_Torrents_ have a true degrading and _scooping_ action upon the earth’s surface, but, by the necessary consequence of the sense which we attach to the word, this action cannot be exercised upon spaces of great extent, for a torrent is a water-course which has a great declivity. Now, on account of the little height which the most elevated summits of the globe have in comparison with the extent of its surface, this action cannot be very extensive; it can only, therefore, produce short and narrow ravines. This action, as all who have visited high mountain chains may have seen, is only often local and instantaneous; it presents no remarkable effect but upon the heaps of debris which cover the declivities of the mountains, and on broken rocks, partially disintegrated by other causes, and lastly on moveable deposits. The results of this action contribute to confine it within narrower limits still, by heaping up at the mouths of torrents in the valleys or plains, the debris carried down by these torrents. The elevation of the soil, which necessarily follows from the accumulation of these debris, diminishes with the declivity, the rapidity, and consequently the power of these water-courses. Great masses of water moving rapidly, have a marked transporting power. Striking examples of this power have but too often been seen in Holland, by the breaking down of the dikes, and in Alpine mountains, in consequence of extraordinary rains during tempests, or from the rupture of some of the natural barriers of lakes. In these latter times (in 1818), the Vallée de Bagne experienced the terrible effects of this devastating power. Masses of ice having fallen towards the upper part of this valley, and accumulated there, raised a dike sufficiently compact and strong to block up the course of the Dranse. The waters of this river, rapid and pent up in certain parts of its course, as are all those of the high Alps, accumulated above this barrier of ice, and formed a lake which attained, at its maximum, 130 metres of mean breadth, from 3000 to 4000 metres of length, and 36 of mean depth, and consequently a volume of water estimated at about 29,000,000 cubic metres. Although, by means of operations conducted with equal skill and courage, about the third part of this volume was let off without danger, the remaining part having suddenly broken through the barrier of ice, was precipitated with an almost unexampled impetuosity of 11 metres in the second, into the Vallée de Bagne. In the first part of its course, and in the space of half an hour which the mass of water took in traversing a league, it carried away trees, dwellings, enormous masses of debris, and rocks _already separated from their mass_, as M. Escher, expressly says; it covered all the broad parts of the valley with rubbish, pebbles and sand, and carried the remainder of the substances which it had swept away, as well to the extremity of the valley, towards Martigny, as into the bed of the Rhone. The