Contemporary literature is full of devices for the mechanical dispersal of animals. Marine currents, warm and cold, were favoured all the more since they showed the probable original homes of the creatures in question. If these could not stand sea-water, they floated upon logs or icebergs, or they were blown across by storms; fishes were lifted over barriers by waterspouts, and there is on record even an hypothetical land tortoise, full of eggs, which colonised an oceanic island after a perilous sea voyage upon a tree trunk. Accidents will happen, and beyond doubt many freaks of discontinuous distribution have to be accounted for by some such means. But whilst sufficient for the scanty settlers of true oceanic islands, they cannot be held seriously to account for the rich fauna of a large continent, over which palaeontology shows us that the immigrants have passed like waves. It should also be borne in mind that there is a great difference between flotsam and jetsam. A current is an extension of the same medium and the animals in it may suffer no change during even a long voyage, since they may be brought from one litoral to another where they will still be in the same or but slightly altered environment. But the jetsam is in the position of a passenger who has been carried off by the wrong train. Almost every year some American land birds arrive at our western coasts and none of them have gained a permanent footing although such visits must have taken place since prehistoric times. It was therefore argued that only those groups of animals should be used for locating and defining regions which were absolutely bound to the soil. This method likewise gave results not reconcilable with each other, even when the distribution of fossils was taken into account, but it pointed to the absolute necessity of searching for former land-connections regardless of their extent and the present depths to which they may have sunk.
That the key to the present distribution lies in the past had been felt long ago, but at last it was appreciated that the various classes of animals and plants have appeared in successive geological epochs and also at many places remote from each other. The key to the distribution of any group lies in the configuration of land and water of that epoch in which it made its first appearance. Although this sounds like a platitude, it has frequently been ignored. If, for argument's sake, Amphibia were evolved somewhere upon the great southern land-mass of Carboniferous times (supposed by some to have stretched from South America across Africa to Australia), the distribution of this developing class must have proceeded upon lines altogether different from that of the mammals which dated perhaps from lower Triassic times, when the old south continental belt was already broken up. The broad lines of this distribution could never coincide with that of the other, older class, no matter whether the original mammalian centre was in the Afro-Indian, Australian, or Brazilian portion. If all the various groups of animals had come into existence at the same time and at the same place, then it would be possible, with sufficient geological data, to construct a map showing the generalised results applicable to the whole animal kingdom. But the premises are wrong. Whatever regions we may seek to establish applicable to all classes, we are necessarily mixing up several principles, namely geological, historical, i.e. evolutionary, with present day statistical facts. We might as well attempt one compound picture representing a chick's growth into an adult bird and a child's growth into manhood.
In short there are no general regions, not even for each class separately, unless this class be one which is confined to a comparatively short geological period. Most of the great classes have far too long a history and have evolved many successive main groups. Let us take the mammals. Marsupials live now in Australia and in both Americas, because they already existed in Mesozoic times; Ungulata existed at one time or other all over the world except in Australia, because they are post-Cretaceous; Insectivores, although as old as any Placentalia, are cosmopolitan excepting South America and Australia; Stags and Bears, as examples of comparatively recent Arctogaeans, are found everywhere with the exception of Ethiopia and Australia. Each of these groups teaches a valuable historical lesson, but when these are combined into the establishment of a few mammalian "realms," they mean nothing but statistical majorities. If there is one at all, Australia is such a realm backed against the rest of the world, but as certainly it is not a mammalian creative centre!
Well then, if the idea of generally applicable regions is a mare's nest, as was the search for the Holy Grail, what is the object of the study of geographical distribution? It is nothing less than the history of the evolution of life in space and time in the widest sense. The attempt to account for the present distribution of any group of organisms involves the aid of every branch of science. It bids fair to become a history of the world. It started in a mild, statistical way, restricting itself to the present fauna and flora and to the present configuration of land and water. Next came Oceanography concerned with the depths of the seas, their currents and temperatures; then enquiries into climatic changes, culminating in irreconcilable astronomical hypotheses as to glacial epochs; theories about changes of the level of the seas, mainly from the point of view of the physicist and astronomer. Then came more and more to the front the importance of the geological record, hand in hand with the palaeontological data and the search for the natural affinities, the genetic system of the organisms. Now and then it almost seems as if the biologists had done their share by supplying the problems and that the physicists and geologists would settle them, but in reality it is not so. The biologists not only set the problems, they alone can check the offered solutions. The mere fact of palms having flourished in Miocene Spitzbergen led to an hypothetical shifting of the axis of the world rather than to the assumption, by way of explanation, that the palms themselves might have changed their nature. One of the most valuable aids in geological research, often the only means for reconstructing the face of the earth in by-gone periods, is afforded by fossils, but only the morphologist can pronounce as to their trustworthiness as witnesses, because of the danger of mistaking analogous for homologous forms. This difficulty applies equally to living groups, and it is so important that a few instances may not be amiss.
There is undeniable similarity between the faunas of Madagascar and South America. This was supported by the Centetidae and Dendrobatidae, two entire "families," as also by other facts. The value of the Insectivores, Solenodon in Cuba, Centetes in Madagascar, has been much lessened by their recognition as an extremely ancient group and as a case of convergence, but if they are no longer put into the same family, this amendment is really to a great extent due to their widely discontinuous distribution. The only systematic difference of the Dendrobatidae from the Ranidae is the absence of teeth, morphologically a very unimportant character, and it is now agreed, on the strength of their distribution, that these little arboreal, conspicuously coloured frogs, Dendrobates in South America, Mantella in Madagascar, do not form a natural group, although a third genus, Cardioglossa in West Africa, seems also to belong to them. If these creatures lived all on the same continent, we should unhesitatingly look upon them as forming a well-defined, natural little group. On the other hand the Aglossa, with their three very divergent genera, namely Pipa in South America, Xenopus and Hymenochirus in Africa, are so well characterised as one ancient group that we use their distribution unhesitatingly as a hint of a former connection between the two continents. We are indeed arguing in vicious circles. The Ratitae as such are absolutely worthless since they are a most heterogeneous assembly, and there are untold groups, of the artificiality of which many a zoo-geographer had not the slightest suspicion when he took his statistical material, the genera and families, from some systematic catalogues or similar lists. A lamentable instance is that of certain flightless Rails, recently extinct or sub-fossil, on the isalnds of Mauritius, Rodriguez and Chatham. Being flightless they have been used in support of a former huge Antarctic continent, instead of ruling them out of court as Rails which, each in its island, have lost the power of flight, a process which must have taken place so recently that it is difficult, upon morphological grounds, to justify their separation into Aphanapteryx in Mauritius, Erythromachus in Rodriguez and Diaphorapteryx on Chatham Island. Morphologically they may well form but one genus, since they have sprung from the same stock and have developed upon the same lines; they are therefore monogenetic: but since we know that they have become what they are independently of each other (now unlike any other Rails), they are polygenetic and therefore could not form one genus in the old Darwinian sense. Further, they are not a case of convergence, since their ancestry is not divergent but leads into the same stratum.
THE RECONSTRUCTION OF THE GEOGRAPHY OF SUCCESSIVE EPOCHS.
A promising method is the study by the specialist of a large, widely distributed group of animals from an evolutionary point of view. Good examples of this method are afforded by A.E. Ortmann's ("The geographical distribution of Freshwater Decapods and its bearing upon ancient geography", "Proc. Amer. Phil. Soc." Vol. 41, 1902.) exhaustive paper and by A.W. Grabau's "Phylogeny of Fusus and its Allies" ("Smithsonian Misc. Coll." 44, 1904.) After many important groups of animals have been treated in this way—as yet sparingly attempted—the results as to hypothetical land-connections etc. are sure to be corrective and supplementary, and their problems will be solved, since they are not imaginary.
The same problems are attacked, in the reverse way, by starting with the whole fauna of a country and thence, so to speak, letting the research radiate. Some groups will be considered as autochthonous, others as immigrants, and the directions followed by them will be inquired into; the search may lead far and in various directions, and by comparison of results, by making compound maps, certain routes will assume definite shape, and if they lead across straits and seas they are warrants to search for land-connections in the past. (A fair sample of this method is C.H. Eigenmann's "The Freshwater Fishes of South and Middle America", "Popular Science Monthly", Vol. 68, 1906.) There are now not a few maps purporting to show the outlines of land and water at various epochs. Many of these attempts do not tally with each other, owing to the lamentable deficiencies of geological and fossil data, but the bolder the hypothetical outlines are drawn, the better, and this is preferable to the insertion of bays and similar detail which give such maps a fallacious look of certainty where none exists. Moreover it must be borne in mind that, when we draw a broad continental belt across an ocean, this belt need never have existed in its entirety at any one time. The features of dispersal, intended to be explained by it, would be accomplished just as well by an unknown number of islands which have joined into larger complexes while elsewhere they subsided again: like pontoon-bridges which may be opened anywhere, or like a series of superimposed dissolving views of land and sea-scapes. Hence the reconstructed maps of Europe, the only continent tolerably known, show a considerable number of islands in puzzling changes, while elsewhere, e.g. in Asia, we have to be satisfied with sweeping generalisations.
At present about half-a-dozen big connections are engaging our attention, leaving as comparatively settled the extent and the duration of such minor "bridges" as that between Africa and Madagascar, Tasmania and Australia, the Antilles and Central America, Europe and North Africa. (Not a few of those who are fascinated by, and satisfied with, the statistical aspect of distribution still have a strong dislike to the use of "bridges" if these lead over deep seas, and they get over present discontinuous occurrences by a former "universal or sub-universal distribution" of their groups.) This is indeed an easy method of cutting the knot, but in reality they shunt the question only a stage or two back, never troubling to explain how their groups managed to attain to that sub-universal range; or do they still suppose that the whole world was originally one paradise where everything lived side by side, until sin and strife and glacial epochs left nothing but scattered survivors?
The permanence of the great ocean-basins had become a dogma since it was found that a universal elevation of the land to the extent of 100 fathoms would produce but little changes, and when it was shown that even the 1000 fathom-line followed the great masses of land rather closely, and still leaving the great basins (although transgression of the sea to the same extent would change the map of the world beyond recognition), by general consent one mile was allowed as the utmost speculative limit of subsidence. Naturally two or three miles, the average depth of the oceans, seems enormous, and yet such a difference in level is as nothing in comparison with the size of the Earth. On a clay model globe ten feet in diameter an ocean bed three miles deep would scarcely be detected, and the highest mountains would be smaller than the unavoidable grains in the glazed surface of our model. There are but few countries which have not be submerged at some time or other.