ANIMAL.] DISTRIBUTION 281 Whales and Seals ; Mr Andrew Murray's Geographical Distribution of the Mammalia.) General Relations of Marine with Terrestrial Zoological Regions. The general facts of distribution of marine animals now adduced accord very well with what we know of those terrestrial changes which have led to the actual distribution of land animals. The great Indo-Pacific region - so well marked in every important group of marine animals - probably owes its individuality to the fact that Australia has been isolated during the whole of the Tertiary, and probably during much of the Mesozoic epoch, while numerous islands in the Indian and Pacific oceans have always afforded an extensive shore-line favourable to the development of aquatic forms of life. The Atlantic has probably been for long periods even more inclosed than it is now, owing to the greater southward extension of South Africa and South America ; while the profound depths of its central channel have served as a barrier between the inhabitants of the shallow waters of its eastern and western shores. In like manner the great trough of deep water which separates the most eastern groups of the Pacific islands from the west coast of America has necessarily led to the establishment of distinct oceanic faunas in these regions ; while this very fact - the remarkable distinctness of the Pacific and West American faunas - tells us plainly that this barrier of deep ocean is one of the ancient features of the earth's surface. We shall find, too, that many of the details, and not a few of the anomalies, of the distribution of marine animals become intelligible from our knowledge of past geographical changes. The considerable affinity between the Crustacea, Mollusca, and fishes of the eastern and western coasts of America exactly corresponds to the fact, clearly established by a consideration of the distribution of living and extinct land animals, that these oceans have been united, at several distinct periods, by two or more channels over what is now Central America, the final union of the two continents being comparatively recent. The fact that the uniting channels were always situated within the same limited area sufficiently explains the considerable amount of generic and specific difference of two faunas ranging over coast-lines running north and south for many thousand miles on the opposite sides of great continents. The curious fact (only recently established) that so deep and extensive an inland sea as the Mediterranean contains but few peculiar marine animals, becomes quite intelligible when we consider that till middle or late Tertiary times it consisted of two inland seas or lakes. Such inland seas are always very poor in animal life ; and it is therefore not surprising that the Mediterranean should now contain hardly any forms but such as it has received from the Atlantic, or from the Red Sea during a submergence of the Isthmus of Suez. The numerous allied or even identical forma in the northern and southern oceans, which are not found in the interven ing warm regions, are more difficult to explain. Mr Darwin believes that such facts are due to the action of the glacial period, which at its height may have cooled certain tracts of the tropical ocean sufficiently to allow temperate forms to cross from the northern to the southern hemi sphere or the reverse. Perhaps, however, the agency of icebergs may have been sufficient without any permanent cooling of the equatorial ocean ; for even now these huge floating glaciers often reach to 40 N. lat. and 35 S., and, Captain Maury assures us, sometimes even reach tbe tropics. We may therefore well suppose that during the height of the glacial period icebergs would not only regularly reach the tropics, but, carried on by currents in definite lines, might often pass across the equatorial zone, carrying with them a girdle of cold water in which many inhabitants of the Arctic or Antarctic seas might safely make the passage to another hemisphere. The fact that many forms of plants peculiar to cold or temperate regions are found scattered on isolated mountain summits in the tropics is, as Mr Darwin has shown, to be explained only by the influence of an extreme glacial period, and this must have produced analogous effects on the inhabitants of the ocean. (Origin of Species, 6th ed. p. 330.) DISTRIBUTION OF ANIMALS IN TIME. This subject will necessarily be treated in some detail under the articles GEOLOGY and PALAEONTOLOGY. Here v/e shall only sketch its outlines and general principles. The past history of living things as revealed by geology is an ever-changing panorama. At each successive stage some forms disappear, while new ones take their place. The farther we go back the more unlike is the general assemblage of animals and plants to that which now exists. If we confine our attention to any one class or order of animals, we find that it makes its first appearance at some definite epoch, and, under successively changing forms, either continues till the present time, or reaches a maximum, diminishes, and finally disappears. Thus some gi oups are altogether modern, others extremely ancient; some have run through all their phases in a comparatively short period, others have lived from the earliest epochs of the earth s history of which we have any record and still survive. If we could be sure that the numerous fossils yet discovered gave us anything like an adequate idea of all the varied forms of life that had ever lived upon the globe, and the order in which they had appeared, we should be in a position to decide as to the truth and value of the development hypothesis. But the more we examine the question, the more certain it becomes that the " geological record, " as it is termed, is extremely imperfect, and that the whole of the extinct animals which we have discovered do not forni any fair representation of the entire series that have lived upon the earth. This is the case even with the more recent deposits and^ those which are richest in animal remains ; but as we go back into the past the record becomes more and more imperfect, till in the Secondary, and still more in the Palaeozoic formations, we only have preserved to us a few scattered fragments, equivalent perhaps to a few pages with here and there a short chapter taken at random out of a voluminous history. The causes of this necessarily imperfect record of the past have beea fully discussed by Sir Charles Lyell and Mr Darwin ; we need only refer here to two general causes of such imperfection. The first is, that every aqueous deposit is formed by the wearing down of previous deposits, so that the records of one age are, to a large extent, necessarily destroyed to provide the records of the next, which in its turn is destroyed in a succeeding age. The other cause of imperfection is, that extensive areas are always sinking (to allow new deposits to be formed over them), and are being subjected to subterranean heat to such an extent as to change their texture and obliterate their fossils, when they become crystalline or metamorphic rocks. The more recent deposits so acted on will rarely have had time to have become raised above the sea-level, and subsequently exposed by denudation ; yet certain Eocene strata in the Alps are stated by Sir C. Lyell to be truly metamorphic (Students Elements of Geology, p. 600). The older a formation is, therefore, the more frequently will it have been exposed in one area or another to this metamorphic action ; and it follows that, going back ward in time, we shall at last come to a period, all the formations antecedent to which will have become, metamor phosed, and their fossils, if any, obliterated. We appear to have almost reached such a state of things at the base of the Palaeozoic rocks ; and there is good reason to believe
VII. - 36