CHAPTER XXVII.

PATH OF THE ICE-SHEET IN NORTH-WESTERN EUROPE AND ITS RELATIONS TO THE BOULDER CLAY OF CAITHNESS.[250]

Character of Caithness Boulder Clay.—Theories of the Origin of the Caithness Clay.—Mr. Jamieson’s Theory.—Mr. C. W. Peach’s Theory.—The proposed Theory.—Thickness of Scottish Ice-sheet.—Pentlands striated on their Summits.—Scandinavian Ice-sheet.—North Sea filled with Land-ice.—Great Baltic Glacier.—Jutland and Denmark crossed by Ice.—Sir R. Murchison’s Observations.—Orkney, Shetland, and Faroe Islands striated across.—Loess accounted for.—Professor Geikie’s Suggestion.—Professor Geikie and B. N. Peach’s Observations on East Coast of Caithness.—Evidence from Chalk Flints and Oolitic Fossils in Boulder Clay.

_The Nature of the Caithness Boulder Clay._—A considerable amount of difficulty has been felt by geologists in accounting for the origin of the boulder clay of Caithness. It is an unstratified clay, of a deep grey or slaty colour, resembling much that of the Caithness flags on which it rests. It is thus described by Mr. Jamieson (Quart. Jour. Geol. Soc., vol. xxii., p. 261):—

“The glacial drift of Caithness is particularly interesting as an example of a boulder clay which in its mode of accumulation and ice-scratched _débris_ very much resembles that unstratified stony mud which occurs underneath glaciers—the ‘_moraine profonde_,’ as some call it.

“The appearance of the drift along the Haster Burn, and in many other places in Caithness, is in fact precisely the same as that of the old boulder clay of the rest of Scotland, except that it is charged with remains of sea-shells and other marine organisms.

“If want of stratification, hardness of texture, and abundance of well-glaciated stones and boulders are to be the tests for what we call genuine boulder clay, then much of the Caithness drift will stand the ordeal.”

So far, therefore, as the mere appearance of the drift is concerned, it would at once be pronounced to be true Lower Till, the product of land-ice. But there are two circumstances connected with it which have been generally regarded as fatal to this conclusion.

(1) The striæ on the rocks show that the ice which formed the clay must have come from the sea, and not from the interior of the country; for their direction is almost at right angles to what it would have been had the ice come from the interior. Over the whole district, the direction of the grooves and scratches, not only of the rocks but even of the stones in the clay, is pretty nearly N.W. and S.E. “When examining the sections along the Haster Burn,” says Mr. Jamieson, “in company with Mr. Joseph Anderson, I remarked that the striæ on the imbedded fragments generally agreed in direction with those of the rocks beneath. The scratches on the boulders, as usual, run lengthways along the stones when they are of an elongated form; and the position of these stones, as they lie imbedded in the drift, is, as a rule, such that their longer axes point in the same direction as do the scratches on the solid rock beneath; showing that the same agency that scored the rocks also ground and pushed along the drift.”

Mr. C. W. Peach informs me that he seldom or never found a stone with two sets of striæ on it, a fact indicating, as Mr. Jamieson remarks, that the drift was produced by one great movement invariably in the same direction. Let it be borne in mind that the ice, which thus moved over Caithness in this invariable track, must either have come from the Atlantic to the N.W., or from the Moray Firth to the S.E.

(2) The boulder clay of Caithness is full of sea-shells and other marine remains. The shells are in a broken condition, and are interspersed like the stones through the entire mass of the clay. Mr. Jamieson states that he nowhere observed any instance of shells being found in an undisturbed condition, “nor could I hear,” he says, “of any such having been found; there seems to be no such thing as a bed of laminated silt with shells _in situ_.” The shell-fragments are scratched and ice-worn, the same as the stones found in the clay. Not only are the shells glaciated, but even the foraminifera, when seen through the microscope, have a rubbed and worn appearance. The shells have evidently been broken, striated, and pushed along by the ice at the time the boulder clay was being formed.

_Theories regarding the Origin of the Caithness Clay._—Mr. Jamieson, as we have seen, freely admits that the boulder clay of Caithness has the appearance of true land-ice till, but from the N.W. and S.E. direction of the striæ on the rocks, and the presence of sea-shells in the clay, he has come to the conclusion that the glaciation of Caithness has been effected by floating ice at a time when the district was submerged. I have always felt convinced that Mr. Jamieson had not hit upon the true explanation of the phenomena.

(1) It is physically impossible that any deposit formed by icebergs could be wholly unstratified. Suppose a mass of the materials which would form boulder clay is dropped into the sea from, say an iceberg, the heavier parts, such as stones, will reach the bottom first. Then will follow lighter materials, such as sand, then clay, and last of all the mud will settle down over the whole in fine layers. The different masses dropped from the various icebergs, will, no doubt, lie in confusion one over the other, but each separate mass will show signs of stratification. A good deal of boulder clay evidently has been formed in the sea, but if the clay be unstratified, it must have been formed under glaciers moving along the sea-bottom as on dry ground. Whether _unstratified_ boulder clay may happen to be formed under water or on dry land, it must in either case be the product of land-ice.[251] Those who imagine that materials, differing in specific gravity like those which compose boulder clay, dropped into water, can settle down without assuming the stratified form, should make the experiment, and they would soon satisfy themselves that the thing is physically impossible. The notion that unstratified boulder clay could be formed by deposits from floating ice, is not only erroneous, but positively pernicious, for it tends to lead those who entertain it astray in regard to the whole question of the origin of drift.

(2) It is also physically impossible that ice-markings, such as those everywhere found on the rocky face of the district, and on the pebbles and shells imbedded in the clay, could have been effected by any other agency than that of land-ice. I need not here enter into any discussion on this point, as this has been done at considerable length in another place.[252] In the present case, however, it is unnecessary, because if it can be shown that all the facts are accounted for in the most natural manner by the theory of land-ice, no one will contend for the floating-ice theory; for it is admitted that, with the exception of the direction of the striæ and the presence of the shells, all the facts agree better with the land-ice than with the floating-ice theory.

My first impression on the subject was that the glaciation of Caithness had been effected by the polar ice-cap, which, during the severer part of the glacial epoch, must have extended down to at least the latitude of the north of Scotland.

On a former occasion (see the _Reader_ for 14th October, 1865) it was shown that all the northern seas, owing to their shallowness, must, at that period, have been blocked up with solid ice, which displaced the water and moved along the sea-bottoms the same as on dry land. In fact, the northern seas, including the German Ocean, being filled at the time with glacier-ice, might be regarded as dry land. Ice of this sort, moving along the bed of the German Ocean or North Sea, and over Caithness, could not fail to push before it the shells and other animal remains lying on the sea-bottom, and to mix them up with the clay which now remains upon the land as evidence of its progress.

About two years ago I had a conversation with Mr. C. W. Peach on the subject. This gentleman, as is well known, has long been familiar with the boulder clay of Caithness. He felt convinced that the clay of that country is the true Lower Till, and not a more recent deposit, as Mr. Jamieson supposes. He expressed to me his opinion that the glaciation of Caithness had been effected by masses of land-ice crossing the Moray Firth from the mountain ranges to the south-east, and passing over Caithness in its course. The difficulty which seems to beset this theory is, that a glacier entering the Firth would not leave it and ascend over the Caithness coast. It would take the path of least resistance and move into the North Sea, where it would find a free passage into deeper water. Mr. Peach’s theory is, however, an important step in the right direction. It is a part of the truth, but I believe not the whole truth. The following is submitted as a solution of the question.

_The Proposed Theory._—It may now be regarded as an established fact that, during the severer part of the glacial period, Scotland was covered with one continuous mantle of ice, so thick as to bury under it the Ochil, Sidlaw, Pentland, Campsie, and other moderately high mountain ranges. For example, Mr. J. Geikie and Mr. B. N. Peach found that the great masses of the ice from the North-west Highlands, came straight over the Ochils of Perthshire and the Lomonds of Fife. In fact, these mountain ridges were not sufficiently high to deflect the icy stream either to the right hand or to the left; and the flattened and rounded tops of the Campsie, Pentland, and Lammermoor ranges bear ample testimony to the denuding power of ice.

Further, to quote from Mr. Jamieson, “the detached mountain of Schehallion in Perthshire, 3,500 feet high, is marked near the top as well as on its flanks, and this not by ice flowing down the sides of the hill itself, but by ice pressing over it from the north. On the top of another isolated hill, called Morven, about 3,000 feet high, and situated a few miles to the north of the village of Ballater, in the county of Aberdeen, I found granite boulders unlike the rock of the hill, and apparently derived from the mountains to the west. Again, on the highest watersheds of the Ochils, at altitudes of about 2,000 feet, I found this summer (1864) pieces of mica schist full of garnets, which seem to have come from the Grampian Hills to the north-west, showing that the transporting agent had overflowed even the highest parts of the Ochil ridge. And on the West Lomonds, in Fifeshire, at Clattering-well Quarry, 1,450 feet high, I found ice-worn pebbles of Red Sandstone and porphyry in the _débris_ covering the Carboniferous Limestone of the top of the Bishop Hill. Facts like these meet us everywhere. Thus on the Perthshire Hills, between Blair Athol and Dunkeld, I found ice-worn surfaces of rocks on the tops of hills, at elevations of 2,200 feet, as if caused by ice pressing over them from the north-west, and transporting boulders at even greater heights.”[253]

Facts still more important, however, in their bearing on the question before us were observed on the Pentland range by Mr. Bennie and myself during the summer of 1870. On ascending Allermuir, one of the hills forming the northern termination of the Pentland range, we were not a little surprised to find its summit ice-worn and striated. The top of the hill is composed of a compact porphyritic felstone, which is very much broken up; but wherever any remains of the original surface could be seen, it was found to be polished and striated in a most decided manner. These striæ are all in one uniform direction, nearly east and west; and on minutely examining them with a lens we had no difficulty whatever in determining that the ice which effected them came from the west and not from the east, a fact which clearly shows that they must have been made at the time when, as is well known, the entire Midland valley was filled with ice, coming from the North-west Highlands. On the summit of the hill we also found patches of boulder clay in hollow basins of the rock. At one spot it was upwards of a foot in depth, and rested on the ice-polished surface. The clay was somewhat loose and sandy, as might be expected of a layer so thin, exposed to rain, frost, and snow, during the long course of ages which must have elapsed since it was deposited there. Of 100 pebbles collected from the clay, just as they turned up, every one, with the exception of three or four composed of hard quartz, presented a flattened and ice-worn surface; and forty-four were distinctly striated: in short, every stone which was capable of receiving and retaining scratches was striated. A number of these stones must have come from the Highlands to the north-west.[254]

The height of Allermuir is 1,617 feet, and, from its position, it is impossible that the ice could have gone over its summit, unless the entire Midland valley, at this place, had been filled with ice to the depth of more than 1,600 feet. The hill is situated about four or five miles to the south of Edinburgh, and forms, as has already been stated, the northern termination of the Pentland range. Immediately to the north lies the broad valley of the Firth of Forth, more than twelve miles across, offering a most free and unobstructed outlet for the great mass of ice coming along the Midland valley from the west. Now, when we reflect how easily ice can accommodate itself to the inequalities of the channel along which it moves, how it can turn to the right hand or to the left, so as to find for itself the path of least resistance, it becomes obvious that the ice never would have gone over Allermuir, unless not only the Midland valley at this point, but also the whole surrounding country had been covered with one continuous mass of ice to a depth of more than 1,600 feet. But it must not be supposed that the height of Allermuir represents the thickness of the ice; for on ascending Scald Law, a hill four miles to the south-west of Allermuir, and the highest of the Pentland range, we found, in the _débris_ covering its summit, hundreds of transported stones of all sizes, from one to eighteen inches in diameter. We also dug up a Greenstone boulder about eighteen inches in diameter, which was finely polished and striated. As the height of this hill is 1,898 feet, the mass of ice covering the surrounding country must have been at least 1,900 feet deep. But this is not all. Directly to the north of the Pentlands, in a line nearly parallel with the east coast, and at right angles to the path of ice from the interior, there is not, with the exception of the solitary peak of East Lomond, and a low hill or two of the Sidlaw range, an eminence worthy of the name of a hill nearer than the Grampians in the north of Forfarshire, distant upwards of sixty miles. This broad plain, extending from almost the Southern to the Northern Highlands, was the great channel through which the ice of the interior of Scotland found an outlet into the North Sea. If the depth of the ice in the Firth of Forth, which forms the southern side of this broad hollow, was at least 1,900 feet, it is not at all probable that its depth in the northern side, formed by the Valley of Strathmore and the Firth of Tay, which lay more directly in the path of the ice from the North Highlands, could have been less. Here we have one vast glacier, more than sixty miles broad and 1,900 feet thick, coming from the interior of the country.

It is, therefore, evident that the great mass of ice entering the North Sea to the east of Scotland, especially about the Firths of Forth and Tay, could not have been less, and was probably much more, than from 1,000 to 2,000 feet in thickness. The grand question now to be considered is, What became of the huge sheet of ice after it entered the North Sea? Did it break up and float away as icebergs? This appears to have been hitherto taken for granted; but the shallowness of the North Sea shows such a process to have been utterly impossible. The depth of the sea in the English Channel is only about twenty fathoms, and although it gradually increases to about forty fathoms at the Moray Firth, yet we must go to the north and west of the Orkney and Shetland Islands ere we reach the 100 fathom line. Thus the average depth of the entire North Sea is not over forty fathoms, which is even insufficient to float an iceberg 300 feet thick.

No doubt the North Sea, for two reasons, is now much shallower than it was during the period in question. (1.) There would, at the time of the great extension of the ice on the northern hemisphere, be a considerable submergence, resulting from the displacement of the earth’s centre of gravity.[255] (2.) The sea-bed is now probably filled up to a larger extent with drift deposits than it was at the ice period. But, after making the most extravagant allowance for the additional depth gained on this account, still there could not possibly have been water sufficiently deep to float a glacier of 1,000 or 2,000 feet in thickness. Indeed, the North Sea would have required to be nearly ten times deeper than it is at present to have floated the ice of the glacial period. We may, therefore, conclude with the most perfect certainty that the ice-sheet of Scotland could not possibly have broken up into icebergs in such a channel, but must have moved along on the bed of the sea in one unbroken mass, and must have found its way to the deep trough of the Atlantic, west of the Orkney and Shetland Islands, ere it broke up and floated away in the iceberg form.

It is hardly necessary to remark that the waters of the North Sea would have but little effect in melting the ice. A shallow sea like this, into which large masses of ice were entering, would be kept constantly about the freezing-point, and water of this temperature has but little melting power, for it takes 142 lbs. of water, at 33°, to melt one pound of ice. In fact, an icy sea tends rather to protect the ice entering it from being melted than otherwise. And besides, owing to fresh acquisitions of snow, the ice-sheet would be accumulating more rapidly upon its upper surface than it would be melting at its lower surface, supposing there were sea-water under that surface. The ice of Scotland during the glacial period must, of necessity, have found its way into warmer water than that of the North Sea before it could have been melted. But this it could not do without reaching the Atlantic, and in getting there it would have to pass round by the Orkney Islands, along the bed of the North Sea, as land-ice.

This will explain how the Orkney Islands may have been glaciated by land-ice; but it does not, however, explain how Caithness should have been glaciated by that means. These islands lay in the very track of the ice on its way to the Atlantic, and could hardly escape being overridden; but Caithness lay considerably to the left of the path which we should expect the ice to have taken. The ice would not leave its channel, turn to the left, and ascend upon Caithness, unless it were forced to do so. What, then, compelled the ice to pass over Caithness?

_Path of the Scandinavian Ice._—We must consider that the ice from Scotland and England was but a fraction of that which entered the North Sea. The greater part of the ice of Scandinavia must have gone into this sea, and if the ice of our island could not find water sufficiently deep in which to float, far less would the much thicker ice of Scandinavia do so. The Scandinavian ice, before it could break up, would thus, like the Scottish ice, have to cross the bed of the North Sea and pass into the Atlantic. It could not pass to the north, or to the north-west, for the ocean in these directions would be blocked up by the polar ice. It is true that along the southern shore of Norway there extends a comparatively deep trough of from one to two hundred fathoms. But this is evidently not deep enough to have floated the Scandinavian ice-sheet; and even supposing it had been sufficiently deep, the floating ice must have found its way to the Atlantic, and this it could not have done without passing along the coast. Now, its passage would not only be obstructed by the mass of ice continually protruding into the sea directly at right angles to its course, but it would be met by the still more enormous masses of ice coming off the entire Norwegian coast-line. And, besides this, the ice entering the Arctic Ocean from Lapland and the northern parts of Siberia, except the very small portion which might find an outlet into the Pacific through Behring’s Straits, would have to pass along the Scandinavian coast in its way to the Atlantic. No matter, then, what the depth of this trough may have been, if the ice from the land, after entering it, could not make its escape, it would continue to accumulate till the trough became blocked up; and after this, the great mass from the land would move forward as though the trough had no existence. Thus, the only path for the ice would be by the Orkney and Shetland Islands. Its more direct and natural path would, no doubt, be to the south-west, in the direction of our shores; and in all probability, had Scotland been a low flat island, instead of being a high and mountainous one, the ice would have passed completely over it. But its mountainous character, and the enormous masses of ice at the time proceeding from its interior, would effectually prevent this, so that the ice of Scandinavia would be compelled to move round by the Orkney Islands. Consequently, these two huge masses of moving ice—the one from Scotland and the much greater one from Scandinavia—would meet in the North Sea, probably not far from our shores, and would move, as represented in the diagram, side by side northwards into the Atlantic as one gigantic glacier.

Nor can this be regarded as an anomalous state of things; for in Greenland and the antarctic continent the ice does not break up into icebergs on reaching the sea, but moves along the sea-bottom in a continuous mass until it reaches water sufficiently deep to float it. It is quite possible that the ice at the present day may nowhere traverse a distance of three or four hundred miles of sea-bottom, but this is wholly owing to the fact that it finds water sufficiently deep to float it before having travelled so far. Were Baffin’s Bay and Davis’s Straits, for example, as shallow as the North Sea, the ice of Greenland would not break up into icebergs in these seas, but cross in one continuous mass to and over the American continent.

The median line of the Scandinavian and Scottish ice-sheets would be situated not far from the east coast of Scotland. The Scandinavian ice would press up as near to our coast as the resistance of the ice from this side permitted. The enormous mass of ice from Scotland, pressing out into the North Sea, would compel the Scandinavian ice to move round by the Orkneys, and would also keep it at some little distance from Scotland. Where, on the other hand, there was but little resistance offered by ice from the interior of this country (and this might be the case along many parts of the English coast), the Scandinavian ice might reach the shores, and even overrun the country for some distance inland.

We have hitherto confined our attention to the action of ice proceeding from Norway; but if we now consider what took place in Sweden and the Baltic, we shall find more conclusive proof of the downward pressure of Scandinavian ice on our own shores. The western half of Gothland is striated in the direction of N.E. and S.W., and that this has been effected by a huge mass of ice covering the country, and not by local glaciers, is apparent from the fact observed by Robert Chambers,[256] and officers of the Swedish Geological Survey, that the general direction of the groovings and striæ on the rocks bears little or no relation to the conformation of the surface, showing that the ice was of sufficient thickness to move straight forward, regardless of the inequalities of the ground.

At Gottenburg, on the shores of the Cattegat, and all around Lake Wener and Lake Wetter, the ice-markings are of the most remarkable character, indicating, in the most decided manner, that the ice came from the interior of the country to the north-east in one vast mass. All this mass of ice must have gone into the shallow Cattegat, a sea not sufficiently deep to float even an ordinary glacier. The ice coming off Gothland would therefore cross the Cattegat, and thence pass over Jutland into the North Sea. After entering the North Sea, it would be obliged to keep between our shores and the ice coming direct from the western side of Scandinavia.

But this is not all. A very large proportion of the Scandinavian ice would pass into the Gulf of Bothnia, where it could not possibly float. It would then move south into the Baltic as land-ice. After passing down the Baltic, a portion of the ice would probably move south into the flat plains in the north of Germany, but the greater portion would keep in the bed of the Baltic, and of course turn to the right round the south end of Gothland, and thence cross over Denmark into the North Sea. That this must have been the path of the ice is, I think, obvious from the observations of Murchison, Chambers, Hörbye, and other geologists. Sir Roderick Murchison found—though he does not attribute it to land-ice—that the Aland Islands, which lie between the Gulf of Bothnia and the Baltic, are all striated in a north and south direction.[257]

Upsala and Stockholm, a tract of flat country projecting for some distance into the Baltic, is also grooved and striated, not in the direction that would be effected by ice coming from the interior of Scandinavia, but north and south, in a direction parallel to what must have been the course of the ice moving down the Baltic.[258] This part of the country must have been striated by a mass of ice coming from the direction of the Gulf of Bothnia. And that this mass must have been great is apparent from the fact that Lake Malar, which crosses the country from east to west, at right angles to the path of the ice, does not seem to have had any influence in deflecting the icy stream. That the ice came from the north and not from the south is also evident from the fact that the northern sides of rocky eminences are polished, rounded, and ice-worn, while the southern sides are comparatively rough. The northern banks of Lake Malar, for example, which, of course, face the south, are rough, while the southern banks, which must have offered opposition to the advance of the ice, are smoothed and rounded in a most singular manner.

Again, that the ice, after passing down the Baltic, turned to the right along the southern end of Gothland, is shown by the direction of the striæ and ice-groovings observed on such islands as Gothland, Öland, and Bornholm. Sir R. Murchison found that the island of Gothland is grooved and striated in one uniform direction from N.E. to S.W. “These groovings,” says Sir Roderick, “so perfectly resemble the flutings and striæ produced in the Alps by the actual movement of glaciers, that neither M. Agassiz nor any one of his supporters could detect a difference.” He concludes, however, that the markings could not have been made by land-ice, because Gothland is not only a low, flat island in the middle of the Baltic, but is “at least 400 miles distant from any elevation to which the term of mountain can be applied.” This, of course, is conclusive against the hypothesis that Gothland and the other islands of the Baltic could have been glaciated by ordinary glaciers; but it is quite in harmony with the theory that the Gulf of Bothnia and the entire Baltic were filled with one continuous mass of land-ice, derived from the drainage of the greater part of Sweden, Lapland, and Finland. In fact, the whole glacial phenomena of Scandinavia are inexplicable on the hypothesis of local glaciers.

That the Baltic was completely filled by a mass of ice moving from the north is further evidenced by the fact that the mainland, not only at Upsala, but at several places along the coast of Gothland, is grooved and striated parallel to the shore, and often at right angles to the markings of the ice from the interior, showing that the present bed of the Baltic was not large enough to contain the icy stream. For example, along the shores between Kalmar and Karlskrona, as described by Sir Roderick Murchison and by M. Hörbye, the striations are parallel to the shore. Perhaps the slight obstruction offered by the island of Öland, situated so close to the shore, would deflect the edge of the stream at this point over on the land. The icy stream, after passing Karlskrona, bent round to the west along the present entrance to the Baltic, and again invaded the mainland, and crossed over the low headland of Christianstadt, and thence passed westward in the direction of Zealand.

This immense Baltic glacier would in all probability pass over Denmark, and enter the North Sea somewhere to the north of the River Elbe, and would then have to find an outlet to the Atlantic through the English Channel, or pass in between our eastern shores and the mass from Gothland and the north-western shores of Europe. The entire probable path of the ice may be seen by a reference to the accompanying chart (Plate V.) That the ice crossed over Denmark is evident from the fact that the surface of that country is strewn with _débris_ derived from the Scandinavian peninsula.

Taking all these various considerations into account, the conclusion is inevitable that the great masses of ice from Scotland would be obliged to turn abruptly to the north, as represented in the diagram, and pass round into the Atlantic in the direction of Caithness and the Orkney Islands.

If the foregoing be a fair representation of the state of matters, it is physically impossible that Caithness could have escaped being overridden by the land-ice of the North Sea. Caithness, as is well known, is not only a low, flat tract of land, little elevated above the sea-level, and consequently incapable of supporting large glaciers; but, in addition, it projects in the form of a headland across the very path of the ice. Unless Caithness could have protected itself by pushing into the sea glaciers of one or two thousand feet in thickness, it could not possibly have escaped the inroads of the ice of the North Sea. But Caithness itself could not have supported glaciers of this magnitude, neither could it have derived them from the adjoining mountainous regions of Sutherland, for the ice of this county found a more direct outlet than along the flat plains of Caithness.

The shells which the boulder clay of Caithness contains have thus evidently been pushed out of the bed of the North Sea by the land-ice, which formed the clay itself.

The fact that these shells are not so intensely arctic as those found in some other quarters of Scotland, is no evidence that the clay was not formed during the most severe part of the glacial epoch, for the shells did not live in the North Sea at the time that it was filled with land-ice. The shells must have belonged to a period prior to the invasion of the ice, and consequently before the cold had reached its greatest intensity. Neither is there any necessity for supposing the shells to be pre-glacial, for these shells may have belonged to an inter-glacial period. In so far as Scotland is concerned, it would be hazardous to conclude that a plant or an animal is either pre-glacial or post-glacial simply because it may happen not to be of an arctic or of a boreal type.

The same remarks which apply to Caithness apply to a certain extent to the headland at Fraserburgh. It, too, lay in the path of the ice, and from the direction of the striæ on the rocks, and the presence of shells in the clay, as described by Mr. Jamieson, it bears evidence also of having been overridden by the land-ice of the North Sea. In fact, we have, in the invasion of Caithness and the headland at Fraserburgh by the land-ice of the North Sea, a repetition of what we have seen took place at Upsala, Kalmar, Christianstadt, and other flat tracts along the sides of the Baltic.

The scarcity, or perhaps entire absence of Scandinavian boulders in the Caithness clay is not in any way unfavourable to the theory, for it would only be the left edge of the North Sea glacier that could possibly pass over Caithness; and this edge, as we have seen, was composed of the land-ice from Scotland. We might expect, however, to find Scandinavian blocks on the Shetland and Faroe Islands, for, as we shall presently see, there is pretty good evidence to prove that the Scandinavian ice passed over these islands.

_The Shetland and Faroe Islands glaciated by Land-ice._—It is also worthy of notice that the striæ on the rocks in the Orkney, Shetland, and Faroe Islands, all point in the direction of Scandinavia, and are what would be effected by land-ice moving in the paths indicated in the diagram. And it is a fact of some significance, that when we proceed north to Iceland, the striæ, according to the observations of Robert Chambers, seem to point towards North Greenland. Is it possible that the entire Atlantic, from Scandinavia to Greenland, was filled with land-ice? Astounding as this may at first appear, there are several considerations which render such a conclusion probable. The observations of Chambers, Peach, Hibbert, Allan, and others, show that the rocky face of the Shetland and Faroe Islands has been ground, polished, and striated in a most remarkable manner. That this could not have been done by ice belonging to the islands themselves is obvious, for these islands are much too small to have supported glaciers of any size, and the smallest of them is striated as well as the largest. Besides, the uniform direction of the striæ on the rocks shows that it must have been effected by ice passing over the islands. That the striations could not have been effected by floating icebergs at a time when the islands were submerged is, I think, equally obvious, from the fact that not only are the tops of the highest eminences ice-worn, but the entire surface down to the present sea-level is smoothed and striated; and these striations conform to all the irregularities of the surface. This last fact Professor Geikie has clearly shown is wholly irreconcilable with the floating-ice theory.[259] Mr. Peach[260] found vertical precipices in the Shetlands grooved and striated, and the same thing was observed by Mr. Thomas Allan on the Faroe Islands.[261] That the whole of these islands have been glaciated by a continuous sheet of ice passing over them was the impression left on the mind of Robert Chambers after visiting them.[262] This is the theory which alone explains all the facts. The only difficulty which besets it is the enormous thickness of the ice demanded by the theory. But this difficulty is very much diminished when we reflect that we have good evidence, from the thickness of icebergs which have been met with in the Southern Ocean,[263] that the ice moving off the antarctic continent must be in some places considerably over a mile in thickness. It is then not so surprising that the ice of the glacial epoch, coming off Greenland and Northern Europe, should not have been able to float in the North Atlantic.

_Why the Ice of Scotland was of such enormous Thickness._—The enormous thickness of the ice in Scotland, during the glacial epoch, has been a matter of no little surprise. It is remarkable how an island, not more than 100 miles across, should have been covered with a sheet of ice so thick as to bury mountain ranges more than 1,000 feet in height, situated almost at the sea-shore. But all our difficulties disappear when we reflect that the seas around Scotland, owing to their shallowness, were, during the glacial period, blocked up with solid ice. Scotland, Scandinavia, and the North Sea, would form one immense table-land of ice, from 1,000 to 2,000 feet above the sea-level. This table-land would terminate in the deep waters of the Atlantic by a perpendicular wall of ice, extending probably from the west of Ireland away in the direction of Iceland. From this barrier icebergs would be continually breaking off, rivalling in magnitude those which are now to be met with in the antarctic seas.

_The great Extension of the Loess accounted for._—An effect which would result from the blocking up of the North Sea with land-ice, would be that the waters of the Rhine, Elbe, and Thames would have to find an outlet into the Atlantic through the English Channel. Professor Geikie has suggested to me that if the Straits of Dover were not then open—quite a possible thing—or were they blocked up with land-ice, say by the great Baltic glacier crossing over from Denmark, the consequence would be that the waters of the Rhine and Elbe would be dammed back, and would inundate all the low-lying tracts of country to the south; and this might account for the extraordinary extension of the Loess in the basin of the Rhine, and in Belgium and the north of France.[264]

_Note on the Glaciation of Caithness._

I have very lately received a remarkable confirmation of the path of the Caithness ice in observations communicated to me by Professor Geikie and Mr. B. N. Peach. The latter geologist says, “Near the Ord of Caithness and on to Berriedale the striæ pass off the land and out to sea; but near Dunbeath, 6 miles north-east of Berriedale, they begin to creep up out of the sea on to the land and range from about 15° to 10° east of north. _Where the striæ pass out to the sea_ the boulder clay is made up of the materials from inland and contains no shells, but _immediately the striæ begin to creep up on to the land_ then shells begin to make their appearance; and there is a difference, moreover, in the colour of the clay, for in the former case it is red and incoherent, and in the latter hard and dark-coloured.” The accompanying chart (Plate VI.) shows the outline of the Caithness coast and the direction of the striæ as observed by Professor Geikie and Mr. Peach, and no demonstration could be more conclusive as to the path of the ice and the obstacles it met than these observations, supplemented and confirmed as they are by other recorded facts to which I shall presently allude. Had the ice-current as it entered the North Sea off the Sutherland coast met with no obstacle it would have ploughed its way outwards till it broke off in glaciers and floated away. But it is clear that the great press of Scandinavian ice and the smaller mass of land-ice from the Morayshire coast converging in the North Sea filled up its entire bed, and these, meeting the opposing current from the Sutherland coast, turned it back upon itself, and forced it over the north-east part of Caithness. The farther south on the Sutherland coast that the ice entered the sea the deeper would it be able to penetrate into the ocean-bed before it met an opposition sufficiently strong to turn its course, and the wider would be its sweep; but when we come to the Sutherland coast we reach a point where the land-ice—as, for example, near Dunbeath—is forced to bend round before it even reaches the sea-shore, as will be seen from the accompanying diagram.

We are led to the same conclusions regarding the path of the ice in the North Sea from the presence of oolitic fossils and chalk flints found likewise in the boulder clay of Caithness, for these, as we shall see, evidently must have come from the sea. At the meeting of the British Association, Edinburgh, 1850, Hugh Miller exhibited a collection of boreal shells with fragments of oolitic fossils, chalk, and chalk flints from the boulder clay of Caithness collected by Mr. Dick, of Thurso. My friend, Mr. C. W. Peach, found that the chalk flints in the boulder clay of Caithness become more abundant as we proceed northward, while the island of Stroma in the Pentland Firth he found to be completely strewn with them. This same observer found, also, in the Caithness clay stones belonging to the Oolitic and Lias formations, with their characteristic fossils, while ammonites, belemnites, fossil wood, &c., &c., were also found loose in the clay.[265] The explanation evidently is, that these remains were derived from an outcrop of oolitic and cretaceous beds in the North Sea. It is well known that the eastern coast of Sutherlandshire is fringed with a narrow strip of oolite, which passes under the sea, but to what distance is not yet ascertained. Outside the Oolitic formation the chalk beds in all probability crop out. It will be seen from a glance at the accompanying chart (Plate VI.) that the ice which passed over the north-eastern part of Caithness must have crossed the out-cropping chalk beds.

As has already been stated in the foregoing chapter, the headland of Fraserburgh, north-eastern corner of Aberdeenshire, bears evidence, both from the direction of the striæ and broken shells in the boulder clay, of having been overridden also by land-ice from the North Sea. This conclusion is strengthened by the fact that chalk flints and oolitic fossils have also been abundantly met with in the clay by Dr. Knight, Mr. James Christie, Mr. W. Ferguson, Mr. T. F. Jamieson, and others.