were abundant in Greece, and several of these appear to have been the ancestors of those now living in Africa; while two species of giraffes also inhabited Greece and North-West India. Equally suggestive is the occurrence in Europe of such birds as trogons and jungle-fowl characteristic of tropical Asia, along with parrots and plaintain-caters allied to forms now living in West Africa.

Let us now inquire what information Geology affords us of changes in land and sea at this period. From the prevalence of early tertiary deposits over the Sahara and over parts of Arabia, Persia, and Northern India, geologists are of opinion that a continuous sea or strait extended from the Bay of Bengal to the Atlantic Ocean, thus cutting off the Peninsula of India with Ceylon, as well as all tropical and South Africa from the great northern continent. At the same time, and down to a comparatively recent period, it is almost certain that Northern Africa was united to Spain and to Italy, while Asia Minor was united to Greece, thus reducing the Mediterranean to the condition of two inland seas. We also know that the north-western Himalayas and some of the high lands of Central Asia were at such a moderate elevation as to enjoy a climate as mild as that which prevailed in Central Europe during the Miocene epoch, and was therefore perhaps equally productive in animal and vegetable life.

Mr. Searles V. Wood, "On the Form and Distribution of the Land-tracts during the Secondary and Tertiary Periods respectively," Philosophical Magazine, 1862.

19 This part of the Himalayas was elevated during the Eocene period, and remains of a fossil Rhinoceros have been found at 16,000 feet elevation in Thibet.

We have, therefore, good evidence that the great Euro-Asiatic continent of Miocene times exhibited in its fauna a combination of all the main features which now characterise the Palearctic, Oriental, and Ethiopian regions; while tropical Africa, and such other tropical lands as were then, like the peninsula of India, detached and isolated from the continent, possessed a much more limited fauna, consisting for the most part of animals of a lower type, and which were more characteristic of Eocene or Secondary times. Many of these have no doubt become extinct, but they are probably represented by the remarkable and isolated lemurs of West Africa and Southern Asia, by the peculiar Insectivora of South Africa and Malaya, and by the Edentata of Africa and India. These are all low and ancient types, which were represented in Europe in the Eocene and early Miocene periods, at a time when the more highly specialised horses, giraffes, antelopes, deer, buffaloes, hippopotami, elephants, and anthropoid apes had not come into existence. And if these large herbivorous animals were all wanting in tropical Africa in Miocene times, we may be quite sure that the large felines and other carnivora which prey upon them were absent also. Lions, leopards, and hyænas can only exist where antelopes, deer, or some similar creatures abound; while smaller forms allied to the weasels and civets would be adapted to a country where small rodents or defenceless Edentata were the chief vegetable-feeding mammalia.

If this view is correct (and it is supported by a considerable amount of evidence which it is not possible here to adduce), all the great mammalia which now

so specially characteristic of Africa-the lions,

leopards, and hyænas,-the zebras, giraffes, buffaloes, and antelopes,-the elephants, rhinoceroses, and hippopotami, and perhaps even the numerous monkeys, baboons, and anthropoid apes,-are every one of them comparatively recent immigrants, who took possession of the country as soon as an elevation of the old Eocene and Miocene sea-bed afforded a passage from the southern borders of the Palearctic region. This event probably occurred about the middle of the Miocene period, and it must have effected a vast change in the fauna of Africa. A number of the smaller and more defenceless of the ancient inhabitants must have been soon exterminated, as surely as our introduced pigs, dogs, and goats, exterminate so many of the inhabitants of oceanic islands; while the new comers finding a country of immense extent, with a tropical climate, and not too much encumbered with forest vegetation, spread rapidly over it, and thenceforth, greatly multiplying, became more or less modified in accordance with the new conditions. We shall find that this theory not only accounts for the chief specialities, but also explains many of the remarkable deficiencies of the Ethiopian fauna. Thus, bears and deer are absent, because they are comparatively late developments, and were either unknown or rare in Europe till late Miocene or Pliocene times; while, on the other hand, the immense area of open tropical country in Africa has favoured the preservation of numerous types of large mammalia which have perished in the deteriorated climate and diminished area of Europe.

Our knowledge of the geology of Africa is not sufficiently detailed to enable us to determine its earlier

history with any approach to accuracy. It is clear, however, that Madagascar was once united with the southern portion of the Continent, but it is no less clear that its separation took place before the great irruption of large animals just described; for all these are wanting, while lemurs, insectivora, and civets abound,the same low types which were once the only inhabitants of the mainland. It is worthy of note, that south temperate Africa still exhibits a remarkable assemblage of peculiar forms of mammalia, birds, and insects,-the two former groups mostly of a low grade of organisation; and these, taken in connection with the wonderfully rich and highly specialised flora of the Cape of Good Hope, point to the former existence of an extensive south-temperate land in which so many peculiar types could have been developed. Whether this land was separated or not from Equatorial Africa, or formed with it one great southern continent, there is no sufficient evidence to determine.

Turning now to tropical Asia, we find a somewhat analogous series of events, but on a smaller scale and with less strongly-marked results. At the time when tropical and South Africa were so completely cut off from the great northern continent, the peninsula of India with Ceylon was also isolated; and it seems probable that their union with the continent took place at a somewhat later period. The ancient fauna of this south-Asiatic island may be represented by the slow Loris a peculiar type of lemurs, some peculiar rats (Murida), and perhaps by the Edentate scaly ant-eater; by its Uropeltidæ a peculiar family of snakes, and by many peculiar genera of snakes and lizards, and a few peculiar amphibia. On

the other hand, we must look upon the monkeys, the large carnivora, the deer, the antelopes, the wild pigs, and the elephants, as having overrun the country from the north; and their entrance must, no doubt, have led to the extermination of many of the lower types.

But there is another remarkable series of changes which have undoubtedly taken place in Eastern Asia in Tertiary times. There is such a close affinity between the animals of the Sunda Islands and those of the Malay Peninsula and Siam; and between those of Japan and of Northern Asia, that there can be little doubt that these islands once formed a southern and eastern extension of the Asiatic continent. The Philippines and Celebes perhaps also formed a part of this continent; but if so, the peculiarity and poverty of their mammalian fauna shows that they must have been separated at a much earlier period.1 The other islands probably

remained united to the continent till the Pliocene period. The result is seen in the similarity of the flora of Japan to that which prevailed in Europe in Miocene times; while in the larger Malay Islands we find, along with a rich flora developed under long-continued equatorial conditions of uniform heat and moisture, a remnant of the fauna which accompanied it, of which the Malay tapir, the anthropoid apes, the tupaias, the galeopitheci or flying lemurs, and the sun-bears, may be representatives.

There is another very curious set of relations worthy of our notice, because they imply some former com

For a full account of the evidence and conclusions as to these islands see the author's Geographical Distribution of Animals, vol. i. pp. 345, 359, 426, 436.