former being more or less fissured by rents. But the contrary is the case on the great glaciers which move on small slopes, and with smaller velocities; and the discovery of this fact rewarded me for the labour of a short visit which I made the Great Aletsch glacier, in July 1844, when I ascertained, not merely the small daily progress of the mass of the glacier, but the astonishing retardation produced by the sides, whilst the surface remained compact and wholly undivided by longitudinal crevasses. In that case, I found that, "whilst the velocity of the ice at 1300 feet, or about a quarter of a mile from the side, is 14 inches in 24 hours, at 300 feet distant from the side it was but 3 inches in the same time; and close to the side it had nearly, if not entirely, vanished."* Now this observation, a hasty one, and which, therefore, I am happy to have confirmed, is more than borne out by the observations on the glacier of the Aar, detailed in the Comptes Rendus, and which were made shortly after. The movement of the centre of the glacier is to that of a point 5 metres from the edge as FOURTEEN to ONE; such is the effect of plasticity! Thirteen-fourteenths of the motion of the glacier of the Aar are due to the sliding of the ice over its own particles, and one-fourteenth only to its motion over the soil.

V. Motion of Glaciers of the Second Order. It is a question of considerable interest to know how those small glaciers, called by De Saussure glaciers of the second order, advance, compared to the great ice masses which fill the bottoms of valleys. These little glaciers, on the contrary, are usually isolated, extending but a small way, occupying a nook or niche in a mountain side, and though persisting in their occupancy, and shewing signs of motion and activity, like other glaciers, yet stretch forward but a small way, then cease abruptly, as if foiled in the struggle to join their icy contribution to the magnificent glacier which often fills the valley immediately below them.† Their isolated

Eighth Letter on Glaciers, Edin. Phil. Journal, Oct. 1844 [p. 62 of this volume].

See a plate, giving a correct idea of a glacier of the second order, in my Travels, Plate ix.

1845.] MOTION OF GLACIERS OF THE SECOND ORDER.

75

position, their great absolute height, and their usually very steep declivity and small surface, give considerable interest to the determination of their rate of motion, at least approximately. Accordingly, I seized the occasion of spending some days in July 1844, at the Hospice of the Simplon (already at a height of 6600 feet above the sea), to examine and measure the progress of the small glacier which hangs from the slope of the Schönhorn, immediately behind it, and 1400 feet higher. I intend to give elsewhere a minute account of this glacier, and my observations upon it; but in the mean time I may state that one of the marks observed, at a point having an inclination of 10o, moved at the rate of 1.4 inches in 24 hours; and another, at an inclination of 20°, moved 1.8 inches in the same time. This small result is quite conformable with the dry and powdery condition of such elevated glaciers, yielding little water, and capable of exerting, on their under parts, a very trifling hydrostatic pressure.

*

Exactly analogous results were obtained by M. Agassiz's coadjutors at a somewhat later period of the same year. The experiments are fully detailed in the Comptes Rendus; † and the conclusions which are deducible from them, are—(1.) That the daily motions of these small glaciers, which rested on beds so highly inclined as from 15° up to 33°, are included between 20 and 72 millimetres (0.79 to 2.84 English inches) per diem. (2.) The observers think that their observations go to prove that when these glaciers are prolonged far enough to meet the main glacier below, and to unite their streams, then the lower part of the tributary glacier, or that nearest the point of union, moves SLOWER than the upper part, or that nearest the origin of the little glacier; but, on the contrary, if the glacier be pendant on the slope, and the lower end decays away without joining the principal, then the inferior extremity moves FASTER than the origin. Now the cause of this variation in the two

*This experiment is briefly mentioned at the close of my Eighth Letter on Glaciers. P. 1303, and two following pages.

cases (should the fact really appear to be general, as is not unikely, provided the lower station be always chosen low enough) seems to be, that the main glacier resists the interference of its tributary with its course, and consequently represses its stream, causing a heaping up in front, such as mere friction on a low inclination alone produces, and is thus in conformity with the viscous theory. In the other case, that of the free glacier of the second order, the difference of velocity at the upper and lower station (one-seventh part only) is not more than the difference of slope (15° and 25°) will readily explain.

VI. Movements of Bas-Névés or Snow-beds. One observation remains which completes my analysis of the measures of M. Agassiz's coadjutors. It is one of considerable interest, and I believe is new. It is the establishment of the fact that the highly inclined beds of old snow, formed by avalanches, which lie unmelted in the ravines, without assuming any external trace of glacial structure, have a proper motion of their own. This, though to me not unexpected, is very interesting; for the most attached advocate of either the dilatation or the sliding theory, will hardly maintain, on the one hand, that the congelation of soft snow could act here as a propelling force, or on the other, that the motion can take place without acceleration in the totality of a mass, inclined (in this case) at an angle even of 43°, over the bed on which it rests: especially since the actual movement under this enormous inclination was only 7 millimetres, or three-tenths of an inch, per day;* or one-thirtieth of that of the great glacier under an inclination of but a few degrees. The velocity increased towards the lower extremity, as in the free glacier of the second order. On the plastic theory, this evidently presents the extreme case of a body, approaching in its nature to a soft heavy powder slightly moistened, which gives way by the yielding of its parts, and so far resembles a fluid (as a bank of earth, slightly glutinous, rather than sand does); and the slowness of movement is in conformity with the imperfectness of the fluid pressure, and with the fact already *Comptes Rendus, p. 1305, line 32; and p 1306.