level of the water is not more than two inches. A line of clear, blue ice marks the boundary of each layer, produced, in the opinion of various writers, by the melting of the top portion of the snow-fall of a previous winter, while the white part represents the unmelted part of the snow-fall. Wilkes estimates the snow-fall in this region at thirty feet in the year, and, as only a very small part of this can be melted in the course of the short summer, an immense accumulation must go on. The small part of each year's fall that is melted will be clear ice, and this is represented by the line of blue in the berg; so that, by counting the number of layers in a berg, some idea can be had of its age.
Sir Wyville Thomson says that the reduction in thickness of the layers from top to bottom is due mostly to compression, and estimates that at a depth of 1,400 feet enough heat will be generated to melt the ice. On the other hand, Croll contends that the temperature of the bottom of the immense icebergs, which sometimes tower 700 and 1,000 feet above the water, is 20° or 22° Fahr., and when we remember that ice floats with from six to seven times its height below water, and that the bottom of a 1,000-foot berg would be 6,000 or 7,000 feet below the surface, it is easily seen that no melting would occur on the bottom of an ice-sheet only 1,400 feet thick. The reason assigned by Croll for the gradual thinning of the ice-layers toward the bottom is, that, instead of being due solely to compression, it is due mainly to what he terms dispersion. In other words, if, at 85° south, a mass of ice covers one square foot of surface, it will, in its gradual passage north, be made to cover two square feet at 80° latitude; at 70° it will occupy four square feet, and, at 60° south, six square feet; that is to say, a stratum which was one foot thick in 85° latitude will be only two inches thick when it has reached 60° latitude.
The discharge of icebergs from the extremity of the ice-field takes place constantly, and they are sometimes huge. Croll has collected notices of bergs which towered 400, 580, 720, 960, and 1,000 feet above the sea-level, and, as six or seven times the bulk above water floats below, some adequate idea can be formed of the contents of a berg three, four, or five miles in length. As the ice-field must have an onward motion to cause this discharge, it has been calculated that one foot in 211 is the smallest slope which will be effectual, and that the ice moves here at the rate of about one quarter of a mile per annum. As it is impossible that thirty or forty feet of snow can be melted in a single short summer's season, it follows that there must be a rapid accumulation all the time. This accumulation will probably be greatest at the pole, and Croll estimates that here the ice has attained a thickness of about seven miles. Allowing this estimate to be correct, and it does not seem excessive, when it is remembered that the thickness of the ice-sheet over Northern New England, during the Glacial period, was 6,000 feet, the question arises, Upon what sort
