Page:Walden, or, Life in the Woods.djvu/327

of Walden on the 6th of March, 1847, stood at 32°, or freezing point; near the shore at 33°; in the middle of Flints' Pond, the same day, at 32½°; at a dozen rods from the shore, in shallow water, under ice a foot thick, at 36°. This difference of three and a half degrees between the temperature of the deep water and the shallow in the latter pond, and the fact that a great proportion of it is comparatively shallow, show why it should break up so much sooner than Walden. The ice in the shallowest part was at this time several inches thinner than in the middle. In mid-winter the middle had been the warmest and the ice thinnest there. So, also, every one who has waded about the shores of a pond in summer must have perceived how much warmer the water is close to the shore, where only three or four inches deep, than a little distance out, and on the surface where it is deep, than near the bottom. In spring the sun not only exerts an influence through the increased temperature of the air and earth, but its heat passes through ice a foot or more thick, and is reflected from the bottom in shallow water, and so also warms the water and melts the under side of the ice, at the same time that it is melting it more directly above, making it uneven, and causing the air bubbles which it contains to extend themselves upward and downward until it is completely honey-combed, and at last disappears suddenly in a single spring rain. Ice has its grain as well as wood, and when a cake begins to rot or "comb," that is, assume the appearance of honey-comb, whatever may be its position, the air cells are at right angles with what was the water surface. Where there is a rock or a log rising near to the surface the ice over it is much thinner, and is frequently quite dissolved by this reflected heat; and