There is another branch of the subject that I cannot so easily leave. It is the answer to the question, What became of the many peculiar tropical American genera of animals and plants, when a great part of the tropics was covered with ice, and the climate of the lower lands much colder than now? For instance, the Heliconii and Morphos are a group of butterflies peculiar to tropical America, containing many distinct genera which, on any theory of descent from a common progenitor, must have originated ages before the glacial period. How is it that such peculiarly tropical groups were not exterminated by the cold of the glacial period, or if able to stand the cold, that they did not spread into temperate regions on the retreat of the ice? I believe the answer is, that there was much extermination during the glacial period, that many species and some genera, as, for instance, the American horse, did not survive it, and that some of the great gaps that now exist in natural history were then made; but that a refuge was found for many species, on lands now below the ocean, that were uncovered by the lowering of the sea caused by the immense quantity of water that was locked up in frozen masses on the land. Mr. Alfred Tylor considers that the ice cap of the glacial period was the cause of a great reduction of the level of the sea, amounting to at least 600 feet.* (* "Geological Magazine" volume 9 page 392.) But if we admit that the ice existed in both hemispheres at the same time, we shall have to speculate on a lowering of the level of the sea to at least 1000 feet. We have many facts tending to prove that during the extreme extent of the glacial period the land stood much higher relatively to the sea than it now does. Professor Hartt believes that during the time of the drift, Brazil stood at a much higher level than at present,* (* "Geology and Physical Geography of Brazil" by Ch. Fred. Hartt page 573.) and we can, on the supposition of a general lowering of the sea all over the world, account for the distribution of animal life over islands now separated by shallow seas. Thus Mr. Bland, in a paper read before the American Philosophical Society, on "The Geology and Physical Geography of the West Indies, with reference to the distribution of Mollusca," states his opinion that Porto Rico, the Virgins, the Anguilla group, Cuba, the Bahamas, and Hayti, once formed continuous dry land that obtained its land molluscs from Central America and Mexico. The land molluscs of the islands to the south, on the contrary, from Barbuda and St. Kitt's down to Trinidad, are of two types, one Venezuelan, the other Guianian; the western side of the supposed continuous land, namely, Trinidad, Tobago, Grenada, the Grenadines, St. Vincent, and St. Lucia, belonging to the first type; the eastern side, from Barbados to Antigua, to the second.* (* Quoted in "At Last" by Charles Kingsley page 305.)