But Mr. A. R. Wallace, who has most recently written on the subject, is of opinion that the original division proposed by Mr. Sclater in 1857 is the most tenable, and he therefore adopts it in the very exhaustive work upon the geographical distribution of animals which he has recently issued. Mr. Sclater's Six Regions are as follows:-- 1.--_The Palæarctic Region_, including Europe, Temperate Asia, and North Africa to the Atlas mountains. 2.--_The Ethiopian Region_, Africa south of the Atlas, Madagascar, and the Mascarene islands, with Southern Arabia. 3.--_The Indian Region_, including India south of the Himalayas, to South China, and to Borneo and Java. 4.--_The Australian Region_, including Celebes and Lombok, Eastward to Australia and the Pacific islands. 5.--_The Nearctic Region_, including Greenland, and North America, to Northern Mexico. 6.--_The Neotropical Region_, including South America, the Antilles, and Southern Mexico. This arrangement is based upon a detailed examination of the chief genera and families of birds, and also very nearly represents the distribution of mammals and of reptiles. Its regions are not, as in other subsequently proposed and more artificial systems, controlled by climate; for they range, in some instances, from the pole to the tropics. It probably approaches more nearly than any other yet proposed to that desideratum, a division of the earth into regions, founded on a collation of the groups of forms indigenous to or typical of them, and upon a selection of those peculiar to them; with a disregard of, or only admitting with caution, any which, though common to and apparently establishing connection between two or more regions, may have in fact but little value for the purpose of such comparison; from the fact of its being possible to account for their extended range by their capability of easy transport from one region to another by common natural agencies.[41] Such an arrangement should be consistent with the retrospective information afforded by palæontology; and, taking an extended view of the subject, be not merely a catalogue of the present, but also an index of the past. It should afford an illustration of an existing phase of the distribution of animal life, considered as the last of a long series of similar phases which have successively resulted from changes in the disposition of land and water, and from other controlling agencies, throughout all time. A reconstruction of the areas respectively occupied by the sea and the land at different geological periods will be possible, or at least greatly facilitated, when a complete system of similar groupings, illustrative of each successive period, has been compiled.