6thly. A portion of the islands of the Indian archipelago might, perhaps, be considered by some geologists as an appendage of the same province. In fact, we find in the large islands of Java, Sumatra, and Borneo, the same genera, for the most part, as on the continent of India, and some of the same species, e. g. the tapir (Tapirus Malayanus), the rhinoceros of Sumatra, and some others. Most of the species, however, are distinct, and each island has many, and even a few genera, peculiar to itself. Between eighty and ninety species are known to inhabit Java, and nearly the same number occur in Sumatra. Of these, more than half are common to the two islands. Borneo, which is much less explored, has yielded already upwards of sixty species, more than half of which are met with either in Java or Sumatra. Of the species inhabiting Sumatra and not found in Java, Borneo contains the greater portion. Upon the whole, if these three large islands were united, and a fusion of their respective indigenous mammalia should take place, they would present a fauna related to that of continental India, and comprising about as many species as we might expect from analogy to discover in an area of equal extent. The Philippine Islands are peopled with another assemblage of species generically related to the great Indian type.
7thly. But the islands of Celebes, Amboina, Timor, and New Guinea, constitute a different region of mammalia more allied to the Australian type, as having an intermixture of marsupial quadrupeds, yet showing an affinity also to the Indian in such forms as the deer (Cervus), the weasel (Viverra), the pig (Sus), the Macaque monkey (Cercopithecus), and others. As we proceed in a south-westerly direction, from Celebes to Amboina and thence to New Guinea, we find the Indian types diminishing in number, and the Australian (i. e. marsupial forms) increasing. Thus in New Guinea seven species of pouched quadrupeds have been detected, and among them two singular tree-kangaroos; yet only one species of the whole seven, viz. the flying opossum (Petauris ariel), is common to the Indian archipelago and the main land of Australia. The greater the zoological affinity, therefore, between the latter and the New Guinea fauna, although it seems in some way connected with geographical proximity, is not to be explained simply by the mutual migration of species from the one to the other.
8thly. When Australia was discovered, its land quadrupeds, belonging almost exclusively to the marsupial or pouched tribe, such as the kangaroos, wombats, flying opossums, kangaroo-rats, and others, some feeding on herbs and fruits, others carnivorous, were so novel in their structure and aspect, that they appeared to the naturalist almost as strange as if they were the inhabitants of some other planet. We learn from the recent investigations of Mr. Waterhouse,[875] that no less than 170 species of marsupial quadrupeds have now been determined, and of the whole number all but thirty-two are exclusively restricted to Australia. Of these thirty-two, nine belong to the islands in the Indian archipelago before mentioned, and the other twenty-three are all species of opossum inhabiting the tropical parts of South America, or a few of them extending into Mexico and California, and one, the Virginian opossum, into the United States.
9thly. It only remains for me to say something of the mammiferous fauna of North and South America. It has often been said that, where the three continents of Asia, Europe, and North America, approach very near to each other towards the pole, the whole arctic region forms one zoological and botanical province. The narrow straits which separate the old and new world are frozen over in winter, and the distance is farther lessened by intervening islands. Many plants and animals of various classes have accordingly spread over all the arctic lands, being sometimes carried in the same manner as the polar bear, when it is drifted on floating ice from Greenland to Iceland. But on a close inspection of the arctic mammalia, it has been found of late years that a very small number of the American species are identical with those of Europe or Asia. The genera are, in great part, the same or nearly allied; but the species are rarely identical, and are often very unlike, as in the case of the American badger and that of Europe. Some of the genera of arctic America, such as the musk ox (Ovibos), are quite peculiar, and the distinctness of the fauna of the great continents goes on increasing in proportion as we trace them southwards, or as they recede farther from each other, and become more and more separated by the ocean. At length we find that the three groups of tropical mammalia, belonging severally to America, Africa, and India, have not a single species in common.
The predominant influence of climate over all the other causes which limit the range of species in the mammalia is perhaps nowhere so conspicuously displayed as in North America. The arctic fauna, so admirably described by Sir John Richardson, has scarcely any species in common with the fauna of the state of New York, which is 600 miles farther south, and comprises about forty distinct mammifers. If again we travel farther south about 600 miles, and enter another zone, running east and west, in South Carolina, Georgia, Alabama, and the contiguous states, we again meet with a new assemblage of land quadrupeds, and this again differs from the fauna of Texas, where frosts are unknown. It will be observed that on this continent there are no great geographical barriers running east and west, such as high snow-clad mountains, barren deserts, or wide arms of the sea, capable of checking the free migration of species from north to south. But notwithstanding the distinctness of those zones of indigenous mammalia, there are some species, such as the buffalo (Bison Americanus), the racoon (Procyon lotor), and the Virginian opossum (Didelphis Virginiana), which have a wider habitation, ranging almost from Canada to the Gulf of Mexico; but they form exceptions to the general rule. The opossum of Texas (Didelphis carnivora) is different from that of Virginia, and other species of the same genus inhabit westward of the Rocky Mountains, in California, for example, where almost all the mammalia differ from those of the United States.
10thly. The West Indian land quadrupeds are not numerous, but several of them are peculiar; and 11thly, South America is the most distinct, with the exception of Australia, of all the provinces into which the mammalia can be classed geographically. The various genera of monkeys, for example, belong to the family Platyrrhini, a large natural division of the quadrumana, so named from their widely separated nostrils. They have a peculiar dentition, and many of them prehensile tails, and are entirely unknown in other quarters of the globe. The sloths and armadillos, the true blood-sucking bats or vampyres (Phyllostomidæ), the capybara, the largest of the rodents, the carnivorous coatimondi (Nasua), and a great many other forms, are also exclusively characteristic of South America.
"In Peru and Chili," says Humboldt, "the region of the grasses, which is at an elevation of from 12,300 to 15,400 feet, is inhabited by crowds of lama, guanaco, and alpaca. These quadrupeds, which here represent the genus camel of the ancient continent, have not extended themselves either to Brazil or Mexico; because, during their journey, they must necessarily have descended into regions that were too hot for them."[876] In this passage it will be seen that the doctrine of "specific centres" is tacitly assumed.
Quadrupeds in Islands.—Islands remote from continents, especially those of small size, are either destitute of quadrupeds, except such as have been conveyed to them by man, or contain species peculiar to them. In the Galapagos archipelago no indigenous quadrupeds were found except one mouse, which is supposed to be distinct from any hitherto found elsewhere. A peculiar species of fox is indigenous in the Falkland Islands, and a rat in New Zealand, which last country, notwithstanding its magnitude, is destitute of other mammalia, except bats, and these, says Dr. Prichard, may have made their way along the chain of islands which extend from the shores of New Guinea far into the Southern Pacific. The same author remarks, that among the various groups of fertile islands in the Pacific, no quadrupeds have been met with except the rat and a few bats as above mentioned, and the dog and hog, which appear to have been conveyed thither by the natives from New Guinea. "Rats are to be found even on some desert islands, whither they may have been conveyed by canoes which have occasionally approached the shore. It is known, also, that rats occasionally swim in large numbers to considerable distances."[877]
Geographical range of the Cetacea.—It is natural to suppose that the geographical range of the different species of Cetacea should be less correctly ascertained than that of the terrestrial mammifers. It is, however, well known that the whales which are obtained by our fishers in the South Seas are distinct from those of the North; and the same dissimilarity has been found in all the other marine animals, of the same class, so far as they have yet been studied by naturalists.
Dispersion of quadrupeds.—Let us now inquire what facilities the various land quadrupeds enjoy of spreading themselves over the surface of the earth. In the first place, as their numbers multiply, all of them, whether they feed on plants, or prey on other animals, are disposed to scatter themselves gradually over as wide an area as is accessible to them. But before they have extended their migrations over a large space, they are usually arrested either by the sea, or a zone of uncongenial climate, or some lofty and unbroken chain of mountains, or a tract already occupied by a hostile and more powerful species.