Conceive a mass of ice lying on a flat horizontal surface, and receiving heat on its upper surface, say from the sun; as the heat passes downwards through the mass, the molecules, acting as conductors, melt and resolidify. Each fluid molecule solidifies in an interstice, which has to be widened in order to contain it. The pressure thus exerted by the continual resolidifying of the molecules will cause the mass to widen out laterally, and of course as the mass widens out it will grow thinner and thinner if it does not receive fresh acquisition on its surface. In the case of a glacier lying in a valley, motion, however, will only take place in one direction. The sides of the valley prevent the glacier from widening; and as gravitation opposes the motion of the ice up, and favours its motion down the valley, the path of least resistance to molecular pressure will always be down the slope, and consequently in this direction molecular displacement will take place. Molecular pressure will therefore produce motion in the same direction as that of gravity. In other words, it will tend to cause the glacier to descend the valley.
The lateral expansion of the ice from internal molecular pressure explains in a clear and satisfactory manner how rock-basins may be excavated by means of land-ice. It also removes the difficulties which have been felt in accounting for the ascent of ice up a steep slope. The main difficulty besetting the theory of the excavation of rock-basins by ice is to explain how the ice after entering the basin manages to get out again—how the ice at the bottom is made to ascend the sloping sides of the basin. Pressure acting from behind, it has been argued by some; but if the basin be deep and its sides steep, this will simply cause the ice lying above the level of the basin to move forward over the surface of the mass filling it. This conclusion is, however, incorrect. The ice filling the basin and the glacier overlying it are united in one solid mass, so that the latter cannot move over the former without shearing; and although the resistance to motion offered by the sloping sides of the basin may be much greater than the resistance to shear, still the ice will be slowly dragged out of the basin. However, in order to obviate this objection to which I refer, the advocates of the glacial origin of lake-basins point out that the length of those basins in proportion to their depth is so great that the slope up which the ice has to pass is in reality but small. This no doubt is true of lake-basins in general, but it does not hold universally true. But the theory does not demand that an ice-formed lake-basin cannot have steep sides. We have incontestable evidence that ice will pass up a steep slope; and, if ice can pass up a steep slope, it can excavate a basin with a steep slope. That ice will pass up a steep slope is proved by the fact that comparatively deep and narrow river valleys are often found striated across, while hills which stood directly in the path of the ice of the glacial epoch are sometimes found striated upwards from their base to their summit. Some striking examples of striæ running up hill are given by Professor Geikie in his “Glacial Drift of Scotland.” I have myself seen a slope striated upwards so steep that one could not climb it.
A very good example of a river valley striated across came under my observation during the past summer. The Tay, between Cargill and Stanley (in the centre of the broad plain of Strathmore), has excavated, through the Old Red Sandstone, a channel between 200 and 300 feet in depth. The channel here runs at right angles to the path taken during the glacial epoch by the great mass of ice coming from the North-west Highlands. At a short distance below Cargill, the trap rising out of the bed of the river is beautifully ice-grooved and striated, at right angles to the stream. A trap-dyke, several miles in length, crosses the river about a mile above Stanley, forming a rapid, known as the Linn of Campsie. This dyke is moutonnée and striated from near the Linn up the sloping bank to the level of the surrounding country, showing that the ice must have ascended a gradient of one in seven to a height of 300 feet.
From what has been already stated in reference to the resolidifying of the molecules in the interstices of the ice, the application of the molecular theory to the explanation of the effects under consideration will no doubt be apparent. Take the case of the passage of the ice-sheet across a river valley. As the upper surface of the ice-sheet is constantly receiving heat from the sun and the air in contact with it, there is consequently a transferrence of heat from above downwards to the bottom of the sheet. This transferrence of heat from molecule to molecule is accompanied by the melting and resolidifying of the successive molecules in the manner already detailed. As the fluid molecules tend to flow into adjoining interstices before solidifying and assuming the crystalline form, the interstices of the ice at the bottom of the valley are constantly being filled by fluid molecules from above. These molecules no sooner enter the interstices than they pass into the crystalline form, and become, of course, separated from their neighbours by fresh interstices, which new interstices become filled by fluid molecules, which, in turn, crystallize, forming fresh interstices, and so on. The ice at the bottom of the valley, so long as this process continues, is constantly receiving fresh additions from above. The ice must therefore expand laterally to make room for these additions, which it must do unless the resistance to lateral expansion be greater than the force exerted by the molecules in crystallizing. But a resistance sufficient to do this must be enormous. The ice at the bottom of the valley cannot expand laterally without passing up the sloping sides. In expanding it will take the path of least resistance, but the path of least resistance will always be on the side of the valley towards which the general mass of the ice above is flowing.
It has been shown ([Chapter XXVII.]) that the ice passing over Strathmore must have been over 2,000 feet in thickness. An ice-sheet 2,000 feet in thickness exerts on its bed a pressure of upwards of 51 tons per square foot. When we reflect that ice under so enormous a pressure, with grinding materials lying underneath, was forced by irresistible molecular energy up an incline of one in seven, it is not at all surprising that the hard trap should be ground down and striated.
We can also understand how the softer portions of the rocky surface over which the ice moved should have been excavated into hollow basins. We have also an explanation of the transport of boulders from a lower to a higher level, for if ice can move from a lower to a higher level, it of course can carry boulders along with it.
The bearing which the foregoing considerations of the manner in which heat is transmitted through ice have on the question of the cause of regelation will be considered in the Appendix.