Island Life/XIX

CHAPTER XIX

ANCIENT CONTINENTAL ISLANDS: THE MADAGASCAR GROUP


 * Remarks on Ancient Continental Islands—Physical Features of Madagascar—Biological Features of Madagascar—Mammalia—Reptiles—Relation of Madagascar to Africa—Early History of Africa and Madagascar—Anomalies of Distribution and How to Explain Them—The Birds of Madagascar as Indicating a Supposed Lemurian Continent—Submerged Islands between Madagascar and India—Concluding Remarks on "Lemuria"—The Mascarene Islands—The Comoro Islands—The Seychelles Archipelago—Birds of the Seychelles—Reptiles and Amphibia—Freshwater Fishes—Land Shells—Mauritius, Bourbon, and Rodriguez—Birds—Extinct Birds and their Probable Origin—Reptiles—Flora of Madagascar and the Mascarene Islands—Curious Relations of Mascarene Plants—Endemic Genera of Mauritius and Seychelles—Fragmentary Character of the Mascarene Flora—Flora of Madagascar Allied to that of South Africa—Preponderance of Ferns in the Mascarene Flora—Concluding Remarks on the Madagascar Group.

We have now to consider the phenomena presented by a very distinct class of islands—those which, although once forming part of a continent, have been separated from it at a remote epoch when its animal forms were very unlike what they are now. Such islands preserve to us the record of a by-gone world,—of a period when many of the higher types had not yet come into existence and when the distribution of others was very different from what prevails at the present day. The problem presented by these ancient islands is often complicated by the changes they themselves have undergone since the period of their separation. A partial subsidence will have led to the extinction of some of the types that were originally preserved, and may leave the ancient fauna in a very fragmentary state; while subsequent elevations may have brought it so near to the continent that some immigration even of mammalia may have taken place. If these elevations and subsidences occurred several times over, though never to such an extent as again to unite the island with the continent, it is evident that a very complex result might be produced; for besides the relics of the ancient fauna, we might have successive immigrations from surrounding lands reaching down to the era of existing species. Bearing in mind these possible changes, we shall generally be able to arrive at a fair conjectural solution of the phenomena of distribution presented by these ancient islands.

Undoubtedly the most interesting of such islands, and that which exhibits their chief peculiarities in the greatest perfection, is Madagascar, and we shall therefore enter somewhat fully into its biological and physical history.

Physical Features of Madagascar.—This great island is situated about 250 miles from the east coast of Africa, and extends from 12° to 25½° S. Lat. It is almost exactly 1,000 miles long, with an extreme width of 360 and an average width of more than 260 miles. A lofty granitic plateau, from eighty to 160 miles wide and from 3,000 to 5,000 feet high, occupies its central portion, on which rise peaks and domes of basalt and granite to a height of nearly 9,000 feet; and there are also numerous extinct volcanic cones and craters. All round the island, but especially developed on the south and west, are plains of a few hundred feet elevation, formed of rocks which are shown by their fossils to be of Jurassic age, or at all events to belong to somewhere near the middle portion of the Secondary period. The higher granitic plateau consists of bare undulating moors, while the lower Secondary plains are more or less wooded; and there is here also a continuous belt of dense forest, varying from six or eight to fifty miles wide, encircling the whole island, usually at about thirty miles distance from the coast but in the north-east coming down to the sea-shore.



The sea around Madagascar, when the shallow bank on which it stands is passed, is generally deep. This 100-fathom bank is only from one to three miles wide on the east side, but on the west it is much broader, and stretches out opposite Mozambique to a distance of about eighty miles. The Mozambique Channel is rather more than 1,000 fathoms deep, but there is only a narrow belt of this depth opposite Mozambique, and still narrower where the Comoro Islands and adjacent shoals seem to form stepping-stones to the continent of Africa. The 1,000-fathom line includes Aldabra and the small Farquhar Islands to the north of Madagascar; while to the east the sea deepens rapidly to the 1,000-fathom line and then more slowly, a profound channel of 2,400 fathoms separating Madagascar from Bourbon and Mauritius. To the north-east of Mauritius are a series of extensive shoals forming four large banks less than 100 fathoms below the surface, while the 1,000-fathom line includes them all, with an area about half that of Madagascar itself. A little further north is the Seychelles group, also standing on an extensive 1,000-fathom bank, while all round the sea is more than 2,000 fathoms deep.

It seems probable, then, that to the north-east of Madagascar there was once a series of very large islands, separated from it by not very wide straits; while eastward across the Indian Ocean we find the Chagos and Maldive coral atolls, perhaps marking the position of other large islands, which together would form a line of communication, by comparatively easy stages of 400 or 500 miles each between Madagascar and India. These submerged islands, as shown in our map at p. 424, are of great importance in explaining some anomalous features in the zoology of this great island.

If the rocks of Secondary age which form a belt around the island are held to indicate that Madagascar was once of less extent than it is now (though this by no means necessarily follows), we have also evidence that it has recently been considerably larger; for along the east coast there is an extensive barrier coral-reef about 350 miles in length, and varying in distance from the land from a quarter of a mile to three or four miles. This seems to indicate recent subsidence; while we have no record of raised coral rocks inland which would certainly mark any recent elevation, though fringing coral reefs surround a considerable portion of the northern, eastern, and south-western coasts. We may therefore conclude that during Tertiary times the island was usually as large as, and often probably much larger than, it is now.

[[Image:Wallace_Island_Life_p415.png|center|495px|thumb| MAP OF THE MADAGASCAR GROUP, SHOWING DEPTHS OF SEA.

In this Map the depth of the sea is shown by three tints; the lightest tint indicating from 0 to 100 fathoms, the medium tint from 100 to 1,000 fathoms, the dark tint more than 1,000 fathoms.

]]

Biological Features of Madagascar.—Madagascar possesses an exceedingly rich and beautiful fauna and flora, rivalling in some groups most tropical countries of equal extent, and even when poor in species, of surpassing interest from the singularity, the isolation, or the beauty of its forms of life. In order to exhibit the full peculiarity of its natural history and the nature of the problems it offers to the biological student, we must give an outline of its more important animal forms in systematic order.

Mammalia.—Madagascar possesses no less than sixty-six species of mammals—a certain proof in itself that the island has once formed part of a continent; but the character of these animals is very extraordinary and altogether different from the assemblage now found in Africa or in any other existing continent. Africa is now most prominently characterised by its monkeys, apes, and baboons; by its lions, leopards, and hyænas; by its zebras, rhinoceroses, elephants, buffaloes, giraffes, and numerous species of antelopes. But no one of these animals, nor any thing like them, is found in Madagascar, and thus our first impression would be that it could never have been united with the African continent. But, as the tigers, the bears, the tapirs, the deer, and the numerous squirrels of Asia are equally absent, there seems no probability of its having been united with that continent. Let us then see to what groups the mammalia of Madagascar belong, and where we must look for their probable allies.

First and most important are the lemurs, consisting of six genera and thirty-three species, thus comprising just half the entire mammalian population of the island. This group of lowly-organised and very ancient creatures still exists scattered over a wide area; but they are nowhere so abundant as in the island of Madagascar. They are found from West Africa to India, Ceylon, and the Malay Archipelago, consisting of a number of isolated genera and species, which appear to maintain their existence by their nocturnal and arboreal habits, and by haunting dense forests. It can hardly be said that the African forms of lemurs are more nearly allied to those of Madagascar than are the Asiatic, the whole series appearing to be the disconnected fragments of a once more compact and extensive group of animals.

Next, we have about a dozen species of Insectivora, consisting of one shrew, a group distributed over all the great continents; and five genera of a peculiar family, Centetidæ, which family exists nowhere else on the globe except in the two largest West Indian Islands, Cuba and Hayti, thus adding still further to our embarrassment in seeking for the original home of the Madagascar fauna.

We then come to the Carnivora, which are represented by a peculiar cat-like animal, Cryptoprocta, forming a distinct family, and having no close allies in any part of the globe; and eight civets belonging to four peculiar genera. Here we first meet with some decided indications of an African origin; for the civet family is more abundant in this continent than in Asia, and some of the Madagascar genera seem to be decidedly allied to African groups—as, for example, Eupleres to Suricata and Crossarchus.

The Rodents consist only of four rats and mice of peculiar genera, one of which is said to be allied to an American genus; and lastly we have a river-hog of the African genus Potamochærus, and a small sub-fossil hippopotamus, both of which being semi-aquatic animals might easily have reached the island from Africa, by way of the Comoros, without any actual land connection.

Reptiles of Madagascar.—Passing over the birds for the present, as not so clearly demonstrating land-connection, let us see what indications are afforded by the reptiles. The large and universally distributed family of Colubrine snakes is represented in Madagascar, not by African or Asiatic genera, but by two American genera—Philodryas and Heterodon, and by Herpetodryas, a genus found in America and China. The other genera are all peculiar, and belong mostly to widespread tropical families; but two families—Lycodontidæ and Viperidæ, both abundant in Africa and the Eastern tropics—are absent. Lizards are mostly represented by peculiar genera of African or tropical families, but several African genera are represented by peculiar species, and there are also some species belonging to two American genera of the Iguanidæ, a family which is exclusively American; while a genus of geckoes, inhabiting America and Australia, also occurs in Madagascar.

Relation of Madagascar to Africa.—These facts taken all together are certainly very extraordinary, since they show in a considerable number of cases as much affinity with America as with Africa; while the most striking and characteristic groups of animals now inhabiting Africa are entirely wanting in Madagascar. Let us first deal with this fact, of the absence of so many of the most dominant African groups. The explanation of this deficiency is by no means difficult, for the rich deposits of fossil mammals of Miocene or Pliocene age in France, Germany, Greece, and North-west India, have demonstrated the fact that all the great African mammals then inhabited Europe and temperate Asia. We also know that a little earlier (in Eocene times) tropical Africa was cut off from Europe and Asia by a sea stretching from the Atlantic to the Bay of Bengal, at which time Africa must have formed a detached island-continent such as Australia is now, and probably, like it, very poor in the higher forms of life. Coupling these two facts, the inference seems clear, that all the higher types of mammalia were developed in the great Euro-Asiatic continent (which then included Northern Africa), and that they only migrated into tropical Africa when the two continents became united by the upheaval of the sea-bottom, probably in the latter portion of the Miocene or early in the Pliocene period.

It is clear, therefore, that if Madagascar had once formed part of Africa, but had been separated from it before Africa was united to Europe and Asia, it would not contain any of those kinds of animals which then first entered the country. But, besides the African mammals, we know that some birds now confined to Africa then inhabited Europe, and we may therefore fairly assume that all the more important groups of birds, reptiles, and insects, now abundant in Africa but absent from Madagascar, formed no part of the original African fauna, but entered the country only after it was joined to Europe and Asia.

Early History of Africa and Madagascar.—We have seen that Madagascar contains an abundance of mammals, and that most of them are of types either peculiar to, or existing also in, Africa; it follows that that continent must have had an earlier union with Europe, Asia, or America, or it could never have obtained any mammals at all.

Now these ancient African mammals are Lemurs, Insectivora, and small Carnivora, chiefly Viverridæ; and all these groups are known to have inhabited Europe in Eocene and Miocene times; and that the union was with Europe rather than with America is clearly proved by the fact that even the insectivorous Centetidæ, now confined to Madagascar and the West Indies, inhabited France in the Lower Miocene period, while the Viverridæ, or civets, which form so important a part of the fauna of Madagascar as well as of Africa, were abundant in Europe throughout the whole Tertiary period, but are not known to have ever lived in any part of the American continent. We here see the application of the principle which we have already fully proved and illustrated (Chapter IV., p. 60), that all extensive groups have a wide range at the period of their maximum development; but as they decay their area of distribution diminishes or breaks up into detached fragments, which one after another disappear till the group becomes extinct. Those animal forms which we now find isolated in Madagascar and other remote portions of the globe all belong to ancient groups which are in a decaying or nearly extinct condition, while those which are absent from it belong to more recent and more highly-developed types, which range over extensive and continuous areas, but have had no opportunity of reaching the more ancient continental islands.

Anomalies of Distribution and How to Explain Them.—If these considerations have any weight, it follows that there is no reason whatever for supposing any former direct connection between Madagascar and the Greater Antilles merely because the insectivorous Centetidæ now exist only in these two groups of islands; for we know that the ancestors of this family must once have had a much wider range, which almost certainly extended over the great northern continents. We might as reasonably suppose a land-connection across the Pacific to account for the camels of Asia having their nearest existing allies in the llamas and alpacas of the Peruvian Andes, and another between Sumatra and Brazil, in order that the ancestral tapir of one country might have passed over to the other. In both these cases we have ample proof of the former wide extension of the group. Extinct camels of numerous species abounded in North America in Miocene, Pliocene, and even Post-pliocene times, and one has also been found in North-western India, but none whatever among all the rich deposits of mammalia in Europe. We are thus told, as clearly as possible, that from the North American continent as a centre the camel tribe spread westward, over now-submerged land at the shallow Behring Straits and Kamschatka Sea, into Asia, and southward along the Andes into South America. Tapirs are even more interesting and instructive. Their remotest known ancestors appear in Western Europe in the early portion of the Eocene period; in the latter Eocene and the Miocene other forms occur both in Europe and North America. These seem to have become extinct in North America, while in Europe they developed largely into many forms of true tapirs, which at a much later period found their way again to North, and thence to South, America, where their remains are found in caves and gravel deposits. It is an instructive fact that in the Eastern continent, where they were once so abundant, they have dwindled down to a single species, existing in small numbers in the Malay Peninsula, Sumatra, and Borneo only; while in the Western continent, where they are comparatively recent immigrants, they occupy a much larger area, and are represented by three or four distinct species. Who could possibly have imagined such migrations, and extinctions, and changes of distribution as are demonstrated in the case of the tapirs, if we had only the distribution of the existing species to found an opinion upon? Such cases as these—and there are many others equally striking—show us with the greatest distinctness how nature has worked in bringing about the examples of anomalous distribution that everywhere meet us; and we must, on every ground of philosophy and common sense, apply the same method of interpretation to the more numerous instances of anomalous distribution we discover among such groups as reptiles, birds, and insects, where we rarely have any direct evidence of their past migrations through the discovery of fossil remains. Whenever we can trace the past history of any group of terrestrial animals, we invariably find that its actual distribution can be explained by migrations effected by means of comparatively slight modifications of our existing continents. In no single case have we any direct evidence that the distribution of land and sea has been radically changed during the whole lapse of the Tertiary and Secondary periods, while, as we have already shown in our fifth chapter, the testimony of geology itself, if fairly interpreted, upholds the same theory of the stability of our continents and the permanence of our oceans. Yet so easy and pleasant is it to speculate on former changes of land and sea with which to cut the gordian knot offered by anomalies of distribution, that we still continually meet with suggestions of former continents stretching in every direction across the deepest oceans, in order to explain the presence in remote parts of the globe of the same genera even of plants or of insects—organisms which possess such exceptional facilities both for terrestrial, aërial, and oceanic transport, and of whose distribution in early geological periods we generally know little or nothing.

The Birds of Madagascar, as Indicating a Supposed Lemurian Continent.—Having thus shown how the distribution of the land mammalia and reptiles of Madagascar may be well explained by the supposition of a union with Africa before the greater part of its existing fauna had reached it, we have now to consider whether, as some ornithologists think, the distribution and affinities of the birds present an insuperable objection to this view, and require the adoption of a hypothetical continent—Lemuria—extending from Madagascar to Ceylon and the Malay Islands.

There are about one hundred and fifty land birds known from the island of Madagascar, of which a hundred and twenty-seven are peculiar; and about half of these peculiar species belong to peculiar genera, many of which are extremely isolated, so that it is often difficult to class them in any of the recognised families, or to determine their affinities to any living birds. Among the other moiety, belonging to known genera, we find fifteen which have undoubted African affinities, while five or six are as decidedly Oriental, the genera or nearest allied species being found in India or the Malay Islands. It is on the presence of these peculiar Indian types that Dr. Hartlaub, in his recent work on the Birds of Madagascar and the Adjacent Islands, lays great stress, as proving the former existence of "Lemuria"; while he considers the absence of such peculiar African families as the plantain-eaters, glossy-starlings, ox-peckers, barbets, honey-guides, hornbills, and bustards—besides a host of peculiar African genera—as sufficiently disproving the statement in my Geographical Distribution of Animals that Madagascar is "more nearly related to the Ethiopian than to any other region," and that its fauna was evidently "mainly derived from Africa."

But the absence of the numerous peculiar groups of African birds is so exactly parallel to the same phenomenon among mammals, that we are justified in imputing it to the same cause, the more especially as some of the very groups that are wanting—the plantain-eaters and the trogons, for example,—are actually known to have inhabited Europe along with the large mammalia which subsequently migrated to Africa. As to the peculiarly Eastern genera—such as Copsychus and Hypsipetes, with a Dicrurus, a Ploceus, a Cisticola, and a Scops, all closely allied to Indian or Malayan species—although very striking to the ornithologist, they certainly do not outweigh the fourteen African genera found in Madagascar. Their presence may, moreover, be accounted for more satisfactorily than by means of an ancient Lemurian continent, which, even if granted, would not explain the very facts adduced in its support.

Let us first prove this latter statement.

The supposed "Lemuria" must have existed, if at all, at so remote a period that the higher animals did not then inhabit either Africa or Southern Asia, and it must have become partially or wholly submerged before they reached those countries; otherwise we should find in Madagascar many other animals besides Lemurs, Insectivora, and Viverridæ, especially such active arboreal creatures as monkeys and squirrels, such hardy grazers as deer or antelopes, or such wide-ranging carnivores as foxes or bears. This obliges us to date the disappearance of the hypothetical continent about the earlier part of the Miocene epoch at latest, for during the latter part of that period we know that such animals existed in abundance in every part of the great northern continents wherever we have found organic remains. But the Oriental birds in Madagascar, by whose presence Dr. Hartlaub upholds the theory of a Lemuria, are slightly modified forms of existing Indian genera, or sometimes, as Dr. Hartlaub himself points out, species hardly distinguishable from those of India. Now all the evidence at our command leads us to conclude that, even if these genera and species were in existence in the early Miocene period, they must have had a widely different distribution from what they have now. Along with so many African and Indian genera of mammals they then probably inhabited Europe, which at that epoch enjoyed a sub-tropical climate; and this is rendered almost certain by the discovery in the Miocene of France of fossil remains of trogons and jungle-fowl. If, then, these Indian birds date back to the very period during which alone Lemuria could have existed, that continent was quite unnecessary for their introduction into Madagascar, as they could have followed the same track as the mammalia of Miocene Europe and Asia; while if, as I maintain, they are of more recent date, then Lemuria had ceased to exist, and could not have been the means of their introduction.

Submerged Islands between Madagascar and India.—Looking at the accompanying map of the Indian Ocean, we see that between Madagascar and India there are now extensive shoals and coral reefs, such as are usually held to indicate subsidence; and we may therefore fairly postulate the former existence here of several large islands, some of them not much inferior to Madagascar itself. These reefs are all separated from each other by very deep sea—much deeper than that which divides Madagascar from Africa, and we have therefore no reason to imagine their former union. But they would nevertheless greatly facilitate the introduction of Indian birds into the Mascarene Islands and Madagascar; and these facilities existing, such an immigration would be sure to take place, just as surely as American birds have entered the Galapagos and Juan Fernandez, as European birds now reach the Azores, and as Australian birds reach such a distant island as New Zealand. This would take place the more certainly because the Indian Ocean is a region of violent periodical storms at the changes of the monsoons, and we have seen in the case of the Azores and Bermuda how important a factor this is in determining the transport of birds across the ocean.

[[Image:Wallace_Island_Life_p425.png|center|375px|thumb| MAP OF THE INDIAN OCEAN.

Showing the position of banks less than 1,000 fathoms deep between Africa and the Indian Peninsula.

]]

The final disappearance of these now sunken islands does not, in all probability, date back to a very remote epoch; and this exactly accords with the fact that some of the birds, as well as the fruit-bats of the genus Pteropus, are very closely allied to Indian species, if not actually identical, others being distinct species of the same genera. The fact that not one closely-allied species or even genus of Indian or Malayan mammals is found in Madagascar, sufficiently proves that it is no land-connection that has brought about this small infusion of Indian birds and bats; while we have sufficiently shown, that, when we go back to remote geological times no land-connection in this direction was necessary to explain the phenomena of the distribution of the Lemurs and Insectivora. A land-connection with some continent was undoubtedly necessary, or there would have been no mammalia at all in Madagascar; and the nature of its fauna on the whole, no less than the moderate depth of the intervening strait and the comparative approximation of the opposite shores, clearly indicate that the connection was with Africa.

Concluding Remarks on "Lemuria."—I have gone into this question in some detail, because Dr. Hartlaub's criticism on my views has been reproduced in a scientific periodical, and the supposed Lemurian continent is constantly referred to by quasi-scientific writers, as well as by naturalists and geologists, as if its existence had been demonstrated by facts, or as if it were absolutely necessary to postulate such a land in order to account for the entire series of phenomena connected with the Madagascar fauna, and especially with the distribution of the Lemuridæ. I think I have now shown, on the other hand, that it was essentially a provisional hypothesis, very useful in calling attention to a remarkable series of problems in geographical distribution, but not affording the true solution of those problems, any more than the hypothesis of an Atlantis solved the problems presented by the Atlantic Islands and the relations of the European and North American flora and fauna. The Atlantis is now rarely introduced seriously except by the absolutely unscientific, having received its death-blow by the chapter on Oceanic Islands in the Origin of Species, and the researches of Professor Asa Gray on the affinities of the North American and Asiatic floras. But "Lemuria" still keeps its place—a good example of the survival of a provisional hypothesis which offers what seems an easy solution of a difficult problem, and has received an appropriate and easily remembered name, long after it has been proved to be untenable.

It is now more than fifteen years since I first showed, by a careful examination of all the facts to be accounted for, that the hypothesis of a Lemurian continent was alike unnecessary to explain one portion of the facts, and inadequate to explain the remaining portion. Since that time I have seen no attempt even to discuss the question on general grounds in opposition to my views, nor on the other hand have those who have hitherto supported the hypothesis taken any opportunity of acknowledging its weakness and inutility. I have therefore here explained my reasons for rejecting it somewhat more fully and in a more popular form, in the hope that a check may thus be placed on the continued re-statement of this unsound theory as if it were one of the accepted conclusions of modern science.

The Mascarene Islands.—In the Geographical Distribution of Animals, a summary is given of all that was known of the zoology of the various islands near Madagascar, which to some extent partake of its peculiarities, and with it form the Malagasy sub-region of the Ethiopian region. As no great additions have since been made to our knowledge of the fauna of these islands, and my object in this volume being more especially to illustrate the mode of solving distributional problems by means of the most suitable examples, I shall now confine myself to pointing out how far the facts presented by these outlying islands support the views already enunciated with regard to the origin of the Madagascar fauna.

The Comoro Islands.—This group of islands is situated nearly midway between the northern extremity of Madagascar and the coast of Africa. The four chief islands vary between sixteen and forty miles in length, the largest being 180 miles from the coast of Africa, while one or two smaller islets are less than 100 miles from Madagascar. All are volcanic, Great Comoro being an active volcano 8,500 feet high; and, as already stated, they are situated on a submarine bank with less than 500 fathoms soundings, connecting Madagascar with Africa. There is reason to believe, however, that these islands are of comparatively recent origin, and that the bank has been formed by matter ejected by the volcanoes or by upheaval. Anyhow, there is no indication whatever of there having been here a land-connection between Madagascar and Africa; while the islands themselves have been mainly colonised from Madagascar, some of them making a near approach to the 100-fathom bank which surrounds that island.

The Comoros contain two land mammals, a lemur and a civet, both of Madagascar genera and the latter an identical species, and there is also a peculiar species of fruit-bat (Pteropus comorensis), a group which ranges from Australia to Asia and Madagascar but is unknown in Africa. Of land-birds forty-one species are known, of which sixteen are peculiar to the islands, twenty-one are found also in Madagascar, and three found in Africa and not in Madagascar; while of the peculiar species, six belong to Madagascar or Mascarene genera. A species of Chameleon is also peculiar to the islands.

These facts point to the conclusion that the Comoro Islands have been formerly more nearly connected with Madagascar than they are now, probably by means of intervening islets and the former extension of the latter island to the westward, as indicated by the extensive shallow bank at its northern extremity, so as to allow of the easy passage of birds, and the occasional transmission of small mammalia by means of floating trees.

The Seychelles Archipelago.—This interesting group consists of about thirty small islands situated 700 miles N.N.E. of Madagascar, or almost exactly in the line formed by continuing the central ridge of that great island. The Seychelles stand upon a rather extensive shallow bank, the 100-fathom line around them enclosing an area nearly 200 miles long by 100 miles wide, while the 500-fathom line shows an extension of nearly 100 miles in a southern direction. All the larger islands are of granite, with mountains rising to 3,000 feet in Mahé, and to from 1,000 to 2,000 feet in several of the other islands. We can therefore hardly doubt that they form a portion of the great line of upheaval which produced the central granitic mass of Madagascar, intervening points being indicated by the Amirantes, the Providence, and the Farquhar Islands, which, though all coralline, probably rest on a granitic basis. Deep channels of more than 1,000 fathoms now separate these islands from each other, and if they were ever sufficiently elevated to be united, it was probably at a very remote epoch.

The Seychelles may thus have had ample facilities for receiving from Madagascar such immigrants as can pass over narrow seas; and, on the other hand, they were equally favourably situated as regards the extensive Saya de Malha and Cargados banks, which were probably once large islands, and may have supported a rich insular flora and fauna of mixed Mascarene and Indian type. The existing fauna and flora of the Seychelles must therefore be looked upon as the remnants which have survived the partial submergence of a very extensive island; and the entire absence of non-aërial mammalia may be due, either to this island having never been actually united to Madagascar, or to its having since undergone so much submergence as to have led to the extinction of such mammals as may once have inhabited it. The birds and reptiles, however, though few in number, are very interesting, and throw some further light on the past history of the Seychelles.

Birds of the Seychelles.—Fifteen indigenous land-birds are known to inhabit the group, thirteen of which are peculiar species, belonging to genera which occur also in Madagascar or Africa. The genera which are more peculiarly Indian are,—Copsychus and Hypsipetes, also found in Madagascar; and Palæornis, which has species in Mauritius and Rodriguez, as well as one on the continent of Africa. A black parrot (Coracopsis), congeneric with two species that inhabit Madagascar and with one that is peculiar to the Comoros; and a beautiful red-headed blue pigeon (Alectorænas pulcherrimus) allied to those of Madagascar and Mauritius, but very distinct, are the most remarkable species characteristic of this group of islands.

Reptiles and Amphibia of the Seychelles.—The reptiles and amphibia are rather numerous and very interesting, indicating clearly that the islands can hardly be classed as oceanic. There are seven species of lizards, three being peculiar to the islands, while the others have rather a wide range. The first is a chameleon—defenceless slow-moving lizards, especially abundant in Madagascar, from which no less than eighteen species are now known, about the same number as on the continent of Africa. The Seychelles species (Chamæleon tigris) also occurs at Zanzibar. The next are skinks (Scincidæ), small ground-lizards with a wide distribution in the Eastern hemisphere. Two species are however peculiar to the islands—Mabuia seychellensis and M. wrightii. The other peculiar species is one of the geckoes (Geckotidæ) named Æluronyx seychellensis, and there are also three other geckoes, Phelsuma madagascarensis, Gehyra mutilata and Hemidactylus frenatus, the two latter having a wide distribution in the tropical regions of both hemispheres. These lizards, clinging as they do to trees and timber, are exceedingly liable to be carried in ships from one country to another, and I am told by Dr. Günther that some are found almost every year in the London Docks. It is therefore probable, that when species of this family have a very wide range they have been assisted in their migrations by man, though their habit of clinging to trees also renders them likely to be floated with large pieces of timber to considerable distances. Dr. Percival Wright, to whom I am indebted for much information on the productions of the Seychelles Archipelago, informs me that the last-named species varies greatly in colour in the different islands, so that he could always tell from which particular island a specimen had been brought. This is analogous to the curious fact of certain lizards on the small islands in the Mediterranean being always very different in colour from those of the mainland, usually becoming rich blue or black (see Nature, Vol. XIX. p. 97); and we thus learn how readily in some cases differences of colour are brought about, either directly or indirectly, by local conditions.

Snakes, as is usually the case in small or remote islands, are far less numerous than lizards, only two species being known. One, Dromicus seychellensis, is a peculiar species of the family Colubridæ, the rest of the genus being found in Madagascar and South America. The other, Boodon geometricus, one of the Lycodontidæ, or fanged ground-snakes, is also peculiar. So far, then, as the reptiles are concerned, there is nothing but what is easily explicable by what we know of the general means of distribution of these animals.

We now come to the Amphibia, which are represented in the Seychelles by two tailless and two serpent-like forms. The frogs are Rana mascareniensis, found also in Mauritius, Bourbon, Angola, and Abyssinia, and probably all over tropical Africa; and Megalixalus seychellensis a peculiar tree-frog having allies in Madagascar and tropical Africa. It is found, Dr. Wright informs me, on the Pandani or screw-pines; and as these form a very characteristic portion of the vegetation of the Mascarene Islands, all the species being peculiar and confined each to a single island or small group, we may perhaps consider it as a relic of the indigenous fauna of that more extensive land of which the present islands are the remains.

The serpentine Amphibia are represented by two species of Cæcilia. These creatures externally resemble large worms, except that they have a true head with jaws and rudimentary eyes, while internally they have of course a true vertebrate skeleton. They live underground, burrowing by means of the ring-like folds of the skin which simulate the jointed segments of a worm's body, and when caught they exude a viscid slime. The young have external gills which are afterwards replaced by true lungs, and this peculiar metamorphosis shows that they belong to the amphibia rather than to the reptiles. The Cæcilias are widely but very sparingly distributed through all the tropical regions; a fact which may, as we have seen, be taken as an indication of the great antiquity of the group, and that it is now verging towards extinction. In the Seychelles Islands there appear to be three species of these singular animals. Cryptopsophis multiplicatus is confined to the islands; Herpele squalostoma is found also in Western India and in Africa; while Hypogeophis rostratus inhabits both West Africa and South America. This last is certainly one of the most remarkable cases of the wide and discontinuous distribution of a species; and when we consider the habits of life of these animals and the extreme slowness with which it is likely they can migrate into new areas, we can hardly arrive at any other conclusion than that this species once had an almost world-wide range, and that in the process of dying out it has been left stranded, as it were, in these three remote portions of the globe. The extreme stability and long persistence of specific form which this implies is extraordinary, but not unprecedented, among the lower vertebrates. The crocodiles of the Eocene period differ but slightly from those of the present day, while a small freshwater turtle from the Pliocene deposits of the Siwalik Hills is absolutely identical with a still living Indian species, Emys tectus. The mud-fish of Australia, Ceratodus forsteri is a very ancient type, and may well have remained specifically unchanged since early Tertiary times. It is not, therefore, incredible that this Seychelles Cæcilia may be the oldest land vertebrate now living on the globe; dating back to the early part of the Tertiary period, when the warm climate of the northern hemisphere in high latitudes and the union of the Asiatic and American continents allowed of the migration of such types over the whole northern hemisphere, from which they subsequently passed into the southern hemisphere, maintaining themselves only in certain limited areas, where the physical conditions were especially favourable, or where they were saved from the attacks of enemies or the competition of higher forms.

Fresh-water Fishes.—The only other vertebrates in the Seychelles are two fresh-water fishes abounding in the streams and rivulets. One, Haplochilus playfairii is peculiar to the islands, but there are allied species in Madagascar. It is a pretty little fish about four inches long, of an olive colour, with rows of red spots, and is very abundant in some of the mountain streams. The fishes of this genus, as I am informed by Dr. Günther, often inhabit both sea and fresh water, so that their migration from Madagascar to the Seychelles and subsequent modification, offers no difficulty. The other species is Fundulus orthonotus, found also on the east coast of Africa; and as both belong to the same family—Cyprinodontidæ—this may possibly have migrated in a similar manner.

Land-shells.—The only other group of animals inhabiting the Seychelles which we know with any approach to completeness, are the land and fresh-water mollusca, but they do not furnish any facts of special interest. About forty species are known, and Mr. Geoffrey Nevill, who has studied them, thinks their meagre number is chiefly owing to the destruction of so much of the forests which once covered the islands. Seven of the species—and among them one of the most conspicuous, Achatina fulica—have almost certainly been introduced; and the remainder show a mixture of Madagascar and Indian forms, with a preponderance of the latter. Five genera—Streptaxis, Cyathoponea, Onchidium, Helicina and Paludomus, are mentioned as being especially Indian, while only two—Tropidophora and Gibbus, are found in Madagascar but not in India. About two-thirds of the species appear to be peculiar to the islands.

Mauritius, Bourbon and Rodriguez.—These three islands are somewhat out of place in this chapter, because they really belong to the oceanic group, being of volcanic formation, surrounded by deep sea, and possessing no indigenous mammals or amphibia. Yet their productions are so closely related to those of Madagascar, to which they may be considered as attendant satellites, that it is absolutely necessary to associate them together if we wish to comprehend and explain their many interesting features.

Mauritius and Bourbon are lofty volcanic islands, evidently of great antiquity. They are about 100 miles apart, and the sea between them is less than 1,000 fathoms deep, while on each side it sinks rapidly to depths of 2,400 and 2,600 fathoms. We have therefore no reason to believe that they have ever been connected with Madagascar, and this view is strongly supported by the character of their indigenous fauna. Of this, however, we have not a very complete or accurate knowledge, for though both islands have long been occupied by Europeans, the study of their natural products was for a long time greatly neglected, and owing to the rapid spread of sugar cultivation, the virgin forests, and with them no doubt many native animals, have been almost wholly destroyed. There is, however, no good evidence of there ever having been any indigenous mammals or amphibia, though both are now found and are often recorded among the native animals.

The smaller and more remote island, Rodriguez, is also volcanic; but it has, besides a good deal of coralline rock, an indication of partial submergence helping to account for the poverty of its fauna and flora. It stands on a 100-fathom bank of considerable extent, but beyond this the sea rapidly deepens to more than 2,000 fathoms, so that it is truly oceanic like its larger sister isles.

Birds.—The living birds of these islands are few in number and consist mainly of peculiar species of Mascarene types, together with two peculiar genera—Oxynotus belonging to the Campephagidæ or caterpillar-catchers, a family abundant in the old-world tropics; and a dove, Trocazza, forming a peculiar sub-genus. The origin of these birds offers no difficulty, looking at the position of the islands and of the surrounding shoals and islets.

Extinct Birds.—These three islands are, however, preeminently remarkable as having been the home of a group of large ground-birds, quite incapable of flight, and altogether unlike anything found elsewhere on the globe; and which, though once very abundant, have become totally extinct within the last two hundred years. The best known of these birds is the dodo, which inhabited Mauritius; while allied species certainly lived in Bourbon and Rodriguez, abundant remains of the species of the latter island—the "solitaire," having been discovered, corresponding with the figure and description given of it by Legouat, who resided in Rodriguez in 1692. These birds constitute a distinct family, Dididæ, allied to the pigeons but very isolated. They were quite defenceless, and were rapidly exterminated when man introduced dogs, pigs, and cats into the island, and himself sought them for food. The fact that such perfectly unprotected creatures survived in great abundance to a quite recent period in these three islands only, while there is no evidence of their ever having inhabited any other countries whatever, is itself almost demonstrative that Mauritius, Bourbon, and Rodriguez are very ancient but truly oceanic islands. From what we know of the general similarity of Miocene birds to living genera and families, it seems clear that the origin of so remarkable a type as the dodos must date back to early Tertiary times. If we suppose some ancestral ground-feeding pigeon of large size to have reached the group by means of intervening islands afterwards submerged, and to have thenceforth remained to increase and multiply unchecked by the attacks of any more powerful animals, we can well understand that the wings, being useless, would in time become almost aborted. It is also not improbable that this process would be aided by natural selection, because the use of wings might be absolutely prejudicial to the birds in their new home. Those that flew up into trees to roost, or tried to cross over the mouths of rivers, might be blown out to sea and destroyed, especially during the hurricanes which have probably always more or less devastated the islands; while on the other hand the more bulky and short-winged individuals, who took to sleeping on the ground in the forest, would be preserved from such dangers, and perhaps also from the attacks of birds of prey which may always have visited the islands. But whether or no this was the mode by which these singular birds acquired their actual form and structure, it is perfectly certain that their existence and development depended on complete isolation and on freedom from the attacks of enemies. We have no single example of such defenceless birds having ever existed on a continent at any geological period, whereas analogous though totally distinct forms do exist in New Zealand, where enemies are equally wanting. On the other hand, every continent has always produced abundance of carnivora adapted to prey upon the herbivorous animals inhabiting it at the same period; and we may therefore be sure that these islands have never formed part of a continent during any portion of the time when the dodos inhabited them.

It is a remarkable thing that an ornithologist of Dr. Hartlaub's reputation, looking at the subject from a purely ornithological point of view, should yet entirely ignore the evidence of these wonderful and unique birds against his own theory, when he so confidently characterises Lemuria as "that sunken land, which, containing parts of Africa, must have extended far eastward over Southern India and Ceylon, and the highest points of which we recognise in the volcanic peaks of Bourbon and Mauritius, and in the central range of Madagascar itself—the last resorts of the mostly extinct Lemurine race which formerly peopled it." It is here implied that lemurs formerly inhabited Bourbon and Mauritius, but of this there is not a particle of evidence, and we feel pretty sure that had they done so the dodos would never have been developed there. In Madagascar there are no traces of dodos, while there are remains of extinct gigantic struthious birds of the genus Æpyornis, which were no doubt as well able to protect themselves against the smaller carnivora as are the ostriches, emus, and cassowaries in their respective countries at the present day.

The whole of the evidence at our command, therefore, tends to establish in a very complete manner the "oceanic" character of the three islands—Mauritius, Bourbon, and Rodriguez, and that they have never formed part of "Lemuria" or of any continent.

Reptiles.—Mauritius, like Bourbon, has lizards, some of which are peculiar species; but no snakes, and no frogs or toads but such as have been introduced. Strange to say, however, a small islet called Round Island, only about a mile across, and situated about fourteen miles north-east of Mauritius, possesses a snake which is not only unknown in Mauritius, but also in any other part of the world, being altogether confined to this minute islet! It belongs to the boa family, and forms a peculiar and very distinct genus, Casaria, whose nearest allies seem to be the Ungalia of Cuba and Bolyeria of Australia. It is hardly possible to believe that this serpent has very long maintained itself on so small an island; and though we have no record of its existence on Mauritius, it may very well have inhabited the lowland forests without being met with by the early settlers; and the introduction of swine, which soon ran wild and effected the final destruction of the dodo, may also have been fatal to this snake. It is, however, now almost certainly confined to the one small islet, and is probably the land-vertebrate of most restricted distribution on the globe.

On the same island there is a small lizard, Scelotes bojeri, recorded also from Mauritius and Bourbon, though it appears to be rare in both islands; but a gecko, Phelsuma guentheri, is restricted to the island. As Round Island is connected with Mauritius by a bank under a hundred fathoms below the surface, it has probably been once joined to it, and when first separated would have been both much larger and much nearer the main island, circumstances which would greatly facilitate the transmission of these reptiles to their present dwelling-place, where they have been able to maintain themselves owing to the complete absence of competition, while some of them have become extinct in the larger island.

Flora of Madagascar and the Mascarene Islands.—The botany of the great island of Madagascar has been perhaps more thoroughly explored than that of the opposite coasts of Africa, so that its peculiarities may not be really so great as they now appear to be. Yet there can be no doubt of its extreme richness and grandeur, its remarkable speciality, and its anomalous external relations. It is characterised by a great abundance of forest-trees and shrubs of peculiar genera or species, and often adorned with magnificent flowers. Some of these are allied to African forms, others to those of Asia, and it is said that of the two affinities the latter preponderates. But there are also, as in the animal world, some decided South American relations, while other groups point to Australia, or are altogether isolated.

No less than 3,740 flowering plants are now known from Madagascar with 360 ferns and fern-allies. The most abundant natural orders are the following:

The flora contains representatives of 144 natural orders and 970 genera, one of the former and 148 of the latter being peculiar to the island. The peculiar order, Chælnaceæ, comprises seven genera and twenty-four species; while Rubiaceæ and Compositæ have the largest number of peculiar genera, followed by Leguminosæ and Melastomaceæ. Nearly three-fourths of the species are endemic.

Beautiful flowers are not conspicuous in the flora of Madagascar, though it contains several magnificent flowering plants. A shrub with the dreadful name Harpagophytum Grandidieri has bunches of gorgeous red flowers; Tristellateia madagascariensis is a climbing plant with spikes of rich yellow flowers; while Poinciana regia, a tall tree, Rhodolæna altivola and Astrapœa Wallichii, shrubs, are among the most magnificent flowering plants in the world. Disa Buchenaviana, Commelina madagascarica, and Tachiadenus platypterus are fine blue-flowered plants, while the superb orchid Angræcum sesquipedale, Vinca rosea, Euphorbia splendens, and Stephanotis floribunda, have been long cultivated in our hot-houses. There are also many handsome Combretaceæ, Rubiaceæ, and Leguminosæ; but, as in most tropical regions, this wealth of floral beauty has to be searched for, and produces little effect in the landscape.

The affinities of the Madagascar flora are to a great extent in accordance with those of the fauna. The tropical portion of the flora agrees closely with that of tropical Africa, while the plants of the highlands are equally allied to those of the Cape and of the mountains of Central Africa. Some Asiatic types are present which do not occur in Africa; and even the curious American affinities of some of the animals are reproduced in the vegetable kingdom. These last are so interesting that they deserve to be enumerated. An American genus of Euphorbiaceæ, Omphalea, has one species in Madagascar, and Pedilanthus, another genus of the same natural order, has a similar distribution. Myrosma, an American genus of Scitamineæ has one Madagascar species; while the celebrated "travellers' tree," Ravenala madagascariensis, belonging to the order Musaceæ, has its nearest ally in a plant inhabiting N. Brazil and Guiana. Echinolæna, a genus of grasses, has the same distribution.

Of the flora of the smaller Madagascarian islands we possess a fuller account, owing to the recent publication of Mr. Baker's Flora of the Mauritius and the Seychelles, including also Rodriguez. The total number of species in this flora is 1,058, more than half of which (536) are exclusively Mascarene—that is, found only in some of the islands of the Madagascar group, while nearly a third (304) are endemic or confined to single islands. Of the widespread plants sixty-six are found in Africa but not in Asia, and eighty-six in Asia but not in Africa, showing a similar Asiatic preponderance to what is said to occur in Madagascar. With the genera, however, the proportions are different, for I find by going through the whole of the generic distributions as given by Mr. Baker, that out of the 440 genera of wild plants fifty are endemic, twenty-two are Asiatic but not African, while twenty-eight are African but not Asiatic. This implies that the more ancient connection has been on the side of Africa, while a more recent immigration, shown by identity of species, has come from the side of Asia; and it is already certain that when the flora of Madagascar is more thoroughly worked out, a still greater African preponderance will be found in that island.

A few Mascarene genera are found elsewhere only in South America, Australia, or Polynesia; and there are also a considerable number of genera whose metropolis is South America, but which are represented by one or more species in Madagascar, and by a single often widely distributed species in Africa. This fact throws light upon the problem offered by those mammals, reptiles, and insects of Madagascar which now have their only allies in South America, since the two cases would be exactly parallel were the African plants to become extinct. Plants, however, are undoubtedly more long-lived specifically than animals—especially the more highly organised groups, and are less liable to complete extinction through the attacks of enemies or through changes of climate or of physical geography; hence we find comparatively few cases in which groups of Madagascar plants have their only allies in such distant regions as America and Australia, while such cases are numerous among animals, owing to the extinction of the allied forms in intervening areas, for which extinction, as we have already shown, ample cause can be assigned.

Curious Relations of Mascarene Plants.—Among the curious affinities of Mascarene plants we have culled the following from Mr. Baker's volume. Trochetia, a genus of Sterculiaceæ, has four species in Mauritius, one in Madagascar, and one in the remote island of St. Helena. Mathurina, a genus of Turneraceæ, consisting of a single species peculiar to Rodriguez, has its nearest ally in another monotypic genus, Erblichia, confined to Central America. Siegesbeckia, one of the Compositæ, consists of two species, one inhabiting the Mascarene islands, the other Peru. Labourdonasia, a genus of Sapotaceæ, has two species in Mauritius, one in Natal, and one in Cuba. Nesogenes, belonging to the verbena family, has one species in Rodriguez and one in Polynesia. Mespilodaphne, an extensive genus of Lauraceæ, has six species in the Mascarene islands, and all the rest (about fifty species) in South America. Nepenthes, the well-known pitcher plants, are found chiefly in the Malay Islands, South China, and Ceylon, with species in the Seychelles Islands, and in Madagascar. Milla, a large genus of Liliaceæ, is exclusively American, except one species found in Mauritius and Bourbon. Agauria, a genus of Ericaceæ, is found in Madagascar, the Mascarene islands, the plateau of Central Africa, and the Camaroon Mountains in West Africa. An acacia, found in Mauritius and Bourbon (A. heterophylla), can hardly be separated specifically from Acacia koa of the Sandwich Islands. The genus Pandanus, or screw-pine, has sixteen species in the three islands—Mauritius, Rodriguez, and the Seychelles—all being peculiar, and none ranging beyond a single island. Of palms there are fifteen species belonging to ten genera, and all these genera are peculiar to the islands. We have here ample evidence that plants exhibit the same anomalies of distribution in these islands as do the animals, though in a smaller proportion; while they also exhibit some of the transitional stages by which these anomalies have, in all probability, been brought about, rendering quite unnecessary any other changes in the distribution of sea and land than physical and geological evidence warrants.

Fragmentary Character of the Mascarene Flora.—Although the peculiar character and affinities of the vegetation of these islands is sufficiently apparent, there can be little doubt that we only possess a fragment of the rich flora which once adorned them. The cultivation of sugar, and other tropical products, has led to the clearing away of the virgin forests from all the lowlands, plateaus, and accessible slopes of the mountains, so that remains of the aboriginal woodlands only linger in the recesses of the hills, and numbers of forest-haunting plants must inevitably have been exterminated. The result is, that nearly three hundred species of foreign plants have run wild in Mauritius, and have in their turn helped to extinguish the native species. In the Seychelles, too, the indigenous flora has been almost entirely destroyed in most of the islands, although the peculiar palms, from their longevity and comparative hardiness, have survived. Mr. Geoffrey Nevill tells us, that at Mahé, and most of the other islands visited by him, it was only in a few spots near the summits of the hills that he could perceive any remains of the ancient flora. Pine-apples, cinnamon, bamboos, and other plants have obtained a firm footing, covering large tracts of country and killing the more delicate native flowers and ferns. The pine-apple, especially, grows almost to the tops of the mountains. Where the timber and shrubs have been destroyed, the water falling on the surface immediately cuts channels, runs off rapidly, and causes the land to become dry and arid; and the same effect is largely seen both in Mauritius and Bourbon, where, originally, dense forest covered the entire surface, and perennial moisture, with its ever-accompanying luxuriance of vegetation, prevailed.

Flora of Madagascar Allied to that of South Africa.—In my Geographical Distribution of Animals I have remarked on the relation between the insects of Madagascar and those of south temperate Africa, and have speculated on a great southern extension of the continent at the time when Madagascar was united with it. As supporting this view I now quote Mr. Bentham's remarks on the Compositæ. He says: "The connections of the Mascarene endemic Compositæ, especially those of Madagascar itself, are eminently with the southern and sub-tropical African races; the more tropical races, Plucheineæ, &c., may be rather more of an Asiatic type." He further says that the Composite flora is almost as strictly endemic as that of the Sandwich Islands, and that it is much diversified, with evidences of great antiquity, while it shows insular characteristics in the tendency to tall shrubby or arborescent forms in several of the endemic or prevailing genera.

Preponderance of Ferns in the Mascarene Flora.—A striking character of the flora of these smaller Mascarene islands is the great preponderance of ferns, and next to them of orchideæ. The following figures are taken from Mr. Baker's Flora for Mauritius and the Seychelles, and from an estimate by M. Frappier of the flora of Bourbon given in Maillard's volume already quoted:—

The cause of the great preponderance of ferns in oceanic islands has already been discussed in my book on Tropical Nature; and we have seen that Mauritius, Bourbon, and Rodriguez must be classed as such, though from their proximity to Madagascar they have to be considered as satellites to that great island. The abundance of orchids, the reverse of what occurs in remoter oceanic islands, may be in part due to analogous causes. Their usually minute and abundant seeds would be as easily carried by the wind as the spores of ferns, and their frequent epiphytic habit affords them an endless variety of stations on which to vegetate, and at the same time removes them in a great measure from the competition of other plants. When, therefore, the climate is sufficiently moist and equable, and there is a luxuriant forest vegetation, we may expect to find orchids plentiful on such tropical islands as possess an abundance of insects adapted to fertilise them, and which are not too far removed from other lands or continents from which their seeds might be conveyed.

Concluding Remarks on Madagascar and the Mascarene Islands.—There is probably no portion of the globe that contains within itself so many and such varied features of interest connected with geographical distribution, or which so well illustrates the mode of solving the problems it presents, as the comparatively small insular region which comprises the great island of Madagascar and the smaller islands and island-groups which immediately surround it. In Madagascar we have a continental island of the first rank, and undoubtedly of immense antiquity; we have detached fragments of this island in the Comoros and Aldabra; in the Seychelles we have the fragments of another very ancient island, which may perhaps never have been continental; in Mauritius, Bourbon, and Rodriguez we have three undoubtedly oceanic islands; while in the extensive banks and coral reefs of Cargados, Saya de Malha, the Chagos, and the Maldive Isles, we have indications of the submergence of many large islands which may have aided in the transmission of organisms from the Indian Peninsula. But between and around all these islands we have depths of 2,500 fathoms and upwards, which renders it very improbable that there has ever been here a continuous land surface, at all events during the Tertiary or Secondary periods of geology.

It is most interesting and satisfactory to find that this conclusion, arrived at solely by a study of the form of the sea-bottom and the general principle of oceanic permanence, is fully supported by the evidence of the organic productions of the several islands; because it gives us confidence in those principles, and helps to supply us with a practical demonstration of them. We find that the entire group contains just that amount of Indian forms which could well have passed from island to island; that many of these forms are slightly modified species, indicating that the migration occurred during late Tertiary times, while others are distinct genera, indicating a more ancient connection; but in no one case do we find animals which necessitate an actual land-connection, while the numerous Indian types of mammalia, reptiles, birds, and insects, which must certainly have passed over had there been such an actual land-connection, are totally wanting. The one fact which has been supposed to require such a connection—the distribution of the lemurs—can be far more naturally explained by a general dispersion of the group from Europe, where we know it existed in Eocene times; and such an explanation applies equally to the affinity of the Insectivora of Madagascar and Cuba; the snakes (Herpetodryas, &c.) of Madagascar and America; and the lizards (Cryptoblepharus) of Mauritius and Australia. To suppose, in all these cases, and in many others, a direct land-connection, is really absurd, because we have the evidence afforded by geology of wide differences of distribution directly we pass beyond the most recent deposits; and when we go back to Mesozoic—and still more to Palæozoic—times, the majority of the groups of animals and plants appear to have had a world-wide range. A large number of our European Miocene genera of vertebrates were also Indian or African, or even American; the South American Tertiary fauna contained many European types; while many Mesozoic reptiles and mollusca ranged from Europe and North America to Australia and New Zealand.

By very good evidence (the occurrence of wide areas of marine deposits of Eocene age), geologists have established the fact that Africa was cut off from Europe and Asia by an arm of the sea in early Tertiary times, forming a large island-continent. By the evidence of abundant organic remains we know that all the types of large mammalia now found in Africa (but which are absent from Madagascar) inhabited Europe and Asia, and many of them also North America, in the Miocene period. At a still earlier epoch Africa may have received its lower types of mammals—lemurs, insectivora, and small carnivora, together with its ancestral struthious birds, and its reptiles and insects of American or Australian affinity; and at this period it was joined to Madagascar. Before the later continental period of Africa, Madagascar had become an island; and thus, when the large mammalia from the northern continent overran Africa, they were prevented from reaching Madagascar, which thenceforth was enabled to develop its singular forms of low-type mammalia, its gigantic ostrich-like Æpyornis, its isolated birds, its remarkable insects, and its rich and peculiar flora. From it the adjacent islands received such organisms as could cross the sea; while they transmitted to Madagascar some of the Indian birds and insects which had reached them.

The method we have followed in these investigations is to accept the results of geological and palæontological science, and the ascertained facts as to the powers of dispersal of the various animal groups; to take full account of the laws of evolution as affecting distribution, and of the various ocean depths as implying recent or remote union of islands with their adjacent continents; and the result is, that wherever we possess a sufficient knowledge of these various classes of evidence, we find it possible to give a connected and intelligible explanation of all the most striking peculiarities of the organic world. In Madagascar we have undoubtedly one of the most difficult of these problems; but we have, I think, fairly met and conquered most of its difficulties. The complexity of the organic relations of this island is due, partly to its having derived its animal forms from two distinct sources—from one continent through a direct land-connection, and from another by means of intervening islands now submerged; but, mainly to the fact of its having been separated from a continent which is now, zoologically, in a very different condition from that which prevailed at the time of the separation; and to its having been thus able to preserve a number of types which may date back to the Eocene, or even to the Cretaceous, period. Some of these types have become altogether extinct elsewhere; others have spread far and wide over the globe, and have survived only in a few remote countries—and especially in those which have been more or less secured by their isolated position from the incursions of the more highly-developed forms of later times. This explains why it is that the nearest allies of the Madagascar fauna and flora are now so often to be found in South America or Australia—countries in which low forms of mammalia and birds still largely prevail;—it being on account of the long-continued isolation of all these countries that similar forms (descendants of ancient types) are preserved in them. Had the numerous suggested continental extensions connecting these remote continents at various geological periods been realities, the result would have been that all these interesting archaic forms, all these defenceless insular types, would long ago have been exterminated, and one comparatively monotonous fauna have reigned over the whole earth. So far from explaining the anomalous facts, the alleged continental extensions, had they existed, would have left no such facts to be explained.

149 See Dr. J. E. Gray's "Revision of the Viverridæ," in ''Proc. Zool. Soc.'' 1864, p. 507.

150 Some of the Bats of Madagascar and East Africa are said to have their nearest allies in Australia. (See Dobson in Nature, Vol. XXX. p. 575.)

151 This view was, I believe, first advanced by Professor Huxley in his "Anniversary Address to the Geological Society," in 1870. He says:—"In fact the Miocene mammalian fauna of Europe and the Himalayan regions contain, associated together, the types which are at present separately located in the South African and Indian provinces of Arctogæa. Now there is every reason to believe, on other grounds, that both Hindostan south of the Ganges, and Africa south of the Sahara, were separated by a wide sea from Europe and North Asia during the Middle and Upper Eocene epochs. Hence it becomes highly probable that the well-known similarities, and no less remarkable differences, between the present faunæ of India and South Africa have arisen in some such fashion as the following: Some time during the Miocene epoch, the bottom of the nummulitic sea was upheaved and converted into dry land in the direction of a line extending from Abyssinia to the mouth of the Ganges. By this means the Dekkan on the one hand and South Africa on the other, became connected with the Miocene dry land and with one another. The Miocene mammals spread gradually over this intermediate dry land; and if the condition of its eastern and western ends offered as wide contrasts as the valleys of the Ganges and Arabia do now, many forms which made their way into Africa must have been different from those which reached the Dekkan, while others might pass into both these sub-provinces."

This question is fully discussed in my Geographical Distribution of Animals (Vol. I., p. 285), where I expressed views somewhat different from those of Professor Huxley, and made some slight errors which are corrected in the present work. As I did not then refer to Professor Huxley's prior statement of the theory of Miocene immigration into Africa (which I had read but the reference to which I could not recall) I am happy to give his views here.

152 The total number of Madagascar birds is 238, of which 129 are absolutely peculiar to the island, as are thirty-five of the genera. All the peculiar birds but two are land birds. These are the numbers given in M. Grandidier's great work on Madagascar.

153 The Ibis, 1877, p. 334.

154 In a paper read before the Geological Society in 1874, Mr. H. F. Blanford, from the similarity of the fossil plants and reptiles, supposed that India and South Africa had been connected by a continent, "and remained so connected with some short intervals from the Permian up to the end of the Miocene period," and Mr. Woodward expressed his satisfaction with "this further evidence derived from the fossil flora of the Mesozoic series of India in corroboration of the former existence of an old submerged continent—Lemuria."

Those who have read the preceding chapters of the present work will not need to have pointed out to them how utterly inconclusive is the fragmentary evidence derived from such remote periods (even if there were no evidence on the other side) as indicating geographical changes. The notion that a similarity in the productions of widely separated continents at any past epoch is only to be explained by the existence of a direct land-connection, is entirely opposed to all that we know of the wide and varying distribution of all types at different periods, as well as to the great powers of dispersal over moderate widths of ocean possessed by all animals except mammalia. It is no less opposed to what is now known of the general permanency of the great continental and oceanic areas; while in this particular case it is totally inconsistent (as has been shown above) with the actual facts of the distribution of animals.

155 Geographical Distribution of Animals, Vol. I., pp. 272-292.

156 The term "Mascarene" is used here in an extended sense, to include all the islands near Madagascar which resemble it in their animal and vegetable productions.

157 For the birds of the Comoro Islands see ''Proc. Zool. Soc.'', 1877, p. 295, and 1879, p. 673.

158 The following is a list of these peculiar birds. (See the Ibis, for 1867, p. 359; and 1879, p. 97.)

159 Specimens are recorded from West Africa in the Proceedings of the Academy of Natural Science, Philadelphia, 1857, p. 72, while specimens in the Paris Museum were brought by D'Orbigny from S. America. Dr. Wright's specimens from the Seychelles have, as he informs me, been determined to be the same species by Dr. Peters of Berlin.

160 "Additional Notes on the Land-shells of the Seychelles Islands." By Geoffrey Nevill, C.M.Z.S. ''Proc. Zool. Soc.'' 1869, p. 61.

161 In Maillard's Notes sur l'Isle de Réunion, a considerable number of mammalia are given as "wild," such as Lemur mongoz and Centetes setosus, both Madagascar species, with such undoubtedly introduced animals as a wild cat, a hare, and several rats and mice. He also gives two species of frogs, seven lizards, and two snakes. The latter are both Indian species and certainly imported, as are most probably the frogs. Legouat, who resided some years in the island nearly two centuries ago, and who was a closer observer of nature, mentions numerous birds, large bats, land-tortoises, and lizards, but no other reptiles or venomous animals except scorpions. We may be pretty sure, therefore, that the land-mammalia, snakes, and frogs, now found wild, have all been introduced. Of lizards, on the other hand, there are several species, some peculiar to the island, others common to Africa and the other Mascarene Islands. The following list by Prof. Dumeril is given in Maillard's work:—

Four species of chameleon are now recorded from Bourbon and one from Mauritius (J. Reay Greene, M.D., in Pop. Science Rev. April, 1880), but as they are not mentioned by the old writers, it is pretty certain that these creatures are recent introductions, and this is the more probable as they are favourite domestic pets.

Darwin informed me that in a work entitled Voyage à l'Isle de France, par un Officier du Roi, published in 1770, it is stated that a fresh-water fish had been introduced from Batavia and had multiplied. The writer also says (p. 170): "On a essayé, mais sans succès, d'y transporter des grenouilles qui mangent les œufs que les moustigues deposent sur les eaux stagnantes." It thus appears that there were then no frogs on the island.

162 That the dodo is really an abortion from a more perfect type, and not a direct development from some lower form of wingless bird, is shown by its possessing a keeled sternum, though the keel is exceedingly reduced, being only three-quarters of an inch deep in a length of seven inches. The most terrestrial pigeon—the Didunculus of the Samoan Islands, has a far deeper and better developed keel, showing that in the case of the dodo the degradation has been extreme. We have also analogous examples in other extinct birds of the same group of islands, such as the flightless Rails—Aphanapteryx of Mauritius and Erythromachus of Rodriguez, as well as the large parrot—Lophopsittacus of Mauritius, and the Night Heron, Nycticorax megacephala of Rodriguez, the last two birds probably having been able to fly a little. The commencement of the same process is to be seen in the peculiar dove of the Seychelles, Turtur rostratus, which, as Mr. Edward Newton has shown, has much shorter wings than its close ally, T. picturatus, of Madagascar. For a full and interesting account of these and other recently extinct birds see Professor Newton's article on "Fossil Birds" in the Encyclopædia Britannica, ninth edition, vol. iii., p. 732; and that on "The Extinct Birds of Rodriguez," by Dr. A. Günther and Mr. E. Newton, in the Royal Society's volume on the Transit of Venus Expedition.

163 See Ibis, 1877, p. 334.

164 A common Indian and Malayan toad (Bufo melanostictus) has been introduced into Mauritius and also some European toads, as I am informed by Dr. Günther.

165 This brief account of the Madagascar flora has been taken from a very interesting paper by the Rev. Richard Baron, F.L.S., F.G.S., in the Journal of the Linnean Society, Vol. XXV., p. 246; where much information is given on the distribution of the flora within the island.

166 It may be interesting to botanists and to students of geographical distribution to give here an enumeration of the endemic genera of the Flora of the Mauritius and the Seychelles, as they are nowhere separately tabulated in that work.

Among the curious features in this list are the great number of endemic shrubs in Mauritius, and the remarkable assemblage of five endemic genera of palms in the Seychelles Islands. We may also notice that one palm (Latania loddigesii) is confined to Round Island and two other adjacent islets offering a singular analogy to the peculiar snake also found there.