and more unequal texture of the rocks of the Highland-like tracts, combined with a greater rainfall and consequent intenser erosion on the western than on the eastern side of the country. The wonderful uniformity of scenery throughout the uplands is only a faithful reflex of the remarkable persistence of the same geological structure. Any marked change in the character and durability of the rocks is sure to make itself apparent in the form of the ground. Thus the bands of hard grit and fine conglomerate, which run in continuous parallel bands in a south-west and north-east direction along the uplands, show themselves in more rugged strips of hill, on the sides and crests of which the naked rock more frequently protrudes in crags and scars. Such are the heights that range south-westwards from Queensberry Hill across Nithsdale into Galloway. The shales, on the other hand, crumble down into finer debris, and consequently produce long smooth declivities, such as the green slopes of the Lowther Hills.

The intrusive igneous rocks likewise give rise to local peculiarities of configuration. The granite hills of Galloway, though never peaked, display their lines of low grey cliff, from which many a block has been scattered over the lower grounds. The detached bosses of felsite, or the necks of ancient Carboniferous volcanoes, stand out as prominent cones, while the sheets of Old Red Sandstone and Carboniferous lava range in lines of terraced escarpment. The best locality for tracing the influence of such igneous rocks upon landscape is in the tract of Border country between Birrenswark and the Merse of Berwickshire. Among the more prominent eminences of this kind are Tinnis Hill, and Watch Hill in Liddesdale, Pike Fell (and part of Arkleton Fell) in Ewesdale, Greatmoor, Maiden Paps, Scawd Law, Leap Hill, and Windburgh Fell at the head of the Slitrig

Valley, Bonchester Hill, Rubers Law, Black Law, Dunian Hill, Lanton Hill, Minto Crags, Peniel Heugh, the Eildon Hills, the numerous little hills between Maxton on the Tweed and the foot of Lauderdale, and the long line of craggy heights that stretch by Smailholm and Stitchill to Greenlaw.

In connection with the eruptive rocks of the region, reference has been made to the basalt-dykes which run across the uplands from south-east to north-west. Though they do not form any striking feature in the topography,

their persistent course across hill and valley throws much light on the history of the denudation of the high grounds of the south of Scotland. The evidence is precisely similar to that adduced from the Highlands. The Tertiary age of these dykes may be considered to be established beyond question. They cross some of the widest and deepest valleys among the Silurian hills, and they therefore prove that these valleys must have been eroded since older Tertiary time. The case of Annandale is particularly interest

ing. I have referred to that valley as being as old as the Permian period. That it was filled up with Permian deposits, and remained so in early Tertiary time, is shown by the great dyke which crosses it near Moffat. Since then the Permian breccia

valley has been cleared out, most of the and sandstone has been removed, and the dyke has been laid bare along the slopes and bottom. The dyke forms. no prominent surface feature at Moffat, but a few miles to the south-east where it crosses the Glengap Burn, near the Wamphray Water, it runs as a prominent rib down each side of the valley (Fig. 68). A more striking proof of the excavation of the valley since the time of the dyke could not be desired. Other dykes cross the valley of the Clyde below Crawford, so that the erosion of that hollow also has been effected since early Tertiary time.

CHAPTER XIV

THE ANCIENT GLACIERS OF THE SOUTHERN UPLANDS

As the rocks of the southern district of Scotland are on the whole less durable than those of the Highlands, they have not preserved quite so faithfully or universally the impress left upon them by the ice-sheets of the Glacial period. That they have been intensely glaciated, however, will be recognised by any one who seeks for proofs of ice-work. In spite of the thick mantle of boulder-clay that covers so much of the valleys and of the lower hill-slopes, and in spite also of the tendency of so many of the rocks to decay and conceal themselves beneath a coating of turf or peat, abundant polished and striated rocks may be found in every district from the headlands of Wigtonshire to those of St. Abb's Head. In some parts of Galloway, indeed, the roches moutonnées are hardly less perfect and conspicuous than in most of the Highlands.

From the direction of the striæ, it is evident that the Southern Uplands formed another centre of dispersion for the southern part of the Scottish ice-sheet.1 A vast mass of ice

1 To speak more accurately, there were several distinct centres of movement of the ice that lay on these uplands, as will be evident from the Glaciation Map. But the southern ice-field may be regarded as one

flowed northwards into the plain of Ayrshire, where, joining the stream that was descending from the Highlands, it bent round to the west and went southwards down the Firth of Clyde. Still thicker and more extensive was the great icefield that crept off the southern side of Galloway into the Solway Firth and the Irish Sea, both of which, at the height of the Ice Age, were filled with ice. Across the eastern part of the uplands, the pressure of the icy stream, which, coming down from the Highlands, spread out over the lowland valley, seems to have driven the southern ice eastward, and the united stream turned away southward along what is now the bed of the North Sea. Proofs of these movements are furnished not only by the direction of the striæ, but by the scattered erratics which have been strewn over the ground. The evidence for them is given in summary form on the Map of the Glaciation of Scotland accompanying this volume.

The boulder-clay tells perhaps even more markedly in the scenery of the Southern Uplands than in the Highlands. It has accumulated to a great thickness in the valleys, where it forms a kind of sloping floor or platform, extending from the base of the declivity on either side, and trenched by the stream which, in winding across it, has exposed great scars of it in the banks. It is more especially banked up on the lee sides of the hills, while the opposite sides, against which the full pressure of the ice came, are comparatively bare. In Peeblesshire and Selkirkshire, where the march of the icesheet was from the west, the west sides of the valleys are thickly spread with boulder-clay, while on the east side the bare rock comes abundantly to the surface.

In these valleys, the boulder-clay has a twofold influence

vast sheet that moved outwards and downwards into the low grounds on all sides.