able facts which he has elicited. Exceptions can be taken to it, of course. It may be asked, at the very outset, Is it absolutely necessary to accept this dogma that the only remaining agent is the denuding power of ice? Have we arrived at the end of all knowledge? And the cogency of the reasoning may be doubted by which the conclusion is derived, that rock-basins have necessarily been excavated by ice, because they are commonly found in districts which were formerly covered by glacier. It may be said that the connection which has been shown between the two[1] may be nothing more than an accidental coincidence, and that, taken by itself, it is scarcely more convincing than that icebergs have made the Arctic seas, because those seas are full of icebergs. Such objections, however, do not touch Professor Ramsay's main arguments; and I think that any one who honestly endeavours to master them will feel that they are very ingenious, and that they are by no means easy to refute.
It is impossible to deny a certain limited power of erosion to glaciers; and it is difficult to see why a great glacier should not make a hollow (a shallow one) if it were to come down upon a plain, and work there for a long time. For example, let a c b d, in the accompanying diagram, be a transverse section of a glacier which is moving over level ground, a g d f b, The glacier would naturally be thickest towards the centre, and its motion would probably be greatest in the same neighbourhood. It should therefore erode its bed to a greater extent at or about the point d than anywhere else; and as the motion and weight of the ice would be greater at or about f and g than at points between f b or g a, so also would the erosion be greater thereabouts. In short, it is reasonable to conclude that in course of time the glacier might form a hollow in its previously level bed, such as is represented by the dotted
- ↑ Professor Ramsay claims to be the first who has pointed out this connection. Professor Dana extends the statement still further:—"Another great fact that belongs to the Drift latitudes on all the continents, and may have the same origin, is the occurrence, on the coasts, of fiord valleys,—deep, narrow channels, occupied by the sea, and extending inward often 50 or 100 miles."—Manual of Geology, 1867, p. 541.
