passing from the island of Bali to that of Lombock, where the two regions are in closest proximity. In Bali we have barbets, fruit thrushes, and woodpeckers; on passing over to Lombock these are seen no more, but we have abundance of cockatoos, honeysuckers, and brush-turkeys, which are equally unknown in Bali or in any island further west. The strait is here fifteen miles wide, so that we may pass in two hours from one great division of the earth to another, differing as essentially in their animal life as Europe does from America.* It we travel from Java or Borneo to Celebes or the Moluccas, the difference is still more striking. In the first, the forests abound in monkeys of many kinds, wild cats, deer, civets and others, and numerous varieties of squirrels are constantly met with. In the latter, none of these occur, but the prehensile-tailed cuscus is almost the only terrestrial mammal seen, except wild pigs, which are found in all the islands, and deer (which have probably been recently introduced) in the Celebes and the Moluccas. The birds which are most abundant in the Western islands are woodpeckers, barbets, trogons, fruit-thrushes, and leaf-thrushes; they are seen daily, and form the great ornithological features of the country. In the Eastern islands these are absolutely unknown, honeysuckers and small lories being the most common birds; so that the naturalist feels himself in a new world, and can hardly realize that he has passed from the one region to the other in a few days, without ever being out of sight of land.

"The inference that we must draw from these facts is undoubtedly that the whole of the islands eastwards,

* This is too vaguely expressed. It would be nearer the inark to say, as Europe does from South America. (O. SCHMIDT.)

beyond Java and Borneo, do essentially form a part of a former Australian or Pacific continent, although some of them may never have been actually joined to it. This continent must have been broken up not only before the Western islands were separated from Asia, but probably before the extreme south-eastern portion of Asia was raised above the waters of the ocean; for a great part of the land of Borneo and Java is known to be geologically of quite recent formation; while the very great difference of species, and in many cases of genera also, between the productions of the Eastern Malay islands and Australia, as well as the great depth of the sea now separating them, all point to a comparatively long period of isolation."

"It is interesting to observe among the islands themselves how a shallow sea always intimates a recent land connection. The Aru islands, Maisol and Waigiou, as well as Jobic, agree with New Guinea in their species of mammalia and birds much more closely than they do with the Moluccas, and we find that they are all united to New Guinea by a shallow sea. In fact, the 100-fathom line round New Guinea marks out accurately the range of the true Paradise birds.

"It is further to be noted-and this is a very interesting point in connection with theories of the dependence of special forms of life on external conditions—that this division of the Archipelago into two regions characterized by a striking diversity in their natural productions, does not in any way correspond to the main physical or climatal divisions of the surface." We will further quote only the following: "Borneo and New Guinea, as alike physically as two distinct countries can be, are

zoologically wide as the poles asunder; while Australia, with its dry winds, its open plains, its stony deserts, and its temperate climate, yet produces birds and quadrupeds which are closely related to those inhabiting the hot, damp, luxuriant forests which everywhere clothe the plains and mountains of New Guinea."

Wallace gives the most specific proofs that, as the parts of this Archipelago approach one another like separated extremities of two continents, they bring with them two entirely different fauna. Similarly, the Mediterranean and West Indian Archipelagos are devoid of any peculiar character, and are completely dependent on the adjacent continents for their animal life and vegetation. We have already discussed Madeira and its land snails. Insular faunas therefore do not require the hypothesis of more centres of creation than are offered by the continents; and Rütimeyer has endeavoured to trace the extraction of birds and mammals to two centres of derivation. A great series of animal-geographical facts is explicable only on the hypothesis of the former existence of a southern continent, of which the Australian mainland is a remnant.

The present Marsu-
Their occurrence

pials are concentrated in Australia. in the south-western portion of the Malay Archipelago, including New Guinea, seems like a radiation from that centre. No single token makes it appear that the Marsupials existing in former periods in the northern hemisphere, from the Jura forwards, had migrated to meet those which were pressing on from the southern continent towards the equator. Only as to the opossum, so widely extended in South America, could a question arise, which is however solved by the examination of a

host of congeners, one and all alien to the population predominant in America, and indicating importation probably in the Tertiary period; unless it be assumed, with Rütimeyer, "that implacental mammals were created out of Australia as well as in it."

Among the first to be mentioned are the wingless birds, that is, those which are anatomically and systematically connected, and which we now find scattered over continents and some of the larger islands. The cassowary of New Holland and America, the extinct giant birds of Madagascar and New Zealand, the African ostrich, which has advanced from the south northwards, cannot have originated in their present isolation. The same considerations are forced upon us by the mammals named Bruta by Linnæus, and by modern zoologists termed Edentata, by reason of their imperfect dentition, among which, accepting the latter definition, must be included the Ornithorhyncus, or duck-mole of Tasmania. These duck-moles incontestibly occupy the lowest grade among the mammals now extant; but the other true Edentata are no less alien to the higher orders, and their occurrence in South America on the one hand, and in South Africa and South Asia on the other, as well as the impossibility of tracing them from a common centre in the northern hemisphere, points to the vanished land of the south, where perhaps the home of the progenitors of the Maki of Madagascar may also be looked for.

Or," says Rütimeyer, " does the hypothesis of a Polar land, once possessing an abundance of animal life, partly covered by the ocean and partly by a coat of ice, appear an unfounded assumption to us who now witness the

elevation of a similar frozen surface in the northern hemisphere, and are surrounded in the Alps by a still existing-in our glacial drift by a scarcely vanishedarctic scene? Or need the conjecture that the almost exclusively graminivorous and insectivorous Marsupials, sloths, armadilloes, ant-eaters, and ostriches, once possessed an actual point of union in a southern continent, of which the present flora of Terra del Fuego, the Cape, and Australia, must be the remains,-need this conjecture raise difficulties at a moment when from their fossil remains Heer restores to our sight the ancient forests of Smith's Sound and Spitzbergen?"

Having ventured to reconstruct the southern continent, with its strange fauna, of which the remains are so widely dispersed, Rütimeyer casts about for more specific evidence in favour of the hypothesis to which the course of the world's formation everywhere gives rise, that fresh-water animals and likewise terrestrial animals came up from the sea. Hence the notably small division of sirenoid fish (Lepidosiren, Protopterus), which breathe air during the dry season of the year, must not be considered reptiles adapting themselves to aquatic life, but the reverse. The organ which in fish served as a hydrostatic apparatus, the swim bladders, becomes in them the lung. Thus we must go back from terrestrial to aquatic tortoises, and from them to those denizens of the sea which are allied to the Enaliosaurians, so frequent in the Jurassic strata. The evolutionary and biographical history of the land crabs shows us in the plainest manner how the inhabitant of the sea becomes a terrestrial animal; a special problem which, as we have already mentioned, Fritz Müller has