The reader will begin to think that we have sufficiently "explained" the Permian revolution. Far from it. Some of its problems are as yet insoluble. We have given no explanation at all why the ice-sheets, which we would in a general way be prepared to expect, appear in India and Australia, instead of farther north and south. Professor Chamberlin, in a profound study of the period (appendix to vol. ii, "Geology"), suggests that the new land from New Zealand to Antarctica may have diverted the currents (sea and air) up the Indian Ocean, and caused a low atmospheric pressure, much precipitation of moisture, and perpetual canopies of clouds to shield the ice from the sun. Since the outer polar regions themselves had been semi-tropical up to that time, it is very difficult to see how this will account for a freezing temperature in such latitudes as Australia and India. There does not seem to have been any ice at the Poles up to that time, or for ages afterwards, so that currents from the polar regions would be very different from what they are today. If, on the other hand, we may suppose that the rise of "Gondwana Land" (from Brazil to India) was attended by the formation of high mountains in those latitudes, we have the basis, at least, of a more plausible explanation. Professor Chamberlin rejects this supposition on the ground that the traces of ice-action are at or near the sea-level, since we find with them beds containing marine fossils. But this only shows, at the most, that the terminations of the glaciers reached the sea. We know nothing of the height of the land from which they started.