GLACIAL PERIOD or PLEISTOCENE ICE AGE, in geology, is the name given to a geological period, probably within the duration of man's occupancy of Europe, characterized, ac cording to the now universal interpretation, by cold conditions approaching arctic severity, the evidence of which is contained in superficial deposits extending over 8,000,000 sq.m. of Europe and North America, and great areas of Asia, and the southern Hemisphere.
The attention of geologists of the 18th and early 19th cen turies was directed to the existence of a series of deposits—clays, gravels, brickearths, sands and the like—that seemed to be an exception to the general orderly arrangement of the geological strata, associated as they were with strangely scored blocks of stone foreign to the districts in which they were found, and with scorings and "dressings" of rock surfaces, and seemed to indicate the operation of agencies different from those to which ordinary geological phenomena were attributed. It thus came about that the ideas involved in the Mosaic cosmogony and the Noachian deluge were readily seized upon and an explanation seemingly adequate was found ready to hand. The deposits were styled Diluvium, a term long since abandoned by British and American geologists, but still used by some writers on the Continent.
The earliest notes of dissent were sounded by Schimper, Venetz and Charpentier, who recognized in the action of Alpine glaciers forces productive of effects generally comparable with those attributed to a universal deluge, allowance being made for dif ferences of physiographic conditions. Though many workers on the Continent adopted and developed this hypothesis. circum stances conspired to delay its acceptance in Britain, the chief, perhaps, being that British geologists were less familiar with the action of glaciers. Another reason is to be found in the fact that, as the general direction of transport seemed to be from north to south and many unfamiliar types of crystalline rocks had ap parently been transported by a hill-and-valley ignoring agency, a great "wave of translation" seemed to be the only agency ade quate to explain the facts. This view found its last advocate in the late Sir Henry Howorth. This hypothesis was followed by one admitting the agency of ice but with the corollary, to which the facts seemed to point, of a great ice-sheet extending in all direc tions from the North Pole. This in turn was modified when it was discovered that in certain parts of Britain marine shells were contained in the deposits. This fact was brought to the knowledge of geologists in 1831 by Joshua Trimmer who found shell frag ments, not only on the low grounds in many places, but also at altitudes up to 1,35o ft. above the sea at Moel Tryfaen near Snowdon. Smith of Jordanhill about the same time recognized that shells found in Till (boulder-clay) of Scotland included species whose present habitat is in the Arctic seas.
Speculation now took the form of a marine submergence suf ficient to cover all the country up to 1.400-1,500 ft. with a sea cumbered with icebergs and floe-ice, and from the melting of these their burden of rock-materials, the boulder-clays, etc., were produced. In 1838 Buckland, who had only a few years before published his Reliquiae Diluvianae (1823), while on a tour in Switzerland made the acquaintance of Louis Agassiz who, though at first sceptical, had been converted to the views of Charpentier and Venetz that the boulders found across the plain of Switzer land and on the flanks of the Jura had been transported by glacier ice, of which the existing glaciers of the Alps were the dwindled representatives. Buckland was at first no less sceptical, but with characteristic open-mindedness he fully accepted Agassiz's con clusions and induced the great Swiss to visit Britain. The product of a joint tour was to confirm Buckland in the belief that the diluvial phenomena were attributable to the action of land-ice. All the phenomena of glacier-action were recognized—terminal and lateral moraines, ice-worn surfaces (one of which in the Blackford hills has been preserved and its significance recorded on a tablet), roches moutonnees, striated and far carried blocks. Agassiz described the famous parallel roads of Glen Roy as the strand-lines of a temporary glacier-dammed lake of the type of the Marjelen See, though its full story was not deciphered until Jamieson's demonstration in 1863.
Buckland's paper marked a new departure in glacial geology in Britain. It was followed by Lyell's memoirs on the Drift of Norfolk, and on the geological evidences of glaciers in Forfar shire. It has already been remarked that the older views still found advocates, and even in the loth century, the marine origin of boulder-clay has found some support. It may, however, be safely asserted that the postulate of a great ice-sheet having its radiant-point near the head of the Gulf of Bothnia is the starting point for all modern work on the glaciation of northern Europe.
Rocky hills or abrupt declivities in the path of a glacier may be moutonnees, and the effect of "plucking" on the lee-side will cause a retreat of the face upstream at the same time that the blocks detached will contribute to the burden of stones carried by the base of the glacier. Where a boss of rock of exceptional hard ness stands in the path of an ice-stream the lower layers of the ice cleave round it and it is usual to find a gully eroded across the front and along the two sides becoming evanescent towards the rear. The lowland valley of Scotland is replete with examples, such as the Castle rock at Edinburgh, and North Berwick Law. In the former case the long slope of the High street has been found to consist of a protected mass of soft coal measure rock forming a characteristic illustration of "crag and tail," but the "tail" in many instances is formed by accumulation of sand, gravel or the like in the "slack-water" in the rear of the crag.
The erosion-effects of ice-streams exhibit themselves in other forms and a great controversy without at present a decisive issue has arisen as to the limit of the scale of magnitude of these effects. Whether, for example, it being admitted that grooves and shallow hollows can be produced, the production of lake-basins and the characteristics of valley contours on a large scale can be ascribed to glacial erosion. Ramsay, James Geikie, Alfred Russel Wallace, Penck, Prof. W. M. Davis and others have held the opinion that they can. Prof. Collet in 1922 remarked "of all the Swiss geographers and geologists, Heim is the only one who will not admit that the great lakes at the foot of the Alps were formed by the action of glaciers." On the other side can be quoted also Bonney, Garwood and Gregory.
Moraines, as regards their structure, commonly show a con fused intermingling of rock-materials of every grade from the minutest of mud-particles up through sand-grains and stones to giant rock-masses of such dimensions that it is difficult to realize that they•are not part of the solid fabric of the country; these will be further mentioned later. Where the clayey element is very abundant a moraine may be said to consist of true boulder clay (the Scottish "Till"). Boulder-clay occurs, however, in other forms than as moraines. Its most common aspect is as sheets of greater or less extent spread over the land, sometimes in undulating relief, but, more commonly, forming extensive plateaux like, for example, that expanse between Manchester and Liverpool or much of Suffolk and north Essex, or the north Ger man plain. This type of boulder-clay is usually tough and hard, indeed in some of the excavations it is necessary to blast it with explosives; but the texture will naturally depend primarily on the nature of the constituents, and in a secondary degree as Sorby showed, upon the pressure to which it has been exposed.
The stones in boulder-clay assume in many cases characteristic attitudes—in one known as the "forced arrangement," the for ward end of the boulder is uptilted (Hinde). The orientation of a stone of large size will of ten serve to indicate the direction of travel, the longer axis being in line of movement and if there is a sharp and a blunt end, the sharp end will be foremost. It is hardly an exaggeration to say that there is no limit to the size of boulders: at Birkenwald near Berlin there is a transported mass of chalk of an estimated volume of 2,000,000 cu. metres, which is believed to have travelled i 5 km., and an erratic of chalk at Great Catworth, Huntingdon, though not quite equalling it in magnitude, must have been carried not less than 7o miles. Composite erratics of a succession of secondary rocks such as the Roswell erratic near Ely—long used as a quarry—may per haps have been originally outliers from an escarpment which have been pushed off their bases. Boulder-clay occasionally assumes the form of semi-ovoid mounds to which the name of "drumlins" has been given. These may have been built up near the melting end of an ice-stream by the accretions of englacial materials, but they may owe their form to the erosion of a belt of boulder-clay transverse to the front of an over-riding ice-sheet or glacier: they may be compared with roches moutonnees, and, like them, have taken on a form stable, at least temporarily, under moving ice.
Intimately connected with outwash plains are the Karnes or Eskers (Scandinavian Asar). These two terms were in their original significance applied in Ireland and Scotland respectively to phenomena of the same type, but American geologists employ them to designate structures different in form and geological re lations. Fairchild defines them in the following terms : "The term `kame' is here used . . . as designating deposits, chiefly sand and gravel, having a general knob and basin topography and formed at the margin or periphery of an ice-sheet. The term `esker' . . . is employed to denote distinct ridges, chiefly gravel, believed to have been deposited in the beds of subglacial streams, being phenomena of radial drainage." The distinction is fundamental —a kame, to an American geologist, is an accumulation parallel to the ice-front ; an esker is radial to it. As understood by British geologists, "eskers are winding ridges or strings of mounds com posed of water-worn sand and gravel. . . . They frequently show in their course across country a distinct disregard for the present surface gradient, thus proving that the streams which deposited them could not have flowed entirely on the surface, or else must have been under sufficient hydrostatic pressure to flow uphill" (Wright). Sollas favours the idea that they were formed in the channels of subglacial streams; but Gregory attributes a morainic origin' to them, in which opinion Charlesworth concurs, and he applies the term dsar to those disposed radially to the ice-front.
The surface beneath the glacial deposits has in general the characteristics of a land-surface: H. B. Woodward has described examples of "piping" of chalk in such situations which he regarded as evidence of sub-aerial erosion. The evidence outside the glaciated areas of the existence of ice-sheets and glaciers or of cold conditions are no less significant than those direct proofs of ice-action, and they have in recent years received a large amount of attention. When a glacier or ice-sheet obstructs the natural drainage of the ice-free country a lake will be formed, but will usually drain away on the removal of the ice barrier, leaving, however, certain signs by which its former existence may be recognized and which may furnish valuable corroboration, not otherwise obtainable, of the position and fluctuations of the ice front. The criteria relied on for their identification are mainly four: (r) strand lines; (2) floor deposits; (3) deltas, and (4) drainage channels cutting spurs or watersheds. The earliest ex ample to be recognized in Britain was the famous parallel roads of Glen Roy. The channels cut by water overflowing from such lakes furnish some of the most picturesque features of a glaciated country—many thousands have been recognized in the British Isles.
It is necessary to amplify a little the allusion to the occurrence of marine shells—a few whole and many more fragments—in the glacial deposits, which retarded for more than half a century the recognition of the true explanation of our glacial phenomena. The shelly Drift can, with few exceptions, be shown to occupy positions shown by independent evidence to have been invaded by ice that had crossed the sea-bed in its progress and thus had an opportunity of incorporating in its lower layers such relics of marine life. It was objected by advocates of the submergence hypothesis that even if the ice could pick up such shells it would inevitably grind them to powder. This objection has been met by Garwood and Gregory's discovery of perfect shells on the sur face of the Ivory glacier in Spitzbergen at a higher elevation than the raised beach from which the glacier had rifled them. Later, Lamplugh and others have described the moraine of the Sef strom glacier, also in Spitzbergen, in which vast numbers of marine shells in perfect condition have been thrust out of the fjord entangled in boulder-clay. Another way in which marine organisms can be uplifted has been described by Debenham. He found on the surface of two of the antarctic glaciers which are partially afloat marine muds with delicate organisms. He attributed their occurrence to the freezing of the glacier on to the sea floor, and melting at its upper surface causing the gradual emergence of the entangled mud.
Some doubt attaches to the geological age of Tillites (the name given to the ancient indurated boulder-clays) in localities in North America, e.g., Alaska, Nova Scotia and Prince Edward's island. At Squantum near Boston (U.S.A.), however, though some doubt exists as to geological age, the deposit is avouched by Coleman to be a typical ancient boulder-clay. After eliminat ing all doubtful records there remains a body of unchallengeable evidence of glaciation in Permo-Carboniferous times on a scale even vaster than that of the Pleistocene period, extending over enormous areas, and approaching in Africa, India, Australia and Brazil so near to the present position of the Equator as to con stitute a most baffling problem and which finds at present only partial solution in the speculations of Wegener.
Glacial deposits have now been recognized in rocks of many geological ages from the indubitable Tillites of the pre-Cambrian rocks of Canada and India to the great glacial series of Cambrian age in South Australia and the Silurian Tillites of Alaska. No authentic signs of glacial conditions can be recognized in the Devonian rocks; the Permo-Carboniferous have been mentioned above. Sporadic signs of glacial conditions have been recognized in rocks of Mesozoic and Cainozoic age, but nothing that could bear comparison with the earlier or with Pleistocene Ice ages.
A second great ice-sheet was constituted by the glaciers of the Alps, which spread in all directions, filling the great valleys and debouching on the plains. The whole area between the Alps and the Jura was filled with ice which left its deposits and erratics at high levels upon the confronting slope, in a few instances thrusting lobes through the passes. To the south it invaded the plains of Lombardy, excavated, as some would say, the great Italian lake-basins, and cast down the colossal moraines of the Dora Baltea and Dora Riparia. The lesser mountain chains of Europe had their own glacier systems, and great extensions of the glaciers of the Himalayas and probably every other mountain chain of Asia have been recognized. It is of interest to note that the famous cedars of Lebanon are growing on morainic material.
In North America three ice-sheets have been identified. The Labradorean on the north-east had its centre, or radiant point, in northern Quebec, whence it spread in every direction. Its southerly flow covered most of eastern Canada and reached as far as New York and Cincinnati. The Keewatin sheet, radiating from relatively low ground to the west of Hudson bay, expanded in all directions so as to come on the one hand into confluence with the Labradorean sheet and on the other to meet the third sheet. The Cordilleran diverges from an axis west of the Rocky mountains, and when it was at its full development lobes were thrust through some of the passes and may have come in contact with the Keewatin ice. In a south-westerly direction it reached Vancouver island and Queen Charlotte's islands and the Pacific ocean. Three hundred miles within the lobate front of the Keewatin-Labradorean sheet lies the Driftless area of Wisconsin —an area several hundred miles in extent left untouched between the two sheets. These three ice-sheets are computed to have covered an area of 4,000,000 sq. miles. Besides extensions of glaciers from some of the other mountainous centres of North America, Greenland was at some stage of the glacial period covered with ice to a greater extent than at present, and con tributed with other areas of arctic America to a total of ice covered land in the Old and New World of the Northern Hemis phere of about 8,000,000 sq. miles.
In the Southern Hemisphere evidence exists of a greater de velopment of glaciers than at the present time. Thus in Australia, New Zealand and Tasmania, glaciers extended much below exist ing limits, in some cases, and in others to areas where no glaciers are now found. In British East Africa Gregory found moraines 5,400 ft. below the present limits. In South America the same facts of greater glacier extension are disclosed. It must be ob served, however, that no proof has yet been adduced that the glaciers of the Southern Hemisphere were at a maximum simul taneously with those of the Northern Hemisphere, and it would be in agreement with some speculations regarding the cause of Ice ages if the northern and southern glaciations were alternate. Nor does it follow that temperature is the only factor governing the growth of glaciers; on the contrary the pole of greatest cold in the Northern Hemisphere, near Verkojansk, has no perennial snow though the subsoil is frozen probably to a depth exceeding 700 ft. as at Irkutsk.
James Geikie in the 2nd edition of his Great Ice Age (1877) was probably the first geologist to attempt a general classification of British deposits and their correlation with those of the Con tinent, but his views were largely coloured by Croll's splendid exposition of an astronomical theory of the Ice age with its necessary implication of glacial alternating with interglacial periods, and, further, by the doctrine of the great submergence which then dominated British geology, and to which he appeared to have adhered with some modification to the end. Geikie was content, in the work cited, to specify one composite "Great suc cession of' Glacial and Interglacial periods," a "Last Interglacial period," and "Last Glacial period," but in his last publication he differentiated the following: 6th Glacial epoch, Upper Turbarian, indicated by the deposits of peat which underlie the lower raised beaches.
5th Interglacial epoch, Upper Forestian.
5th Glacial epoch, Lower Turbarian, indicated by peat deposits overlying the lower f orest-bed, by the raised beaches and carse clays of Scotland, and in part by the Littorina-clays of Scan dinavia.
4th Interglacial epoch, Lower Forestian, the lower forests under peat beds, the Ancylus-beds of the great freshwater Baltic lake and the Littorina-clays of Scandinavia.
4th Glacial epoch, Mecklenburgian, represented by the moraines of the last great Baltic glacier, which reach their southern limit in Mecklenburg; the 10o ft. terrace of Scotland and the Yoldia-beds of Scandinavia.
3rd Interglacial epoch, Neudeckian, intercalations of marine and freshwater deposits in the boulder-clays of the southern Baltic coasts.
3rd Glacial epoch, Polandian, glacial and fluvio-glacial forma tions of the minor Scandinavian ice-sheet ; and the "upper boulder clay" of northern and western Europe.
2nd Interglacial epoch, Helvetian, interglacial beds of Britain and lignites of Switzerland.
2nd Glacial epoch, Saxonian, deposits of the period of maximum glaciation when the northern ice-sheet reached the low ground of Saxony, and the Alpine glaciers formed the outermost moraines.
1st Interglacial epoch, Norfolkian, the forest-bed series of Norfolk.
1st Glacial epoch, Scanian, represented only in the south of Sweden, which was overridden by a large Baltic glacier. The Chillesford clay and Weybourne crag of Norfolk and the oldest moraines and fluvio-glacial gravels of the Arctic lands may belong to this epoch.
This classification was subjected to a searching criticism by Lamplugh in 1906. He pointed out that by the method adopted, except as regards the uppermost members, the sequence involved a continual change of locus—thus, for relics of the 1st Glacial epoch it was necessary to go to Sweden, where no trace of the Norfolkian 1st Interglacial epoch, is found, and so on through the series. Lamplugh could further quote geologists of each country in turn who disclaimed belief in the order or position in the sequence of each member. He ascribed the assumed sequences to the temporary oscillations of an ice-sheet, but disputed the suggestion of complete, or even of considerable, deglaciation. It may also be remarked that differences of alimentation of different portions of an ice-sheet having an area of 2,000,000 sq.m. must have caused large fluctuations in the extension of the ice. Charles worth has shown in a recent discussion of the glaciation of the northwest of Ireland how variations of pressure brought first one then another of contending ice-streams over the same spot. Geikie's 4th Interglacial, 5th Glacial, 5th Interglacial and 6th Glacial stages are all based on variations in the plant constituents of the peat beds of Scotland and the north of England, and whatever may be the value of this evidence it would seem quite inadequate to sustain the burden of two glacial and two inter glacial epochs, which, if Croll's astronomical theory of the Ice age were accepted, would comprise in all four periods of 10,5o0 years each. The beautiful methods of de Geer, to be mentioned later, give an approximation to an actual chronology of the time, in years, that has elapsed since the Yoldia beds of Scandinavia were laid down. It is 12,500 years and into that brief space must be crowded Geikie's four climatic phases, as well as the first 18 centuries of the present era, which can hardly be regarded as part of the 6th Glacial epoch, and will therefore constitute a "6th Interglacial epoch." Penck and Briickner's Classification.--A classification which, unlike Geikie's, has the merit of a more limited range of locality, and of being based on the observations of a pair of geologists working in concert, is that of Penck and Bruckner (Die Alpen im Eiszeitalter). To accord with the stratigraphical succes sion their table is inverted—the oldest at the bottom: Daun Stadium, snowline 30o metres lower than at present. Geschnitz Stadium, snowline boom. lower than at present. Buhl Stadium, snowline 900m. lower than at present.
4th Wdrm Glaciation (divided by Achen interstadial), snow line 1,200m. lower than at present.
Riss-Wurm Interglacial, brief period warmer than at present. 3rd Riss Glaciation, maximum glaciation of France, Switzer land, Po and Rhine valley, snowline about 1,3oom. below present. Mindel-Riss Interglacial, period of uplift, very long duration. 2nd Mindel Glaciation, maximum glaciation of eastern, north eastern and south-eastern Alps, snowline 1,3oom. below present. Giinz-Mindel Interglacial, doubtful.
Gunz Glaciation, not well represented—snowline probably 1,200 m. lower than at present.
There are here four recognized glacial periods, and, with some doubt, three, or with the Achen interstadial dividing the W unz period, four interglacial. This classification, though designed only for the Alpine region, has been applied by many geologists to other districts, or even continents, and with some degree of probability has been adopted as a standard to which other classi fications are made to conform.
In Yorkshire, while inland there are an older and newer boulder clay, never seen either in superposition, or even in the same district, on the coast three, or possibly four, boulder-clays can be identified, not only by the test of superposition, but by character istics of colour and the assemblages of the far transported erratics which they contain. In ascending or chronological order they are: (I) Basement clay; (2) Purple clay, sometimes in two divisions separated by sands and gravels; (3) Hessle clay. The first of these is a leaden-coloured clay with occasional included patches of sea-bottom. It contains a few Scandinavian rocks, and rocks from the Lake Distict and the Pennine chain. The Purple clay is characterized by a redder colour and by the presence of numer ous large blocks of Shap granite and Carboniferous limestones. The Hessle clay is by far the best characterized, as it contains in great profusion rocks from the Tweed valley and the Cheviots. Between the Purple clay and the Hessle clay, that normally suc ceeds it, there is found at one locality in Lincolnshire at an elevation of about 8oft. above sea-level a bed of estuarine silt with marsh plants and brackish-water shells. This is the only "interglacial" bed in Britain that is beyond challenge.
The ice of the Hessle-clay phase at its maximum failed to cross the Cleveland hills, but the comparative obstruction of its move ments in the North sea, by the opposition of the Scandinavian ice-sheet, deflected much of its flow into the Vale of York and a great moraine at Escrick, sm. S. of the city of York, seems to mark its utmost extension. Beyond that line scattered shreds of a much older, greatly denuded, boulder-clay are found. The contrast between the two glaciations is very striking and the dif ference between the sharp and well-preserved topography of the later (Hessle) and the almost complete obliteration of that pro duced by the earlier glaciation gives a measure of their relative ages. The same contrast will be seen in America.
The most complete succession of deposits is naturally to be found in the peripheral areas and the classification of deposits in the Mississippi valley is as follows: The Nebraskan Till and Aftonian Interglacial are not ad mitted by all American geologists. Each Till sheet overlaps its predecessor from north to south, but the Kansan is considered to have been laid down by the Keewatin ice, which was confined to the states on the western side of the Driftless area of Wisconsin, viz., Nebraska, Iowa, Kansas and Missouri, and to have retreated in a northerly direction, before the advance of the Labradorean ice, which in parts of Illinois and Wisconsin overrode the Kansan drift. Between these two sheets of Till, beds of peat and other subaerial deposits are traceable over an area indicating an advance of the Labradorean ice for not less than iso years over the country laid bare by the Kansan retreat. The Iowan marks a re advance of the Keewatin ice but its relations to the Labrador ice is not definitely settled. The so-called Iowan Till is in some places separated from the Illinoian by a terrestrial surface upon the weathered Illinoian Till, of which the Iowan is by some con sidered to be an upper division. The Wisconsin Till, whether a single sheet or in two stages is still uncertain, is underlain by a weathered terrestrial surface in many sections.
It is evident that many and extensive fluctuations of the ice margins must have taken place but there is no means of deciding whether this amounted to a complete deglaciation or only a partial shrinkage. Coleman has described sections near Toronto, discovered by Hinde, which give support to the supposition of complete disappearance of the ice, or at least its withdrawal from the area of the Great Lakes and subsequent re-advance in great force. At the Scarboro Heights sections show three beds of boulder-clay separated by sedimentary beds. Fossiliferous silts intervening between the older boulder-clay and the next have yielded a large suite of freshwater shells, and an abundant flora comprising many timber-trees such as oaks, elms and maples, some of which are now extinct, while others are limited to lati tudes four or five degrees south of Toronto. It is unfortunate that the associated boulder-clays cannot be referred to equivalents south of the Great Lakes. Coleman is of the opinion that these deposits are unmistakable evidence of, at the least, one complete removal of the ice from the Labradorean centre.
The use of biological and anthropological data stands on a different footing from these and is, in fact, the application of the geological test of "characteristic fossils." It is in many ways more trustworthy, though it too has its limitations, for latitude and the distribution of land and sea no less than the character of the vegetation and conditions of the terrain, have had their influence upon the distribution of the Pleistocene mammalia which are adopted as criteria of age. Some of the same factors must also have operated to control the dispersal of early man. To cite but a single instance—Magdalenian man who hunted the reindeer in southern Europe during the closing stages of the Ice age is represented to-day by the Eskimo.
Using the two classes of evidence separately, or conjointly where possible, a sequence of human culture and in some cases of race, can be correlated with the contemporary mammalian fauna and these in turn with the more or less direct products of the ice, or of the deglaciated land, and with deposits of rivers outside the glaciated areas.
A mere enumeration of the mammalia associated with these stages of culture would furnish an imperfect basis for classifica tion of the deposits apart from the implements and weapons of man, but one decisive contrast is afforded between the Chellean and later faunas, viz., the presence in Chellean deposits of the straight-tusked elephant (Elephas antiquus) which does not occur in Acheulian or later deposits being replaced by the mammoth (E. primigenius) which survived in Europe till the close of the Magdalenian period. Similarly the soft-nosed rhinoceros (R. leptorhinus,) gives place to the woolly rhinoceros (R. A hippopotamus closely related to the existing Nilotic species is characteristic of Chellean deposits in this country. The Chellean fauna has a distinctly warm aspect and it furnishes perhaps the best indication of geological age of any of the Pleistocene faunas.
It is fortunate that where the cultural criteria are lacking, the mammalian evidence may fix the age of deposits and vice versa : thus the mammalian remains are the sole evidences in the north of England, while in the south many river-terraces, though lack ing the mammalia, rarely fail to reward prolonged search for implements. The terraces of rivers in the country beyond the glaciated area can be brought into relation with the phases of the Ice age by these means.