With regard to the diminution of the _glaciers reservoirs_, or, in other words, to the manner in which the ice disappears, notwithstanding the continual additions made to it, we have the following remarkable passage:--"In seeking the cause of the diminution of glaciers, it has occurred to my mind that the ice, notwithstanding its hardness and its rigidity, can only support a given pressure without breaking or being squeezed out. According to this supposition, whenever the pressure exceeds that force, there will be rupture of the ice, and a flow in consequence. Let us take, at the summit of Mont Blanc, a column of ice reposing on a horizontal base. The ice which forms the first layer of that column is compressed by the weight of all the layers above it; but if the solidity of the said first layer can only support a weight equal to 100, when the weight exceeds this amount there will be rupture and spreading out of the ice of the base. Now, something very similar occurs in the immense crust of ice which covers the summits of Mont Blanc. This crust appears to augment at the upper surface and to diminish by the sides. To assure oneself that the movement is due to the force of pressure, it would be necessary to make a series of experiments upon the solidity of ice, such as have not yet been attempted."[G] I may remark that such experiments substantially verify M. Rendu's notion. But it is his observations and reasoning upon the _glaciers d'ecoulement_ that chiefly interest us. The passages in his writings where he insists upon the power of the glaciers to mould themselves to their localities, and compares them to a soft paste, to lava at once ductile and liquid, are well known from the frequent and flattering references of Professor Forbes; but there are others of much greater importance, which have hitherto remained unknown in this country. Regarding the motion of the Mer de Glace, Rendu writes as follows:-- [Sidenote: MEASUREMENT OF MOTION.] [Sidenote: THE SIDES OF THE GLACIER RETARDED.]