stated further than to mention that the opinion of Prof. Forbes, that it flows as a "viscous substance," is not accepted. Wax, we may say, is both plastic and viscous, and yields equally to pressure and to tension. Ice yields to pressure, but not to tension. It cannot be stretched. Glaciers maintain their cohesion under pressure as plastic bodies, but, wanting viscosity, they break into profound chasms where the tension is great. In ice, therefore, one property is wanting to render it a viscous substance. In Fig. 9 is beautifully shown the opening of crevasses on the margin of a glacier, where the flow is retarded by friction. A like phenomenon occurs when a glacier falls in a cascade over a precipice. Then chasms appear of startling
Fig. 9.
Marginal Crevasses.
depths, and gigantic blocks of ice are thrown into the widest confusion. From the chaos come sounds which indicate what is going on below. The nether air is filled with echoes from the murmur or roar of water, the falling of bowlders, and crashing of ice. At the foot of the fall the broken fragments of ice are crowded together, become solid by regelation, and the mass moves on.
The terminus of a glacier may be many thousand feet below the limit of perpetual snow before its disintegration is complete. But the wonderful fabric falls at last, as heat destroys its molecular framework, and is lost in the turbid flood which forever pours from beneath its portals. But the sediment of the incipient rivers thus formed was no part of the crystalline structure, for crystallization casts out impurities, and gathers neither soil nor stain in its beautiful textures. The sediment arises from abrasion of solid matters held in the under-surface of the glacier upon the rocks of its bed. In Fig. 10
