I intend to give in this chapter a general outline of the subject which will be discussed in the subsequent ones. This will include a brief history of the great events, in recent geological times, which have modified the evolution of the European fauna by the influence which they have exerted on the course of the successive streams of migration.

The composition of the European fauna is the first item which will have to be taken into consideration. But not only must the existing species of animals be dealt with: the extinct ones, too, at least those which have lived in Europe during late Tertiary times, will be useful for our inquiries. A knowledge of the past faunas is a most important factor in tracing the original home of the European animals.

Where a species first originated, whether this was in one or several places, or, in other words, where it first had its home, cannot be determined with absolute certainty in the present state of our knowledge, but as a rule it can be indicated approximately with a fair amount of precision. In a few instances, species may possibly have had a dual origin. The majority of naturalists doubt that there are any such, but it seems to me that almost the same forces may have acted in different localities on certain forms so as to produce, in very exceptional circumstances, similar species. The vast majority of animals, however, have no doubt originated in one locality; or, we might say, almost all species have but one home.

We may assume that every animal gradually extends its range by migration, as the result of the natural increase of the species necessitating a search for fresh feeding grounds. Every species thus tends to slowly take possession of all the habitable parts of the globe to which it has access. They would all naturally spread from their original homes in every direction, unless prevented by an impassable barrier. We have already learned that to all land animals, the sea acts as such a barrier. Mountains and rivers act also in a similar way, but not to the same extent. It is not difficult to understand also that a forest may be a formidable barrier to a typical inhabitant of the open country and vice versÂ, whilst a desert is impassable to almost all terrestrial organisms. Some species are scarcely affected by climate, and flourish equally well in the tropics and in temperate or cold countries; the majority, however, are greatly influenced by it. "No more striking illustration," remarks Merriam (p. 38), "could be desired of the potency of climate compared with the inefficiency of physical barriers, than is presented by the almost total dissimilarity of the North American Tropical and Sonoran Regions, though in direct contact, contrasted with the great similarity of the Boreal Regions of North America and Eurasia, now separated by broad oceans, though formerly united, doubtless, in the region of Behring Sea."

To return to the composition of the European fauna, we now know positively that a number of the mammals and birds inhabiting Central and Eastern Europe are of Siberian origin. How they came, and when, will form the subject for discussion in Chapter V. At present it will suffice to mention that in the superficial deposits belonging to the Pleistocene series of the North European plain have been discovered the remains of many typical members of the Siberian Steppe-fauna. Some of these, such as the Saiga-Antelope (Saiga tartarica), Fig. 2, still inhabit portions of Eastern Europe, whilst others have retreated to their native land. But it might be asked, how is it known that these species did not originate in Europe, and thence migrate to Siberia? Because if they had originated on our continent, they would have spread there. They would have invaded Northern and Southern Europe, and they would probably have left some remains in Spain, Italy, or Greece. They would also have left some of their relations in Europe; but all their nearest allies, too, are Asiatic. Moreover,—and this completes, I think, the proof of their Siberian origin,—the Pleistocene remains of these animals in Europe become less abundant, and the number of species likewise decreases, as we proceed from east to west. With these remains of Steppe animals are generally associated those of others, which we must also look upon as Siberian emigrants, such as the Pikas or tailless Hares belonging to the genus Lagomys, the pouched Marmots (Spermophilus), and others. Some of them, as I have mentioned, still inhabit Central and Eastern Europe, whilst others have a wider distribution on our continent.

This migration must have been an unusually large one. It has been suggested that the Glacial period had some connection with it, and there can be little doubt, as we shall see later on, that a change of climate probably brought about this great Siberian invasion of Europe. But other causes might tend in the same direction, such as want of sufficient food after a few years of great increase of any particular species. It is not known to what we owe the periodic visits of the Central Asiatic Sandgrouse (Syrrhaptes paradoxus), Fig. 3, but certain it is that immense flocks of these birds invade Europe from time to time at the present day, just as those mammals may have done in past ages.

The Siberian migrations will be spoken of in the subsequent pages, as the Siberian element of the European fauna. These migrations, however, are not the only ones which reached Europe from Asia. The sixth chapter deals with migrations which have influenced our fauna far more than the Siberian. The latter did not last long, nor did they affect the whole of Europe. But what I may call the Oriental migrations spread to every corner of Europe and certainly lasted throughout the whole of the Tertiary Era. The Oriental element came probably from Central and Southern Asia, and in its march to Northern Europe it was joined by local European migrations. For on our continent, too, animals originated and spread in all directions from their centres of dispersal. A separate chapter has been given to the Alpine fauna, and another to that of South-western Europe, which will be known by the name of the Lusitanian element. Finally, animals have also reached us from the north, and in the fourth chapter the history of that remarkable migration will be fully discussed under the title of the Arctic element of the European fauna.

It is generally believed that Africa played an important rÔle in the peopling of our continent, but this is quite a mistake. The eminent Swiss palÆontologist RÜtimeyer was quite right in saying (p. 42) that it is much more probable that Morocco, Algeria, and Tunis were stocked with animals by way of Gibraltar, and perhaps also by Sicily and Malta, from Europe, than the South of Europe from Africa.

I have already referred to what are known as "centres of dispersion" of animals, but before continuing to explain the general outline of this book, it will be necessary to make a few additional remarks on the subject.

Since every animal naturally tends to spread in every direction from its original home—that is to say, from the place of its origin—the latter should correspond with the centre of its range. And in any particular group of animals the maximum number of species should be formed in the area or zone which is the centre of its distribution. In the great majority of instances this is probably the case, in the higher animals perhaps less so than in the lower; still the rule must hold good that the original home of a species is generally indicated by the centre of its geographical distribution.

Take for example our familiar Badger (Meles taxus). It inhabits Europe and Northern Asia. It is absent apparently from many parts of Central Asia, but it appears again farther south in Palestine, Syria, Persia, Turkestan, and Tibet. West Central Asia would be about the centre of its range. That this corresponds to its place of origin is indicated by the fact that the only three other Badgers known—viz., M. anakuma, M. leucurus, and M. albogularis—are confined to Asia. If we examine the fossil history of the genus, we find that the two most ancient instances of the existence of Badgers have been discovered in Persia, where M. Polaki and M. maraghanus occur in miocene deposits. The latter had migrated as far west as Greece in miocene times; no other trace of the Badger, however, is known from Europe until we come to the pleistocene beds. There are a good many cases known among mammals where the centre of dispersion would indicate to us a similar origin. On the other hand, there may be no fossil evidence of the occurrence of a species, or of its ancestors, in Asia, whilst such has been discovered in Europe. I think, however, that the present range of a species forms a safer criterion for the determination of its original home, as the Asiatic continent is still practically unworked from a palÆontological point of view. In a letter which I received from Professor Charles DepÉret, he advocates the view that the wild Boar (Sus scrofa) is probably of European, and not, as I maintained (c, p. 455), of Asiatic origin; because there seemed to be a direct descent from Hyotherium of the middle miocene of Europe, through the upper miocene Pig of the Mount LebÉron (Sus major) and of Eppelsheim (Sus antiquus), and the pliocene Pigs of Montpellier (Sus provincialis) and of the Auvergne (Sus arvernensis). No doubt this appears rather a strong case in favour of the European origin of the wild Boar, but although the Tertiary strata of Asia, as I remarked, are as yet little known, a number of fossil pigs are known from India, Persia, and China, the oldest being the upper miocene Persian Pig (Sus maraghanus). Pigs are therefore as old in Asia as in Europe, and as a direct intercourse between the two continents probably never ceased since miocene times, it is not surprising that this genus should occur in both. Even if the genus had its origin in Europe, it is quite possible that in later Tertiary times, the active centre of origin was shifted to the neighbouring continent, and that henceforth many new species issued forth from Asia, some of which may subsequently have been modified on reaching our continent. The wild Boar (Sus scrofa), however, to judge from its general range, I must look upon as merely an immigrant in Europe. I have no doubt that it originated somewhere in Asia, probably in the south.

The view I take of the origin of our European Boar is also supported by Dr. Forsyth Major's recent researches. He was led to a re-investigation of the history of the Pig while examining a large number of fossil skulls in the Museum at Florence, and came to the conclusion that only three or four species of recent wild pigs can be clearly distinguished (b, p. 298). One of these, viz., Sus vittatus, he thinks, is traceable in slight modifications from Sardinia to New Guinea and from Japan to South Africa. The centre of distribution of this species lies in Southern Asia. Of the three remaining species, two, viz., Sus verrucosus and S. barbarus, are entirely confined to the great islands which form part of the Malay Archipelago. Finally, Sus scrofa, our Central European wild Boar, is so closely related to S. vittatus that the Sardinian Boar might be looked upon as a variety of either the one or the other. At any rate, Dr. Major recognises clearly in Sus vittatus the representative of the ancestral stock of which Sus scrofa is a somewhat modified offshoot.

The fauna of Europe consists, as I have mentioned, to a large extent of immigrants from the neighbouring continents. This is especially noticeable among the higher animals. When we come to the lower, such as the amphibia, we find a larger percentage, and among the land mollusca the great majority, to be of European origin. The foreigners are, as we learned, called Orientals, Siberians, and Arctics. For the sake of convenience, only two of the great European centres of origin have a chapter devoted to themselves, namely, the Alpine and the Lusitanian centres. There is another, however, of almost equal importance which lies in the east.

In the British Islands there is only an exceedingly small and insignificant group of species which are peculiar, and which we may consider to have had their origin there. Almost the whole of the British fauna is composed of streams of migrants which came from the north, south, and east, though many of these immigrant species have since their arrival been more or less distinctly modified into varieties or local races.

The eminent French conchologist Bourguignat (a, p. 352) was of opinion that, as far as terrestrial mollusca were concerned, there are in Europe three principal centres of creation or dispersion—all situated in mountainous countries and not in the plains. He distinguished the Spanish, Alpine, and Tauric centres, and believed that almost all species known from Europe had originated in one of these three, and that each of them possessed quite a distinct type of its own. This theory seems to agree very well with the facts of distribution. Let us take, for instance, the genus Clausilia, a pretty turret-shaped snail, which abounds on old ruined walls. Only two species, viz., Cl. laminata and Cl. bidentata, are met with in Ireland. In England we find the same species with the addition of two others, Cl. biplicata and Cl. Rolphii. Crossing over the Channel to Belgium, these four species occur again, and also several others not known in England. In Germany the list of ClausiliÆ mounts up to twenty-five species, including all those found in the British Islands. As we proceed eastward the number of species of this genus increases steadily, and when we reach the Caucasus or the Balkan Peninsula the conchologist is able to make a collection of several hundred different kinds, whilst farther east again they diminish. This clearly indicates there is in South-Eastern Europe a powerful centre of creation of ClausiliÆ, from which the species have spread all over Europe. But it is by no means certain that this centre was always in our continent, for in South-Eastern Asia and the Malay Archipelago ClausiliÆ increase once more. It is interesting to note, however, that almost all these eastern forms belong to the sub-genus PhÆdusa (vide Boettger), which had only been known as a fossil genus from a few species in the Eocene and Oligocene of Southern Europe. The first centre of origin, therefore, may possibly have been in Southern Asia, and in these early Tertiary times a second centre may have become established in Southern Europe from which the sub-genus Garnieria went eastward, Macroptychia southward, and Nenia westward across the Atlantic to South America. Only a few remnants of these primitive ClausiliÆ are now left in Europe, such as the interesting Cl. (Laminifera) Pauli.

As an example of a genus which has its centre of distribution in South-Western Europe we might take that to which our common brown garden slug belongs, viz., Arion. Dr. Simroth, who was the first to point out that the species of Arion had spread over our continent from South-Western Europe (p. 5), is inclined to the belief that the ArionidÆ had originated on the old land-bridge between Europe and North America, which is generally known by the name of "Atlantis." From this a branch went westward to the New World and another eastward as far as Southern Asia, but Arion and a number of other genera are more or less confined to South-Western Europe. Only a few species of Arion have a wide range in Europe, one of them, A. subfuscus, crossing the borders of our continent into Siberia. In the British Islands and in Western Germany, which are about equi-distant from the supposed creative centre of the genus, there are found five species. In France six or seven species are met with, and in Spain and Portugal about ten. Towards the east, Arions diminish in number. This genus, therefore, forms part of a migration which I have designated as "Lusitanian" from Lusitania, the name applied by the Romans to what we now call Portugal. Another genus of slugs, Geomalacus, is interesting from the fact that one species occurs in the British Islands, being otherwise confined to the Lusitanian province. Parmacella, a slug-like animal bearing a tiny shell at the extremity of its tail, has probably likewise had its origin in this part of Europe. All this, however, will be more fully referred to in the seventh chapter, which deals with the Lusitanian fauna.

As regards the Alpine centre of origin, Dr. Kobelt considers three groups of mollusca as especially characteristic of the Alps, viz., the sub-genus Campylaea of the great and widely-spread genus Helix, and the genera Pomatias and Zonites. The latter, which is not to be confounded with our British Hyalinia (formerly united with Zonites), does not extend very far south or north of the Alps. There may be others too, which owe their origin to these mountains, but most of the terrestrial mollusca are exceedingly ancient, and many genera have existed long before the Alps had made their appearance above the surface of the early Tertiary seas. It should be remembered that Hyalinia and Pupa, both British genera, are known from carboniferous deposits in forms which closely approach those living at the present day, and in these and a great number of other instances, it is quite impossible to determine the original home of the genus.

This little digression on centres of dispersion will help us to understand in what manner the indigenous element of the European fauna joined in with the alien members as they arrived in our continent. The species confined to South-Eastern England need not necessarily have come to us from Eastern Europe or Siberia. Alpine species spread northward probably at the same time as the Siberian animals went westward. An Alpine form may therefore have joined a batch of the latter and entered England with them. Even a Lusitanian animal may have mingled with these migrants, so that all three elements may occur together in one locality.

But these are exceptions. The migrations have, as a rule, not joined to any great extent; indeed, all those naturalists who have carefully examined the problem of the origin of the European fauna, have felt that it was composed of elements which arrived at different times.

The great Russian naturalist, the late Professor Brandt, distinguished five phases in the history of the Eurasian mammalian fauna (pp. 249-254). During the first phase—an uncertain period of long duration—the mammals held intact their position in the northern half of Asia. The Mammoth, the Hairy Rhinoceros, Bison, Musk Ox, Wild Sheep, Reindeer, and perhaps Tigers, HyÆnas, etc., lived then, with numerous peculiar Rodents, under such climatic conditions, according to Brandt, that they were able to extend their range along with tree vegetation to the extreme north of the Asiatic continent. This, he thinks, seems to have been the case especially with the Reindeer, Mammoth, Rhinoceros, and Musk Ox. The second phase was characterised by the dispersion of the Northern Asiatic mammalian fauna towards Central, Southern, and Western Europe, and this period lasted until the complete extermination of the Mammoth. The third phase dates from the time when the Mammoth and the Hairy Rhinoceros had become extinct, whilst the fourth commenced with the disappearance of the Reindeer in Europe, and terminated when the Wild Ox in the feral state had become unknown. Finally, the last phase constitutes the present time. Lartet held similar views, and also believed that Europe was peopled by successive migrations from Asia.

Botanists have worked at the problem of the European flora much more systematically, and our knowledge of the origin of that flora has been greatly increased within the last twenty years, chiefly by the researches of Professor Engler. More recently, detailed studies have been made in Scandinavia by Professor Blytt, in the Alps by Dr. Christ and Mr. Ball, in Germany by Professor Drude, Dr. Schulz, and many others. Dr. Schulz (p. 1) is of opinion that the great majority of the European plants have either migrated to or have originated in our continent since the beginning of the Pliocene epoch, and that the original home of the immigrants must be looked for in Asia and in Arctic America. From the latter an almost uninterrupted migration must have taken place during the greater part of Tertiary times up to the commencement of the Pliocene epoch, partly over a direct land-connection with Europe by way of Greenland, Iceland, and the Faroes, and also vi Spitsbergen, Franz Josef Land and Novaya Zemlya, and partly by an indirect one across the Behring Straits between Alaska and Kamtchatka.

A good deal of work still requires to be done before zoologists have acquired the same intimacy with the European fauna as botanists have with the flora. However, the view that our animals all come from Asia, as was long ago believed, has been abandoned for some time. The first to bring under the notice of naturalists the hypothesis, that there must have been two distinct migrations of northern animals to Central Europe—one from the north, and another from the east—was the late Mr. Bogdanov. The Arctic species, of which remains have been discovered in the Pyrenees—namely, the Reindeer, Arctic Hare, Willow Grouse, etc., he thought had nothing to do with those which invaded Europe from Siberia during the Glacial period. He maintained that the former had quite a distinct origin, and came from Scandinavia (p. 26).

As I shall deal with this problem more fully in a subsequent chapter, I need only mention that I fully agree with the view expressed by Mr. Bogdanov that two distinct migrations of northern species to Central Europe can be traced.

No one, I think, has done more in fostering a careful study of the migrations of animals than our distinguished geologist Professor Boyd Dawkins. He did not follow Bogdanov in distinguishing two Arctic migrations; however, he did more in constructing a very ingenious chart (a, p. 111) representing the geography of Europe during the last and most recent geological epoch—the Pleistocene—and indicating on it the probable extent, during that time, of an eastern and a southern migration of mammals. The map is very instructive, and is the first ever published giving a clear idea of a southern and an eastern migration to Europe. He believed that the migration of the southern mammals northward, took place conjunctly with the westward movement of the eastern species. Having once reached Europe, the southern species are supposed to have passed northward in summer time, whilst the eastern forms (he calls them northern) would swing southwards. The two migrations would thus occupy, at different times of the year, the same tract of ground (a, p. 113). From the mingling of the remains of the HyÆna with those of the Reindeer and Hippopotamus in the Kirkdale Cavern, he infers that the former preyed upon the Reindeer at one time of the year, and on the Hippopotamus at another. He argues that in such a manner might be explained the curious mixture of northern and southern types which we find in the British pleistocene and in cave deposits.

Besides mammals, the only European animals which have received some attention with a view to a study of their origin, are the Butterflies and the Land-Snails. The entomologists who have taken up the problem have in so far scarcely produced satisfactory results, as they all seemed to be bound down to the hypothesis that practically all the butterflies had been destroyed in Europe during the Glacial period. Hofman, in his interesting little work, comes to the conclusion (p. 50), that only in Greece and Spain could a small remnant of the butterflies have survived the extreme rigours of climate. Greece was at that time connected with Asia Minor, and Spain with North Africa; and the author supposes that the semi-alien fauna inhabiting these tracts was mainly responsible for the re-stocking of Southern Europe, but that the main mass of our butterflies are post-glacial Siberian immigrants.

The work published by Messrs. Speyer deals only with the origin of the Central European Butterflies. The period during which our European species originated is not specified, but the authors believe that they had their home either in Southern Russia or Central Asia. The fact that the number of butterflies decreases very considerably as we proceed north-westward in Europe appears to them to substantiate these views. The apparent dislike evinced by butterflies to the damp Atlantic Coast climate, they think, clearly indicates that they had originated in a dry and more continental climate. The history of the North European Butterflies and Moths has been carefully described by Mr. Petersen. He adopts Hofman's theory as to the almost total extinction of the Lepidoptera in Europe during the Glacial period. The chief immigration to Europe after that period is, he thinks, Siberian.

At first there appeared species which belonged to a cold climate, and which now live in elevated regions; then came forms suited to a milder climate, which established themselves on the north-easterly slopes of the Alps. The most recent addition which our continent has received from Siberia is, according to Mr. Petersen, the present Scandinavian fauna. Scandinavia has obtained a larger number of species than the European plain, because to this last migration were added such as prefer a northern or Alpine climate.

As a contribution to the history and composition of the European fauna, by far the most important work ever published is that of Dr. Kobelt, the eminent German conchologist. Whilst the researches into the origin of the Lepidoptera, above described, have been marred by the prevalent prejudice as to the deleterious effects of a glacial climate on the butterflies, the present author boldly works out the problem on independent lines. He shuns theories and speculations almost altogether. His great work, as yet practically unknown, the result of a lifetime of the most painstaking labour, ranks among the most important contributions to zoogeography. I shall have frequent occasion to refer to it throughout these pages. Meanwhile some of his more remarkable conclusions may be mentioned. "Comparing all classes of animals as to their zoogeographical importance, the highest rank must undoubtedly be accorded to the land-snails" (i., p. 7). "The Pleistocene, and with it the land and fresh-water molluscan fauna of the present day has been gradually evolved from the Tertiary one, and its roots can be traced through the Cretaceous to the Jurassic epoch. During the whole of that time no sudden appearance of a new fauna can be demonstrated. Quite slowly, step by step, the Cretaceous is succeeded by the Tertiary fauna, and one after the other of the characteristic palÆarctic genera appear—first the fresh-water, then the land forms" (p. 141). "The division of the North Alpine from the South Alpine fauna must be older than the Glacial period; and the present Central European fauna had already become developed from the Pliocene in all its details of form and distribution before the commencement of the Ice Age" (p. 162). "We must draw the conclusion from the preceding remarks, that the present (palÆarctic) molluscan fauna in its distribution is older than the Glacial period, and that the latter produced merely a retreat of the fauna from the most inhospitable regions of Europe with a subsequent re-immigration, but did not cause its destruction" (i., p. 169).

A few attempts have also been made by naturalists to trace the origin of the fauna of some smaller European areas. Thus RÜtimeyer, in dealing with the mammalian fauna of Switzerland, remarks (p. 31) "that it seems certain that, in spite of many local disturbances, the continuity of generations was never interrupted throughout the whole of the Tertiary period until the present day."

An even more interesting memoir is that of Mr. KÖppen on the origin of the Crimean fauna. It is only recently, according to this author, that this peninsula has become connected with Southern Russia. And it is for this reason that the Squirrel and a number of other animals, and also plants, present in Russia, are absent from the Crimea. Originally the latter probably formed a westward continuation of the Caucasus, and at that time it was surrounded by the sea on all other sides. "Much later," he continues, "after and in consequence of a local subsidence, the country between the Caucasus and the Crimea became interrupted. The latter existed for a long time as an island, and only much later, in recent geological times, did it become united with Southern Russia by means of the isthmus of Perekop."

There is, on the whole, a great diversity of opinion as to how the European fauna has originated; however, except in Dr. Kobelt's work, no attempt has hitherto been made to collect together all the available information, and to include in the inquiry more than one class of animals. The little work which I venture to bring before the public will not by any means exhaust the subject, nor is our knowledge of the European fauna sufficient to give more than a mere sketch of many of the animal groups mentioned. As we have learned in the introduction, different classes of animals are not all of equal importance in indicating the changes which have taken place in the distribution of land and water. While Dr. Kobelt is of opinion that the land-snails are by far the most important in such an inquiry, Mr. Lydekker believes that mammals afford the safest and truest indications of such changes. Mr. Beddard puts in a claim for earthworms, as even a narrow strait of sea-water forms an insuperable barrier to their dispersion. Dr. Wallace agrees with Mr. Lydekker, and goes so far as to say (p. 74) that "whenever we find that a considerable number of the mammals of two countries exhibit distinct marks of relationship, we may be sure that an actual land-connection, or at all events an approach to within a very few miles of each other, has at one time existed." Besides the groups referred to, I claim that particular attention should be devoted to Amphibia, which, contrary to Wallace, I hold do not possess special facilities for dispersal; and also to spiders and to all wingless animals leading a subterranean life, such as some of the wood-lice, planarian worms and apterous beetles.

A thorough knowledge of the changes in the distribution of land and water is desirable in order to appreciate the extent and variations of former migrations. A study of the British fauna, for example, teaches us that the British Islands were once connected with one another and with the continent of Europe between England and France. It was Professor James Geikie, I believe, who first pointed out, many years ago, that the area now covered by the Irish Sea was formerly in all probability a fresh-water lake. This had its outlet at the southern extremity in the form of a stream into which most likely flowed the smaller rivers from the south-east of Ireland, and which was joined from the east by the Severn, and finally debouched into the Atlantic (Fig. 4). The range in the British Islands of those species which have migrated to them from the south, indicates that whilst the Atlantic Ocean had gradually crept up and flooded the area between Ireland and Wales, and had turned the fresh-water lake into a bay, communication between Scotland and Ireland was still possible. The occurrence of many Scandinavian species in Scotland which are absent on the continent of Europe, indicates that these two countries also were united formerly. Most geologists hold that such a connection, if it existed, must have broken down in Pliocene times. Professor Judd, however, has expressed his belief (p. 1008) that it still existed until after the appearance of man in Northern Europe, and that our forefathers might have been able to walk dry foot from Scotland to Norway.

That Gibraltar was connected with Morocco, and Sicily with Southern Italy and Greece on the one hand, and with Tunis on the other, is more generally recognised; whilst Professor Suess has shown (vol. i., p. 442), on purely geological grounds, that the Egean Sea was dry land up till quite recently—certainly, he thinks, till after the appearance of man. This supposition enables us to understand, as will be more fully discussed in the sixth chapter, how the Oriental fauna entered Europe. Such minor zoogeographical problems as the occurrence of the Wild Goat of Asia Minor (Capra Ægagrus) on the islands of Crete and on some of the Cyclades now almost explain themselves. The Sea of Marmora is probably a modern formation, so that Asia Minor extended not long ago beyond the Turkish capital, but Dr. Kobelt believes that an arm of the Black Sea communicated up till recent times along the lower course of the Maritza with the Gulf of Saros. It can be shown also that Sardinia and Corsica formed part of the continent of Europe, and that their present fauna and flora reached them by migration on land.

The Russian naturalists, Brandt and KÖppen, believed that at no very distant date a sea extended right across Eastern Russia from the Caspian to the Arctic Ocean, whilst Professor Boyd Dawkins expressed himself in very similar language as follows (c, p. 35): "Before the lowering of the temperature in Central Europe the sea had already rolled through the low country of Russia, from the Caspian to the White Sea and the Baltic, and formed a barrier to western migration to the Arctic mammals of Asia." These naturalists based their opinions on distributional evidence, but additional facts will be brought forward in the fifth chapter to substantiate these views.

These are some of the more important geographical events which will be dealt with in detail in the subsequent chapters in connection with the history of the migrations of the European fauna.

A separate chapter has been devoted to the British fauna and its origin, since it plays a very important part in the evolution of that of our continent. So essential is a thorough knowledge of this fauna, that I think it would be difficult to understand, without it, the main features of the great migrations; and I have before now expressed the opinion that the British fauna forms the key to the solution of the problem of the origin of European animals. We know that our British species came to us by land—at least the bulk of them. But we want to know what direction they came from, and at what time they arrived. When Ireland became disconnected from Great Britain, and the latter from Scandinavia and France, is another interesting problem. Professor Boyd Dawkins has indicated to us a method of the special line of research to meet such inquiries. "The absence," he says (b, p. xxix), "of the beaver and the dormouse from Ireland must be due to the existence of some barrier to their westward migration from the adjacent mainland, and the fact that the Alpine hare is indigenous, while the common hare is absent, implies that, so far as relates to the former animal, the barrier did not exist."

Many members of the great Siberian invasion reached England, but Ireland remained entirely free from these migrants. The assumption therefore seems not unreasonable, that the latter country at the time of their arrival was no longer joined to England. The great bulk of the Irish fauna is composed of Lusitanian, Alpine, and Oriental immigrants, and there is besides a distinctly Arctic or North American element. All these, of course, must have established themselves in Ireland before the Siberian fauna set foot in England, since it has been shown that a continuous land-surface was necessary for their migration. Owing to the perfect preservation of the remains of the Siberian migrants in recent continental deposits, the history of that migration can be clearly followed, and it is possible even to determine the date of its arrival in England—in geological language at any rate. The time of the colonisation of Ireland can be thus approximately fixed as having taken place at a period prior to the arrival of the Siberian migrants in England.

All those who have seriously studied the problems presented by our British fauna—notably the late Professor Forbes, and more recently Mr. Carpenter and myself—are agreed that the Lusitanian element is the oldest, and that the newest is that which has come to us from the east.

The sequence of events in the British Islands was probably as follows:—The first comers were the members of that fauna which issued from South-western Europe; then came the Alpine, and at the same time probably the Arctic and the Oriental; and finally the Eastern or Siberian. The migrations of all but the last continued, uninterruptedly, for very long periods.

The study of these migrations has convinced me that, though climate was a powerful factor in the evolution or history of the European fauna, the geographical changes which took place on our continent in later Tertiary times exerted a yet stronger influence. The principal climatic disturbance is generally supposed to have been the so-called "Ice Age." So firmly rooted is the conviction, among naturalists of the present day, of the enormous destruction which this period produced on our European fauna, so that all animal life practically disappeared from large areas of our continent, that it is desirable that we should now shortly review the history of that remarkable period in order to ascertain in how far these views are corroborated by facts. Frequent reference, moreover, will be made throughout this work to the theories connected with the Glacial period.

It has been stated by an eminent geologist that during part of the Glacial period the climate was such that neither plants nor animals could have existed in the British Islands. If that had been so, it is evident that very few organisms could have even survived in France, though a number of Arctic species might have dragged on an existence in Southern Europe. At any rate, on the return of more genial conditions, the Arctic species would undoubtedly have been the first to gain admission to the British Islands, to re-people the arid wastes. Our supposition that the Lusitanian element in the British fauna is the oldest would therefore be wrong. From early Tertiary times onward, the climate of Europe, which was then semi-tropical, gradually became more and more temperate; until finally the Ice Age or Glacial period arrived, during which, according to Professor J. Geikie—one of our highest authorities on this subject—a great part of Northern Europe became practically uninhabitable owing to the severity of the climate.

To enable us to judge better of the true value of the many hypotheses which have been advanced to account for this supposed extraordinary fall of temperature during the "Ice Age," we must compare the views of other authorities with the one just quoted. I do not propose to discuss the causes which have led to the production of the Glacial period—those interested in these questions should consult the writings of Dr. Croll, Professor J. Geikie, Professor Bonney, Mr. Falsan, and others—but merely to give the climatic aspects from a physical, zoological, and botanical point of view.

According to Professor Penck (a, p. 12), the nature of the glacial climate can be determined by comparing the snow-line of the Glacial period with that of the present day. The position of the snow-line is dependent on two climatic factors—viz., precipitation and temperature. We know the height at which snow must have lain permanently during the Glacial period, or during the maximum phase of glaciation. If the Ice Age had been produced solely by an increase of snowfall, as has been suggested, Professor Penck tells us that then it must have snowed three or four times as much as it does now. But he does not adopt the view that the Ice Age is due to an increase of snowfall alone. His calculations, based upon the height of the snow-line, tend to show that a general decrease of temperature to the extent of from 4-5 degrees Centigrade (all other atmospheric conditions remaining the same as now) would be sufficient to give us back the Glacial period.

Professor Neumayr (p. 619) adopted a similar principle in determining the temperature which prevailed in Europe during the Glacial period. Snow now lies in the Pyrenees 1000 metres higher than it did then, 1,200 metres higher in the Alps, and 800 metres higher in the Tatra mountains. Since the temperature in Central Europe decreases by half a degree Centigrade for every 100 metres of elevation, it follows that if the glacial phenomena had only been brought about by a decrease of temperature without an increase of moisture, we should have had a reduction of temperature during the Glacial period of six degrees Centigrade in the Pyrenees, of seven degrees in the Alps, and of four in the Tatra mountains. The general lowering of the temperature of Europe, says Professor Neumayr, could not have amounted to more than six degrees Centigrade. Moreover, he is of opinion that the very low snow-line in the British Islands proves that even during the Ice Age a comparatively mild climate prevailed there, and that the climatic conditions generally, in the different parts of Europe, were relatively about the same as they are now.

Professor J. Geikie does not give us his views as to the temperature of the Glacial period, but he maintains that a lowering of the temperature is evinced not only by the widespread phenomena of glaciation, but by the former presence in our temperate latitudes of a northern fauna and flora.

Mr. Charles Martins, who based his calculations on the temperature during the Glacial period on the glaciers of Chamounix, concluded that it only needed a lowering of the temperature to the extent of four degrees Centigrade to bring the glaciers down to the plain of Geneva, and in fact give us back the Glacial period. It need not surprise us, therefore, that the French geologist, Mr. Falsan, the author of La pÉriode glaciÈre, is of opinion (p. 230) that the mean annual temperature of France during the Glacial period was approximately from 6-9 degrees Centigrade, perhaps more. Close to the immense glaciers of the Rhone, it might have been about six degrees. This is the actual mean annual temperature of the South-west of Sweden and Norway, or the North of Scotland.

Although all these investigations tend to show that the climate of Europe during the Glacial period was by no means so severe as we are often led to believe, yet there exists also a school of geologists who maintain there was actually a higher temperature than at present. The inconsistency of mentioning heat in connection with ice and snow is more apparent, however, than real, for we must remember Tyndall's original remark on this subject. It is the snow, he says, which feeds the glaciers. But the snow comes from the clouds, and these again originate from the vapours which the sun causes to be absorbed from the ocean. Without the sun's heat, we should have no water vapour in the atmosphere; without vapour, no clouds; without clouds, no snow; without snow, no glaciers. The ice of glaciers, therefore, owes its origin indirectly to the sun's heat. It has been supposed that if the sun's heat diminished, larger glaciers would form than those existing to-day, but the diminution of the solar heat would infallibly reduce the amount of water vapour in the air, and would thus stop the very source of glaciers.

Mr. Falsan even admits that without a change of the mean annual temperature (p. 201) of Europe, the central portions of our continent might at this period have enjoyed an insular climate. This more equable and humid climate could, within certain limits, favour the development of the ancient glaciers by increasing the snowfall and slackening the summer rate of melting.

It seems evident then, according to these views, that with a comparatively slight change of the atmospheric conditions in the British Islands, we might have glaciers back again on all our highest mountain ranges in England, Scotland, and Ireland. But a widespread belief seems to prevail that the presence of glaciers implies a very low temperature. Snow and ice, however, are formed as soon as the temperature falls below freezing point; it does not matter whether there be 1 or 20 degrees of cold. Winters with a few degrees of frost will be just as favourable for the growth of glaciers as winters with the most severe cold.

Let us now see what the fauna and flora, as far as we know it, tell us of the climate of the Glacial period. At the very outset of our inquiry we are confronted with one very serious difficulty in the problem, and that is the supposed occurrence of inter-glacial mild phases alternating with colder ones during the Ice Age. At first, when traces of a temperate flora and fauna were discovered intercalated between two layers of boulder clay, their presence was explained by the supposition of a mild inter-glacial period. The famous Forest-bed on the east coast of England was also pronounced to be an inter-glacial deposit, though not coming precisely under this definition. In a few places one such bed was found, in some two or more, and in others none at all. Professor James Geikie discovered the evidences of no less than five of such inter-glacial epochs (p. 612) in Europe. Lest a reader of that author's remarkable work on the Ice Age might carry with him the idea that his hypotheses had met with general acceptance, a few quotations from almost equally high authorities on glacial matters will be useful. "That the glaciers," remarks Professor Bonney (p. 245), "were liable to important oscillations seems to be proved, but whether the evidence suffices to establish inter-glacial epochs, in the usual sense of the words, is more doubtful. When the snow-fields, as in the Alps, were much more extensive than they are at present, the glaciers which radiated from them would be more sensitive to minor climatal change. Even now they oscillate considerably. But during a Glacial epoch, an inch, either more or less, of precipitation might mean a considerable advance or retreat of the ice in the lowlands." French geologists look with even less favour on Professor Geikie's theories. Mr. Falsan (p. 212) says that he agrees with Messrs. Favre, de Saporta, Lory, de Mortillet, Desor, de Lapparent, Lortet, Chantre, Benoit, Fontannes, DepÉret, and many other geologists, that there was only a single Glacial period, which, according to each particular region, might be divided into several phases, or into their equivalents—viz., one or more extensions of the ancient glaciers. But, on the whole, the view that there was at least one inter-glacial phase in the Glacial period meets with more general acceptance among geologists, I think, though the other opinion agrees much better with the nature of the fauna and flora as it has been revealed to us from the pleistocene deposits.

The occurrence of the remains of such arctic species of mammals as the Musk-Ox, Arctic Fox, Glutton, Lemming, and many others in these deposits, is frequently held up to us by geologists as a proof of the prevalence of an arctic climate while these beds were laid down. And indeed this appears at first a most satisfactory explanation of the phenomenon. But we must not judge the climate of Europe by their presence alone. As I shall explain more fully in Chapter V., these species invaded Europe owing to two circumstances. Firstly, because the climate of Siberia was becoming colder, necessitating a southward movement, with a consequent over-population in a reduced area; secondly, because a new short route to Europe had been opened up for them about the same time (see p. 221). An invasion of Europe therefore took place from east to west. Similar invasions occur even at the present day, though not caused by a change in our climate, for every now and then immense flocks of the Siberian Sandgrouse emigrate to our continent. The mammalian migrants referred to are not to be looked upon as constituting the whole of our fauna at that time. Europe had a fauna of its own, and these invaders merely mingled with our animals. There was, no doubt, a keen struggle for existence, as the result of which the weaker in many cases succumbed. The hypothesis, however, that these Siberian migrants occupied an empty continent, forsaken by its pre-glacial inhabitants, is not supported by any facts.

All those who have investigated the pleistocene fauna have been struck by the extraordinary mixture of northern and southern types of animals. Professor Dawkins attempted to explain these facts by the supposition (p. 113) that "in the summer time the southern species would pass northwards, and in the winter time the northern would sway southwards, and thus occupy at different times of the year the same tract of ground, as is now the case with the elks and reindeer." "In some of the caverns," he continues (p. 114), "such as that of Kirkdale, the hyÆna preyed upon the reindeer at one time of the year, and the hippopotamus at another."

A similar mingling of northern and southern faunas has also been observed in France. Mr. Falsan tells us (p. 236), that the remains of the mammals gathered and determined by Lartet and Gaudry belong partly to species which have been wrongly regarded as indications of a severe climate, and partly to such as are accustomed to a relatively mild temperature. In several localities in France, viz., at Levallois, St. Acheul, and Arcy, the remains of the Hippopotamus have occurred together with those of the Reindeer; whilst, according to Sir H. Howorth, the Lion has been found together with northern Voles at BicÊtre, near Paris. It is stated by the same authority (p. 115) that much the same conditions exist in Germany. "The lion and the spotted hyÆna, the mammoth and rhinoceros, were found with the marmot, the suslik, the lemming, the pica, and the reindeer." At another locality near Thiede, remains of the Mammoth, woolly Rhinoceros, Horse, Ox, Reindeer, Arctic Fox, Lemming, and Pica are met with in the same deposit. In quoting the presence of these northern animals in Europe as evidence of an arctic climate, we commit a fatal mistake. Indeed, breeders of animals and those acquainted with zoological gardens know perfectly well that it is much easier to keep a northern species in a southern climate, than a southern species in a northern one. If in a Central European deposit occur a mixture of northern and southern forms of animals, the presence of the latter is more remarkable than that of the former. Logically, we should look upon the occurrence of southern species in the north, therefore, as supporting the view that a mild climate had induced them to travel northward. The only indication, indeed, of the presence of a Monkey in the British Isles in former times comes to us from the very same strata which have also yielded the remains of the Siberian mammals.

Before I conclude the consideration of the pleistocene fauna, it may be of interest to hear what Mr. Lydekker, one of our highest authorities on fossil mammals, has to say on this subject. "The most remarkable feature connected with this fauna is the apparently contradictory evidence which it affords as to the nature of the climate then prevalent. The Glutton, Reindeer, Arctic Fox, and Musk-Ox are strongly indicative of a more or less arctic climate; many of the Voles (Microtus), Picas (Lagomys), and Susliks (Spermophilus), together with the Saiga Antelope, appear to point equally strongly to the prevalence of a Steppe-like condition; while the Hippopotamus and Spotted HyÆna seem as much in favour of a sub-tropical state of things. Many attempts have been made to reconcile these apparently contradictory circumstances; one of the older views being that while the tropical types of animals lived during a warm interlude, they migrated southwards with the incoming of colder conditions to the arctic type of fauna. Since, however, it has now been ascertained that the remains of both tropical and arctic forms have been found lying side by side in the same bed, it is perfectly certain that such an explanation will not meet the exigencies of the case" (p. 300).

In Germany the remains of the Siberian mammals occur to a large extent in a pleistocene deposit known as "loess," and the theory has of late years gained ground that the latter is the fine dust-like sand accumulated during an intensely arctic dry climate. That many of the mammals discovered in the "loess" now inhabit the dry steppes of Eastern Europe and North-Western Asia seems to lend support to this supposition; but besides the mammals there are also land and freshwater shells in this deposit. The molluscan fauna certainly indicates no steppe-character, according to Dr. Kobelt (b, i. p. 166).

The attempt to utilise the Siberian migrants to Europe as indicators of a severe climate there, certainly fails to establish conviction. But it may be asked, surely the remains of the Alpine and Arctic plants which have been found in pleistocene deposits must decide this question in favour of one or the other hypothesis? Let us test it.

Plants being more directly affected than animals by changes of temperature and rainfall, remarks Mr. Clement Reid (p. 185), give evidence of the highest value when we inquire into former climatic conditions. The severity of the climate during the Glacial period is often assumed from the occurrence in pleistocene strata of such plants as Dryas octopetala, some species of willow, the dwarf birch, and others, which are now found in high latitudes and in the Alps, but are, as a rule, absent from the plain of Northern Europe. Professor J. Geikie goes so far as to state (p. 398) that it was unlikely that southern England during the climax of the glacial cold had much if any vegetation to boast of, and continues, "It is certain, however, that it was clothed and peopled by an Arctic flora and fauna when the climatic conditions were somewhat less severe, relics of that flora having been detected at Bovey Tracey." He believes, therefore, that an Arctic flora took possession of England as soon as the climate enabled it to live in the country. Arctic plants, according to this explanation of the sequence of events, were the first immigrants to reconquer the dreary, plantless wastes and make them habitable for mammals.

Fortunately these views do not at all agree with those of many of our leading European botanists and others entitled to have a voice in the matter. Professor Warming is of opinion that the main mass of the present flora of Greenland survived the Glacial period in that country (p. 403); whilst Professor Drude has shown (p. 288) that all plant life could not possibly have been destroyed in northern countries. He maintains that the greater part of the Arctic floral elements which unite Greenland and Scandinavia must have survived the Glacial period in these countries in sheltered localities. Indeed, he justly remarks, where at the present moment do we find such plantless wastes? Greenland, Franz-Josef Land, and Grinnell Land, situated in high Arctic latitudes, all have a flora composed of flowering plants and cryptograms. "I cannot understand," he continues (p. 286), "why a flora, possibly mixed with northern forms but in the main points agreeing with our present floral elements, should not have persisted throughout the Ice Age even in the heart of Germany." "To my mind," says Col. Feilden, the well-known Arctic traveller (b, p. 51), "it seems indisputable that several plants now confined to the polar area must have originated there, and have outlived the period of greatest ice-development in that region." The theory in favour of a survival of the pre-glacial flora has been especially strengthened by the late Mr. Ball (than whom probably no botanist possessed a better knowledge of Alpine plants), who was strongly in favour of this view as far as the Alps are concerned. "Is it credible," he says (p. 576), "that in the short interval since the close of the Glacial period hundreds of very distinct species and several genera have been developed on the Alps, and, what is no less hard to conceive, that several of these non-Arctic species and genera should still more recently have been distributed at wide intervals throughout a discontinuous mountain chain some 1,500 miles in length, from the Pyrenees to the Eastern Carpathians?" Mr. Ball's remarks, indeed, just touch upon a very important characteristic of all the so-called Alpine plants. In Europe they chiefly occur in Scandinavia and the central and southern mountain ranges, whilst they are mostly absent from the intervening lowlands. Again, we find a large number of species in the mountains of Central Asia and in some of the North American mountains. Almost all species of Alpine plants, in fact, are examples of discontinuous distribution; and this, as every naturalist knows, is always, in both animals and plants, a proof of antiquity.

The glacial or Alpine flora is very old, and must have originated long before the Ice Age. But it might be urged, why should these plants be now almost confined to the Arctic regions and the higher mountain ranges, where the temperature undoubtedly is very low, if they had originated during a pre-glacial period probably much milder than the present? The answer can be given by those who have made Alpine plants their special study, and who have attempted to grow them by administering to them a temperature and such climatic conditions as to be most conducive to good health. We should all expect these plants to be very robust, and especially to be able to stand extremely low temperatures. But, strange to say, the very opposite is the case. Professor Blytt tells us (p. 19) that "Arctic and Alpine species in the Christiania Botanic Gardens endure the strongest summer heat without injury, while they are often destroyed when not sufficiently covered during winter." The English climate then, one would think, ought to suit these plants, since the winters are not too cold; but we find that at Kew Gardens the large collection of Alpine plants have to be wintered in frames under glass in order to keep them in good health; and Professor Dyer, the Director of the Gardens, thinks they are mostly intolerant of very low temperatures (compare also pp. 161-164).

Such being the constitution of Alpine plants, how could they possibly have originated during the Glacial period and wandered from the mountains into the plains, across numbers of formidable barriers, often exposed to icy winds, for thousands of miles? As a matter of fact, Alpine plants have survived in the high North and in the Alps because they are there permanently protected during winter by a covering of snow from very low temperatures, and they are at the same time prevented from drying up. If they are given sufficient moisture and a constant, mild temperature they seem to do very well. Such conditions are afforded them in many parts of the British Islands, and we find indeed the Mountain Avens (Dryas octopetala), one of the most typically Arctic plants, growing wild in profusion on the coast of Galway, in Ireland, at sea-level. The winter temperature of that part of Ireland resembles that of southern Europe, being no less than 12° Fahr. above freezing point. This fact appears to strengthen the view not only that the Alpine flora is of pre-glacial origin, but that the climate of Europe during the Glacial period was mild.

Having now shortly reviewed the state of our knowledge with regard to the former presence in our temperate latitudes of Arctic animals and plants, it still remains for me to give a succinct statement of the light thrown by this fauna and flora on the widespread phenomena of glaciation. It is necessary to do so, because, though the greater development of glaciers on the mountains of Europe in former times does not presuppose the prevalence of an Arctic climate, the survival through the Ice Age of a fauna and flora could not possibly have taken place in northern Europe if the theories of glaciation now so much in vogue are really true. Professor Geikie reminds us, in speaking of his native country (p. 67), that "we must believe that all the hills and valleys were once swathed in snow and ice; that the whole of Scotland was at some distant date buried underneath one immense mer de glace, through which peered only the higher mountain tops." That under such conditions no fauna or flora to speak of could have survived in Scotland is evident. Then again he argues (p. 426) that because in the great plain of Europe we meet occasionally with striated rock-surfaces and roches moutonnÉes very similar to those produced by the glaciers of Switzerland, it must have been traversed by "inland ice" flowing from Scandinavia and the Baltic southward. The boulder clay of Germany is supposed to have accumulated underneath this vast "mer de glace," as he calls it. There is no question here of a simple local development of glaciers, such as could have existed under a mild and moist climate; practically all the plants and animals would have been annihilated in northern Europe under such conditions, as there were no areas free from ice. A more vivid idea of the state of Europe during the epoch of maximum glaciation will be obtained by looking at Professor Geikie's map (p. 437). The whole of Scandinavia, Iceland, Scotland, Ireland, and Switzerland is there represented as having been completely enveloped in ice, and also the greater part of Russia, Germany, and England. In speaking of Scandinavia (p. 424) he remarks that "the whole country has been moulded and rubbed and polished by one immense sheet of ice, which in its deeper portions could hardly have been less than 5000 feet or even 6000 feet thick." The greater portion of the area indicated as having been underneath a sheet of ice is thickly covered with superficial accumulations of gravel, sand, and clay. The latter is generally spoken of as "boulder clay," and, with the associated sand and gravel, it may be observed equally well in Russia or Germany, in England or Ireland. As a rule these stony clays thicken out as they are traced from the high-lying tracts to the low grounds; and especially near the mountains the rock-surfaces are often polished and striated. "For many years it was believed," continues Professor Geikie (p. 432), "that all those superficial deposits were of iceberg origin. The low grounds of Northern Europe were supposed to have been submerged at a time when numerous icebergs, detached from glaciers in Scandinavia and Finland, sailed across the drowned countries, dropping rock-rubbish on the way. Such was thought to have been the origin of the erratics, stony clay, and other superficial accumulations, and hence they came to be known as the 'great northern drift formation.'" "But," he adds (p. 433), "when the phenomena came to be studied in greater detail and over a wider area, this explanation did not prove satisfactory. The facts described in the preceding paragraphs—the occurrence of striated surfaces and roches moutonnÉes, the disturbed appearances associated with the till, and the not infrequent presence of giants' kettles—convinced geologists that all the vast regions over which boulder-clay is distributed were formerly occupied by the 'inland ice' of Scandinavia."

I think Professor Geikie over-estimates the value of the evidences which appear to be in favour of his theory. His treatise on the Ice Age leaves one under the impression that the older view of the marine origin of the boulder-clay is not only done with for good and all, but that no geologists nowadays believe in it. If a more careful study of the glacial phenomena has led most geologists to abandon what I might call the "marine view" in favour of the terrestrial one, a more careful study of the fauna and flora will, I venture to think, have the opposite effect. However, it appears that even from a purely geological point of view more can be said in favour of the old theory than Professor Geikie and his school are ready to admit. Thus we are told by Professor Bonney (p. 280), in referring to the boulder-clay, that "the singular mixture and apparent crossing of the paths of boulders are less difficult to explain on the hypothesis of distribution by floating ice than on that of transport by land-ice, because, in the former case, though the drift of winds and currents would be generally in one direction, both might be varied at particular seasons. So far as concerns the distribution and thickness of the glacial deposits, there is not much to choose between either hypothesis; but on that of land-ice it is extremely difficult to explain the intercalation of perfectly stratified sands and gravel and of boulder-clay, as well as the not infrequent signs of bedding in the latter." "Anything," writes Professor Cole (p. 239), "that keeps open the position maintained by Lyell and others, that extensive glaciation is compatible with mild and sheltered nooks and corners, and that much of the distribution of boulder-clay was performed in seas and not on land, may be welcomed by rationalists, at any rate until further research has been carried on among the Arctic glaciers. At present every year brings evidence of modern marine boulder-clays in high latitudes, and removes us farther and farther from belief in a moraine profonde." That foraminifera are occasionally found in boulder-clay has been known for a long time, but it is only within recent years that these marine organisms have been shown to occur in so many localities, that Mr. Wright, who examined a large number of samples, says (p. 269), "I am forced to the conclusion that the Scottish as well as the Irish boulder-clay is a true marine sedimentary deposit."

In the fourth and fifth chapters I shall return to this subject again, and mention a number of facts of distribution which appear to me much easier of explanation by means of the marine than by the land-ice theory. But I do not propose to go into further geological details in this volume, as I think I have clearly conveyed my position in this controversy.

Before concluding this short review of the glacial problem, so far as it affects the origin of the European fauna, I should like to refer to the opinion of one who has devoted years to the study of the glacial phenomena in the Arctic Regions, viz., Col. Feilden. "To a certain extent," he says (a, p. 57), "all boulder clays at home are fragmentary when compared with the boulder-bearing beds of Kolguev, which we may safely assume are 50 miles in length by 40 in width, with a thickness of not less than 250 feet, probably far more, all lying in one undisturbed mass. It is suggestive that all the glacial deposits which I have met with in Arctic and Polar lands, with the exception of the terminal moraines now forming above sea-level in areas so widely separated as Smith's Sound, Grinnell Land, North Greenland, Spitsbergen, Novaya Zemlya, and Arctic Norway, should be glacio-marine beds. Throughout this broad expanse of the Arctic Regions I have come across no beds that could be satisfactorily assigned to the direct action of land-ice; that is to say, beds formed in situ by the grinding force and pressure of an ice-sheet. On the contrary, so far as I can judge, the glacial beds which I have traced over the extensive area mentioned above have all been deposited subaqueously and re-elevated."

One of the strongest arguments that can be used against the view of the marine origin of the glacial phenomena in Northern Europe seems to me the fact that we find polished rock-surfaces far removed from the source of glaciers, and so exactly resembling those produced at the present day by our Alpine glaciers as to appear identical to the experienced eye. Most of such striated and polished rocks occurring in the higher mountain ranges of Scandinavia, and also of the British Islands, have no doubt been actually produced by glaciers, whilst those in the plain, sometimes hundreds of miles away from the mountains, must have originated in a similar manner; that is to say, by a heavy mass of material containing stones being slowly dragged over the rock-surfaces. The weight which causes the stones to polish the latter is generally ice, but it is quite conceivable that any other substance, especially if it is in a semi-solid state, must act and operate in much the same way. All polished rock-surfaces are carved by glaciers, because we can see them done by glaciers every day, is the argument commonly used nowadays. It was not so formerly. But Mr. Mallet and his views are almost forgotten now; his name does not even appear in our great modern works on the Ice Age. His argument was that as the land rose out of the glacial sea, the mud which had accumulated round the shore slipped downward in a direction determined by the contour of the surrounding valleys and mountains. The moment the land rose above water-level, the large mass of gravel and mud lying upon it slipped downward. During a steady rising of the land there would therefore be produced a continuous sliding down of this mud-glacier, which would groove and polish the rock underneath it, in the same manner as the ice-glaciers do in the Alps (p. 47). Professors Sedgwick and Haughton became strong adherents of Mr. Mallet's theory at the time, but it seems later on to have fallen into disfavour with geologists, who may not even be thankful to have it brought to light again.

SUMMARY OF CHAPTER II.

I have endeavoured to show in this chapter how we can determine approximately the original home of an animal. By this means we are able to study the component elements of the European fauna, which is found to consist to a large extent of migrants from the neighbouring continents. There is a Siberian, an Oriental, and an Arctic element in it. The remainder of the fauna is derived from local centres of dispersal. What was formerly believed to have been one great northern migration now resolves itself, on closer study, into two very distinct ones—the Siberian and the Arctic. The mammals have received most attention hitherto, because their remains are so frequently met with, thus enabling us more easily to investigate their past history; but butterflies and snails have not been neglected, and at least one very remarkable work on the latter has been published dealing with their origin in Europe and in the remainder of the PalÆarctic region.

The former distribution of land and water is intimately connected with the origin of the European fauna, and the changes which have taken place in this respect may be best traced by the present distribution of mammals, snails, and earthworms. In this manner the British Islands may be shown to have been connected with one another and with the Continent; Spain with Morocco across the Straits of Gibraltar; Greece with Asia Minor, and so forth.

The British fauna has played such an important part in the evolution of the European fauna, that it forms the key to the solution of the wider problem. In it five elements are recognisable, of which the Lusitanian element is the oldest, and the Siberian the most recent. It has been deemed advisable to conclude this chapter with a short review of the history of the Glacial period in its climatic effects on the animals and plants of Europe. A number of writers are quoted who have conducted special researches in determining the temperature of our continent at the time. The fauna of Europe is frequently described as having been of an Arctic nature, but as a matter of fact there existed during the Ice Age a striking and most remarkable mingling of a northern and a southern fauna. The presence of Siberian mammals in Europe is said to have been due to the prevalence of a dry steppe climate, but this view is not supported by other evidence. The Alpine flora in a wide sense is probably pre-glacial in origin, and appears to have survived the Ice Age where it is now known to exist. A few words on the phenomena of glaciation are added before bringing the chapter to a close.