198. And now for the Léchaud branch. On August 1 we fix ten stakes across this glacier above the point where it is joined by the Talèfre. Measured on August 3, and reduced to twenty-four hours, the motion was found to be-- MOTION OF GLACIER DE LÉCHAUD. Seventh Line: K K' upon Sketch. Stake 1 2 3 4 5 6 7 8 9 10 Inches 5 8 10 9 9 8 6 9 7 6 199. Here our conjecture is still further verified, the rate of motion being even less than that of the Glacier du Géant. § 28. _Motion of Top and Bottom of Glacier._ 200. We have here the most ample and varied evidence that the sides of a glacier, like those of a river, are retarded by friction against its boundaries. But the likeness does not end here. The motion of a river is retarded by the friction against its bed. Two observers, viz. Prof. Forbes and M. Charles Martins, concur in showing the same to be the case with a glacier. The observations of both have been objected to; hence it is all the more incumbent on us to seek for decisive evidence. 201. At the Tacul (near the point _a_ upon the sketch plan, p. 83) a wall of ice about 150 feet high has already attracted our attention. Bending round to join the Léchaud the Glacier du Géant is here drawn away from the mountain side, and exposes a fine section. We try to measure it top, bottom, and middle, and are defeated twice over. We try it a third time and succeed. A stake is fixed at the summit of the ice-precipice, another at 4 feet from the bottom, and a third at 35 feet above the bottom. These lower stakes are fixed at some risk of boulders falling upon us from above; but by skill and caution we succeed in measuring the motions of all three. For 24 hours the motions are:-- Top stake 6 inches. Middle stake 4½ " Bottom stake 2⅔ " 202. The retarding influence of the bed of the glacier is reduced to demonstration by these measurements. The bottom does not move with half the velocity of the surface. § 29. _Lateral Compression of a Glacier._ 203. Furnished with the knowledge which these labours and measurements have given us, let us once more climb to our station beside the Cleft under the Aiguille de Charmoz. At our first visit we saw the medial moraines of the glacier, but we knew nothing about their cause. We now know that they mark upon the trunk its tributary glaciers. Cast your eye, then, first upon the Glacier du Géant; realise its width in its own valley, and see how much it is narrowed at Trélaporte. The broad ice-stream of the Léchaud is still more surprising, being squeezed upon the Mer de Glace to a narrow white band between its bounding moraines. The Talèfre undergoes similar compression. Let us now descend, shake out our chain, measure, and express in numbers the width of the tributaries, and the actual amount of compression suffered at Trélaporte. 204. We find the width of the Glacier du Géant to be 5,155 links, or 1,134 yards.