and African coasts. The Manatee could hardly live to cross the Atlantic. It is only found close to the coast, in estuaries and rivers, where it browses on sea-weeds and other vegetable food in shallow water. How did it travel from America to Africa (or vice versa), unless there were a continuous shore-line between them? The same may be said of the Monk-Seal (Monachus), of which one species lives in the Mediterranean and on the African coast and islands, and another in the West Indies. We can hardly believe that these creatures could easily traverse the whole Atlantic. The hypothesis of a former barrier of land between Africa and America, which we know is supported by other facts of distribution,1 would alone explain the difficulty.

On the other hand, in the Pacific we find no such break between the north and south. The aquatic mammals of Notopelagia have evidently had free access to the whole of the Pacific for a long period, and have well availed themselves of this facility.

Again, while the great Southern Ocean exhibits a considerable uniformity of marine mammalian life, we see the Northern waters divided into two distinctly recognizable Regions by the interposed masses of land. All these facts, with the one exception of the supposed Atlantic barrier, would tend in favour of the now generally accepted doctrine that the principal masses of land and water are not of modern origin, but have existed mainly in their present shapes throughout all ages.

1 Cf. Wallace, Geogr. Distrib., vol. i., p. 156.

CHAPTER IX

DISTRIBUTION OF MONKEYS AND LEMURS

SECTION I.-INTRODUCTORY REMARKS

HAVING now completed the discussion of the great Zoological Regions of terrestrial and marine mammals and their leading peculiarities, it is proposed to take the subject in another way, and considering the Orders, or great primary groups of mammals, one after the other, to sketch out the mode in which their leading forms are distributed over the surface of the world. We shall then see whether the conclusions thus arrived at appear to lead to similar results to those attained by making geographical divisions our primary subject of study. In doing this it will be more convenient to take the most highly organized groups of mammals first, and to descend gradually to the lowest, reversing the arrangement used in the previous chapters of this work. The divisions and names now employed are taken from the last edition of the "List of Vertebrated Animals in the Zoological Society's Gardens," (1896), but do not differ materially from those used in Flower and Lydekker's "Mammals Living and Extinct," which we have hitherto mainly followed. These

divisions, commonly called "Orders," are fourteen in

The Order of Quadrumana, or Monkeys, of which about 200 species are now recognized, is generally divided by zoologists into four families: (1) the Simiidæ or Apes, (2) the Cercopithecida or Old-World Monkeys, (3) the Cebida or New-World Monkeys, and (4) the Hapalidæ or Marmosets. Of these families the two first are exclusively inhabitants of the tropical or sub-tropical districts of the Old World, while the two last are as severely restricted to the hotter portion of the New World, and form, in fact, two of the most characteristic groups of the Neotropical Region. We thus see that the division of the Quadrumana into families according to their structure is in complete accordance with the distinctness of the geographical areas in which they are found.

The two first and highest families of Monkeys, it should be remarked, are much more nearly allied to one another than they are to the two families which inhabit the New World. They are sometimes called "Catarrhines," on account of the narrowness of the nasal septum and the consequent downward direction of the nostrils. They all