A NIGHT ENCAMPMENT, 1 93 formed the island, to ascertain, if possible, something of its structure. A little bank, produced by a landslip, enabled him to step down to the level of the sea, and from there he was able to look closely at the steep wall which formed the coast. Where he stood the soil rose scarcely three feet above the water. The upper part consisted of a thin layer of earth and sand mixed with crushed shells ; and the lower of hard, compact, and, if we may so express it, " metallic" ice, strong enough to support the upper soil of the island. This layer of ice was not more than one foot above the sea- level. In consequence of the recent fracture, it was easy to see the regular disposition of the sheets of ice piled up horizontally, and which had evidently been produced by successive frosts in comparatively quieter waters. We know that freezing commences on the surface of liquids, and as the cold increases, the thickness of the crust becomes greater, the solidification proceeding from the top downwards. That at least is< the case in waters that are at rest ; it has, however, been observed that the very reverse is the case in running waters — the ice forming at the bottom, and subsequently rising to the surface. It was evident, then, that the floe which formed the foundation of Victoria Island had be^n formed in calm waters on the shores of the North American continent. The freezing had evidently commenced on the surface, and the thaw would begin at the bottom, according to a well-known law; so that the ice-field would gradually decrease in weight as it became thawed by the warmer waters through which it was passing, and the general level of the island would sink in proportion. This was the great danger. As we have just stated, Hobson noticed that the solid ice, the ice- field properly so called, was only about one foot above the sea-level ! We know that four-fifths of a floating mass of ice are always sub- merged. For one foot of an iceberg or ice-field above the water, there are four below it. It must, however, be remarked that the density, or rather specific weight of floating ice, varies considerably according to its mode of formation or origin. The ice-masses which proceed from- sea water, porous, opaque, and tinged with blue or green, according as they are struck by the rays of the sun, are lighter than ice formed from fresh water. All things considered, and making due allowance for the weight of the mineral and vegetable layer above the ice, Hobson concluded it to be about four N
