At first geologists were disposed to attribute all the phenomena of mountain-folding to the progressive cooling of the earth. Although this sphere has already lost a large part of the heat with which it was in the beginning endowed, it is still very hot in its deeper parts, as is shown by the phenomena of volcanoes. This internal heat, which to the present day at the depth of a hundred miles below the surface is probably greater than that of molten iron, is constantly flowing away into space; probably enough of it goes away on the average each day to melt a hundred cubic miles or more of ice, or, in more scientific phrase, the amount of heat rendered latent by melting that volume of frozen water. J.R. Meyer, an eminent physicist, estimated the quantity of heat so escaping each day of the year to be sufficient to melt two hundred and forty cubic miles of ice. The effect of this loss of heat is constantly to shrink the volume of the earth; it has, indeed, been estimated that the sphere on this account contracts on the average to the amount of some inches each thousand years. For the reason that almost all this heat goes from the depths of the earth, the cool outer portion losing no considerable part of it, the contraction that is brought about affects the interior portions of the sphere alone. The inner mass constantly shrinking as it loses heat, the outer, cold part is by its weight forced to settle down, and can only accomplish this result by wrinkling. An analogous action may be seen where an apple or a potato becomes dried; in this case the hard outer rind is forced to wrinkle, because, losing no water, it does not diminish in its extent, and can only accommodate itself to the interior by a wrinkling process. In one case it is water which escapes, in the other heat; but in both contraction of the part which suffers the loss leads to the folding of the outside of the spheroid. Although this loss of heat on the part of the earth accounts in some measure for the development of mountains, it is not of itself sufficient to explain the phenomena, and this for the reason that mountains appear in no case to develop on the floors of the wide sea. The average depth of the ocean is only fifteen thousand feet, while there are hundreds, if not thousands, of mountain crests which exceed that height above the sea. Therefore if mountains grew on the sea floor as they do upon the land, there should be thousands of peaks rising above the plain of the waters, while, in fact, all of the islands except those near the shores of continents are of volcanic origin--that is, are lands of totally different nature.