But before I attempt to explain the non-occurrence of Skye rocks in the till of the Outer Hebrides, let me show in a few words what the glaciation of the Long Island, Skye, and the north-west Highlands teaches us as to the general aspect presented by the ice-sheet. The height reached by the surface of the ice in Ross-shire and the Long Island respectively indicates of course that the main movement was from the mainland. We must conceive of an immense sheet of solid ice filling up all the inequalities of the land, obliterating the glens, and sweeping across the hill-tops; and not only so, but occupying the wide basin of the Minch to the entire exclusion of the sea, the surface of the ice rising so high that it overtopped the whole of the Outer Hebrides, and left only the tips of a few of the higher mountains uncovered. The slope of the surface was persistently outwards from the mainland, and the striation of the Long Island indicates clearly that the dip or inclination of that surface was towards the north-west. Nay, more than this, we are now enabled for the first time to say with some approach to certainty what was the precise angle of that inclination. If we take the upper surface of the ice in Ross-shire to have been 3000 feet (and it was not less), then the slope between the mainland and the Outer Hebrides was only 25 feet in the mile, or about 1 in 210. It is quite possible, however, and even probable, that the actual height attained by the ice-sheet in the north-west Highlands was more than 3000 feet. I think it may yet turn out to have been 3500 feet, and if this were so it would give an inclination for the surface of the ice of about 35 feet in the mile. In either case the slope was so very gentle that to the eye it would have appeared like a level plain. Over the surface of this plain would be scattered here and there a solitary big erratic or two, while in other places long trains of large and small angular boulders would stream outwards. All these would be derived from such mountain in Skye and the mainland as were able to keep their heads above the level of the ice-flow; while a few also might be dislodged by the frost and rolled down upon the glacier from the tips of the Clisham and the Langa in Harris, and Hecla and Beinn Mhor in South Uist. Every such block, it is evident, would be carried across the buried Hebrides, out into the Atlantic in the direction indicated by the glaciation of the Long Island--that is, towards the north-west.