produced ; and we must remember that in the most conspicuous of those described above, the most compact rock forms the lowest stage. To this question of cliffs in the bed of a glacier I shall return again; for the present we may venture to assert that it is mechanically impossible for a glacier to sink into its bed so as to excavate a cirque on so gigantic a scale. (6) But may they have been produced at some bend in the ice-stream like the concavities in a river's course ? The Creux de Champs and Am Ende der Welt certainly cannot be thus explained, because they are at the head of deep well-marked lateral glens about a mile and a half long. The two cirques near the Blacken Alp are inexplicable on this theory ; for the spurs from the Blackenstock and Rothschutz would effectually shelter them from the action of any glacier descending from the Surenen Pass (7562 feet); and indeed the general contour of the ground suggests that their recesses would be among the feeders of its neve, the rise from the Blacken Alp to the last ridge leading to the Pass being comparatively gentle. The Per-a-Cheval would offer less resistance to this explanation than any other cirque, as it is on the concave side of the elbow of a valley ; but as existing glaciers do not appear to produce such marked effects in turning corners in their present beds, we may fairly require much corroborative evidence before venturing to apply this explanation here.
There remains then the wider question to consider, Are we justified in supposing glaciers to have been the principal agents in the excavation of the Alpine valleys ? To prevent mistakes, let me state that I do not now purpose to deal with that particular case of this theory which is advocated by several distinguished foreign geologists, and with so much ability in this country by Professor Ramsay, viz. the excavation of lake-basins by glaciers, but rather with that wider view to which, among others, Professor Tyndall has more than once given the support of his experience and talent, especially in a paper published in the Philosophical Magazine*.
If we assign to glaciers any large share in the excavation of valleys, we may fairly ask that the main contours should lend themselves readily to this explanation, just as those of a river- valley, although somewhat modified by other meteoric action, suggest running water as the tool that has principally fashioned them. Now, as a rule, in the upper parts of Alpine valleys, where the glacial action may be supposed most recent, and therefore least affected by other denuding agents, we are struck by the steepness of the last few miles up to the watershed, as compared with the inclination of the rest of the bed. I will only mention a few instances out of many — for the fact must have struck all travellers — and will select the first from the Dauphine Alps, not only because it is a district very familiar to myself, but also because the valley of the Veneon runs, for its whole extent, with one most unimportant exception, through remarkably hard crystalline rocks. Measured on the map, this valley is about twenty miles in length from the spot where the river debouches into the small plain above the Romanche, near
- Phil. Mag. Ser. 4. vol. xxiv. p. 169.
VOL. XXVII. PART I. Z
