the bottom of all lies a gnarled highly crystalline gneiss, called by Sir Roderick the Fundamental Gneiss,' inasmuch as he has shown it to be the oldest rock in the British Isles, and the foundation on which all the others are built up. Resting on the upturned worn edges of this truly primæval rock rise huge masses of gently-inclined red sandstone and conglomerate, supposed to be of the same geological age as the Cambrian rocks of Wales. These strata, formerly described and mapped as Old Red Sandstone, form the lofty pyramidal mountains on the Atlantic borders of Sutherland and Ross, and are confined to the north-west of the island. They are overlaid with thick quartz rocks and bands of limestone which pass under flaggy micaceous gneisses and schists, and contain in their under-portions Lower Silurian fossils. Hence we learn that the rocks which overlie the red Cambrian sandstones, even though they are crystalline and wholly changed from the soft loose strata in which they were laid down on the sea-bed, yet do not belong to the age of chaos, as used to be supposed, nor to an epoch anterior to the creation of life, but are really the equivalents in time of the Lower Silurian rocks of the south of Scotland, of Cumberland, and of Wales. A reference to the long section on the map from Skye to the Cheviot Hills will show

1 He has also called it "Laurentian " from its probable identity with a gneiss occupying a similar position in Canada.

the reader that it is these metamorphosed Lower Silurian strata which form the greater part of the Highlands, and that they are arranged in great folds, whereby the lower portions of the series are brought up to the surface again and again.1

The strata of sand and mud which accumulated to a depth of thousands of feet over the sinking floor of the old Silurian ocean have been crumbled up into endless folds and puckerings of which, as may be seen on the map, the long axis, or strike, runs generally in a north-easterly and south-westerly direction. When the wind blows freshly from the north-west the sea is roughened with long broken lines of wave, stretching from south-west to north-east, and rolling in towards the south-east. So over the Scottish Highlands the gneissose and schistose rocks have been tossed, as it were, by a long swell from the north-west into numerous wave-like plications that follow each other, fold after fold, and curve after curve, from Cape Wrath to the Lowland border.

Thus, instead of a scene of undistinguishable chaos, the Scottish Highlands are found to be governed by the same laws of structure as other hilly regions formed of stratified rocks. We are not left, therefore, to speculate wholly in the dark upon the origin

1 See the details of this structure, given in a Memoir on the subject by Sir Roderick Murchison and the author in the Quarterly Journal of the Geological Society, vol. xvii. 171.

Connexion of Geological Structure and Scenery. 95

of the Highland mountains and glens, as if they had been upheaved among rocks of which the mode of formation and the actual structure were alike unknown. If we were still vaguely surmising that the gneiss and the schist had been thrown down upon the floor of a primæval thermal ocean, and had been broken up when that hardened floor was upheaved into the first dry land, there might be some excuse for a belief that the ancient convulsions had been the means of throwing up our northern mountains and tearing open their glens. But when these rocks are discovered to be only a modification of ordinary sedimentary deposits, and to reveal their geological age by their enclosed organic remains, all such vague conjectures must cease, and the rocks must be examined and determined by the ascertained laws which govern the arrangement of masses of stratified rocks.

Connexion of Geological Structure and Scenery.When it is said that the rocks of which the Highlands are made have been thrown into oft-repeated foldings it might be supposed that these undulations of the strata have given rise to the present contour of the surface, each mountain-chain corresponding with an arch or anticlinal axis of the gneiss and schists, and each line of strata or glen with a trough or synclinal axis. On the contrary, the very reverse of this arrangement is often found to hold. What in a geological sense are basins or troughs, frequently rise

into rugged and lofty mountains, while the arches, on the other hand, are occupied by deep valleys. A striking example of this feature is to be found in Ben Lawers. That wide-based broad-shouldered moun

FIG. 6.-SECTION OF BEN LAWERS.

a Quartzose rocks. b Limestone. c Schistose rocks.
(See also the section No. 2 in the Map.)

tain rises from the valley of Loch Tay on the one side, and sinks into Glen Lyon on the other. It forms thus a huge dome-shaped mass between two deep valleys. But instead of owing this form to an upward curving of the schists, it actually lies in a basin of these rocks which dip underneath the mountain on the banks of Loch Tay, and rise up again from its further skirts in Glen Lyon. Thus Ben Lawers is in reality formed of a trough of schists, while the valley of Loch Tay runs along the top of an anticlinal arch. Hence that which, in geological structure, is a depression has by denudation become a great mountain, while what is an elevation has been turned into a deep valley.

The present heights and hollows of the Highlands, therefore, are not to be traced to any of the original

convolutions of the old crystalline rocks. Nor can they be assigned, as they often are in a popular way, to grand primeval eruptions of granite. That rock, it is true, covers a considerable space in the Highlands, and rises up among the highest mountain groups. But there are also wide spaces of low ground abounding in granite. The long lonely Moor of Rannoch, for instance, lies in large measure on granite; while the range of mountains that bounds its south-eastern margin consists, not of granite, but of quartz rock. Indeed, there is good reason to believe that granite is not an igneous rock in the ordinary sense, but that instead of bursting through and upheaving the gneiss and schist, it is itself only a further stage of the metamorphism of these rocks.

Until the Highland tracts are surveyed in minute detail, it will not be possible to ascertain how far they are traversed by lines of fault, nor to what extent such features have shewn themselves at the surface, and have served to guide the excavation of the valleys. After a long and detailed examination of the contorted rocks of the Silurian uplands of the southern counties, I have been led to believe that the faults and the folds of the strata on the whole have had only a secondary influence in originating the present irregularities of the surface. And this is probably the case also with the contorted and metamorphosed Silurian rocks of the Highlands. The

H