sun than she is now, the climate of the northern hemisphere would be very different from what it is.
One such period of great eccentricity occurred about two million five hundred thousand years ago. Fifty thousand years later there was another. Again, eight hundred and fifty thousand years ago there was a third, and two hundred thousand years ago a fourth. Those periods were characterized by cold such as we have no conception of. More than arctic winter lingered far on into the spring, and unmelted ice of one year accumulated through the next, till from the pole to the south of Scotland the earth was covered with a vast icecap, probably several miles in thickness.
Now, in Europe and America, wherever in fact any records are left of the glacial epoch, it is remarked that a general subsidence of the land followed closely on the appearance of the ice. This fact led certain geologists to conclude that there was some physical connection between the two phenomena, and Mr. Jamieson suggested to the Geological Society that the crust of the earth might have yielded under the enormous weight of the ice. Mr. Croll, however, gives a different explanation; and the more it is understood the more it appears to gain ground with those capable of forming an opinion. He says that the surface of the ocean always adjusts itself in relation to the earth's centre of gravity, no matter what the form of the earth happens to be. If a large portion of the water of the ocean were formed into solid ice, and placed round the north pole, its weight would naturally change the centre of gravity of the earth. The centre would be changed a little to the north of its former position. The water of the ocean would then forsake its old centre, and adjust itself with reference to the new. The surface of the ocean will therefore rise toward the north pole, and fall toward the south. The land will not sink under the sea, but, what amounts to the same thing, the sea will rise upon the land. The extent of submergence will be in proportion to the weight of the ice.
It is easy to see that glaciation would not be contemporaneous on both hemispheres. One hemisphere would be covered with ice and snow, while the other would be enjoying a perpetual spring. A glacial epoch resulting from the eccentricity of the earth's orbit would extend over a period of a hundred thousand years. But, for the reason given above, the glaciation would be transferred from one hemisphere to another every ten thousand years. A glacial epoch extending over a hundred thousand years would therefore be broken up into several warm periods. The warm period in one hemisphere would coincide with the cold one in the other, and there would be elevation of the land during the warm period and subsidence during the cold.
This cause would be quite sufficient to effect the alternate upheaval and depression. During the successive ages that each pole alternately was subjected to glaciation, the winter ice, unmelted by the brief sum-
