes to beds of sand and gravel, and signs of the polishing and furrowing of the rocks in the counties of Kerry and Killarney, as high as 2500 feet above the sea, and supposes (perhaps with good reason) that the land was depressed even to that extent. He observes that above that elevation (2500 feet) the rocks are rough, and not smoothed, as if by ice. Some of the drift was traced as high as 1500 feet, the highest hills there exceeding 3400 feet. Mr. Jukes, however, is by no means inclined to insist on submergence to the extent of 2500 feet, as he is aware that ice, like that now prevailing in Greenland, might explain most, if not all, the appearances of glaciation in the highest regions.

Although the course taken by the Irish erratics in general is such that their transportation seems to have been due to floating ice or coast-ice, yet some granite blocks have travelled from south to north, as recorded by Sir R. Griffiths, namely, those of the Ox Mountains in Sligo; a fact from which Mr. Jamieson infers that those mountains formed at one time a centre of dispersion. In the same part of Ireland, the general direction in which the boulders have travelled is everywhere from north-west to south-east, a course directly at right angles to the prevailing trend of the present mountain ridges.

MAPS ILLUSTRATING SUCCESSIVE REVOLUTIONS IN PHYSICAL GEOGRAPHY DURING THE PLEISTOCENE PERIOD.

(FIGURE 39. MAP OF THE BRITISH ISLES AND PART OF THE NORTH-WEST OF EUROPE, SHOWING THE GREAT AMOUNT OF SUPPOSED SUBMERGENCE OF LAND BENEATH THE SEA DURING PART OF THE GLACIAL PERIOD.

The submergence of Scotland is to the extent of 2000 feet, and of other parts of the British Isles, 1300. In the map, the dark shade expresses the land which alone remained above water. The area shaded by diagonal lines is that which cannot be shown to have been under water at the period of floating ice by the evidence of erratics, or by marine shells of northern species. How far the several parts of the submerged area were simultaneously or successively laid under water, in the course of the glacial period, cannot, in the present state of our knowledge, be determined.)

(FIGURE 40. MAP SHOWING WHAT PARTS OF THE BRITISH ISLANDS WOULD REMAIN ABOVE WATER AFTER A SUBSIDENCE OF THE AREA TO THE EXTENT OF 600 FEET.

The authorities to whom I am indebted for the information contained in this map are—for:

SCOTLAND: A. Geikie, Esquire, F.G.S., and T.F. Jamieson, Esquire, of Ellon, Aberdeenshire.

ENGLAND: For the counties of: Yorkshire, Lancashire, and Durham: Colonel Sir Henry James, R.E. Dorsetshire, Hampshire, and Isle of Wight: H.W. Bristow, Esquire. Gloucestershire, Somersetshire, and part of Devon: R. Etheridge, Esquire. Kent and Sussex: Frederick Drew, Esquire. Isle of Man: W. Whitaker, Esquire.

IRELAND: Reduced from a contour map constructed by Lieutenant Larcom, R.E., in 1837, for the Railway Commissioners.)

(FIGURE 41. MAP OF PART OF THE NORTH-WEST OF EUROPE, INCLUDING THE BRITISH ISLES, SHOWING THE EXTENT OF SEA WHICH WOULD BECOME LAND IF THERE WERE A GENERAL RISE OF THE AREA TO THE EXTENT OF 600 FEET.

The darker shade expresses what is now land, the lighter shade the space intervening between the present coastline and the 100 fathom line, which would be converted by such a movement into land. The original of this map will be found in Sir H. de la Beche's "Theoretical Researches" page 190, 1834, but several important corrections have been introduced into it from recently published Admiralty Surveys, especially: 1st. A deep channel passing from the North Sea into the entrance of the Baltic. 2nd. The more limited westerly extension of the West Coast of Ireland.)

The late Mr. Trimmer, before referred to, has endeavoured to assist our speculations as to the successive revolutions in physical geography, through which the British Islands have passed since the commencement of the glacial period, by four "sketch maps" as he termed them, in the first of which he gave an ideal restoration of the original Continental period, called by him the first elephantine period, or that of the forest of Cromer, before described. He was not aware that the prevailing elephant of that era (E. meridionalis) was distinct from the mammoth. At this era he conceived Ireland and England to have been united with each other and with France, but much of the area represented as land in the map, Figure 41, was supposed to be under water. His second map, of the great submergence of the glacial period, was not essentially different from our map, Figure 39. His third map expressed a period of partial re-elevation, when Ireland was reunited to Scotland and the north of England; but England still separated from France. This restoration appears to me to rest on insufficient data, being constructed to suit the supposed area over which the gigantic Irish deer, or Megaceros, migrated from east to west, also to explain an assumed submergence of the district called the Weald, in the south-east of England, which had remained land during the grand glacial submergence.

The fourth map is a return to nearly the same continental conditions as the first—Ireland, England, and the Continent being united. This he called the second elephantine period; and it would coincide very closely with that part of the Pleistocene era in which Man co-existed with the mammoth, and when, according to Mr. Trimmer's hypothesis previously indicated by Mr. Godwin-Austen, the Thames was a tributary of the Rhine.* (* Joshua Trimmer, "Quarterly Journal of the Geological Society" volume 9 1853, Plate 13, and Godwin-Austen, ibid. volume 7 1851 page 134 and Plate 7.)

These geographical speculations were indulged in ten years after Edward Forbes had published his bold generalisations on the geological changes which accompanied the successive establishment of the Scandinavian, Germanic, and other living floras and faunas in the British Islands, and, like the theories of his predecessor, were the results of much reflection on a vast body of geological facts. It is by repeated efforts of this kind, made by geologists who are prepared for the partial failure of some of their first attempts, that we shall ultimately arrive at a knowledge of the long series of geographical revolutions which have followed each other since the beginning of the Pleistocene period.

The map, Figure 39, will give some idea of the great extent of land which would be submerged, were we to infer, as many geologists have done, from the joint evidence of marine shells, erratics, glacial striae and stratified drift at great heights, that Scotland was, during part of the glacial period, 2000 feet below its present level, and other parts of the British Isles, 1300 feet. A subsidence to this amount can be demonstrated in the case of North Wales by marine shells. In the lake district of Cumberland, in Yorkshire, and in Ireland, we must depend on proofs derived from glacial striae and the transportation of erratics for so much of the supposed submergence as exceeds 600 feet. As to central England, or the country north of the Thames and Bristol Channel, marine shells of the glacial period sometimes reach as high as 600 and 700 feet, and erratics still higher, as we have seen above. But this region is of such moderate elevation above the sea, that it would be almost equally laid under water, were there a sinking of no more than 600 feet.

To make this last proposition clear, I have constructed, from numerous documents, many of them unpublished, the map, Figure 40, which shows how that small amount of subsidence would reduce the whole of the British Isles to an archipelago of very small islands, with the exception of parts of Scotland, and the north of England and Wales, where four islands of considerable dimensions would still remain.

The map does not indicate a state of things supposed to have prevailed at any one moment of the past, because the district south of the Thames and the Bristol Channel seems to have remained land during the whole of the glacial period, at a time when the northern area was under water. The map simply represents the effects of a downward movement of a hundred fathoms, or 600 English feet, assumed to be uniform over the whole of the British Isles. It shows the very different state of the physical geography of the area in question, when contrasted with the results of an opposite movement, or one of upheaval, to an equal amount, of which Sir Henry de la Beche had already given us a picture, in his excellent treatise called "Theoretical Researches."* (* Also repeated in De la Beche's "Geological Observer.") His map I have borrowed (Figure 41), after making some important corrections in it.

If we are surprised when looking at the first map, Figure 40, at the vast expanse of sea which so moderate a subsidence as 600 feet would cause, we shall probably be still more astonished to perceive, in Figure 41, that a rise of the same number of feet would unite all the British Isles, including the Hebrides, Orkneys, and Shetlands, with one another and the Continent, and lay dry the sea now separating Great Britain from Sweden and Denmark.

It appears from soundings made during various Admiralty surveys, that the gained land thus brought above the level of the sea, instead of presenting a system of hills and valleys corresponding with those usually characterising the interior of most of our island, would form a nearly level terrace, or gently inclined plane, sloping outwards like those terraces of denudation and deposition which I have elsewhere described as occurring on the coasts of Sicily and the Morea.* (* "Manual of Geology" page 74.)

It seems that, during former and perhaps repeated oscillations of level undergone by the British Isles, the sea has had time to cut back the cliffs for miles in many places, while in others the detritus derived from wasting cliffs drifted along the shores, together with the sediment brought down by rivers and swept by currents into submarine valleys, has exerted a levelling power, filling up such depressions as may have pre-existed. Owing to this twofold action few marked inequalities of level have been left on the sea-bottom, the "silver-pits" off the mouth of the Humber offering a rare exception to the general rule, and even there the narrow depression is less than 300 feet in depth.

Beyond the 100 fathom line, the submarine slope surrounding the British coast is so much steeper that a second elevation of equal amount (or of 600 feet) would add but slightly to the area of gained land; in other words, the 100 and 200 fathom lines run very near each other.* (* De la Beche, "Geological Researches" page 191. )

The naturalist would have been entitled to assume the former union, within the Pleistocene period, of all the British Isles with each other and with the Continent, as expressed in the map, Figure 41, even if there had been no geological facts in favour of such a junction. For in no other way would he be able to account for the identity of the fauna and flora found throughout these lands. Had they been separated ever since the Miocene period, like Madeira, Porto Santo, and the Desertas, constituting the small Madeiran Archipelago, we might have expected to discover a difference in the species of land-shells, not only when Ireland was compared to England, but when different islands of the Hebrides were contrasted one with another, and each of them with England. It would not, however, be necessary, in order to effect the complete fusion of the animals and plants which we witness, to assume that all parts of the area formed continuous land at one and the same moment of time, but merely that the several portions were so joined within the Pleistocene era as to allow the animals and plants to migrate freely in succession from one district to another.

SOUTHERNMOST EXTENT OF ERRATICS IN ENGLAND.

In reference to that portion of the south of England which is marked by diagonal lines in Figure 39, the theory of its having been an area of dry land during the period of great submergence and floating ice does not depend merely on negative evidence, such as the absence of the northern drift or boulder clay on its surface; but we have also, in favour of the same conclusion, the remarkable fact of the presence of erratic blocks on the southern coast of Sussex, implying the existence there of an ancient coast-line at a period when the cold must have been at its height.

These blocks are to be seen in greatest number at Pagham and Selsea, 15 miles south of Chichester, in latitude 50 degrees 40 minutes north.

They consist of fragments of granite, syenite and greenstone, as well as of Devonian and Silurian rocks, some of them of large size. I measured one of granite at Pagham, 27 feet in circumference. They are not of northern origin, but must have come from the coast of Normandy or Brittany, or from land which may once have existed to the south-west, in what is now the English Channel.

They were probably drifted into their present site by coast ice, and the yellow clay and gravel in which they are embedded are a littoral formation, as shown by the shells. Beneath the gravel containing these large erratics, is a blue mud in which skeletons of Elephas antiquus, and other mammalia, have been observed. Still lower occurs a sandy loam, from which Mr. R.G. Austen* (* "Quarterly Journal of the Geological Society" volume 13 1857 page 50.) has collected thirty-eight species of marine shells, all Recent, but forming an assemblage differing as a whole from that now inhabiting the English Channel. The presence among them of Lutraria rugosa and Pecten polymorphus, not known to range farther north in the actual seas than the coast of Portugal, indicates a somewhat warmer temperature at the time when they flourished. Subsequently, there must have been great cold when the Selsea erratics were drifted into their present position, and this cold doubtless coincided in time with a low temperature farther north. [Note 30.] These transported rocks of Sussex are somewhat older than a sea-beach with Recent marine shells which at Brighton is covered by Chalk rubble, called the "elephant-bed" which I cannot describe in this place, but I allude to it as one of many geological proofs of the former existence of a seashore in this region, and of ancient cliffs bounding the channel between France and England, all of older date than the close of the glacial period. [Note 31.]

In order to form a connected view of the most simple series of changes in physical geography which can possibly account for the phenomena of the glacial period, and the period of the establishment of the present provinces of animals and plants, the following geographical states of the British and adjoining areas may be enumerated.

First, a continental period, towards the close of which the forest of Cromer flourished: when the land was at least 500 feet above its present level, perhaps much higher, and its extent probably greater than that given in the map, Figure 41.

Secondly, a period of submergence, by which the land north of the Thames and Bristol Channel, and that of Ireland, was gradually reduced to such an archipelago as is pictured in map, Figure 40; and finally to such a general prevalence of sea as is seen in map, Figure 39. This was the period of great submergence and of floating ice, when the Scandinavian flora, which occupied the lower grounds during the first continental period, may have obtained exclusive possession of the only lands not covered with perpetual snow.

Thirdly, a second continental period when the bed of the glacial sea, with its marine shells and erratic blocks, was laid dry, and when the quantity of land equalled that of the first period, and therefore probably exceeded that represented in the map, Figure 41. During this period there were glaciers in the higher mountains of Scotland and Wales, and the Welsh glaciers, as we have seen, pushed before them and cleared out the marine drift with which some valleys had been filled during the period of submergence. The parallel roads of Glen Roy are referable to some part of the same era.

As a reason for presuming that the land which in map, Figure 41, is only represented as 600 feet above its present level, was during part of this period much higher, Professor Ramsay has suggested that, as the previous depression far exceeded 100 fathoms (amounting in Wales to 1400 feet, as shown by marine shells, and to 2300, by stratified drift), it is not improbable that the upward movement was on a corresponding scale.

In passing from the period of chief submergence to this second continental condition of things, we may conceive a gradual change first from that of Map 39 to Map 40, then from the latter phase to that of Map 41, and finally to still greater accessions of land. During this last period the passage of the Germanic flora into the British area took place, and the Scandinavian plants, together with northern insects, birds, and quadrupeds, retreated into the higher grounds.

Judging from the evidence at present before us, the first appearance of Man, when, together with the mammoth and woolly rhinoceros, or with the Elephas antiquus, Rhinoceros hemitoechus, and Hippopotamus major, he ranged freely from all parts of the Continent into the British area, took place during this second continental period.

Fourthly, the next and last change comprised the breaking up of the land of the British area once more into numerous islands, ending in the present geographical condition of things. There were probably many oscillations of level during this last conversion of continuous land into islands, and such movements in opposite directions would account for the occurrence of marine shells at moderate heights above the level of the sea, notwithstanding a general lowering of the land. To the close of this era belong the marine deposits of the Clyde and the Carses of the Tay and Forth, before alluded to.

In a memoir by Professor E. Forbes, before cited, he observes, that the land of passage by which the plants and animals migrated into Ireland consisted of the upraised marine drift which had previously formed the bottom of the glacial sea. Portions of this drift extend to the eastern shores of Wicklow and Wexford, others are found in the Isle of Man full of arctic shells, others on the British coast opposite Ireland. The freshwater marl, containing numerous skeletons of the great deer, or Megaceros, overlie in the Isle of Man that marine glacial drift. Professor Forbes also remarks that the subsequent disjunction of Ireland from England, or the formation of the Irish Channel, which is less than 400 feet in its greatest depth, preceded the opening of the Straits of Dover, or the final separation of England from the Continent. This he inferred from the present distribution of species both in the animal and vegetable kingdoms. Thus, for example, there are twice as many reptiles in Belgium as in England, and the number inhabiting England is twice that found in Ireland. Yet the Irish species are all common to England, and all the English to Belgium. It is therefore assumed that the migration of species westward having been the work of time, there was not sufficient lapse of ages to complete the fusion of the continental and British reptilian fauna, before France was separated from England and England from Ireland.

For the same reason there are also a great number of birds of short flight, and small quadrupeds, inhabiting England which do not cross to Ireland, the Irish Channel seeming to have arrested them in their westward course.* (* E. Forbes, Fauna and Flora of British Isles, "Memoir of the Geological Survey" volume 1 1846 page 344.)

The depth of the Irish Channel in the narrower parts is only 360 feet, and the English Channel between Dover and Calais less than 200, and rarely anywhere more than 300 feet; so that vertical movements of slight amount compared to some of those previously considered, with the aid of denuding operations or the waste of sea cliffs, and the scouring out of the channel, might in time effect the insulation of the lands above alluded to.

TIME REQUIRED FOR SUCCESSIVE CHANGES IN PHYSICAL GEOGRAPHY IN THE PLEISTOCENE PERIOD.

The time which it would require to bring about such changes of level, according to the average rate assumed in Chapter 3, however vast, will not be found to exceed that which would best explain the successive fluctuations in terrestrial temperature, the glaciation of solid rocks, the transportation of erratics above and below the sea-level, the height of arctic shells above the sea, and last, not least, the migration of the existing species of animals and plants into their actual stations, and the extinction of some conspicuous forms which flourished during the Pleistocene ages. When we duly consider all these changes which have taken place since the beginning of the glacial epoch, or since the forest of Cromer and the Elephas meridionalis flourished, we shall find that the phenomena become more and more intelligible in proportion to the slowness of the rate of elevation and depression which we assume.

The submergence of Wales to the extent of 1400 feet, as proved by glacial shells, would require 56,000 years, at the rate of 2 1/2 feet per century; but taking Professor Ramsay's estimate of 800 feet more, that depression being implied by the position of some of the stratified drift, we must demand an additional period of 32,000 years, amounting in all to 88,000; and the same time would be required for the re-elevation of the tract to its present height. But if the land rose in the second continental period as much as 600 feet above its present level, as in Figure 41, this 600 feet, first of rising and then of sinking, would require 48,000 years more; the whole of the grand oscillation, comprising the submergence and re-emergence, having taken about 224,000 years for its completion; and this, even if there were no pause or stationary period, when the downward movement ceased, and before it was converted into an upward one.

I am aware that it may be objected that the average rate here proposed is a purely arbitrary and conjectural one, because, at the North Cape, it is supposed that there has been a rise of about 5 feet in a century, and at Spitsbergen, according to Mr. Lamont, a still faster upheaval during the last 400 years.* (* "Seasons with the Sea-Horses" page 202.) But, granting that in these and some exceptional cases (none of them as yet very well established) the rising or sinking has, for a time, been accelerated, I do not believe the average rate of motion to exceed that above proposed. Mr. Darwin, I find, considers that such a mean rate of upheaval would be as high as we could assume for the west coast of South America, where we have more evidence of sudden changes of level than anywhere else. He has not, however, attempted to estimate the probable rate of secular elevation in that or any other region.

Little progress has yet been made in divining the most probable causes of these great movements of the earth's crust; yet what little we know of the state of the interior leads us to expect that the gradual expansion or contraction of large portions of the solid crust may be the result of fluctuations in temperature, with which the existence of hundreds of active and thousands of extinct volcanoes is probably connected.

It is ascertained that solid rocks, such as granite and sandstone, expand and contract annually, even under such a moderate range of temperature as that of a Canadian winter and summer. If the heat should go on increasing through a thickness, say only of 10 miles of the earth's crust, the gradual upheaval of the incumbent mass may amount to many hundreds of feet; and the elevation may be carried still farther, by the complete fusion of part of the inferior rocks.

According to the experiments of Deville, the contraction of granite, in passing from a melted, or as some would say its plastic condition, to a solid state, must be more than 10 per cent.* (* "Bull. Societe Geologique France" 2nd series volume 4 page 1312.) So that we have at our command a source of depression on a grand scale, at every period when granitic rocks have originated in the interior of the earth's crust. All mineralogists are agreed that the passage of voluminous masses, from a liquid or pasty to a solid and crystalline state, must be an extremely slow process. It may often happen that, in the same series of superimposed rocks, some are expanding while still solid or while partially melting, while others are at the same time crystallising and contracting; so that the alterations of level at the surface may be the result of complicated and often of conflicting agencies. The more gradually we conceive such changes to take place, the more comprehensible they become in the eyes of the chemist and natural philosopher who speculates on the changes of the earth's interior; and the more fertile are they in the hands of the geologist in accounting for revolutions on the habitable surface.