CHAPTER VIII.

PROOFS FROM GEOGRAPHICAL DISTRIBUTION OF ORGANISMS.

It is well known that the kinds of organisms found in widely-separated countries differ more or less conspicuously. The traveler in Australia or in Africa finds all, the traveler in Europe nearly all, the animals and plants wholly different from those he has been accustomed to see at home. Even the visitor from the Atlantic to the Pacific coast, if he observes at all, will find nearly all organisms strange to him. The facts of geographical diversity of organisms are so numerous and complex that, at first sight, they seem utterly lawless. Only recently this subject has been redeemed from chaos and reduced to something like order and law by the light thrown upon it by the theory of evolution. We will give, in very brief outline, the most important facts, and then show how they may be explained.

=Geographical Faunas and Floras.=--The group of animals and plants inhabiting any locality, whether peculiar to that locality or not, is called, in popular language, its fauna and flora. But, in a true scientific sense, a fauna and flora is a _natural_ group of animals and plants in one place, _differing_ more or less conspicuously from other groups in other places, and _separated from them by physico-geographical boundaries, or by physical conditions of some kind._ The members of such a group can only exist in certain harmonic relations with external conditions, and with one another. These relations with one another are often complex and nicely adjusted, so that change in one term is propagated through the whole series of terms, giving rise often to the most unexpected results, until finally a new equilibrium is established. Thus, the destruction of certain insectivorous birds, in mere wanton sport, may give rise to the multiplication of insect pests, and this to the destruction of certain kinds of plants, and this to the diminution of certain herbivores, and this in its turn to the disappearance of certain carnivores. It is well known that the introduction of rabbits into New Zealand and Australia has produced the most unexpectedly disastrous effect upon certain crops, on account of the absence of the fierce and active carnivores which keep in check their excessive multiplication in Europe.

Now, among the physical conditions which limit faunas and floras, and separate them from each other, the most important and universal is temperature.

=Temperature-Regions.=--If we travel from equator to pole, we pass through mean temperatures varying from 80° to 0°. This gives rise to a very regular zonal arrangement of plant-forms: 1. We have first a region in which palms and palm-like forms are abundant and characteristic, and which therefore may be called the region of palms. It corresponds with the tropic zone. 2. We next have a region in which hard-wood foliferous trees are most abundant and characteristic; first mostly evergreens and then deciduous trees, and therefore may be called the region of hard-wood forests. This corresponds with the temperate-zone. 3. Then we find a region characterized predominantly by pines and pine-like trees and birches, and may be called the region of pines. This is the sub-Arctic region. 4. Then a region without trees, but only shrubs and summer plants. This is the Arctic region. 5. And, finally, an almost wholly plantless region of perpetual ice--the polar region.

These regions are determined wholly by temperature, and therefore, in going up a mountain-slope to snowy summits, we pass through similar regions in smaller space. For example, in going from sea-level to the summits of the Sierra, 14,000 to 15,000 feet high, we commence in a region of predominantly hard-wood trees; but at 3,000 feet the forests become almost wholly coniferous, at 11,000 to 12,000 feet the vegetation becomes shrubby, and at 13,000 feet we reach perpetual snow.

We have taken plants first, because these, being fixed to the soil and incapable of voluntary seasonal migrations, are more strictly and simply limited by temperature--i. e., the arrangement of different kinds in zones is more simple and conspicuous. But the same rule holds also for animals. In passing from equator to pole, animal kinds also change frequently, so that there are many temperature-faunas in which the animals are all very different. In both animals and plants, species, genera, families, etc., are limited by temperature. These are familiar facts; we recall them to the reader in order that we may base thereon a clearer definition of these limits.

=More Perfect Definition of Regions.=--1. The area over which any form spreads is called its _range_. Now, the range of a species is more restricted than that of a genus, because, when a species is limited by temperature, another species of the same genus may carry on the genus. For the same reason the range of a family is usually greater than that of a genus, and so on for higher classification-groups. For example, pines range on the slopes of the Sierra from about 2,000 feet to 11,000 feet, but not the same species. In ascending, we meet first the nut-pine (_Pinus Sabiniana_), then the yellow-pine (_P. ponderosa_), then the sugar-pine (_P. Lambertiana_), then the tamarack-pine (_P. contorta_), and last, the _Pinus flexilis_, etc.

2. Where two contiguous temperature-regions come in contact, there is no sharp line between; on the contrary, they _shade gradually_, almost imperceptibly, into one another, the ranges of species overlapping and interpenetrating, and the two species coexisting on the borders of their ranges. This is represented by the diagram (Fig. 64), in which the horizontal lines represent the north and south ranges of species of two groups, A and B, separated by the dotted line.

3. Species also pass out gradually on the borders of these ranges and others come in gradually, so _far as number and vigor of individuals are concerned_. If _a a′_ and _b b′_ (Fig. 65) represent the north and south range of two species, and _b a′_ their overlap or area of coexistence, then the height of the curves A and B will represent the number and vigor of the individuals in different parts of the range.

4. While, therefore, there is a shading of contiguous groups into each other by overlap of species-ranges; while there is also a gradual passing out of species so far as number and vigor of individuals is concerned, yet, in _specific characters_ we observe usually no such gradation. Species seem to come in on one border with all their specific characters perfect, remain substantially unchanged throughout their range, and pass out on the other border, still the same species. In other words, one species takes the place of another, usually by _substitution_, not by _transmutation_. It is _as if_ species had originated, no matter how, each in its own region, and had spread in all directions as far as physical conditions and struggle with other species would allow. This important subject will be more fully discussed later.

5. We have thus far spoken of species as limited by temperature alone, but they are limited also by _barriers_. If, then, there be an east and west barrier, such as a high mountain-range, or a wide sea or desert, there will be no shading or gradation of any kind, because the barrier prevents overlapping, interpenetration, and struggle on the margins. For example: The species north and south of the Himalayas, or north and south of Sahara, are widely different. It is, again, _as if_ they originated each where we find them and spread as far as they could, but the physical barrier prevented mingling and shading.

6. There are temperature-regions south as well as north of the equator. Now, although the climatic conditions are quite similar, the species of corresponding temperature-regions north and south are wholly different. It is, again, as if they originated where we find them, and were kept separate by the barrier of tropical heat between. If carried over, they often do perfectly well.

_Continental Faunas and Floras._

If the land-surfaces were continuous all around the globe, there is little doubt that each temperature region with its characteristic species would also be substantially continuous. There would, it is true, be some local variations dependent upon soil and humidity, etc., but substantially the same species would exist all around. The distribution would be almost wholly zonal. But the intervening oceans are complete barriers to continental species. Hence we ought to expect, and do find, that the faunas and floras of different continents are almost totally different. Each apparently originated on its own continent, and did not spread to other continents, only because they could not get there. It is necessary to explain this in more detail.

Fig. 66 represents a polar view of the earth, showing the eastern and western continents, and the five temperature zones already described. Now, if we examine the species in each region, commencing at the pole, we find that those of Nos. 5 and 4 are almost identical all around. The reason is obvious. The continents come close together there, with ice-connection if not land-connection all around. There is but one circumpolar region. But, as soon as we come down to No. 3 and No. 2, the species on the two continents are nearly all different, because there is an impassable barrier between, either in the form of ocean or of Arctic cold. For example, the animals and plants inhabiting the United States are almost wholly different from those in Europe, not only in species, but even largely in genera and to some extent in families. There are some exceptions to this rule, but these are of the kind which prove the rule, or rather the principle on which the rule is founded. These exceptions are mainly of three kinds: 1. _Introduced species._--All our weeds, many garden-plants, and many animal pests are of this kind. They were not found here when America was discovered, only because they could not get here; for, when brought here, they do so well that they often overrun the country and dispossess the native species, as we ourselves have done the Indians. 2. _Hardy or else wide-migrating species._--Hardy species have wide range; they may belong to No. 4 as well as No. 3. If so, they range down to No. 3 on both continents. Migrating birds, such as ducks and geese, etc., breed in summer in No. 4, and migrate southward in winter on both continents from the common circumpolar ground. 3. _Alpine species._--It is a curious fact that species on tops of snowy mountains in temperate regions of the two continents are wonderfully similar, though so completely isolated. We are not yet prepared to discuss this point. We shall do so later. Suffice it to say now that it can be completely explained.

In region No. 1 the continental diversity is still greater. Not only species and genera, but whole families and even orders, are peculiar to each continent. The great pachyderms--elephant, rhinoceros, hippopotamus--are peculiar to the Eastern; the edentates--sloths and armadillos--to the Western. The humming-birds, those gems of the forests, of which there are over four hundred species, and the whole cactus family, are peculiar to America, while the tailless monkeys are equally characteristic of the Eastern Continent.

The continents do not come together again toward the south, and, therefore, as might be expected, the great difference between the two persists to the southern points. The faunas of the southern points of South America, Africa, and Australia are very different.

=Subdivisions of Continental Faunas and Floras.=--Besides the subdivisions of continental faunas, north and south, determined by temperature as already explained, if there be in any continent an impassable barrier running north and south, there will be a corresponding difference in the species on the two sides, east and west. We give but one example: The North American Cordilleras or Rocky Mountains, with their high ranges and desert plains, constitute a very great barrier between the eastern and western portions of the United States. Hence, we find an extraordinary difference between the species inhabiting California and those found in the eastern portion of the country. Speaking generally, all the species and many of the genera are peculiar. The exceptions, too, are significant. Leaving out introduced species, of which there are many, they are mostly strong-winged or widely-migrating birds, such as the turtle-dove, the turkey-buzzard, the bald eagle, and, of course, many water-birds.

=Special Cases.=--If any body of land is widely separated from all other lands by deep seas, we invariably find a corresponding peculiarity of its species. Thus, the species inhabiting Australia and Madagascar are perhaps the most peculiar in the world. We do not dwell further on these, because we will discuss them hereafter. There is a little group of very small islands--the Galapagos--about six hundred miles off the western coast of South America, and surrounded on all sides by deep sea. These islands are stocked with a collection of curious animals not found elsewhere on the surface of the earth; but among them are no mammals at all. We might multiply examples without limit. Even the rivers emptying in the same sea sometimes have each its peculiar species of mussels. In the Altamaha River there are several species of unios--such, for instance, as the wonderful spinous unio--not found elsewhere. How came they there? Howsoever they may have come there, they are now kept isolated there by barriers of land and of salt water.

Many other curious details will come up in our discussion of the origin of diversity.

=Marine Species.=--Precisely the same principles apply here; but diversity in the case of marine species is perhaps less marked, and certainly less general, because of the universal oceanic connection. Open-sea species are therefore almost universal. But many marine species are confined to shallow water, and therefore to shore-lines. The species on the two shores of the same ocean, or the two coasts of the same continent, are different, being isolated east and west by barriers of deep sea or of land, and north and south by temperature. Also about isolated lands, like Australia and Madagascar, the species are peculiar.

Thus, then, species, genera, etc., are limited in every direction; north and south by temperature, and in all directions by barriers, in the form of oceans, deserts, and mountain-chains. Add to these, peculiar climates and soils, and we see that, from this point of view, the whole surface of the earth may be divided and sub-divided into regions, sub-regions, provinces, etc. It would carry us too far to explain the primary and secondary divisions adopted by Mr. Wallace, and the somewhat different ones suggested by Mr. Allen. Our main object is to discuss the _cause_ of this diversity, and especially to show the light shed upon it by the theory of evolution. We have only given a sketch of the facts sufficient for this purpose.

_Theory of the Origin of Geographical Diversity._

It will be observed that all along we have assumed a sort of provisional theory. We have said in every case, it is _as if_ organic forms originated where we find them, and have gone thence wherever they could--as far in every direction as physical conditions and struggle with competing species would allow. This view has been formulated as the “theory of specific centers of origin.” There would be less objection to this as a first provisional theory did it not assume a supernatural _mode_ of origin. But, in the minds of those who hold it, it has usually assumed expressly or tacitly the form of “_specific centers of creation_,” thus implying the immutability of specific types and the supernaturalism of specific origin (page 68). In this latter or usual form it completely fails to account for the facts given above. For, if this were the mode of origin, each species ought in every case to be perfectly adapted to its own environment, and to no other. But, on the contrary, introduced species often flourish better than in their own country, and better than the natives of their new homes. In the less objectionable form of “specific centers of origin,” without defining the mode of origin, it accounts well for many of the more obvious facts of geographical diversity, as it _now_ exists, but not all. According to this view, the amount of diversity ought to be in strict proportion to the completeness of isolation, or impassableness of the separating barriers; but this is not _exactly_ true. There is another element, not yet mentioned, which is just as important as impassableness, but which until recently has been left entirely out of account. This is the element of _time_--the amount of time since the barrier was set up, or during which it has continued to exist. These two elements, it is true, are closely connected with each other; for, since all changes in physical geography have taken place very slowly--since barriers in the form of mountain-ranges and seas have increased by slow process of growth--it is evident that impassableness is, to some extent, a measure of time. But they are by no means in strict proportion. The one or the other may predominate.

Now, this time-element connects geographical distribution with changes of physical geography and climate in _geological_ times, and especially with the _latest_ of these changes, viz., those occurring during the _Glacial epoch_. During that remarkable epoch extraordinary changes of climate, from extreme Arctic rigor to great mildness, enforced wide migrations of species southward and northward; while concomitant changes of physical geography, by elevation of the earth’s crust over wide areas, opened highways between previously-isolated continents, permitting migrations in various directions, and by subsequent depression again isolating the migrated species in their new homes. It is evident, then, that the recognition of the element of almost unlimited time at once introduces into the question of geographical distribution the _idea of evolution_. If the study of geographical distribution, as _it now exists_, and as a part of science of physical geography, gave rise naturally to the theory of “specific centers of origin,” the study of the same, in connection with geological time, and as a part of geological science, now demands its explanation by the theory of evolution.

It must be borne in mind, then, that geographical diversity of organisms is not a question of the present epoch only. There has been geographical diversity in every previous geological epoch; it is, therefore, a question of geology as well as of biology. It is probable, however, that diversity has increased with the course of geological times, and is greater now than ever before. In other words, in the evolution of the organic kingdom, the law of differentiation has prevailed here, as in other departments of biology. A clear statement of the causes of the _present_ distribution of organisms must embrace also the causes of geographical diversity _generally_. We give, therefore, at once a brief statement of what seems to us the most probable view, and shall then proceed to show how it explains the present distribution.

=Most Probable View of the General Process.=--Bearing in mind, then, this time-element, the phenomena of geographical diversity are best explained by the following suppositions: 1. A gradual progressive movement (evolution) of the organic kingdom, marching, as it were, abreast, at equal rate along the whole line--i. e., in all parts of the earth, and throughout all geological times, under the action of all the forces or factors, and following all the laws, of evolution already explained (pages 19 and 73). If this were all, there would be no _geographical_ diversity, although _organic diversity_ might be as great as it is now. There would be differentiation of forms and structure everywhere, but no differentiation of groups in different localities. 2. Under the influence of different conditions in different places, more or less isolated from one another by climatic or physical barriers, the onward movement (evolution) of organic forms takes different directions and different rates, and gives rise to local groups, which become more and more differentiated, without limit as time goes on. This element, acting by itself throughout all geological times, would ere this have produced an extreme geographical diversity, such as does not anywhere exist. 3. From time to time, at long intervals, extensive changes of physical geography and climate, produced by crust elevations, partly enforce by change of temperature, and partly permit by opening of gateways, extensive migrations and dispersals of species, by which mingling and struggle for life and final readjustment takes place, and extreme diversity is prevented. Such mingling of different faunas and floras on the same ground, and the severe struggle for life that thus ensues, and the survival of the fittest in many directions, are, as already shown, among the most powerful factors of evolution. They tend to _increase organic_ diversity, but to _diminish geographical_ diversity. 4. At the close of such great periods of change as indicated in the last, by contrary movement of the earth-crust--i. e., subsidence--new barriers are set up and new isolations are produced, and the process of divergence again commences and increases steadily so long as the barriers continue to exist.

Now, the last of these periods of great changes and extensive migrations, and subsequent isolations, was the Glacial epoch. It was this epoch, therefore, which mainly determined the present geographical distribution of species. Thus, the present distribution is a key to the directions of the last great migrations, and therefore to the nature of the changes in physical geography and climate which then occurred; and, conversely, the character of these changes, determined in other ways, _furnishes the only key to the present distribution of species_.

Before applying the foregoing principles in the explanation of special cases, it may be well to give a very brief outline of the condition of things during the Glacial epoch.

In America, during this epoch, by increasing cold the southern margin of the great northern ice-sheet crept slowly southward, until it reached the latitude of about 38° to 40°. Arctic species were thus driven southward slowly, from generation to generation, until they occupied the whole of the United States, as far as the shores of the Gulf, while temperate species were forced still farther south, into Central and South America. This period of extreme rigor and southward migration was followed by a period of great mildness, during which the ice and its accompanying Arctic conditions retreated northward, followed by Arctic species. More than one advance and retreat, apparently, occurred during this time. Again, during the same time, brought about by northern elevation, there was broader connection than now exists between North and South America, and free migrations between, in both directions, enforced by extreme changes in temperature. Also, during this or previous time, there were broad connections between North America and Asia, in the region of Behring Strait, and between America and Europe, in high-latitude regions, and extensive migrations of faunas and floras between were thus permitted. The necessary result of all these migrations of species, partly enforced by changes of climate, partly permitted by opening of gateways since closed, was exceptionally rapid changes in organic forms. This was the result of two causes: First, the severer pressure of a changing physical environment; and, second, a severer struggle for life between the natives and the invaders.

In Europe, during the same time and from similar causes, there were at least three or four different faunas struggling together for mastery on the same soil. First, there were the Pliocene indigenes, who had, if any, pre-emption right to the soil; second, invaders from Arctic regions, driven southward by increasing cold; third, invaders from Asia, permitted by the removal of the old sea-barrier which once extended from the Black Sea to the Arctic, and of which the Caspian and Aral are existing remnants, and thus opening a gateway for migration which has remained open ever since; fourth, invaders from Europe and Asia into Africa, and sometimes back again into Europe, by opening of gateways through the Mediterranean, which have been since closed. One of these highways was through Gibraltar, and one from Italy to Africa through Sicily. As in America, so here, in even greater degree, the severe pressure of changing environment and the severe struggle for life produced rapid changes of organic forms. Many species were destroyed; others saved themselves by modifications adapted more perfectly to the changed conditions. There is little doubt that man came into Europe with the Asiatic invasion, and was one of the principal agents of change, especially in the way of destruction of many old forms.

Such is a very brief outline of the last great geological change and its general results. Being the last, this one has left the strongest and most universal impress on the _present_ geographical distribution. But similar changes by crust oscillations, if not also by extreme changes of climate, have repeatedly occurred in geological times, and some of the most remarkable geographical faunas and floras are the result of these earlier geological changes. We will now give a few examples illustrating these principles:

1. _Australia_ is undoubtedly more peculiar in its fauna and flora than any other known country. Not only are all its species peculiar, not found elsewhere on the face of the earth, but its genera, its families, and even many of its orders of animals and plants, are also peculiar. These facts are so familiar that it is unnecessary to dwell on them. I need only mention, among plants, the whole of the simple-leaved acacias, already mentioned on page 86, of which there are so many species, and the whole family of the eucalyptids, of which there are several hundred species. Among animals I need mention only the order of monotremes, or egg-laying mammals, and nearly the whole order of marsupials, or pouched animals, of which there are over two hundred species. On the other hand, the true typical mammals are entirely absent, with the exception of a few bats and a few rats, which have evidently been accidentally introduced from abroad.

Another very noteworthy fact, which must be taken in connection with the last, is that Australian forms are far less advanced in the race of evolution than those of any other country--i. e., that many old forms which have long ago become extinct elsewhere are still retained there. A few examples will suffice. The marsupials just mentioned are an old form once universally distributed, but now nearly extinct everywhere, except in Australia; the cestracion, or Port Jackson shark, and the ceratodus, are Palæozoic and Mesozoic forms retained only in Australia.

What is the explanation of these remarkable facts? We find the sufficient answer in the fact that Australia has been long isolated from all other countries. While geographical changes in geological times have mingled more or less the organic forms of other countries, and the sharp struggle for life has produced more rapid advance and the production of many new and higher forms better armed for the battle of life, Australia has remained isolated from competition, and therefore comparatively unprogressive.

Can we tell when Australia was finally isolated? Approximately we can. The class of mammals is divided into two groups, which differ widely from each other; so widely, that they are called sub-classes. These are placental mammals, or true typical mammals, and non-placental or reptilian mammals. The non-placentals include only the marsupials and the monotremes (ornithorhyncus and echidna). The monotremes actually lay eggs and incubate them. In the marsupials the embryo has no placental connection with the mother, and is born in a very imperfect condition, utterly unfit for independent life, and placed in the pouch (marsupium), and _permanently_ attached there to the teat until it is capable of independent life; after which only it voluntarily nurses like other new-borns. In other words, the gestation commenced in the womb is completed in the pouch. The uterine gestation in the opossum is only seventeen days, while the marsupial gestation is about two and a half months. In a kangaroo seven feet high in sitting position the embryo at birth is only one inch long--a pink, hairless, almost amorphous mass. The monotremes are pure oviparous animals, like birds and reptiles. The marsupials might well be called _semi-oviparous_. In pure egg-layers the whole embryonic development is outside of the body; in pure young-bearers the whole is within the body; in marsupials it is partly within and partly without. Now--1. The monotremes are found nowhere but in Australia and the neighboring New Guinea. 2. The marsupials are also all confined to the Australian region, except a few oppossums in America. 3. There are some two hundred and thirty species of non-placentals in the Australian region. 4. As already said, there are no true mammals at all in Australia, except a few bats and rats which have come accidentally from abroad. 5. But non-placentals existed abundantly in _Mesozoic times everywhere_, both in Europ-Asia and in America, while true mammals did not appear at all on the surface of the earth until the _Tertiary_, when they almost immediately became very abundant everywhere, except in Australia. _Evidently, therefore, Australia was isolated before the Tertiary._ The enormous difference between its fauna and flora and those of other countries is due to at least three things: 1. So long an isolation necessarily produced great divergence of forms. This alone, however, would not affect the _grade of organization_. 2. Saved from wide migrations, and especially invasions from Eurasia, the great field of competitive struggle, it was left far behind in the race of evolution. Hence many of its forms are archaic; its mammalian fauna, for instance, is still in the Mesozoic stage. 3. Its distance from other large continents is so great that accidental colonization has been very slight, only extending to a few bats and a few rats.

I stop a moment to insist on the effect of competitive struggle in developing organic forms strong for the battle of life. Of all the continents, Eurasia has been the scene of most frequent geological changes, and therefore the arena of fiercest competitive struggle through wide and frequent migrations. Eurasian species, therefore, are the strongest of all. They have conquered wherever they have gone. Species in isolated regions are usually the weakest. The great moas and the dodo could not have continued to exist unless protected in a sort of bomb-proof. Kangaroos would now be quickly exterminated by the introduction of fierce Eurasian carnivores.

2. _Africa._--The fauna of that part of Africa north of Sahara is essentially Mediterranean--i. e., a sub-group of the Eurasian. Sahara, rather than the Mediterranean Sea, is the true intercontinental barrier. The true African region, therefore, is south of Sahara. Now, according to Mr. Wallace, whom I mainly follow here, the true African mammalian fauna consists of two very different groups of animals. The one is a group of very small, curious animals, mostly low forms of insectivores and lemurs, very peculiar to this region, though more resembling those of Madagascar than of any other region; the other is a group of large and powerful animals which dominate the region. These latter are similar to, though not identical with, those which inhabited Eurasia in Pliocene times. The great carnivores, pachyderms, and ruminants of the region are examples of this group. Now, the explanation of these facts is as follows: The indigenes of Africa are the animals of the first group. Africa, in Tertiary times, was isolated from the great field of combat, Eurasia, and therefore its animals were small, of low grade, and peculiar. During later Tertiary (Pliocene) times, then, Africa was inhabited by animals of the first group, while Eurasia was dominated by animals of the second group. These two groups were then separated by the Desert of Sahara, or else by a sea in that region. Some time during the Glacial epoch geographical changes removed this barrier, and climatic changes drove the Eurasian animals southward into Africa, where, finding congenial climate, they took possession of the continent, dominating the feebler natives. Subsequently they were isolated there by the formation of the desert, and the process of divergence commenced, and has gone on to the formation of many new forms. Meanwhile the change, partly by extinction and partly by modification, has gone on still more rapidly in Eurasia, but in a different direction. Hence, Africa is regarded as one of the primary faunal regions.

3. _Madagascar._--This, next to the Australian, is probably the most peculiar faunal region known. There is probably not a single mammalian species found there which is known to occur anywhere else. It is remarkable also as the principal home of that strange, generalized, ancient form of monkeys--the lemurs. And yet its animals, though very different, have a distant resemblance to those of Africa; not, however, to the present dominant type, but to those we have called the indigenes. Not one of the northern invaders is found there. The obvious conclusion from these facts is, that Madagascar was formerly united with Africa, and both were occupied by the same mammalian fauna (which may be called African indigenes, although they were considerably different from their descendants of the present day), but became separated before the northern invasion. The effect of this invasion was to hasten the steps of change in the indigenous fauna of Africa, partly by extermination, partly by modification, while the isolated portion in Madagascar went on at the usual slow rate of change in isolated regions. The time since the separation (which was certainly during the Tertiary period) has been sufficiently long to produce very great divergence in both, but _especially in the African indigenes_. In the fauna of Madagascar, therefore, we have a nearer approach to the original fauna of both. On account of this long isolation, we have here many ancient types which are extinct elsewhere. The lemurs are such an ancient type. These are a wonderfully-generalized type of monkeys--a connecting link between monkeys and other mammals, especially insectivores. As might be supposed, from the law of differentiation, already explained (page 11), they are the earliest form, the progenitors, of monkeys. In fact, in early Tertiary times, they were found not only in Africa and Madagascar, but all over the earth, as the only representatives of the monkey family. The true monkeys were not introduced until the mid-Tertiary. In Eurasia and in America (which at that time was probably connected with Eurasia) wide migrations and frequent conflicts of faunas produced comparatively rapid evolution of new and higher forms, while in isolated Africa old types continued until the invasion. Madagascar was spared this invasion, and therefore old types are still preserved there. At present, at least three quarters of all lemurs are confined to Madagascar, although a few species are still found in Africa and in the great East Indian islands.

4. =Island-Life.=--Mr. Wallace has divided islands into two kinds, continental and oceanic islands. The division is undoubtedly a good one, although we may not always be able to refer an example with certainty to the one or the other class. _Continental_ islands are those on the borders of continents, and separated from the latter only by _shallow water_. _Oceanic_ islands are those, usually very small, found in the midst of the ocean, with abyssal depth all about. Continental islands may be regarded as appendages to the neighboring continent--as outliers of continents separated by submergence, and have, in fact, been thus formed. Oceanic islands have been formed geologically recently by volcanic action building up from the sea-bottom. Continental islands have a continental structure--i. e., they are composed of stratified as well as of igneous rocks. Their structure is a record of geological history, like that of the neighboring continent. Oceanic islands are composed wholly of volcanic rocks; or, if there be any stratified rocks, these are only of the most recent date. As examples of continental islands we have New Zealand as an appendage of Australia, the great East Indian (Borneo, Java, Sumatra, etc.) and the Japanese Islands, etc., as appendages of Asia; the British Islands, appendages of Europe; the West Indian Islands, appendages of America; Madagascar, an appendage of Africa, etc., etc. As examples of oceanic islands we have the Azores and Bermudas in the Atlantic, and the Polynesian islands in mid-Pacific.

_a. Continental Islands._--Now, the fauna of continental islands, as might be expected from the mode of origin of these islands, is similar to, though not identical with, that of the neighboring continent; the amount of difference being in proportion to the _length of time since_ they were separated and the _width of the separation_. _Madagascar_, for example, has been long separated from its parent continent, and by a wide and deep channel. Its fauna, therefore, differs greatly from that of Africa, although resembling it more than that of any other country. The separation of _New Zealand_ from Australia has been not quite so long, and the divergence, therefore, is not so great. These two will be sufficient illustrative examples of long separation, and therefore of great differentiation of forms.

On the other hand, the British Isles are an excellent example of comparatively recent separation. These isles have probably been several times united and separated from Europe, but we are here concerned only with the more recent. They are now separated from the continent and from one another only by shallow seas. An elevation of less than six hundred feet--geologically a very small change--would bare the bottoms of the Irish and English Channels and the North Sea, and connect these islands with one another and with the continent (Fig. 67). Now, it is well known that there were during the Glacial epoch, and subsequently, several oscillations of level sufficient to connect and separate these islands. In the mid-Glacial epoch the British Islands, by submergence, were nearly obliterated, being reduced to an archipelago of small islets representing the high mountains of Wales and Scotland. The Pliocene fauna and flora were, therefore, largely exterminated. During the close of that epoch they were elevated above the present condition and broadly connected with the continent (Fig. 67), and the newly-exposed land was taken possession of by European species, man among the number. Still later--i. e., at the beginning of the present epoch--the islands by subsidence were again separated, but not widely, from the continent. This is the condition now. What, then, was the result? 1. The fauna and flora of the British Isles are substantially the same, but _less rich_ in species than that of Continental Europe, some of the European species being wanting. This shows that the last connection was not a long one; the colonization had not been completed before re-isolation. 2. This poverty of species is more conspicuous in Ireland, because colonization is progressive in space as well as in time. Some species had not reached so far when Ireland was re-isolated from England. The conspicuous absence of snakes, for example, is thus accounted for. There is, we all know, another theory to account for this, but we prefer the natural one. 3. The difference between British and European fauna and flora is very small, it is true, but there is some difference, varietal if not specific. The reason is, that the time since separation is too small to produce much divergence, and the width of the existing barriers not great enough to prevent colonization by accidental causes.

The continental islands of the southern coast of Asia are good examples of an intermediate condition as to the length of time since separation, and of the consequent degree of differentiation of the faunas and floras.

_Coast-Islands of California._--We give one more example, and dwell upon it a little, because it occurs on our own coast.

The recent studies of Mr. E. L. Greene on the flora of the islands off the coast of California have brought to light some facts which are an admirable illustration of the principles laid down above.

On looking at a good map of California, any one will observe eight or ten islands, some of them of considerable size, strung along the coast from Point Conception southward, and separated from the mainland by a sound twenty to thirty miles wide. They are in structure true continental islands--outliers of the mainland separated by a subsidence of a few hundred feet. Moreover, the date of their separation is known. They were certainly connected with the mainland during the later Pliocene and early Quaternary, for bones of the mammoth, characteristic of that time, have been found on one of them.[28] They were therefore separated during the Glacial epoch.

The main peculiarities of the flora of these islands are the following:

1. Out of nearly three hundred species of plants gathered by Mr. Greene, about fifty are wholly peculiar to these islands. 2. Of the remaining two hundred and fifty species, nearly all are distinctively Californian. In other words, the distinctively Californian forms are very abundant, while the common American forms are rare--i. e., the island flora is distinctively Californian, with many peculiar species added.

I explain these facts as follows: The whole coast-region of California is geologically very recent, having emerged from the sea as late as the beginning of the Pliocene epoch. As soon as emerged it was of course colonized from adjacent parts. Since that time its peculiar flora has been formed by gradual modification. The environment has been sufficiently peculiar, the isolation sufficiently complete, and the time sufficiently long, to make a very distinct group of organisms. It is one of Mr. Wallace’s primary divisions of the Ne-arctic region.

During late Pliocene and early Quaternary times, as already said, the islands were still a part of the mainland, and the whole was occupied by the same species, viz., the distinctively Californian species now found in both, together, as I suppose, with the peculiar island species. During the oscillations of the glacial times the islands were separated by subsidence of the continental margin. Simultaneously with this subsidence, or subsequently thereto, came the invasion of northern species, driven southward by glacial cold. Then came the mingling of invaders with natives, the struggle for mastery, the extermination of many forms--viz., the peculiar island species--and the slight modification of others, and the final result is the California flora of to-day. But the island flora was spared this invasion by isolation. Therefore the invading species are mostly wanting, the distinctive island species were saved, and the result is the island flora of to-day. The island flora, therefore, somewhat nearly represents the Pliocene indigenes of both.

It will be observed that this case is somewhat like that of Madagascar, but with a characteristic difference. In the case of Madagascar, the separation has been long. The extreme peculiarity of its fauna is the result partly of progressive divergence and partly of many forms saved by isolation. In the case of the coast-islands of California, the time has not been long enough for any great divergence by modification. The peculiarity of its species is due almost wholly to species saved by isolation.[29]

_b. Oceanic Islands._--We have seen that faunas and floras of continental islands are somewhat similar to those of the neighboring continent, though with varying degrees of difference--the amount of difference, or divergence by evolution, being in proportion to the amount of time and the impassableness of the separating barriers. But oceanic islands have never been connected with any continent. They are new land formed in the midst of the ocean by volcanic action. When they first appeared they were, of course, without inhabitants of any kind, animal or vegetal. How were they peopled? We answer by _waifs_ from here and there--by _castaways from other lands_. The dominance of particular kinds will depend on the direction of winds and currents, bringing from some lands more than others, and upon the kinds of animals or seeds of plants most liable to be successfully carried across wide seas. Their faunas and floras, therefore, are characterized by a mixture of species resembling, though not usually identical with, those of various lands, with a predominance of those of some one land, and by the singular and complete absence of mammals and amphibians, these being unlikely to be transported by floating timber, as are small reptiles and insects, etc. Among mammals, however, there is a significant exception in favor of bats, the reason being both their power of flight and their habit of concealment in hollow trees, etc. To this explanation, however, we must add that divergence by isolation will meanwhile go on in proportion to time. The Azores, for example, have been peopled from Europe, Africa, and America, but mostly from Europe, on account of the prevailing winds and currents being favorable to colonization from that direction. There are many curious peculiarities in the species, however, because colonization is very slow, and divergent variation has been going on _pari passu_. The Bermudas, on the other hand, have been colonized mainly from America, because of the current of the Gulf Stream.

These few examples are sufficient for our purpose, which is only to illustrate the causes of geographical distribution. If any one desires to pursue this interesting subject, we would refer him to that most fascinating book, Mr. Wallace’s “Island-Life.”

5. =Alpine Species.=--These afford an admirable illustration of the fact that in isolated faunas and floras the amount of difference is proportioned not only to the completeness of isolation, but also and mainly to the _time of isolation_.

It is well known that Alpine species--i. e., those species inhabiting the region bordering the perpetual snow of lofty mountains--are very similar to one another, even in the most distant localities, where their isolation from one another is as complete as possible; as, for example, in the high Alps of Europe, the high mountains of Colorado and California. Why is this? We find the key to this mystery in the additional fact that _they are similar also to Arctic species_. A somewhat full explanation is here necessary.

During Miocene times, magnolias and taxodiums (bald cypress), like those in forests and swamps of Carolina and Louisiana, and sequoias and libocedrus like those now in California, and many other temperate-region forms of plants, grew abundantly in Greenland, and northward certainly to 75° north latitude. At that time there could not have been any perpetual polar ice, and therefore no Arctic species, unless on _high mountains in polar regions_. In Pliocene times perpetual polar ice, and therefore Arctic species, probably commenced to appear. As the cold of the Glacial epoch came on and increased in severity, the polar ice extended southward as a general ice-sheet, until it reached in America 40° and in Europe about 50° north latitude. In the United States its margin can be traced as a distinct moraine through Long Island, middle New Jersey, middle Pennsylvania; thence, less distinctly, following the Ohio River, crossing the Mississippi; thence following the Missouri, on its south side, into Montana. By the increasing cold, Arctic species were driven slowly southward, generation after generation, until they occupied the whole of the United States to the Gulf, and the whole of Europe to the Mediterranean. As these species on the two continents came from a _common home_ in polar regions, they were similar to one another, except in so far as some slight divergent modification may have been produced during their southward travel. When the glacial rigor declined, and the ice-sheet gradually retreated to its present position, Arctic species, following the snow-edge, went also northward, on both continents, to their present home in polar regions. But there was an alternative way of migration left open which was embraced by certain plants and insects. While on both continents most individuals went northward, some of them went upward, following the snow-edge into high mountains, and were left _stranded there_. Thus it has come to pass that the plants and insects of high mountains in temperate regions of different continents, though so widely separated and impassably isolated, are extremely similar to one another. But, though similar, they are rarely identical. The time has been long enough for some but not very great divergent modification. It is impossible to conceive a more beautiful illustration of the principles we have been trying to enforce.

* * * * *

Thus, then, undoubtedly all the phenomena of geographical distribution of species are most rationally explained on the principle of slow evolution--changes, different in different places, and increasing with the time of isolation and its completeness.

=Objection.=--The only objection which can be raised against this view is the manner in which contiguous geographical faunas and floras pass into one another when they are _limited not by barriers but by temperature_. In passing from equator to poles, over continuous land, we of course pass through many successive faunas and floras, limited wholly or mainly by temperature. Now, if species are indeed indefinitely modifiable, then on the borders of contiguous faunas or floras, where one species disappears and another closely allied but adapted to a colder temperature takes its place, the one species (say the anti-evolutionists) ought to be gradually _transmuted_ into the other, so that all the gradations may be traced. But this is certainly not usually the fact. On the contrary, a species may indeed pass out gradually, and another come in gradually, so far as _number and vigor of individuals are concerned_; but, in _specific character_, they may be said, usually at least, to come in suddenly, with all their characters perfect, remain unchanged throughout their whole range, and pass out suddenly at its borders. Another species takes its place, overlapping in range and coexisting on the borders of both; this also continues unchanged, as far as it goes, and so on. The change from one fauna to another is apparently not by _transmutation_ of one species _into_ another by gradations, but by _substitution_ of one perfect species _for_ another perfect species. As a broad general statement, the condition of things is precisely such as would be the case if specific types were substantially immutable by physical conditions, but were originated in some inscrutable way (created) in the regions where we now find them, and have spread in every direction as far as physical conditions and struggle with other species would allow them--their ranges therefore interpenetrating and overlapping one another on their borders.

Two characteristic examples will make our meaning clear. There is not a more characteristic tree known than the sweet-gum, or liquidambar. This tree grows from the borders of Florida to the shores of the Great Lakes. It may indeed be most numerous and vigorous somewhere in the middle region, and may die out gradually in number and vigor of individuals on the borders of its range, but in specific character it is substantially the same throughout, easily recognizable by its dense wood, its winged bark, its five-starred leaf, its spinous burr, and its fragrant gum. Physical conditions may diminish its number and vigor, and limit its extension, but seem powerless to essentially modify its specific character. It seems to give up its life rather than change its nature.

Another striking example: The sequoias (redwood and big-tree) are entirely confined to California, and there are only two species now existing, viz., the redwood (_S. sempervirens_) of the Coast Ranges, and the big-tree (_S. gigantea_) of the Sierra Nevada. Doubtless they are most numerous and vigorous somewhere in the middle of their range, and die out gradually in number and vigor on the borders north and south, being replaced there by other genera better adapted to the physical conditions; but in specific character they remain essentially unchanged throughout. They are everywhere the same--easily recognizable by wood, bark, leaf, and burr. Both in this case, and in the previous one of the sweet-gum, it is as if they were created perfect in their present localities, and have spread in all directions as far as physical conditions and the struggle with other competing species would allow; but physical conditions seem powerless to change them into any other species by adaptive modification.

=Answer.=--We have, we believe, stated the objection fairly. The answer is, that the elements of _time_ and of _migrations_ have not been taken into the account. In fact, this objection was conceived and formulated before the idea of geological time was fully assimilated by the human mind, and our theories of origin adjusted to it. If these species did indeed originate where we now find them, and _in the present geological epoch_, the argument might at least be entertained; but this is not the fact. We know something of the geological history of all these species, and the history of the migrations of some of them. We know that sweet-gums were abundant and of _many species_ in the United States in Tertiary times, and all have become extinct except this remnant. Whatever of modifications there were must be looked for at or about the time of its origin in Tertiary times, not now. Species, like individuals, are plastic only when young. This one has already become rigid, and all the more so as it is a remnant widely separated from other species. For competition is strongest and most effective with nearest allies. Present species are mostly isolated remnants--terminal twiglets of the tree of life. Twiglets are of course widely separated at their visible ends. Their points of union with other twiglets must be sought below.

In the case of the sequoias, we know something also of the history of their migrations. In Miocene times they were abundant, and of many species in circumpolar regions. Some twenty-four species of fossil sequoias are known, fourteen of which are Tertiary. By the cold of the Glacial epoch they were driven slowly southward, both in America and in Europe--in America as far as Southern California. After the Glacial epoch, and the return of temperate conditions, they doubtless attempted to go northward again; but these great changes were too much for them; they were wholly exterminated in Europe, and nearly so in America. A few were left stranded high up on the slopes of the Sierra Nevada, and on the cool, moist slopes of the Coast Ranges. The species now in California are not identical with those found in the Miocene strata of Greenland; but the difference is only what we might expect after such extensive migrations and such long and severe struggle for life. Further, it is noteworthy that the Miocene species fall into two groups, viz., the yew-like leaved and the cypress-like leaved. These are represented to-day in California, the one by the redwood, the other by the big-tree. They are evidently direct descendants of the Miocene species, though somewhat modified.

But it will be objected that there ought to be some cases of transitional forms showing transmutation--in fact, there ought to be some cases of species now forming under our eyes. There are, we believe, examples of such cases. But intermediate forms are not likely to be maintained long, especially if migrations occur to give rise to severe conflict of forms. In that case the intermediate forms are soon eliminated, and species become distinct. This important point will be discussed more fully in the next chapter.