here southwards from North America, where their remains (or those of closely allied species) have been likewise found in the nearly contemporaneous formations of the United States.

Moreover, it would seem that in these bygone days, not only did the northern forms move southwards, but that also some of the southern forms emigrated northwards. This is evidenced by the fauna of the so-called "Equus-beds" and "Megalonyx-beds" of a slightly later date in the United States, which contain a composite Mammal - fauna of northern forms mixed with forms usually considered to be exclusively South American— such as the gigantic armadillo-like Glyptodon, the Capybara (Hydrochorus), Toxodon, and others.

Finally, in the age of the "Pampas" beds, the peculiar South American Mammal-fauna seems to have reached its culminating-point, and to have far exceeded that of the present day both in number of species and in the size of the individuals. This great increase in size, which is, as a rule, accompanied by an extreme specialisation of individual organs, seems to have had a fatal effect on its possessors, as none of the larger Edentates or Toxodonts appear to have outlived the end of the pampas formation. Along with most of the larger arrivals from the north, such as Mastodon and Equus, they became extinct. All the conclusions to be derived from this much-abbreviated account of the extinct Mammals of South America, confirm in a remarkable way the evidence of the present fauna as to this history of the Neotropical Region. Up to the last period of the Tertiary epoch, South America was certainly isolated from the rest of the world, and the connections with Australia and with Africa, if they ever did exist, must have been previous to this

MAP OF THE NEOTROPICAL REGION
SHOWING ITS DIVISION INTO 4 SUB-REGIONS.

period. At the beginning of Pliocene time, during the deposition of the Araucanian formation in Argentina and the "Equus-beds" in the United States, a wide bridge between North and South America, affording an easy road to migrating animals, must have existed, and this again seems to have become considerably narrowed to form the present Isthmus of Panama.

LIST OF THE PRINCIPAL AUTHORITIES REFERRED TO
IN CHAPTER III.

(1) ALLEN, J. A.-"The Geographical Distribution of North American Mammals." Bull. Amer. Mus. N.H., iv., p. 199, 1892.

(2) ALLEN, J. A.-"On a Small Collection of Mammals from the Galapagos Islands, collected by Dr. G. Baur."

N.H., iv., p. 47, 1892.

Bull. Amer. Mus.

(3) ALSTON, E. R.-"Biologia Centrali-Americana, Mammalia. London, 1879-82.

(4) CHAPMAN, F. M.-"Notes on Birds and Mammals observed near Trinidad, Cuba, with remarks on the Origin of West Indian Bird-life." Bull. Amer. Mus. N.H., iv., p. 279, 1892.

(5) COPE, E. D.-"Description of two large extinct Rodents from Anguilla, West Indies, with remains of human art associated." Proc Amer. Philos. Soc. Philad., xi., 1871, p. 183.

(6) LEIDY, J.-"Remarks on Mylodon." Proc. Acad. Nat. Sci. Philad., 1885, p. 49.

(7) SALVIN, O.—"On the Avifauna of the Galapagos Archipelago.' Trans. Zool. Soc., ix., p. 447 (1876).

(8) SCLATER, P. L.-"Address to Section D. (Biology)." " Report of Forty-fifth Meeting of the British Association, at Bristol," p. 85 (1876).

(9) THOMAS, O.-"On Some Mammals from Central Peru." P. Z. S., 1893, p. 333.

(10) TROUESSART, E. L.-" Note sur le rat musqué (Mus pilorides) des Antilles." Ann. Sci. Nat. Zool. (6), xix., No. 5 (1885).

(11) WALLACE, A. R.—“The Geographical Distribution of Animals." 2 vols. London, 1876.

(12) ZITTEL, K. A. VON.-"The Geological Development, Descent, and Distribution of the Mammalia." S. B. k. bayer. Akad. Wiss., xxiii., p. 137, 1893. Translation in Geol. Mag. (3), x., p. 401 (1893).

CHAPTER IV

THE ETHIOPIAN REGION

(PLATE IV., p. 122)

SECTION I-BOUNDARIES OF THE ETHIOPIAN REGION

THE Ethiopian Region (see Map, Plate IV.) contains the whole of Africa south of the Sahara, together with Southern Arabia and the island of Madagascar. As in all other cases where there is a long land-frontier between two neighbouring Regions, so here it is impossible to lay down anything but an approximate line of demarcation between the Ethiopian and Palearctic Regions.

The boundary usually adopted is the line of the Tropic of Cancer, which strikes Africa between Morocco and Senegambia, runs through the middle of the Sahara, crosses the Nile between the first and second cataracts, and passes through Arabia to the neighbourhood of Oman, on the Persian Gulf. Most of the country through which this line passes is desert, and its mammalian fauna is consequently meagre. Mr. O. Thomas (6) has recently published an account of a collection of mammals received at the British Museum from Oman, which shows, as would naturally be expected, that "the geographical relationships of this district are about equal with Africa and India; three of the species being distinctly African in affinities, three Indian, and the remainder either peculiar or widely spread and of no special significance." On the whole,