Il iV 'i ' M CORNELL UNIVERSITY LIBRARY BOUGHT WITH TflE INCOME OF THE SAGE ENDOWMENT FUND GIVEN IN 189I BY HENRY WILLIAMS SAGE WM 'm, li^J- im. I - ■ • » -rvJ' ra^^asmiiv 'ii 1 Date Du® 19- 9 ftl • J- « •1 y 'j i6 2. Jerboa {Alactaga jaculus) . J III. The Arctic Fox {Cams lagopus) . . „ i8 IV. The Musk Ox (Ovibos moschatus) , . „ 20 V. The Glutton (Gulo borealis) . . ,, 22 VI. The Alpine Ibex {Ibex alpinus) -s 1 „ VII. The Mammoth {Elephas primigenius) J ■* 5 VIII. Polar or Arctic Willow (Salix polaris) To face page 26 IX. Dwarf Birch (Betula nana) . ^ . „ 28 X. Mountain Avens {Dryas octopetald) . „ jo XI. The Saiga- Antelope {Saiga iartarica) . „ 32 XII. The Kiang or Dzeggetai {Equus hemionus) . „ 32 XIII. The Marmot {Arctomys marmotd) . ■ u 34 XIV. The Bison {Bison europceus) . ■ >! 34 XV. Palaeolithic Implements : Chellean and Acheulian . . . . „ 40 XVI. Palaeolithic Implements : Mousterian and Aurignacian . . . . „ 42 XVII. Pateolithic Implements : Solutrdan and Magdalenian . . » . „ 44 XX LIST OF ILLUSTRATIONS XVIII. The Aletsch Glacier, Switzerland .• XIX. Do. do. . XX. Monte Rosa and Lyskamm . XXI. The Lauterbrunnental, Switzerland,. To face page 142 ., 144 „ 146 „ 160 ILLUSTRATIONS IN TEXT I. Diagram of a Limestone Cave . , 3. Trilobite Cave (after I'Abbe Parat) 3. Cave of Sirgenstein (after Dr R. R. Schmidt) 4. Great Ofnet Cave (after Dr R. R. Schmidt) 5. Rock-shelter of Krapina (after Prof. Kramberger) . 6. Drift-deposits in Somme Valley (after Pro'f. Commont) 7. Over-deepened Alpine Valley 8. Section across the Lauterbrunnental 9. Section at La Celle-sous-Moret (after Prof. Obermaier) 71 75 79 83 109 157 161 233 MAPS A. Europe during the Second Glacial Epoch B. Europe in Interglacial Times C. Europe during the Third Glacial Epoch D. Europe during the Fourth Glacial Epoch To follow page 317 THE ANTIQUITY OF MAN LECTURE I THE TESTIMONY OF PLEISTOCENE F.AUNAS AND FLORAS Geological Estimates of Past Time are Indefinite. Oldest Authentic Human Relics are of Pleistocene Age. Doubtful Origin of the so-called " Eoliths." Discovery of supposed Primitive Types of Palaeolithic Implements in Pliocene of England. The Testimony of Animals and Plants as to Climatic Conditions of the Pleistocene Period. Southern and Temperate Group" of Mammals and Asso- ciated Plants and Molluscs. Tundra of Snow-loving Fauna and Contemporaneous Flora. Steppe Fauna and Flora. Climatic Changes of Pleistocene Period. In the present course of lectures- I purpose dealing with the question of the antiquity of the human race from a purely geological point of view. There are, as everyone knows, other points of view than this from which the subject has beeij considered. The researches of anthropologists and archaeologists have thrown a flood of light upon the q^iestion, and left us in no doubt that the advent of man must be assigned to a very remote period. But at almost every step in their investigations they have been compelled to appeal to geological evidence in support of their 2 PLEISTOCENE FAUNAS AND" FLORAS conclusions. They can tell us what kind of men lived in prehistoric times, but without calling in the aiH of geology they can say little about his environ- ment. They have shown us that the prehistoric occupants of Europe belonged probably to several races, and that these races exhibited different degrees of culture, the earliest comers being in many respects more primitive than their successors. But not a few of the results arrived at have been obtained by follow- ing geological methods of investigation. It is safe to say, indeed, that the successive stages in the history of the ancient peoples who flourished in Europe long before the dim beginnings of modern civilisation, have been established mainly by geological research. Not only so, but geology alone can tell us to what particular horizon in the long past the first appear- ance of man in our Continent must be assigned. Do hot expect from geology, however, any precise state- ment as to the date of that interesting event. It is not at all likely, indeed, that the stony science will ever be able to tell us just how many years have passed away since then. Her answer to all inquiries of the kind will doubtless continue to be that of the old man in the ballad : — " I cannot tell, I do not know. But 'twas a long-, long time ago." And yet, as I hope to show in the course of these lectures, all the evidence points to the conclusion that our race has inhabited Europe for many thousands of years. Unfortunately, the evidence GEOLOGICAL ESTIMATES OF TIME 3 that appeals most strongly to the geologist is not of a kind which others can readily appreciate. Should he venture to offer a rough estiniate of the date in question, and admit at the same time that his estimate may be several thousand, years beyond or under the truth, it is needless to tell him that com- putations so indefinite must be quite unreliable. A traveller who had painfully made his way across some trackless continental area, without maps or any means of measuring exactly the distance traversed, could not tell us precisely how many miles he had walked, but would nevertheless feel assured that the journey had been a very long one. The reality of a great distance covered would be just as' patent to him as if his route had been carefully mapped and accurately measured. Even so is it with th$j geologist. Judg- ing from what is known of the operations of nature in our own day- — reasoning, in short, from the present to the past — he is convinced that the history of the human race cannot be comprised within a few thousand years. Impressed by the long succession of geological changes of which man has been a witness — the notable modifications of the surface brought about by denudation, thef mutations of land and sea, the remarkable oscillations of climate, the revolutions of faunas and floras — the geologist is quite convinced of the great antiquity of man, even although he cannot give the precise date of his apparition. The earliest human relics and remains, the 4 PLEISTOCENE FAUNAS AND FLORAS authenticity of which is beyond dispute, belong to that stage of geological history known as the Pleistocene period. Of late years, it is true, certain evidence has been adduced, which if substantiated would carry back man's advent to a much more remote date. I refer to the so-called "eoliths"- — rudely fractured flints which, according to a number of excellent observers, show evidence of design, and are believed by them to be man's, handiwork. That view, however, has not been generally accepted ; many archaeologists, indeed, would put it aside as entirely negligible. They can see no evidence of intelligent workmanship in the supposed implements, the breaking and chipping of which they attribute to natural causes. In weighing the conflicting state- ments and arguments of experts for and against the human origin of eoliths, the only conclusion one can reasonably come to is that of the judicious Sir Roger de Coverley — " there is much to. be said on both ■sides." A Scottish jury's decision of "not proven" would seem indeed to meet the case. There is a geological side to the question, however, which can- not be ignored. It would appear that eoliths occur only in places where flint pebbles are commonly present, as in Tertiary gravel-beds of all ages. In such beds eoliths often abound, whereas in regions where flint gravels are wanting eoliths are similarly absent. This of itself, although not conclusive against the eoliths as artifacts, certainly inclines one to be sceptical. We may readily admit that the manu- SUPPOSED PRE-PAL^OLITHIC ARTIFACTS 5 facture of true Palceolithic implements must have been preceded by that of simpler' or more primitive types. Even the most conservative geologist has never questioned the possibility of very rude artifacts occurring in older deposits than Pleistocene river- drifts and cave-accumulations. But when he is asked to believe that artifacts of such a character occur not only in Pliocene, but in Miocene, and even in Oligocene and Eocene formations, he is naturally staggered. Since the Eocene period, which must date back several millions of years, the whole mammalian fauna has undergone manifold modifications and changes, continuous evolution having resulted in the more or less complete transformation of numerous types, while many others have long been extinct. And yet, if we accept the eoliths ias proofs of man's existence in Eocene and Oligocene times, we must admit that in his case — and in his case alone — evolution must have been at a standstill during a prodigiously extended period. For. it must be under- stood that the eoliths of the older Tertiary formations cannot be distinguished from those met with in Miocene, Pliocene, and even Pleistocene deposits. It is quite possible, however, nay, it is even probable, that man was in existence in Pliocene times. Only the other day, indqed, Mr J. Reid Moir recorded the occurrence of human artifacts of pre- Palaeolithic types in the Pliocene of England. Should this discovery be fully confirmed, our estimates of the antiquity of man must be greatly 6 TLEISTOCENE FAUNAS AND FLORAS increased, Mr Moir's announcernent has not sur- prised geologists ; but until the artificial character of the "specimens" in question has been established beyond doubt, we cannot assert that man lived in Europe before the Ice Age. (See Note i.) In the present course of lectures, therefore, our attention must be confined to the history of the Pleistocene period, during which man certainly occupied our Continent. The geijeral succession of events that marked the passage of that period has now been fairly well ascertained, and even a rapid review of the geological changes — the climatic and geographical mutations — experienced by our race can hardly fail to convey some impression of its prolonged duration. The investigator in this field df inquiry meets at the outset with certain difficulties, which do not often trouble those who study the records of very much older periods. It seems paradoxical, but is nevertheless true, that the history of so vastly remote an era as the Carboniferous, is in some respects more easily deciphered than that of the relativ45ly recent accumu- lations of Pleistocene age. In the case of the Carboniferous we have an enorrhous thickness of strata, extending more or less continuously over thousands of square miles. The oldest deposits occur at the bottom of the series, and overlying these comes a great succession of strata, which may be studied, bed after bed, until we reach the topmost and youngest of the series. If, therefore, we have PLEISTOCENE FORMATIONS 7 skill enough to interpret the record, we can have no doubt as to the sequence of events. With Pleistocene accumulations — and more espdcially with those containing human relics and remains — it is quite otherwise. As a rule the latter never attain a great thickness, and seldom extend continuously over a wide area. They occur, in fact, as sporadic or isolated and often widely-separated patches, the relation of which, one to another, it might seem at first almost impossible to determine. It is only occasionally that we encounter a continuous succession of strata, from which the history of a long sequence of events can be extracted. But if Pleistocene accumulations are thus infrequently arranged in consecutive series, nevertheless we are not without evidence — often indirect, it is true, but sufficiently cogent notwithstanding — which enables us to surmount the difficulties of interpretation caused by the scattered and interrupted nature of the deposits. If cave-deposits and river-alluvia containing human relics occupy no extensive areas, such is not the case with certain other accumulations of Pleistocene age. The well-known glacial formations often spread continuously over enormous tracts, and the tale they have to tell has a most direct bearing on that revealed by a study of the cave- and river-deposits — for both sets of accumulations belong to the same geological period. It is obviously important, therefore, that we should ascertain the precise relation of these several groups of formations to each other. The 8 PLEISTOCENE FAUNAS AND FLORAS history of one group we shall find practically involves that of all the others, and it matters little whether we begin our inquiry into the antiquity of man in Europe by studying the wide-spread deposits of the Ice Age, or by passing in review the testimony yielded by our cave-accumulations and riveridrifts. Each line of investigation has its own advantages, but probably it will serve our purpose best to begin with the consideration of those cave-deposits and old river- alluvia from which have been obtained most abundantly the relics and remains of our ancient predecessors. As an introduction to our proposed line of study, the present lecture will be devoted to the question of Pleistocene climate, so far as that is revealed by the character of the faunas and floras that were contemporaneous with prehistoric: man. As these are for the most part still living species their evidence cannot be gainsaid. The conditions under which they flourish in our own day must afford an approximately true picture of Pleistocene times. Even the extinct species have something definite to tell us, their testimony supporting that of their still surviving congeners. The Pleistocene mammals beldng to three more or less well-marked groups, narrigly : — a Southern and Temperate Group, a Tundra or Snow-loving Group, and a Steppe Group. These several groups I shall consider seriatim, but shall at the same time have frequent occasion to refer .to other types of EVIDENCE OF GENIAL CONDITIONS 9 animal life, and to the contemporaneous plants of the period. I. Southern and Temperaje Group. — This group includes both existing and extinct types. Among the former are hippopotamus, at present confined to Africa, where its range has been con- siderably restricted within historical times, for it formerly abounded in the Delta of the Nile. A smaller species, it may be added, occurs in some of the rivers of western Africa. The remains of a large hippopotamus, which cannot be distinguished specifically from the present African form, lived plentifully in Pleistocene Europe, ills range extending as far north at least as Yorkshire. A smaller species and even a dwarf type also flourished in southern Europe in Pleistocene times, their remains being specially abundant in some of the Mediterranean islands. Several extinct types of elephant roamed over Pleistocene Europe. Of these the most notable was the southern elephant [Elephas meridionalis) — the largest of all the European species, attaining an estimated height at the shoulder of twelve feet nine inches. Another species was trhe straight-tusked elephant {E. aniiguus), which ranged from north Africa over all central and western Europe as far north as Yorkshire. It is further notable that pigmy elephants — one species hardly exceeding three feet in height — were associates of the dwarf hippopotamus in Sicily and Malta. 10 PLEISTOCENE FAUNAS AND FLORAS Rhinoceroses of extinct speci6s were not less prominent members of the Pleistocene mammalian fauna. Among conspicuous southern types that ranged from the Mediterranean into the British area were the relatively small Etruscan rhinoceros and the megarhine or broad-nosed rhinoceros (Ji. merckii). The pachyderms I have mentioned could only have flourished under genial conditions. Their former presence in what are now temperate latitudes does iiot, however, indicate a sub-tropical climate, for they were associated with many mammals that still flourish in the forests and meadow-lands of Europe. Not a ievf of their congeners, it is true, are extinct, but among the living forms are red deer, roe deer, fallow deer, bison, wild boar, horse, beaver, and many others. Among the carnivores, contemporaneous with that fauna, were a number of southern types, such as sabre-toothed tiger (now extinct), caffer cat (a native of Egypt and south Africa), serval (a south African species), spotted hysena (a native of Africa, south of the Sahara, Plate I), striped hysena (common in India, whence it ranges west as far as the Caucasus, Arabia, Syria, Palestine, and northern Africa), lion (at present confined to Africa and south-west Asia, although in historical times it had a much wider range). Other carnivores were various bears, wolves, foxes, martens, otters, etc. Many of the Pleistocene carnivores of existing species attained a larger size than their representatives in our day. Probably this was owing PLATE I [To face page 10. EVIDENCE OF GENIAL CONDITIONS 11 to the favourable conditions under which they flourished^a genial, temperate climate, and a land well stocked with numerous herbivora. The testimony of the Southern and Temperate group of Pleistocene mammals is strongly supported by that of the plants and land-shells. The evidence supplied by the flora, indeed, is even more convincing, and enables us to realise quite clearly the climatic conditions under which prehistoric man and his associates lived for a time in Europe. In certain Pleistocene deposits in Tuscany, for instance, we find a remarkable assemblage of plant remains — indigenous species being commingled with forms, some of which, although still living in Europe, are not now natives of Tuscany, while others are exotic, and yet others are extinct. Amongst these last are several known to have existed in Europe before Pleistocene times — they are survivals from the Tertiary. Along with these occurs the laurel of the Canary Islands, a variety of t-he common laurel. This plant does not now grow spontaneously in Italy, and was at one time believed to be an extra- European species. It still lives, however, near Toulon, on the French shores of the Mediterranean, where the orange is cultivated in the open air. It' grows also near Blidah in Algeria, but its headquarters are the Canary Islands, where it flourishes luxuriantly in the woody regions with a northern exposure, between r6oo feet and 4800 feet above the sea — regions which, as Saporta remarks, are nearly always bathed 12 FLEISTOCENE FAUNAS AND FLORAS in steaming vapours, and exposed to the heavy rains of winter. During the greater part of the year the temperature keeps above 69° F., rarely falling in the winter months below 59' or 60°, and only on the coldest days reaching 49°. The common laurel and the beech are frequently associated in the Pleistocene of Tuscany, showing that they formerly grew side by side in that region. This, however, is no longer the case ; the laurel demands more shade than it could find under present conditions, while the beech has retired to the northern flanks of the Apennines to obtain the fresh, cool climate now denied to it in the low grounds of Tuscany. In the same deposits occur a number of other trees which still characterise the flora of the Mediterranean regions. Certain tufas of Provence have yielded a flora similar in all respects to that of the Tuscan deposits, save that the Tertiary and extinct species met with in the latter are absent. Among the plants of the tufas of Provence are the Canary laurel, which is associated with a number of species still indigenous to the neighbourhood. Commingled with these, how- ever, are others no longer natives of Provence, such "as Salzmann's pine, Pyrenean pine, and dwarf or mountain pine, which have abandoned the low ground of that region, and found refuge in the mountains. The same is the case indeed with certain other associated types which are now only met with in forest-clad hilly districts, and chiefly in situations with a northern exposure, wherd they can obtain EVIDENCE OF GENIAL CONDITIONS 13 the requisite shade and coolness. We may note further that the Aleppo pine and the olive are entirely wanting in the tufas of Provence. These species obviously demand considerable summer heat rather than a humid climate, and are characteristic of the Mediterranean region to-day. Even the ever- ^green oak is absent, and very rarejy occurs in any of ;the tufas of southern Europe. Looked at broadly, the Pleistocene flora of southern :France is most remarkable for the intimate associa- tion of still indigenous species with species which have ceased to be so— some of these last having retreated because unable to support the cold of winter, while others have retired to the mountains to escape the dryness of summer. The climate of southern France, therefore, like that of central Italy, must have been exempt from extremes — the winters were milder than now and the summers not so dry. An examination of the plant-reuiains of Pleistocene tufas in northern France leads to similar conclusions. In the tufa of La Celle, not far from Paris, for example, we get a group of plants indiqative of a former geographical distribution very different from that now obtaining. Among these are Canary laurel, fig-tree, box, and others, which no longer grow in the country round Paris, Thei Canary laurel, as already mentioned, exists in a wild state only in the most southern part of the Department of Van The fig-tree is not indigenous north of Provence, while the box hardly passes north beyond Lyons. Mingled 14 PLEISTOCENE FAUNAS AND FLORAS with these and other southern forms are certain species that no longer Hve side by side with their former associates, although still common in central and northern France, some of them ranging into middle and northern Europe. In this remarkable assemblage of plants we have evidence that a genial, humid, and equable climate formerly characterised northern France. The pres- ence of the Canary laurel suffices to show that the winters must have been mild, for this plant flowers during that season, and repeated frosts would prevent it reproducing its kind. It is a mild winter rather than a warm summer which the laurel demands, and the same may be said of the fig- tree. That the summer in north France was more humid and not so hot as at present,, is shown by the presence in the Pleistocene tufas of several species which cannot endure a hot arid climate, but flourish abundantly in the shady woods of northern France and Germany. The tufas of Germany contain the remains of a flora that differs less from that of southern Europe in Pleistocene times than it does from, the present flora of central Europe. Several of the species met with in the German tufas (as beech, lime, maple, sycamore, etc.) occur to-day in the Mediterranean region, but only in the mountains. In Pleistocene times, how- ever, they lived upon the low grounds of central Italy. The German Pleistocene flora is distinguished from that of southern Europe, chiefly by the presence EVIDENCE OF GENIAL CONDITIONS 15 of firs, and by the absence of the more southern forms, as vine, fig, judas-tree, laurustinus, etc. Nevertheless, nearly half of the species occurring in the Pleistocene tufas of Germany are common to the tufas of the low grounds of southern Europe. These facts would thus seem to indicate for central Europe more equable and humid conditions than the present, and it is remarkable that the Pleistocene flora of central Russia similarly implies a climate rather insular than continental. So again the plant-remains yielded by certain Pleistocene deposits met with at high elevations in the Alps, have close affinities with a flora that flourishes to-day along the southern coasts of the Black Sea. When the Pontic alpine rose {^Rhododen- dron ponticum) and its associates clothed the lofty mountain-slopes of the Inn Valley, the climate could not have been less genial than that experienced at present upon the flanks of the Alps in Italy. These Italian Alps were likewise favoured with somewhat more clement conditions than they now enjoy. The vine then grew, as it still does, along the banks of Lake Iseo, the box abounded, the chestnut flourished at an elevation of 2600 feet, and the Pontic alpine rose — now no longer an alpine plant — was very widely distributed. Without going into more details of the evidence furnished by the flora, enough has been advanced to justify the conclusion that during some stage of the Pleistocene period the climate differed considerably 16 PLEISTOCENE FAUNAS AND FLORAS from that of the present. It was essentially oceanic or insular. The summers and winters were not so strongly differentiated as they are now. The climate, in a word, was milder and more equable. I must not omit to mention that the same con- clusions may be drawn from the character of the shells which occasionally appear in large numbers along with plant-remains in Pleistocene tufas. In the tufa of La Celle, near Paris, for instance, about fifty species of shells occur, most of them still living in the district. Others, however, have retired frOm that district, although met with elsewhere in France ; yet others, no longer natives of France, are found elsewhere in Europe. Lastly, several extinct or entirely exotic species have been recognised. Among the emigrated species, some have retired to the hillier parts of France, while one is not now a native of western Europe, but flourishes over wide regions in eastern and north-eastern Europe. The most notable of the extinct forms is a zonites, belonging to a group foreign to northern France and similar latitudes in our Continent. The two forms to which it most nearly approaches are natives, the one of Austrian Tyrol and the other of Croatia. The geographical distribution of the shells, so different from the present, and (he former wider diffusion of certain forms, lead to the conclusion that the climate of northern France was formerly more equable, so as to permit species, now widely separated, to live together. That it was also a humid climate PLATE 11 [To face page 10. EVIDENCE OF COLD CONDITIONS 17 is indicated by the general facies of the shells, nearly all of them land-shells, and the large majority such as live in damp and shady places. Certain German tufas have yielded a similar assemblage of shells, and thus tell the same tale as the French tufas. The Pleistocene alluvia of many other parts of Europe now and again contain fresh- water shells, some of which no .longer live in the districts where these alluvia occur, but in more southerly or south-easterly regions. It is needless, however, to go into greater detail, for again and again the facts point to the same conclusion — namely, that during some part of the Pleistocene period the climate was more genial and equable than now. Clement winters and cool summers formerly permitted the wide diffusion and intimate association of plants which to-day have a very different range. Temperate and southern species (as ash, poplar, sycamore, fig, judas-tree, etc.) Overspread all the low grounds of France as far north at least as Paris. It was under such conditions that elephants, rhinoceroses, and hippopotami, and vast herds of deer and wild cattle ranged over Europe, from the shores of the Mediterranean up to the lati- tude of Yorkshire, and probably even farther north still ; and from the borders of Asia to the Western Ocean. (See Note 2.) II. Tundra or Snow-loving Group. — We may now turn our attention to another series of mammals which has a very different tale to tell us — the tundra or snow - loving group. I think we shall better 18 PLEISTOCENE FAUNAS AND FLORAS understand their testimony and. that of the con- temporaneous flora, if we first endeavour to have a clear conception of what is meant by tundra conditions. Geologists can only interpret the past by constant reference to the present, and I make no apology, therefore, for a rapid sketch of those regions of the globe where tundra conditions prevail. The arctic lands of Eurasia ind North America show two well-marked zones — a zone of treeless wastes bordering the Arctic Ocean, and a coniferous forest zone lying immediately to the south. The treeless wastes are known as tundras in Europe and Asia, and as " barren grounds " in North America. These form plains of immense extent, but of very unequal width from north to south. In Eurasia they lie for the most part north of the Arctic Circle, while in North America they range upon the whole considerably farther south, reaching the 6oth parallel on the western shores of Hudson Bay. Their southern boundary, however, is in both Old and New Worlds exceedingly irregular. Where the flat lands are exposed to the full sweep of the northern blasts, tundra conditions advance far to the south, invading the forest-zone in narrower or broader stretches. Indeed, even within the region of arctic forests isolated patches and wider afeas of tundra are encountered. In other places more sheltered from the fierce winds coming from the polar seas, the arctic forests in their turn encroach upon the tundras, so as nearly to reach the shores of the PLATE III [To face page IS. EVIDENCE OF COLD CONDITIONS 19 frozen ocean. Such is the case in the valleys of the Yenesei, the Khatanga, the Olenek, the Lena, and other north Siberian rivers. Similarly, in North America the arctic forests straggk" down the valleys of the Mackenzie and other rivers to beyond the Arctic Circle. Mosses and lichens form the prevailing vegetation of the tundras — marshes and bogs extending over vast areas in spring and summer, while the less marshy tracts are carpeted with grey lichens. Here and there, too, in sheltered spots, dwarf birch and willow-scrub sprinkle the surface or flourish in denser masses, and ever and anon more or less wide stretches of meadow put in an appearance. Now and again the interminable plains give place to rolling ground, the low hills and knolls being not infrequently clothed with dwarf trees. No hard and fast line, indeed, can be drawn between the tundras and the arctic forests. The two regions not only interdigitate, but numerous oases of trees are encountered in the tundras along their southern margin, while equally numerous patches of tundra, as already mentioned, are met with farther south within the forest zone. It may be added, that in northern Siberia bare rocky hills and mountains — highly fissured, and showing many gullies, ravines, and debris-strewn valleys — now and again break the uniformity of a tundra landscape. A word or two now as to the characteristic animals of the tundras and barren grounds. First among 30 PLEISTOCENE FAUNAS AND FLORAS these come the lemmings (Plate II, i). They feed on grass-roots and stalks, mosses, reindeer-lichens, and the shoots of the dwarf birch, for which in winter they tunnel through the turf or under the snow. The banded lemming is an especially characteristic form, since it is confined to the maritime tracts of Eurasia and the adjacent islands, and the corre- sponding areas of North America, and is never met with in the forest-zone. The Obi lemming has a similar distribution, but ranges somewhat farther south, and not quite so far north, as the banded lemming. The arctic fox (Plate III) is another characteristic member of the tundra fauna, having a high northern range. It occasionally wanders south to the 6oth parallel, but that is only in treeless regions, for it every- where avoids the forests, seeming to prefer the barest and most sterile lands. Another common denizen of the tundras is the arctic or mountain hare. This is the same species so commonly met with above the limits of the forests in the mountains of temperate Europe. A closely allied form (polar hare) frequents the barrens of North America. The reindeer must also be included in the tundra fauna, atthough in winter it ranges far into the forest-zone. The musk-ox (Plate IV), formerly a native of Eurasia, is now confined to North America. Like the arctic fox it avoids the forests, ranging north of these from the 6oth parallel up to the highest latitudes. Such are the most characteristic mammals of the tundras. There are many other animals, however. PLATE IV P [ 2'o /ace 2}age 20. EVIDENCE OF COLD CONDITIONS 21 which frequent the same regions, more especially in summer. Among these may be mentioned glutton, voles, ermine, weasel, wolf, common fox, and brown bear. The summer visitors also fnclude a vast host of birds, especially water-birds. The climate of all these northern plains is extreme — the winter temperature falling upon an average to 27° below zero, while in summer the average temperature is about 50° F. The actual range in certain regions is of course considerably greater. These conditions necessarily give rise to annual migrations. Only a few mammals brave the long winter of the tundras, when river and lake are often frozen solid, and the whole land fe sheeted in snow. During the great frosts the air is remarkably still, but as winter draws to a close, storms of wind and snow become frequent. Wide regions are then often swept bare, and the snow is blown into every abrupt hollow and depression in the plains, and into the gullies and ravines of the hills, where it becomes so -beaten as often to bear the weight of a man. Not only snow, but sand and dust, are thus swept forward. The sand and dust are no doubt largely obtained from the great river-valleys and deltas, but no inconsider- able proportion is derived also from the bare rocky hills and mountains, which in many places diversify the surface of the circumpolar plains. Frost is a great pulveriser of rocks, not only splitting them into fragments, but disintegrating their surfaces into grit, sand, and dust. It is remarkable how in the ^2 PLEISTOCENE FAUNAS AND FLORAS highest northern latitudes the surface of the snow often becomes discoloured with fine sand and dust derived in this way from exposed rock-surfaces. Having now glanced at the climatic conditions of our existing tundras and the character of their fauna, I may refer shortly to the wide distribution of that fauna in Pleistocene Europe. Two of the most characteristic tundra forms, as we have seen, are the banded and the Obi lemmings. Now remains of both these species have been met with again and again over all central Europe — in Russia, Poland, Austria- Hungary, North and South Germany, northern Switzerland, France, Belgium, and England. Sometimes they occur in single specimens, at other times they are extremely numeroils, the remains of several hundreds having been obtained at various localities. In many places both species of lemmings are found together ; elsewhere etther one or other may occur alone. The banded lemming, as a rule, has left its remains most abundantly in hilly and upland tracts, while those of the Obi lemming are met with more frequently in low-lying areas — a distribution quite in keeping with that which obtains at present in the tundras. That these arctic animals were not mere passing or occasional visitors is shown by the fact that young and full-grown individuals occur together in hundreds at various places, and are associated with the remains of other characteristic arctic animals which bred in the same regions. Thus well-preserved skeletons of arctic fox, having their PLATE V [To face page 22 EVIDENCE OF COLD CONDITIONS 23 milk-teeth, have been found lying side by side with the bones of the lemmings. As the arctic fox breeds in June, it is obvious that those young individuals must have died in summer. Our knowledge of the former distribution of the arctic lemmings is no doubt not so full as it will yet be, but already we have ascertained that these creatures ranged as far south as central France and the base of the Alps, in Switzerland, and as far west as Somerset, in England. Besides the arctic fox, many other northern forms were congeners of the lemmings in middle and western Europe, such as mountain hare, musk-ox, reindeer, glutton (Plate V), various voles, ermine, weasel, wolf, common fox, ibex (Plate VI), and the now extinct mammoth (Plate VII) and woolly rhinoceros. A number of northern birds have also been recorded from the same deposits as those which have yielded relics of the tundra animals. I need mention only ptarmigans, buntings, snow-owls, ducks, geese, and swans — all of which are in harmony with the arctic character of the mammals, since the same forms are in our day constant summer visitants in the circumpolar treeless lands. The arctic fauna does not seem to have found its way into Italy, but true alpine species — ibex, chamois, alpine hare, and marmot — then ranged far south in the peninsula. We may note further that just as there is evidence of the former occupation of middle and western Europe by an arctic fauna, so we have abundant traces in the same regions of a well-marked arctic flora. High 24 PLEISTOCENE FAUNAS AND FLORAS northern species of mosses, polar willow (Plate VIII), dwarf birch (Plate IX), mountain avens (Plate X), and various other northern plants have been met with in superficial deposits over a very wide area, extend- inig from southern, Sweden and England across middle Europe to the foot of the Alps. We cannot doubt, therefore, that true tundra conditions have formerly prevailed at relatively low latitudes in Europe. The wide-spread distribution of the arctic animals and plants just mentioned points clearly to that and to no other conclusion. We may, therefore, reasonably infer that the climate of middle Europe must at one time have approximated in char- acter to that of northern Siberia- — the seasons being doubtless strongly contrasted, and thus compelling annual migrations. With the advfent of summer the home of the arctic lemmings was invaded by troopS of visitants — by mammoth, woolly rhinoceros, wild horse, saiga-aritelope, and many others, and by numerous birds. An arctic-alpine vegetation clothed the low grounds, which in the warm season doubtless showed wide stretches of bog and marsh and many shallow lakes. Here and there flourished patches and wider tracts of birch and willow-scrub, but the land was practically treeless. Man, we know, was an .occupant of middle Europe at this time. Perhaps, like the mammoth and the woolly rhinoceros, he may have been rather a summer visitor than a constant denizen, departing for more clement regions at the approach of winter. We shall probably not err in PLATE VI The Alpine Ibex {Ibex altinus). PLATE VII [Bbiween pages 24 and 25. EVIDENCE OF STEPPE CONDITIONS 25 supposing that the winter would have much re- semblance to that now experienced in northern Siberia' — long spells of still weather with intense frost, interrupted now and again (especially at the changes of the seasons) by fierce snowstorms, in which the wild animals could hardly fail occasion- ally to perish in large numbers. III. Steppe Group. — We come next to consider the steppe group of mammals, and as I have prefaced my account of the tundra group with a short sketch of the conditions that obtain in circumpolar regions, I may now take a similar brief glance at the steppe lands of Europe and Asia. The regions included under this head show considerable variety. Some steppes are mere desert wastes, while others are fertile tracts capable of high cultivation. Many are low plains, others are elevated plateaus — the former having a sub-arctic, the latter a sub-tropical climate ; and between low and high steppes many gradations are met with. All are more or less characterised by an extreme range of temperature. The steppes with which we are at present concerned, however, are the generally low grassy plains known as the sub-arctic steppes. These occupy wide areas in south-east Russia and south-west Siberia, extending between the middle course of the Volga and that of the Irtysch. It is quite a mistake to suppose that these steppes are throughout all their extent treeless plains. In many places chains and irregular groups of hills diversifv the surface, while here and there trees of 26 PLEISTOCENE FAUNAS AND FLORAS various kinds, such as pine, larch, birch, oak, Hme, alder, willow, wild-apple, and others, are more or less plentiful. Many of the woods are mere oases, extending along the banks of riv^s and streams, or clustering round the margins of fresh-water lakes. In south-east Russia the boundary between the steppes and the forest lands is very irregular — the two regions constantly interdigitate. The climate of the sub-arctic steppes is quite con- tinental — the summer being relatively warm, and the winter relatively cold. The average temperature in January hardly exceeds 30° F., while that of July is at least 70°. Again, the rainfall js very uncertain. In some years it is excessive, in others meagre, while occasionally it fails altogether. With the approach of spring vegetation rapidly develops, becoming rank and luxuriant, but with the heat af summer it quickly fades and withers away. Severe frost, and frequently heavy snowstorms, characterise the winter. In such areas as are more or less wooded, the climate is somewhat less continental — the summers being relatively less dry, and the winters not so cold. But even in those wooded regions the seasons are strongly contrasted. In general, we may say, the steppe lands in summer are practically rain- less. The ground is thus parched and burnt up, so that sand and dust rise with every wind : and as the open plains are often swept by summer storms, vast quantities of loose materials are transported from place to place, and her? and there accumulate in PLATE VIII [To Jaoe page W. EVIDENCE OF STEPPE CONDITIONS 27 hollows and depressions, or come to rest in the lee of sheltering rocks and hills. In winter, if little snow has fallen, the unprotected ground is similarly scoured by tempests — dust, sand" and even small stones being carried forward. Thus, both in summer and in winter sand- and dust-storms play an import- ant r6le, and loose materials are piled up to great depths in valleys, and in the ravines, fissures, and crevices of the rocky hills. As a rule these heaps and sheets of drifted sand and dust show little or no arrangement, although now and again some trace of bedding may appear. Should they chance to become well covered with snow in winter, then, when warmth returns and the snow gradually melts away, plants quickly spring up, and the heaps become fixed and cease to drift. It is obvious that not infrequently land-shells, and often enough the remains of mammals, must be entombed in such wind-blown materials. In winter, however, it is snow more commonly than dust that drifts before the wind. The great snowstorms of the sub-arctic steppes are quite as terrible as those of the tundras. No life can with- stand the fury of the blizzards, and many are the disasters on record. In 1827, for example, all the flocks and herds that wandered over the steppes between the Volga and the Urals perished in one great storm. According to the Government Report the loss sustained by the Kirghiz amounted to 10,500 camels, 280,500 horses, 30,480 cattle, and 28 PLEISTOCENE FAUNAS i^ND FLORAS 1,012,000 sheep. Not many years pass without some disaster of this kind, and when the snow has melted away, hundreds of cattle, often far strayed, may be found huddled together in one place — some suffocated, frozen, or starved to death, others drowned in the creeks and ravines in whieh they had vainly sought for refuge from the blast. Now we can readily conceive how the carcasses might eventually be buried under drifted sand and dust, and the bony skeletons thus become preserved for an indefinite period. Among the most characteristic animals of the sub- arctic steppes are jerboas (Plate II, 2), pouched marmots, bobac, pika or tailless hare, small hamster rat, various voles, corsac, caragan fox, manul cat, saiga-antelope (Plate XI), dzeggetai (Plate XII), wild horse, etc. Besides these, many other animals are met with in the steppes, but are hardly so characteristic, since they range into adjacent regions, to which they more properly belong. Amongst them may be mentioned lynx, wild cat, tiger, wolf, jackal, common fox, martens, ermine, weasel, otter, glutton, badger, brown bear, beaver, common hare, mountain hare, wild boar, elk, reindeer, roe-deer, stag, etc. Several hundred species of birds frequent the steppes, among which may be mentioned great and little bustards, larks, grouse, buzzards, eagles, owls, etc. All the animals already mentioned as most charac- teristic of the sub-arctic steppes are represented in the PLATE IX Dwarf Birch {Beiula nana). From the Rock Garden, Royal Botanic Gardens, Edinburgh. Tlioto : Ptojcshsot Bayley Bal/otkr. [2'o /ace paffe 28. EVIDENCE OF STEPPE COJSDITIONS 29 caves and alluvial deposits of west aind middle Europe. Jerboas, pouched marmots, bobacs, and marmots (Plate XIII), tailless hares and others, all formerly flourished in those latitudes. Besides these most characteristic steppe animals, occurred many other forms which were not restricted to steppe lands, such as mam- moth and woolly rhinoceros, marsh-lynx, cave-lion, hysena, wolf, common fox, ermine, weasel, badger, rein- deer, urus, bison (Plate XIV), etc. Many birds also were 'present — all of them species which in our own day frequent the steppes of south-east Russia. Laud- shells are also very often found in less or greater abundance along with the relics of the steppe animals just mentioned, most of the shells representing forms that now live in dry steppes, while some are natives of wooded regions. The plant-remains associated with relics of the steppe fauna are quite in keeping \yith the latter, but are upon the' whole seldom met with, the conditions not being favourable to their preservation. Trunks and branches of trees occur very rarely, the most common remains being a few thir( layers and seams of peaty matter, apparently consisting chiefly of grasses. Nevertheless, we need have no doubt that a steppe flora formerly flourished in middle Europe, for (as Engler, Ascherson, Petry, and other botanists have shown) many well-known steppe plants survive in the existing flora of that region. Among the animals associated with the true steppe forms were some which, as we have seen, 30 PLEISTOCENE FAUNAS A^B FLORAS invaded central Europe in tundra times. Of these perhaps the most notable are the mammoth and the woolly rhinoceros. Probably they were only summer visitors, but under steppe conditions they became truly indigenous and very abundant. The broad valleys and open spaces of central Europe were at that time treeless plains, although woods seem to have existed here and there, especially along the margins of lakes and streams. Thfe climate, we need not doubt, was much like that of the sub-arctic steppes of south-east Russia and south-west Siberia, regions which, like the tundras, are much exposed to wind action. We may be sure, therefore, that while dry steppe conditions prevailed throughout central Europe, dust-storms and snowstorms must have been of common occurrence. We have seen how in existing tundras and Steppes the semi- domesticated and wild animals of those regions are now and again overwhelmed in storms and smothered in snow. Similar catastrophes, it may well be believed, must have happened again and again in the tundras and steppes of prehistoric times. And we are not left in this matter to mere conjecture, for the carcasses of some of the more notable animals of those days, now extinct, have been preserved to the present in the frozen snows — the famous ice-formations of northern Siberia. So perfectly preserved, indeed, was the mammoth discovered by Mr Adams, that its flesh was devoured by wolves and bears, and from the appearajices presented by PLATE .^ EVIDENCE OF STEPPE CONDITIONS 31 it and others, we cannot doubt that the animals had perished in snow-drifts. Brandt records, for example, that the congested veins and capillary vessels in the head of a rhinoceros examined by him were charged with coagulated blood, as if the animal had died of suffocation ; and Schrenck says of another described by him, that the distended nostrils and gaping mouth were highly suggestive of a similar death. It is probable that these animals were sumrher visitors to the tundras, overtaken by autumnal snowstorms. If perfectly preserved carcasses are rare, such is not the case with skeletal remains. In many places throughout Siberia the bones of various mammals occur in enormous quantities, huddled together, as it were, in very limited spots. It seems impossible to account for such hecatombs on any other sup- position than that they are the silent records of great blizzards. Even in our own time herds of wild reindeer, with their young, are overcome by snowstorms in the tundras, while jn North America great flocks of sheep and cattle frequently perish in the same way. Professor Garman, who has drawn attention to the disastrous results of blizzards in the great prairie lands of that region, is of opinion that the extraordinary heaps of skulls and other remains of the bison that are met with here and there in northern Colorado and Wyoming, are the remains of herds which have been suffocated in snow-drifts. It is not necessary to suppose that all the relics and remains of the mammoth and its congeners in 32 PLEISTOCENE FAUNAS AN© FLORAS Siberia are evidence of the desfructive effect of blizzards. The animals doubtless met their death under many different circumstances.- Sometimes they would appear to have been bogged in swampy holes and morasses. I have referred to" the peculiar ice- formations of the arctic coast-lands. These are sheets of ice of unknown thickness, preserved under more or less thick accumulations of earthy and loamy materials. The ice is believed to represent the blown or drifted snows of prehistoric times, which here and there have been protected from complete dissolution by soil and subsoil flowing over and accumulating upon them, under the influence of thaw, in spring and summer. Such movements of superficial materials are indeed of common occurrence in high latitudes at the present day. The surface of the buried ice-strata is very uneven, being furrowed and trenched by deep ruts and hollows. These depressions are filled up with frozen mud, etc., Containing vegetable debris and abundant mammalian remains, including those of mammoth and woolly rhinoceros. Probably a large number of the bones may simply have been introduced into the hollows by the flowing soil in spring — they may have been lying originally scattered over the surface. In other cases, however, the animals themselves seem to have fallen or sunk into the depressions. All the evidence leads to the inference that in the warm season these high northern regions were visited abundantly by mammoths, rhinoceroses, horses, bisons, wapiti, and PLATE XI w H Z <; < u W [To face page 82. PLATE XII iwmmmf)*, ^ii [Tu face pi"jc 32. CLIMATIC CHANGES 33 others. Such being the case, it is not hard to understand how the bulkier animals' might now and again become trapped in the treacherous bogs and subjacent muds that covered and concealed the ice-formations and their deep clefts and depressions. Thus a general review of the evidence afforded by the animals and plants that lived in Europe during Pleistocene times leads to the conclusion that the period in question was characterised by very considerable changes of climate. The theory of great annual migrations, formerly advanced to account for the presence of southern, temperate, boreal, and arctic types of mammals in the Pleistocene of Europe generally, can no longer be sustained. Even if we could suppose it possible that hippopotami and reindeer might have wandered to and fro across the whole breadth of our Continent in one season, we must admit that no migrations can account for the strongly contrasted floras which have left their remains in the Pleistocene deposits of northern, central, and southern Europe. They at least could not have indulged in such feats of travel. Now a wide-spread and decided change of climate does not take place suddenly — a revolution of the kind affecting a whole continent can only be very gradually accomplished. In subsequent lectures I shall adduce evidence to show that the climatic changes of the Pleistocene were carried on through a long series of ages — that there was no abrupt transition from one kind of climate to another. For 34 PLEISTOCENE FAUNAS Ai^JD FLORAS the present, however, we have not got beyond the threshold of our inquiry — all I have tried to show is the fact that prehistoric man lived through a succession of climatic revolutions. That, then, may be taken as our first argument in favour of the great antiquity of the human race. When we reflect on the simple fact that no marked change in the physical conditions of the Old World has been effected within historical times, we are prepared to believe that the strongly contrasted climates of the Pleistocene must have extended over many thousands of years. Most of the more or less local climatic changes or modifica- tions brought about within historical times have resulted from the action of man himself. The wholesale destruction of forests,, the draining of swnmps and marshes, the planting of trees have aill duubtless had their effect. But we have no reason to believe that the average temperature of Europe has either risen or fallen since the earliest times of which any written records have been preserved. (See Note 3.) PLATE XIII The Marmot {Arciomys mannota). {To face page 34. PLATE XIV "^1^ ^ vft^^^a^- The Bison (^Bison europarus). {To face page 34. LECTURE 11 THE TESTIMONY OF TH^ CAVES Archseological Stages of Culture : Palaeolithic, Neolithic, Bronze, and Iron Ages, Types of Paleeolithlc Man. Origin of Caves. General Character of Cave-deposits containing Relics of Prehistoric Man. Kent's Cavern and Brixham Cave — their Testimony as to the Contemporaneity of Man and the Pleistocene Fauna, and the Pro- longed Duration of Pleistocene Times. In the preceding lecture I discussed the subject of Pleistocene climate. Confining attention exclusively to the evidence of faunas and floras, I showed that their testimony put it beyond doubt that the Pleisto- cene period was characterised by profound climatic revolutions. Such revolutions necessarily imply the lapse of a very long time, and as prehistoric man was certainly a witness of them, the great antiquity of our race must be admitted. Fcom the evidence I have adduced we cannot, it is true, arrive at any definite conclusion as to the a:bs6lute length of the Pleistocene period. So far as we have advanced in our inquiry, all we can be sure of is its prolonged duration. We are almost wholly ignorant of the causes which induce climatic changes on such a grand scale as those we have been considering. Nevertheless, we are not without evidence of a 36 THE TESTIMONY OF THE CAVES more definite character which, so far as it goes, gives some precision to our conception of Pleistocene time. The deposits containing human relics and remains have yielded certain data, which may eventu- ally enable us to form tentative! estimates of the duration of the period in question. Of the deposits referred to some occur in caverns, while others appear at the surface as alluvial or subaerial accumulations. To the study of these records of the past I shall presently draw your attention, but before doing so I may very briefly sum up what archaeology has to tell us about the relics of Pleisto- cene man and their classification. That classification is based chiefly on the characters of implements and ornaments, and prehistoric time, has been divided into three periods, termed respectively the Stone Age, the Bronze Age, and the Iron Age. Of these the earliest is the Stone Age, when irnplements and ornaments werej fashioned exclu- sively of stone, horn, and bone. The use of metals for such purposes was then unknown. To the Stone Age succeeded the age of Bronze, at which time cutting-instruments, such as swords, knives, and axes, began to be made of copper, and an alloy of that metal and tin. When in the course of time iron replaced brotize for cutting-instruments, the Bronze Age came to an end and the Iron Age supervened. This classification has received the strongest support from independent geological investigations, and has long been generally accepted. ARCHAEOLOGICAL PERIODS 37 These archsological periods arfe" doubtless just so many phases of civilisation, and it is conceivable that Stone, Bronze, and Iron Ages may have been con- temporaneous in different parts of one and the same continent. Although there is nothing improbable in such a supposition, it has nevertheless been well ascertained that, so far as Eurofie is concerned, a true Stone Age to which the use of metals was altogether unknown endured throughout our Continent for a prolonged period. And it has likewise been proved that after the knowledge of bronze had become general in Europe, our ancient predecessors continued through many centuries ignorant of the use of iron. The close of the Stone Age was not marked by the total abandonment of stone for bronze artifacts, for stone continued to be used for some kinds of implements well into the Bronze A^e, and even down to historic times. The substitution of metal for stone cutting-implements might have been very slowly effected in some parts of Europe : and one can readily believe that in certaip countries bronze might come to be almost exclusively employed for such purposes, while elsewhere it remained much longer either only partially in use or quite unknown. One can hardly doubt, for instance, that long after the natives of southern Europe had commenced to attack their enemies with bronze swords and daggers, and to decorate their own persons with trinkets of the same alloy, the inhabitants of the secluded mountain- valleys of Scotland and the outlying islands might 38 THE TESTIMONY OF THE" CAVES still be living in a Stone Age. It is very improbable, in a word, that there was any abrupt transition from an age in which only stone implements were used to one in which bronze was exclusively employed. Further, we cannot infer that the Stone and Bronze periods of one country are necessarily strictly con- temporaneous throughout with the similar stages in the archaeological history of all other parts of Europe. The closing scenes of the Stone Age in north-west Europe may be synchronous with the beginning of the Bronze period in the south-east of the Continent. This would necessarily follow if it be the case, as many believe, that the knowledge of metals came from the East. The difference in point of antiquity, then, between the commencement of the Bronze Age in two such countries as Greece and Britain, let us say, would simply be measured by the length of time the natives of the latter country remained ignorant of bronze after the former had come to know it. But that time, however long it may have been, is too trifling to be taken into consideration, when periods of such duration as those of prehistoric archaeology are being dealt with. Moreover, the passage from the true Stone Age into the Bronze Age may have been actually sudden, if, as is not unlikely, metallurgical knowledge came in with one of those great folk- waves which have successively sv\fept over Europe. But however that knowlege was acquired, it is certain that, long after cutting-instruments of bronze had come into use in every part of our; Continent, stone ARCHAEOLOGICAL PERIODS 39 artifacts continued to be employed for many purposes, and are found commingled with relics of bronze in the "finds" that belong to that period. With the Bronze and Iron Ages, however, the geologist has but little to do. No great revolutions of climate or geographical conditions have super- vened in Europe since the close of the true Stone -period. That period of archaeological history com- prises two subdivisions, namely, the Palaeolithic or Old Stone Age, and the Neolithic or New Stone Age. The stone implements pertaining to. the former, how- ever rudely fashioned or well-finished they may be, are merely chipped into shape, arrd never ground or polished. The stone weapons and instruments of the later age, on the other hand, are often finely worked, being frequently ground to a sharp edge or point, or polished all over. But the general simplicity and rudeness of its artifacts are not the most distinguish- ing features of the Palaeolithic Age. As we shall see presently, the relics of that age are most frequently met with in positions that plainly argue for them a much greater antiquity than can be assigned to the oldest relics of Neolithic times. Not only so, but Palaeolithic man was the congener of many great mammals, that became either locally or wholly ex- tinct before the appearance of his Neolithic successor in Europe. The animals with which the latter was contemporaneous belong, for the most part, to species still living in our Continent — the forms, in short, are familiar, although a few are now locally extinct, as 40 THE TESTIMONY OF THE CAVES urus, wild-boar, wolf, and beaver, in Britain, all of which have vanished within historic times.'' I have referred to the stone tools used by Paleo- lithic man. It is remarkable that nearly all those implements are fashioned of flint and chert, and chiefly of the former ; but he employed also granite, hard sandstone, grit, etc., as hammers or pounders, probably for mashing roots, brea-king and crushing bones, and other purposes. Many artifacts, however, were formed of bone and horn. Among these are harpoon-heads, barbed on one or both sides, awls, pins, and needles, with well-formed eyes. More remarkable than these are the engraved and sculp- tured fragments of bone, horn, ivory, and stone, which some of the Palaeolithic folk" have left behind them. The engravings have befen etched with a sharp-pointed implement, and are often wonderfully clever, and even artistic representations of the creatures they portray. Amongst many others we find characteristic representations of cave-bear, rein- deer, mammoth, or woolly elephant, great Irish deer, ibex, bison, horse, etc., and occasionally even man. Besides etchings or gravings we meet now and again with carvings on the handles of daggers and other implements, such as representations of reindeer and mammoth. More notable still are ingenious statuettes in ivory of men and women, sometimes artistic, often crude. Again, the walls of certain 1 The foregoing paragraphs, much abridged and otherwise modified, are taken from Prehistoric Europe, Chapter II* PAL.EOLITHIC IMPLEMENTS. PLATE XV 4- 5 Types of the Coup de poing or Hand-axe. Chellean (r, 2) \b shows sinuous cutting edge of la. Acheuliaii (3-5) 4, 5, preserve at the base portions of the original surface of the nodules. All the tigures on the plate are \ of the actual size. From Musie Prehistorique by MM. G. and A. de Mortillet. \To face page 40. ARCH-EOLOGICAL PfiRIODS 41 caves (occupied towards the close of Paleolithic times) are decorated with numerous engraved draw- ings, and even coloured representations of bison, horse, and other animals. It is needless to say that the various implements and works of art thus brieflv mentioned must belono- to very difterent stages of the Palaeolithic period. By comparing and contrasting the artifacts encountered in a large number of caves, archaeologists were struck by the lact that the implements and ornaments occur- ring in one set of caves were indicative of a higher, or, as it might be, of a lower grade of culture than the artifacts characteristic of another series. It was inferred, therefore, that the caves in which the ruder implements were found had probably been occupied earlier than those from which the better finished implements had been obtained. Xow and again, in fact, distinct stages of culture were observed in one and the same cave- — the lower relic-beds showing simpler and ruder implements, while the upper deposits were charged with artifacts of a more advanced type. Of late years evidence of this, kind has greatly increased, and the general conclusions of archccologists have been thereby strongly supported. But however satisfactory and suggestive the evidence of cavern- deposits may be, it is, nevertheless, imperfect or deficient. Each cave, no doubt, contains the record of some portion of Pleistocene times, but no in- dividual cave furnishes a cortiplete consecutive 42 THE TESTIMONY OF THE CAVES history. Fortunately, however, the evidence of the eaves is supplemented and confirmed by that of ancient fluviatile and other accumui&.tions, so that the succession or classification of culture - stages, which we owe chiefly to the genius of G. de Mortillet and other French archaeologists, is now well established. As I shall have frequent occasioa to refer to these culture-stages, I may name them now, but it will not be possible to give any particular account of their distinguishing characters. To do so would take up more time than I can spare, for the geological tale I have to tell is at once long and some.what complicated, and it is the eventful history of the Pleistocene and its relation to the antiquity of our race with which I am chiefly concerned. The archaeological stages are as follows, beginning with the oldest ; — Older Palaeolithic. I. 2. Lhellean Acheulian . 3- Mousterian . 4- Aurignacian 5- Solutr^an 6. Magdalenian 7- Azilian . . 8. Neolithic. 9- Bronze Age. o. Iron Age. Younger Palaeolithic. (Transition Stage.) The Chellean stage is so named from Chelles, a town east of Paris, where the Pleistocene river-drifts PLATE XVr PALAEOLITHIC IMPLEAIENTS. Mousterian ■ i, §; 2, J ; 3, g. After MM. de Mortillet. Aurignacian : 4, i ] S « ^, i i 6 a i, ■. After MM*. Breuil and Bouyssonie. ITo face page iz ARCH^OLOGICAL PERIODS 43 rest directly on deposits of Tertiary age. The im- plement most characteristic of this stage is a large pointed artifact — the coup de poing (Plate XV, i, 2) of French writers — which was not hafted, but appears to have been held in the hand, and is therefore some- times called a "hand-axe" by English archaeologists. The Chellean artifacts are associated with remains of the southern mammals. Some archseologists re- cognise two earlier stages than the Chellean, namely, the " Mesvinian " and the " Strepyan." The former of these is represented by simple flakes of flint, evidently man's handiwork, and is further characterised by the absence of the co^tp de poing or hand-axe. The " Strepyan," on the other hand, is marked by the presence not only of simple flakes but of primitive forms of the Chellean coup de poing. By other archaeologists the Mesvinian and the Strepyan are not ranked as independent stages, but included in the Chellean, as "early Chellean." The AcHEULiAN — named from: St Acheul in the valley of the Somme — is represented by artifacts of the same type as those of the Chellean, but thinner and lighter, more shapely, and altogether better finished (Plate XV, 3-5). The fauna associated with the Acheulian artifacts includes both southern and northern forms, as the straight -tusked elephant {Elephas antiquus), and the mammoth or woolly .elephant {^E. primigenitis). The MousTERiAN takes its name from the caves of Le Moustier in the valley of the Vdz^re. The im- 44 THE TESTIMONY OF THE CAVES plements of this stage show a greater variety of form and are usually more finally chipped than those of the preceding stages. The coup de poing or hand-axe still occurs, but is rare, and would ^seem to have gone out of use in early Mousterian tinges. The fauna of this period was represented by mammoth, woolly rhinoceros, musk-ox, reindeer, lemmings, and other northern forms, together with brown bear, lion, hyaena, and leopard. In some cases, however, Mousterian artifacts are associated with remains of the southern mammalia. It may be added that while Chellean and Acheulian relics rarely occur in caves, Mousterian implements are most frequently met with in caves and rock-shelters — the men of that stage having been true troglodytes (Plate XVI, 1-3). The AuRiGNACiAN (so-called from the grotto of Aurignac, Haute Garonne) is represented by stone implements of many varied forms, and by artifacts of bone, horn, and ivory. To the same stage belong numerous works of art, such as figurines of women carved in ivory, besides etchings or engravings of various animals, and now and again the walls of caves occupied by Aurignacian man show painted designs. Among the mammalian remains of this stage are mammoth, woolly rhinoceros, bison, Irish deer, reindeer (not common), cave-bear, hyaena, etc. (Plate XVI, 4-6). The SoLUTR^AN (from Solutr6, Saone-et- Loire) is characterised by beautifully shaped and finely finished flint implements, such as the so-called "laurel-leaf" PLATE XVII PALAEOLITHIC IMPLEMENTS. Soluirian : i, 2, 3, all J ; 4, \. Magdahnian : 5, 6, 7, 8, 9, 10, all %. All figures from MM. de Mortillet's Musie Prchistovijue. [To face page 44. ARCH^OLOGICAL PERIODS 45 and "willow-leaf" lance heads (Plate XVII, 1-4). Artifacts of bone and horn (sometimes engraved) and sculptures in stone are met with. Remains of horse and reindeer, but more especially those of the horse, are abundant. At the station of Solutre so numerous were the latter that it was believed they must represent about 100,000 individuals. Other animal remains are those of mammoth, urus, hyaena, wolf, cave-bear, etc. The Magdalenian (named from the rock-shelter of La Madeleine in the valley of the V6zere) is notable for the great variety of its artifacts of bone, horn, and ivory (Plate XVII, 5-10). Stone implements were still employed, but they exhibit on the whole inferior workmanship to those of the preceding stage. But the artistic efforts of Magdalenian man far surpassed those of his predecessors of Aurignacian and Solutr^an times. His engravings and sculptures, his drawings and monochrome and polychrome paintings evince a skill and power which are simply astonishing. The walls and sometimes even the roof of his caves are decorated with frescoes representing chiefly the animals he hunted — bisons, horses, deer, etc. He has also left behind him characteristic engravings of the more notable carnivores and pachyderms of the time — cave-lion, cave-bear, mammoth, and woolly rhinoceros. The AziLiAN stage I shall refer to more particularly in the sequel. Its chief interest arises from the fact that it helps to bridge over the gap which in many 46 THE TESTIMONY OF THE CAVES places separates the Palaeolithic from the Neolithic age. But to the consideration of this matter I shall return in a subsequent lecture. It must not be supposed that the several archaeo- logical stages represent well-defined periods of time, or that they were all of equal duration. Neither do they necessarily represent a succession of different races. The human remains which now and again have been met with show, indeed, that one particular stage, the Mousterian, was characterised throughout by the presence of at least one special type of man. Several races, however, seem to have lived in Europe during subsequent Palaeolithic times, but whether these represent just so many d.ifferent folk-waves coming into our Continent, or whether one race may not in some cases have been evolved from another, we do not know. The oldest human skulls hitherto discovered are of Mousterian age — the earliest find of a tolerably complete skeleton having been made in 1857 in the Neandertal, not far from Dusseldorf. It was at first doubted whether these remains were really representative of a race. The skull had a certain primitive aspect, but it was thought that its simian characters might simply indicate an abnormal condition. So many discoveries of the same type, however, have since been made in different parts of Europe, that anthropologists are now quite assured that throughout Mousterian tijnes there lived in our Continent a race characterised by a low retreating forehead, enormous brow ridges, large PALAEOLITHIC RACES 47 round eye-orbits, massive jaws, and feebly developed chin. The heavy limb bones are curved, and the whole structure of the skeleton indicates a strong muscular man, some 5 feet 3 inches in height. Up till recently no skeletal remains oP earlier date than these were known. A few years ago, however, an exceptionally massive lower jaw, with pronounced simian characters but a decidedly human set of teeth, was obtained from deposits near Heidelberg, which may date back to the Chellean or even to an earlier stage. Again, only the other day a frag- mentary skull of very primitive type was discovered at Piltdown in Sussex, which may possibly be as old as or even older than the Heidelberg specimen. It is probable, therefore, that the Neandertal or Mousterian race was not the earliest to appear in Europe. Of Aurignacian man several skeletons have been preserved. He was of a more advanced type than his Mousterian predecessor — the skull being highly developed, with a large and lofty forehead, less pro- truding jaws, and a rudimentary chin. He was also taller (5 feet 10 inches), the limbs' being, in striking contrast with those of the earlier race, slender as in modern man. Another race of Aurignacian age is represented by two skeletons found in a cave near Mentone, which are described as being somewhat negroid in character. Magdalenian man appears to have been of small stature (5 feet 2 inches to 5 feet 4 inches). His 48 THE TESTIMONY OF THE CAVES skull, however, is shapely and lofty, the eye-orbits only moderately prominent, the chin well-formed, and the face in profile orthognathic. (See Note 4.) After this rapid glance at the general results arrived at by archseologists and anthropologists, we may next consider the nature of the evidence furnished by caves and rock-shelters as to the antiquity of man. For our present purpose it is hardly necessary to enter into any detail as to the formation of caves. Most of the large and importg,nt caves occur in limestone, those excavated in other kinds of rock being usually of inconsiderable size. Nearly all owe their origin to the chemical and mechanical action of rain and running water, while a few have been formed in other ways. The caves of most interest to the student of Palaeolithic times are of two kinds — some having been formed at the surface and in the light of day, while others Have been hollowed out below ground by the action of water. Those of the former class are commonly of smaller size than the others, and are typically represented by the hollows that occur at the base of many inland cliffs, and by the sea-caves so commonly met with along existing coast-lines and in connection with old raised sea-beaches. Usually the hollow at the base, or at any inter- mediate level between the base and top, of an inland cliff marks the outcrop of sorrfe relatively soft or readily disintegrated rock. Should such a soft rock happen to be washed by stream or river, a hollow ORIGIN OF CAVES 49 of considerable size may be scoo'ped out. For the formation of such hollows, however, it is not necessary that the strata should consist of unequally yielding materials. Cliffs of homogeneous, material are often undercut by streams, simply by mechanical erosion. The mode of formation of sea-caves is familiar to all, and need not be considered ; they and the more or less shallow rock-shelters that occur in the face and at the foot of inland cliffs and steep slopes, may of course be excavated in almost any kind of rock. The second class of caves includes all the most extensive underground galleries, many of which ramify in various directions, winding tortuously about and often opening on either side into similar intricate grottos. All these cavities owe their origin to the action of underground water. The chemical composition of springs might well lead us to expect that the more soluble rocks of the earth's crust must often be honeycombed and excavated to a very considerable extent, for the amount of mineral matter brought to the surface in solution is enormous. It is not surprising, therefore, that now and again the ground subsides where soluble rocks occur. Such subsidences often take place in regions of calcareous strata — these rocks being usually tunnelled and undermined by water descending from the surface by fissures and other natural division-planes. Owing to its content of carbonic acid the water dissolves the rock, and thus widens the fissures through which it makes its way ; and as these increase in capacity 50 THE TESTIMONY OF THE CAVES and give passage to larger bodies of water, they tend to be still further enlarged by the filing action of the stones, grit, and sand swept along. Most of our great caves, therefore, are simply the deserted channels of subterranean streams and rivers. Many such underground watfer-courses are well known, and the direction of sortie of them can be traced by swallow-holes, chasms, and sinks, which indicate where the roofs of the cavities have fallen in, or have been pierced by acidulated water. In certain regions nearly all the drainage is conducted underground. Rivers after flowing for a considerable distance at the surface suddenly disappear, to follow a hidden course for many mileSr it may be, before they again emerge. Sometimes, indeed, they never reappear, but enter the sea by subterranean channels. As the streams work their way doWn through massive limestone, many modifications of the underground labyrinths are effected ; new channels being gradually opened up, while old courses are abandoned. The latter then become more or less dry galleries, and owing to the gradual lowering of the surface by subaerial denudation, may eventually become access- ible by one or even by several openings. As might have been expected, therefore, limestone-caves are of common occurrence in valleys, and frequently open on steep hill-slopes or in river-cliffs. It may be well now to give some account of the deposits met with in caves. These are of several kinds, but I shall describe only those that are CONTENTS OF CAVES 51 common to most of the caves that contain relics of Palaeolithic man. The deposits In question consist usually of gravel, sand, and silt, of reddish earth (known as "cave-earth"), of stalactites anrd stalagmites, and of fallen rock-fragments. (See Fig. i, p. 53.) {a) The gravel, sand, and silt are obviously water- formed, and in many cases have been introduced by adjacent streams or rivers when these flowed at a higher level. Not infrequently deposits of this kind nre the oldest met with in a cave, and when such is the case they rest directly on the rocky floor. After the neighbouring stream or river had cut its bed to a lower level, the time came when it could no longer reach the cave, even during flood. The cave then remained dry, and might become the abode of man or of wild beasts. (6) The cave-earth consists of the insoluble residue of limestone. In bare uplands and plateaus limestone usually presents a more or less corrugated surface — irregular ridges of naked rock running in various directions, with intervening hollows of equally irregular form. This characteristic configuration is mainly due to the chemical action of rain which removes the carbonate of lime in solution, leaving behind the insoluble residue of the rock — a kind of reddish earth, which thus tends to accumulate in superficial hollows. But during heavy rain or the melting of snow, much of that earth may be swept down through sinks and swallow-holes into underground cavities of all kinds ; while loose stones and skeletal remains 52 THE TESTIMONY OF THE CAVES lying at the surface of the ground have doubtless often been introduced in the same way. (c) Stalactites and stalagmites are invariably present in limestone caves. Carbonated water oozing out on the roof is exposed to evaporation, and compelled to part with some of the calcareous matter held in solution, which adheres to the roof. When the gathering drops fall to the floor they are subject still further to evaporation; and give up the remainder of their calcareous content. This action carried on for a considerable time results in the for- mation of long pendants hanging from the roof like giant icicles, and of bosses and conical hummocks rising as it were from the floor to meet them. Not in- frequently indeed stalactites and stalagmites do meet, and then have the appearance of fantastic pillars that seem as if they had been designed to support the roof. The stalagmites may consist of pure carbon- ate of lime, but are often stained throughout with red earth. Their rate of growth is very variable. A layer nearly a quarter of an inch is said to have accreted in one year in a cave at Ingleborough. That must be an exceptional case, however, for with such rapid accretion all limestone caves must long ago have been filled up. If the rate of accretion in any particular cave had never varied, we should have no difficulty in ascertaining the« age of its stalag- mites. The rate is no doubt determined by physical and climatic conditions, and as these have probably remained much the same since eaVly historical times, o w Z. o < e > i o to ."t. <: Q 1. J3 CO ?: ^ 1^ o T3 -13 fi. ^ g CONTENTS OF CAVES 55 estimates based on the present rate of growth might be fairly reliable for the past two or three thousand years. But to measure the rate of accumulation throughout the far distant Pleistocene period by that of the present would be an unreliable extra- polation. During the exploration of the famous Kent's Cavern, near Torquay, a number of names and dates carved on a bed of stalagmite were detected. Some of these go back to the beginning of the seven- teenth century, but although calcareous matter has been continually accreting on them ever since, it has' been at so slow a rate that the inscriptions are still perfectly legible. On the surface of a large boss that rises above the general level of the floor of the cave, and thus marks a spot where the drip has been more continuous, and the growth, therefore, more rapid than elsewhere in the cave, there is this inscrip- tion — "Robert Hedges of Ireland, Feb. 20, 1688." Yet the film of stalagmite formed over it is not more than the twentieth of an inch in thickness ; nor is there any direct evidence to show that accretion of this particular boss was more rapid in earlier times. The bed of which it forms a part is of very variable thickness, being hardly an inch in some places, while in others it swells out to as much as five feet. If, therefore, we took the rate at which the large boss has accreted during the past two centuries as a reliable standard of measurement, we should be compelled to infer that this layer of stalagmite began to form about 56 THE TESTIMONY OF THE CAVES 240,000 years ago. On the same computation still greater would be the age of an underlying bed of stalagmite, occurring in the same cave. The latter attains a thickness of twelve feet, and at the present rate of accretion must have required 576,000 years for its growth. These estimates, however, are unreliable, for they are based on the assumption that the con- ditions have remained unchanged throughout the whole period of accretion. But s\ich we know has not been the case. Not only ha-s the region been greatly modified by denudation, but several climatic revolutions have taken place since the formation of the stalagmites commenced. It might quite well be that stalagmitic growth was fortierly much more rapid than is now the case. Nevertheless, even on the most extravagant assumption- as to the former rate of accretion, we are forced to admit a period of many thousands of years for the formation of beds of stalagmite as thick as those 1 have mentioned. (d) Large and small fragments of limestone, obviously fallen from the sides and roof, are of common occurrence in all cave-deposits. These seem to be present most abundantly in the chambers of galleries that open directly to the day, or which can be shown to have formerly had some such direct communication with the external atmosphere. The uppermost layer in which traces of prehistoric man and the Pleistocene mammals occur, is frequently sprinkled with numerous fallen masses, and some- times with a more or less thick breccia or solidified CONTENTS OF CAVES 57 aggregate of fragments of limestone of all sizes, by which the entrance to a cave may be blocked up. These fragments have been detached from the rock in various ways. Gradual widening of joints or fissures by the corrosive action of percolating water loosens large and small fragments from time to time. By this means rock-falls may take place at any moment, in any part of a cave, and under any conditions of climate. Not improbably the tremors and vibrations caused by earthquakes may occasionally shake down loosened blocks. But neither of these explanations accounts for the more abundant presence of blocks and debris in the vestibules of caves. Obviously those parts of a cave are most subject to the influence of the external atmosphere, and we shall probably not be far wrong in attributing the dislodgment of the blocks and debris in question to the action of frost, which during the prevalence of tundra climatic con- ditions must have been very effective. In the deeper recesses of the caves the cold would be less intense, and perhaps that is the reason why falls from the roof are less notable in these parts. We must now study a few typical examples of caves from which relics of prehistoric man have been obtained. The caverns and rock-shelters ransacked by archaeologists and geologists are" numerous indeed. Our own islands, Belgium, France, Spain, Germany, Switzerland, Austria, and most of the Mediterranean coast-lands of southern Europe have furnished excel- lent examples. It is impossible, therefore, to do more 58 THE TESTIMONY OF TllE CAVES than select a few which may be considered as more or less representative. I shall consequently limit myself to a short description of such as have special interest from a geological point of view — ^^those, namely, that afford the most direct testimony to the antiquity of man. We shall find that these caves not only yield clear evidence of those successive stages of culture which are of so much interest to archceologists, but are often eloquent of climatic and geographical changes. Kent's Cavern.^ — The first example of a cave I take has already been mentioned in connection with the origin of stalagmites. Beginning with the lowest and therefore the oldest deposit met with on the floor of this notable cave, we have the following succession : — 1. Breccia. — An aggregate of subangular and rounded stones, in a red sandy matrix. This deposit, consisting as it does of water- worn stones, etc., foreign to the hill in which the cave occurs, proves that from time to time water coming from some adjacent higher ground invaded the cave. The presence in the breccia of remains of the bear and a few Chellean stone implements indicates that now and again the cave was occupied by bears, and occasionally visited by man. 2. Crystalline Stalagmite. — When this stratum began to form water had ceased to flow through the cave. The stalagmite bespeaks a prolonged period of gradual accretion, for it attains a, maximum thiclf- KENT'S CAVERN 59 ness of twelve feet. There is evidence, however, to show that after a considerable mass had accumulated, it was undermined by torrential waters which had again invaded the cave. The stalagmite was thus to some extent broken up, and considerable portions of it and the underlying breccia were swept away. The cave, as before, continued at intervals to be occupied by bears. 3. Cave-earth. — This stratum was crowded with animal remains, both of extinct and living species, and it also yielded numerous relics of man's handi- work. Among the animals represented are sabre- toothed tiger, lion, wild-cat, hyaena, bear, mammoth, woolly rhinoceros, reindeer, Irish deer, red-deer, urus, bison, wolf, fox, badger, glutton, etc. Artifacts were abundant, including not only flint implements and cores or nuclei of flint, but bone-harpoons, awls, pins, and a bone needle. It is obvious that during the accumulation of the earth, which is often more or less strongly calcified, man frequently occupied the cave, while at other times (as the abundance of gnawed and broken bones sufficiently testifies) it became a hyaena's den. 4, Black Band. — Artifacts of stone, bone, and horn were very plentiful in this layer, which consisted mainly of charcoal. With these ai;tifacts were associ- ated burnt bones, and remains of ox, deer, horse, badger, bear, fox, woolly rhinoceros, hyaena, etc. The black band was met with only in one part of the cave, near the entrance, and represents the old 60 THE TESTIMONY OF THE CAVES hearths round which PalEeolithic man gathered to warm himself, and to do such primitive cooking as he required, which included the roasting of large bones for the sake of their juicy marrow. 5. Granular Stalagmite. — In this stratum, which varied from one inch to five feet in thickness, were found "stones of various kinds, shells of cockles and cuttle- fish, impressions of ferns, charcoal, bones and teeth of bears, mammoth, hysena, woolly rhinoceros, horse, fox, and flakes and cores of flint." The granular stalagmite is discoloured with cave-earth, and tells much the same tale as the underlying black band. 6. Black Mould. — This deposit is of comparatively recent origin. It yielded articles of stone and bronze, and remains of a number of animals, all of which are still indigenous. A great hiatus therefore separates the black mould from all the underlying beds with their extinct and no longer indigenous species. The mould lay between and amongst large blocks of limestone, some of which rested: upon it. I may add that similar fallen blocks, large and small, were encountered throughout all the cave-accumulations. The chief results obtained frdm the examination of the deposits in Kent's Caverns may thus be shortly summarised : — 1. The co-existence of Palceolithic man and the Pleistocene fauna is demonstrated.- 2. The prolonged duration of the period involved is proved not only by the great thickness attained by the stalagmites, but by the mechanical deposits KENT'S CAVERN 61 introduced by running water. It is evident that the whole configuration of the neighbourhood became eventually so modified by erosion, that the streams which at intervals had invaded the cave and its galleries were at last excluded. 3. The human relics obtained from the upper Palseolithic strata indicate a more advanced culture stage than those occurring in the breccia at the bottom. The latter, consisting of chert and flint only, are more massive, more rudely-formed, and less symmetrical than those met with in the cave-earth. They are probably of Chellean age, and have been formed by operating directly on nodules, of which portions of the original surface usually remain. The stone implements in the overlying beds on the other hand have been struck from blocks of flint and chipped into the desired shape — the cores from which the flakes had been obtained occurring now and again. We seem to have in the beds overlying the breccia representatives of at least two culture-stages — the Mousterian and the Magdalenian. 4. What kind of climate may have prevailed during the formation of the breccia there is no evidence to show. But we may safely conclude that the hunters and fishers, whose relics occur so abundantly in the cave-earth, the stalagmites, and the black band, lived under cold climatic conditions. 5. Lastly, as already indicated, a great break or hiatus separates the Palaeolithic deposits from the abruptly succeeding black mould with its modern fauna. !62 THE TESTIMONY OF THE CAVES Brixham Cave. — The next cagve I shall refer to is that of Brixham, which occurs ih the same part of England as Kent's Cavern, but op the opposite side of Torbay. The accumulations met with at Brixham are in descending order as follows : — 1. Stalagmitic floor — a few inches up to one foot in thickness. 2. Breccia of angular fragments of limestone which filled up the northern entrance to the cave but thinned off rapidly inwards. 3. Black bed — a peaty calcareous earth, one inch to one foot thick. 4. Cave-earth, two feet to four feet thick. — It contained numerous fragments of limestone — very small pieces up to blocks weighing one ton. Fragments of stalagmite or portions of an old "floor" also occurred. 5. Shingle, consisting of pebbles of several rocks, all, with the exception of limestone fragments, being foreign to the hill. All the accumulations in this cavern, save the black bed, were more or less fossiliferous — the cave- earth being much the richest repository of bones. The animals represented are mammoth, woolly rhinoceros, horse, urus, red-deer, reindeer, roebuck, lion, hysena, bears (several species), fox, badger, hare, rabbit, lemming, water-rat, shrew. '^ number of stone implements were scattered through the deposit. The succession of changes indicated may be briefly stated. BRIXHAM CAVte 63 The shingle at the bottom of the section has been introduced by water flowing over the country lying to the west of Brixham. Occasionally, however, the cave ceased to be thus invaded, and remained for longer or shorter periods dry. The reiics of mammoth, horse, and ox, occurring in the shingle, may have been brought in during those dry intervals by lions and hysenas, for the bones have evidently been gnawed. At this period in the history of the cave the valley of Brixham and its tributaries must have been seventy to eighty feet less deep than at present. After the cave had become largely choked with shingle, invasion by running water ceased, probably owing to the deepening of the channels outside. It was during that period of quiet that a first floor of Stalagmite gradually accreted over the shingle. A time came, however, when this stalagmite was broken up and the surface of the shingle-bed on which it rested was lowered. This change would seem to have been caused by a final irruption of water. After that event the cave remained habitually dry 1 — the only deposit laid down being red earth — the insoluble residue of the limestone itself. It is probable that to a considerable extent this material may have been introduced from the outside. The cave had now become the resort of hysenas a-nd bears, and was occasionally visited by PalaeoHthic man. Eventually the drip from the roof succeeded in forming another stalagmitic floor, and the remains of carnivores along with those of their prey became sealed up. 64 THE TESTIMONY OF THE CAVES Finally, the entrances to the cave were closed with an accumulation of limestone debris, and from that time it remained unused and untrodden until its discovery in 1858. Here, then, we have evidence of the contempor- aneity of man and the Pleistocene mammalia, and of the extreme antiquity of the period during which they were in joint occupation of southern England. At the time when the cave began to be visited by lions and hysenas and Palaeolithic man, the valley of Brixham was considerably less deep than now. So long a time has elapsed since then that the small streams of the district have been able to excavate their beds in hard rock to a depth not far short of 100 feet. (See Note 5.) In conclusion, it may be noted that the artifacts occurring in the caves of England belong chiefly to the Mousterian and Magdalenian stages — the former being best represented. Chellean implements are rarely met with, and the same is the case with Solutrean artifacts, while there would seem to be no indubitable trace of the Aurignacian stage of culture. LECTURE III THE TESTIMONY OF THE CAVES — continued Continental Examples, Belgian Caves contain Relics chiefly of Mousterian and Magdalenian Culture-stages. French Caves yield Relics of all the Archaeological Stages from Chellean to Neolithic, Bronze, and Iron. Examples : Caves of Le Moustier, La Chapelle- aux-Saints, La Ferassie, Grimaldi (Italy). Caves of Middle Europe : Caves of Sirgenstein and Ofnet representing all Culture-stages from Mousterian to Neolithic and Later Times. Rock-shelter of Krapina (Croatia). Rock-shelter of Schweizersbild (Switzerland). In this lecture I continue my description of certain caves which appear to be somewhat representative ; at all events they serve to illustrate the essential features of the geological evidence furnished by cave-accumulations in general. The caves of Belgium, France, Italy, Germany, and other countries in Europe have yielded, one may say, very much the same kind of evidence as that supplied by the cavern-deposits of England. In certain of the river-valleys of Belgium, particularly in those of some of the tributaries of the Meuse, a number of fine caves and rock-shelters occur. They appear in rocky escarpments at different levels above the streams, from a few yards up to 200 feet, one of the best known being that of Spy, in which were 66 E U THE TESTIMONY OF THE CAVES discovered two skeletons of Mousterian man. The floor-deposits in most of these caves consist usually of alternations of fluviatile sediments and cave-earth, with layers of stalagmite. To these accumulations the relics and remains of Palseolithic man and the Pleistocene mammalia are restricted. Above the ossiferous deposits comes a mass of yellow clay and angular fragments of limestone. Relics and remains of Neolithic man appear at and near the surface of this yellow clay, but they never occur underneath it. The stony yellow clay thus takes its place in the series between the deposits of the Old Stone period and those which mark the later or New Stone period. It is obvious that here, as in many other caves of north-west Europe, a break or hiatus separates the Neolithic from the Palaeolithic ages I need not enumerate the mammals, remains of which occur in the Belgian caves — they agree generally with the mammalian fauna of the Devon- shire caves. It may be pointed out, however, that the lower and upper strata are distinguished by a certain notable feature. The lower beds are char- acterised by the great abundance of remains of the mammoth, while in the upper beds those of reindeer are much the most numerous, extinct forms being present in greatly reduced numbers. Belgian geol- ogists, therefore, have recognised- two stages in the Paleolithic period — an early or Mammoth period, and a later or Reindeer period ; which correspond to the Mousterian and Magdalenian stages. BELGIAN AND FRENCli CAVES 67 It may be added that during the earlier stage many of these Belgian caves were liable to be invaded by the rivers in flood-time. Long before the close of the Reindeer period, however, the valleys had been so deepened by erosion, that the caves in question could no longer be reached by the flooded streams. The Trou de I'Erable, for instance, which occurs at a height of some 200 feet above the river Molignee, was liable during the Mammoth period to be flooded, but before the end of the Reindeer period this was rendered impossible — the river having in the interval eroded its valley to a depth of over 170 feet. Interesting and important as the researches in England and Belgium undoubtedly are, it must be admitted that the caves of France, which are probably more numerous than those of any other country in Europe, have yielded the most prdlific and important results. Cave-exploration carried on through a long series of years by many devoted experts across the Channel has indeed given rise to a most voluminous literature, which it would be impossible to review in a short course of lectures like the present. And if to the work of French archaeologists and geologists we add that of similar experts in other countries of Europe, it is obvious that all I can do is to select for description one or two caves as representative examples. From the caves of Dordogne and Perigord, and those occurring in or near the Pyrenees, have been 68 THE TESTIMONY OF THE CAVES obtained an abundant archaeological harvest. It is by the careful study of all this material, as we have seen, that French archaeologists have been able to construct a succession of culture-stagfes, which is found to harmonise in every essential with the results of cave-exploration in other lands. The establishment of such sequences is of special importance to the geologist, for the invaluable aid they afford in his attempts to unravel the complicated history of Pleistocene times. In the following brief sketch of the evidence obtained by French archaeologists, therefore, our attention must be largely restricted to the order of succession of the cave-deposits, and to the proofs they yield of climatic changes. The abundant details gathered as to the life and doings of the ancient inhabitants of France are doubtless of surpassing interest. Into these, however, I cannot enter, but must keep as closely as I can to my special line of inquiry, namely, to the history of the Pleistocene period, and the evidences of the antiquity of man. Not the least notable of the French caves are those of Le Moustier — a little plateau of horizontally- bedded limestone and shale that overlooks the river V^zere in Perigord. Here Mousterian man appears to have found a home for a prolonged period, and has left behind him very numerous and interesting relics. His chief occupations were hunting and fishing, and from the abundance of the remains of horse, wild cattle, and reindeer, we may infer that these animals FRENCH CAVES 69 were his mainstay, supplying him at once with food and raiment. But he hunted also many others, especially musk-ox, ibex, chamois,, and red deer, and now and again mammoth and woolly rhinoceros ; while occasionally bears, lion, hy^na, glutton, arctic fox, and other carnivores were overcome. In the caves and rock-shelters of Le Moustier, we find the usual evidence of old hearths and of the manufacture of stone implements — flakes and nuclei or cores of flint being richly represented. The " lower grotto " of Le Moustier is particularly notable for the discovery there in 1907 of a human skeleton, that of a youth of sixteen years, which had obviously been covered up in Palaeolithic times. This is one of several similar discoveries recorded from the caves of France. In the cave of La Chapelle-aux-Saints in the Department of Correze, for instance, the skeleton of a man about fifty years of age was revealed, after the removal of a considerable thickness of overlying undisturbed clay a,nd cave-earth. These deposits were crowded with human relics of the same character as those met with in the grottos of Le Moustier. The body had obviously been buried in a grave, and not merely covered over with debris. Agaifi^ in the Cave of La Ferassie, Dordogne, yet another human skeleton has been unearthed. This discovery is particularly interesting, inasmuch as there is evidence in the cave- deposits of a succession of culture-stages. At the very bottom occurred a bed of red sand, above which came a layer containing Acheulian implements, and 7:0 THE TESTIMONY OF THE CAVES overlying that a stratum characterised by the presence of Mousterian artifacts. It was at. the bottom of this upper stratum that the skeleton lay. The succeeding layers contained artifacts differing from those of Mousterian type, and belonging to the Aurignacian stage of culture. In like manner the so-called Trilobite Cave, near Arcy-sur-Cure, has yielded a succession of culture-stages, including Mousterian, Aurignacian, Solutr^an, Magdalenian, and Neolithic. (See Fig. 2.) At Grimaldi, on the border of Italy, near Mentone, occur a number of caves, obviously excavated by the sea at a time when the land stood sixty-five feet lower than at present. The exploration of these caves by MM. E. Riviere, M. Boule, E. Cartailhac and Abb^ de Villeneuve has been fruitful in results. In one, known as the Grotte des Enfants, we have clear evidence of changing climatic conditions. On the rocky floor of this cave a hearth was exposed, and Palaeolithic man would appear to have been in occupation not long after the sea had retreated. The animal bones found at this level were those of the broad-nosed rhinoceros, lion, and ibex. The old Mousterian hunters having vacated the cave for a time, it was next taken posses- sion of by hysenas, who turned up the ashes of the hearth and gnawed the bones left by the hunters. The presence of hysena coprolites is additional evidence of the presence of that animal ; Palaeolithic man, however, by and by returned," probably after no long interval, for layers of ashes or hearths appear S GRIMALDI CAVES 73 immediately above the deposit in which the hyaena coprolites occur. Eventually he abandoned the cave for a protracted time, during which a thickness of three and a half feet of weathered rock-material slowly accumulated upon its floor. Resting upon this bed of barren rock-material or red clay is a well-marked hearth, showing that the cave again became the abode of man, who had now advanced to the Aurignacian culture-stage. On top of this layer of ashes and charcoal two skeletons were found — those of an old woman and a youth. The bodies had obviously been interred with care, for large stones had been so adjusted as to protect them. After this burial the cave was long deserted — a thickness of three feet of cave-earth having slowly accumulated before man reOccupied the place. His tenancy, however, was once more interrupted by a burial — that of a full-grown man, the body having been protected in the same manner as in the case of the old woman and youth. The deposits overlying this level are of consider- able thickness and consist principally of cave-earth, with a number of old hearths or layers of ashes occur- ring at irregular intervals. Thus it would seem that the cave was alternately occupied and abandoned by man through a lengthened period. The hearths indicate more or less continuous residence, but from the occurrence of artifacts throughout the beds of cave-earth it may be inferred that man often visited the place. Continuous residence was not without 74 THE TESTIMONY OF THE CAVES its dangers, as is shown by the extensive rock-falls from the roof that have taken place. Near the top of the cave-deposits a great hearth or thick layer of ashes was met with, in connection with which two interments occurred — one that of an old woman, and the other that of two young children. It is notable that many personal ornaments were found in all the graves — relics of armlets, necklaces, and decorations for head and breast. From the children's grave thousands of perforated shells were obtained, suggesting that the bodies had probably been clad in highly decorated garments. An adjoining cave (Grotte du Cavillon) has yielded a succession of deposits, similar to those of the Grotte des Enfants. Towards the bottom of the series several old hearths occurred, with bones of broad- nosed rhinoceros, cave-bear, and elephant [E. antiquus ?). Higher up the relics and remains agreed with those occupying the same position in the Grotte des Enfants — the artifacts of stone, bone, and horn being similar, and the animal remains representing cave-bear, brown bear, lion, hyaena, marmot, woolly rhinoceros, wild horse, urus, elk, red deer, wapiti, etc. The skeleton of a man was dis- interred from these upper deposits. Round the skull lay many perforated shells, and 3 number of teeth of the red deer similarly pierced.. Several inter- ments of a like kind have been discovered in other caves at Grimaldi — all apparently belonging to the same Aurignacian stage of the Palaeolithic period, < S J3 Pi Q 5 .„i-i 2 Id r1 J>^' M r7! ^^ T3 ■- U) „ o W XI s ^ > < -^^ o !^ rt ^ -0) ,t; H ^ K O "!&! w «o ^ p '" t^ O •c^ a CM) ? ) < ) S /J .> B en •c (U n "to 3 III O GRIMALDI CAVES 77 for the artifacts and the mammalian remains are the same throughout. The Grotte du Prince, the last of the Mentone caves I shall refer to, affords even clearer evidence of changing climatic conditions than the Grotte des Enfants. The deepest stratum ia this cave was of marine origin, charged with shells of still living Mediterranean species. Upon it rested an old hearth, with bones and teeth of elephant {E. antiquus), broad- nosed rhinoceros, wild horse, red deer, wild ox, brown bear, and hyaena. Higher up similar hearths or layers of ashes contained the bones of hippopotamus, horse {Equus sfenonis), ibex, cave-bear, hysena, panther, etc. The artifacts associated with this southern fauna were typical Mousterian. The deposits con- taining these remains were buried under great blocks of rock, a fall from the roof, which covered nearly the whole floor of the cave. The next folk to occupy the cave were accompanied by quite a different fauna — the mammalian remains representing horse, stag, roe, ibex (very abundant), wild cattle (not common), wolf, brown bear, cave-bear, panther, lynx, hyaena, and, most notable of all, reindeer, chamois, and marmot— a fauna obviously indicative of cold climatic conditions. The accompanying arti- facts are of a rather indeterminate character, some suggesting the Aurignacian and others the Solutrdan culture-stage, but none that can; be recognised as definitely Magdalenian. Within the past few years certain caves in 78 THE TESTIMONY OF THE CAVES Germany have yielded very interesting and import- ant evidence, which confirms and an>plifies the results obtained from the study of the caves of France and Italy. Thus, according to Professdr R. R. Schmidt, the cave of Sirgenstein — a prominent crag of lime- stone overlooking the Achtal in Wurtemberg — contains a remarkably complete^ succession of Paleolithic culture-stages : Mousterian, Aurignacian, Solutrean, and Magdalenian occurring successively, one above another. (See Fig. 3, p. 75.) Again, in the Ofnet cave, near Ndrdlingen (Bavaria), the same explorer has found the following succession of stages : — Aurignacian, Solutrean, Magdalenian, Azilian, Neolithic, Bronze, and later periods. (See Fig. 4-) In many other German caves traces of several culture-stages have been detected- — implements of a very primitive type, when these are present, always occurring in deposits at or near the bottom, while Mousterian or later artifacts appear in overlying accumulations. In this respect, Such caves have much the same tale to tell as not a few French, Belgian, and English caves, and do not therefore call for further notice. There is one famous Austrian- cave or rock- shelter, however — that of Krapina^, in Croatia (see Fig. 5, p. 83) — of which some accouiit must be given. It was explored by Dr Gorjanovic - Kramberger, who has published the results of his investigation in the Proceedings of the Anthropological Society , ^ 1^1. 13 ^ i gradual pro- gression of our r.ioe. from an vige when only very simple and rudely fashioned stone implements were used, to a time when finely-formed artifacts of bone and horn came to be manufactured by folk possessed of notable artistic talents. There are certain links in the chain of evidence, however, which are still w.inting. The evidence o\' the caves, as I h.ue said, is incomplete. But as I hope to show in my next lecture, abundant traces o\ man's h.mdiwork occur elsewhere than in caves, and these help to fill up the lacuna:', and so enable the arch.i'ologist to follow the career of our race more or less confidently through the whole course of FaLvolithic tini:s. If the arch.x\ilogical evidence of the ca\ es is not so full and complete as it might have been, the same is the case, and in larger degree, with the geological evidence. We meet with certain proofs of ch.uiging climatic conditions, as supplied by the .\hern.ue appearance and disappearance of more or less strongly contrasted mammalian faun.is. But highly suggestive 9.S that evidence may be. there is vet much more tluu Ave should like to know. Fortunately this additional knowledge has been obtained in no small measure from a study of those alluvial and superficial t'ormations "which form the subject of succeeding lectures. (See NoTv: 6.) LECTURE I\^ THE TESTIMONY OF THE RIVER-DRIFTS General Character of Pleistocene River-drifts. The Typical Succession of Deposits in the Somme Valley. Drifts of the Thames \'alley. E\idencc of Great Denudation and of Climatic Changes : Genial, Temperate, and Arctic Conditions. The- "Trail" or "Contorted Drift," and its Significance. The Succession of Archasological Culture-stages confirmed by the Evidence of the Ri\-er-drifts. The Loess : Tundras and Steppes of Middle Europe. In my last lecture I discussed the general nature of the evidence yielded by cave-accuniulations as to the antiquity of man. That evidenqS leaves us in no doubt that our PaK-eoIithic predecessors lived under climatic conditions that must sometimes have differed very considerably from those of modern Europe. In certain caves we find human relics associated with a mammalian fauna clearly indicative of warm or grenial conditions ; while in othec caves the animal remains are those of northern or even of arctic types. Now and again, indeed, representatives of southern, temperate, and northern faunas occur at different levels in one and the same cave.. The changes of climate thus vouched for necessarily imply a very long lapse of time. And the same conclusion is 101 102 THE TESTIMONY OF THE RIVER-DRIFTS forced upon us when the nature SCnd origin of the cave-accumulations themselves are considered — the red earths and stalagmites, namely, in and underneath which human relics occur. These formations, it is true, do not enable us to formulate any precise or definite estimate of the time required for their accumulation. But they at least assure us that the caves must have sheltered Palaeolithic man for many thousands of years. Nor does their evidence stand alone ; it is supported by that derived from a con- sideration of the amount of valley-erosion effected by adjacent streams and rivers since the caves first came into occupation. However important the evidence of the caves may be, it is obviously fragmentary. No one cave has yielded a complete record of Palaeolithic and later rimes. Nevertheless, by comparing the human relics from a large number of caves, archaeologists have succeeded, as we have seen, in showing that these belong to different phases of culture and are suggestive of gradual progression. It has now been definitely established that during Pleistocene times the race advanced from a very primitive stage, when man used only the simplest and most rudely-fashioned flint-implements, to a stage when finely-formed artifacts of stone, bone, horn, and ivory came to be manufactured by a notably artistic folk. We have next to learn that the evidence of the caves is supported and supplemented by that of certain other Pleistocene formations, Not only so, RIVER-DRIFl^S NOT POST-GLACIAL 103 but from a study of the latter we have acquired a clearer conception of the climatic and geographical conditions that obtained in Europe during its occupation by our savage predecessors. The formations I refer to are of very diverse origin. Amongst them are freshwater and marine deposits, and enormous sheets and heaps of glacial debris. With the history of all those varied formations, the question of the antiquity of man is intimately bound up. It is only of late years, however, that this has been recognised. For a long time archaeologists and geologists were of opinion that our cave-accumulations and certain ancient river-gravels and loams which had likewise yielded human relics and remains, all belonged to a later date than the Pleistocene glacial accumula- tions. That such is not the case I shall subsequently endeavour to demonstrate, but for the present we may restrict our attention to those ancient alluvia in which Palaeolithic relics occur along with the remains of extinct and exotic mammals. The deposits in question are well developed in such valleys as those of the Thames and other streams in the south of England, and in many similar valleys in France and Belgium. As they have been long studied in France and England, the river-drifts of those countries are w^ll known, and may be taken as more or less representative. The valleys occupied by them are obviously of fluviatile origin, and it is worthy of note that when the process of excavation had been completed, and the valleys had 104 THE TESTIMONY OF THE RIVER-DRIFTS attained their greatest depth, the land must have been somewhat more elevated than now. This is proved by the simple fact that the bottoms of the valleys lie at considerable depths below the present rivers. In the London district, for example, the floor of the old Pleistocene valley underlies the Thames at a depth of at least seventy feet, while that of the Seine in the neighbourhood of Paris is similarly depressed. The river-drifts about to "be described often cloak the valley-slopes more or 'less smoothly; not infrequently, however, they present the aspect of somewhat ill-defined platforms or terraces. Occa- sionally they appear as isolated patches or irregular interrupted sheets, especially at the higher levels ; at lower levels they are usually more continuous. Although the deposits differ considerably in character, two more or less well-marked kinds may nevertheless be distinguished — namely, "gravels and sands," and "loams and brick-earths." The gravels and sands are clearly of fluviatile origin, for they consist of materials derived from the drainage-areas in which they occur, arranged in layers parallel to the inclination of the valley-bottom. Although the stones are usually rounded and "water-worn, and the deposits distinctly stratified, such is not always the case. Occasionally the stones are rough and angular, and the confused aspect of the deposits suggests some- what tumultuous action. Now and again, indeed, large erratic blocks occur, such as might have been GENERAL CHARACTER OF RIVER-DRIFTS 105 carried by river-ice, while the bedding may be much disturbed, as if from the grounding of ice-rafts. The loams and brick-earths, while consisting :nainly of fine-grained materials, occasionally show lines and layers of so-called "angular gravel" — that is, rough stones, unlike the smoothly rounded stones that form the bulk of the gravel' and sand series. These loamy deposits frequently cloak the valley-slopes continuously, extending from the plateau above down to the modern alluvium of the river, underneath which they pass. They thus overlie the gravels and sands, the one series being as a rule sharply marked-off from the other. Their origin has been much discussed by geologists, who are far from being agreed on the subject. The deposits in question differ in character, and cannot all have been formed- in the same way. In many cases there can be little or no doubt that they are of the nature of flood-accumulations — being, like the gravel and sand beds, true river-formations. But in other cases they would appear to have had quite a different origin. Although the river-drifts of England and France have been carefully studied for many years, it must be admitted that the precise succession of events chronicled by these drifts has not yet been definitely ascertained. The recent researches of M. Commont in the valley of the Somme, however, have thrown much needed light on the subject. He has succeeded in unravelling the structure and discovering the true succession of the Pleistocene deposits of that famous 106 THE TESTIMONY OF THE RIVER-DRIFTS region, and his results cannot but have a far-reaching influence. It is interesting to reflect that the river- drifts of the Somme valley were the earliest of the kind to attract the attention of geologists and arch^ologists. This was due to the discoveries made many years ago by an enthusiastic French antiquarian, Boucher de Perthes, to whom must be assigned the merit of having been the first to direct the attention of the scientific world to the occurrence of human implements associated with remains of extinct animals in beds of undisturbed Pleistocene gravel and sand. It was not, however, until after cave-explorations had convinced geologists of the antiquity of man, and his contemporaneity with the Pleistocene mammals, that the importance of the discoveries in the Somme valley was realised. Over fifty years have elapsed since then, and it need hardly be said that in the interval our know- ledge of river-drifts has greatly increased. So abundant is the evidence, indeed, that even the briefest summary of the more important phenomena would occupy the time of many lectures. The most I can attempt at present is to sketch in outline the more notable features of the evidence furnished by certain typical or representative valleys. The full meaning of that evidence, however, will not appear until we have attained a comprehensive view of the Pleistocene formations as a whole. When we have learned what these have to tell us of the climatic and geographical conditions of Europe during Pleistocene RIVER-DRIFTS OF SOMME* VALLEY 107 times, the significance of the evidence we are about to consider will become more apparent. Nevertheless, apart from that wider outlook which we may hope to gain, the phenomena of the old rKyer-drifts have yet an intelligible tale to unfold, which may be followed without reference to the collateral evidence supplied by other Pleistocene formations. Within the last few years the river-drifts of the Somme have again claimed the special attention of geologists and archaeologists on account, as I have said, of the remarkable results obtained by M. Commont, some account of which must now be given. The valley of the Somme has been ex- cavated in chalk, the surface of which is of course largely concealed under river-drifts of one kind or another. At relatively high levels, however, the bare chalk appears, forming short and relatively abrupt slopes. These, as Professor Prestwich showed long ago, are the frontal edges of plg,tforms of erosion. The drifts lying on such a platform at a consider- able elevation constituted his "high-level" series, while those occupying the bottom of the valley formed his " low-level " series. Where no such high- level platform was visible, he had necessarily some difficulty in distinguishing between one series and the other — their relative height above the river, rather than any distinguishing characters of the drifts them- selves, determining whether they should be classed as "high-" or "low-level." For a number of years M. Gommont has with 108 THE TESTIMONY OF THE RIVER-DRIFTS great zeal worked out the structure of the drifts of the Somme valley, and shown that the continuity of the valley-slopes is interrupted by. three successive platforms of erosion, which, although often cloaked and concealed by the drifts, he has yet been able to follow by means of borings and openings of one kind and another. (See Fig. 6.) The uppermost terrace occurs at a height of about lOO feet above the Somme, the middle terrace at about seventy feet, while the lowest, beginning at a height of forty feet, passes underneath the river. Beds of fluviatile gravel and sand rest upon these platforms of chalk — to which indeed they are confined. The deposits of each of the three terraces have yielded artifacts — the middle terrace being much the richest repository. The lowest terrace is for the most part inaccessible, passing as it does under the modern alluvium of the river ; a few artifacts, however, have been obtained from it. In the highest terrace implements rarely occur. It is important to note that all the artifacts met with in the gravels and sands of the terraces are Chellean — the most primitive forms occurrmg in the coarse gravels that repose immediately upon the chalk, while the better formed artifacts of the same type seem to be restricted to the overlying sands. Since Chellean artifacts occur upon all three terraces, it is clear that the excavation of the Somme valley must have been practically completed in early Pleistocene times. The Chellean stage is thus shown to have been of prolonged duration, for however rapidly i t M s O H S < > s -< ti> RIVER-DRIFTS OF SOMME VALLEY 111 the Somme may have lowered its bed, it must never- theless have taken a very long time indeed to have deepened its valley for more than ido feet. The loams and brick-earths are of later age than the gravels and sands of the terraces, which they over- lap and conceal- — extending more or less continuously from the highest levels down to the bottom of the valley. They differ in origin from the terrace-deposits which, as we have seen, are river-ac,cumulations. The loams and brick-earths, on the other hand, are some- what of the nature of rain-wash, and consist of materials which have come down from the- adjacent plateaux and hill-slopes. They show traces of bedding, often indicated by lines and layers of angular gravel, but these dip directly down the valley-slope. Not only are they thus steeply-inclined, but they often contain land-shells and plant-remains. All the appearances they present, in short, are suggestive of the action of surface-waters, such as rain or melting snow, which have carried away and re-arranged over the flanks of the hills and the sides of the valley, the loams and other residual products resulting from the long- continued weathering of the chalk. M. Comment shows that in the Somme valley there are three separate and distinct "loams," one overlying the other. The lowest and oldest {limon fendilld') is characterised by the presence of Acheulian artifacts, while the overlying loam [ergeron) with its associated angular gravels, contains Mousterian im- plements — Magdalenian artifacts appearing only 112 THE TESTIMONY OF THE RIVER-DRIFTS towards the top of the same accumulation. The uppermost or superficial loam, on the other hand, is a brick-earth of quite recent origin, due partly to the washing of rain, and partly to the- weathering of the immediately underlying deposits. M. Commont's results, therefore, are highly important in several respects. In the first place, they confirm the evidence of the caves as to the succession of the several archaeological stages from the Chellean to the Magdalenian — the only stages not represented being the Aurignacian and Solutr^an. In the second place they lead us to believe that when the river- drifts of such great valleys as those of the Seine and the Thames have been as assiduously investigated, they also will show a definite succession of stages. In my first lecture I made reference to the calcareous tufa of La Celle, near I^aris. In the section (Fig. 9, p. 233), copied from one by Professor Obermaier, the position of that particular tufa in the Pleistocene series is clearly indicated. The gravels A may safely be assigned to the warm Chellean stage, while the over-lying tufa shows the passage from that very genial stage to the succeeding cooler Conditions of Acheulian times. The tale told by the Pleistocene deposits of the Thames valley, although not quite so clear as that obtained by M. Commont from his, researches in the valley of the Somme, is yet full of interest to archaeologists and geologists alike, for they have yielded many organic remains and an abundant RiVER-DRIFtg OF THAMES VALLEY ll3 harvest of artifacts. The mammals represented agree generally with those recorded from our cave- accumulations, and we may safely- conclude therefore that the old river alluvia and the cave-deposits are approximately of the same age. Among the species we note hippopotamus, broad-nosed rhinoceros, and straight-tusked elephant — members of the southern group — as well as notable members of the northern group, such as reindeer and musk-ox. Only now and again, however, do these strongly contrasted faunas occur at separate and distinct horizons in the alluvia. Very frequently the remains are commingled, and this formerly led to the belief that great seasonal migrations had taken place — the hippopotamus and its congeners wandering north in summer, and the reindeer and its associates coming south in winter, so that the same area was occupied alternately during the year by the two faunas. That view, however, has been abandoned, and geologists now recognise that the intermingling of mammalian remains in the Pleistocene alluvia is most probably due to the conditions under which such alluvia are deposited. Rivers are constantly cutting into and redistributing their accumulations, so that any organic relics they may contain are liable to be disinterred and reburied, side by side it may be, with others belonging to a later date. Not only so, but now and again an active river will trench its gravels more or less extensively and again fill up the excavations it has made. In this way sand and gravel have become banked against H 114 tHE TESTIMONY OF THE RIVER-DRIFTS similar deposits of greater age, and the line of junction between the two sets of beds cannot always be recognised — the one sometimes seeming to shade into the other. But let it be noted that where no such disturbances of the bedding have taken place — where the alluvia have obviously been regularly accumulated, bed above bed, their mammalian remains are found to belong usually to one and the same fauna — sometimes to the southern and temperate group, sometimes to the northern. And the same is the case with the shells enclosed in the alluvia. Now and again, for example, these represent species which no longer live in English rivers [Corbicula fluminalis and Unio littoralis) but in those of more southerly latitudes. On the other hand, the shells present in undisturbed alluvia are frequently indicative rather of cold- temperate, or even pf arctic conditions, and a similar tale is told by the plp,nt-remains. It would thus appear that the river-drifts in question are the records of a long period of time, during which considerable climatic changes super- vened. The length of the period is indicated not only by those climatic revolutions, but by the enormous amount of work done by the Pleistocene rivers. The formation of the gravels with their associated beds of sand and loam must have been a protracted process, for they not only attain a great thickness but over- spread broad areas. In short, all the phenomena of the drifts themselves, not to speak, of the modifications of surface experienced by the wide regions traversed RIVER-DRlFrS OF THAMES VALLEY 115 by the valleys, demonstrate how great has been the denudation of the land since it was first visited by Palaeolithic man. And all that denudation — the excavation of the valleys and the lowering of the land-surface — it must be remembered, was practically completed before the apparition of his Neolithic successor. It is not too much to say that the modification of surface effected since the beginning of the New Stone age is too inconsiderable to be noticed, when one thinks of the great changes brought about during the preceding Palaeolithic period. The Pleistocene river-deposits, — the gravels and sands — occasionally appear in the valleys of the Thames and its tributaries as tolerably well-defined terraces, rising one above another. More frequently, however, this terraced aspect Is obscure. Nevertheless we can hardly doubt that the gravels were deposited during the gradual excavation of the valley, so that those occurring at the higher levels must be the oldest. The associated loams and brick-earths often overlap the gravels, and consequently extend to a higher level. The latest of these loamy accumulations are certain overlying tumultuous sheets and heaps of earthy materials and angular gravels, which spread downwards from the heights overldoking the valleys. The oldest river-deposits of the Thames have yielded remains of the southern and temperate group of mammals (hippopotamus, broad-nosed rhinoceros, straight-tusked elephant, mammoth, deer, horse, lion, etc.), but no definitely arctic types, for the mammoth 116 THE TESTIMONY OF THE RIVER-DRIFTS had a wide range, and although chiefly associated with northern forms, seems not infrequently to have invaded the provinces occupied by the temperate and southern species. Corbicula fluminalis — a southern shell — appears in the same deposits, while man is represented by artifacts of types characteristic of the earlier culture-stages — Chellean in the lower beds, Acheulian higher up. The river-drifts of somewhat later age are of particular interest. From them have come immense quantities of implements all belonging apparently to the Mousterian culture-stage. It is notable that among these later river-deposits several old land- surfaces have been discovered. These are known as "Palaeolithic floors," and consist" of a few inches of angular gravel, crowded in places with unabraded implements and flakes, which obviously occur just where they were left by the Palaeolithic workmen. One such floor has been traced by Mr Worthington Smith over a wide area. He believes it must formerly have extended throughout East Middlesex into Herts as far as Hertford and Ware, and on both sides of the Thames from London to the Nore. Plant-remains occur plentifully on that old land-surface, but they are often so fragile and badly preserved that their deter- mination is not always possible. Impressions of portions of leaves, and stems of grass, rushes, and sedges are not uncommon ; and birch, alder, pine, yew, elm, and hazel have been recognised. The common male fern is of frequent occurrence, while RIVER-DRIFTS OF LEA VALLEY 117 the royal fern {Osmunda regalis) is present in such profusion as to suggest that the fronds might represent the litter or bedding on which the Mousterian savages had rested. Upon the whole, therefore, this assem- blage of plants indicates a temperate climate. That conclusion is not gainsaid by the mammalian remains occurring in the brick-earth, which, although they include mammoth and reindeer, are chiefly of temperate types. (See Note 7.) The drift gravels we have hitherto been considering occur at relatively high levels — seventy to eighty feet above the sea — and belong, therefore, to the oldest series of Pleistocene alluvia. Of younger age are the low-level gravels that appear at heights of forty feet and less above the sea. Amongst these Mr S. H. Warren recently discovered in the valley of the Lea an arctic plant-bed, that varied from one to two feet in thick- ness. It occurs at a depth from the surface of fifteen to seventeen feet, and is sometimes exposed in con- tinuous section for twenty or thirty yards. Among the plants recognised are a number of alpine-arctic mosses, now either entirely extinct in Britain, or confined to the summits of the higher Scottish mountains, while all the remaining species are just such as might at present be seen flourishing in the arctic mainland of Europe. The testimony of the accompanying shells agrees with that of the plants, for they indicate a climate like that of Lapland or southern Iceland. The low-level valley-gravels in which this notable plant-bed occurs have yielded 118 THE TESTIMONY OF THE RIVER-DRIFT'S remains of mammoth, woolly rhinoceros, reindeer, horse, ox or bison, lemming, etc., but the only human implements met with are remanii forms — that is to say, they have been derived from the higher and older terraces. A long interval must have elapsed between the deposition of the high-level gravels and the formation of these low-level deposits. This is at once indicated by the fact that the latter only began to be formed after the river had gradually deepened its valley by some sixty or seventy feet below the level at which it had flowed when the Mousterian tool-makers wandered along its shores. That cold climatic conditioris supervened sub- sequent to the formation of the high-level gravels and brick-earths has been known for many years. Almost everywhere these deposits are overlaid by amorphous materials, underneath which the river alluvia are much disturbed, flexed, folded, and con- torted. These are the phenomena known to geologists as " Trail," and they are well exemplified and explained by the so-called " flowing soils '" of high arctic lati- tudes. In such regions the frozen soils and subsoils, when exposed to the heat of the stin In summer, tend to become semi-fluid and to creep and flow more or less sluggishly down the slope§. In the London district sheets and masses of similar origin, creep- ing outwards from the heights into the valleys, have ploughed their way into the old river-drifts, and every- where produced much confusion. Embedded at all CONTORTED RIVER-DRIFTS 119 angles in this muddy or earthy material are found abraded and whitened flint implements, which have all the appearance of having been long exposed to the weather, and were doubtless caught up and carried down from the old land-surface that overlooked the valleys. These remarkable sheets of "contorted drift " represent the closing stage of the Paleolithic period, so far as the valleys of the Thames and its tributaries are concerned. No Palaeolithic relics orcur in any deposits that may chance to overlie the " contorted drift " — the oldest artifacts met with in any such younger undisturbed formations are of Neolithic age. Thus the evidence, so far as I have traced it, would seem to show that certain high-level river-gravels were deposited at a time when the climatic conditions were upon the whole temperate ; while arctic conditions prevailed during the accumulation of the low-level alluvia. But, as already indicated, the high-level gravels contain now and again large stones and blocks which could not have been transported by river-action alone, but have apparently been carried by ice-rafts. Now and again, also, the alluvia have yielded relics of the northern mammalian fauna, as reindeer, woolly rhinoceros, and mammoth. But the testimony of these remains is so far negatived by the occurrence in the same deposits of characteristic temperate forms. This commingling of discordant faunas is probably largely due to the mode of forma- tion of the beds in which the remains occur. Nothing 120 THE TESTIMONY OF THE RIVER-DRIFT^S is more difficult, as Lyell long ago recognised, than to settle the chronology of fluviatile deposits. " It is almost equally difficult," he remarks, "to avail our- selves of the evidence of organic remains or of the superposition of the strata, for we may find two old river-beds on the same level, in juxtaposition, one of them perhaps many thousands of years posterior in date to the other." For all that we can tell, there- fore, there may have been several oscillations of climate during the gradual accumulation of the high-level alluvia. It is enough, however, for our present purpose to realise the fact that the oldest alluvial deposits are characterised by the presence of the southern and temperate group — while the ac- companying human relics are among the most primitive recognised by archaeologists. The low-level gravels, on the other hand, which must be many thousands of years younger than those of the higher level, were unquestionably accumulated under very cold conditions. The plant-remains indeed leave us in no doubt that at the time of the formation of the low-level terrace of the Lea valley, southern England must have been a region of tundras. Hitherto no evidence is forthcoming to show that man was an occupant of our country at that epoch. From the Pleistocene alluvia of the London district, therefore, we cannot as yet educe a complete con- secutive history. They tell us of a Thames, the bed of which in early Palaeolithic times was considerably more than loo feet above that of the modern river. DEEPENING OF THAMES VALLEY 121 Through the long Pleistocene period it continued slowly to deepen its valley — at one time flowing as a broad placid stream, now and again as a torrential river, rising high above its ordinary level and flood- ing extensive regions. And thus ds the erosion pro- ceeded, the slopes of the gradually deepening valley became mantled with gravel and swathed in broad sheets of flood-loam. When the river had sunk its channel some seventy feet below its earlier bed, a very cold climate supervened ; the land was then sparsely clothed with arctic mosses and dwarf willows and birches, while ever and anon moving masses of mud and stones made their way into the valley from the heights above. Ajpd with this picture of arctic conditions we reach the end of the Pleisto- cene story, so far as that is revealed by the old alluvia of the Thames. The excavation of the valley, as already remarked, had been practically completed long before Neolithic man appeared upon the scene. Having now shortly outlined the more important evidence derived from the old riveij alluvia of France and England, we may next briefly consider some of the most significant phenomena displayed by the valleys of middle Europe, and for this purpose it will suffice to confine your attention to one particular kind of valley-drift, which, although not unknown in western Europe, yet attains its most conspicuous development in the central region of the Continefnt. The evidence I have cited from the river-drifts proves that Palaeo- lithic man was contemporaneous in western Europe 122 THE TESTIMONY OF THE RIVER-DRIFTS sometimes with the southern and temp.erate mammalian group, and at other times with the jiorthern forms. Did time suffice I might adduce similar evidence from the river-deposits both of central and s'outhern Europe ; but I shall content myself by describing very shortly the peculiar formation known to geologists as loss. Loess is certainly one of the most remarkable superficial accumulations of central and west central Europe. Typically it is a fine-grained, yellowish, calcareous, sandy loam, consisting very largely of minute grains of quartz, with some admixture of argillaceous and calcareous matter. Upon the whole the quartz-grains are well rounded, although often enough sharply angular. Frequently the accumulation shows a porous structure, and is penetrated by long, approximately vertical root-like tubes or canals, lined with calcareous matter, which cause the deposit to cleave or divide in vertical planes. Hence it usually forms upright bluffs upon the margins of streams or rivers which intersect it. It is usually unstratified, except now and again towards the bottom of the deposit, where iniercalated layers, and sometimes even thick beds of s^and, make their appearance. The loess is essentially a deposit of the low grounds, and is well developed in the broad river- valleys of western and central Europe, as in those of the Seine, the Garonne, the Rhone, the Meuse, the Moselle, the Rhine and its tributaries, the Danube and many of its affluents, such as* the Drave, the Save, the Morava, and the Theiss. It also extends LOESS MAINLY OF vEOLIAN ORIGIN 123 as a narrow belt along the southern margin of the great plains of north Germany. It is in southern and south-eastern Russia, however, where it attains its widest development, covering as it does an immense tract between the valleys of the Pruth and the Volga. Throughout this vast region it is usually very dark in colour, forming what is known as the Black Earth. I shall not at present consider the source of the materials of which the loess is composed. It is obvious enough that the accumulation has in some places been arranged by water. Thus here and there, especially 4t or towards the bottom of the deposit, distinct traces of bedding may be seen, and the beds have yielded fresh-water shells. This, however, is exceptional. Loess is, for the most part, a subaerial accumulation — a wind-blown deposit. This is shown not only by the rounded character of its minute constituents and by the general absence of bedded arrangement, but by the abundant presence of land-shells and the frequent occurrence of mammalian remains. Its organic contents are essentially terrestrial. Moreover, its particular distribution — the mode in which it occurs — points clearly to the action of prevalent winds. Thus, although it is widely developed over low-lying regions, it nevertheless sweeps up to heights of 200 to 300 feet and more above the bottoms ©f the great river- valleys. Not only so, but ever an4 anon it extends across the hills and plateaus between adjacent valleys, wrapping the whole land, in short, like a mantle. Again, in many places, we find it heaped up in the 124 THE TESTIMONY OF THE RIVER-DRIFrS lee of hills, the exposed windward slopes of which bear no trace of it, while in certain valleys it shows a similar partial distribution. Among the organic remains yielded by the loess are some that indicate arctic conditions, while others are strongly suggestive of a steppe climate, and yet others tell us of forest lands. It is impossible that a,ll the creatures referred to could have lived side by side in the same region, and annual migrations will not wholly explain their appearance in the same deposit. The evidence leads to the conclusion that the accumulation of the loess must represent a long period of time during which climatic changes took place. Fortunately now and again the loessic accumu- lations exhibit a succession of faunal zones — different suites of organic remains occurring at different levels. And a similar and corresponding succession has been discovered in not a few caves in middle Europe. A tundra fauna is the earliest of which we have any record in the loess. In my first lecture I gave a general account of that fauna and its former distribution throughout what are now the temperate latitudes of our Continent. It ranged, as you may remember, from Russia as far west as England, and from northern Germany south to the forelands of the Alps— its remains occurring not only in caves, but often abundantly in loess. In the latter, just as in the rock-shelter at Schweizersbild, the remains in question are most plentiful towards the bottom of the deposits, and gradually become less numerous in the beds above, BLIZZARDS OF PREHISTOBIC TIMES 125 until they finally disappear. Before the last trace of the arctic forms has vanished, however, relics of the steppe fauna begin to occur, and eventually predominate. In a word, there was no sudden dying- out of one fauna and precipitate appearance of another, but a gradual replacement, consequent doubtless upon changing climatic conditions. One of the notable characteristics of tundra and steppe climates, let me remind you, is the frequent occurrence of tempestuous winds and blizzards, during which much destruction of animal life may take place. Abundant evidence is forthcoming to show that the wild animals of our prehistoric steppes and tundras were often done to death in their hundreds and thousands. Again and again immense heaps and accumulations of their skeletal remains have been encountered in the loess — appearances strongly recalling the similar bone-finds of Siberia and North America. The deposits in which the European bone-finds occur are of wind-blown origin, and we seem justified, therefore, in concluding that the animals perished in snowstorms. In these low latitudes, however, we could not expect to meet with ice-formations like those of Siberia. That drifted snows did formerly accumulate in middle Europe, however, and were preserved for long periods under coverings of sand and other materials, we have good reasons for believing. Indeed, even at the present day the drifted snows in south-east Russia are occasionally buried under sand and .^o persist for years. 126 THE TESTIMONY OF THE RIVER-DRIFTS In one case recorded by Borszcdff, what appeared to be an ordinary sandhill proved to be a mass of congealed snow cloaked in sand about a foot in thickness. Immediately under the surface the snow was granular and n^vd-like, but a little deeper it was firm and solid ice. This was in one of the tributary valleys of the Ilek, in the steppes south of Orenburg, about the 50th parallel — a relatively dry region. If snow can be thus preserved in a low-lying region so far south, we may readily conceive how in the steppe epoch of middle Europe snow-drifts, similarly protected might now and again have persisted for years. But it was during the preceding tundra epoch that this would be most commonly the case. And much interesting evidence is forthcoming to show that in many places thick sheets of congealed snow did accumulate and become buried and! preserved at that time. In a subsequent lecture I shall give some account of the so-called " rubble-drifts " of central and southern Europe. For the pfresent, however, I need only say that the drifts in question consist of rock-rubbish, which has moved down gentle hill- slopes and spread itself over adjacent low grounds — phenomena that indicate the former presence of deep snow-drifts, in and upon which the rocky debris travelled. These were not glaciers, but merely sheets of snow and neve, charged with and covered by earthy matter and rock- fragments, and set in motion in spring and summer when thaw ensued. The occurrence in the earthy debris of SNOW-DRIFTS OF PREHISTORIC TIMES 127 bones of the reindeer and other Pleistocene mammals shows that the deposits belong to prehistoric times. Again, certain phenomena connected with the river gravels of the same period lead to the conviction that the drainage was often interfered with by snow-drifts. As Mr Darwin suggested to me many years ago, the river valleys would seem to have become filled in places with alternate sheets of congealed snow or ice and layers of gravel and shingle. Long afterwards, when the interbedded strata of ice melted slowly away, the associated river detritus settled quietly down, and owing to the differential movement of the subsiding materials, the longer stones arranged themselves in lines of least resist- ance — so that now we find them most usually standing on end in the gravel beds. (See Note 8.) Thus, apart from the evidence supplied by the bone-accumulations of the loess, we may reasonably infer that snow-drifts were of common occurrence in middle Europe in prehistoric times. Doubtless most of the snow which covered the plains of our Continent in winter melted and disappeared in summer, just as is the case in the tundras and steppes of our own day. The carcasses of animals that may have perished in blizzards would thus most frequently become uncovered in spring, to be devoured by hyaenas, wolves, and bears, and the disarticulated skeletons might often be bleached and weather-worn before they were finally buried in loess. Nor was it only in plains and open valleys that sudden death may have 128 THE TESTIMONY OF THE RIVER-DRIFTS overtaken large numbers of animals. In tundras and steppes alike the wild and semi-wild denizens of the plains seek refuge from the drifting snow in the fissures, caves, gullies, and ravines of the hills and mountains, where they are sometirries frozen to death or smothered. Herbivorous and carnivorous animals thus often perish together, for ih the presence of a common danger, whether it be prairie or forest fire, or flood or blizzard — natural antipathies and animosities are in abeyance, and all alike struggle to escape. Man, as I have mentioned, lived in middle Europe in tundra times, and we have abundant evidence of his presence there throughout the succeeding steppe epoch. Again and again human relics have been met with at all levels in the loess throughout central Europe. Thus in the valleys of the Danube and some of its tributaries, they have been discovered in undisturbed loess at depths of from 20 to nearly 100 feet from the surface. Not a few of these finds evidently represent old prehistoric camping stations — marked by the presence of quanti- ties of charcoal and ashes, burnt and calcined bones, together with worked flints, bones, and ivory. Among the animal remains are those of mammoth, woolly rhinoceros, musk ox, reindeer, elk, horse, lion, glutton, bear, wolf, arctic fox, common fox, and hysena. In concluding this brief review of the tale told by the alluvial and ceolian deposits of the Pleistocene, I may remark that while this tale clearly supplements that of the cave-accumulations, it nevertheless does LAND-BRIDGES ACROSS MEDITERRANEAN 129 not give us a complete history of* the period. It is true I have omitted all direct reference to certain geographical changes which may- be inferred from some of the phenomena we have been considering. For instance, I might point out how the distribution of Pleistocene faunas and floras is indicative of con- siderable changes in the relative level of land and sea. At one or more stages of the period notable land- bridges must have connected Europe with Africa across the Mediterranean. It was doubtless by such connections that the southern mammals invaded our Continent, and the former presence of the same species in Britain likewise bespeaks a union between our islands and the Continent. At various epochs of the Pleistocene period Palaeolithic man might have walked dry shod from Africa into Europe, and across what are now the English Channel and the North Sea into Britain. But the evidence for such changes is not derived only from the geographical distribution of the Pleistocene animals and plants, and I prefer, there- fore, to pass from the subject for the present. In a later part of this course I shall return to it after having discussed the evidence supplied by the glacial formations. Just one other point, and I have done. It must be kept in view that the river alluvia containing relics and remains of Palaeolithic man and his congeners occur for the most part in middle and southern Europe. In Britain they are hardfy met with north of the Thames valley, and they are similarly absent 130 THE TESTIMONY OP THE RIVER-DRIFTS from the corresponding latitudes on the Continent. But we encounter them in all the great valleys of middle and southern Europe. It is quite otherwise with the younger alluvia containing relics of Neolithic man. These may be said to appear in every latitude from the extreme north to the extreme south of Europe, and from the shores of the Atlantic in- definitely eastwards. LECTURE V THE TESTIMONY OF THE GLACIAL FORMATIONS I. — Glacial Action Glacial Action. Formation of Screes. Flowing Soils of Arctic Regions. Breccias of Gibraltar, their Character and Origin. Alpine Glaciers. Ice-cap Glaciers of Norway. Arctic and Antarctic Ice-sheets. Geological Work of Glaciers and Ice-sheets. Superficial and Ground- or Bottom-Moraines. Rock-basins and other Evidence of Glacial Erosion. Over-deepening of Valleys by Glacial Action. In preceding lectures I have discussed the general phenomena of the Pleistocene cave-accumulations and river alluvia with their relics and remains of man. From the study of these we have learned something not only of prehistoric man himself, but of the climatic conditions under which he lived. We have seen that for some time at least he hunted the rein- deer in the south of France, when a large part of what is now temperate Europe must have been a dreary tundra — a region clothed with an arctic-alpine flora ; swathed in snow during winter, subject to severe storms and blizzards at the changes of the seasons ; and to excessive floods in spring and summer. Reflecting upon all this we naturally m 132 TESTIMONY OF THE GLACIAL FORMATIONS ask : — .What relation do the cave-accumulations and river-alluvia bear to the glacial formations ? Did Paleeolithic man live in Europe before, during, or after the Ice Age? Obviously to satisfy ourselves on this point, we must devote attention to the records of that age ; for we are well assured that the glacial formations belong to the Pleistocene period — to that period, in fact, when Palaeolithic man first appeared in Europe. These formations present us with so many com- plex phenomena — so many difficult problems are suggested by them — that before entering on their consideration I think it well to devote this lecture to the study of existing glacial action. The present, as geology insists, is the only key to the past, and at the risk of wearying those who may already be familiar with glacial phenomena, I can hardly avoid describing such as seem to bear most directly on the origin of the Pleistocene glacial accumulations. Glacial action is a somewhat com- prehensive term, however, for it signifies not only the work of glaciers or ice-rivers, but that of frost and snow, which, as we all know, may characterise regions where no glaciers exist. As these several kinds of ice action must have operated in Pleistocene times, it is necessary that we should understand how they work and what they can do. First, then, let us consider the action with which we are most familiar- — that of frost". Frost pulverises our soils and causes rock-surfaces to crumble down. ACTION OF FROST ISS The modus operandi is simple enough. All rocks are more or less porous, and thus tend to absorb water, which when it freezes causes their constituent particles to separate. Again rock-beds and rock-masses of every kind are traversed by natul"al division-planes, along which water makes its way from the surface. When water freezes in the joints and fissures these are of course suddenly widened — in. a word, the rocks are disintegrated and ruptured, and when thaw follows, the loosened particles, blocks, and fragments tend to fall asunder. In temperate latitudes like ours the results of this action are most conspicuous at high elevations, where the naked mountain-tops are often entirely concealed under their own ruins. Should rupturing take place in cliffs or upon steep slopes, the debris shoots downwards, either to accumulate in heaps below, or to cloak the declivities with rough sheets or "screes." Nor does the process of destruction stop here, for the screes are themselves attacked — their rock-fragments being further dis- integrated and ruptured. In this way much of the material eventually becomes reduced to a kind of rough gravel, grit, and earth, that sinks or is washed down between the larger blocks. During heavy rain the debris may be saturated to such an extent that not infrequently it creeps and flo^s down hill ; and the same movement often takes pla'ce when the snows are melting in spring. It will be readily understood that the phenomena described attain a great development in arctic regions. 134 TESTIMONY OF THE GLACIAL FORMATIONS Frozen hard during winter and covered with thick snow, the screes in those regions remain steadfast till summer returns. Then the snow melts, the earthy debris thaws, becomes semi-fluid, and moves sluggishly downward — frequently spreading far out over low-lying and relatively level or gently-inclined ground. The movement may be arrested at night by the freezing of the sodden material;. Even then, how- ever, motion does not at once cease, for while freezing goes on expansion takes place, and the masses are compelled to move in the direction of least resistance — that is, of course, down hill. Similar phenomena have been observed at high elevations in temperate regions ; certain valleys in the Rocky Mountains, for example, are thickly covered with earthy debris which, when the snow melts on the mountains, becomes saturated and creeps slowly forward. Hedin, the famous Swedish explorer, has given a graphic description of the great moving sheets of water- saturated detritus of the Tibetan, highlands, which his caravan had the utmost difficulty in crossing. It may be remembered that in my last lecture I referred to the "flowing soils" of. Arctic regions as helping us to explain the origin of those remarkable sheets of confused earthy and stony materials that are so frequently associated with the ancient alluvia of Pleistocene times. In all parts of Europe forma- tions of the kind occur, in illustration of which I may give a short account of the massive breccia or solidi- fied rock-rubble of Gibraltar. BRECCIAS OF GIBRALTAR 135 The Rock of Gibraltar, as everyone knows, is a narrow arete or sharp ridge, that" trends nearly due north and south. At its north eijd it terminates in a steep wall-like front, some 1350 feet in height. On the east side the declivities a,re hardly less pre- cipitous, while on the west the gradients are not nearly so steep. The culminating" point of the ridge reaches about 1400 feet above the sea. As regards the geological structure of Gibraltar, all that need be said is that the ridge coincides with the outcrop of a thick limestone or series of limestone-beds that dip towards the west at a high angle, so as to pass under certain dark shales which make their appearance towards the base of the Rock. The overlying breccia is a confused aggregate of angular fragments and blocks of limestone, embedded in a matrix of calcified grit and earth — the whole accumulation forming a mass as solid as the limestones over which it has been spread. It varies in thickness from a few feet up to thirty or forty yards, being thinnest and occur- ring only in patches on the steeper slopes of the Rock, but thickening out upon the lower grounds, as it approaches and enters the sea. It has manifestly been derived from the breaking-up of the limestone, for it contains no stones foreign to the place. That the breccia is no longer forming becomes obvious at a glance. Everywhere- it has been worn, furrowed, and trenched by rain and torrents, in pre- cisely the same way as the limestone where that is exposed. In short, the breccia presents the roughly 136 TESTIMONY OF THE GLACIAL FORMATIONS corrugated surface characteristic of calcareous rocks which have been long subjected to the chemical action of rain. The red earth or insoluble residue has as usual collected in the numerous irregular grooves and hollows of the rugged surface. In further proof of the great antiquity of the breccia, it may be mentioned that long after it had become thoroughly consolidated, subsidence of the region supervened and the Rock of Gibraltar was partially submerged. Under those conditions the sea carved out certain well-marked horizontal terraces which are seen traversing the face of the Rock. These old sea- margins, it may be noted, have been hewn sometimes in limestone, sometimes in consolidated rock-rubble or breccia. All the appearances presented by the breccia lead to the belief that it is an ancient scree due to the action of frost and thaw. Only severe frost, indeed, could have wedged out the larger blocks in the mass, some of which measure several yards in diameter, and must weigh twenty or thirty tons at least. The larger blocks and smaller fragments are all alike sharply angular, which would not have been the case had the accumulation owed its origin to waves and breakers. Nor can it be due to the action of torrents. Torrents might have swept down small fragments to the base of the Rock, but they could hardly carry these and still less the larger blocks across the low grounds, for a distance of 500 or 600 yards, over which the inclination of the ground is not BRECCIAS OF GIBRAVrAR 137 more than 8° or 9", and in places does not exceed o o 2 or 3 . How then was the debris spread over those low- lying areas ? Frost might well rupture and shatter the limestones of the ridge, and the steep slopes of the Rock might be sprinkled with fragments and covered with screes, formed in the usual way. We may even admit that the impetus acquired by blocks detached by frost along the summit of the Rock might suffice to bring these down the more abrupt declivities, but it could not possibly transport them across the low grounds to the sea. The most reasonable explanation of the breccias is at once suggested by the arctic phenomena of "flowing soils and debris " to which some reference was made in my last lecture. Gibraltar, where at present frost is unknown, would seem formerly t©. have experienced severe winters, when the limestones along the crest of the arete were broken up and shattered. At that time heavy snow may well have swathed the flanks of the Rock and extended over the low grounds. Under such conditions scree materials, slipping and rolling down, would be scattered over the snow- covered slopes, and possibly snow-shoots or avalanches might now and again occur. Judging from what takes place in many high mountain tracts, there might even have been a slow down- ward movement of snow and nev^ with much in- cluded rock-rubbish. In one way or another heaps of commingled snow and debris would tend to collect 138 TESTIMONY OF THE GLACIAL FORMATIONS about the base of steep declivities and precipices. With the return of spring and summer the snow would melt, and the sodden rock-rubbish would then move sluggishly forwards like the so-called "earth- glaciers " of the Rocky Mountains. The limestone-breccias of Gibfaltar are only one of many similar ancient sheets and heaps of rubble that were abundantly developed in Pleistocene times throughout Europe. The distinguishing character of these old screes is their worn and denuded aspect. No longer in the process of formation, they are gradually being worn and wasted away. More- pver they frequently make their appearance in regions where, as at Gibraltar, frost and snow are at present unknown. But of more importance toi the student of Pleistocene geology is the action of true glaciers or ice-rivers. Everyone is familiar with the fact that in the Alps and other regions wBere glaciers occur there is a line above which more snow accumulates than is melted or evaporated. Needless to say the snow cannot augment in thickness to an indefinite extent. From abrupt declivities great shreds are constantly dropping away as avalanches, but however extensive and numerous these may be, they do not sufifice to drain the vast supplies of snow which collect in the profound amphitheatrical recesses of the mountains. This work is effected by a regular system of ice-rivers. At low temperatures snow is dry and powdery. When the temperature rises FORMATION OF GLACIERS 139 melting begins, and drops of water set free at the surface sink into the subjacent layer, where they again freeze, forming envelopes around the snow crystals. Eventually all the snow becomes in this way granular, the air being partially expelled in the process. At the same time the pressure exerted by the accumulating snow tends still further to squeeze out the air entangled in the underlying opaque nev^, which thus gradually passes into' the condition of transparent ice. The ice now begins to creep outwards from the capacious cirque in which it originated, and makes its way downward, flowing very much as if it were a viscOus body. Shortly after starting on its journey, it may be joined by similar flows proceeding from adjacent cirques. The young glacier thus developed may eventually coalesce with one or more ice-streams coming from tributary valleys, and as it descends to lower levels may itself become the tributary of some great trunk glacier, formed by the union of many smaller ice-flows. The movement of a glacier is differential- — it is greatest about the middle or centre of the flow, but decreases towards the sides and bottom, where the motion is retarded by friction. The rate of move- ment varies ; in small glaciers it may be only a few inches up to a foot or so in a day, but in very large ice-flows, such as those of arctic regions, it sometimes exceeds fifty feet in twenty-four hours. The rate depends partly on temperature, partly on the inclina- tion of the ground, and partly on the volume of the 140 TESTIMONY OF THE GLACIAL FORMATIONS ice. Hence the motion is greater in summer than in winter, and during day than during night. Again, other things being equal, a glacier moves more rapidly on a steep than on a gentle slope ; while very thick ice-streams flow faster than glaciers of relatively insignificant dimensions. The glaciers to which I have been referring are typically represented by those of the Alps- — the distinguishing character of which is that they have their source in deep mountain-recesses whence they make their way down mountain valleys. But there are other types of ice-flow of great interest, especially from a geological point of view. These may be called ice-cap glaciers, of which excellent examples are seen in Norway. The configuration of that country differs essentially from that of the Alps. In the latter we have a complicated series of mountain- groups and crested ranges, everywhere overlooking the profound cirques and deep valleys in which glaciers are generated. Norway, on the other hand, is a lofty plateau, deeply trenched, it is true, by mountain valleys and fiords, but showing between these depressions extensive tracts with a relatively flat or undulating surface, which are often not dominated by peaks or ridges. Upon these plateaus snow accumulates, becomes granulated, and passes into ice, just as in the Alpine cirques. In short, lenticular ice-caps or sheets are formed, which attain their maximum thickness about the centre, whence they thin away as the outskirts of the plateaus are INLAND ICE OF GREENLAND 141 approached. The ice, it need hardly be said, is not passive but flows outwards in all directions. Here and there it projects longer or shorter tongues or lobes into valleys ; in other places, where no valleys are present, the ice-sheet may terminate in a con- tinuous ice- wall on the verge or brink of a fiord. The essential features of a Norwegian ice-cap are repeated, but on a much larger' scale, by the great "' inland-ice" of Greenland. There is good reason to believe that Greenland, like Norway, is a tableland. With the exception of a narrow rparginal strip the whole country is covered by a vast dome-shaped ice-sheet. Here and there, however, as the coastal tracts are approached, mountain-tops peer through the ice. But elsewhere no such " nunataks," as they are called, can be seen, and the ice may terminate with a steep wall-like front. A slow general movement of this enormous mer de glace takes place from the centre towards east and west. But the great tongues that protrude from it into the fiords (some of which they entirely fill) flow much more rapidly. One of these — the Humboldt Glacier — is forty-five miles wide where it enters Bafiin Bay. Little water is discharged from under- neath the wall-like margins of the inland ice ; but it is quite otherwise with the large effluent glaciers, from which rivers escape all the year round. The arctic ice-cap is surpassed in extent by that of the antarctic regions, but so far as the phenomena of chief significance to the geologist are concerned, 142 TESTIMONY OF THE GLACIAL FORMATIONS the southern ice-sheet is simply an enlarged replica of the inland ice of Greenland. There are certain differences, however. For instance, the antarctic inland ice covers the land almost completely, extend- ing generally into the sea. In Victoria Land, however, the appearances recall those of Greenland — the coastal tracts showing here and there mountainous nunataks, between which great effluent glaciers descend to the sea. In Kaiser Wilhelm Land, on the other hand, no marginal tract ijs left uncovered, the ice flowing out to sea as one continuous mer de glace, and terminating in a vertical wall. There are certain other features presented by the ice-world of the Antarctic (such as the enormous floating " ice- shelf") which are apparently peculiar to those regions, but as their particular significance to the geologist is not very clear, they need not be discussed. We may now briefly consider the nature of the geological work done by glaciers and ice-sheets. In the case of the ordinary alpine glacier, we are at once impressed by the appearance of the great heaps and parallel ridges of frost-riven debris" or moraines, as they are called, which line its surface. The origin of that debris is obvious. From its source in a cirque or cirques to its termination, an alpine glacier is overlooked by more or less precipitous heights, from which frost is continually detaching smaller and larger rock-masses, the production of debris being aided by the more or less frequent descent of avalanches. The margins of the glacier become in PLATE XVIII [2'o /ace page 14y. ORIGIN OF MORAINES 143 that way loaded with angular rock-rubble and large blocks and masses of all shapes, the whole forming what are known as lateral moraines. The long ridges and banks of debris that extend down the middle of a glacier are termed medial moraines, and these, it is hardly necessary to say, result from the union of two or more glaciers. When the glacier reaches its termination, all that superficial rock- rubbish is shot down in front of the ice, where it forms end- or terminal moraines. (See Plates XVIII- XX.) As a glacier flows with a differential motion, it is subject to stress and strain, to which it yields by cracking across ; so that deep clefts or crevasses come into existence : and the same results are brought about by the irregularities of surface over which the ice makes its way. Wide crevasses often open across a glacier, and when a superficial moraine reaches the brink of such a great gash it is abruptly truncated, and would seem to be finally engulfed, for the surface of the ice on the farther side of the crevasse may be entirely devoid of rock-rubbish. As the glacier flows on its way, however, the engulfed materials begin gradually to reappear at the surface, until by and by the moraine is reconstructed. Its reappearance is due partly to the ablation or superficial melting of the ice, and partly to the curving upwards of the lines or planes of flow. While superficial moraines are characteristic of alpine glaciers, they never appear upon ice-caps that 144 TESTIMONY OF THE GLACIAL FORMATIONS are not overlooked by mountains. Now and again an effluent glacier proceeding from, a Norwegian ice- cap may have rock-debris showered upon it as it descends its mountain valley, but its course is usually short and the rate of flow comparatively rapid, so that conspicuous lateral moraines cannot be accumulated. The same phenomena characterise the glaciers of Greenland. The great inland-ice from which they come is entirely devoid of superficial debris, but the effluent glaciers when they enter mountain valleys and fiords tend, of course, to become sprinkled with debris along their margins. There is yet another kind of moraine of which some account must be given. This is known as bottom- ot ground-moraine, from the fact that it occurs underneath glaciers and ice-caps. The moraine in question varies much in character, but consists essentially of boulders and smaller stones of every shape and size, commingled with grit, sand, and clay. There are certain peculiarities that serve at once to distinguish it from superficial morainic matter. While the rock-fragments in the latter are invariably sharply angular and unworn, those in bottom-moraine are blunted, rubbed, and worn, and frequently scratched or striated, smoothed, and even polished. Stones and boulders of this kind, along with grit, sand, and clay, often crowd the basal portion of an alpine glacier, which then resembles a breccia — the ice acting as a binding material. Whence has this rock- rubbish been derived ? Did it originally lie on the PLATE XIX The Aletsch Glacier, showing Medial Moraine. Photo : Photoohrom Co. [To face page lU. ORIGIN OF MORAINES 145 surface of the glacier, and subsequently find its way through crevasses to the bottom ? At first sight that seems a plausible explanation of its origin. Surface- moraines, as we have seen, are often engulfed in crevasses ; but the debris thus swallowed up is sooner or later disgorged, and reappears at the surface further down the valley. If crevasses penetrated the whole thickness of a glacier, debris plunging into them might readily reach the rocky bed and become enclosed in the basal portion of the ice. It may be doubted, however, whether crevasses ever extend to the bottom of an alpine glacier, except near the sides where the ice is relatively thin. In the case of moderately sized glaciers the plasticity of ice must at some depth from the surface cause the fractures to heal up, so that unless a crevasse Ibe kept open and deepened by ablation it probably never actually reaches the rock-head. The introduction of super- ficial moraines In this way to the bottom of a glacier must therefore be exceptional, and cannot wholly explain the origin of ground-moraines. This con- clusion is confirmed by the fact that summit-glaciers, which are often so situated that no rock-debris can be showered upon them, nevertheless extrude more or less conspicuous bottom-moraines. And it is the same with ice-sheets, large and small alike — these always disclose moraines of this kind, although the whole surface of the ice above mg.y be totally devoid of rock-debris. In such cases it is self-evident that the ground- or bottom-moraine has been derived K 146 TESTIMONY OF THE GLACIAL FORMATIONS directly from the rocky pavement underneath the ice. The great inland ice of Greenland, it is well known, develops enormous sheets and masses of ground- moraine. The lower strata of that* vast mer de glace, for a thickness of fifty or seventy feet above its base, often contains layers and irregular sheets of clay, mud, sand, stones, and boulders, all of infraglacial origin, while the upper and much thicker mass of ice shows no such inclusions. When the materials enclosed in the ice below attain a certain mass, the friction caused by their presence impedes the move- ment of the ice, and deposition or accumulation of bottom-moraine then takes place. The formation of ground-moraine assures us, therefore, that glaciers are most effective eroding agents. The rock-head over which they advance is ruptured or fractured, ground dowh, and smoothed. The included stones are not only themselves rubbed and striated, but the surface over which they are dragged becomes similarly modified. Hence, rough excrescences are gradually reduced and rounded off — their surfaces displaying long ruts and grooves where they have been scored by the travelling stones and blocks, and finer striae and polishing where they have been scratched by hard gritty particles, and rubbed and smoothed by the /abrading sand and clay of the bottom-moraine. Such grooving, striation, and polishing appear most conspicuously on the side of prominent or projecting rocks facing the direction of ice-flow. But when the glacier-pavement shows no PLATE XX GLACIAL EROSION 147 abrupt protuberant rocks, the whole surface may be equally smoothed and striated. Notinfrequently, how- ever, the rock-head instead of being thus smoothed is jumbled and broken, and larger and smaller blocks are wrenched out of place. Reefs and slabs of every shape and size thus dislodged are dragged away, while now and again bedded rock-mksses are abruptly bent over in the direction of thet flow, dislocated, crushed, and shattered. The rivers that escape from glaciers are invariably turbid. This is another proof of the fact that the grinding of rocks is constantly going on under moving ice. For the mud that renders these rivers turbid is simply fine rock-flour or powder. Many estimates of the amount of this finely reduced rock-matter have been made. It has been calculated, for instance, that the river coming from the A4r glacier carries away daily 280 tons of mud in Suspension. The Justedal glacier in Norway, draining an ice-field 820 square miles in extent, discharges in a summer day 1968 tons of sediment, while from the Vatna- jokull (Iceland) which drains an ice-field ten times larger than that of the Justedal, the river discharges annually nearly fifteen million tons. A glacial river strews its valley with enormous quantities of gravel, derived partly from its ground- moraines and partly from its terminal moraines. The gravels are coarsest, as one might expect, in the upper reaches of the valley ; lower down the stones are smaller, while sand and eventually mud 148 TESTIMONY OF THE GLACIAL FORMATIONS become better developed as the distance from the glacier increases, and the valley widens and flattens out. It is worth noting, however, that when high floods occur, considerable deposition of fine sediment may take place over the whole floor of a broad valley. In the case of the Alps, the large lakes act as settling reservoirs, so that most of the rock-flour produced by glacier-grinding is intercepted in close proximity to the mountains. But during the Ice Age, as we shall learn, the sites of all those lakes were occupied by enormous glacifers, the sediments derived from which were swept away from the mountain-region and distributed over widely flooded tracts in the low grounds beyond. There are many other phenomena connected with glaciers and glacial action to which* I have made no reference. Our attention has been* confined, in fact, to those which enable us to realise the conditions that obtained in Europe when certain Pleistocene formations were being accumulated. From that limited point of view, the most important conclusion to be drawn from the study of modern glacial action is that frost and glacier-ice are potent geological workers — eroding and denuding, and necessarily producing much waste material in the process. No one has ever doubted the vigorous activity of frost — its effects are too patent in the mountains of temperate latitudes and at all levels in circumpolar regions. But it is otherwise with the action of glaciers. Everyone admits of course that glaciers GLACIAL EROSIO> 149 carry vast quantities of rock-debris upon their backs. But to what extent do they erode the valleys in which they lie? Unfortunately we cannot see what is going on underneath glaciers and ice-sheets. Only here and there, along the side of a glacier or occasionally within its terminal ice-cave, can we examine its pavement and note how it works'. It is therefore just in places where the work of erosion must be at a minimum and deposition at a maximum that direct observations are possible. The snout of a glacier not infrequently may be seen resting upon its terminal moraine, and this has led some observers to infer that glaciers cannot be very effective eroding agents. But if our study of the action of rivers were similarly restricted to their plain-tracks where denudation is at a standstill, what should we know of river-erosion ? It is only when we have learned that the alluvia in the plain-track of a river often attain a very great thickness and occupy a most extensive area, that we begin to acquire some notion of the potency of fluviatile erosion. The reflection that all these vast accumulations of sediment — these far stretching plains and deltas — consist of river-alluvia, soon convinces us that, if little or no denudation takes place in the plain-track, great erosion must be effected further up the valley. For it is manifest that deltas only grow at the expense of the drainage-areas from which their materials are derived. If that be self-evident in the case of rivers, it is not less obvious in the case of glaciers, and jce-sheets. As a river-valley begins 150 TESTIMONY OF THE GLACIAL FORMATIONS to flatten out the erosive power of the river gradually decreases, while at the same time deposition of sediment progressively augments. It is the same with glaciers and ice-sheets — as these approach their termination erosion diminishes while accumulation increases. When that terminatio,n is reached their direct action of course ceases, and the further trans- portation of rock-material is given .over, as it were, to the rivers that escape from them. Hitherto we have been considering such glacial action as can be studied at the present day, and have come to the justifiable conclusion that the ability of glaciers and ice-sheets to erode is as well established as that of streams and rivers. But our knowledge of what moving ice can do has been greatly increased by a study of the deserted beds of ancient glaciers. We ourselves live upon the abandoned floor of a mighty ice-sheet which has left iadelible evidence of its former presence. The great valleys of the Alps, as we shall learn, were at one time occupied by glaciers, compared with which those of our time are the merest pygmies. Throughout the whole of northern Europe, indeed, and In nearly all the mountains of the central and southern parts of the Continent we meet with similar evidences of intense glacial action. And from an examination of the phenomena thus so abundantly displayed, we gain some notion of what must be taking place underneath existing glaciers and ice-sheets. My next two lectures, ^therefore, will be devoted to that subject, but I am GLACIAL LAKES 151 tempted on the present occasion to make brief reference to certain particular features of the glaciation which, ahhough only indirectly concerned with my main thesis, are nevertheless of very considerable interest. One of the most striking features of formerly glaciated lands is the profusion of lakes, both large and small. Consider for a moment the numerous lakes of our own country, of Scandinavia, of Finland, and the Alpine lands, and contrast these regions with lands that have not been glaciated. Were we to colour upon a map of Europe all the countries which have passed through glacial conditions, we should find that outside of the coloured area only a very small number of lakes would be left. Some of these, we should find, occupy the craters of extinct volcanoes. Others are barrier-lakes of one kind or another, caused by the damming-up of streams by landslips or rock-falls. Yet others are due to the action of underground waters, which bring up in solution immense quantities of mineral matter from below, and thus induce sinkings of the surface of less or greater extent. But the multitudinous lakes of glaciated lands cannot owe their origin to any such causes. With very few exceptions, they are the result of glacial erosion, and the irregular distribution of glacial and fluvio-glacial formations. Very many occupy hollows excavated by ice, others are ponded behind dams of detritus, while, in the case of a considerable number, the basins are due partly to erosion and partly to accumulation. 152 TESTIMONY OF THE GLACIAL FORMATIONS The first geologist to call attention to the origin of the lakes so common in glaciated lands, was Sir A. C. Ramsay, who maintained that the rock-basins occupied by them had been ground out by ice. In the case of a glacier, as we have seen, erosion is carried on throughout the whole extent of its bed. Obviously, however, the process of grinding and rupturing must be unequally developed — all parts of the rock-floor cannot be worn down at the same rate. Erosion we should expect to be most vigorously carried on where the ice is thickest, and to be less effective where the ice is thinnest. But the erosion, however great it may be, will not necfessarily result in the formation of a rock-basin. For instance, a glacier that descends and terminates on a steeply-inclined surface may be a most effective eroding agent, but it cannot grind out an elongated basin. It is only when it reaches and flows for a longer or shorter distance upon a gently-inclined or relatively flat surface that the conditions become favourable for that kind of formation. After a glacier emerges from the steeper part of its course — from the " torrent-track " of the valley — and deploys upon the less inclined track beyond, the ice, owing to the greatly diminished gradient, heaps up, as it were, and by and by attains its maximum thickness. As it flows on its way, the glacier eventually begins to lose in thickness, and the rate of movement progressively decreases with the gradual attenuation of the ice. Now although "we cannot observe what goes on underneath a glacier, GLACIAL LAKE$ 153 we nevertheless know that erosion does take place, and that it must be greatest where the ice is thickest. We infer, therefore, that in time the glacier must grind out a long trough in the bottom of the valley, the depth and extent of which will be proportionate to the size of the glacier, as also to its duration, and no doubt to the hardness and toughness of the rocks over which it flows. Towards its terminal front erosion ceases and accumulation begins — morainic matter being dumped upon and distributed over the valley-bottom, the surface of which may thus be considerably raised. When the glacier finally dis- appears, its bed will be marked by the presence of a hollow or depression, due partly to erosion and partly to the accumulation of detritus at its lower end. Needless to say, this basin will be occupied by a lake from which a river will be discharged. In course of time the river may cut its way down through the detritus at the end of the lake and thus lower the level of the latter. When it reaches the solid rock, however, the lowering of the level will not be so readily effected. It is conceivable, indeed, that the lake might be entirely silted up long before the outgoing stream could succeed in. draining it. Many of these glacial lakes attain an astonishing depth, and one is not surprised, therefore, that Ramsay's theory of their origin should have com- mended itself at first to only a few geologists. It was hard to believe that basins, reaching a depth in some cases of several hundred feet, could possibly 154 TESTIMONY OF THE GLACIAL FORMATIONS have been slowly excavated by glacial action. But the basins in question are not abrupt and sudden depressions. When we take into consideration their extent, it becomes obvious that they are relatively shallow pans or troughs. The descent from the lower or upper end of such a basin to its deepest part is extremely gradual. Loch Lomond, for instance, is 630 feet deep, but its length is one hundred and seventy-six times greater than its depth. Loch Ness, again, is one hundred and thirty-six times longer than it is deep ; and the great lakes of the Alps are all in like case — their depth is only a small fraction of their length. Could the water be removed from any one of these depressions, we should hardly be able to recognise its basin-shape without careful measurement. The rock-basins of our mountain valleys are not the only hollows which must be attributed to glacial action. In highly glaciated lands like Scotland, we frequently encounter the phenomena known as " crag- and-tail," which are well illustrated by the Castle Rock of this city. Here we have an abrupt rock- mass looking westward, with the long ridge of the High Street sloping gently in the opposite direction. A horseshoe-shaped depression passes round the west base of the Rock, and is prolonged through the Princes Street Gardens on the north side and the Grassmarket on the south side. This hollow is the result of glacial scour at a time when our district was overflowed from west to east by a mer de glace CRAG-AND-TAIL 155 some 2000 or 3000 feet in thickness, which drowned the Castle Rock, Calton Hill, and Arthur Seat, just as a river covers the boulders in its bed. The bottom part of the ice-flow, stemmed by the Castle Rock, was compelled by the current constantly advancing from the west, to exert greater erosive power at the base of the obstruction, and hence in .time hollowed out a depression, while behind the obstruction — that is, on the High Street ridge — morainic debris was accumulated. Very much the same action may be observed taking place on the bed of a river immediately in front of a large boulder. The water heaped against the boulder is forced downwards, and thus sweeps out a hollow in front, while the sand and gravel removed by the process tend to come to rest in the rear. The Castle Rock is only a small example of the phenomena of crag- and-tail. In my next lecture I shall point out that deflection-basins on a very much larger scale came into existence elsewhere under the same great mer de glace. Our conception of glacier ice as an eroding agent has thus been greatly extended by a study of the effects produced by the vast ice-flows of the Glacial period. Students of glacial geology have learned much, however, since Ramsay advanced his famous theory. While the glacial excavation of rock-basins is no longer considered improbable, there are many observers who go considerably further in their appreciation of glacial erosion, and maintain that it 156 TESTIMONY OF THE GLACIAL FORMATIONS alone can account for the great depths attained by the main trunk valleys of the Alps. These, it is well known, have frequently been deepened to a much greater extent than their tributaries. In the long established river-systems of regions which have never been glaciated, we usually find that lateral or tributary valleys gradually widen and flatten out as they approach the main valley, Were such a valley- system to be so far submerged as to allow the sea to occupy the bottom of the main or trunk valley, it is obvious that inlets of less or greater importance would at the same time invade the lower reaches of the lateral valleys. In a highly glaciated mountain area like the Alps, however, the relation of the tributary valleys to a trunk valley is often very different. Take, for instance, the valley of the Rhone between Visp and Martigny. Here we have a broad and deep flat-bottomed valley with more or less precipitous sides — a gigantic trench, in fact — that extends for many miles into the heart of the mountain land. Numerous tributary streams, join the Rhone, but only two of these — the Drance and the Visp — occupy valleys that open widely upon the main valley. All the others shoot down through a succession of abrupt, narrow, and profound gorges. Follow one of these upwards to a height of 800 or 1000 feet, and we find ourselves spmewhat suddenly in a relatively wide and open valley. Obviously the latter has been truncated or cut across by the great trench of the Rhone. (See Fig. 7.) Fig. 7. — Diagrammatic Plan and Sections of an Over-deepened Alpine Trunk-valley, with its Tributary Valleys. O, Over-deepened valley ; V, V, relics of preglacial valley ; A, tributary valley draining extensive ice-field ; B, tributary valley draining less extensive ice-field, abruptly truncated by main valley. OVER-DEEPENING OF f ALLEYS 159 There is good evidence to show that before glacial times this great disparity of depth between the valley of the Rhone and the high-level valleys from which its tributaries come did not exist. At heights of 800 feet to nearly 1000 feet above the Rhone we encounter terraces and flat embayments in the mountains, which we have no difficulty in recognising as remnants of its old preglacial valley. Not infrequently these terraces can be followed into the high-level tributary valleys, with which indeed they are continuous. In preglacial times, therefore, the river-system of the Rhone was normal — the tributary valleys widening and flattening out as they opened into the main valley. Had nothing occurred to interfere with the ordinary action of the rivers, the relation between the Rhone and its affluents would doubtless have remained normal. But the main valley has been over-deepened to the extent of 800 to 1000 feet or thereabout, and that over- deepening could not have been the work of the Rhone, otherwise the tributary valleys would have been deepened pari passu. The trenching of the Rhone valley and the truncation of the valleys that formerly opened upon it in the usual way, must be attributed to glacial action. The main valley being one of the chief lines of drainage was at one time the bed of the most gigantic of the old Swiss glaciers, the thickness of which between the two points Visp and Martigny exceeded 4000 feet. That enormous ice- river, like the great glaciers of Greenland, may have flowed rather rapidly, and if so, must have been a 160 TESTIMONY OF THE GLACIAL FORMATIONS correspondingly powerful eroding agent. Its numerous tributaries have also scoured theirvalleys, but owing to their smaller size, and to the fact that they were to some extent dammed back by the enormous Rhone glacier, they could not be equally effective. It was only in such valleys as those of the Visp and the Drance, which drained very extensive ice-fields, that erosion comparable to that of the Rhone valley could be accomplished. The over-deepening of valleys by glacial action is conspicuous in many other places throughout the Alps — one of the most instructive examples being that of the Lauterbrunnental (see Plate X'XI), where the old preglacial valley forms the well-mdrked lofty terraces of Wengen and Miirren. In the bottom of that valley has been cut the steep-sided (J -shaped trench through which the Weisse Liitschine now flows. We have here, 4n short, a valley within a valley. (See Fig. 8.) That the younger valley has not been excaivated by the river, is shown by the fact that the tributary brooks and streams have not yet succeeded in cutting out courses equal in depth to that occupied by the Liitschine. The relation of these tributaries to the river is abnormal. Instead of flowing in hollows that widen and flatten out until they open upon the main valley, many of them (such as the well-known Staubbach) cascade over the precipices that bound the latter, while others descend through narrow and steep gorges in a succession of waterfalls and tumultuous rapids. As the bounding walls of the younger valley rise to a PLATE XXI CQ m [To face page 160. OVER-DEEPENING OF VALLEYS 163 height of looo feet or thereabout,* they enable us to realise what is meant by the "over-deepening" of valleys by glacial erosion. In this account of the work done by glaciers I may seem to have wandered far from my main subject — the geological evidences of the antiquity of man. But, as I hope to show you in subsequent lectures, that is not the case. The history of the Ice Age is to a large extent the history of the Pleistocene period which witnessed the apparition of Palaeolithic man in Europe. The excavation of rock-basins, and the over-deepening of great mouniain-valleys by hundreds of feet, must of necessity have occupied a very long time. And as these striking modifications of the land were accomplished in whole or in part while man occupied our Continent, they serve to increase our conception of the great antiquity of our race. (See Note 9.) LECTURE VI THE TESTIMONY OF THE GLACIAL FORMATIONS — continued II. — Glaciation of Northern Europe The Glaciation of Northern Europe. Extent of Scottish Ice and Directions of Ice-flow. Submarine Rock-basins. Southern Ter- mination of Ice-sheet in England. Glacial Excavation on the Floor of the Irish Sea. Rock-rubble in Non-glaciated Areas of England. Scandinavian Mer de Glace. Thickness of Ice-sheet that invaded North Germany. Over-deepening of Norwegian Fiords, and Results of Glacial Action in Germany. Contemporaneous Conditions in the Atlantic. In no part of Europe can the evidence of former ice- action be better studied than in pur own country. For this reason I shall describe the general phenomena exhibited by Scotland in some little detail, since they may be taken as typical of all the other glaciated regions of northern Europe. The most notable of our glacial formations is undoubtedly the till or boulderTclay. This is a more or less gritty clay, usually crdwded with stones and blocks of all shapes and sizes. These are not arranged in layers, but confusedly scattered through the clay, in which we look in vaih for any trace of 161 TILL OR BOULDER CLAY 165 what geologists call stratification. The accumulation, in short, shows no evidence of having been gradually deposited in water and spread out in beds. The stones are subangular, blunted, often smoothed and striated, and resemble in every particular the similar stones and boulders occurring in the bottom-moraines of existing glaciers and ice-sheets; The clay, again, is quite unlike any alluvial deposit. Alluvial clay consists of materials which have been long exposed to the chemical action of rain and superficial water ; while the clay of our till is wholly unweathered. Obviously it has been accumulated under conditions which excluded the chemical action that affects rocks of every kind exposed at the surface of the ground. In all these respects it is precisely similar to the clay of a true bottom-moraine. Usually it is very tough and much compressed, often showing lines or planes of shearing, and now and again presenting the appearance of a kind of rude bedding, as if it had been gradually accumulated. But this rude bedding can never be mistaken for thaj of an aqueous formation. One may note further that the character of the till is somewhat local. That is to say, a large number of the stones and boulders in any given mass have Come no great distance, but belong: for the most part to the drainage-area in which the deposit occurs. Closely associated, however, with these stones of more or less local origin we detect others which have travelled much farther. The till of our own district, 166 TESTIMONY OF THE GLACIAL FORMATIONS for example, consists very largely of materials derived from the Carboniferous strata which extend for many miles to the west of Edinburgh. But among this local material we note not a few stones and boulders that have wandered all the way from the Highlands. Far-travelled erratics are usually the most worn and rounded, while not infrequently fragments which have come but a short distance are quite rough and even sharply angular. In some places, indeed, the bottom part of the till is made up chiefly of large masses and shattered debris of the immediately underlying rocks. In other places, again, the outcrops of the beds below the till have been forcibly bent' over or dragged forward in some particular direction. Water- worn sediments are now and again enclosed in the till. These may consist of mere lines and layers of gravel, sand, and fine clay, or of relatively thick beds. Such inclusions are frequently more or less crumpled and contorted, and curiously involved with the till, and it may be noted that precisely similar inclusions of water-formed beds often accompany the bottom-moraines of the Alps and Arctic lands. A few words now as to the distribution of the till in Scotland. In the Highlands it is largely confined to wide flat-bottomed valleys and gently-inclined hill- slopes. It is rarely seen in narrow glens, where a few patches, however, may sometimes be found nest- ling in the rear of prominent rocks that look up the glens. In the Southern Uplands it has a similar distribution, being best developed in the broader ORIGIN OF BOULDER-CLAY 167 valleys. As the hills of that region, however, have on the whole a softer outline and th,e valley-slopes are less steeply inclined, the till frequently puts in a strong appearance and covers wide areas. The bottoms of many of the valleys, for example, are continuously paved with it, and when such is the case it has been trenched by streams and rivers, and thus flanks these in bluffs and scauffS. It reaches its greatest development, however, in the Lowlands, \yhere it sometimes attains a thickness of loo feet or more, and often extends over wide, regions. Usually it presents a broadly undulating surface, but in certain districts, as in Teviotdale, Tweeddale, and Nithsdale, it is drawn out, as it were, in a series of parallel banks and hollows, which coincide in direction with that of the broad valleys in which they occur. Followed from the Lowlands to the mountains it tends to become stonier and coarser, often passing into a rough morainic earth. These facts thus shortly stated suffice to prove that till or boulder-clay is of the nature of a bottom- moraine, and the trend of the ice-flow to which it owes its origin is indicated by the various features I have described. Its true nature, for instance, is shown by the directions in which the included stones and boulders have travelled, by the alignment of the parallel banks or "drumlins," as they are called, and by the phenomena of crag-and-tail. To these proofs of glaciation we add the testimony furnished by the striated and smoothed rock-surfaces of Scotland, such 168 TESTIMONY OF THE GLACIAL FORMATIONS as are so often exposed when the till has been stripped away. It need only be said these exactly recall the glaciated rocks of the Alps and Greenland. But strise and roches moutonndes are met with abundantly elsewhere than underneath boulder-clay. Wherever bare rock appears at the surface, either in Lowlands, Uplands, or Highlands, it may exhibit traces of glaciation. The striae no doubt have frequently dis- appeared, but the rounded and flowing contour characteristic of glacial work often remains. So long a time has elapsed since the Glacial period, however, that frost, rain, and superficial action generally have certainly removed much of the evidence. Neverthe- less, that evidence is still so abundant that geologists have no difficulty in making out the direction of ice- flow, not only on the mainland but in the adjacent islands. Rock-strise and the characteristic glaciated contour have been traced all over the land, from the sea-level up to heights of 3000 and 3500 feet. It would seem, therefore, that the whole of Scotland, with the exception of the tops of our highest moun- tains, has been smothered in ice. So copious and detailed are the proofs, that we can even estimate the approximate thickness attained by the Scottish ice-sheet, and determine the slope of its upper surface as it flowed out into the Atlantic. In Wester Ross, for instance, glaciation may be traced up to a height of 3500 feet, while in the Outer Hebrides it does not go higher than 1600 feet. Now the ice-sheet which advanced from Ross in a general THE SCOTTISH MER DE GLACE 169 north-west direction filled up the Minch and over- flowed the Outer Hebrides. The distance from the Torridon Mountains in Ross to the Cliseam in Harris ■being fifty-six miles, the slope of the mer de glace is readily obtained; and found not to have exceeded thirty-five feet per mile — a gentle inclination. In the central part of the Minch, off the ^st coast of Harris, the depth of the sea averages 250 feet approximately. Over that region the surface of the ice-sheet must have reached a height of 2550 feet above the present sea-level, and if we add to that the depth of the water ( = 250 feet), we get for the mer de glace a thickness of 2800 feet. Nearer the coast of the mainland the ice-sheet would, of course, be still thicker, probably more than 3500 feet. (See Note 10.) It will be readily understood that the configuration of the land — the disposition of high-lying and low- lying areas — must, as a rule, have controlled the direction of the Scottish mer de glace as it flowed outwards to the sea. In a general way, therefore, the ice would follow the trend of the principal valleys ; but as we trace its spoor down to the east coast we find that, instead of going straight out to sea, it has invariably turned aside. For instance, the ice that occupied Tweeddale flowed in the same general direc- tion as the present river until it approached the sea, when it gradually wheeled round to pursue a south- easterly course. The same change of direction took place in the depression now occupied by the Firth of Forth, the coast-lands as far as St Abbs having been 170 TESTIMONY OF THE GLACIAL FORMATIONS traversed from west to east, after which the ice-flow swung round to south-east. So again the coastal tracts of the Firth of Tay are striated in the general direction of the Firth, but beyond Dundee the rock- strise bend away towards north-east, and the same is the case, one may say, along the entire coast-line of Forfar, Kincardine, and Aberdeen. Obviously, the ice-flow descending from the Highlands towards the east coast must have encountered some insuperable obstruction, which compelled it to follow what might at first seem to be an abnormal course. A still more striking example is exhibited by the glacial phenomena of Caithness, along the east coast of which the striae are directed from south-east to north-west. This shows that the ice streaming from the Highlands into the Moray Firth was deflected from its path, and actually forced to overflow the north-east part of the county. And this invasion from the sea is further proved by the notable fact that the till of Caithness contains many broken and striated sea-shells. Nor is this the only evidence of ice-invasion from the North Sea, for both the Orkney and the Shetland Islands are glaciated from east to west. In my last lecture I gave some account of the phenomena known as crag-and-tail, and mentioned the fact that in front of the "crag" a hollow frequently appears. The same phenomena, but on a much larger scale, are of common occurrence on the sea- floor off the north-west coast of Scotland. The vur de glace coming from the mainland was so thick, DEFLECTION-BASINS ON SEA-FLOOR 171 as I have pointed out, that it succeeded in overflowing not a few of the islands of the Inner and Outer Hebrides. Many of these islands, however, attain a considerable height and must have peered like nunataks above the surface of the mer de glace. Nor can we doubt that to some extent they obstructed the ice-flow. The latter as it made for the Atlantic followed a north-westerly and westerly direction, and those islands that rose abruptly from the sea-floor to mountainous elevations would necessarily stem the current pressing against them, and thus induce excessive glacial erosion. We are not surprised, therefore, that in front of such stdep islands more or less well-marked depressions or basins should occur, some of them reaching a great depth. It is note- worthy that the profoundest depressions of the kind appear off" such islands as directly confronted the ice-flows proceeding from the larger sea-lochs. Take, for example, the case of the Island of Arran. That island stood in the path of the iee descending from the Argyllshire Highlands by Loch Fyne and the Firth of Clyde, and the charts: show that it is surrounded on north, west, and east by a great horseshoe-shaped depression, the maximum depth of which (444 feet) is attained, where it might have been expected to occur, off" the north coast. But the most notable example of the kind is that of the long series of basins in the Minch, extending from south-west to north-east in front of the Outer Hebrides — a direction at approximately right angles 172 TESTIMONY OF THE GLACIAL FORMATIONS to that of the ice-sheet by which those islands were so largely overwhelmed. It was only the upper strata of the mer de glace, however, that could traverse the islands. The latter rise so steeply from the bed of the Minch that the lower strata of the ice could not possibly surmount them, but were deflected to north-east and south-^est as an "under- tow." Not before the mountain-barrier declined in height into the low grounds in t"he north of Lewis was that undertow able gradually to invade the island, and to cover it with its bottom-moraine which, like that of Caithness, is charged with crushed and broken sea-shells, taken up doubtless from the floor of the Minch. The deep hollows ranged along the inner margin of the Outer Hebrides are similar in all respects to the depression that so nearly surrounds Arran — they are "deflection-basins," due to glacial erosion, and ranging in depth from 400 feet to 800 feet. "We cannot, of course, tell whether those basins are wholly excavated in rock, or whether they may not owe some of their depth to unequal accum- ulation of glacial and marine deposits. But their form and disposition, and the whole configuration of the sea-floor so exactly recall the aspect of the ice-worn low grounds of the Outer Hebrides, the rocky coast-lands of north-west Scotland, and similar glaciated regions in other lands^ that we can hardly doubt the bottom of the Minch and adjacent areas owes its characteristic features to glaciation — that the deep troughs hugging the shores of the rocky GLACIATION OF ENGtAND 173 islands that face the mainland have been ground out by the great mer de glace on its passage into the Atlantic." The glacial phenomena of England tell much the same tale as the boulder-clay and rock-striae of Scotland ; but while Scotland was wholly drowned in its mer de glace, large areas in the sister country escaped glaciation. The Yorkshire Moors and a considerable portion of the Pennine Chain, for example, were surrounded but not overwhelmed. The terminal front of the ice-sheet extended south to the valley of the Thames at and near London, whence it seems to have turned away towards north- west into the Midlands, and thereafter swung round by the valleys of the Avon and the Severn to the Bristol Channel. Although the southern extremity of England escaped invasion by the ice-sheet, it could not fail to experience a severe climate. This is proved by the appearance in many places of great sheets of rock-rubble which have travelled down hill- slopes and overspread wide reaches of low ground. Excellent examples are furnished by the so-called " head " so frequently seen capping the sea-cliffs of Devon and Cornwall. This accumulation recalls the aspect of the "flowing soils and debris" of Arctic lands. It consists of a more or less coarse agglomeration of angular rock fragments of all shapes and sizes set in an earthy matrix. Here and there it has a rudely bedded appearance, but not such as would suggest 174 TESTIMONY OF THE GLACIAL FORMATIONS arrangement by running water. "Coombe-rock " is the name given to similar sheets of rock-rubble, which extend from the South Downs for some eight miles over the adjacent low grounds. These notable accumulations are made up, as might have been expected, of debris derived from the Downs, and consist, therefore, of "unstratified or obscurely stratified flints, battered but not rolled, and embedded in a matrix of chalky paste and pieces of chalk." Further south the deposits in question pass into an accumulation of flints in a loamy matrix, and eventually change into a brick-earth with scattered angular flints, from which have been obtained teeth of elephant and horse, broken and apparently decayed before they were embedded, and a few Palaeolithic implements. Rubble drifts of very much the pame character, it may be added, occur in the Channel Islands and the north of France. But, to return to the ice- sheet, I would remind you of the remarkable deflections of ice-flow which took place along the coastal tracts of the North Sea. The Scottish ice as it issued from Tweeddale gradu- ally turned away towards south-east, and the same deflection has left its marks in the maritime regions of Northumberland and Durham. South of the York- shire Moors, however, the direction of ice-flow changed from south-east to south-west throughout all East Anglia. The cause of such notable deflections was the presence of a great Scandinavian mer de glace in the basin of the North Sea. The ice streaming out EROSION ON FLOOR OF IRISH SEA 175 from Scotland and the north-east of England was suffi- ciently massive to protect those regions from invasion by the inland-ice of Scandinavia ; but it was otherwise in East Anglia, where the foreign mer de glace over- spread the land with its bottom-moraine, in which fragments of characteristic Scandinavian rocks occur in large numbers. Ireland, it need hardly be said, was, like Scotland, drowned in a mer de glace of its own, sufficiently massive to preserve the island from invasion by any alien ice-flow. The Scottish ice, however, closely approached the north-east coast of Ireland, against which it divided — one branch flowing north-west and the other south-east. A mighty mer de glace derived chiefly from Scotland, Ireland, and Wales, but partly also from the mountains of Cumberland and West- moreland, filled the basin of the Irish Sea. As this great confluent ice-flow passed south it was divided by the high grounds of Wales — -orte branch streaming south-east across the low grounds of Lancashire and Cheshire, to unite with the Scandinavian ice in the Midlands — the other and larger mass flowing through St George's Channel on its way ou:t to the Atlantic. The Scottish and Irish ice-flows, uniting on the bed of the North Channel, could not escape mutual constriction, and their erosive action was thus Correspondingly increased. The result of this is disclosed by the soundings on the Admiralty Charts, which indicate great depressions on the sea-floor just where we should have expected them to occur. These 176 TESTIMONY OF THE GLACIAL FORMATIONS basins are quite comparable, in short, to the deep hollows that fringe the inner margin of the Outer Hebrides. Were the British area to be elevated for 600 feet or thereabout, the North Channel and the Irish Sea would be converted into land ; but a great elongated lake would appear, extending for some 240 miles from the Scottish Highlands southwards to the region between Wales and Wicklow county in Ireland, with a maximum depth of nearly 600 feet. Turning now to the Continent, we meet with glacial phenomena the same in kiad as those of the British area, but developed on a far grander scale. An ice-sheet covered practically all northern Europe — the centre of dispersion having been the Scandinavian plateau. From that region the ice-flow, as we have learned, filled the North Sea and was confluent with the British mer de glace. It overspread the major portion of the Low Countries and north Germany, extending to the foot of the Harz, the Riesengebirge, the Erzgebirge, and the Sudetic ranges. It swept across Finland and occupied all norithern and central Russia, broad lobes advancing south to within 200 miles of the Black Sea and the Sea of Azov respec- tively. In the far north-east it was confluent with the ice that descended from the Urals into the valley of the Pechora. It thus extended from north to south for a distance of at least 1500 miles, and about 2500 miles from west to east. It is not hard to get some conception of the thickness attained by that endrmous ice-sheet. GLACIATION OF NORTHERN EUROPE 177 The average elevation of the inland-ice within the Scandinavian peninsula, from which it radiated, may be taken as 7000 feet above the present sea- level. From this central area of dispersion it flowed south and dropped its erratics and morainic debris upon the northern flanks of the Harz and other mountains of middle Europe, most 'of which supported snow-fields and glaciers of their own. On the Harz the blocks brought by the ice-sheet from the north occur up to a height of 1350 feet. Now the distance between the area of dispersion and the Harz is 530 jniles or thereabout, so that the surface of the mer de glace had the very gentle inclination of ten or eleven feet per mile. In southern Jutland, therefore, that surface must have been 2800 feet above the present level of the sea, and if we deduct from this the height of the land, which would be about 300 feet, we obtain a thickness in round numbers of 2500 feet for the ice that passed over Denmark. Again, we get a distance of about 700 miles, when we measure from the site of the Scandinavian ice-shed to the lower slopes of the Riesengebirge, against which the ice-sheet terminated at a height of 1300 feet above sea-level. Along that line, therefore, the surface-slope of the ice could have been little more than eight feet per jnile. This would give a thickness of 2900 feet for the sheet in the south of Sweden, and of 1300 feet or thereabout for the ice in the neighbourhood of Berlin. A mer de glace of that extent and thickness could not fail to do much work. Scandinavia and Finland M 178 TESTIMONY OP THE GLACIAL FORMATIONS were severely scoured and abraded — 'they constituted the area of dominant erosion, while the low-lying plains of Denmark, Holland, Prus§ia, and central and northern Russia formed the area of dominant accum- ulation. Throughout this latter region, therefore, the glacial formations attain so great a thickness that the underlying rocks are very seldom exposed. The track of the ice-flow is thus traced chiefly by means of the erratics it carried, but now and again, as in Prussia and Saxony, their testimony is confirmed by that of rock-strise and roches moutonndes. In the Scandinavian peninsula, on the other hand, the path of the ice-sheet is clearly indicated by the strongly glaciated rock- surfaces and the evidence everywhere of extreme erosion. Nowhere, perhaps, is this so apparent as in the great fiords of Norway. " Norway, as everyone knows, is an ancient plateau, deeply incised and cut up, as it were, into irregular segments. These segments vary much in extent and form — sometimes the surface of the fjeld is flat and undulating, elsewhere! it is scarped and worn into irregular groups and masses of variously- shaped mountains and ridges without any determinate arrangement. The orography is everywhere in strong contrast to that of the Alps with their extended parallel chains and longitudinal galleys. " Not less strong is the contrast between the fiord-valleys of Norway and the V-shaped valleys of the Alpine Chain —the fiord-valleys being U-shaped and resembling in this respect the over-deepened trunk-valleys of the Alps. OVER-DEEPENING OF FIORDS 179 The principal channels of erosion in Norway are deep and trench-like valleys, the tributaries of which are relatively insignificant. The aiain stream, flowing through a deep mountain valley, eventually enters the sea at the head of a fiord. Below this point, however, few or no side-valleys as a rule break the continuity of the fiord-walls. Numerous tributary waters, some of which are hardly less important than the head-stream, do indeed pour into the fiord, but they have not yet eroded for themselves deep trenches. After winding through the plateau-land in broad and shallow valleys their relatively gentle course is suddenly interrupted, and they at once cascade over the precipitous rock-walls to the sea. The side-valleys that open upon a fiord are thus truncated by the steep mountain-wall as abruptly as if, to use the words of the late Dr Richter, they had been cut across with a knife. Mountain valleys of the V-shaped alpine type are not wanting in the fjeld, but as they are followed inland they soon lose that character and acquire softer features. The valleys of the fjeld lands are for the most part broad and open, many lakes being strung along the courses of the streams. We are here dealing in fact with a plateau lake-land, a region in which glacial erosion has been in excess of accumulation. It is through this gently-undulating, highly ice-worn plateau-land, with its shallow valleys, that the profound chasm-like fiord-valleys have been cut to depths of 3000 to 6500 feet. 180 TESTIMONY OF THE GLACIAL FORMATIONS A fiord is simply a partially drowned land-valley — the work of erosion of one kind or another. But why should the formation of a main or fiord-valley be so immeasurably in advance of thaC of its tributaries? Obviously there must have been a time when the process of valley-deepening proceeded more rapidly along the lines of the present fiords and their head- valleys than in the side-valleys which open upon these from the fjelds. Valley-erosion by river-action could not have been carried on equally throughout the land, otherwise a completely developed or normal hydro- graphic system would have beeil the result. We cannot doubt that the present valley - system of Norway must have originated in the usual way by river-erosion in ages long prior to. the Glacial period. Its present abnormal character, however, cannot be attributed to the action of running water alone, but must be assigned largely to that of the great glaciers of the Ice Age. The fiord-valleys and fiords were the chief lines along which the Scanttinavian inland ice made its way. The fjelds were nq. doubt at the same time scoured and abraded, but the ice attained a much greater thickness, and doubtless flowed most rapidly in the main valleys. The conditions were essentially the same as in the Alps — over-deepening took place in the great trunk-valleys. The fiords with their precipitous rock-walls are comparable to the deep trench of the Rhone between Visp and Martigny, while the truncated side-valleys of the fjelds correspond to the similarly truncated tributary valleys of the DISTURBANCES DUE TO GLACIATION 181 Rhone. The narrow gorges, with their numerous waterfalls and cascades that plunge into the over- deepened valley of the Rhone and pther great Alpine valleys, find their counterpart in the abundant fosses, that shoot from the Norwegian fjelds into the fiords. If Swiss glaciers succeeded in deepening their valleys to the extent of looo feet or morfe, we need feel no surprise that the enormous ice-flaws descending the trunk-valleys of Norway should have produced similar effects upon even a larger scale. It goes without saying that the excessive erosion — the deepening of a fiord by many hundreds of feet — must have been a most protracted process, extending, over a long period of time. As additional evidence of the powerful action of the great ice-flows of the Glacial period, mention rhay be made of the extensive contortion, dislocation, and displacement of rocks which now and again took place under the Scandinavian inland ice. The disturbances referred to are on so large a scale that they have often been attributed to subterranean action. Good examples of the phenomena are well exposed in the cliffs of certain islands in the Baltic (Moen and Rugen), where the Chalk, several hundred feet in thickness, has been ruptured, overturned, and displaced so as now and again to overlie and to enclose the bottom-moraine. Again, in Mecklenburg- Schwerin, where the ground is thickly covered with glacial deposits, great masses of Chalk here and there protrude as hills above the surface, just as 182 TESTIMONY OF THE GLACIAL FORMATIONS if they were in situ, but they are liow known to be merely huge erratics. On a still larger scale are the rock-displacements seen in Saxony, where the Tertiary Brown Coal formation is often confusedly mixed up with bottom-moraine—wide stretches of strata having been forced out of place so as to appear intercalated in the boulder-clay, as if they formed part of one and the same series, in which position the Tertiary coal-beds hi,ave actually been mined. No doubt these are somewhat exceptional cases, for the Chalk and relatively unconsolidated Tertiary strata would yield to the intense pressure exerted by an advancing ice-sheet more readily than denser or harder and more resistant strata. Even the latter, however, have been bent over in the direction of ice-flow, and often fractured and forced out of place. The glacial deposits covering the peripheral area of maximum accumulation in Denmark, Holland, northern Germany, etc., consist very largely of rock-debris derived from the Scandinavian peninsula and Finland — the areas of maximum erosion. They attain a great thickness, and some attempts have been made to form a rough estimate of the amount of material transferred from north to south. Accord- ing to Dr Helland, if the abundant erratic materials which at present cover most extensive tracts in Russia and north Germany were . to be transferred to Finland and Sweden from which they originally came, they would raise the general surface of those FLUVIO-GLACIAL DEPOSITS 183 countries by 255 feet. It ought to be stated, how- ever, that all these materials were not the product of one ice-sheet. As we shall afterwards learn, there have been several glacial invasions of the low grounds of northern Europe, each of which had its share in the formation of the morainic accumulations of those regions. The terminal front of the great Scandinavian ice-sheet was margined by a zone of less or greater breadth, over which gravel and sand were distributed by water coming from the surface and from under- neath the ice. Where the latter abutted upon rnountain-slopes, this zone of gravel and sand was of limited width ; but where it terminated upon low- lying ground and the water could readily escape, aqueous deposits overspread correspondingly broader areas. Sometimes the gravel and ^and are heaped together so as to form multitudinous hills and hillocks, dotted with large erratics, but more usually they appear as undulating or flat sheets often of vast extent. That much loamy material or rock- flour — the result of glacial grinding — was washed out from the ice we can readily believe. Some of this would doubtless come to rest in the abundant temporary lakes of the low grounds, but much must have been carried by the flooded rivers out to sea. To this subject, however, I shall return in my next lecture, after I have discussed the glacial phenomena of the Alpine lands. Although extensive areas of middle Europe must 184 TESTIMONY OF THE GLACIAL FORMATIONS have been inundated during the Glacial period, especially in summer, nevertheless much of the relatively low-lying land escaped, gind was clothed with an arctic and sub-arctic flora, and roamed over by a tundra fauna. But the conditions of the land at this time will be better understood after we have considered the glacial phenomena of other parts of Europe. Hitherto I have been referring chiefly to the general aspect of the great inland ice, and discussing the effects it produced upon the land, and we may now turn our attention for a little to the action of ice in the sea. The Scandinavian ice was so thick that it entirely filled up the North Sea, and compelled the British mer de glace out of what might have seemed to be its normal course. Many geologists, therefore, have supposed that glacial conditions may have co- incided in northern and north-western Europe with a general elevation of the land — that, in a word, the North Sea basin at the beginning of the Ice Age was dry land, and the British Islands formed part of the Continent. But there is really no evidence of any such general uplifting of the land in glacial times. Indeed, such evidence as we have points the other way about — for it proves that the North Sea certainly existed just before East Anglia was invaded by the Scandinavian ice, the approach of which was heralded, so to say, by floating ice, and by the presence in our seas of a well-marked arctic fauna. "The notion of a great elevation of the land appears to have been ICEBERGS OF THE GLACIAL PERIOD 185 suggested partly by the exigencies of the now dis- carded theory, that a general uplift was required to bring about glacial conditions ; knd partly by the belief that an ice-sheet could not have flowed from Scandinavia to Britain had the North Sea been in existence. But that sea is quite shallow, and could have formed no obstacle to the advance of an ice- sheet several thousand feet thick. It was simply dispossessed, just as in our day the polar seas are dispossessed for miles by the inland ice flowing out- wards from the Arctic and Antarctic areas. It is not difficult to picture to ourselves the con- ditions that obtained along the borders of the ice- covered tracts of north-west Europe during the Glacial period. Could anyone have approached our Continent from the west at that time he would have encountered, some fifty or sixty miles outside of the British Islands, a great wall of ice — the terminal front of the mer de glace, which may well have advanced to what is now the loo-fathoms Hne. When the increas- ing depth of the ocean prevented the further advance of the great sheet, the calving of icebergs would begin, and tabular masses of larger and smaller dimensions would forthwith sail away. Probably no Gulf-stream Drift washed the western shores of Europe at that time, the set of pcean currents and dominant winds alike being towards the south. How far south icebergs may have floated we do not know, but they certainly sailed much further away than is now the case, for some of them stranded and left their 186 TESTIMONY OF THE GEACIAL FORMATIONS erratics on the shores of the Azores, when these islands • ere slightly depressed below their present level. That a cold oceanic current passed along the west shores of Europe is further evidenced by certain Pleistocene deposits in Italy, whfch have yielded a large number of molluscan species that no longer live In the Mediterranean. Amongst these are many North Atlantic forms, and others that are exclusively boreal and even arctic — all of which must, of course, have entered the Mediterranean from the Atlantic. The occurrence of these northern types so far south 'of their present habitat is quite in keeping with the appearance of marmots and other, alpine animals in the low grounds of Italy, and of reindeer and mammoths in southern France and Spain. The* .Glacial period, in short, was marked by wholesale migrations of floras and faunas from north to south. LECTURE VII THE TESTIMONY OF THE GLACIAL FORMATIONS — continued III. — Glaciation of the Alps, etc. Glaciation of the Alps and other Mountains. The Old Glacier of the Rhone. Old Glaciers of the Rhine, the I Her, the Inn, &c. Glacial Phenomena of Upper Engadine. Old Glaciers on South Side of the Alps. Evidence of Glacial Erosion — Lake-basins ; Over- deepened Valleys ; Great Extent of Ground-moraines, and Fluvio- glacial Deposits. Glaciation of other Mountains in Middle and Southern Europe. Foehn Winds and the Origin of the Loess. Among the most conspicuous evidences of formerly widespread ice-action in the Alpine lands are the large travelled blocks so lavishly scattered everywhere. From an early period these attracted the attention of intelligent travellers, some of whom expressed surprise when they saw that the isolated blocks did not consist of the same kind of rock as that of the mountains on which they lay. The numerous gigantic erratics — some as big as cottages — that strew the flanks of the Jura looking towards the Alps, are often referred to by old writers as specially remarkable. Obviously these had been carried to 187 188 TESTIMONY OF THE GLACIAL FORMATIONS where they are now found — but by what mysterious agency? Some, probably all, had come from the Alps, and had therefore travelled many miles. Amongst other vague conjectures it had been suggested that the transporting agent might have been water — that the blocks might have been swept down from the Alps by powerful ddbicles or cataclysms. Such views, however, were never con- sidered quite satisfactory and convincing — a more reasonable explanation of the phenomenon had yet to be found. It is interesting to Scotsmen to know that the first to discover the solution of the problem — to divine the true origin of the erratics — was a fellow-countryman, John Playfair, formerly Professor of Natural Philosophy in the University of Edinburgh, and a devoted disciple of James Hutton, one of the founders of the present system of geology. During a visit to Switzerland in 1815, Playfair was much impressed by the great wandered blocks of the Jura. One of these, a mass of granite,, he estimated to weigh 2520 tons, and his sagacious reflections are Worth quoting, inasmuch as his recognition of the glacial origin of the erratics in question was in point of fact the first step taken to woifk out the history of the Ice Age. " When we consider," he remarks, "that the present point where the granite is to be found in its native place is at a distance of seventy miles, it will appear no easy matter to assign a conveyance by which this block could have performed such a journey over hills and valleys without consider- FOUNDERS OF GLACIAL GEOLOGY 189 able injury. A current of water, however powerful, could never have carried it up an acclivity, but would have deposited it in the first valley it came to, and would in a much less distance have rounded its angles, and given to it the shape so characteristic of stones subjected to the action of water. A glacier which fills up valleys in its course, and which conveys the rocks on its surface free from attrition, is the only agent we now see capable of transporting them to such a distance, without destroying that sharpness of the angles so distinctive of these masses." Some fifteen years, however, were to elapse before tlie study of the glacial formations of the Alps began to be seriously attacked by scientific men. It is chiefly to Charpentier and L. Agassiz that we owe the first clear outhne of the several phenomena which unite to prove beyond any question that the Alpine glaciers were formerly much more extensive. These early observers indeed may be truly said to have laid the foundations of glacial geology. During the many years that have passed since the results of their investigations were given to the world, hosts of geologists from every country have visited the Alpine lands and increased our knowledge of their glaciation. But the next most notable advance was made when Morlot in 1854 and Heer in 1858 discovered that there had been more than one great extension of the glaciers. These two investigators were the earliest to recognise the existence of certain accumulations which have come to be known as 190 TESTIMONY OF THE GLACIAL FORMATIONS " interglacial" formations. Still later it was reserved for a Scotsman, Andrew Crombie Ramsay, to demon- strate that the great Alpine lakes occupy basins of glacial erosion. His views were for years pertinaciously contested, but many of his opponents, who thought that he had greatly exaggerated glacial erosion, are to-day constrained to admit that glaciers are much more effective agents of erosion than Ramsay himself had suspected. In short, it is now maintained, as we have learned, that many Alpine valleys have been throughout deepened by glacial action, and that the present rock-basins, profound and capacious as they may be, are yet of subordinate irnportance. They are, in short, relatively shallow depressions hollowed out in the bottoms of valleys, which in many cases had been already over-deepened by glacial scour and excavation. Many other questions connecteid with the glacial history of the Alpine lands have interested geologists since the appearance of Ramsay's paper, On the Glacial Origin of Certain Lakes in. Switzerland ; but important as these are, it must be admitted that by the work of the earlier observers the fundamental conclusions of glacial geology had already been established. Shortly stated, these conclusions are as follows: {a) the former greater extension of the glaciers ; (^) the periodical return of such extensive glaciation ; and {c) the effective action of glacier-ice as a modifier of the earth's surface. Nearly thirty years ago Dr Penck, in his well-known work on the OLD GLACIERS OF THE ALPS 191 glaciation of the German Alps, recognised that these conclusions summed up the chief results of glacial research hitherto obtained, and in his and Professor Bruckner's recent great work — Dh Alpen im Eisseit- alter- — the same opinion is expressed with regard to the present position of glacial geology. It is needless to say, however, that the problems considered by the older glacialists have since their time been looked at .from other points of view, and treated in a different manner, while many subsidiary problems of much interest and importance have been discussed. Only those who are conversant with the literature of the science can realise the great advances made within the past thirty or forty years. By the" following of new lines of research and the employment of improved methods of investigation, our knowledge of the history of the Ice Age of Europe has been in a manner revolutionised. And no researches of the kind have been more fruitful in results than those pursued in the Alpine lands. The Alpine glaciers of the Ice Age attained gigantic proportions as compared with their puny successors of to-day, but it must not be supposed that the mountain-land was ever so continuously covered with ice as is the case with Greenland, The ancient glaciers did not, like the vast Arctic glaciers, draw their supplies from one uninterrupted snowfield, but each was fed from its own particular n6v6-basin. Nevertheless neighbouring glaciers were in many cases not so independent, so sharply separated from 192 TESTIMONY OF THE GLACIAL FORMATIONS each other as in our day. Not infrequently they coalesced across what are now dividing watersheds. This was especially the case on the northern slopes of the chain, as in Switzerland, North Tyrol, and Upper Bavaria. In those regions the glaciers reached the low forelands, where they united to form a continuous ice-sheet. It was otherwise, however, with the glaciers that descended from the higher Alps towards the east and south-west. These were clearly separated from each other, and did not coalesce even upon the low grounds ; many indeed dropped their terminal moraines well within their mountain valleys. The same to some extent was the case with the ice-flows that drained the southern flanks of the chain. Several of these, however, deployed upon the forelands — the glaciers that occupied the sites of the great lakes Maggiore, Lugano, and Como uniting outside of the mountain valleys to form a continuous ice-covering. It would appear, therefore, that the middle section of the Alps, between Switzerland and Upper Bavaria on the one hand, and the region of the Italian lakes on the other, was' the area of maximuiri glaciation. The elevated central part of that area was covered with continuous ice, from which glaciers flowed north, while others trended south. It is worthy of note, however, that the ice-shed separating those two sets of streams nowhere coincided with, but lay north of the water- shed. The same fact has been observed in connection with the glaciation of the Scandinavian peninsula and the Scottish Highlands — in neither of those regions SNOW^LINE OF GLAClAt TIMES 193 did the ice-shed of glacial times coincide with the watershed. In general terms it may be said that the glaciers of the Ice Age were simply exaggerations of their present successors. The latter are fed from the same neve-basins as those from which the glaciers of the Ice Age drew their supplies. It would appear, more- over, that these basins were not, save in the central part of the chain, more deeply filled in Pleistocene times than they are now. According to Professor Penck, the Alps above the existfrig snow-line must have much the same appearance as they had in the Glacial period. If this be the case, then it would seem that the former vast development of glaciation was due, not so much to increased precipitation of snow, as to a lower rate of ablation or melting. In other words, extreme glacial conditions were the immediate result of a general lowering of the tempera- ture. It is further notable that the precipitation of glacial times bore a close relation to that of the present. The areas of maximum and minimum pre- cipitation during the Ice Age and in our own day are the same — the existing snow-linej running approxi- mately parallel to that of the Glacial period, but at an average elevation of 1200 metres above it. The general facies of the organic remains, met with here and there in the moraines and fluvio-glacial gravels of the period, are quite in keeping with these conditions. On the north side of the Alps the great confluent glaciers terminated in a dreary tundra-like N 194 TESTIMONY OP THE GLACIAL FORMATIONS region, lying some 400 to 600 metres below the depressed snow-line. From the ice-front escaped numerous glacial streams, which distributed broad sheets of shingle and gravel over the low forelands. During summer these gravel-flats would necessarily be traversed by a network of water-courses which in winter-time would be mostly dried up. The foreland of the Alps would thus seem to have resembled the tracts extending outwards from the Vatnajokull in Iceland. It was, in short, a desolate region, but sufficiently clothed with vegetation to tempt thither the mammoth, the woolly rhinoceros, and the reindeer. The conditions on the south side of the Alps were, according to Penck, less forbidding. There the snow-line was higher, and forests extended up to a height of 800 to 1000 metres, so that many of the glaciers must have invaded tree-covered areas ; and it is even quite likely that trees may here and there have grown upon the moraine-covered glaciers them- selves, just as is the case to-day in Alaska and the Himalaya. The gravel-flats spreading out in front of the southern ice-flows were not so sterile as those on the opposite side of the mountains. In certain places, not liable to be flooded by the glacial waters, a some- what rich molluscan fauna flourished. Forests, how- ever, did not extend everywhere along the foot of the mountains ; on the contrary, wide stretches of marsh and peat-bog covered considerable tracts, the pools in which nourished alpine diatoms. The great CONDITIONS IN ALPINE FORELANDS 195 mammals that roamed the tundras on the north side of the chain appear not to have frequented the forest- lands of the south. The teeth of mammoth, which are not uncommon fossils in the northern gravel-beds, ^occur very rarely in the south, one specimen alone having been obtained, while remains of the reindeer have been met with only at Mentone on the west, and in Carinthia on the east side of the Alps. Of the woolly rhinoceros no trace has been found. It is further notable that the arctic element, so conspicuous in the glacial fauna of the north, is not present in the south, where the fauna is characteristically alpine — chamois, ibex, alpine hare, and marmot having ranged far south in the penin- sula. According to Professor Penck, therefore, the southern foreland of the Alps must have resembled the south coastlands of Alaska rather than the drearier wastes of Iceland. Towards the east and south-west extremities of the chain, the larger glaciers must likewise have descended into forest-clad land. Many of the smaller ice-flows of those regions, however, were confined to the upper reaches of the mountain valleys. The general aspect of middle Scandinavia probably reproduces, not inaptly, the conditions that characterised the two ends of the Alpine chain during glacial times— the summits of the south-west Alps appearing then not unlike the present snowy heights in the interior of New Zealand. I have said that the Pleistocene glaciers of the Alps must have resembled their ptmy representatives 196 TESTIMONY OF THE GLACIAL FORMATIONS in almost every respect except size ; and to bring this all-essential difference clearly before you, I may contrast certain existing Alpine glaciers with their Pleistocene predecessors. The present glacier of the Rhone, as everybody knows, is one of the smaller glaciers of the Alps. At its source,, some 10,200 feet above the sea, It is surrounded by peaks and ridges that range in height from 500 feet to about 2000 feet higher. After a course of less than six miles it terminates at a height of some 5700 feet. Now contrast this with the ancient glacier. At its source the latter reached on the flanks of the Schneestock a height of 11,600 feet, or 1400 feet above the surface of the existing glacier. At Visp, sixteen miles further down the valley, the upper limits of glaciation occur at a height of over 5500 feet above the river. Opposite Leuk (twenty-five miles below Visp) the flanks of the lUhorn are ice-worn* up to more than 4900 feet. Again, at Martigny (thirty-five miles below Visp), glaciation can be traced on I'ArpIlle to a height of 5300 feet above the bottom of the great valley. After leaving its- relatively narrow course, the ancient glacier spread itself over the site of the Lake of Geneva, and all the adjacent low grounds, and abutted upon the southern flanks of the Jura, only the higher ridges of which rose above the surface of this great mer de glace. On the slopes of Le Chasseron above Yverdon, at the head of Lake Neuchatel, the upper limits of glaciation occur at a height of 4300 feet, or 3100 feet above the THE OLD GLACIER OF THE RHONE 197 surface of the Lake of Geneva. " The higher ridges of the Jura overlooking the Swiss Lowlands seem, in short, to have formed islets or nunataks in the broad sea of ice which overflowed through passes in the mountains, reaching Ornans, some' thirty miles north- east of the Le Chasseron, and Salins, a similar distance in a westerly direction." It is hardly necessary to say that this overflow through the passes of the Jura vvas from the upper portion of the great glacier, the lower strata of ice could not mount the rocky barrier,^ but were deflected by it to right and left- — one branch of the sheet flowing away towards the north, and the other in the opposite direction. As the latter " passed out of Switzerland, it was joined by the glacier of the Arve and other ice- streams coming from the mountains of Savoy, and flowed at first due south, being prevented following a westerly course by the opposing mass of the Grand Colombier. On the flanks of this mountain, opposite Culoz, thirty miles below Geneva, glaciation goes up to a height of 9300 feet. Rounding the Grand Colombier, the united ice-streams of Rhone and Savoy, now joined by the glaciers of Dauphiny, fanned out upon the low grounds of France — showing a frontage that extended from Bourg, Trevoux, and Lyons to Vienne, and thence by Beaurepaire to Vinay in the valtey of the I sere, a distance of more than 100 miles. It is interesting to note that between Culoz and Grenoble, all the hills under a height of 3900 feet are rounded 198 TESTIMONY OF THE GLACIAL FORMATIONS and ice-worn and sprinkled witfi bottom-moraine. After crossing the secondary chain of the Grand Chartreuse and the Dent du Chai, the old mer de glace rapidly lowered its surface as it deployed upon the plains of Dauphiny and the Dombes." The heights attained by glaciation enable us to form some conception of the inclination of that surface from the source of the glacier in the neve-fields under the Schneestock to its termination in the low grounds of France. From its source to Oberwald, a distance of nine miles, the descent was steep, for the bed of the valley between those two points has an average fall of probably not less than 700 feet per mile. The slope of the surface of the old glacier in this steep part of its course was therefore some 300 feet per mile. From Oberwald to Leuk (thirty-nine miles) the surface slope declined to seventy-five feet per mile. From Leuk down to Martigny the inclination of the surface was not quite two feet. This slight fall points to the gorging of the valley by such glaciers as those of the Aletsch and other tributaries. From Martigny to the Lake of Geneva the surface-slope increased, but over the lowlands of Switzerland the surface must have approached horizontality. The inclination, how- ever, again increased to forty-two feet per mile as the ice crossed the secondary chain to fan out upon the loWj'ground beyond. The total length of the glacier being 245 miles, its surface-inclination from source to termination was rather more than forty-three feet per mile. But from the head of the Lake of Geneva to OLD GLACIERS OF THE RHINE, ETC. 199 Lyons — i.e., the lower course of the glacier — the fall of surface hardly exceeded thirty-five feet per mile. The north branch of the Rhone glacier need not be followed in detail. I may just note that, like the southern branch, it sent longer and shorter ice-streams through the chain of the Jura. Flowing away to north-east it was joined successively by the glaciers of the Aar, the Reuss, and the Linth, and thereafter this united ice-flow became confluent with the great glacier of the Rhine. How far down the valley of the Rhine the massive glacier extended is somewhat uncertain. It appears to have been confluent with the glaciers of the Black Forest, but the great valley is so deeply covered with fluvio-glacial and alluvial accumulations that we cannot say where the ice- stream actually terminated. We know, however, that the enormous Rhine glacier occupied the site of Lake Constance, and extended north into the valley of the Danube for nearly forty miles beyond the northern shores of the lake. Immediately to its right it seems to have coalesced with the glacier of the I Her, the terminal moraines of which occiir about eighteen miles north of Kempton, and therefore well out upon the foreland. Next, passing towards north-east, we encounter the terminal moraines" of the confluent glaciers of the Lech and the Isar^— the moraines of the latter extending some nine miles north of the Ammer See, The Isar glacier coalesced on its right with the glaciers of the Inn and the Salzach. Thus at the climax of the Glacial period the northern 200 TESTIMONY OF THE GLACIAL FORMATIONS forelands of the Alps were covered by a continuous mer de glace that extended from the valley of the Rhine through Baden, Hohenzollern, Wurtemberg, Swabia, and Upper Bavaria to the valley of the Salzach — a distance, not following the sinuosities of the ice-front, of some 250 miles. Before leaving these glaciers I may take special note of one — that of the Inn — on account of the apparently abnormal course it followed in the upper reaches of its valley. The river Inn rises in the Silsersee at the head of Upper Engadine, whence it holds a general north-easterly course until it leaves the mountains. The Engadine, aS the upper portion of the valley is called, is broad and trench-like, resembling in that respect the similar longitudinal valley of the Rhone above the Lake of Geneva. The most notable feature of the Inn valley is the remark- able width it attains at its very head. Near Maloja it is as broad as for many miles further down — the head is reached not in a cirque or deep mountain gorge but in a broad lake. Immediately above the upper end of this lake we cross a narrow ridge of no great height, and find ourselves on the verge of a precipice that sinks abruptly into the deep Val Bregaglia. Obviously the Engadine and the Val Bregaglia are different portions of one and the same -great longitudinal hollow — the former trending north- east into Tyrol, and the latter south-west towards Chiavenna and Lake Como. The divide between the two valleys is far too narrow ahd too low to have THE OLD GLACIER OF THE INN 20 L formed a gathering ground for snow to feed the vast glacier which we know occupied the Inn valley. Yet the evidences of glaciation are conspicuous throughout the whole extent of that long hollow from its source to its termination. Ground-moraine in many places paves the floor of the valley, striated rocks, and roches moutonn^es appear again and again, while the mountain slopes on either side are severely glaciated up to a height of 2000 feet above the river. The line separating the glaciated rocks below from the rough non-glaciated rocks above is particularly well- marked^ — side moraines and erratics occurring at and below that line. The descent of the valley from the Maloja is quite gentle. To Celerina or Samaden (opposite the mouth of the Pontresina valley), a distance of ten miles, the fall is not more than 230 feet, or twenty- three feet per mile, and most of that fall takes place between St Moritz and Celerina — the descent from Maloja to Campfer (seven miles) not exceeding three feet per mile. The glacier of the Inn, having no independent n6v6-field of its own, must have been fed entirely by its numerous tributaries. Only a brief glance at a good map of the district suffices to show that this must have been the case, and the fact becomes very evident when we trace the direction of ice-flow. The trend of rock striae and roches moutonnSes, and the "carry" of erratics, demonstrate that the glacier, instead of flowing down, flowed up the valley from Celerina to Maloja. The cause of 202 TESTIMONY OF THE GLACIAL FORMATIONS this apparent anomaly is not far to seek. The Pontresina valley opening upon the valley of the Inn from the south-east was gorged with ice, supplied chiefly by the Morteratsch and Roseg glaciers. That ice flowing into the Inn valley, nearly at right angles to its direction, was deflected to right and left. The left branch was thus compelled to make its way for ten miles up the valley, being joined by several lateral tributaries before it reached Maloja, after which the united ice-flows poured into the Val Bregaglia, and subsequently joined the massive glacier that flowed by Como and Lecco info the low grounds of Italy. The right branch, on the other hand, followed a normal course down the Inn valley through Tyrol, and eventually fanned out upon the foreland in Bavaria, with a frontage of nearly seventy miles — the point of the great ice-lobe extending forty-five miles beyond the mouth of the mountain valley. The glaciers that descended the south side of the Alps did not, for obvious reasons, attain the extraordinary dimensions of such ice-flows as those of the Rhone, the Rhine, and the Inn. The latter drained a much greater extent of mountainous land, while the climatic conditions in the north and south differed then as they differ now. The snow- and n6v6-lines of the present descend to a lower level on the north than on the south side.t)f the Alps, and the same was the case in glacial times. Yet the glaciers that invaded the valley of the Po were of considerable size and not less important than some OLD SOUTH ALPINE GLACIERS 203 of their northern contemporaries. The ice-flow that occupied the valley of the Dora Riparia — the deep and broad depression traversed by the Mont Cenis railway — deployed upon the plains of Lombardy, and piled up its large terminal moraines at Rivoli, eight miles west of Turin. This, although one of the smaller Italian glaciers, had an ice- front some six miles in extent. The glacier of the great valley of Aosta, draining a far more extensive and loftier tract of mountains, necessarily attained much more notable dimensions, for it flowed for twenty miles beyond the mouth of the mountain valley, and terminated with an ice-front nearly forty miles in extent. Its end moraines at Ivrea are correspond- ingly enormous, constituting as they do a ridge of hills that rise several hundred feet above the level of the plains. Coming further easlf we next encounter the moraines of a grand mer de glace formed by the union of several important glaciers, the beds of which are indicated by the basins of Lakes Orta, Maggiore, Lugano, Como, and Lecco. These confluent lobes of ice had a width of not less than sixty miles. The glacier of Lake Iseo flowed out upon the low grounds for seven miles beyond the mouth of its mountain valley ; while the larger glacier of Lake Garda advanced for some nine miles beyond the termination of the lake — the ice-lobe measuring over seventeen miles in width between Castiglione and Sommacampagna. Further east the glaciers were not so large. Several, however, 204 TESTIMONY OF THE GEACIAL FORMATIONS deployed upon the low grounds,^ as, for instance, those of the Brenta-Piave and thef Taghamento, It is difficult to realise the aniount of work done by the ancient glaciers of the Alps. The excavation of the rock-basins now occupied fey lakes must have occupied a very long time, no matter how relatively rapid the process may have been. But the formation of these basins, extensive and deep as many of them are, hardly impresses one so much as the general over-deepening of the main valleys. When we think of those valleys being deepened throughout a large part of their course for several hundred feet, we cease to be surprised at the great extent and thickness attained by bottom-moraines in the low-lying parts of the Alpine lands. Naturally these moraines are seldom thick in the areas of dominant erosion, save here and there where the configuration of the surface and the conditions of ice-flow favoured their accumu- lation. It is upon the low grounds that spread out from the foot of the mountains that bottom- moraine reaches its greatest development. "All the wide tract between the Alps and the Jura, for example, is more or less thickly covered with it. It likewise extends in broad sheets along the northern front of the Alps, overspreading very considerable areas in south Germany. Similarly it cloaks large parts of the low grounds of Savoy and Dauphiny, and the middle valley of the Rhone in France, while round the lower ends of the great lakes bf Italy it puts in a prominent appearance." In the case of the existing DETRITAL ACCUMULATIONS OF PO VALLEY 205 »uny glaciers some portion of their subglacial detritus ir bottom-moraine may possibly be derived from ,bove, but such could hardly have been the case dth the gigantic ice-flows of the Glacial period. It s inconceivable that during the climax of glaciation iny notable amount of rock-rubbish derived from iuperficial sources could have made its way down ;hrough the thick ice-masses, and entered into the "ormation of their bottom-moraines. The latter, in I word, are the direct product of glacial erosion. It would be a great mistake, however, to suppose hat these bottom-moraines are the only products )f glacial erosion. Immense quantities of material nust have been swept out frop underneath the incient glaciers by the large rivers escaping from hem. The enormously thick and wide-spread iheets of gravel that cover so much of the forelands )n the north side of the Alps consist to a large ixtent of subglacial detritus, water-worn and re- irranged. And the same is the case, but in a lesser legree, with the deep gravelly deposits that stretch rom the foot of the Alps into the plains of the Po. ^ere, however, superficial morainic matter was lumped in front of the ancient glaciers to a much greater extent than was the case with the terminal noraines on the other side of the mountains, and onsequently a considerable portion of the old river- gravels of the Po valley may well consist of materials i^hich originally travelled down the mountain valleys ,s surface moraines. 206 TESTIMONY OF THE GLACIAL FORMATIONS The subglacial rivers swept along not only much shingle, gravel, and sand, but vast quantities of fine materials which could not come to rest in the immediate neighbourhood of the mountains. What became of that fine muddy sedim.ent? We cannot doubt that most of it discharged from the south side of the mountains would be transported seawards. But the conditions were very different in the low grounds that spread northwards from the Alps, which were traversed then as now by rivers coming from a glaciated area. We can well believe that during the summers of the Ice Age the swollen muddy rivers would inundate immense areas of low-lying land. The valleys of the Rhine and the Danube, for instance, must have been flooded to an extent impossible under existing conditions. The slack waters and temporary flood-lakes, formed in that way, would then become settling reservoirs, in which fine sediments would accumulate from year to year'. Such is believed by many geologists to be the origin of the materials which I described in a former lecture as loess. The loess, in short, consists in large measure at least of flood-loam which, as already explained, was at a later stage blown about and redistributed by the wind. From this very rapid summary of the evidence derived from a study of the great lake-basins, the over-deepened valleys, the vast accumulations of bottom-moraine, of fluvio-glacial gravel, and flood- loams, we cannot but conclude that a Glacial epoch must have been one of prolonged duration. OLD GLACIERS OF THE JURA, ETC. 207 Lest it should be supposed that I have xaggerated the extent of the snow-fields and ■laciers of the Alps during Pleistocene times, I shall ,sk you to consider the conditidns that obtained ontemporaneously in other parts of middle and outhern Europe. I have mentioned the fact that he ranges of the Jura were traversed in several )laces by an overflow from the great mer de glace of he Rhone, underneath which the low grounds of Switzerland were drowned. But the Jura supported now-fields and had large independent glaciers of its wn. The same was the case with the Vosges and he Black Forest, the moraines of which have long een known. In like manner the Harz and all the lountains of middle Germany, together with the 'arpathians, nourished glaciers, some of which ttained a considerable size. It is needless to say lat the Ural Mountains also supported independent tiow-fields and glaciers, more especially towards the orthern end of the range. More notable, however, are the glacial phenomena ccurring in regions further south. The Balkans nd other hilly tracts in south-east Europe exhibit bundant evidence of former snow-fields and glaciers, id it may be added that similar traces occur again nd again in the mountainous regions of Asia Minor. ;e-worn rocks and moraines testify likewise to the iistence during Pleistocene times of glaciers in the puan Alps, and more notable still in the less evated Apennines — regions which in our day are 208 TESTIMONY OF THE GLACIAL FORMATIONS devoid of perennial snow-fields and glaciers. So, again, the mountain valleys of Corsica have their ancient moraines, while the sierras of central and southern Spain bear like witness to the former presence of considerable glaciers. Evidence to the same effect is forthcoming from the Serra da Estrella of Portugal, and from the Asturian Moun- tains and the Pyrenees. The last-named mountains still carry snow-fields and glaciers, but these occur only in the loftier parts of the chain, the glaciers forming mere patches, as it were, on the flanks of the highest mountains. The largest of these glaciers, after a course of two miles, terminates at a height of 7200 feet above the sea. "Viewed from the south the crest of the chain appears bald and naked, and even the northern slopes of the mountains through- out long stretches show, in the height of summer, no snowy covering. Only here and there are the highest ridges and peaks flecked with snow and ispotted with small glacier-patches. During the Glacial period, however, these mountains presented a very different appearance, great glaciers flowing north and south from the dominant crest." The largest of these was the glacier of the Garonne which, coming from a height of 8400 feet, flowed north for a distance of forty-five miles to near Montrejean, where its moraines occur at 1500 feet above" sea-level. In its lower reaches this glacier was over 2600 feet thick, and formed a broad mer de glace, from which branches overflowed through passes into adjacent valleys. OLD GLACIERS OF THE PYRENEES 209 A^nother large glacier was that of the Ariege, which bad a course of thirty-eight miles. After its con- fluence with the glacier of Vic-Dessos, the united :ce-flow exceeded 1300 feet in thickness. The last Df these ancient French glaciers that need be mentioned is that of Argeles, which, having its source at a height of 9000 feet, flowed for a distance of thirty-four miles and terminated at Ade, some 1300 feet above the sea. The greatest thickness it attained was 3000 feet. The glaciers on the Spanish side of the Pyrenees were less extensively developed, thus presenting the same contrast that obtained in thd Alps between the ylaciers that flowed down the northern and southern slopes of that great mountain-land. And the causes jf the contrast were doubtless the same in both ;ases— ^he southern exposure and more rapid descent nto the low grounds of Italy and Spain respectively. The largest glacier of the southern Pyrenees — that of ;he Gallego valley — was thirty-eight miles in length, 3ut the others were neither so numerous nor so mportant as those on the French side of the moun- ;ains. According to Dr Penck, the Spanish glaciers lad a mean length of under twenty-five miles, and :erminated at the level of 2600 feet above the sea ; ivhile the average length of the Fi-ench glaciers was )ver thirty-eight miles — their terminal fronts occurring It a height of 1600 feet above the sea. During the Ice \ge the snow-line of the Pyrenees" appears to have jeen depressed some 3600 feet belojv its present level, o 210 TESTIMONY OP THE GLACIAL FORMATIONS and, just as in our day, it gradually rose from west to east. In the more elevated tracts of southern and central France, now all below the snow-line, traces of former glaciatlon are often met with. In the Cevennes and the plateaus of Aubrac and Auvergne old moraines are frequently encountered. The evidence of former glacial action is especially well displayed in the high plateau region of Auvergne, so famous for its numerous and often well-preserved volcanic cones, of which the most important are Puy de Ddme (4806 feet), Mont Dore (6188 feet), and Cantal (6093 feet). From each of these volcanic mountains considerable glaciers descended and spread far out upon the surrounding plateau. It is interesting to note, that Acheulian implements are frequently met with lying on the surface of the morainic accumulations laid down by these ancient ice-flows. Further north in France evidences of former glaciation occur at considerably lower levels. In the mountains of Beaujolais and Lyonnais, for example, several glaciers of considerable size have left indisputable traces behind them. The largest of these had its source in the mountain-mass of Saint- Rigaud, some 3300 feet in height. Snow -fields, however, existed in the same region at levels not exceeding 2600 feet, from which glaciers descended. A little further south rises the isolated mountain-group of Mont Pilat (4900 feet), in the valleys opening out from which moraines of important ice-flows may be RUBBLE-DRIFT IN HILLY REGIONS 211 seen. More notable still are the relics of glaciation met with in the plateau of Morvan, which reaches an extreme height of 2960 feet. Large erratics and morainic debris occur plentifully round the Morvan, showing that glaciers have streamed outwards in many directions from the dominant heights. Great sheets of fluvio-glacial gravel spread down all the valleys of that region and extend even into the Paris basin, where erratics derived from the Morvan have long been recognised. In many other parts of middle and southern Europe where no true moraines may occur, we are nevertheless not without other evidence of cold climatic conditions. Almost everywhere we meet with heaps and sheets of rock-debris of essentially the same character and origin as the limestone-breccias of Gibraltar and the "head" of southern England. They are of course most notable in "hilly regions, as in the Vosges, the Black Forest, the Swabian and Franconian Jura, the Carpathians, the Balkans, the Apennines, the Sierras of Spain and Portugal, the high-lying parts of southern and central France, etc. The rubble-heaps referred to not only cloak the hill- slopes, but extend down the valleys or spread outwards for considerable distances over adjacent low-lying tracts. They have everywhere the same character : — they consist of local materials, have been formed under conditions very different from the present, md are no longer accreting but gradually wasting iway. 212 TESTIMONY OF THE GLACIAL FORMATIONS The evidence now adduced may suffice to convince you that the climate of Europe during a glacial period must have been very cold. Before concluding this rapid review of the evidence, let me once more refer to the wide-spread loamy deposits known as loess, in which remains of the tundra and steppe faunas and relics of Palaeolithic man frequently appear, I described these loamy deposits as having been rearranged by the action of wind, but nothing was said as to the cause of the dry climatic conditions of the steppe period, or of the source of the materials which the wind distributed so lavishly over southern Russia. Some have supposed that when an ice- sheet covered northern and north-western Europe, it must have acted as a powerful condenser upon the moist winds from the Atlantic. Precipitation, it has been inferred, would take place before the clouds could reach central Europe, and the climate there would consequently be very dry. France could not be much affected in that way, but= the dryness would steadily increase as the heart of the Continent was approached. We have no reason to believe, however, that the prevalent winds in glacial times came from the west, and even if they had it is not easy to see how or whence they could have obtained sufficient materials for the formation of the loess. During the climax of glacial times the inland ice of Scandinavia flowed south to the foot of the m'ountains of middle Germany and thus covered much of the area presently occupied by loess. We must reftiember, moreover. ORIGIN OF THE LOESS 213 that most extensive regions througHout middle Europe must have been inundated by the waters escaping from that northern ice-sheet and from the great glaciers of the Alps and other moujitains. Doubtless much loam would then be deposited in the slack waters and temporary lakes of the time. It could only be during winter, however, when these waters were much reduced in extent, that loamy materials could be exposed at the surface over any considerable areas. Moreover, when we keep in view the great extent and thickness attained by the loess throughout central and especially south-east Europe, it is impossible to believe that its materials could have been caught up by the wind from the relatively limited tracts left dry in the winter of a glacial period. In southern Russia alone the loess extends from 55° 55' N, lat. to the Black Sea, and reaches even as far south as the Caucasus. The late Dr Croll was of opinion that the prevalent winds during a glacial period would not be westerly as they are at present, but northerly and north-easterly. He thought they would be exceedingly powerful, in consequence of the great difference in temperature between high latitudes and the equator. More recently M. Tut- koffski has come to a somewhat similar conclusion. He believes that over the inland ice an enormous anticyclonic system existed which induced constant centrifugal winds. These winds would die away in the circumjacent regions, but far beyond the limits of the ice-sheet. They would thus partake of the -214 TESTIMONY OF THE GLACIAL FORMATIONS character of foehns, having a temperature elevated in proportion to the descent from the high central part of the ice-sheet towards its much lower periphery. They would also be very dry winds. Such conditions are to some extent realised at present in polar regions, being most pronounced where glaciation is most extensive, as in the north of Greenland. The geological action of the foehns of Pleistocene times could not come into play while the ice-sheet was advancing or had reached its limits. Extensive areas in front of the mer de glace would at that time be inundated, while the regions beyond would be protected from the action of wind by the compact covering afforded by the vegetation of the tundras. In western Europe, therefore, great sheets of sand and gravel, liable to frequent inundation, would form a broad marginal zone in front of the ice-sheet ; in middle Europe the waters of ablation, dammed up to a large extent by the mountains of Germany, would have more difficulty in making their escape ; but in eastern Europe no such barriers existed, the wide valleys of the Dnieper, the Don, and the Volga opening a way directly to the south. A broad zone in southern Russia, therefore, would be subject to repeated inundations, and would become clothed with extensive sheets of sand. While these conditions obtained, the formation of dust-deposits must have been precluded. In a word, the loess of middle and south-eastern Europe could not have been generated and spread out over ORIGIN OF THE LOESS 215 those regions during the climax- of glaciation- — its formation took place at a later stage. According to M. Tutkoffski, with whom geologists generally will agree, the loess could only begin to accumulate when the great mer de glace was retiring. As the latter continued its retreat it left deserts behind it — wide regions overspread with sand and boulder-clay. For a long time, Tutkoffski remarks, these regions must have been devoid of vegetation, owing to the absence of a fertile soil and to the desiccating effect of the glacial foehn. Subject to insolation, to chemical and physical decomposition and disintegra- tion, and to the mechanical action of the wind, the deserted bed of the retiring mer de glace could not have been otherwise than a barren desert ; Tutkoffski, therefore, speaks of it as the "zone of deflation." The glacial foehn constantly scouring that zone swept up the fine powder or dust of the desiccated morainic and fluvio-morainic accumulations, and transported it far away to south-east, south, and south-west, finally depositing it over a broad tract bordering the zone of deflation. That region of dust accumulation Tutkoffski terms the " zone of inflation." Meteorological considerations and the evidence of organic remains compel us to cpnclude that this last-named zone must have been- a vast region of continental steppes, clothed with vegetation and peopled with a true steppe fauna. Here it was that the foehn, owing to its diminished gradient, gradually died out, depositing its fine dust or loess, in which the 216 TESTIMONY OF THE GLACIAL FORMATIONS remains of that fauna and the relics of Paleeolithic man have been preserved. As the vast mer de glace continued to retire, the zones of deflation and inflation necessarily shifted northwards also. Hence in time the zone of deflation, characterised by its desiccated moraines, and barren sands with their glyptoliths or wind-polished stones, was gradually buried under loess. In short, as the zone of deflation moved northwards it was encroached upon part passu by the zone of inflation. But when the margin of the retreating ice reached the northern regions of Europe, and the Baltic and the North Sea began to appear, the climate of the zone of deflation would be rendered more humid or less continental. At the same time as the area covered by the inland ice was being gradually restricted, the foehns flowing out from it would become correspondingly feeble and inconstant, until eventually deflation and the seoHan transport of dust would cease. It is for this reason, Tutkoffski remarks, that the range of the loess towards the north is strictly limited. Thus, as he says, the normal loess, which, with its abundant evidence of steppe conditions covers deposits that bear witness to tundra and desert conditions, is as truly a product of the old ice-sheet as its erratics, moraines, and fluvio-glacial accumulations. The difference between it and these deposits is simply this, that the latter tell us of a time when the great mer de glace was in the heyday of its might and vigour, while the former is the witness of its decay and dissolution. ORIGIN OF THE LOESS 217 Tutkoffski's views apply specially to the enormous accumulations of loess in south-east Europe, which are almost exclusively of aeolian origin. The loess of middle and western Europe, however, is not entirely wind-blown. Not infrequently it would appear to be rather of the nature of a flood-loam, and here and there it contains fresh-water shells. The valleys of the Danube, the Rhine, and other rivers descending from the Alps were subject, as we know, to frequent inundations, from which loam must have been abundantly deposited. Nevertheless, the great valleys referred to were likewise swept by winds, and the loess throughout wide tracts has all the characteristics of a wind-blown formation, having yielded, as we have learned, many relics of Palseo- lithic man and plentiful remains of the steppe fauna. The wind-blown loess of central Europe, like that of south-east Russia, did not begin to form until the tundra conditions of the Glacial period were passing away. But whether it was distributed entirely by northerly winds is doubtful. Under existing con- ditions foehns descend from the snow-covered Alps, and it is not unreasonable to infer that similar dry but more powerful winds may have swept down from the Alps and other snow-clad mountains during the slow retreat of the great glaciers. LECTURE VIJI THE TESTIMONY OF INTKRGLACIAL FORMATIONS Evidence of Changes of Climate. Different Types of Interglacial Forma- tions. Examples from Scotland, England, France, Northern Germany, and Central Russia. Interglacial Deposits of the Alps — in Tyrol, Switzerland, and North Italy. Contrast between the Interglacial Floras of Northern and Southern Alpine Lands. Duration of Interglacial Epochs, as measured by Extent of River Erosion. Gibraltar Interglacial Deposits. From my description of the glacial conditions of the Pleistocene period, you must have had no difficulty in concluding that the arctic animals and plants of which I spoke In my first lecture must have lived in Europe during the Ice Age. That indeed is demon- strated by the fact that their remains occur in the morainic and fluvio-glacial deposits of what are now our temperate latitudes. We feel assured, then, that in glacial times the middle tracts of our Continent must have existed as bleak and barren tundras ; and that when the great Alpine glaciers and the Scandinavian inland ice began to retire, these tundra conditions slowly changed and were gradu- ally succeeded by a steppe climate, not unlike that of south-east Russia and south-west Siberia in our own day. 218 EVIDENCE OF CLIMATIC CHANGES 219 But you will remember that warm or genial conditions also characterised some stage or stages of the Pleistocene period, when, hippopotamus, elephant, and other animals of a southern habitat roamed over our Continent. The evidence obtained from cave-accumulations and ancient river alluvia clearly indicates, therefore, that alternations of climate took place during Pleistocene times, and we have next to inquire whether the glacial formations have any similar tale to tell. Well, they have ! We know now that the so- called Ice Age was not one continuous period of intense cold, but was again and again interrupted by long spells of more genial conditions. The Pleistocene period was characterised above all by these cHmatic oscillations. It consisted, in short, of a cycle or succession of glacial and interglacial epochs. Before proceeding to the description of particular interglacial deposits, it may be well to define what is meant by the term interglacial. Originally it was applied to fossiliferous deposits, usually of fresh-water but sometimes of marine origin, which rested upon glacial accumulations of one kind or another, and were covered by similar products of glacial action. For instance, when lacustrine sand and silt, containing fresh-water shells, mammalian remains, and peaty layers with temperate plants were found resting upon boulder-clay and overlaid by another boulder- clay, they were said to be interglacial,. The inference 220 TESTIMONY OF INTERGLACIAL FORMATIONS to be drawn from such a succession of deposits seemed obvious. The underlying boulder-clay indicated the former presence of a great glacier or an ice-sheet, which subsequently melted away, and the land then became clothed with temperate vegetation and occupied by large animals. Subsequently the climate again changed and glacial or arctic conditions supervened, as witnessed by the overlying boulder- clay. Geologists who had not themselves seen such successions of glacial and interglacial beds were for some time sceptical as to the interpretation put upon them by their discoverers. Many were the ingenious attempts to explain them away. Sometimes, for instance, it was suggested that the overlying boulder- clay might have slipped en masse from some adjacent hill-slope, and so covered up an old lake-bed. When this was shown to be impossible,, then doubts were expressed as to whether the upper boulder-clay was really what it seemed to be — a glacial deposit. Eventually, however, when discoveries of interglacial beds had greatly increased — when such successions as I have instanced had been met with in almost every country in Europe which had formerly been glaciated — scepticism could no longer prevail. It was then admitted generally that interglacial beds do occur. The only question that remained to be settled was the extent of the climatic changes they signified. Now, it is obvious that if interglacial beds contained only the remains of a. tundra flora and fauna, their presence need not imply any marked ADVANCE AND RETREAT QF GLACIERS 221 climatic change. Existing glaciers, as we all know, have their periods of advance and retreat. After a succession of mild and dry seasons with a diminished snowfall, alpine glaciers begin to retire, and vegeta- tion by and by invades the area fol-merly occupied by ice. Eventually severe seasons with a heavy snow- fall succeed, and the glaciers readvance and recover the ground they had vacated. In doing so, it is obvious their moraines may here and there come to overlie the soil and plant-remain? belonging to the period of temporary glacial retreat. Such being the case with valley-glaciers, it is probable that similar minor fluctuations of the ice- front may have characterised the' great mer de glace of northern Europe, and that relics of a tundra flora might now and again come to be enclosed in morainic accumulations. But when interglacial beds are marked by the presence of a flora and fauna indicative of warm-temperate or very genial con- ditions, it is quite clear that they cannot be thus explained. When the relics of such a flora and fauna are met with in the heart of a formerly glaciated region, embedded in deposits that rest upon and are covered by boulder-clay, the succession can only point to decided climatic changes. The term interglacial, however, is frequently applied to deposits which are neither underlaid nor overlaid by glacial or morainic materials. Cold climatic conditions, as we have learned, may be indicated by organic remains. Should we encounter in France or 222 TESTIMONY OF INTERGLAGIAL FORMATIONS south Germany, for instance, the reHcs of a tundra fauna and flora embedded in silt or clay, we should conclude that the deposits belonged to the Ice Age. And should two such series of deposits occur in the same place, separated the one from the other by beds charged with the remains of temperate and southern plants and animals, the intercalated beds would be correctly described as interglacial. It will be understood, therefore, that true inter- glacial formations may be encountered not only in regions where glacial accumulations occur, but in tracts which we have no reason to believe were ever invaded by glaciers or ice-sheets. To illustrate the mode of their occurrence in glaciated and non-glaciated territories, I may now describe a few typical examples. Throughout the low grounds of a large part of Scotland two sheets of boulder-clay are recognised, of which the upper and younger sheet is the better developed, extending continuously over wide areas. The lower and older is by no means so continuous, being often entirely wanting. Frequently, however, both boulder-clays occur together — the line of demarcation between them being usually conspicuous, when the face of a section is well exposed. Unfortun- ately the natural exposures in river-cuttings are often obscured under the debris that falls or is washed down from the upper part of a section. Nevertheless clear sections have been obtained, in which the younger clay sometimes rests directly upon the older clay, the upper part of the latter for a few feet down- SCOTTISH INTERGLACIAL BEDS 223 wards often showing a "weathered" aspect. That is to say, it is somewhat discoloured, as if from the chemical action of rain, its ferruginous constituents having become more or less oxidised. This would seem to indicate that the lower boulder-clay had for some time formed a land- surface, before the upper boulder-clay came to be deposited. Still more frequently the two clays are separated by beds of sand, gravel, and silt, which may vary in thickness from a few feet up to many yards. The exact significance of these intercalated Heds cannot always be determined. In some cases they may be of infra-glacial origin {i.e., accumulated underneath a mer de glace) ; in other cases they may be torrential or fluviatile, laid down in the light of day at a time when the mer de glace represented by the lower boulder-clay was melting away, or spread out at a later period in front of the advancing ice-sheet represented by the upper boulder-clay. Deposits of the kind, therefore, certainly show that there were pauses in the formation of boulder-clay, but they afford no clear evidence as to the duration of the pauses. Now and again, however, the water-formed beds that separate the two clays have a much more definite tale to tell us. From time to time they have yielded organic remains. Some good examples were formerly disclosed in our own neighbourhood, especially near Slateford, during quarrying operations. Unfortun- ately these interesting relics of interglacial conditions 224 TESTIMONY OF INTERGLACIAL FORMATIONS have gradually been removed or obliterated as the quarries have extended. Excellent sections were exposed both at Redhall Quarry and Hailes Quarry, where the two boulder-clays wece separated by a variable thickness of gravel, sand, and alluvial silt with peat. Among the species obtained from the peat were hazel, the nuts of which abounded, oak, alder, and Scots fir, and many small flowering plants. Along with these occurred elytra of beetles and cadis- cases — the whole assemblage of remains indicating a temperate climate like that of Scotland at present. The succession of changes indicated by the sections referred to would seem, therefore, to have teen as follows : — The lower boulder-clay is the bottom- moraine of an ice-sheet which flowed from west to east, as the carry of its included stones and the trend of glacial strise in the neighbourhood sufficiently attest. Now, that easterly movement was due to the union of two ice-sheets — one flowing south from the Highlands, and the other streaming northwards from the Southern Uplands. At the time the lower boulder-clay was being accumulated, there- fore, all Scotland must have been covered by its ice-cap. The thick gravel and sand bejis, with their large angular and subangular stones and boulders which rest immediately on the trenched -and eroded surface of the lower till, speak to us of a time when the mer de glace melted away, and when the waters derived -from the dissolving ice and snow denuded the bottom- BURIED FORESTS IN PEAT 225 moraine and rearranged detritus of one kind or another which the ice-sheet had l6ft scattered over its deserted bed. The interglacial beds occur in hollows of these fluvio-glacial deposits and are indicative of a temperate climate, when perennial snow and ice had either disappeared entirely, or were restricted to the higher elevations of our country. Immediately overlying the interglacial beds tor- rential fluvio-glacial deposits appeared. These were probably laid down at a time when the mer de glace represented by the upper boulder-cfay was beginning to overwhelm the land. That boulder-clay, just like its predecessor, is the bottom-moraine of an easterly flowing ice-sheet, and implies, therefore, that all the Scottish Lowlands were once more smothered in ice. Similar interglacial beds have been observed in other parts of our country, and the*, tale told by them is much the same as that just outlined. But climatic oscillations are not evidenced only by the occurrence of fossiliferous deposits intercalated between boulder- clays. In most peat-bogs throughout Scotland we find a succession of floras which tells a like tale of climatic changes. One very common phenomenon is the occurrence in those peat-mossds of two "buried forests," the lower separated from' the upper by a variable thickness of peat. Over forty years ago I pointed out that the wide distribution of such buried forests throughout the peat-bogs of the British Islands p 226 TESTIMONY OP INTERGLACIAL FORMATIONS and north-west Europe could only be accounted for by climatic changes. The peat in which the "forest beds" are entombed is composed of the bog-moss {Sphagnum) and other moisture-loving plants — the whole assemblage of plants being indicative of humid and relatively cool conditions, while, the buried forests, on the other hand, imply a drier and more genial climate. Within the past few years these general conclusions have been strongly 'Supported by the researches of a well-known botanist, Dr F. Lewis, who, during a careful examination of the peat-bogs in every part of our islands, has detected a definite layer of arctic-alpine plants intercalated in the peat that separates the lower from the upper forest-bed. We can no longer doubt, therefore, that climatic changes took place during the formation of our peat-bogs. I may now cite one or two examples of similar phenomena met with in England. There, just as in Scotland, we encounter interglacial deposits amongst the accumulations of the Ice Age. Near Hull, for instance, occurs a thick series of gravel and sand, charged with fresh-water shells, kmongst which the old river-shell, Corbicula fluminalis, is conspicuous. That species is now extinct in Europe, but still lives in the Nile and various rivers of Asia. It is of common occurrence, it may be remembered, in the Pleistocene river-alluvia of the Thames and the Seine, where it is associated with extinct and exotic mammalia. The fresh-water beds at Hull are known to be younger than the oldest boulder - clay of NORFOLK FOREST-BED 227 Yorkshire, and are directly overlaid by another and later boulder-clay. They are thus of interglacial age. The Pliocene and Pleistocene deposits of Norfolk have long been famous amongst geologists, and are of special interest as containing the records of the first approach of glacial conditions. The Pliocene strata are for the most part of marine origin — the fossils met with in the oldest member of the series clearly indicating a warm-temperate climate. But as we pass upwards in the series, northern types of life begin to appear and gradually increase in numbers, while at the same time the southern types, so dominant in the older strata, gradually vanish. When at last we reach the upper member of the series (Weybourn beds), we find the "marine molluscan fauna presenting a thoroughly arctic aspect. Over- lying these arctic shell-beds certain fresh-water and estuarine strata known as the " Forest-bed series " are next encountered. The deposits in question are essentially of fluviatile origin, and are often crowded with drift-wood and tree-stumps. The latter were formerly supposed to occupy the place of growth and to indicate an old land-surface ; hence the name of the series, but the stumps in question are now believed to have been drifted. Traces of an old soil, however, do occur in the series. Plant remains are plentifully present — all belonging to living species, and with few exceptions still indigenous to Norfolk. The general assemblage shows that the climate of East Anglia had now become mild and moist, and little if at all 228 TESTIMONY OF INTERGLAGIAL FORMATIONS colder than it is to-day. The associated mammalian remains are of great interest : they include those of elephants {Elephas meridionalis, E. antiquus), hippopotamus, rhinoceros (not the woolly rhinoceros), horse, bison, boar, and many kinds of deer, bears, sabre-toothed tiger, hysena, wolf, and others. Above the Forest-bed series come marine deposits containing arctic shells, while these arctic marine beds are in turn succeeded upwards by a fresh-water stratum charged with the relics of an arctic flora, amongst which we note the polar willow [Salix polaris) and other northern plants. Here again we have clear evidence of climatic oscillations. The Weybourn beds underlying the Forest-bed series introduce us to a North Sea tenanted by a well-marked arctic fauna. The Forest- bed series, on the other hand, implies decidedly temperate conditions : while the succeeding deposits just as certainly bespeak a relapse to an arctic climate. The next example I cite is of peculiar interest from our present point of view, inasmuch as it shows the relation of Palaeolithic man to those changing climatic conditions. At Hoxne, in Norfolk, a ie^ miles from Diss, a brick-clay has been worked for many years. So far back as 1797 stone imple- ments were recorded and figured as having been found in this clay-pit, and the figures: are clearly those of Palaeolithic tools. This interesting discovery, neglected for upwards of sixty years, was recalled DEPOSITS AT HOXNE 229 in 1859 when British geologists and archaeologists were flocking to the Somme valley to inspect the remarkable " finds " of similar implements in the Pleistocene river-gravels of that region. On visiting Hoxne, Mr (afterwards Sir John) Evans and Professor Prestwich found that the implements had come from one of a series of fresh-water beds occupying a hollow in the surface of a boulder-clay. From time to time since then the old clay-pit has been examined by geologists, and arctic and temperate plants have been obtained from the deposits ; but it was not till 1896 that a thorough examination of the place was made by a committee appointed by the British Association. Thanks to the efforts of that committee we now know the precise conditions under which the fresh-water beds were deposited, and their relation to the boulder-clay., A series of trial pits and borings showed that the fresh-water beds occupied a narrow channel excavated through glacial deposits, and were underlaid throughout by boulder- clay. The bottom beds of the fresh-water series consisted of lacustrine clays contaming abundance of plant-remains, and were succeeded upwards by a bed of lignite, over a foot thick. The facies of the flora was decidedly temperate — a large proportion of the species being still indigenous. Above that lignite came thirteen feet of dark carbonaceous loam and sand, charged with the relics of an arctic flora. Temperate conditions had thus eventually given place to a climate probably not unlike, that of the cold 230 TESTIMONY OF INTERGLAGIAL FORMATIONS treeless regions of North America and Siberia. At the top of the whole series lies the brick-clay which has so frequently yielded flint implements. A few mammalian bones and several s|iecies of molluscs occur in this deposit, but none of these are specially characteristic either of northern or southern habitats. The molluscs, with a single excerption, are all still natives of the district, while the few mammals are widely migrating species. We need only note that while they certainly do not point specially to an arctic climate, they are quite in keeping with temperate conditions. From the facts thus briefly recapitulated we learn that after the final dissolution of the great mer de glace represented by the bottom-moraine known as the "great chalky boulder-clay," the climate of East Anglia eventually became quite temperate. Whether man lived in the region at that tirne we do not know — no trace of his presence having hitherto been met with in the beds containing the abundant temperate flora. Neither have any Palaeolithic implements been discovered in the succeeding arctic-plant beds. It is only in the overlying brick-earth beds that such relics appear. Let me now recall for a moment the occurrence of the arctic plant-bed in the Pleistocene river-drifts of the Thames valley. That plant-bed, it will be remembered, overlies river-gravels from which Palaeolithic implements have frequently been obtained, and these gravels certainly rest upon and are there- TUFA OF LA CELLE 231 fore younger than the boulder-clay of the same region. I shall not at present stop to consider whether the arctic plant-beds of Hoxne and the Thames valley are of the same age. Certain con- siderations show that they can hardly be so ; it is enough, however, to recognise the. fact that great alternations of climate took place in Pleistocene times, and that man was certainly present during at least some of these remarkable changes. We may next turn our attention to the inter- glacial deposits of the Continent. The river-drifts of the Seine and the Somme have yiqlded, as we have seen, the same assemblage of organic remains and Palaeolithic relics as the corresponding deposits of the Thames valley, and must therefore be assigned to glacial and interglacial times. In my first lecture I referred to the tufa of La Celle (see Fig. 9, p. 233)— so famous for its abundant plant-remains, which include fig, box, judas-tree, Canary laurel, and many other species indicative of a climate as warm as that of the Dalmatian coast-lands. This tufa overlies the Pleistocene river-gravels, and is itself covered by a sheet of loess — there can be no doubt, therefore, as to its geological position. It is #orth noting that the plants in the upper part of the tufa indicate a somewhat cooler climate than those in the lower portion of the same formation. From the upper part Acheulian artifacts have been obtained. Although no glacial accumulations occur in the valley of the Seine near Paris, we do not hesitate to 232 TESTIMONY OF INTERGLACIAL FORMATIONS describe the tufa of La Celle as an interglacial formation — for its geological position is determined by the character of its organic remains. There are many other cases of interglacial deposits which occur in regions outside of the glaciated areas. But for my present purpose I may confine attention to such as are closely associated with boulder-clay and other morainic accumulations. Throughout a wide region, extending from Denmark and the Low Countries through north Germany into central Russia, several boulder-clays are recognised — often separated the one from the other by aqueous deposits, in wlich remains of the Pleistocene mammalia, shells, and plants have now and again been detected. For instance, in the neighbourhood of Berlin, beds of sand and gravel, resting upon one boulder-clay and covered by another, have yielded remains of southern types of elephant and rhinoceros, together with horse, urus, bison, Irish deer, stag, caribou, wdf, bear, mammoth, woolly rhinoceros, and musk-ox. Here we have a mingling of southern and temperate forms with a i'^v,- northern types. The deposits would seem to represent a long interglacial epoch, during which the climate changed from genial and temperate to cold- temperate or boreal. In other places the interglacial beds contain many plant-remains, the general fades of which indicates for north Germany and central Russia a temperate climate of a more Insular character than the present. Sometimes in these 3 J O ". tie ua O n M ■fc ^ g fl- H In =1 w ^ Ph o Ol S Ji ^ a !K T3 tJ ^ g «; H ^ N w tuO '^ o ^'^ S •'J 0-3 en O ;;s t? o ^ 2I o ° § ■a u <: J ■^ S3 f- ^s w S?, 71 1 1 •^ ^ 0^ S ■« iJ3^ m M C ~ •3 « S M . S ^ rf ^ >- J3 O 3 ■f;^'« o INTERGLACIAL BEDS OF GERMANY, ETC. 235 interglacial beds a gradual change in the facias of the flora may be noted as the deposits are followed from below upwards. The bottojn beds may yield abundant traces of a mixed forest vegetation, of a temperate character, consisting of such trees as hornbeam, birch, fir, along with pine, maple, lime, hazel, oak, holly, and willows. The assemblage of plants met with in the next succeeding layers is often suggestive of gradually changing conditions, for northern or cold-temperate mosses begin to appear, and still higher up we come upon the dwarf birch, but find no trace of lime, maple, hazel, oak, or holly — all of which had vanished at an early stage. I must not omit to mention, however, that the flora occur- ring in certain beds, resting upon and covered by glacial deposits, has throughout quite a northern or boreal character. Perhaps the most notable feature of the flora yielded by a large number of the interglacial beds of north Germany is that to which I have already alluded. Its facies implies a well-marked insular or oceanic climate, the occurrence of the holly in the heart of the Continent, where it no longer grows wild, being very noteworthy. Enough evidence perhaps has now been advanced to show that glacial conditions in north Europe must have been occasionally interrupted. Obviously the great ice-sheet had its periods of advance and retreat. Moreover, the changes 6f climate implied were so pronounced that we cannot suppose their 236 TESTIMONY OF INTERGLAGIAL FORMATIONS effects were confined to northern Europe. If arctic and temperate climates have alternated In the north, we might well expect to encounter evidence of such changes in regions further south. It will be necessary, therefore, to see how far those anticipa- tions are borne out by the records of the Ice Age in the Alps and other mountain-regions of middle and southern Europe. In the great valleys of the Alps, just as in the low grounds of northern Europe, several distinct boulder-clays or bottom-moraines occur, and these, like the corresponding formations of the north, are often separated by aqueous deposits which are sometimes fossiliferous. In shorty interglacial beds have been detected in many places throughout the Alpine lands, one or two examples of which I shall now describe. At Hotting, on the left bank of the Inn opposite Innsbruck, occurs a thick sheet of breccia of angular and subangular rock-fragments, which obviously owes its origin to the action of mountain-torrents. Beginning at a height of several thousand feet above the valley it sweeps down the mountain-slopes, and spreads over a well-marked terrace that rises some 430 feet above the Inn. As this breccia clearly over- lies a bottom-moraine and is covered by a younger morainic accumulation of the same nature, it is obviously interglacial. The underlying bottom-moraine tells us of a time when the Inn valley was occupied by an HOTl'ING BRECCIA 237 enormous glacier, to which local ice-streams descended from all the lateral valleys and mountain-slopes. Professor Penck estimates the snow-line at this stage to have been nearly 4000 feet below its existing level. The whole of the Inn valley was then occupied by a gigantic glacier which, escaping from the mountains, dropped its terminal moraines far out upon the low grounds beyond. The interglacial deposits overlying the bottom- moraine of that enormous ice-flow have yielded a large number of plants. Many of these, having been buried in fine-grained material, are in an excellent state of preservation. Over forty* species have been determined, twenty-nine of which are still native to the region or occur in other Alpine valleys. Six species flourish at present in north Tyrol, but are no longer found growing as formferly at a height of 4000 feet above the sea. More notable still are certain forms, six in number, which are now quite foreign to the Alpine lands. One of these exotic plants most abundantly present in the interglacial beds is the Pontic alpine rose (^Rhododendron ponticum). This species at present grows wild in south-west Spain, in the Caucasus, and in regions on the southern borders of the Black Sea., In the countries which are now its habitat, the snow-line occurs at a height of nearly 9800 feet. We could hardly expect it to flourish, therefore, in the Inn valley under existing conditions, when the snow-line does not exceed 3700 feet. One of its associates in the 238 TESTIMONY OF INTERGLAQAL FORMATIONS interglacial beds is the box [Buxus sempervirens), a species which cannot endure a severe climate. It grows wild in southern and south,-east Europe, and in the regions south of the Black Sea, where it flourishes in the woods, 5800 feet at least below the snow-line. The presence, therefore, of those two plants in the interglacial beds is clearly indicative of a warmer climate than now characterises the Inn valley. Other associated plants tell a similar tale, while the entire absence of the boreal and alpine types that now characterise that region is very striking. It is interesting to compare the mean annual temperature of the hills where the Pontic alpine rose now lives with that of the Inn valley. In the former the annual temperature is 57° to 65° F., while at Innsbruck (1883 feet above the sea) it is only 47°. The Hotting flora, however, flourished at elevations of 3600 feet to nearly 4000 feet, in places where the mean annual temperature does not now exceed 40°. Were the climatic conditions of that interglacial epoch to return, most of the Alpine glaciers would vanish — only a few would be left clinging to the slopes of the loftiest mountains. The boulder-clay or bottom-moraine that overlies the interglacial deposits tells the same tale as its predecessor. Once more a gigantic ice-flow filled the Inn valley and deposited its terminal moraines outside of the mountain region. Interglacial deposits, as I have said, have been SWISS INTERGLACIAL BEDS 239 -detected in many places in the Alpine lands. Among the earliest recognised were those of Switzerland. At Unter-Wetzikon and Diirnteti, for example, a few miles north of Lake Zurich, a lignite or brown coal with associated fresh-water beds has long been worked. The deposits rest upon and are overlaid by morainic accumulations, so that their interglacial position is clearly defined. Among the plants re- cognised are common Swiss fir, Scots fir, yew, white birch, sycamore, hazel, water-lily [Brasenia purpurea) — no longer a European plant — ^marsh trefoil, etc. This flora is indicative of a wariner climate than now characterises the northern foot -slopes of the Swiss Alps, and its evidence is supported by that of the associated fauna which includes straight- tusked elephant, broad-nosed rhinoceros, urus, elk, and stag. On the south side of the Alps similar evidence of climatic changes is encountered. . Thus in the Val Borlezza are found the relics of an old interglacial lake, resting upon and covered by glacial accumula- tions. Among the plants occurring in the lacustrine deposits are the box and the Pontic alpine rose — the .general assemblage of plants having a somewhat more southern aspect than that of Hotting. Associated with this flora occur remains of the broad-nosed rhinoceros and the stag. The origin of the lake is interesting. The mouth of the Borlezza valley appears to have been dammed up for a long time by the great glacier which occupied Lake Iseo. 240 TESTIMONY OF INTERGLAClAL FORMATIONS Into the barrier-lake thus formed mud and glaciated stones were washed from the side of the glacier. Eventually a great embankment of morainic detritus gathered along the flanks of the glacier, so that when the latter at last melted away, the waters of the lake were still held up. The morainic barrier, consisting largely of limestone debris, became solidified into a hard conglomerate, and endured for a long time after the retreat of the great glacier. A genial climate had now supervened- — the surrounding hills being clothed with maple, hornbeam, hazel, elm, box, and the Pontic alpine rose, while the broad- nosed rhinoceros became a native of the district. Eventually, however, these genial conditions came to a close, a great glacier reappeared; and dilating into the Val Borlezza, deposited its moraines upon the surface of the interglacial deposits. Finally, the advent of milder conditions, and the dissolution of the glacier, permitted the river Borlezza to cut the narrow steep trench through which its waters presently rush to join Lake Iseo. The examples now adduced frOm the Alpine lands may suffice to show that during Pleistocene times these regions experienced changes of climate as strongly marked as those of more northern latitudes. There are certain differences between the interglacial floras of the northern and southern sides of the Alps worthy of notice. The forests that clothed the mountain-slopes in the north were of the Baltic type, consisting mainly of conifers, oaks, maples, birches, INTERGLACIAL RIVER-EROSION 241 and hazels. Associated with these, however, were the yew, the box, the water-chestnut, and other types, the presence of which shows that the sriow-Hne could not have been lower than it is to-day. On the southern flanks of the Alps the flora had a marked Illyrian aspect ; chestnuts flourished at a height of 2600 feet, the vine grew as it still does along the banks of Lake Iseo, the box abounded, and the Pontic alpine rose (now no longer an Alpine plant) was Hkewise very widely distributed. Nowhere was this interglacial flora associated with arctic-alpine types. The most notable mammals of the time were extinct forms of elephant (not the mammoth) and rhinoceros (not the woolly species), and the stag. The changes from glacial to interglacial conditions and back again to glacial necessarily imply a long lapse of time. And this conclusion is still further confirmed by the considerable amount of valley erosion that was effected by the rivers of interglacial times, some of them during such a stage having cut their way down through hard rocks to a depth of eighty feet or even of 100 feet. Similar evidence of inter- glacial erosion occurs in many of the hilly regions outside of the Alpine area. In central France, for instance, it has been shown that .during interglacial times, valleys have been deepened* by the rivers for several hundred feet. But these and other evidence of the prolonged duration of interglacial conditions I may leave for the present. In a former lecture some account was given Q 242 TESTIMONY OF INTERGLACIAL FORMATIONS of the great limestone- breccias of Gibraltar, and I endeavoured to show that these remarkable accumula- tions were due to the action of frost, which shattered the rock of the culminating ridge — the debris being carried down the slopes and over the low ground by n6ve and melting snow and rain. The formation of that notable breccia or solidified rock-rubble took place at two distinct periods, separated from each other by a long epoch of milder conditions. The accumulation of the breccia of the first cold stage had ceased, and the loose agglomeration of grit and large and small blocks had become cemented into an indurated mass long before the formation of the later breccias. Not only so, but torrents had worn deep gullies in the older breccia, and acidulated water percolating through crannies and fissures had gradu- ally opened out a series of subterranean galleries and caves, that penetrated both the breccia and the underlying limestone. All this occurred at a time when Spain projected farther into the Mediterranean than it presently does, and when the cUmate was mild or genial, as is shown by the character of the mammalian remains that occur in the caves. At the period referred to Gibraltar must -have appeared as a verdure-clad alp, towering above the surface of a wide expanse of undulating country that stretched south towards the Barbary coast, with which, indeed, it may actually have been connected. The Rock was then tenanted by the ibex in large numbers, and visited from time to time by rhinoceros, elephant, horse, boar, GIBRALTAR LIMESTONE-BRECCIAS 243 and deer, and by bears, wolves, hyaenas, lions, leopards, lynxes, and servals, some of which may have made their lair in one or other of its numerous caves. Now and again torrents flowing down the mountain-slopes swept the carcasses and bones of animals into gullies and underground galleries, where they gradually accumulated along, with other super- ficial debris, and became in time sealed up by the action of carbonated water. Eventually submergence ensued, and the land subsided to the depth of 700 feet or thereabout below its present level. This movement seems to have been interrupted by longer or shorter pa'uses, during which the sea cut terraces or shelves on the flanks of the Rock — shelves which have been qxcavated partly in limestone and partly in the old limestone-breccia. By and by subsidence ceased, .and the land was re-elevated, probably in as gradual a manner as it went down. The old sea-shelves were then partially covered by shelly gravel and sand. So far as can be made out, the shells of these beds belong to the same species as are now living in the neighbouring sea, from which it may be inferred that the temperature of the sea during the period of submergence was much the same as it is to-day. To what extent the land was re -elevated we do not know. It certainly attained a greater height than at present, but whether or not Spain extended as far seaward as it did in the preceding period, can only be conjectured. The sand that clothed the 244 TESTIMONY OF INTERGLACIAL FORMATIONS flanks of the Rock was now acted upon by wind and rain, and to some extent re-arranged^ — " top-dressed," as it were — so as to form long sloping curtains or taluses, the surface of which here and there became hardened by the action of rain, dissolving the cal- careous matter of the shells, and again redepositing it between the grains of grit and sand. Next cold climatic conditions returned, when the limestone again began to break up under the action of frost, and rock-rubble was carried down as before to the low ground. Large blocks and smaller fragments toppling from the lofty cliffs that face the east, fell upon the sand-slopes^^ — here and there plunging into them, where the process of induration had not been far enough advanrced to enable the sand to withstand the force of the impact. In this way the shelly sands were eventually buried completely under a thick accumulation of angular blocks and debris, which having since b^scome thoroughly cemented now presents the same character as the older breccia. Underneath this newer breccia the shelly sands are often quite indurated into hard rock which has been quarried for building purposes. There still remains a long talus of sand, however, sweeping down from the great cliffs that overlook Catalan Bay. This sand is even yet only partially cemented, for it apparently contains a relatively small percentage of calcareous matter. The blocks discharged upon this slope, therefore, merely sank into it, and now PLEISTOCENE CLIMATIC CHANGES 245 project above its surface, while many doubtless must have rolled downward to rest at the base of the Rock in regions that are now submerged. From the samples of interglacial phenomena given you must already have come to the conclusion that they tell the same tale as the cave-deposits and old river-alluvia. It is quite obvious, indeed, that the glacial and interglacial formations must be the equivalents in time of the latter. They are all products of the Pleistocene period — some being indicative of cold, others of genial conditions. The period, in short, was characterised throughout by climatic changes, and we may naturally ask, how many such revolutions occurred, and at what particular stage in the series did prehistoric man first appear in Europe? Fortunately more or less definite answers to these questions are forthcoming. By correlating the geological and palaeontological evidence, the general succession of events that took place during Pleistocene times has been ascertained. In my next two lectures, therefore, I purpose giving an outline of that history. LECTURE IX THE HISTORY OF THE PLEISTOCENE PERIOD Climatic Conditions of Late Pliocene or Preglacial Times. First Glacial Epoch, or " Scanian " of Northern Europe, and " Giinzian " of the Alps. First Interglacial Epoch : Oldest Human Remains. Second Glacial Epoch, or " Saxonian " of Northern Europe, and " Mindelian " of the Alps. Second Interglacial Epoch ' Chellean and Acheulian Stages. Third Glacial Epoch, or " Polonian " of Northern Europe, and " Rissian " of the Alps : Mousterian Culture-stage. Third Interglacial Epoch : Mousterian Culture - stage ends. Fourth Glacial Epoch, or " Mecklenburgian " of Northern Europe and " Wiirmian " of Alps ; Aurignacian, Solutrean, and Magdalenian Culture-stages. In the Pliocene period— that is, the age immediately preceding the Pleistocene period — Europe experienced climatic conditions differing somewhat from those of the present. From the character of the Pliocene flora it may be inferred that the climate was not only genial but very equable. That flora was richer in genera, and some of those genera were richer in species, than is the case with our existing flora. The forests of Pliocene times were singularly uniform, the same species of trees flourishing contemporaneously in Tuscany and southern and central France. The great forests of the period, according to Saporta, 246 CLIMATE OF PREGLACIAL TIMES 247 occupied the plains, the margins qf rivers and lakes, and extended up the valleys to the crests of the mountains, without much change of character. The ivy, the platanus, the liquidambar, various maples, and many walnut trees, elms, hornbeams, laurels, sassafras, and others ranged from central Italy to the heart of France. It was an abundant vegetation, composed for the most part of great trees, some of which were destined erelong to become extinct ; some again were special forms belonging' to genera that are now exotic ; others were species which have survived to the present in more southern and eastern regions ; while yet others are still represented in Europe by identical or very closely allied species. Among the great mammals of the time were dinotherium and mastodon. Various extinct forms of elephant and rhinoceros also flourished towards the close of the period, and with these were associated hippopotamus, primitive horses, and the sabre-toothed tiger. During the same period the shores of north- west Europe were washed by genial waters, in which lived many molluscs that are now confined to the seas of more southern latitudes. It is in the marine deposits of the Pliocene of East Anglia that we read the evidence, of that gradual deterioration of climate which eventually culminated in the first glacial epoch. In the older Pliocene strata southern forms of life are present in great force, but these gradually become less numerous as we follow them into the overlying beds ; while at the 248 THE HISTORY OF THE PLEISTOCENE PERIOD same time immigrants from boreal regions begin to appear, and continue to increase in number as we approach the later accumulations of the Pliocene sea, until at last the whole molluscan fauna becomes thoroughly arctic. The occasional occurrence of erratics in the earlier Pliocene stralta shows that ice- rafts, or possibly icebergs, floated in the North Sea before its fauna had become completely arctic. The Scanian or First Glacial Epoch The Pleistocene period thus begins with the advent of an arctic climate — the deposits known as the Chillesford Clay and Weybourne Crag of Norfolk representing the Scanian or First Glacial epoch. At this early stage a cold current coming from the Northern Ocean must have passecj southward along our western shores, just as in our day the Labrador Current washes the east coast of Canada and the New England States. Under such conditions perennial snow-fields and glaciers could hardly fail to have flourished in Britain, although (fof reasons that will presently appear) we are unable to point in this country to any morainic accumulations belonging to this earliest glacial epoch. Elsewhere, however, such deposits are met with, as in Scandinavia, northern Germany, and the Alpine lands, from the geographic distribution of which we gather that the development of ice-action at this first glacial stage of the Pleistocene was of very considerable importance. The Scandi- navian peninsula seems at that time to have supported FIRST INTERGLACIAL EPOCH 249 a great ice-sheet that not only occupied the basin of the Baltic, but overflowed Scania — the southern part of Sweden — and extended as far south as Hamburg and Berlin. Again, in the Alpine lands strong evidence is forthcoming to show that at the com- mencement of the Pleistocene period glaciers passed down all the great mountain valleys to the low grounds of the foreland — a state of glaciation implying a depression of the snow-line below its present level of about 4000 feet. This is known as the Giinzian stage of the Alpine lands. It is certain, therefore, that the First Glacial epoch was one of pronounced severity. Owing, however, to succeeding glacial invasions, the records of that earliest cold stage in Britain have either been destroyed or buried under- neath later glacial formations — at all events they have not yet been recognised. The Norfolkian or First Interglacial Epoch The Norfolkian epoch is so termed since its most important records are furnished by the Forest-bed series of Norfolk. These deposits have already been described, and it may be remembered that they consist of fresh-water and estuarine beds which overlie certain marine strata containing arctic shells. The position occupied by the Forest-bed shows that at the beginning of the earliest interglacial stage the sea had retired from the English coast-lands. All the evidence, indeed, leads to the belief that after the disappearance of that cold sea with its arctic 250 THE HISTORY OF THE PLEISTOCENE PERIOD fauna, the southern part of the North Sea basin had become land. Britain was joined' to the Continent, and the Rhine, with the Thames as a tributary, continued its course northwards over regions which have long been submerged. It is not improbable, indeed, that the fresh-water Forest-bed series repre- sents the alluvia of that ancient Rhine. The general facies of the Forest-bed flora implies a temperate climate, somewhat similar to that of the present, but probably a little warmer ; and the testimony of the plants is in keeping with that of the fauna, amongst the more notable animals being hippopotamus, elephants i^Elephas meridionalis, E. antiguus), Etruscan rhinoceros, great beaver, primitive horses, giant fallow deer, bison, and the sabre-toothed tiger. The earliest interglacial fauna thus included several well-known Pliocene types, some of which appear not to have survived to any later date. Traces of the Norfolkian stage are met with here and there on the Continent, as in tBe Low Countries, France, Germany, and Italy. The deposits referred to are recognised chiefly by the character of their mammalian remains, which are essentially the same as those of our Forest-bed series. Certain other alluvia of very early Pleistocene age ought perhaps to be assigned to the Norfolkian interglacial epoch. Of these by far the most notable are the sand-beds at Mauer, near Heidelberg, from which have been obtained the oldest skeletal remains of man hitherto met with on the Continent, This fossil— a lower SECOND GLACIAL EPOCH 251 jaw — was discovered at a depth of seventy-eight feet from the surface, in river-deposits from which have come several Pliocene mammals— namely, Etruscan rhinoceros, two species of bear, a primitive horse, and other forms recorded from our Forest-bed series, such as the straight-tusked elephant {£. antiquus\ bison, great beaver, etc. Dr Schotensack describes the mandible as being very massive and without chin projection — quite ape-like in fact. — while the teeth, he says, are without doubt human. The geological and palaeontological evidence, although not quite decisive, seems to favour the reference of this ancient human type to the First Interglacial or Norfolkian stage. The Saxonian or Second Glacial Epoch The deposits immediately overlying the Norfolk Forest-bed are marine, and indicate therefore a readvance of the sea. At first the temperature of the water was such as to favour a: marine molluscan fauna similar to that now living off the English coast. By and by, however, the temperature fell, another invasion of arctic molluscs followed-, and the sea rose some fifty feet above its present level. Soon there- after, however, the sea again retired from the Norfolk coast-land, for above the arctic shell-beds we encounter a fresh-water deposit, from which leaves of the arctic willow and other northern plants have been obtained, as well as several land and fresh-water shells, The change of climate, therefore, from that 252 THE HISTORY OF THE PLEISTOCENE PERIOD of the Norfolkian interglacial For,est-bed to that of the fresh - water arctic bed was most pronounced. The lowering of the temperature could not have been less than 20° — a change as great as we now experience in passing from the south of England to the North Cape. Next in succession to the arctic fresh-water bed come the morainic accumulations that mark the Saxonian epoch — the epoch of maximum glaciation — when the Scandinavian inland ice invaded the low grounds of Saxony, and the vast glaciers of the Mindelian stage of the Alpine succession piled up huge terminal moraines on the forelands (see Map A). The general conditions that obtained in Europe at that stage, having been described in a previous lecture, need not be re-stated. It may suffice to remind you that the Saxonian glacial epoch is well represented in all the mountain ranges and elevated plateaus of the Continent by moraines and fluvio- glacial gravels, while heaps and sheets of rock-rubble and breccia point to the action of severe climatic conditions at lower levels. Flood-loams were then abundantly distributed over the broad valleys and low-lying tracts outside of the glaciated areas. To the same epoch must be referred those Pleistocene marine beds of Sicily and Italy which are charged with a northern fauna. Obviously at that time a cold current washed the west coasts of Europe. Iceland and the Farce Islands each supported an ice-sheet, and bergs and floes drifted as far south SECOND INTERGLACIAL EPOCH 253 and east as the Azores, on the depressed coast-lands of which they stranded and dropped their erratics, while boreal and even arctic molluscs invaded the Mediterranean. The climax of cold having been reached, mer de glace and glaciers eventually began to retire, and the low grounds of our present temperate latitudes, relieved of their icy covering, were invaded by a tundra vegetation and roamed over by reindeer, musk-ox, mammoth, and woolly rhinoceros. In middle Europe dry steppe-conditions by and by supervened ; but with the continuous retreat of the ice-fields and the improvement of the climate, sand- and dust-storms ceased, and the Tyrolian or Second Interglacial epoch was inaugurated. (See Note ii.) The Tyrolian or Second Inte-rglacial Epoch A temperate flora and fauna now began to clothe and people our Continent — the arctic, northern, and sub-arctic forms retiring up the mountains, and returning northward and eastward to their old homes. When this genial epoch reached its climax, a mild oceanic climate prevailed far into the interior of ■Europe — the snow-fields and glaciers of the Alps being then much less extensive than they are now. Britain at that stage formed part of the Continent, the English Channel and probably a large part of the North Sea being dry land. The equable character of the climate is evidenced by flora and fauna alike. The Pontic alpine rose and its congeners 234 THE HISTORY OF THE PLEISTOCENE PERIOD flourished at elevations in the Alpine lands, where under present conditions they could not exist : while hippopotamus, elephants, rhinoceroses, cervine and bovine animals, and many carnivores ranged over the major portion of Europe. And with this teeming mammalian fauna Palaeolithic ni^n was contempo- raneous. Just as the land in those days extended farther into the Atlantic than is presently the case, so in the south there would seem to have been one or more land-bridges across the Mediterranean, over which the southern forms passed to and fro (see Map B). While these conditions endured, a larger body of warm water seems to have flowed into the North Atlantic. This may be inferred from the fact that southern molluscs then invaded the Mediterranean just as their boreal predecessors had done in the preceding glacial epoch, when colder water washed the western shores of our Continent. It was at the climax of this, genial interglacial epoch that the Chellean culture-stage prevailed. The race that now occupied mfddle and southern Europe, and many wide regions in Asia and Africa, are represented by the most primitive of Palaeolithic artifacts. These rudely fashioned stone implements seldom occur in caves, but are often met with in the older Pleistocene river-drifts. From this it has been inferred that Chellean man probably lived in the open, for the climate was clement and equable, the seasons not being so strongly contrasted as in our day. The margins of the rivers were apparently SECOND INTERGLACIAL EPOCH 255 favourite haunts of the race — the coarse gravels affording an inexhaustible supply of the stones required for implements. The genial conditions of the protracted Tyrolian epoch eventually began slowly to give way. The hippopotamus and its associates, the straight-tusked elephant and the southern elephant, migrated south- ward, and the last-named species, it may be added, never returned. Temperate and northern types now became the most notable denizens of north-west and middle Europe, the mammoth and the woolly .rhinoceros being especially characteristic. Mean- while the Chellean had been succeeded by the Acheulian culture-stage. The records of the Tyrolian stage are met with at many places in Britain — both in caves and river- alluvia. In Victoria Cave, near Settle, for instance, remains of the southern and temperate mammalia occur, as hippopotamus, rhinocero's (not the woolly rhinoceros), straight-tusked elephant, bison, red deer, Tiysena, etc. The beds containing these remains are overlaid by the fluvio-glacial deposits of the succeed- ing Polonian ice-sheet. In certain caves in Wales glacial and fluvio-glacial deposits of the same age similarly overlie accumulations which have yielded remains of the Pleistocene mammalia. Sometimes it is the southern group which is represented, while in other cases northern and temperate forms occur, such as mammoth, woolly rhinoceros, reindeer, and horse. Certain caves have obviously been hysena-dens, 256 THE HISTORY OF THE PLEISTOCENE PERIOD while in several caves Palaeolithic implements have been discovered. These finds apparently pertain to the transition stage between the Tyrolian and the succeeding glacial epoch. Before the close of Tyrolian' times the coast- lands of the Baltic had sunk in the sea, and the low grounds of central and northern Britain had likewise been submerged. A certain amount of depression had also been experienced in the Mediterranean region, for to this stage belong the marine terraces cut in the older breccias of Gibraltar. It is probable that the depression in question affected regions lying further to the east, and that the land-bridges which had formerly connected Europe with Africa were now submerged. The evidence seems to point to the Tyrolian having been the warmest, as it was apparently the most protracted interglacial epoch. This is indicated by the character of the fauna and flora, and by the very considerable geographical changes just referred to — changes which affected both northern and southern Europe. Its prolonged duration is further shown by the profound modifications brought about during the epoch by the slow operation of denudation. This latter is the kind of evidence which appeals most strongly to the geologist — for its meaning cannot be misunderstood. It is just possible that the crustal movements of Tyrolian times may have been more rapidly accomplished than we suppose. But we have np reason to believe THIRD GLACIAL EPOCH 257 that rain and rivers were more active during an interglacial epoch than they are in our own day. We know how slowly they work, and how after the passing of many generations the face of the land remains practically unchanged. We cannot but be astonished, therefore, when we realise how greatly the configuration of the land became modified during the occupation of our Continent ty Chellean man. The whole surface was lowered — in some places by not less than fifty or one hundred feet — while the excavation of many river-valleys was far advanced, and in certain cases well-nigh completed in Chellean times. The Polonian or Third Glacial Epoch The cold conditions that characterised the close of the Tyrolian stage obviously signalled the approach of another glacial epoch — the Third or Polonian. Doubtless the transition from the Tyrolian to the Polonian was very gradual, and it would seem that before the climax of the latter had been reached the men of the Mousterian stage of culture had come to occupy the caVes of north-west, -central, and southern Europe. In England, Belgium, Prance, and Germany he was eventually con- temporaneous, not only with mammoth and woolly rhinoceros, but with reindeer, glutton, arctic fox, and other members of the tundra fauna. He must have witnessed the gradual invasion of the low grounds of north-western and northern Europe by 258 THE HISTORY OP THE PLEISTOCENE PERIOD the third great Scandinavian mer de glace. North Britain and Scandinavia would appear at that time to have been almost as deeply buried in ice as during the Saxonlan epoch, but the mer de glace did not reach the limits attained by its predecessor in England, north Germany, and Russia, The direc- tion of ice-flow differed in some important respects from that of the Saxonian ice-sheet. While the Baltic basin seems to have had relatively little influence upon the course followed by the latter, the direction of the succeeding Polonian ice-sheet was to some extent controlled by that great depression. Similar notable differences in the trend of the two ice-sheets have been elsewhere detected. In southern Scotland, for instance, the track of the later mer de glace did not coincide= with that followed by the earlier and greater ice-sheet, but in several places crossed it at nearly right angles. So again, while the Saxonian ice-sheet flo^ved approximately south across Denmark and the southern part of the Baltic basin into Holland and north Germany, the Polonian ice-sheet traversed thosd regions in a more westerly or south-westerly direction. The Saxonian mer de glace, as already indicated, had reached south in England to- the valley of the Thames, but the succeeding Polonian ice-sheet did not extend beyond the Midlands. We are not without direct evidence, howeveV, that an arctic climate at the same time characterised southern England. The succession of deposits at Hoxne in THIRD GLACIAL EPOCH 259 Suffolk shows quite clearly that such was the case. Here we see certain fresh-water beds, charged with plentiful remains of a well-marked temperate flora, resting directly upon the bottom-moraine of the great Saxonian ice-sheet. These particular beds are in their turn overlaid by a series of loamy deposits which have yielded a characteristically arctic flora. We cannot doubt, therefore, that the under- lying beds with their temperate plants are of interglacial age — belonging probably to a late stage of the Tyrolian epoch — while it may be safely inferred that the superjacent arctic-plant beds are of Polonian age. The latter indicate, in short, that tundra conditions prevailed in the south of England when the northern parts of Britain were shrouded in ice. While this notable recrudescence of glacial conditions was in progress in the north, the snow- fields of the Alps likewise increased in extent, and mighty ice-flows once more filled the mountain valleys and deployed upon the forelands. This third cold epoch of the Alpine lands is known as the " Rissian stage." It is remarkable that th« Rissian glaciers in certain valleys were larger than those which had occupied the same valleys in the preceding Mindelian epoch. In other valleys, however, the reverse was the case, the Mindelian glaciers having attained a greater development than their successors in the later or Rissian epoch. In the valleys of the Inn, the Salzach, and the I Her,, for example, the 260 THE HISTORY OF THE PLEISTOCENE PERIOD Mindelian glaciers were the greatest, but such was not the case with the Isar glacier, which was most extensive in the Rissian epoch. So again in the Rhine region, in Switzerland, in the French Alps, and in the Po valley the Rissian glaciation was the more extensive. Professor Penck thinks it not unlikely that these differences may have been the result of those differential crustal movements which are admitted to have affected the Alps during glacial times. If Switzerland experienced a move- ment of elevation in the interglacial epoch that followed the Mindelian glaciation, the Rissian glaciers in the west and south-west Alps would naturally attain larger dimensions than their predecessors of the earlier glacial epoch. Should subsequent research establish this conclusion, it would follow, according to Professor Penck, that the greatest depression of the snow-line must have taken place in the Mindelian epoch — in other words, we should assign the maximum cold of the Glacial period to that stage. So far as northern Europe is concerned, there is no doubt that the Polonian ice- sheet, enormous as it was, did not occupy so wide an area as its Saxonian predecessor. (See Map C.) The Polonian or Rissian epoch has left its mark lipon all the mountain groups and ranges of middle and southern Europe, in the valleys of which more or less conspicuous terminal moraines of that age appear. The physical conditions' which had char- acterised the non-glaciated tracts of our Continent THIRD GLACIAT, EPOCH 261 in the Saxonian or Mindelian epoch were repeated during the subsequent Polonian or Rissian epoch. The Rock of Gibraltar, for instance, had been re- elevated, and thereafter a new formation of breccia took place — the screes of the period covering the beach deposits of the preceding interglacial stage. In many other extra-glacial regions the accumulation of rock-rubble went on apace, and under the influence of frost and thaw soils and subsoils slipped and flowed. With each recurring summer the turbid rivers rose in flood, and inundated wide districts, and gravel, sand, and silt were deposited in enormous quantities over all low-lying regions. When the rigour of the secular winter began to abate, and the snow-fields and mers de glace to retire from the low grounds, the advance of arctic flora and fauna naturally followed. Tundra conditions, which had characterised wide tracts of middle Europe during the climax of the Polonian or Rissian stage, now spread more and more as ice-sheet and glacier continued their retreat, and fluvio-glacial floods and inundations diminished in importance. The lemming and its congeners occupied what are in our days the most fertile regions of western and central Europe, ■while the marmot, the ibex, and other alpine types began to retire from the low grounds of Italy. Rein- deer, mammoth, woolly rhinoceros, and Mousterian man were now the most notable denizens of our Continent. The Mousterian culture - stage thus appears to have continued throughout the Polonian 262 THE HISTORY OF THE PLEISTOCENE PERIOD or Rissian cold epoch, and, as we shall see presently, flourished to even a later date. Gradually the climate continued to improve, the arctic flora and fauna slowly migrating northward. Britain again formed part of the Continent, and the western shores of Europe seem to have extended considerably farther into the Atlantic than at present. As tundra conditions waned a steppe flora and fauna gradually invaded our Continent from the east. For a long time relatively dry conditions prevailed over the plains and low grounds of middle Europe. The climate being somewhat extreme and resembling that which now characterises south-east Russia and south- west Siberia, the low-lying tracts in question were doubtless subject to dust-storms in summer and to blizzards in winter. But with the continued retreat of the northern ice-sheet and the diminishing extent of the snow-fields of the Alpine lands, steppe-conditions eventually passed away. The Durntenian or Third Interglacial Epoch The climate now steadily became less markedly continental and more humid, forests extending in all directions until they covered vast areas. A true forest-fauna now dominated in middle Europe, as it probably had done for some time in the western and southern regions of the Continent. With these wide- spread forest conditions, the Third Interglacial epoch reached its climax. Remains of the straight-tusked elephant and the broad-nosed rhinoceros — both THIRD INTERGLACIAL EPOCH 263 southern forms — occur in several intergkcial accumu- lations of this age. Thus at Pianico in north Italy and at Flurlingen, near Schaffhausen, the bones of the rhinoceros are met with, while remains of both the rhinoceros and the elephant have been found in the lignite-beds at Durnten, near Zurich. How genial the conditions were in the Alpine lands at this stage is shown by the evidence derived from the cave of Wildkirchli, occurring at a height of over 4800 feet on the Ebenalp, near Santis. The cave has yielded abundant remains of a well-marked forest fauna along with numerous Mousterian artifacts. It is noteworthy that characteristic members of th^ tundra fauna are entirely absent, in place of which we have cave-bear, cave-lion, wolf, ibex, stag, etc., all the species named being found in the caves at Mentone, associated there with straight-tusked elephant, broad-nosed rhinoceros, hippopotamus, and Mousterian artifacts. When Mous- terian man followed the chase from Wildkirchli, at a height of over 5000 feet, the climate could not have been less temperate, but was probably even more genial than the present ; and we need not be surprised, therefore, to find his relics associated with remains of the southern mammalia in the coast-lands of the Mediterranean. Such evidence aS: we have, however, would seem to indicate that the Third was not so warm as the Second Interglacial epoch. (See Note 12.) Some considerable portion of the river-drifts of the Thames valley ought probably to be assigned to the Third Interglacial stage. The oldest beds of the 264 THE HISTORY OF THE PLEISTOCENE PERIOD series, it may be remembered, are characterised by the presence of Chellean artifacts, and the remains of the southern and temperate group of mammals. To these earliest river-drifts succeeds a very- vari- able series of gravel, sand, and loam, from which Mousterian artifacts have been obtained in large numbers. The associated mammalian fauna is somewhat mixed, but now and again remains of the reindeer are quite abundant, while elsewhere mammoth, woolly rhinoceros, and horse are well represented, the evidence of the mammalia pointing on the whole to somewhat cold climatic conditions. It must not be forgotten, however, that the Palaeolithic "floors" described by Mr Worthington Smith as occurring in the high-level river-drifts are likewise of Mousterian age, and are nevertheless characterised by the presence of a well-marked temperate flora. It would thus appear that during the formation of the older river-drifts of the Thames district several changes of climate took place. The Chellean stage we should assign to the Second Interglacial epoch, and the Acheulian to the close of that epoch ; while the later gravels with Mousterian artifacts represent the two succeeding epochs, namely, the Third Glacial and the following Interglacial. The records in the coast-lands of the Baltic of that Third Interglacial epoch consist chiefly of marine deposits — the included shells belonging to species that still live in the western Baltic, the Kattegat, and the North Sea. It would thus appear FOURTH GLACIAL EPOCH 265 that considerable depression affefcted those regions during some part of the interglacial epoch, most probably when temperate conditions were beginning to pass away. Certain fresh-water deposits in East and West Prussia, occurring on the same geological horizon, have yielded remains of mammoth, woolly rhinoceros, urus, horse, and the great Irish deer. The Mecklenburgian or Fourth Glacial Epoch At the commencement of this epoch large glaciers descended all the Scottish mountain valleys, and occupied many of the sea-lochs. The land was less extensive than it is now — the sea, tenanted by an arctic fauna, reaching to 130 feet at least above its present level. Our estuaries werS in winter largely frozen over, while in spring and early summer the ice, broken up into floes, often ran aground in shallow water, and confused and contorted the marine sedi- ments accumulating there. Many erratics were by the same agency distributed over the sea-floor, and a similar transport of erratics and rock-rubbish was effected by the icebergs calved by our Highland glaciers. In the Southern Uplands very considerable ice - streams likewise existed, but none of these succeeded in reaching the sea. An arctic flora, comprising the polar willow {Salix polaris), dwarf birch {Betula nana), mountain diVQns{Dryas octopetala), and other well-marked northern forms, occupied the Lowlands ; while in the fresh-water lakes Apus glacialis (a phyllopod now met with only in Green- 266 THE HISTORY OF THE PLEISTOCENE PERIOD land and Spitzbergen) flourished exceedingly. The mountains of the sister countries likewise nourished considerable snow-fields and glaciens, while the tundra conditions of the Scottish lowlands were continued far into the south of England, where Mr Warren has lately discovered in the youngest series of Pleistocene gravels in the valley of the Lea, an arctic plant-bed. The occurrence of an arctic flora so far south is sufficiently striking, and shows, as Mr Warren remarks, that the land must have been bleak and almost treeless — "with little more considerable in the way of vegeta- tion than the stunted bushes of the arctic willow and the arctic birch, even these being half hidden in the growth of moss." This arctic plant-bed is not the only proof of severe climatic conditions having obtained at so late a period in the south of England. For many years it has been known that the Palseolithic river-drifts of the Thames valley are overlaid by much confused sheets of earthy rubble, which have all the appear- ance presented by the "flowing soils" of Arctic regions. This "contorted drift," as it is called by Mr Worthington Smith, has not only disturbed the upper part of the Pleistocene brick-earths, but has in several places ploughed through them, so as to contort and displace the famous Mousterian "floor." It may be added that we have no direct evidence that Palaeolithic man ever re-visited the district, after that cold epoch had passed away. The conditions on the Continent were not less FOURTH GLACIAL EPOCH 267 severe. The Scandinavian peninsula and Finland were shrouded in ice, and a great iner de glace occupied the basin of the Baltic, and piled up large terminal moraines in Denmark, Schleswig-Holstein, the northern provinces of Germany and Russia. The Scandinavian ice-sheet did not, however, as in earlier times, become confluent with that of Britain, the glaciers of Norway now calving their icebergs at the mouths of the great fiords. Somewhat later the Scandinavian peninsula experienced considerable depression, and a cold sea with an arctic fauna communicated across southern Sweden with the Baltic. (See Map D.) Turning our attention to the Alpine lands, we meet with like evidence of a very considerable re- crudescence of glaciation after the close of the Third Interglacial epoch. That Fourth Glacial stage in the Alpine region is known as the "Wurmian." It is represented by great moraines and abundant fluvio-glacial gravels, indicative of gigantic ice-flows, none of which, however, were quite equal to the still mightier glaciers of Mindelian and Rissian times. The younger archseological stages — the Aurig- nacian, the Solutrean, and the Magdalenian- — are closely related to this epoch, thg mammalian fauna indicating for the two first-named stages a somewhat cool climate (as in the caves at Mentone and else- where), and for the Magdalenian even colder conditions. Probably the two first-named stages 268 THE HISTORY OF THE PLEISTOCENE PERIOD should be assigned to the dawn of the Wiirmian^ — to the period of transition from the preceding interglacial epoch, while the Magdalenian belongs essentially to the succeeding glacial epoch. It would seem that in the Alpine lands the Wiirmian glaciers had commenced to retire before the Magdalenian reindeer hunters appeared in those regions. But for Europe generally we can hardly doubt that the last "reindeer period" coincided with the Wiirmian or Mecklenburgian Glacial epoch — the climax of that epoch being reached when Magdalenian man flourished in southern France and Spain. His artifacts, however, occur in many other countries — in England, Belgium, Germany, Switzerland, Austria, Russian Poland, etc., and it is not likely that all those lands were occupied contemporaneously. It would seem most probable that when glacial conditions were becoming more and more pronounced in Britain and northern Europe, man would gradually abandon these latitudes for the less rigorous climate of France. But wide regions in middle Europe where tundra conditions prevailed were roamed over by the Magdalenian nomads for a prolonged period— their associates being as usual reindeer, mammoth, woolly rhinoceros, and their arctic-alpine congeners. Eventually, as the climate changed and the snow- fields began to diminish, the same regions were invaded from the east by the steppe fauna. But Magdalenian man long continued to follow the chase in the dusty wind-blown tracts of central Europe. THE SUPPOSED "HIATUS" 269 The evidence, however, shows that after a time the northern and steppe animals were gradually replaced by a fauna that betokens the presence of woods and meads. The Fourth Glacial epoch had now passed away, and the Magdalenian hunters had finally disappeared. Much discussion has arisen as to the cause of this apparently sudden disappearance. Some have held, and many still hold, the opinion that a distinct hiatus separates Palaeolithic from Neolithic times. Others again believe that no hiatus exists. They point to certain cave accumulations which they think yield evidence of a transition stage. It is an interesting question, which in the present state of our knowledge cannot be definitely solved. Before we consider the matter, however, it will be necessary to study the geological records of the period that followed upon the close of the Mecklenburgian or Wtirmian stage, which I purpose doing in my concluding lecture. LECTURE X THE HISTORY OF THE PLEISTOCENE PERIOD — Continued Later Pleistocene or Post-Mecklenburgian Deposits of Scotland. Fourth Interglacial or " Lower Forestian " Epoch in North-west Europe. Fifth Glacial or " Lower Turbarian " Epoch. Neolithic Man in Scotland. Fifth Interglacial or " Upper Forestian " Epoch. Sixth Glacial or "Upper Turbarian" Epoch. Passage from Neolithic to Bronze and Iron Ages. Post-Wiirmian Stages in the Alps. Successive Advances and Retreats of Glaciers. Advent of Neolithic Man. Hiatus between Palaeolithic and Neolithic periods, now supposed to be bridged by "Azilian." Reliable and Unreliable Estimates of Geological Time. Conclusion. The records of later Pleistocene times are nowhere better preserved or more readily deciphered than in Scotland, and I shall therefore treat of them first. Before doing so let me remind you that the Mecklenburgian or Fourth Glacial epoch is repre- sented in this country by a series of great moraines and fluvio-glacial gravels — the accumulations of considerable district ice-sheets and large valley- glaciers. Our seas, tenanted by an arctic fauna, then stood some 130 feet or thereabout above their present level, while the lowlands were treeless tundras clothed with arctic plants. The raised beaches or marine terraces of the epoch in question are now 270 LATER PLEISTOCENE OF SCOTLAND 271 conspicuous in our broad lowland valleys, as in those of the Tay, the Forth, arjd the Clyde, and may be easily recognised at many other places round our coasts. Amongst the more prominent formations of later Pleistocene age are certain other Raised Beaches, besides Moraines, Alluvia, and Peat-mosses, the relation of which to the accumulations of the Mecklenburgian stage must be carefully noted. The oldest of these younger raised beaches occurs at a height of forty-five to fifty feet above the present sea-level. Like its predecessor (the loo-foot beach) It is best developed in the great estuarine valleys of the Tay, the Forth, and the Clyde. Usually it assumes the form of well-marked terraces of gravel, sand, clay, and silt ; but on the more open sea-coasts it is now and again represented hy- ledges or benches cut in the solid rock. Most of the shells, etc., it contains belong to still indigenous species. Obviously a considerable interval of time separated the formation of this beach from that of the Mecklenburgian stage. Before the forty-five- to fifty-foot beach began to be formed, the characteristic arctic species of the older beach had disappeared from our coasts. Further, there is evidence to show that after the lOO-foot beach had been lifted out of the water, it was for a lengthy period subjected to severe erosion, especially in our estuarine valleys. Moreover, it is quite clear that this erosion was effected by rain and rivers when the land stood at a relatively higher level than it 272 THE HISTORY OF THE PLEISTOCENE PERIOD does to-day, and at a date long prior to the formation of the forty-five- to fifty-foot beach. In the valleys of the Tay and Earn, for instance,; the accumulations of the lOO-foot beach have been extensively trenched and swept away from broad tracts', so that they now form terraces, the bluffs of which overlook the later carse-deposits of the forty-five- to fifty-foot beach. That the erosion referred to was not the work of the sea in which these younger estuarine beds were formed, is proved by the simple fact that the latter do not rest directly upon the denuded deposits of the loo-foot beach. On the contrary, they are ■separated from these by a widespread sheet of peat, and this is directly underlain by river silt, clay, sand, and gravel. It is clear, then, that the Mecklenburgian stage was accompanied or followed by a change in the relative level of land and sea. The sea retreated to a lower level than the present, while rivers cut their way down through the deposits of the loo-foot beach, and in time formed broad ^alluvial flats which were overlooked on either side by the bluffs and terraces of the denuded shelly clays. By and by these younger " haugh-lands " were overspread with dense vegetation, the general character of which betokens a climate not less temperate than the present, the dominant species of trees being oak, alder, hazel, birch, etc. The old land-surface is well represented in the valleys of the Tay and the Earn by a thick layer of woody peat or "lignite, from which LATER PLEISTOCENE OF SCOTLAND 273 rootlets pass downwards into the underlying fluviatile alluvia. The peat, I may add, occurs at and below the present sea-level, and is directly overlaid by the deposits of the forty-five- to fifty-foot beach, which towards their base are crowded with leaves, branches, twigs, and occasional trunks of the trees just men- tioned. Obviously, therefore, the Scottish area again experienced submergence — the sea invading the land up to a height of forty to fifty feet above the present coast-line. In the lower reaches of the Tay valley the carse-deposits belonging to this epoch consist largely of laminated clay and silt. As they are traced up the valley, however, they gradually become more and more arenaceous, until eventually they merge into ordinary river-alluvia, the materials of which become increasingly coarser as they approach the mountains. Thus it is evident that the passage from Mecklenburgian times was marked by the retreat of the sea and the appearance of a wider land-area than now obtains. And this geographical change was accompanied by a decided amelioration of climate. Subsequently, the land again subsided, and we have striking evidence to show that this movement was accompanied or followed by a relapse to cold conditions. At the head of Loch Torridon, for example, well-formed terminal moraines rest directly upon the forty-five- to fifty-foot beach. To this positive evidence may be added certain negative 274 THE litsrORY OF THE PLiitSTOCENE PERIOD evidence. The forty-five- to fifty-foot beach not in- frequently fails to appear at the heads of certain sea-lochs in the north-west Highlands, although it may be well developed in their lower reaches. Either, therefore, glaciers at that time occupied the heads of the fiords in question and thus pre- vented the formation of a beach ; or the beach, having been formed before the glaciers reached the coast, may have been subsequently destroyed by those glaciers when they advanced into the sea. That Neolithic man lived in Scotland during the formation of this beach is proved by the frequent occurrence in it of his relics. At Perth, for example, a dug-out canoe of pine was met with towards the bottom of the carse-clays ; and similar finds have frequently been recorded from the contemporaneous deposits in the valleys of the Forth and the Clyde. The latest conspicuous raised beach occurs at an average level of twenty-five to thirty feet above the sea. The only shells it has yielded belong to still indigenous species, while here and there, as at Oban and in Oronsay, Neolithic artifacts are associated with it. Nowhere, so far as known, do the deposits of this beach merge inland into fluvio- glacial gravels, nor apparently are they anywhere associated with moraines. Often, however, the beach contains trunks and stools of pine and other trees of large size ; while now and again it directly overlies peat with trunks and stools of trees that are rooted in an underlying soil. It is hard to say, LATER PLEISTOCENE OF SCOTLAND 275 however, whether these ancient land-surfaces may "not, in some cases, at least, occupy the same geological horizon as the woody peat met with in the .Tay valley below the older carse-clays of the forty- live- to fifty-foot level. Postponing for the present any further remarks on the evidence supplied by these younger raised beaches, I would shortly direct attention to certain other accumulations which unquestionably belong to later times than the closing phase of the Mecklen- burgian epoch. I refer to our peat-mosses. Every one knows that these frequently contain the remains of trees. In many places throughout Scotland — as well in high as in low grounds, the peat-mosses cover two forest-beds. Typically, the older forest layer occurs at the base of the peat, while the younger tier of trees rests upon and is covered by a variable thickness of the same material. In some bogs only a few feet may separate the two forest- beds, while elsewhere the intervening peat may attain a thickness of many yards. Long ago I endeavoured to show that the general occurrence of these phenomena was indicative of climatic changes. The forest-beds, I maintained, were the products of relatively dry or continental conditions, while the intervening and overlying sheets of peat were evidence of colder and wetter conditions. I further drew attention to the fact that^our peat-mosses are not now in a flourishing condition, but are all more or less rapidly decaying and being denuded by rain 276 THE HISTORY OF THE PLEISTOCENE PERIOD and wind. Under favourable conditions peat is doubtless forming here and there, but this is exceptional, the rate of growth is much exceeded by the general rate of decay and denudation. From this striking fact I inferred that the climate of Scotland had become drier since the formation of the peat that overlies our "upper forest-bed." The earlier writers on the origin of the Scottish peat did not recognise the influence of climatic changes in the overthrow of the old forests. Accord- ing to them it was the destruction of the forests that had induced the formation of the peat-mosses, the work of destruction being attributed chiefly to man's hand, although it was allowed that other causes, such as tempestuous winds, might have had their share. The wholesale overthrow of the forests, it was believed, had obstructed the natural drainage of the land and thus brought about marshy conditions favourable to the growth of sphagnum and its allies. More recently it has been suggested that in certain cases the drainage may have been interrupted by the heaping up of banks of sand, clay, etc., across broad valleys, whereby the forests ovef wide areas may have been destroyed by stagnant water, and thus eventually replaced by bogs. This is a somewhat far-fetched explanation. If it had any evidence in its favour, that should not be hard to point out. Where, one might ask, are those supposed bars or banks behind which the stagnant water is imagined to have accumulated ? LATER PLEISTOCENE OF SCOTLAND 277 That none of these explanations can be accepted as sufficient to account for the phenomena of our peat bogs in general, is shown by the mere fact that the buried forests are not confined to the peat of low-lying and gently undulating ground — to posi- tions, namely, where the drainage might possibly have been disturbed by one or o^ther of the causes suggested. On the contrary, they occur just as constantly in the peat of mountain-slopes and hill- tops, where, owing to the forrn of the ground, interruptions of the drainage could not possibly take place. Moreover, the almost invariable appearance, in the peat of low and high grounds alike, of two buried forests obviously points to the operation of some widely-acting recurrent cause. Conclusions similar to mine were subsequently advocated by the late Professor Blytt, who, after a careful study of the peat-mosses of Norway, was convinced that these gave evidence of a well-marked alternation of wet and relatively dry climatic conditions having obtained after the low grounds of that country had been vacated by the "inland ice " of the Glacial period. I need only add that the phenomena of successive buried forests have long been recognised as characterising the peat-bogs of northern and north-west Europe. The occurrence of these trees, however, has been variously interpreted, some authors upholdiijg views that are practically the same as those I had previously ventured to set forth, while others would attribute 278 THE HISTORY OF THE PLEISTOCENE PERIOD the origin of the peat-mosses to the overthrow of the forests by the various causes already referred to. When we come to inquire into the relation of our Scottish peat-mosses to the glacial deposits, we have no difficulty in ascertaining that they belong to a later date than the Mecklenburgian stage. This is shown by the fact that the peat with its buried trees overspreads the fluvio-glacial gravels and moraines of that epoch. Hence it would appear that our oldest peat-mosses occupy precisely the same geological horizon as the lignite and alluvia which underlie the carse-clays of the forty-five- to fifty-foot level, and directly overlie the deposits of the lOO-foot beach. We are thus led to the conclusion that the ancient land-surface buried underneath the carse-clays of the forty-five- to fifty-foot beach is contemporaneous with the Lower Forest-zone of our inland peat-mosses. This correlation gives us the key to the history of all the later climatic and geographical changes experienced by our country. Summing up the evidence thus briefly stated, we may recognise the following succession of stages in the later Pleistocene history of Scotland : — The Lower Forestian or Fourth Intp:rglacial Epoch. After the Mecklenburgian glaciers had dis- appeared, the land gained in extent, the sea eventually retreating considerably beyond the present coast-line. FORESTIAN AND TURBARIAN STAGES 279 The climate at the same time gradually improved until genial conditions supervened, and a strong forest growth covered the low grounds and extended upwards to elevations which trees in our country no longer attain. The relics of that great forest epoch we find in the Lower Forest-zone of our peat-mosses. Hence I term it the Lower Forestian. The Lower Turbarian or Fifth Glacial Epoch. Next followed partial subsidence of the land, accompanied ere long by a relapse to cold conditions. Snow-fields now reappeared, and considerable glaciers descended our mountain valleys, and in some places reached the sea. The climate was wet and ungenial, -the forests decayed, and bog-mosses gradually usurped their place. To this stage we assign the Lower Peat of our inland bogs, as also the forty- five- to fifty-foot beach, and certain moraines and fluvio-glacial gravels. The Upper Forestian or Fifth Interglacial Epoch. The succeeding stage was characterised by re- elevation of the land, and the retreat of the sea beyond the present coast-line. But the land was probably not so extensive as during the earlier forest epoch. This geographical change was marked by the disappearance of perennial snow and ice, and by a return to dry genial conditions, apparently similar to 280 THE HISTORY OF THE PLEISTOCENE PERIOD those that formerly obtained. Forests again clothed the land — flourishing in many places over the surface of the now desiccated peat-mosses. This stage is represented by the Upper Forest-zone of our inland peat, and by the trees which occur under the deposits of the twenty-five- to thirty-foot beach. It constitutes the Upper Forestian. The Upper Turbarian or Sixth Glacial Epoch. Once more partial subsidence ensued, and the climate again became somewhat cold and wet. Over wide areas, the forests, as before, began to decay, and were eventually buried under the continually extending peat-mosses. We cannot actually demonstrate that snow-fields and glaciers reappeared at this stage. The latest beach we are able to correlate with the Upper Peat, but that beach is nowhere associated with moraines or glacial gravels. Nevertheless we are not without evidence suggestive of the' appearance at this time of inconsiderable glaciers amongst our highest mountains. These small glaciers undoubtedly existed at a later date than the glaciers that dropped their moraines on the forty-five- to fifty- foot beach. It is therefore only reasonable to infer that the high-level corrie-glaciers in question were probably contempo- raneous with the formation of the twenty-five- to thirty- foot beach and the Upper Peat of our inland bogs. But the chief evidence of cold wet conditions is un- questionably that furnished by the Upper Turbarian RECENT AND PRESENT EPOCH 281 itself. It covers the Upper Forestian in precisely the same way as the Lower Turbarian overlies the Lower Forestian. The Recent and Present Epoch. The final stage witnessed the retreat of the sea to its present level. The climate now became drier, and peat-mosses ceased to flourish as they had done in the immediately preceding epoch. Thus the final phase of Pleistocene history may be said to be characterised especially by the general decay and denudation of our peat-mosses, the vegetation growing upon which is almost invariably of a drier type than that found in the peat itself. Did time permit I might follow other lines of evidence, all leading to the conclusion that oscillations of climate marked the closing stages of Pleistocene times. For example, the phenomena presented by the alluvial terraces of our larger river valleys might be cited. It would not be hard to show that during the period in question our rivers have at some stages been most active as eroding agents, while at other stages their chief work has been the transportation and deposition of sediment. During the genial epochs, when the land stood at a higher level than now, the rivers busied themselves especially in deepening and widening their courses, in sweeping away the glacial and fluvio-glacial detritus with which their valleys had been so largely choked. During the cold wet epochs the land was depressed below its present level, 282 THE HISTORY OF THE PLEISTOCENE PERIOD and the larger rivers were then chiefly engaged in fining up the lower reaches of their valleys with abundant sediment. In a word, epochs of dominant erosion alternated with epochs of dominant deposition. In support of my general conclusions, I might also appeal to certain facts relating to the present and past geographical distribution of animals and plants. The appearance, for example, in the North Atlantic of isolated colonies of southern types of molluscs, surrounded on all sides by boreal and cold-temperate forms, and the occurrence now and again of similar no longer indigenous molluscs in the raised beaches of Nova Scotia, Greenland, Spitzbergen, and Scandinavia^ are alike strongly suggestive of warmer conditions having at a very recent period characterised the North Atlantic. Quite in keeping with these phenomena is the fact that the beaches in question are often crowded with southern types which, although still lingering on in these northern seas, do not now attain so large a size as their predecessors of late Pleistocene times, while they are obviously much less abundant. But the discussion of these and other lines of evidence (especially those afforded by the geographical distribution of plants in temperate Europe) would call for more space than is at my disposal. Although the more obvious pr„oofs of alternating genial and ungenial climates supplied by our peat- mosses seem to me too strong to be resisted, fortified as they are by the evidence yielded by our raised PLANT REMAINS OF SCOTTISH PEAT 283 beaches and recent morainic accumulations, I have yet long been of opinion that they would probably be strengthened by a systematic botanical examination of the peat itself. I could not don,bt that a study of the several layers of vegetable matter of which peat- mosses are composed would throw light on the particular character of the climate that obtained during their accumulation. It was obvious to me, although no expert in botany, that bog-mosses were not the only constituents of peat : in many sections I could see what appeared to be a succession of layers made up of the remains of different kinds of plants. And often have I regretted my inability to interpret what that succession meant ; for I felt sure it had some interesting tale to tell. Fortunately the work of interpretation was at last undertaken by an accomplished botanist — Dr Francis J. Lewis. His researches were carried on for a- number of years, during which he made a careful botanical survey of the peat-mosses of our country, the results of which he brought before our Royal Society in a series of masterly monographs. Space will not allow me to give more than a meagre outline of these results. Suffice it to say that Dr Lewis has demonstrated that our peat-mosses show a definite succession of well- marked stages, corresponding to those I had already established on purely geological evidence. In illus- tration of the results obtained by= Dr Lewis, I must content myself with a brief rdsumd of such of his observations as help us to understand the oscillating 284 THE HISTORY OF THE PLEISTOCENE PERIOD climatic conditions of late Pleistocene times. It may- suffice to cite from his account of the peat-mosses of the Southern Uplands and the Highlands. The Southern Uplands is the general term applied to that belt of hilly and mountainous country which extends from the coasts of South Ayrshire and Wigtownshire to the high grounds that terminate on the East Coast between the valleys of the Tweed and the Tyne. Throughout this= broad tract peat- mosses abound, large areas of the higher ground being of a dominantly moorland character. Nowhere are the peat-mosses better developed than in the mountainous district of Merrick in Galloway, and the lofty region in which the river Tweed takes its rise. In these two typical areas the peat-mosses bear the same relation to the glacial and fluvio-glacial deposits —they everywhere overlie the rnoraines and morainic detritus of the Mecklenburgian stage. According to Dr Lewis, the first well-defined zone at the base of the peat in the Southern Uplands is a solid layer of the remains of white birch (Betula alba), mixed with such plants as heather {Calluna vulgaris) and willow [Salix repens). Dr Lewis thinks it is hardly possible that this zone represents the primitive vegetation which covered the Uplands immediately after the disappear- ance of glacial conditions. The first comers would naturally be arctic types, the preservation of which, however, would entirely depend upon climatic and local conditions. In the Merrick district Dr Lewis observed that in many places a thin layer of peat PLANT REMAINS OF SCOTTISH PEAT 285 occurred immediately underneath the birch zone ; but unfortunately the material was in too decomposed a condition to allow of the identification of any particular plants. This structureless peat might, as he thought at the time, represent the primitive vegetation of the district — a view which has been strengthened by his later researches in other parts of Scotland, where a definite layer of arctic-alpine plants occurs at the very bottom of the oldest peat-mosses, resting upon morainic materials (Mecklenburgian) and covered by the Lower Forest-bed. The birch zone at the bottom of the peat-bogs in the Southern Uplands is directly overlaid by a thick stratum of peat, composed entirely of bog-moss {Sphagnum). This succession is constantly repeated throughout the region, alike in the peat at the bottoms of valleys, and in that upon steep hill-sides and flat hill-tops. The Sphagnum-bed thus bears witness to a general increase of precipitation ; it represents, in short, a change from birch-forest conditions to wet moorland. As successive layers of the peat are followed upwards, Dr Lewis finds that the bog-moss gradually gives place to cotton-grass {Erzophorum vaginatum) and rushes {Scirpus). After these plants had flourished for some considerably time, a decided change of climate supervened. In the Merrick district the cotton-grass peat is covered by a dense layer of the stems of crowberry {Empeirum nigruiti) and two characteristic arctic willo.ws {Salix herbacea 286 THE HISTORY OF THE PLEISTOCENE PERIOD and 5. reticulata). The same zone is represented in the peat of Tweedsmuir by the crowberry and the creeping azalea {^Loiseleuria procumbens), the latter being a typical arctic form. The constant appearance of this remarkable zone throughout the Southern Uplands can have only one meaning — it points unmistakably to a decided decrease of tempera- ture. It indicates a stage during which the valleys of southern Scotland were characterised by a climate as rigorous as that now experienced on the summits of our loftiest mountains. The gradual dying away of this cold epoch and the reappearance of forest vegetation are, according to Dr Lewis, faithfully chronicled in the peat. The crowberry, the arctic willows, and the creeping azalea give place above to cotton-grass, and this in its turn to bog-moss or sphagnum- — a succession common to the peat throughout the Southern Uplands. Eventually the wet moorland conditions indicated by the sphagnum-peat passed away ; the bogs dried up and were invaded by trees, by forests of pine in the Merrick district, and forests of white birch in Tweedsmuir. Finally, the conditions again became adverse to forest-growth, and the trees of this upper zone were gradually buried under a stratum of peat, consisting chiefly of rushes, bog-moss, and cotton-grass. In comparing the peat-mosses of the Southern Uplands with those of the Northern Highlands, Dr Lewis finds that the latter, in many places, begin PLANT REMAINS OP SCOTTISH PEAT 287 their history at a later stage than the former. Thus at high levels none of the beds underlying the second zone of arctic plants in the Southern Uplands puts in an appearance. The reason for this is obvious. The recurrence of cold conditions, indicated by the arctic plants of the Southern Uplands was more strongly marked in the Northern Highlands. In those elevated regions considerable snow-fields and glaciers reappeared, and all peat-beds representing the Lower Forest-zone of southern Scotland were swept away. As these glaciers in the north began to retreat, a vegetation like that of the tundras invaded the formerly glaciated tracts. Arctic willows {Salix reticulata and S. herbaced) at first were dominant forms, but these gradually gave place to sub-arctic types (Salix arbuscula, Betula nana, EmpetruTn nigrum, etc.). By and by this sub-arctic brushwood disappeared, and was succeeded by a close growth of cotton-grass and bog-moss, interspersed with some scraggy birch. That humid conditions eventually became intensified is suggested by the fact that the birch in its turn vanished, and sphagnum alone continued for a long time to occupy the ground. These wet moorland conditions next passed away, the thick sphagnum-peat drying up, and eventually supporting a forest of large pines with an under- growth of common heather. The great pines of this upper zone, it may be mentioned, flourished at elevations between 2000 and 3 500 feet above the present sea-level. Finally they decayed and were 288 THE HISTORY OF THE PLEISTOCENE PERIOD gradually buried under peat, consisting chiefly of bog-moss and rushes. At lower levels in the Highland area the series is more complete — the tale told by the peat-bogs beginning with the first Arctic plant-bed and Lower Forest-bed, after which comes the Lower Turbarian with the second Arctic plant-bed, followed in succession by the Upper Forest-bed and the Upper Turbarian. It may be noted further that iri Shetland the first Arctic plant-bed and the Lower Forest-bed make their appearance, while the Upper Forest-bed is absent. My limits will not allow me to enter into other interesting evidence adduced by Dr Lewis. I may mention, however, that he agrees with me that existing conditions no longer favour the general growth of peat. On hill-top, hillside, and in upland valleys alike, the peat, he says, is almost without exception being rapidly wasted away. The vegetation at present covering the peat is nearly always of a drier type than that occurring at slightly greater depths — a fact, he remarks, not without its bearing upon the present denuded state of the "mosses." Before leaving Dr Lewis's work, I may just in a word point out how strongly the conditions of Forestian and Turbarian times contrast with the present. The upper limits reached in our day by pine and birch may be taken roughly at 2000 feet, although in many districts it lies considerably lower than this. But, according to Dr Lewis, these trees GEOGRAPHICAL CHANGES 289 flourished in Upper Forestian tifties to close upon 3500 feet. The upper Hmits of the trees of the Lower Forestian stage cannot be determined, since at the higher elevations of the land the local glaciers of the Lower Turbarian epoch appear to have destroyed all records of the Lower Forest-bed. As regards the range of the arctic-alpine flora we note that at the present time the vegetation in question is confined to heights at and above 2000 feet. But in the Turbarian epochs it descended to within 150 feet of the sea-level. These are striking contrasts vdiich can only be accounted for by considerable climatic changes. In concluding this brief summary of the climatic vicissitudes of late Pleistocene timjes, I may shortly indicate some of the more important geographical changes that marked the passage of those times throughout north-west Europe. The Lower Forestian epoch — one of genial conditions — was ushered in by what appears to have been a widely extended move- ment of elevation. The British Islands were not only united to themselves but to the Continent, while the uplift of the Scandinavian lands converted the Baltic Sea into a great fresh-water lake, the old shores of which are readily traced. The succeeding Lower Turbarian epoch was marked by depression, the British area being separated from the Continent. At the same time considerable subsidence affected the Scandinavian lands, and the Baltic once more became connected with the North Sea. Temperate conditions continued for a time, but eventually the climate 290 THE HISTORY OF THE PLEISTOCENE PERIOD deteriorated, becoming wet and cold. The climax of this epoch was attained when considerable glaciers here and there reached the sea in the north-west Highlands, and an arctic-alpine flora grew down to nearly sea-level. In Upper Forestian times re- elevation of the land ensued, but to what extent is not known. Certainly the British Coasts advanced on all sides further than at present, but there is no reason to suppose that our area again became continental. The climate was genial, the forests clothing our Highlands up to a height of 3500 feet — that is, 1500 feet above the present limit reached by pine and birch. The Upper Turbarian epoch, on the other hand, was characterised by cold and wet conditions, during the prevalence of which the horizontal and vertical range of the forests becam.e much restricted. The climate, in a word, appears to have been much the same as during Lower Turbarian times, as may be inferred from the fact that the arctic-alpine flora descended nearly to sea-level. But if the small corrie glaciers of our Highland mountains be referable to this stage, as is most probable, this would seem to indicate less severe conditions than obtained when the valley-glaciers of the Lower Turbarian stage reached the sea and dropped their moraines on the forty- to forty-five-foot beach. The present or closing phase of the lon^ cycle of changes is marked by a higher temperature and drier con- ditions than those of the Upper Turbarian. Having now outlined the history of late Pleistocene « RUCKZUGSTADIEN " 291 times in north-west Europe, typically represented as I believe that history to be by the superficial accumula- tions of our own country, I would next direct your attention to the contemporaneous records of the Alpine lands. Let me remind you that the Fourth Glacial epoch of those regions is that known as the Wiirmian, which corresponds to the Mecklenburgian of northern Europe. The passing away of the Wiirmian stage is indicated by a succession of large moraines and associated river-gravels, which seem to show that the final retreat of the last great glaciers was interrupted by at least thre.e long pauses or " Rtickzugstadien." The moraines in question are separated from one another by what are termed " interstadial " deposits, which in character and position simulate interglacial accumulations. Pro- fessors Penck and Bruckner (who are our chief authorities on the glacial history of the Alps) are very guarded in their interpretation of these phenomena, but are clearly of opinion that the moraines indicate successive advances of the glaciers, each advance having been preceded by a retreat of unknown extent. Named from places at which they are typically displayed, the three series (beginning with the oldest) are known as the Blihlstadium, the Gschnitzstadium, and the Daunstadium. From many observations we learn that during the earliest of these stadia the average height attained by the snow-line was 900 metres below its present level, while in the subsequent stadia it rose successively — reaching in the Gschnifz- 292 THE HISTORY OF THE PLEISTOCENE PERIOD stadium a height of 600 metres below the existing snow-line, and in the Daunstadium rising some 300 metres higher. The interstadial stages imply long periods of time, during which the glaciers retired up their valleys for considerable, if indeterminate, distances. So far, therefore, they are comparable to the interglacial epochs. In like manner, the moraines of the Buhl, Gschnitz, and Daun stadia are comparable to the similar moraines of the preceding glacial epochs, since each series of moraines, old and young alike, indicates a distinct readvance of the glaciers. Penck and Bruckner fully recognise all this, and are even willing to admit that certain interstadial accumula- tions may eventually come to be recognised as of interglacial importance. In general, however, the evidence is not decisive — the "interstadial" deposits are wanting in any clear indications of true inter- glacial conditions. Their plant-rernains have not yet been exhaustively studied, and uptil this work has been done, it is considered safer to look upon the deposits in question as indicating less important climatic changes than the interglacial accumulations of earlier Pleistocene times. It is to be hoped, therefore, that the plant-bearing beds will ere long receive the careful attention of competent botanists. But if the closing stages of the Pleistocene period in northern Europe were characterised by several well-marked climatic changes, we can hardly doubt that similar oscillations occurred contemporaneously THE SUPPOSED "HIAITUS" 293 in the Alpine lands. (See Note i^.) And the stadial and inter-stadial phases recognised by Penck and Bruckner would thus seem to represent our Lower and Upper Forestian and Turbarian stages. According to Penck, the Biihlstadium marks a temporary re- advance of the great Wlirmian glaciers, and it is interesting to learn that this readvance took place in Magdalenian times. The sucfceeding Gschnitz and Daun stadia may not improbably represent our Lower and Upper Turbarian stages. It would appear, however, that in the Alpine lands the earliest traces of Neolithic man are of later date than the Daun stadium. Thus the interval between the Blihl a,nd Daun stadia represents, according to Penck, the "hiatus" between the Palseolithic and Neolithic ages. It is never quite safe, however, to rely on negative evidence. Neolithic man was certainly living in Scotland towards the close of the Lower Forestian or the beginning of the Lower Turbarian epoch, and by that time, therefore, he must already have spread over a large part of Europe. So far as we have geological evidence to guide us, it would seem as if his advent did not take place until all the extinct and exotic mammals of Pleistocene times had disappeared. The very earliest Neolithic relics are associated with the remains of a forest and field mammalian fauna, which we recognise as being essentially the same as that now occupying Europe. The last representatives of the Paleeolithic period, on the other hand, were the reindeer hunters of 294 THE HISTORY OF THE PLEISTOCENE PERIOD Magdalenian times, who had as their congeners the tundra and steppe groups to which I have so frequently referred. In short, Palaeolithic man closed his career in Europe under dry climatic conditions, at a time when vast steppes extended through the heart of the Continent, while Neolithic man did not begin his career until these conditions had passed or were passing away and great forests had invaded the former steppe-lands. It goes without saying that profound changes of climate do not take place in the twipkling of an eye. The passage from the steppe conditions of late Magdalenian times to the temperate forest-climate which awaited the arrival of Neolithic man, implies most probably a lapse of several thousand years. Are there any geological records of the human race having lived in Europe during that long interval ? Up till a few years ago that question could only be answered in the negative. Wherever accumulations of Neolithic age had been found resting upon deposits containing relics and remains of Palseolithic man, a very distinct break always separated the two. Caves, river- and lake-alluvia — all told the same tale of a hiatus in the history of our race. Recently, however, certain discoveries have served to some extent to fill up that apparent blank. The first of these discoveries was made by a well-known French expert, the late M. Piette, near Mas d'Azil, a little town at the foot of the Pyrenees, distant about forty V miles south-south-west from Toujouse. Here the THE AZILIAN STAGE 295 river Arize winds for a quarter of a mile through a lofty natural tunnel. The road from St Girons to Carcassonne traverses the tunnel, which is thus readily accessible. For a long time it has- been known that accumulations of Magdalenian age are present in the cave and in the side-galleries opening into it, but M. Piette's researches demonstra.ted the occurrence above these Magdalenian beds of younger deposits, representing a culture-stage hitherto unrecognised. The accumulations dug through by him were over twenty feet in thickness. Towards the bottom of his cutting two culture-beds of Magdalenian age were encountered, the one separated from the other by five feet of laminated river-sand. Both culture- beds contained the usual artifacts ; but it was noted that while reindeer remains were very plentiful in the lower bed, and scarce in the upper bed, the remains of red deer abounded in the latter. Over- lying this upper bed came four feet of laminated fluviatile silt, and immediately above these river- deposits appeared certain layers which, according to Piette and archaeologists generally, represent the transition from Magdalenian to Neolithic times. The lower layer of these transition beds was termed "Azilian" by Piette. The various implements it contained showed no trace of the fine work so characteristic of the artifacts occur/ing in the under- lying Magdalenian culture-beds. The beautifully fashioned spear-heads, slender needles, engraved and carved objects of the pure Magdalenian culture were 296 THE HISTORY OF THE ri.KlSTOCENE PERIOD entirely wanting — almost the only {ijiplements of bone and horn met with in the Azilian consisting' of a few of the commonest and simplest fqrms, such as pins and polished awls or pointers. The most notable Azilian artifacts, however, are the barbed harpoons, which were present in large numbers, Piette having collected over a thousand specimens. These are flat-shaped and made exclusively of stag-horn, and thus differ from the Magdalenian type which was cylindrical and fashioned of reindeer-horn. This is readily accounted for by the nature of the materials operated upon. Stag-horn is internally of a looser or spongier texture than reindeer-horn, and con- sequently it was only the outer compact rind of the former that could be utilised for harpoons. Most of the Azilian harpoons are perforated at the butt end with a round or oval hole, which is not the case with the older Magdalenian forms. Personal decoration is suggested by shells and the teeth of stag, wild boar, and bear, which are perforated as if for the purpose of being strung and used as necklets. Besides these were found a large: number of water- worn pebbles or gravel-stones, painted in red and brown with various designs, sudh as round dots, simple lines and bands, zigzag fines, crosses, and devices that have a strange resemblance to letters. What these curious markings sigriffy no one can say — although many guesses have beeh made. The mammalian fauna of the Azilian is entirely modern — not a trace of the reindeer or its arctic- THE SO-CALLED "ARISIAN" STAGE 297 alpine associates being met with. The most abundant remains are those of the red de'er ; other species represented being roe-deer, brown bear, wild boar, badger, wild-cat, and beaver, besides various birds. The Azilian layer passed upwards into an immedi- ately overlying "transition bed" or "Arisian," as it has been termed by Piette. This culture-bed contained a large quantity of land - shells — Helix nemoralis, the brown-lipped snail^ — and remains of horse, ox, stag, and wild boar. Artifacts of stone, bone, and horn occurred, but stag-horn harpoons were rare : the most notable relics, however, were fragments of pottery. From the great abundance of the brown-lipped snail, which no longer lives in the neighbourhood, it has been inferred that the climate was formerly moister than is now the case. The so-called Arisian stage is in its turn overlaid by deposits containing relics of later Neolithic, Bronze, Iron, and Romano-Gallic periods, and do not therefore call for further notice. From the evidence adduced by Piette, we can have no hesitation in assigning his Azilian either to the close of the Palseolithic or to the dawn of the Neolithic age. The reindeer, which had apparently become scarce in late Magdalenian times, had now vanished, and the true forest fauna had taken its place in the Pyrenean foreland. Of this we are well assured from the appearance of the Azilian in many other caves, A number of these caves occur, like that of Mas d'Azil, at the base of the 298 THE HISTORY OF THE PLEISTOCENE PERIOD Pyrenees, and one has been discovered in the Spanish province of Santander. Deposits of the same age are met with further north in several of the Departments of F" ranee, as in Landes, Dordogne, Drome, and Aisne. The AziUan has Hkewise been detected in Germany, as in caves near I stein and Klelnkems (Baden), where the associated mammalian remains indicate a forest fauna. Other localities in Germany where it has been recorded are the cave of Ofnet, near Nordlingen (Bavaria), and certain caves in Saxe-Meiningen and Westphalia. It has been suggested that the human relics found in the MacArthur Cave, near Oban, may possibly be of the same age. The inference is based on the remarkable similarity of the bone and horn artifacts of the Scottish cave to those of Mas d'Azil — the barbed staghorn harpoons especially closely resemble those of the French Azilian. This is a notable coincidence no doubt, but can hardly be accepted as a proof of contemporaneity. The geological evidence, in point of fact, demonstrates that the Oban reHcs cannot belong to any earlier date than our Upper Turbarian, while the Azilian must be assigned to the Lower Forestian. The interval between these two epochs thus represents a prolonged period, during which two very considerable changes of climate supervened. (See Note 14.) From what I have said it will be gathered that the Azilian is at present not known to occur in Europe north of the 51st parallel of latitude. In ANTIQUITY OF MAN 299 the many caves of England, Belgium, and middle Europe generally, which have yielded both Palaeolithic and Neolithic deposits, the latteV are invariably sharply differentiated from the former— there is a decided unconformity or hiatus between the accumula- tions of the two periods. For the present we can only speculate as to what this means. The evidence of the French Azilian shows that Magdalenian man disappeared while the arctic-alpine fauna was gradually being replaced by the forest-fauna. And the same conclusion, as we have seen, is suggested by the testimony of the Schweizersbild rock-shelter. It is quite possible that Palaeolithic man and his Neolithic successor may have come into contact in southern Europe — or, as some suppose, the two may be the same race. On the other hand, it has been suggested that Neolithic man may have been evolved in some extra-European region under a genial climate, and only entered our Continent when the conditions suited his needs. Whether he absorbed or killed out the earlier occupants of the land, or whether they had previously emigrated with the gradually retreating northern fauna we cannot tell, and may never be able to discover. In the course of these lectures I have now and again referred to certain phenorhena, the obvious explanation of which implies the great antiquity of Palaeolithic man. Indeed, I shali have spoken to little purpose if the evidence I have laid before you has not prepared you to admit that the history of 300 THE HISTORY OF THE PLEISTOCENE PERIOD the human race in Europe must embrace a vast period. When we reflect on the many geographical changes that man has witnessed — the submergence and re-elevation of enormous tracts — the erosion of valleys and general lowering of the surface by denudation ; when we consider that he has lived through a succession of stupendous climatic revolu- tions ; that he has seen widely contrasted floras and faunas alternately occupying our Continent — tundras, steppes, and great forests succeeding each other again and again — we must feel convinced that the few thousand years that have elapsed since the downfall of Babylonian, Assyrian, and Egyptian empires are as nothing compared with the long eons that separate the earliest times of history from the apparition of Palaeolithic man in Europe. As already indicated, there have been many attempts to form an approximate estimate of the duration of the Pleistocene as of other geological periods. No geologist has overmuch confidence in such estimates, but, as I have s^id, they serve to give some precision to our conception of geological time. And this is especially the case when the computations are based on definite data by experts, who have a competent knowledge of the records of the period involved. It is not altogether improbable, indeed, that we may yet be able to arrive at a reliable conclusion as to the number of centuries that have elapsed since the beginning of the Ice Age. There can be little doubt that the great cycle of ANTIQUITY OF MAN 301 climatic changes of which I have been speaking owed their origin to cosmic causes, and that some day astronomers and physicists* may succeed in discovering what these causes were and how they worked. In the absence of an exact chronological basis, such as astronomy may yet supply, the absolute duration of the Pleistocene cann.ot be determined. But, as Professor Penck has shown, a careful estimate of the amount of geological work done during the several interglacial epochs may enable us to form some conception of the time involved. If it be justifiable to infer that the greatest amount of work required the longest time for its accomplish- ment, then it would appear that the genial interval between the Mindel and the Riss glaciations greatly exceeded in duration the preceding and succeeding interglacial epochs. On various grounds the Wiirm glaciation is conjectured by Penck to have reached its maximum about 20,000 years ago. The evidence goes to show that the immediately preceding Riss- Wurm Interglacial epoch was not less than three times as long as the post-Wurmian epoch, while the Mindel -Riss Interglacial stage was twelve times longer than that epoch. Thus we; have 20,000 years for the duration of post-Wurmian times, 60,000 years for the Third or Riss- Wiirm Interglacial epoch, and 240,000 years for the Second or Mindel-Riss Inter- glacial epoch. The data for determining the duration of the '. First or Gunz-Mindel Interglacial 302 THE HISTORY OF THE PLEISTOCENE PERIOD epoch are not so ample — all the evidence, however, leads to the belief that while not so long as the second, it was much longer than the Third Interglacial epoch. We may provisionally assume its duration to have been about 100,000 years, and we thus obtain 400,000 years for the first three interglacial epochs : to which we may add 20,000 j'ears to cover the interstadial stages of post-Wiirmian times. For various reasons it is more difificult to estimate the duration of the cold or glacial epochs. The extent of their respective moraines and fluvio-glacial gravels lead to the inference that the Mindel and Riss epochs were longer than the Wurm epoch. But there are so many other considerations to be kept in view that such inferences cannot be advanced with any confidence. It is further quite impossible to say whether the duration of the several glacial epochs bore any relation to that of the interglacial epochs. Judging from the amount of glacial erosion, however, we can hardly doubt that the duration of the several glacial and stadial epochs combined must have largely exceeded 200,000 years. That nuniber is of course a mere guess, and probably, as I have said, much under the truth. It is only a third, indeed, of the time estimated by Dr Penck for two interglacial epochs. But taking his figures and adding those I have just given, we have a minimum period of 620,000 years for the duration qf Pleistocene times. Quite recently Professor Penck has expressed the ANTIQUITY OF MAN 803 opinion that the Glacial period with all its climatic changes may have extended over half a million to a million years, and as the Chellean stage dates back to at least the middle of the period, this would give somewhere between 250,000 and 500,000 years for the antiquity of man in Europe. But if, as recent discoveries would seem to indicate, man was an occu- pant of our Continent during th^ First Interglacial epoch, if not in still earlier times, we may be compelled greatly to increase our estimate of his antiquity. I have referred specially to Penck's opinion for the simple reason that it is based upon a profound knowledge of the geological data, which are best displayed in a mountainous region like the Alps, where all the evidence of glacial ansd fluviatile erosion and accumulation can be studied more satisfactorily than elsewhere. I might have cited other computa- tions of the duration of the Pleistocene period, most of which are under Penck's estimate' — the amount of time allowed by them varying from 100,000 to 400,000 years. But as these computations are based on the assumption that the Pleistocene occupied only one-thirtieth, or it may have been one-tenth, of the time required for the formation of all the preceding Tertiary systems, they are obviously unreliable. It is no doubt true that Pleistocene time was merely a fraction of the extensive period covered by Tertiary history, which has been variously estimated as equal to 3,000,000 or even 6,000,000 years, and it may quite well have been longer. The absolute 304 THE HISTORY OF THE PLEISTOCENE PERIOD duration of the Tertiary is in short quite conjectural, and therefore any estimate of the fraction of that period required for Pleistocene history cannot be seriously considered. In concluding I would again remind you of what was said In my first lecture as to geological measure- ments of time. In the very nature of things they cannot be definite. Penck's estimate makes no pretension to be other than a rough approximation ; but as It is based on actual observation of the amount of geological work accomplished during the Pleistocene it is on that account justifiable, and undoubtedly helps us to realise how greatly extended are the periods of time embraced by glacial history. So far removed from us are early Pleistocene times that our sympathetic Interest In the life of Palseolithic man cannot but be fg.Int. And yet he was very human : doubtless at the outset of his career a beastially selfish and merciless savage, but gradually developing finer traits with the passing of the ages. It Is not without emotion that we look at the beautiful art-work of the Magdalenlan reindeer hunter. And when we remember the conditions under which he lived — exposed to a severe climate and the attacks of many formidable wild beasts, his home a dark cave or rude rock-shelter — we may well be astonished at his attainments as an engraver, a sculptor, and an animal painter. With the simplest of tools and appliances his best efforts rival, if they do not some- times excel, those of our modern art-schools, and ANTIQUITY OF MAN 305 must ever be a marvel to critics who may have nourished the belief that such attainments are only possible in a civilised community. Neither can we contemplate without a certain reverential awe the profound antiquity of our race, nor cease to wonder at the preservation of the humble records of Palaeolithic man. When we think of the countless multitudes of unknown nameless folk — of the innumerable generations of primitive peoples passed into oblivion ages beforei the dawn of the earliest civilisations known to 'history — we recall the words of the grand old physician of Norwich : "The number of the dead long exceedeth all that shall live. The night of time far surpasseth the day, and who knoweth when was the Equinox." And yet of all those primitive peoples, those vanished races, we should have remained absolutely ignorant but for the endurance of their lost and discarded relics. " Time which antiquates antiquities, and hath an art to make dust of all things', hath yet spared these minor monuments. In vain we hope to be known by open and visible conservatories, when to be unknown was the means of their preservation, and obscurity their protection." u NOTES I. Pre-Pal:EOlithic Artifacts. Sir E. Ray Lancaster has since described and figured the supposed artifacts discovered by Mr Moir in the PHocene of East Anglia {Philosophical Transactions, 1912, Ser. B., p. 283). See papers by Mr Moir in Proceedings of the Prehistoric Society of East Anglia, vol. i., part i ; Bedrock, 1913, p. 165. The same author has recorded the occurrence of flint implements from the glacial deposits of Suffolk {Proc. Prehist. Soc. East- Anglia, vol. i., part 3; Bedrock, 1914, p. 490). This last paper gives good illustrations of the several kinds of " pre-Palceolithic implements" discovered by Mr Moir. His opinion as to their human Origin has not met with general acceptance, and archaeologists will doubtless engage in much discussion before the matter is definitely settled. 2. Southern and Temperate Faunas and Floras. The short account of these is abridged from the descriptions given in Prehistoric Europe, Chaps. II. and III., where references to original sources of information will be found. 3. Tundra and Steppe Faunas and Floras. For my sketch of tundras and steppes I have made free use of a paper " On the Tundras and Steppes Qf Prehistoric Europe " (Scottish Geographical Magazine, vol. xiv.), the details in which are derived in large measure from Professor Nehring's well-known work, Ueber Tundren und Steppen der Jetzt und Vorzeit, 1890. 4. Culture-Stages of Pleistocene Times. One of the best introductions to the sthdy of prehistoric relics is MM. G. and A. de Mortillet's Musee Prehistorique, 1903. This 308 NOTES work contains a large series of plates illustrative of the artifacts characteristic of the successive culture-stages. The English reader may refer to Dr Munro's PalcBoUthic Man and Terramara Settle- ments in JEurope, 1912, Chap. II., for a summary account of these culture-stages, and to Chaps. V.-VIII. and Chap. X. of the same work for descriptions of the several types of " fossil man " discovered in our Continent. See also Dr Duckworth's excellent little treatise, Prehistoric Man, 1912. The Piltdown skull mentioned on page 47 is described in a paper by Mr C. Dawson and Dr A. Smith Woodward {Quart. Journ. Geol. Soc, 1913, p. 117). Unfortunately, its precise geological horizon has not been determined. 5. English Caves. Kent's Cavern was explored between i860 and 1880 under the superintendence of Mr Pengelly. The results of his research are published in the yearly Reports of the British Associa- tion. An interesting sketch of the exploration is given by him in a lecture entitled " Kent's Cavern " (Science Lectures for the People, 1872); see also a pamphlet entitled The Ancient Cave-men of Devonshire, published at Torquay. An account of the earlier investigations by MacEnery and others will be found in Transactions of the Devonshire Association for i868, 1869. Brixham Cave. — The exploration of this cave was conducted by a committee of the Geological Society, and under the super- intendence of Mr VengeWy {Philosophical Transactions, 1873, p. 471). Many other English caves have been explored, among which may be mentioned Kirkdale Cavern, in Yorkshire, described long ago by Dr Buckland as a hyaena den (Reliquia DiluviancB, 1823); Victoria cave, near Settle, Yorkshire, explored chiefly by Mr R. H. Tiddeman {Brit. Assoc. Reports, 1874, 1875; Geol. Mag., 1873, p. 11); Cresswell Crags Caves, Derbyshire, explored by Rev. J. Magens Mello {Quart. Journ. Geol. Soc, vols, xxxi.-xxxiii., and vol. XXXV.), and certain caves in the Vale of Clwyd, described by Dr Hicks {Quart. Journ. Geol. Soc, vol. xlii., p. 3 ; Geol. Mag., 1885, p. 510). For references to other caves see H. B. Woodward's The Geology of England and Wales, p. 541. The most recent contribution to our knowledge of British caves is an interesting and instructive lecture by Professor Sollas : "Paviland Cave: an Aurignacian Station in Wales" {Journ. Royal Anthrop. Institute, vol. xliii., 1913). In my second Lecture (p. 64) NOTES 309 I have stated that no indubitable trace of tfie Aurignacian stage of culture has hitherto been recognised in Britain. Professor SoUas has now put it beyond doubt that Aurignacian man flourished here just as on the Continent. 6. Caves of the Continent. The literature of cave-exploration on the Continent is very abundant, but the reader will find full and instructive summaries of the evidence in Professor Obermaier's recently published work, " Der Mensch der Vorzeit," Der Mensch aller Zeiten, vol. i., 191 2-13. Another recent work, Die diluviale Vorzeit Deutschlands, 191 2, by Dr R. R. Schmidt, gives an admirable and copiously illustrated description of the caves of Germany. Of the many important works dealing with special caves mention may be made of the splendid monograph by MM. E. Cartailhac and H. Breuil, La Caverne d'Altamira, 1906, the admirably executed figures and plates of which exhibit the artistic attainmerfts of Magdalenian man in mural drawing and painting. Another magnificent monograph in course of publication, Les Grottes de Grimaldi, is devoted to a description of the exploration of those caves, to three of which I have made some reference in the text. Both of these elabora-te works we owe to the munificence of Prince Albert I. of Monaco. 7. Palaeolithic Floors. The phenomena of these interesting " flcjors " are fully described by Mr Worthington Smith in Man, the Primeval Savage, 1894. 8. Stones "on End" in Pleistocene River-Gravels. For Mr Darwin's explanation of these -see Life and Letters of C. Darwin, vol. iii., p. 313. 9. Over-deepening of Valleys by Glacial Erosion. The amount of over-deepening attributed to the action of great valley-glaciers may appear almost incredible. But that glaciers are powerful agents of erosion must be admitted, and just for the same reason as streams and rivers are allowed to be so. If streams in time can cut deep gorges for hundreds of feet through hard rock, and rivers can excavate profound canons like those of the Colorado, 310 NOTES why should glaciers be incapable of deepening their beds ? Surely a massive glacier flowing for a prolonged period must inevitably deepen its valley in proportion to its size and its duration. When we realise that a glacial epoch embraced many thousands of years, and that such epochs returned again and again during the Pleistocene period, the over-deepening of Alpine valleys and Norwegian fiords will cease to surprise us, lo. Thickness of Pleistocene Glaciers. In estimating the probable thickness attained by these glaciers, geologists usually measure from the bottom of a valley to the upper limits reached by glaciation, which may often yield a sufificiently reliable result. It is obvious, however, that allowance should be made for glacial erosion. In the case of Over-deepened mountain valleys and fiords, for instance, the apparent thickness of the ice as indicated by the upper limits of glaciation is probably, as a rule, considerably in excess of the true thickness, for the surface of a glacier must be gradually lowered as the valley is deepened by erosion. II. Nomenclature of Glacial and Interglacial Epochs, Some years ago (^Journ. of Geology, vol. iii., p. 241) I suggested tentative names for the several glacial and interglacial stages of the Pleistocene. At that time the succession of the several glacial epochs had been more or less definitely ascertained, but the precise position of many of the interglacial deposits of our Continent was still doubtful. For instance, the well-known interglacial deposits of Switzerland were then supposed to belong to the Second Interglacial, and I therefore proposed to designate that epoch the " Helvetian." But the careful researches of MM. Penck and Bruckner have proved these particular interglacial beds to be on the horizon of the Third Interglacial stage, and the term "Helvetian" is therefore no longer appropriate. I now venture to suggest Tyrolian as a better designa- tion for the Second Interglacial epoch, after the famous plant-bearing breccias of Hotting, near Innsbruck (Tyrol),= described by Professor Penck and assigned by him to that epoch. The Third Interglacial, formerly termed by me " Neudeckian," I have now renamed DUrntenian, in recognition of the fact that the deposits of that stage are best represented by the well-known lignite-beds of Diirnten, NOTES 311 etc., in Switzerland. The only other change I would suggest is the substitution of the more euphonious Folonian for " Polandian," as a name for the Third Glacial epoch. 12. DiJRNTENIAN OR ThIRD InTERGLACIAL EpOCH. Amongst the interglacial formations which have been assigned to this stage are those of Krapina and the fluvio-lacustrlne deposits of Taubach, in Saxe Weimar. At Krapina remains of the broad- nosed rhinoceros abounded, and were associated with those of members of the temperate group of mammals, while tundra and steppe faunas had no representatives, The southern and temperate group occurs on two horizons — namely, the Tyrolian or Second Interglacial, and the Diirntenian or Third Interglacial stage. So far, then, as the evidence of its fauna is concerned the Krapina rock-shelter might belong to either of these interglacial horizons. The accompanying artifacts, it would seem, are indeterminate : they may be either Chellean or Mousterian-. The human remains, however, are unquestionably those of Neandertal man, and since all the skeletons of that race hitherto discovered are of Mousterian age, and never occur along with Chellean implements, there can be little doubt that the Krapina station belongs to the Diirntenian or Third Interglacial stage. The Taubach deposits overlie gravel-beds containing erratics of northern derivation, and are capped by loess, so that their inter- glacial character is obvious. The gravel-beds are of fluvio-glacial origin, deposited at a time when a great northern ice-sheet extended far south in Germany. The mer de glace in question was, in all probability, that of the Third Glacial or Polonian stage, and, if so, the Taubach beds with their southern fauna and relics and remains of Paleolithic man must be of Diirntenian age. On the other hand, the underlying fluvio-glacial gravels may yet prove to be those of the Saxonian epoch, in which case the Taubach beds would take their place in the Tyrolian or Second Interglacial stage. Unfortunately, the artifacts associated with the southern fauna at Taubach are apparently as indeterminate as those of Krapina, and cannot be certainly assigned either to a Mousterian or a Chellean horizon. It is interesting to note that M. Commont has recently discovered Mousterian artifacts associated with the southern or "warm" fauna in the river-drifts of the Somme Valley, at Montieres-les-Amiens. 312 NOTES The forms include straight-tusked elephant, hippopotamus, and broad-nosed rhinoceros. (See " Mousterien a faune chaude dans la vallee da la Somme," etc., Congres Internat. d' Anthropologic, etc., Geneva, 1912, t. i., p. 291.) M. Comment thinks that the deposits in question belong to the end of the Riss-Wiirm or Third Inter- glacial stage. But that is highly improbable — they almost certainly must be assigned to the culmination of that genial epoch. Unfortunately, M. Comment, following Professor Boule, has crowded all the Older Palaeolithic culture-stages into the Third Interglacial, and he is doubtless embarrassed by the occurrence of the warm fauna on two horizons in one and the same interglacial epoch. But the fauna in question is not confined to one stage — it characterises both Second and Third Interglacial stages. M. Commont's important discovery is quite in keeping with the appearance in the Thames valley of the Mousterian "floor," the plant- remains of which clearly indicate temperate conditions ; and furthermore it strongly supports Professor Penck's contention that Mousterian man was contemporaneous in middle Europe with the tundra fauna at one time and with the southern fauna at another. 13. Interstadial Phases of the Alpine Glacial Succession. The climatic oscillations of post-Wiirmian times were obviously of inconsiderable extent when contrasted with those of the earlier Pleistocene, and the climatic changes indicated by the corresponding stages of the post-Mecklenburgian in Britain were likewise less important than those which had preceded them. The earlier glacial and interglacial phases were not orsly more prolonged, but more strongly contrasted than the post-Mecklenburgian stages. Nevertheless, the climatic oscillations of Forestian and Turbarian times were of very considerable amplitude, and must have been experienced over a large part of our Continent. They were neither unimportant nor merely local phenomena, and there is no reason, therefore, why they should not be included in the great glacial cycle and described as glacial and interglacial. When the conditions that obtained in Turbarian times are contrasted with those of our day, this contention must appear to be well founded. The arctic-alpine flora occurs in Scotland at and above a height of 2000 feet; in Turbarian times it descended to the sea-level. More- over, during the Lower Turbarian stage cons>idcrable glaciers reached NOTES 313 the sea in our West Highlands, while in the later Turbarian stage many high-level corries would seem to have been occupied by small glaciers. Surely such conditions contrast strongly with the present, and still more strongly with the conditions of Forestian times, when our forests extended to a height of 3500 feet, that is, some 1500 feet above the present limit of tree-growth. The formations of late Pleistocene age have not been so assiduously studied in the low grounds of middle Europe as they will yet be. It can hardly be doubted, however, that climatic oscillations so pronounced as those to which our raised beaches and peat-mosses testify will eventually be found to have affected a large part of the Continent. The stadial and interstadial stages of the Alpine lands have been cited as the equivalents in time of our post-Mecklenburgian stages, but even stronger evidence of the amplitude of the later climatic oscillations has recently been obtained from the same region. Dr Hans Schreiber, who has of late years carried on researches in ■ the peat-bogs distributed along the northern border of the East Alps, has come to the conclusion that this tract in post-Wiirmian times experienced climatic changes, comparable in all respects to those that marked the closing phases of the glacial cycle in Scotland. He describes a succession of five stages which he parallels with the Scottish series and the correspond- ing stadial and interstadial stages of Penck and Bruckner. In his opinion the Biihl-Gschnitz interstadial stage ( = Lower Forestian) was very warm and dry — warmer and drier, in fact, than the present — so that the snow-line at that time must have Been 100 metres higher than now (H. Schreiber, " Vergletscherung und Moorbildung in Salzburg," etc., Verlag des Deutsch-bsterreichischen Moorvereins). Similar results have been obtained by V. Zailer from researches in the peat occurring within the basin of the river Enn. ("Die Enstehungs-geschichte der Moore," etc., Zeitschrift fi'ir MoorkuUur und TorfverwSrhing, 19 10. An outline of Schreiber's and Zailer's .Work is given by Professor Bruckner in Zeitschrift fur Gletscherkunde, Bd. vii., p. 334.) According to Penck and Bruckner the snow-line of the Biihl stadium stood 900 metres lower than now, while Schreiber believes it rose in the succeeding interstadial epoch 100 metres above its present level. A rise of the snow-line of 3250 feet or thereabout is surely indicative of a very considerable change of climate, and we should be justified in distinguishing as interglacial the epoch during which it occurred. 314 NOTES 14. The Azilian Stage and the supposed "Hiatus" between Palaeolithic and Neolithic Epochs. Archfeologists usually assign to the Azilian stage certain minute flint artifacts that show peculiar geometrical forms, as rhomboid, trapezoid, triangular, segment of a circle, etq. From the abundance of these microlithic implements at Fere-en-Tardenois (Aisne), accumulations characterised by their presence have been termed Tardenoisian. Such Tardenoisian artifacts have been recorded from a number of Azilian stations and also, it may be added, from various Neolithic accumulations in England. According to archaeologists, the Azilian or Azilian-Tardenoisian culture-stage bridges over the supposed break in the continuity of the human occupation of our Continent. This confident opinion, however, is hardly warranted by the evidence. So far as that goes, it proves that towards the close of Magdalenian times climatic conditions were gradually changing, and it- seems to indicate that in southern Europe no long interval occurred between the exit of Palaeolithic man and the entrance of his Neolithic successor. The discoveries at Mas d'Azil and elsewhere, therefore, would appear to have left the question of the "hiatus" just where it was over thirty years ago. At that time M. Piette had recently examined the cave of Gourdan in the Pyrenees, and described certain appearances which led him to infer that the Paleolithic and Neolithic races had come into contact; and the results obtained by MM. Lartet and Duparc in the rock-shelter at Sorde seemed to favour the same conclusion. The researches and conjectures of M. de Quatrefages and other anthropologists likewise tended to support the view that no "hiatus" existed. Commenting on the evidence available at the time, I ventured in 1881 to write as follows; "It is quite clear that a wide interval separates the Palceohthic and the Neolithic ages everywhere in central and north-western Europe ; but it is less certain that this interval was as prolonged in the south of France. Possibly, therefore, the Palaeolithic and the Neolithic races may have commingled in Perigord at a time whe'n the reindeer was still living, but in greatly diminished numbers in the valleys of the Pyrenees. But there is nothing to show that any of the Paljeolithic tribes were 'carried away in the pursuit of the reindeer' to its present home in Scandinavia. A prolonged period intervened between the close of the Ice Age \i.e., the Mecklenburgian or NOTES 315 Wiirmian stage) and the reappearance of Man in central and north- western Europe" {Prehistoric Europe, p. 553). That statement, I believe, still sums up all that is certainly known as to the passage from the Old Stone to the New Stone period- The bone artifacts belonging to the so-called " transition-beds " at Mas d'Azil are supposed to be decadent Palseojithic, but the awls, pointers, and spatulse in question are all very simple forms, and might quite well be of Neolithic age. They are indefinite in character and not distinctively Magdaleniajn. No other evidence supplied by the "transition-beds" is clearly suggestive of pre- Neolithic conditions ; on the contrary, it seems rather to point in the other direction. The stag-horn harpoons are certainly Neolithic types, for exactly similar implements occuj in the 25- to 30-foot beach in Scotland; while the water- worn stones, with their mysterious markings, and the heaps of snail shells, together with the fragments of pottery obtained in the Arisian "transition-bed," are all strongly suggestive of a Neolithic horizon. Moreover, it must be remembered that the Azilian is: separated from the Magdalenian beds below by four feet of river"alluvia. This need not indicate a long interval, but it certainly implies a "break" of some kind and not a gradual transition. Again, we note that not a single trace of the reindeer or any of its arctic-alpine associates has hitherto been met with in the Azilian or Azilian-Tardenoisian either of France or Germany. It has been long known to geologists and archaeologists that in early Neolithic times a primitive people, living chiefly on shell-fish and the produce of the chase, overspread wide regions in north- western Europe. Their " kitchen-middens '"• are met with again and again in Britain in connection with ancient raised beaches; and similar accumulations occur in Finland, Denmark, and elsewhere. In Scotland the oldest of these prehistorici "middens" belong to our 45- to 50-foot beach {i.e., Lower Turbarian), while the youngest are associated with the much later 25- to 3q-foot beach (i.e.. Upper Turbarian). If, as some archfeologists suppose, the Oban deposits are Azilian, then that stage must be long posterior in date to the Magdalenian, and cannot possibly represent the transition from Palseolithic to Neolithic times. It may quite well be that Neolithic man appeared in southern Europe before Palseolithic man had vanished from the Pyrenean region, and the^ two races may possibly have there come in contact. But such evidence as we have goes to 316 NOTES show that the older race had departed from central and north-west Europe long before the advent of Neolithic man in those latitudes. The "hiatus" recognised by geologists more than thirty years ago has not yet been bridged over. 15. River-drifts of the Somme Valley. Quite recently M. Comment has recorded the occurrence of Aurignacian and Solutrean artifacts from the river-drifts of the Somme, so that the succession of culture-stages is even more complete than is stated in the text (see pp. 111-112). The artifacts in question have been obtained from the upper part of the "ergeron." It may be as well to mention here that M. Commont's sections of the drifts are much more detailed than I have represented them to be. Thus he has several sub-divisions of each of the " loams," and of the river-gravels and sands of the terraces. These it was not necessary for me to particularise, as only the general succession of the culture-stages was required for my purpose. His "Strepyan," therefore, is included by iiie in the Chellean, and his sub-division of the later culture-stages into " Lower " and "Upper" Acheulian, Mousterian, etc., have been necessarily ignored. (See V. Commont, " Chronologie et stratigraphie des Industrie protohistoriques," etc., Congres Internat. d' Anthropologic, etc., Compte Rendu de la XlVe Session, 1912, t. i., p. 239.) 16. Maps of Europe during Glacial and Interglacial Epochs. Map a shows the areas subjected to glaciation -during the Saxonian or Second Glacial epoch. The limits reached by the great northern mer de glace have been fairly well ascertained, so far as the land-area is concerned. How far out into the Atlantic the ice flowed can only be conjectured, but it is not likely to have extended much beyond the loo-fms. line. Map B represents the greatest extension of land which is inferred to have taken place in interglacial times. It is designed chiefly to indicate the position of the land-bridges which formerly united Europe and Africa, and the probable geographical conditions in the north-west when the British Islands formed part of the Continent. The map, it will be understood, does not represent the distribution of land and sea during any particular interglacial stage. NOTES 317 Elevation of the land seems to have characterised each interglacial phase, but the extent and amount of elevatidn would seem to have varied. Probably the widest extension of the land took place during the Second (Tyrolian) and Third (Diirntenian) Interglacial epochs. But the British area was certainly continental in Lower Eorestian times. Map C shows the probable limits reached by the ice of the Third (Polonian) Glacial epoch in the British area, in Holland, north Germany, and west Poland. But the boundary line through Poland and Russia to the Arctic Ocean is largely conjectural, as indicated by the query marks. The limits given for the ice-sheet in the Atlantic are equally conjectural. Map D. — The area occupied by the northern ice-sheet of the Fourth (Mecklenburgian) Glacial epoch was considerably less extensive. Its limits are indicated by numerous lines of terminal moraines — the map having been compiled chiefly from papers by the late Dr Ussing and Professor Wahnscliaffe, which embody the views of many eminent geologists in Scandinavia, Germany, and Finland. ECOND GLACIAL EPOCH MAP A O -JS 30 35 -U) 1h 50 55 60 70 1 :. Glaciated Areas shown white 'TaknJbiurtliaLauitw ct Co., Edint To follow page 318. NTERGLACIAL TIMES MAP R .TQ"hti BiwLkoknninv S- CaSSjr^. To follow page. 318. ^ 20 J^ 30 35 -10 45 SO bb GO GE. 70 Tb 80 E THIRD GLACIAL EPOCH MAP C Glaciated Areas shown white .IrihBa/ij'tifiLiBrf.„-A Ct. ,EJU To follow page 318. E FOURTH GLACIAL EPOCH MAP D IS 20 35 30 35 40 K &0 55 60 -'-' \ \ \ V^ \ \ _^ ^-^v^ \ "^ \ \ Glac ated Areas shown white Sea] e uf EuJliakililBS ""^ '"" ^^--' "--' \ \ JJin BFirtiolomBiv-.t G-> ."Edin"" /"o follow fjitge 318. INDEX Aar valley, ancient glacier of, 199 Accumulation, area of dominant, 178 Acheulian stage, 43, 255 ; fauna of, 43 Adams, Mr, cited, 30 ^olian origin of loess, 21 5 Agassiz, L., cited, 189 Alder, 26, 224 Aleppo pine, 13 Alpine lands, climate of, in Pleisto- cene times, 15 Alps, glaciation of, 187; ancient glaciers of, 192 ; general con- ditions of, in glacial times, 193- '95 ; glaciers of northern, ig6 ; glaciers of southern, 202 Altamira Cavern, 309 Antarctic glacial phenomena, 141 Anticyclonic winds in glacial times, 213 Antiquity of man, estimates of, 301-304 Aosta, ancient glacier of, 203 Apennines, ancient glaciers of, 207 Apple-tree, 26 Apuan Alps, glaciation of, 207 Aj)us glacialis, 265 Archasological culture-stages, 36, 42 Arctic forests, 18 Arctic fox, 20, 22, 23, 69, 85, 128 Arctic glacial phenomena, 140 Arctic hare, 20, 23 819 Arctic plant-beds, at Hoxne, 229, 259 ; in Scottish peat, 226, 285, 289 ; in valley of the Lea, 117, 266 Argeles, ancient glacier of, 209 Arifege, ancient glacier of, 209 Arisian culture-stage, 297, 314 Arran, submarine depression in front of, 171 Arthur Seat, 155 Ascherson, Dr, cited, 29 Ash-tree, 17 Asturian mountains, glaciation of, 208 Aubrac, fnoraines of, 210 Aurignacian stage, 44, 308 ; fauna of, 44 ; geological horizon of, 267 Auvergne,, moraines of, 210 Azalea, creeping, 286 Azilian culture-stage, 45, 294-299, 314 Azores, erratics in, 253 Badgek, 28, 29, 59, 62, 95, 297 Balkans, glaciation of, 207 Balticbasin, glacier of, 267 ; former fresh-water lake of, 289 Barren grounds, N. America, 18 Bears, 21, 28, 44, 45> 58, 59, 62, 69, 74,i n-, 81, 85, 86, 95, 128, 228, 243, 297 Beaujolais, ancient glaciers of, 210 Beaver, ib, 28, 85, 91, 95, 250, 251, 297 320 INDEX Beech, 12, 14 Belgium, caves of, 65 Betula alba, 284 Betulanana, 24. 265, 284 Bison, 10, 29, 44, 59, 81, 118, 228, 250, 25', 255 Black Forest, ancient glaciers of, 207 Blizzards, 27, 30, 31, I25-128 Boar, 10, 85, 91, 95, 228, 297 Bobac, 28, 29 Bone accumulations in loess, 125 Borszcoff, M., cited, 126 Boucher de Perthes cited, 106 Boulder-clay, general character of Scottish, 164 ; distribution of, 166 ; stratified deposits in, 223 Boule, Professor, cited, 70, 312 Box-tree, 13, 15, 238, 239, 240 Brandt, Professor, cited, 31 Brasenia purpurea, 239 Breccia of caves, 58 ; of Gibraltar 135-138, 242, 261 Breuil, Prof, cited, 309 Britain, former continental con- dition of, 129, 253, 289 Brixham Cave, 62, 308 Bronze Age, 36, 297 Bruckner, Professor, cited, 191, 291, 310, 313 Buckland, Dr, cited, 308 Biihlstadium, 291, 313 Buntings, 23, 87 Buried forests, 225 Bustards, 28 Buzzards, 28 Caffer cat, 10 Caithness, shelly till of, 170 Calluna vulgaris. See Heather Canary laurel, 11, 12, 13, 14, 231 Caragan fox, 28 Carnivores, Pleistocene, 10 Carpathians, glaciation of, 207 Cartailhac, M., cited, 70, 309 Castle Reck, Edinburgh, 154 Catalan Bay, 244 Cave-earth, origin of, 5 1 Caves, general character and origin :of, 48 ; deposits in, 51; of England, 58, 255, 308 ; of Belgium, 65 ; of France, 67 ; of Italy, %p ; of Germany, 78, 309 ; of Austria, 78 Chalk of Baltic islands much dis- turbed, 181 Chamois,= 23, 69, 77 Charpentier, J. de, cited, 89 Chellean stage, 43i 254 ; valley- erosion during, 108 Chestnut, 15 Chillesford clay, 248 Climate, Pleistocene, 8 ; warm, 12- 17 ; cold, 22-25 ; steppe, 25-33 ; changes of, 33 ; cause of steppe, 213 , Clyde, raJsed beaches of, 271 Colorado, blizzards in, 31 Commont, M., on river-drifts of the Somme, 105, 311, 316 Contorted drift of Thames valley, 119, 266 Coombe-rock, 174 Corbicula fluminalis, 1 16, 226 Corrie glaciers of UpperTurbarian, 290 Corsac, 28 Corsica, glaciation of, 208 Cotton grass, 285 Coup de poing, 43 Crag-and:tail, 154, 167, 170 Cresswell Crags caves, 308 CroU, Dr, cited, 213 Crowberry, 285 Culture-stages, 36, 307 ; geological horizons of, 254, 257, 264, 267, 295 Darwin, Mr, on Pleistocene gravels, =127, 309 INDEX 321 Daunstadium, 291 Dawson, Mr C, cited, 308 Deflation, zone of, 215 Deflection -basins, 155, 172 Deflection of ice-flow in Scotland, 169 ; in England, 174 Dinotherium, 247 Dora Riparia, ancient glacier of, 203 Dormouse, 93 Drumlins, 167 Ducks, 33, 86 Duckworth, Dr, cited, 308 Duparc, M., cited, 314 Diirnten, interglacial beds of, 239 Durntenian epoch, 262, 310, 317 Dust-storms, in tundras, 21 ; in steppes, 30 Dwarf pine, 12 ^Dzeggetai, 28, 90 Eagles, 28, 91 Earn valley, lignite of, 372 Ebenalp, Santis, 263 Elephants, southern {Elephas meridionalis), 9, 228, 250 ; straight-tusked {E. antiquus), 9, 43, 74, n, 113, ii5> 228, 239, 250,251,255,312. 5^« Mammoth Elevation, supposed cause of glacial conditions, 184 Elk, 28, 74, 128, 239 Elm, 247 Empetivm nigrum, 285, 287 Engadine, glaciation of, 200 Engler, A., cited, 29 Eoliths, uncertain origin of, 4 Equus stenonis, yy Ergeron of Somme valley, iii, 316 Eriophoruni vaginatum, 285 Ermine, 21, 23, 28, 29, 86, 93 Erosion, glacial, 146-163; area of dominant, in N. Europe, 178 ; in Alps, 204 ; by streams in Palaeo- lithic times, 51, 60, 63, 67, 108, 114, 118, 121, 241, 256 Erratics in river-drifts, lig Estimates of geological time, 2, 301 Evans, Sir J., cited, 229 Falcon, 87 Faroe Islands, glaciation of, 252 Fauna, Arisian, 297 ; Azilian, 297 ; Durntenian, 262 ; forest, 91, 95 ; Polonian, 257, 261 ; Steppe, 25, 87, 125 ; Tyrolian, 254 ; Tundra, 22, 86, 124 Faunas, mixed, in river-drifts, 113 Fieldfare, 91 Fifth glacial epoch, 279 Fifth interglacial epoch, 279 Fig-tree, 13, 15, 231 Fiords, over-deepening of, 178 Fir, Scots. See Pine Firs, 15, 17, 239 First glacial epoch, 248 First interglacial epoch, 249 Firths of Forth, Tay, and Moray, direction of ice-flow in, 169 Fjelds ofNorway, 180 Flood deposits, 252 Flora, Pleistocene, 11 ; preglacial, 245 ; interglacial, 117, 224 ; Steppej 29 ; Tyrolian, 253 ; Tundra, 23, 120 Flowing soils of Arctic regions, 118, 134, 173 Flurlingen, interglacial deposits of, 263 Fluvio-glacial deposits, 147, 183, 205, 211 Foehn winds of glacialperiod, 214- 217 Forelands (Alpine), invaded by glaciers, 200, 203 Forest-bed (Norfolk), 227, 249 Forest beds in peat-bogs, 225 Forest fauna, 91, 95 Forests, interglacial, of Alps, 240 Forth valley, raised beaches of, 271 Fosses, Norwegian, 181 X 322 INDEX Founders of glacial geology, 1 88, 189 Fourth glacial epoch, 265, 317 Fourth intcrglacial epoch, 278 Fox, arctic, 20; common, 21, 23, 29, 59, 62, 86, 128 France, central, ancient glaciers of, 210; Pleistocene flora of, 11, 13 Frost, action of, 23, 132 Gallego valley, ancient glacier of, 209 Garda, lake, ancient glacier of, 203 Garonne valley, ancient glacier of, 208 Geese, 23 Geographical changes in Pleisto- cene times, 256, 272, 273, 289 ; indicated by past and present distribution of faunas and floras, 282 Germany, tufas of, 14 Gibraltar, limestone breccias of, 134 ; no longer forming, 135 ; marine terraces eroded in, 136 ; origin of, 136 ; two periods of formation of, 242, 261 Glacial action, 13 1 Glacial erosion, 146 Glacial lakes, 151 Glaciation, of Scotland, 164 ; of England, 173 ; N. Europe, 176 ; extent of great inland ice, 176; thickness of the ice, 177 ; dis- placement of rock-masses by ice-sheet, 181 ; areas of dominant erosion and accumulation, 182 ; morainic and fluvio-glacial deposits in front of ice-sheet, 183 Glaciers, ancient, confluence of, on Alpine forelands, 199, 203 Glaciers, ancient, of Jura, Black Forest, Vosges, Carpathians, etc., 207 Glaciers, formation of, 138 ; differ- ential motion of, 139 ; alpine, 140 ; Norwegian, 140; geological action of, 142 ; of Turbarian times, ?90 Glutton, 21, 23, 28, 59, 69, 86, 128 Glyptoliths in morainic sands, 216 Goat, 95 Gourdan, Cave of, 314 Greenland, glaciation of, 141 Grimaldi, caves of, "jo^ 309 Grotte de Carillon, 74 Grotte des Enfants, 70 Grotte du Prince, "]"} Grouse, 28, 86 Gschnitzstadium, 291 Giinzian epoch, 249 HamsteRj 85, 87, 91, 95 ; little, 87 Hand-axe, 43 Hare, arctic or mountain, 20, 23, 82, 86, 88 ; common, 28, 62, 93, 95 Harpoons of stag-horn, 296, 315 Harz, glaciation of, 207 Hazel, 224, 235, 239, 240 "Head" of S. England, 173 Heather, 284 Hearths, Paleolithic, 59, 70, •j'^, 74, 75, 77,81, 88,94, 128 Hebrides, Outer, submarine basins in front of, 171 Hedin, Sven v., cited, 134 Heer, Professor, cited, 1S9 Heidelberg, Mauer jaw, 47 Helix nemoralis, 297 "Hiatus," 61, 66, 99, 121, 269, 293, 29s, 299. 314 Hicks, Dr, cited, 308 Highlands, Scottish, peat-bogs of, 286 High-level river-drifts, 107 Hippopotamus, 9, tj, 113, 115, 228,247, 250, 255, 312 Holly, 235 INDEX 323 Hornbeam, 235, 240, 247 Horse, 10, 24, 28, 45, 49, 59, 62, 74,77,87,95, 115, "8, 128, 247, 25;, 297 ; primitive type of, 77 Hotting, interglacial deposits of, 237 Hoxne, glacial and interglacial beds at, 228, 259 Hull, interglacial deposits at, 226 Human skeletal remains in caves, 66, 69, 70, 74, 81 Humboldt glacier, 141 Hysena, 10, 29, 44, 45, 59, 62, 69, 70, 74, n, 128, 228, 243, 255 Ibex, 23, 70, 77, 90 Icebergs of glacial times, 185 Iceland, glaciation of, 252 Ice-shed of Alps in glacial times, 192 Ice-shelf of Antarctic region, 142 Iller valley, ancient glacier of, 199, 259 Inflation, zone of, 215 Innsbruck, interglacial deposits near, 238 Inn valley, ancient glacier of, 199, 200, 259 Interglacial formations, 218; definition of term, 3 19 ; evidence of climatic changes, 220 ; occur both in glaciated and non- glaciated regions, 222 ; Scottish, 222 ; English, 226 ; French, 231 ; German, etc., 232 ; Alpine, 236 ; of Gibraltar, 242 Interments, Neolithic, 94 Interstadial stages of Alps, 291, 292, 312 ; correlated with Fores- tian stages of Scotland, 293, 312 Ireland, glaciation of, 175 Irish deer, 44, 59, 85 Irish Sea, depressions on floor of, 175 Iron Age, 36, 297 Isar valley, ancient glacier of, igg, 260 Iseo basin, ancient glacier of, 203, 239 Istein, cave of, 298 Italy, marine pleistocene deposits of, vifith boreal and arctic shells, 252 Ivy, 247 Jackal, 28, 29 Jerboa, 28, 29 Judas-tree, 15, 17, 231 Jura, ancient glaciers of, 207 Justedal glacier, 147 Kaiser Wilhelm Land, 142 Kent's Cave, 55, 308 Kiang. See Dzeggetai Kirkdale cave, 308 Kollmann, Professor, cited 96 Kramberger, Professor, Gorjan- ovic-, cited, 78 Krapina, rock-shelter of, 78, 311 La Celle, tufa of, 13, 11 1, 231 La Chapelle-aux-Saints, cave of, 69 La Ferassie, cave of, 69 Lake Iseo, 239 Lakes of glaciated regions, 151, 196, 2&3 Land-bridges across Mediter- ranean, 129, 316 Lancaster, Sir E. R., cited, 307 Larch, 26 Lark, 28, 91 Lartet, ]V|., cited, 314 Laurel Of Canary Islands, ii ; common, 12, 247 Laurustinus, 15 Lauterbrunnental, 160 Lea-valley, arctic plant-bed of, 117 Lech valley, ancient glacier of, 199 Lemmingi, 20, 22, 44, 62, 86, 87 324 INDEX Le Mouslier, caves of, 68 Leopard, 243 Lewis, Dr, on plant remains of Scottish peat-bogs, 226, 283-89 Lewis, shelly boulder- clay of, 172 Lime-tree, 14, 26, 235 Limon fendilU of Somme valley, III Linth valley, ancient glacier of, 199 Lion, 10, 29, 59, 62, 69, 70, 74, 128, 243 Liquidambar, 247 Little hamster, 87 Loch Lomond, 154 Loch Ness, 154 Loess, character of, 122 ; geo- graphical distribution of, 123 ; chiefly aeolian, 123; organic remains of, 124, 125 ; preserva- tion of drifted snow under, 125 ; bone-accumulations in, 127 ; of Alpine derivation, 206 ; origin of, 212-17 Loiseleuria procumbens, 286 Lower Forestian, 278, 313; flora of, 284 ; geographical changes of, 289, 317 Lower Turbarian, 279, 315 ; flora of, 285, 287 ; depression during, 289 ; glaciers of, 290 Low-level river-drifts, 107 Lyell, Sir C, cited, 120 Lynx, 28, 77, 243 Lyonnais, ancient glaciers of, 210 M'Arthur Cave, Oban, 298 Magdalenian stage, 45 ; fauna of, 45, 86 ; geological horizon of, 267 ; closing period of, 295 ; art of, 45i 304 Mammals, Pleistocene, 8 ; southern and temperate forms, 9, 43, 44 ; tundra forms, 17 ; steppe forms, 25 Mammoth, 23, 24, 30, 43, 44, 45, 59, 62,,69, IIS, "7, "8, 128, 255 Mammoth period in Belgium, 66 Manul cat, 28, 90 Maple, 14=, 23s, 240, 247 Maral, Persian, 90 Marmot, 23, 29, 74, 77, 85 Marsh-lynx, 29 Marsh-trefoil, 239 Marten, 28, 85, 90, 93, 95 Mas d'Azil, 294, 314 Mastodop, 247 Mauer beds, 250 Mecklenburgian epoch, 265, 270, 317 Mediterranean Sea, boreal and arctic shells in pleistocene of Italy ; land-bridges across, 242 Mello, Rev. J. Magans, cited, 308 Mentone,: caves near, 70 Mer de glace, thickness of Scottish, 168 ; directions of movement of, 169 ; confluence of British with Scandinavian, 174, 176 ; extent of, on the Continent, 176 ; ap- proximate thickness of, 177 ; displacement of rock-masses by, 181 Merrick, peat-bogs of, 284 Mesvinian, 43, 316 Middendorf, A. v., cited, 88 Migration in Pleistocene times, 24, 33) i'i3 Minch, submarine basins on floor of the, 171 Moen island, 181 Mole, 93, 95 Molluscs,' Pleistocene, 16 Moraines, superficial, 142 ; bottom- or ground-, 145 Moray Firth, direction of ice-flow in, 170 Morlot, M,, cited, 189 Morteratsch glacier, 202 Mortillet, G. de, cited, 42 INDEX 325 Morvan, ancient moraines of, 2ii Mountain avens, 265 -Mousterian stage, 43 ; fauna of, 44 ; man of, 46 ; geological horizon of, 257, 263, 311 Munro, Dr R., cited, 308 Musk-ox, 20, 23, 44, 69, 86, 113, 128 Musk-shrew, go Neandertal man, 46, 311 Negroids, Aurignacian, 47 Nehring, Prof., cited, 307 Neolithic relics in 45- to 50-ft. beach, Scotland, 274 ; inArisian, 297 Norfolk, Pliocene and Pleistocene deposits of, 227 Norfolkian epoch, 249 Norway, ice-cap glaciers of, 140 ; over-deepening of fiords of, 179 Niiesch, Dr, on rock-shelter of Schweizersbild, 85 ; his estimate of age of the deposits, 95 Oak, evergreen, 13 ; common, 26, 224, 235 Oban, Neolithic remains in raised beach at, 274, 315 Obermaier, Professor, cited, 112, 309 Ofnet, cave of, 78, 298 Orkney Islands, 170 Oronsay, Neolithic remains in raised beach of, 274 Otter, 28, 85 Over-deepening of valleys by glacial action, 156, 179, 309 Owl, 28 Ox, 95, 118 Paleolithic camping stations in loess, 128; P. floors, 116, 264; implements, 40 ; P. man, 46, 251 ; geographical and climatic changes witnessed by him, 300 Pallas's oat. See Manul Cat Panther, 77 Paviland' cave, 308 Peat-bogs, evidence of climatic changes, 226, i^^-itj, 284, 312- 313; buried trees in, 275 ; Nor- wegian, 277 Penck, Professor, cited, 190, 209, 237, 260, 291, 301-303, 310, 313 Pengelly, Mr, cited, 308 Petry, Dr, cited, 29 Pianico, 263 Piette, M., cited, 294, 314 Pika, 28,087 Piltdown skull, 47, 308 Pine marten, 90 Pines, 12, 13, 26, 239 Playfair, Professor J., cited, 188 Pleistocene, general character of deposits, 7 ; warm climate of, 12-17; cold climate of, 17-25; steppe climate of, 25-33 Pliocene, conditions of Europe during, 246 Polar willow, 24, 228, 265 Polonian epoch, 257, 311 Pontic alpine rose, 15, 237, 238, 239, 240 Poplar, 17 Portugal,; ancient glaciers of, 208 Post-Mecklenburgian stages in Scotland, 271 Post-Wujmian stages of Alps, 313 Pouched marmot, 28, 29 Pre-palaeolithic artifacts, 5 Prestwich, J., cited, 107, 229 Provence, tufas of, 12 Ptarmigan, 23, 86, 95 Pyrenean pine, 12 Pyrenees, glaciation of, 208 QUATREFAGES, M. de, cited, 314 Rabbit, 62 Races, P«ila2olithic, 46 326 INDEX Raised beaches, 270, 271, 315 Ramsay, Sir A. C, cited, 152, 190 Recent and Present epoch, 281, 290 Red-deer, 28, 59, 62, 74, ^^, 85, 91, 95. 239; 255; 297 Reid, Mr J., cited, 5 Reindeer, 20, 23, 28, 29, 44, 45, 59, 62, 77, 86, 91, 93, 95, 113, 117, 118, 128, 255 Reindeer period in Belgium, 66 Reuss valley, ancient glacier of, 199 Rhine valley, ancient glacier of, 199 Rhinoceros, broad -nosed (/?. menkii), 10, 70, 74, yy, 81, 113, 115, 228, 239, 240, 312; Etruscan {R. etruscus), 10 ; woolly {R. tichorhimis), 23, 24, 29, 30, 43, 44, 59, 62, 74. 87, 118, 128, 255 Rhododendron po7tticum. See Pontic alpine rose Rhone valley, ancient glacier of, 196-198 ; over-deepening of, 158 Richter, Dr, cited, 1 79 Rissian epoch, 259 River-action, alternation of erosion and accumulation, 281 River-drifts, loi ; general char- acter of, 104 ; of the Somme, 105 ; contrasted faunas occurring in, 113; indicative of climatic changes, 115-121 ; Palaeolithic and Neolithic relics in, 129 Riviere, E., cited, 70 Roches moutonnhs, 168 Rock-basins, origin of, 152, 190; submarine, 171 Rock-falls in caves, 56 Rock-strise, 168 .Roe-deer, 28, 62, 77, 85, 91, 95, 297 Roseg glacier, 202 Rubble-drifts, 126, 133-38, 173, 211, 242, 252, 261 Rtickzugstadien, 291 Rufous suslik, 90, 91 Riigen island, 181 Russia, Pleistocene flora of, 1 5 Sabre-toothed tiger, 10, 59, 228, 247, 255) Saiga-antelope, 24, 28 Salix arbuscula, 287 ; S. herbacea, 285, 287 ; 5. repens, 284 ; S. reticulata, 286, 287 ; S. polaris. See Polar Willow Salzach valley, ancient glacier of, i99> 259 Salzmann's pine, 12 Saporta, G. de, cited, 11 Sassafras, 247 Saxonian epoch, 251, 316 Scandinavian inland ice invades England, 175 Scanian epoch, 248 Schmidt, Dr R. R., cited, 78, 309 Schotensack, Dr, cited, 25 1 Schreiber, Dr H., cited, 313 Schrenck, L. v., cited, 31 Schweizersbild rock-shelter, 85 ; age of deposits in, 97 Scotland, glaciation of, 164 ; inter- glacial deposits of, 222; late Pleistocene deposits of, 271 Screes, formation of, 133 ; Pleisto- cene, 135-138 Second glacial epoch, 251, 316 Second interglacial epoch, 253, 317 Serval, 243 Sheep, 95 Shetland Islands, glaciation of, 170 Shrew, 62, 93 Siberia, ice-formations of, 30, 32 ; their mammalian remains, 32 Sirgenstein, cave of, 78 Sisel or suslik, 90 Sixth glacial epoch, 281 Smith, Mr Worthington, cited, n6, 264, 26)5^ 309 Snow-drifts of tundras and steppes, 30; of 'prehistoric times, 125 INDEX 327 Snow-line of glacial times, in Alps, 193, 291, 313 ; in Pyrenees, 209 Snow-loving mammals, 17 Snow-owl, 23 Sollas, Prof., cited, 308 Solutrean epoch, 44 ; fauna of, 45 ; geological horizon of, 267 Somme valley, river-drifts of, 105, 311,316 Southern fauna, 9, 43, 70, 74, 77, 81, 113, 311 Southern Uplands, Scotland, peat- bogs of, 284 Spanish sierras, glaciation of, 208 Spy, cave of, 66 Squirrel, 91, 93, 95 Stadial stages of Alps, 293 Stag. See Red-deer Stalagmites, 52 ; rate of growth of, 55 ; of Kent's Cave, 58, 60 Staubbach, 160 Steppe flora, 29 Steppe mammals, 25 Steppes, physical conditions of, 25 ; characteristic animals of, 28 Stone Age, 36 Stones-on-end in river-drifts, 127, 309 Strepyan, 43, 316 Striated rocks, 168 Submergence, Pleistocene, 243, 256, 264, 267, 273, 289 Swan, 23 Switzerland, interglacial deposits of, 239, 310 Sycamore, 14, 17, 239 Tagliamento valley, ancient glacier of, 204 Tailless hare, 28, 29, 87 Tardenoisian, 314 Taubach, interglacial beds at, 311 Tay valley raised beaches, 271 ; lignite of, 272 Tertiary brown coal, disturbed under boulder-clay, 182 Tertiary era, duration of, 303 Tertiary plants in Pleistocene tufas, 12 Thames valley, river-drifts of, 112, 263 Thickness of tner de glace, 169, 177, 196, 310 Third glacial epoch, 257, 317 Third interglacial epoch, 262, 317 Tiddeman, Mr R. H., cited, 308 Tiger, 28 Time, geslogical, estimates of, 1,301 Torridon, Loch, moraines on raised beach, 273 "Trail" of Thames valley, 118 Transition from Palaeolithic to Neolithic, 295, 314 Trilobite cave, 70 Tundra fauna and flora, range of, in Pleistocene times, 22 Tundras of Eurasia and N. America, 18 ; flora and fauna of, 19 ;■ climate of, 21 Tuscany, Pleistocene flora of, 11 Tutkofifski, M., on origin of loess, 213 Tyrol, interglacial deposits of, 236 Tyrolian epoch, 253, 310, 317 Unter-Wetzikon, 239 Upper Forestian epoch, 279, 286, 287 ; limits of tree - growth during; 289 ; extension of land during, 290, 315 Upper Turbarian epoch, 280, 286, 287,290,315 Urals, ancient glaciers of, 207 Urus, 29, 45, 59, 62, 74, 95, 239 Ussing, Dr, cited, 317 Val Borlezza, 239 Valley-erosion in Chellean times, 108 328 INDEX VatnajokuU, 1 47 Victoria Cave, Yorkshire, 255 Victoria Land, 142 Vine, 15 Voles, 21,23, 28, 86, 87, 93 Vosges, ancient glaciers of, 207 Wahnschaffe, Prof., cited, 317 Wales, caves of, 255 Wapiti, 74 Warm fauna. See Southern fauna Warren, Mr S. H., cited, 117,266 Water-lily, 239 Water-rat, 62, 86, 93, 95 Weasel, 21, 23, 28, 29, 86, 93 Weybourn beds, 227, 228, 248 Wild-apple, 26 Wild-ass. See Dzeggetai Wild-boar, 28 Wild-cat, 28, 59, 85, 297 Wild-horse, 85 Wildkirchli, cave of, 263 Wild-ox, 77 Willows, ;26, 284, 285, 286. See Polar WjUow Wolf, 21, 23, 28, 29, 45, 59, 77, 85, 86, 95, 228 Woodward, Dr Smith, cited, 308 Woodward, Mr H. B., cited, 308 Wiirmian epoch, 267, 291 Wyoming, blizzards of, 31 Yew, 239 Zailer, Dr v., cited, 313 PBTNTBD BY OLIVER AND BOYD, EDINBDRQH