Jurassic.

Creta. ceous.

Basaltic

tiles, but which at the same time cannot be satisfactorily separated from similar underlying strata which contain Upper Old Red Sandstone fishes. There occur also below the Lias on some parts of the west coast unfossiliferous red sandstones, conglomerates, and breccias which may possibly belong to the same system. These rocks attain their greatest thickness at Gruinard Bay on the west coast of Ross, where they must be several hundred feet thick. On the cast side of the country, where so many fragments of the Secondary rocks occur as boulders in the glacial deposits. a large mass of strata was formerly exposed at Linksfield near Elgin containing fossils which appear to show it to belong to the Rhætic beds at the top of the Trias. But it was not in place, and was almost certainly a mass transported by icc. Rhætic strata no doubt exist in situ at no great distance under the North Sea.

The Jurassic system is well represented on both sides of the Highlands. Along the east coast of Sutherland good sections are exposed showing the succession of strata. Among these the Lower and Middle Lias can be identified by their fossils. The Lower Oolite is distinguished by the occurrence in it of some coal-seanis, one of which, 3 feet in thickness, has been worked at Brora. The Middle Oolite consists mostly of sandstones with bands of shale and limestones and includes fossils which indicate the English horizons from the Kellaways Rock up to the Coral Rag. The lower part of the Kimmeridge Clay is probably represented by sandstones and conglomerates, forming the highest beds of the series in Sutherland. On the west side of the Highlands Jurassic rocks are found in many detached areas from the Shiant Isles to the southern shores of Mull. Over much of this region they owe their preservation in great measure to the mass of lavas poured over them in Tertiary time. They have been uncovered, indeed only at a comparatively recent geological date. They comprise a consecutive series of deposits from the bottom of the Lias up to the Oxford Clay. The Lower Middle and Upper Lias consist chiefly of shales and shelly limestones, with some sandstones, well seen along the shores of Broadford Bay in Skye and in some of the adjacent islands. The Lower Oolites are made up of sandstones and shales with some limestones, and are overlaid by several hundred feet of an estuarine series of deposits consisting chiefly of thick white sandstones, below and above which lie shales and shelly limestones. These rocks form a prominent feature underneath the basalt terraces of the east side of Skye, Raasay, and Eigg. They form the highest members of the Jurassic series, representing probably some part of the Oxford clay. The next Secondary rocks (Cretaceous) succeed

then unconformably.

Rocks belonging to the Cretaceous system undoubtedly at one time covered considerable areas on both sides of the Highlands, but they have been entirely stripped off the eastern side, while on the western they have been reduced to a few fragmentary patches, which have no doubt survived because of the overlying sheets of basalt that have protected them. Some greenish sandstones containing recognizable and characteristic fossils are the equivalents of the Upper Greensand of the south of England. These rocks are found on the south and west coasts of Mull and on the west coast of Argyllshire. They are covered by white sandstones and these by white chalk and marly beds, which represent the Upper Chalk of England. Enormous numbers of flints and also less abundant fragments of chalk are found in glacial deposits bordering the Moray Firth. These transported relics show that the Chalk must once have been in place at no great distance, if indeed it did not actually occupy part of Aberdeenshire and the neighbouring counties.

Above the highest Secondary rocks on the west coast come plateaus. terraced plateaus of basalt, which spread out over wide areas in Skye, Eigg, Mull, and Morven, and form most of the smaller islets of the chain of the Inner Hebrides (GEOLOGY, vol. x. p. 362). These plateaus are composed of nearly horizontal sheets of basalt-columnar, amorphous, or amygdaloidal-which in Mull attain a thickness of more than 3000 feet. They are prolonged southwards into Antrim (Ireland), where similar basalts overlying Secondary strata cover a large territory. Occasional beds of tuff are intercalated among these lavas, and likewise seams of fine clay or shale which have preserved the remains of numerous land-plants. The presence of these fossils indicates that the eruptions were subaerial, and a comparison of them with those elsewhere found among older Tertiary strata shows that they probably belong to what is now called the Oligocene stage of the Tertiary series of formations, and therefore that the basalt eruptions took place in early Tertiary time. The volcanic episode to which these plateaus owe their origin was one of the most important in the geological history of Great Britain. It appears to have resembled in its main features those remarkable outpourings of basalt which have deluged so many thousand square miles of the western territories of the United States. The eruptions were connected with innumerable fissures up which the basalt rose and from numerous points on which it flowed out at the surface. These fissures with the basalt that solidified in them now form the vast assemblage of dykes which cross Scotland, the north of England, and the north of Ireland (GEOLOGY, vol. x. p. 312). That the

volcanic period was a prolonged one is shown by the great denudation of the plateaus before the last eruptions took place. In the Isle of Eigg, for example, the basalts had already been deeply eroded by, river-action and into the river-course a current of glassy lava (pitchstone) flowed. Denudation has continued active ever since, and now, owing to greater hardness and consequent power of resistance, the glassy lava stands up as the prominent and picturesque ridge of the Scuir, while the basalts which formerly rose high above it have been worn down into terraced declivities that slope away from it to the sea. A remarkable feature in the volcanic phenomena was the disruption of the basaltic plateaus by large bosses of gabbro and of various granitoid rocks. These intrusive masses now tower into conspicuous groups of hills,-the Coolins in Skye, the mountains of Rum and Mull, and the rugged heights of Ardnamurchan. Under the Post-tertiary division come the records of the Ice Age, Youngest when Scotland was buried under sheets of ice which ground down, forma striated, and polished the harder rocks over the whole country and tions. left behind them the widespread accumulations of clay, gravel, and sand known as glacial deposits. The nature of the evidence and the deductions drawn from it have been already stated (GEOLOGY, vol. x. pp. 365-368). The youngest geological formations are the raised beaches, river-terraces, lake-deposits, peat-mosses, and other accumulations, which are related to the present configuration of the country and contain remains of the plants and animals still living on its surface (GEOLOGY, vol. x. pp. 256, 290, 369).

PHYSICAL FEATURES.

The physical features of Scotland may be best realized by regarding the country as composed of three distinct belts of territory, differing from each other in their geological structure and consequently presenting striking contrasts in their scenery.

1. The Highlands, for convenience of description, are here re- The garded as embracing all that part of the country which lies west Highand north of a line drawn along the Firth of Clyde, and thence lands. diagonally in a north-easterly direction from the mouth of the river Clyde to the east coast at Stonehaven. Nearly the whole of this region is high ground, deeply trenched with valleys and penetrated by long arms of the sea. The only considerable arca of lowland lies in the north-eastern counties, embracing the eastern part of Aberdeenshire and the northern parts of Banff, Elgin, and Nairn. Along both sides of the Moray Firth a strip of lower land intervenes between the foot of the hills and the sea, while farther north the county of Caithness is one wide plain, which is prolonged into the Orkney Islands. Seen from beyond its southern margin, the arca of the Highlands presents a well-defined chain of hills, which risa abruptly from the plains of the Lowlands. This is best observed in Strathmore, but it is also conspicuous in the estuary of the Clyde,, where the low hills on the south contrast well with the broken line of rugged mountains to the north. From any of the islands of tho chain of the Inner Hebrides the Highlands along their western scafront rise as a vast rampart, indented by many winding fjords and rising up to a singularly uniform general level, which sinks hero and there and allows glimpses to be had of still higher summits in the interior. The northern margin is hardly less striking when looked at from the Moray Firth, or from the plains of Caithness o Orkney.

From a commanding summit in the interior the Highlands are Highseen to differ from a mountain chain such as the Alps, not merely land in their inferior clevation, but essentially in their configuration and moun structure. They are made up of a succession of more or less nearly taius. parallel confluent ridges, which have, on the whole, a trend from north-east to south-west. These ridges are separated by longitudinal valleys, and each of them is likewise furrowed by transverse valleys. The portions of ridge thus isolated rise into what are termed mountains. But all the loftier eminences in the Highlands aro only higher parts of ridges along which their geological structure is prolonged. It is singular to observe how the general average of level of the summits of the ridges is maintained. From some points of view a mountain may appear to tower above all the surrounding country, but, looked at from a sufficient distance to take in its environment, it may be found not to rise much above the general uniformity of elevation. There are no gigantic dominant masses that must obviously be due to some special terrestrial disturbance. A few apparent exceptions to this statement rise along the western seaboard of Sutherland, in Skye, and elsewhere, but an examination of their structure at once explains the reason of their prominence and confirms the rule.

The general surface of the Highlands is rugged. The rocks pro- Surface ject in innumerable bosses and crags, which roughen the sides and of Highs crests of the ridges. The forms and colours of these roughnesses lands. depend on the nature of the rock underneath. Where the latter is hard and jointed, weathering into large quadrangular blocks, the hills are more especially distinguished for the gnarled bossy character of their declivities, as may be seen in Ben Ledi and the chain of heights to the north-cast of it formed of massive grits and mica chists. Where, on the other hand, the rock decays into sniller débris, the hills are apt to assume smoother contours, as in

Valleys.

Clyde to Stonehaven, on the south side by the pastoral uplands that stretch from Girvan to Dunbar: The simplest conception of the general aspect and structure of this important part of the kingdom is obtained by regarding it as a long trough of younger rocks let down by parallel dislocations between the older masses of the high grounds to the south and north. The lowest of these younger rocks are the various sedimentary and volcanic members of the Old Red Sandstone. These are covered by the successive formations of the Carboniferous system. The total thickness of both these groups of rock cannot be less than 30,000 feet, and, as most of them hear evidence of having been deposited in shallow water, it is manifest that they could only have been accumulated during a prolonged period of depression. The question arises whether this depression affected only the area of the midland valley itself, or whether it extended also over the regions to the north and south. Materials do not yet exist for a definite answer to this question; but so far as the evidence now before us goes there is ground for the inference that, while the depression had its maximum along the line of the lowlands, it also involved some portion at least of the high grounds on either side. In other words, the Old Red Sandstone and Carboniferous rocks, though chiefly accumulated in the broad lowland valley, crept also over some part at least of the hills on either side, where a few outliers are left to tell of their former extension. The central Lowlands of Scotland are thus of great geological antiquity. During and since the deposition of the rocks that underlie them the tract has been the scene of repeated terrestrial disturbances. Long dislocations, running like the ridges of the Highlands and the southern uplands from south-west to north-east, have sharply defined its northern and southern margins. By other fractures and unequal movements of upheaval or depression portions of the older rocks have been brought up within the bounds of the younger, and areas of the younger have been enclosed by the older. On the whole, these terrestrial disturbances have followed the same prevalent north-easterly trend, and hence a general tendency may be observed among the main ridges and valleys to run in that direction. The chains of the Ochil, Sidlaw, Pentland, Renfrew, Campsie, and Fintry Hills, and the valleys of Strathmore, Firth of Tay, and the basin of Midlothian, may be cited as examples. But, undoubtedly, the dominant cause in the determination of the topographical prominences and depressions of the district has been the relative hardness and softness of the rocks. Almost the whole of the eminences in the Lowlands consist of hard igneous rocks, forming not only chains of hills like those just referred to and others in Ayrshire and Lanarkshire, but isolated crags and hills like those of Stirling Castle, Edinburgh Castle, and others conspicuous in the scenery of Fife and the Lothians.

Of the three chief valleys in the central Lowlands two, those of the Tay and the Forth, descend from the Highlands, and one, that of the Clyde, from the southern uplands. Though on the whole transverse, these depressions furnish another notable example of that independence of geological structure already mentioned.

We now proceed to consider the leading physical features of the country with especial reference to their distinctive aspects and their respective modes of origin. Though an eminently hilly country, Scotland is not dominated by any leading mountain chain on which all the other topographical features are dependent. Its leading features are not the monotonous ridges of the high grounds but the valleys that have been opened through them. If these valleys were filled up, the high grounds would once more become what they probably were at first, elevated plains or plateaus, with no strongly marked features,- -no eminences rising much above nor hollows sinking much below the general surface.

Valleys.-Even apart from any knowledge of their origin, the valleys of the country are thus seen to be its fundamental topographical element, and to deserve the first consideration in any attempt to describe and explain its physical features. The longitudinal valleys, which run in the same general direction as the ridges-that is, north-east and south-west-have had their trend defined by geological structure, such as a line of dislocation (the Great Glen), or the plications of the rocks (Lochs Ericht, Tay, and Awe; and most of the sea lochs of Argyllshire). The transverse valleys run north-west or south-east and are for the most part independent of geological structure. The valley of the Garry and Tay crosses the strike of all the Highland rocks, traverses the great fault on the Highland border, and finally breaks through the chain of the Ochil Hills at Perth. The valley of the Clyde crosses the strike of the Silurian plications in the southern uplands, the boundary fault, and the ridges of the Old Red Sandstone, and pursues its north-westerly course across the abundant and often powerful dislocations of the Carboniferous system. Origin of That valleys are essentially due to erosion and not to dislocation Scottish or subsidence of the earth's surface is a fact which has now been valleys. demonstrated by so overwhelming a mass of evidence from all parts of the globe that it may be accepted as one of the axioms of geology. The plications of the earth's crust which folded the rocks of the Highlands and southern uplands not improbably upraised above the sea a series of longitudinal ridges having a general north-easterly

direction. The earliest rain that fell upon these ridges would run off them, first in transverse watercourses down each short slope and then in longitudinal depressions wherever such had been formed during the terrestrial disturbance. Once chosen, the pathways of the streams would be gradually deepened and widened into valleys. Hence the valleys are of higher antiquity than the mountains that rise from them. The mountains in fact have emerged out of the original bulk of the land in proportion as the valleys have been excavated. The denudation would continue so long as the ground stood above the level of the sea; but there have been prolonged periods of depression, when the ground, instead of being eroded, lay below the sea-level and was buried sometimes under thousands of feet of accumulated sediment, which completely filled up and obliterated the previous drainage-lines. When the land reappeared a new and independent series of valleys would at once begin to be eroded; and the subsequent degradation of these overlying sediments might reveal portions of the older topography, as in the case of the Great Glen, Lauderdale, and other ancient valleys. But the new drainage-lines have usually little or no reference to the old ones. Determined by the inequalities of surface of the overlying mantle of sedimentary material, they would be wholly independent of the geological structure of the rocks lying below that mantle. Slowly sinking deeper and deeper into the land, they might eventually reach the older rocks, but they would keep in these the lines of valley that they had followed in the overlying deposits. In process of time the whole of these deposits might be denuded from the area. The valleys would then be seen running in utter disregard of the geological structure of the rocks around them, and there might even remain no trace of the younger formations on which they began and which guided their excavation. This is probably the explanation of the striking independence of geological structure exhibited by the Tweed and the Nith. Among the valleys of Scotland certain prevailing characteristics General have been recognized in the popular names bestowed upon them. characteristics. "Straths" are broad expanses of low ground between bounding hills usually traversed by one main stream and its tributaries,Strath Tay, Strath Spey, Strath Conon. The name, however, has also been applied to wide tracts of lowland which embrace portions of several valleys, but are defined by lines of heights on either side; the best example is afforded by Strathmore-the " great strath between the southern margin of the Highlands and the line of the Ochil and Sidlaw Hills. This long and wide depression,, though it looks like one great valley, strictly speaking, includes portions of the valleys of the Tay, Isla, North Esk, and South Esk, all of which cross it. Elsewhere in central Scotland such a wide depression is known as a "howe," as in the Howe of Fife between the Ochil and Lomond Hills. A "glen" is usually a narrower and steeper-sided valley than a strath, though the names have not always been applied with discrimination. Most of the Highland valleys are true glens. The hills rise rapidly on either side, sometimes in grassy slopes, sometimes in rocky bosses and precipitous cliffs, while the bottom is occupied by a flat platform of alluvium through which a stream meanders. Frequently the bottom of some part of the valley is occupied by a lake. In the south of Scotland the larger streams flow in wide open valleys called "dales," as in Clydesdale, Tweeddale, Teviotdale, Liddisdale, Eskdale, Nithsdale. The strips of alluvial land bordering a river are known as "haughs," and where in estuaries they expand into wide plains they are termed carses. The carses of the Forth extend seawards as far as Borrowstounness and consist chiefly of raised beaches. The Carse of Gowrie is the strip of low ground intervening between the Firth of Tay and the line of hills that stretches from Perth to Dundee.

66

"

"

River-gorges are characteristic features in many of the valleys of RiverScotland. In the Old Red Sandstone they are particularly promi- gorges. nent where that formation has lain in the pathway of the streams sweeping down from the Highlands. In the basin of the Moray Firth some fine examples may be seen on the Nairn and Findhorn, while on the west side of the Cromarty Firth some of the small streams descending from the high grounds of the east of Ross-shire have cut out defiles in the conglomerate, remarkable for their depth and narrowness. On the south side of the Highlands still more notable instances of true "cañons" in the Old Red Sandstone are to be seen where the Ericht, Isla, and North Esk enter that formation. The well-known gorge in which the Falls of Clyde are situated is the best example in the midland valley.1

"

Types of Mountain and Hill.-While the topography of the Flatcountry is essentially the result of prolonged denudation, we may topped reasonably infer that the oldest surfaces likely to be in any measure mounpreserved or indicated are portions of some of the platforms of tains. erosion which have successively been produced by the wearing away of the land down to the sea-level. Relics of these platforms seem to be recognizable both in the Highlands and among the southern uplands. Allusion has already been made to the remarkable flattopped moorlands which in the eastern Grampians reach heights 1 For the principal rivers, the Tay, Spey, Forth, Clyde, and Tweed, see the separate articles, and for the Dee (Aberdeen, Kirkcudbright), &c., see articles on the respective counties.

of 3000 to 4000 feet above the sea. Their most familiar example perhaps is the top of Lochnagar, where when the level of 3500 feet has been gained the traveller finds himself on a broad undulating moor, more than a mile and a half long, sloping gently southwards towards Glen Muick and terininating on the north at the edge of a range of granite precipices. The top of Ben Macdui stands upon nearly a square mile of moor exceeding 4000 feet in elevation. These mountains lie within granite areas; but not less striking examples may be found among the schists. The mountains at the head of Glen Esk and Glen Isla, for instance, sweep upward into a broad inoor some 3000 feet above the sea, the more prominent parts of which have received special names,-Driesh, Mayar, Tom Buidhe, Tolmount, Cairn na Glasha. It would hardly be an exaggeration to say that there is more level ground on the tops of these mountains than in areas of corresponding size in the valleys below. That these high plateaus are planes of erosion is shown by their independence of geological structure, the upturned edges of the vertical and contorted schists having been abruptly shorn off and the granite having been wasted and levelled along its exposed surface. They look like fragments of the original tableland of erosion out of which the present valley-systems of the Highlands have bee: carved. Among the southern uplands traces of a similar tableland of erosion are in many places to be detected. The top of Broad Law in Peeblesshire, for example, is a level moor comprising between 300 and 400 acres above the contour line of 2500 feet and lying upon the upturned edges of the greatly denuded Silurian grits and shales. An instructive example of the similar destruction of a much younger platform is to be found in the terraced plateaus of Skye, Eigg, Canna, Muck, Mull, and Morven, which are portions of what was probably originally a continuous plain of basalt. Though dating back only to older Tertiary time, this plain has been so deeply trenched by the forces of denudation that it has been reduced to mere scattered fragments. Thousands of feet of basalt have been worn away from many parts of its sur face; deep and wide valleys have been carved out of it; and so enormously has it been wasted that it has been almost entirely stripped from wide tracts which it formerly covered and where only scattered outliers remain to prove that it once existed.

It is a curious fact, to which allusion has already been made, that broad flat-topped mountains are chiefly to be found in the eastern parts of the country. Traced westwards these forms gradually give place to narrow ridges and crests. No contrast, for instance, can be greater than that between the wide elevated moors of the castern Grampians, and the crested ridges of western Inverness-shire and Argyllshire-Loch Hourn, Glen Nevis, Glencoe or that between the broad uplands of Peeblesshire and the precipitous heights of Galloway. No satisfactory reason for these contrasts can be found in geological structure alone. Perhaps the key to them is to be sought mainly in differences of rainfall. The western mountains, exposed to the fierce dash of the Atlantic rains, sustain the heaviest and most constant precipitation. Their sides are scamed with torrents which tear down the solid rock and sweep its detritus into the glens and sea lochs. The castern heights, on the other hand, experience a less rainfall and consequently a diminished rate of erosion. There is no reason to doubt that the present preponderance of rainfall in the west has persisted for an enormous luration of time.

Regarding the existing flat-topped heights among the eastern Grampians as representing what may have been the general character of the surface out of which the present Highlands have been carved, we can trace every step in the gradual obliteration of the tableland and in the formation of the most rugged and individualized forms of isolated mountain. In fact, in journeying westwards across the tops of the Highland mountains we pass, as it were, over successive stages in the history of the origin of Highland scenery. The oldest types of form lie on the east side and the newest on the west. From the larger fragments of the denuded tableland we advance to ridges with narrow tops, which pass by degrees into sharp rugged crests. The ridges, too, are more and inore trenched until they become groups of detached hills or mountains. In the progress of this erosion full scope has been afforded for the modification of form produced by variations in geological structure. Each ridge and mountain has been cut into its shape by denudation, but its actual outlines have been determined by the nature of the rocks and the manner in which they have yielded to decay. Every distinct variety of rock has impressed its own characters upon the landscapes in which it plays a part. Hence, amid the monotonous succession of ridge beyond ridge and valley after valley, consider able diversity of detail has resulted from the varying composition and grouping of the rocks.

The process by which the ancient tablelands of the country have been trenched into the present system of valleys and confluent ridges is most instructively displayed among the higher mountains, where crosion proceeds at an accelerated pace. The long "screes or talus-slopes at the foot of every crag and cliff bear witness to the continual waste of the mountain sides. The headwaters of a river cut into the slopes of the parent hill. Each valley is consequently

Usu

lengthened at the expense of the mountain from which it descends. Where a number of small torrents converge in a steep mountain recess, they cut out a crescent-shaped hollow or half-cauldron, which in the Scottish Highlands is known as a corry.' Whether Corries the convergent action of the streams has been the sole agency con- and cerned in the erosion of these striking concavities, or whether snow glens. and glacier-ice may have had a share in the task, is a question that cannot at present be satisfactorily solved. No feature in Highland scenery is more characteristic than the corries, and in none can the influence of geological structure be more instructively seen. ally the upper part of a corry is formed by a crescent of naked rock, from which long trails of débris descend to the bottom of the hollow. Every distinct variety of rock has its own type of corry, the peculiarities being marked both in the details of the upper cliffs and crags and in the amount, form, and colour of the screes. The Scottish corries have been occupied by glaciers. Hence their bottoms are generally well ice-worn or strewn over with moraine stuff, Not infrequently also a small tarn fills up the bottom, ponded back by a moraine. It is in these localities that we can best observe the last relics left by the retreat of the glaciers that once overspread the country. Among these high grounds also the gradual narrowing of ridges into sharp, narrow, knife-edged crests and the lowering of these into cols or passes can be admirably studied. Where two glens begin opposite to each other on the same ridge, their corries are gradually cut back until only a sharp crest separates them. This crest, attacked on each front and along the summit, is lowered with comparative rapidity, until in the end merely a low col or pass may separate the heads of the two glens. The various stages in this kind of demolition are best seen where the underlying rock is of granite or some similar material which possesses considerable toughness, while at the same time it is apt to be split and splintered by means of its numerous transverse joints. The granite mountains of Arran furnish excellent illustrations.

Where a rock yields with considerable uniformity in all directions Formato the attacks of the weather it is apt to assume conical forms in tion the progress of denudation. Sometimes this uniformity is attained of moun by a general disintegration of the rock into fine débris, which rolis tain down the slopes in long screes. In other cases it is secured by the cones. intersection of joints, whereby a rock, in itself hard and durable, is divided into small angular blocks, which are separated by the action of the elements and slide down the declivities. In many instances the beginning of the formation of a cone may be detected on ridges which have been deeply trenched by valleys. The smaller isolated portions, attacked on all sides, have broken up under the influence of the weather. Layer after layer has been stripped from their sides, and the flat or rounded top has been narrowed until it has now become the apex of a cone. The mountain Schiehallien (3547 feet) is a noble instance of a cone not yet freed from its parent ridge. Occasionally a ridge has been carved into a series of cones united at their bases, as in the chain of the Pentland Hills. A further stage in denudation brings us to isolated groups of cones completely separated from the rest of the rocks among which they once lay buried. Such groups may be carved out of a continuous band of rock which extends into the regions beyond. The Paps of Jura, for instance, rise out of a long belt of quartzite which stretches through the islands of Isla, Jura, and Scarba. In many cases, however, the groups point to the existence of some boss of rock of greater durability than those in the immediate neighbourhood, as in the Cuchullins and Red Hills of Skye and the group of granite cones of Ben Loyal, Sutherland. The most impressive form of soli. tary cone is that wherein after vast denudation a thick overlying formation has been reduced to a single outlier, such as Morven in Caithness and the two Ben Griams in Sutherland, and still more strikingly the pyramids of red sandstone on the western margin of Sutherland and Ross-shire. The horizontal stratification of some of these masses gives them a curiously architectural aspect, which is further increased by the effect of the numerous vertical joints by which the rock is cleft into buttresses and recesses along the fronts of the precipices and into pinnacles and finials along the summits. Solitary or grouped pyramids of red sandstone, rising to heights of between 3000 and 4000 feet above the sea, are mere remnants of a once continuous sheet of red sandstone that spread far and wide over the western Highlands.

In

Stratified rocks when they have not been much disturbed from Escarptheir original approximate horizontality weather into what are meuts. called "escarpments, "-lines of cliff or steep bank marking the edge or outcrop of harder bands which lie upon softer or more easily eroded layers. Such cliffs may run for many miles across a country, rising one above another into lofty terraced hills. Scotland the rocks have for the most part been so dislocated and disturbed as to prevent the formation of continuous escarpments, and this interesting form of rock-scenery is consequently almost entirely absent, except locally and for the most part on a comparatively small scale. The most extensive Scottish escarpments are found among the igneous rocks. Where lava has been piled up in successive nearly horizontal sheets, with occasional layers of tuff

or other softer rock between them, it offers conditions peculiarly favourable for the formation of escarpments. In the wide basalt plateaus of the Inner Hebrides these conditions have been manifested on a great scale. The Carboniferous lavas of the Campsie and Fintry Hills and of the south of Dumfriesshire and Roxburgh shire likewise rise in lines of bold escarpment.

Lakes. These important features in the landscapes of Scotland present the general characters of the water-basins so profusely scattered over the northem parts of Europe and North America. They may be classified in four groups, each of which has its own peculiar scenery and a distinct mode of origin-(1) glen lakes, (2) rock-tarns, (3) moraine-tarns, (4) lakes of the plains.

(1) Glen lakes are those which occupy portions of glens. They are depressions in the valleys, not due to mere local heaping up of detritus, but true rock-basins, often of great depth. Much discussion has arisen as to their mode of origin. They have been regarded as caused by special subsidence of their areas, open fissures of the ground, general depression of the central part of each mountain district from which they radiate, and by the erosive action of glacier ice. That they are not open fissures and cannot be explained by any general subsidence of a neighbouring region is now generally admitted. That glaciers have occupied the glens where these lakes exist and have worn down the rocks along the sides and bottom cannot be doubted, but whether the ice would be capable of eroding hollows so deep as many of these lakes is a question which has been answered with equal confidence affirmatively and negatively. On the other hand, to suppose that each of these hollows has been caused by a special local subsidence would involve a complex series of subterranean disturbances, for which some better evidence than the mere existence of the basins is required. Under any circumstances it is quite certain that the lakes must be of recent geological date. Any such basins belonging to the time of the plication of the crystalline schists would have been filled up and effaced long ago. So rapid is the infilling by the torrents which sweep down detritus from the surrounding heights that the present lakes are being visibly diminished, and they cannot, therefore, be of high geological antiquity. It is worthy of remark that the glen lakes are almost wholly confined to the western half of the Highlands, where they form the largest sheets of fresh water. Hardly any lakes are to be seen east of a line drawn from Inverness to Perth. West of that line, however, they abound in both the longitudinal and the transverse valleys. The most remarkable line of them is that which fills up so much of the Great Glen. Loch Ness, the largest, is upwards of 20 miles long, about 1 miles broad, and not less than 774 feet deep in the deepest part. This great depression exceeds the general depth reached by the floor of the North Sea between Great Britain and the opposite shores of the Continent. Other important longitudinal lakes are Lochs Tay, Awe, Ericht, and Shiel. The most pictur esque glen lakes, however, lie in transverse valleys, which being cut across the strike of the rocks present greater variety, and usually also more abruptness of outline. Lochs Lomond, Katrine, and Lubnaig in the southern Highlands, and Lochs Maree and More in the north, are conspicuous examples.

(2) Rock-tarns are small lakes lying in rock-basins on the sides of mountains or the summits of ridges, and on rocky plateaus or plains. Unlike the glen lakes, they have no necessary dependence upon lines of valley. On the contrary, they are scattered as it were broadcast over the districts in which they occur, and are by far the most abundant of all the lakes of the country. Dispersed over all parts of the western Highlands, they are most numerous in the north-west, especially in the Outer Hebrides and in the west of Ross-shire and Sutherland. The surface of the Archæan gneiss is so thickly sprinkled with them that many tracts consist almost as much of water as of land. They almost invariably lie on strongly ice-worn platforms of rock. Their sides and the rocky islets which diversify their surface have been powerfully glaciated. They cannot be due to either fracture or subsidence, but are obviously hollows produced by erosion. They have accordingly with much probability been assigned to the gouging action of the sheets of land-ice by which the general glaciation of the country was effected. In the southern uplands, owing probably to the greater softness and uniformity of texture among the rocks, rock-tarns are comparatively infrequent, except in Galloway, where the protrusion of granite and its associated metamorphism have given rise to conditions of rock-structure more like those of the Highlands. Over the rocky hill-ranges of the central Lowlands rock-tarns occasionally make their appearance. (3) Moraine-tarns-small sheets of water ponded back by some of the last moraines shed by the retreating glaciers-are confined to the more mountainous tracts. Among the southern uplands many beautiful examples may be seen, probably the best known and certainly one of the most picturesque being the wild lonely Loch Skene lying in a recess of Whitecoomb at the head of the Moffat Water. Others are sprinkled over the higher parts of the valleys in Galloway. None occur in the central Lowlands. In the Highlands they may be counted by hundreds, nestling in the bottoms of the corries. In the north-western counties, where the

glaciers continued longest to descend to the sea-level, lakes retained
by moraine-barriers may be found very little above the sea.
(4) The lakes of the plains. lie in hollows of the glacial detritus
which is strewn so thickly over the lower grounds. As these
hollows were caused by original irregular deposition rather than!
by erosion, they have no intimate relation to the present drainage.
lines of the country. The lakes vary in size from mere pools up to
wide sheets of water several square miles in area. As a rule they
are shallow in proportion to their extent of surface.. Though still
sufficiently numerous in the Lowlands, they were once greatly
more so, for, partly from natural causes and partly by artificial
means, they have been made to disappear. The largest sheets of
fresh water in the midland valley are of this class, as Loch Leven
and the Lake of Menteith.

Coast-Linc.-The eastern and western seaboards of Scotland Coast. present a singular contrast. The former is indented by a series of line broad arms of the sea, but is otherwise tolerably unbroken. The land slopes gently down to the margin of the sea or to the edge ot cliffs that have been cut back by the waves. The shores are for the most part low, with few islands in front of them, and cultivation comes down to the tide-line. The western side of the country, on the contrary, is from end to end intersected with long narrow sea lochs or fjords. The land shelves down rapidly into the sea and is fronted by chains and groups of islands. This contrast has sometimes been erroneously referred to greater erosion by the waves on the western than on the eastern coast. The true explanation, however, must be sought in the geological structure of the land. The west side of Scotland, as we have seen, has been more deeply eroded than the eastern. The glens are more numerous there and on the whole deeper and narrower. Many of them are prolonged under the sea; in other words, the narrow deep fjords which wind so far into the land are seaward continuations of the glens which emerge from their upper ends. The presence of the sea in these fjords is an accident. If they could be raised out of the sea they would become gleus, with lakes filling up their deeper portions. That this has really been their history can hardly admit of question. They are submerged land-valleys, and as they run down the whole western coast they show that side of the country to have subsided to a considerable depth beneath its former level. The Scottish sea lochs must be viewed in connexion with those of western Ireland and of Norway. The whole of this north-western coast-line of Europe bears witness to recent submergence. The bed of the North Sea, which at no distant date in geological history was a land surface across which plants and animals migrated freely into Great Britain, sank beneath the sea-level, while the Atlantic ad. vanced upon the western margin of the continent and filled the sea. ward ends of what had previously been valleys open to the sun. No improbably the amount of subsidence was greater towards the west Nearly the whole coast-line of Scotland is rocky. On the eas side of the country, indeed, the shores of the estuaries are gener ally low, but the land between the mouths of these inlets is mor or less precipitous. On the west side the coast is for the most part either a steep rocky declivity or a sea-wall, though strips of lower ground are found in the bays. The sea-cliffs everywhere vary in their characters according to the nature of the rock out of which they have been carved. At Cape Wrath precipices nearly 300 feet high have been cut out of the Archæan gneiss. The varying tex ture of this rock, its irregular foliation and jointing, and its ramifying veins of pegmatite conspire to give it very unequal powers of resistance in different parts of its mass. Consequently it projects in irregular bastions and buttresses and retires into deep recesse and tunnels, showing everywhere a ruggedness of aspect which is eminently characteristic. In striking contrast to these precipices are those of the Cambrian red sandstone a few miles to the east. Vast vertical walls of rock shoot up from the waves to a height of 600 feet, cut by their perpendicular joints into quadrangular piers and projections, some of which even stand out alone as cathedrallike islets in front of the main cliff. The sombre colouring is. relieved by lines of vegetation along the edges of the nearly flat beds which project like vast cornices and serve as nesting-places. for crowds of sea-fowl. On the west side of the country the most notable cliffs south from those of Cape Wrath and the Cambrian sandstones of Sutherland are to be found among the basaltic islands, particularly in Skye, where a magnificent range of precipices rising to 1000 feet bounds the western coast-line. The highest cliffs in the country are found among the Shetland and Orkney Islands. The sea-wall of Foula, one of the Shetland group, and the western. front of Hoy in Orkney rise like walls to heights of 1100 or 1200 feet above the waves that tunnel their base. Caithness is one wide moor, terminating almost everywhere in a range of sea-precipices. of Old Red Sandstone. Along the eastern coast-line most of the cliffs are formed of rocks belonging to the same formation. Beginning at Stonehaven, an almost unbroken line of precipice varying up to 200 feet in height runs southwards to the mouth of the estuary of the Tay. The southern uplands plunge abruptly into. the sea near St Abb's Head in a noble range of precipices 300 to 500 feet in height, and on the western side the same high grounds.