In 1840, M. Agassiz, with some scientific friends, Messrs. Desor, Vogt, and Nicolieb, established themselves under a great overhanging slab of rock on the same moraine, and by means of side walls, and other appliances, constructed a rough abode which, because some of these men of science came from Neufchâtel, they named the “Hôtel des Neuchâtelois.” In two years after its erection, Agassiz discovered that it had moved downwards no less a distance than 486 feet. These and some similar measurements brought to light a very important fact. The reader will observe that the _middle numbers_, corresponding to the central portion of the glacier, are the _largest_: hence it was obvious that the _centre of a glacier, like that of a river, moves more rapidly than the sides_. Owing to the greater central motion of a glacier, its crevasses invariably assume a curved outline, of which the convexity advances towards the bottom of the valley. It has also been ascertained that the superficial part of a glacier moves more rapidly than its base. Again: Tyndall and Hirst, by employing instruments of great precision, have demonstrated that the maximum of motion is not to be found exactly in the centre, but that, according to the windings of the valley through which the glacier flows, it moves sometimes to the right of the centre, and sometimes to the left. Now, the progression of a river exhibits all the characters we have just enumerated, and the truth foreshadowed by Rendu has been confirmed in every detail. The glacier _is_ a “river of ice.” The reader will naturally ask, How can a substance of such apparent rigidity as ice obey, as it _does_ obey, the same laws which regulate the movement of fluids? I can understand, he may say, how water flows in such and such a manner: it is a liquid, and its molecules are deficient in the property of cohesion; but that so solid, and firm, and unimpressible a substance as ice should be capable of motion seems impossible. I can understand very easily that a mass of ice, when loosened or detached from its resting-place, will glide downwards until arrested by some adequate obstacle; but this is not the kind of motion you are describing. According to your explanations, every constituent portion of the glacier moves, and the central faster than the lateral, and the surface faster than the base. These objections were advanced by men of science when the motion of glaciers was first put forward as a theory; and the answer given by Scheuchzer was, that a glacier might be compared, in the summer season, to a sponge saturated with water, which, when afterwards congealed by the cold temperature of autumn and winter, expanded, and produced a dilatation of the mass in every direction. Then, as it could not recede, as it could not _reascend_ its valley-slope, the augmentation of size would necessarily take place in its lower portion.