Another sign of ice agency, of which Mr. Prestwich has given a good illustration in one of his published sections, and which I myself observed in several pits at St. Acheul, deserves notice. It consists in flexures and contortions of the strata of sand, marl, and gravel (as seen at b, c, and d, Figure 22), which they have evidently undergone since their original deposition, and from which both the underlying Chalk and part of the overlying beds of sand Number 3 are usually exempt.

In my former writings I have attributed this kind of derangement to two causes; first, the pressure of ice running aground on yielding banks of mud and sand; and, secondly, the melting of masses of ice and snow of unequal thickness, on which horizontal layers of mud, sand, and other fine and coarse materials had accumulated. The late Mr. Trimmer first pointed out in what manner the unequal failure of support caused by the liquefaction of underlying or intercalated snow and ice might give rise to such complicated foldings.* (* See chapter 12.)

When "ice-jams" occur on the St. Lawrence and other Canadian rivers (latitude 46 degrees north), the sheets of ice, which become packed or forced under or over one another, assume in most cases a highly inclined and sometimes even a vertical position. They are often observed to be coated on one side with mud, sand, or gravel frozen on to them, derived from shallows in the river on which they rested when congelation first reached the bottom.

As often as portions of these packs melt near the margin of the river, the layers of mud, sand, and gravel, which result from their liquefaction, cannot fail to assume a very abnormal arrangement—very perplexing to a geologist who should undertake to interpret them without having the ice-clue in his mind.

Mr. Prestwich has suggested that ground-ice may have had its influence in modifying the ancient alluvium of the Somme.* (* Prestwich, Memoir read to Royal Society, April 1862.) It is certain that ice in this form plays an active part every winter in giving motion to stones and gravel in the beds of rivers in European Russia and Siberia. It appears that when in those countries the streams are reduced nearly to the freezing point, congelation begins frequently at the bottom; the reason being, according to Arago, that the current is slowest there, and the gravel and large stones, having parted with much of their heat by radiation, acquire a temperature below the average of the main body of the river. It is, therefore, when the water is clear, and the sky free from clouds, that ground ice forms most readily, and oftener on pebbly than on muddy bottoms. Fragments of such ice, rising occasionally to the surface, bring up with them gravel, and even large stones.

Without dwelling longer on the various ways in which ice may affect the forms of stratification in drift, so as to cause bendings and foldings in which the underlying or over-lying strata do not participate, a subject to which I shall have occasion again to allude in the sequel, I will state in this place that such contortions, whether explicable or not, are very characteristic of glacial formations. They have also no necessary connection with the transportation of large blocks of stone, and they therefore afford, as Mr. Prestwich remarks, independent proof of ice-action in the Pleistocene gravel of the Somme.

Let us, then, suppose that, at the time when flint hatchets were embedded in great numbers in the ancient gravel which now forms the terrace of St. Acheul, the main river and its tributaries were annually frozen over for several months in winter. In that case, the primitive people may, as Mr. Prestwich hints, have resembled in their mode of life those American Indians who now inhabit the country between Hudson's Bay and the Polar Sea. The habits of those Indians have been well described by Hearne, who spent some years among them. As often as deer and other game become scarce on the land, they betake themselves to fishing in the rivers; and for this purpose, and also to obtain water for drinking, they are in the constant practice of cutting round holes in the ice, a foot or more in diameter, through which they throw baited hooks or nets. Often they pitch their tent on the ice, and then cut such holes through it, using ice-chisels of metal when they can get copper or iron, but when not, employing tools of flint or hornstone.

The great accumulation of gravel at St. Acheul has taken place in part of the valley where the tributary streams, the Noye and the Arve, now join the Somme. These tributaries, as well as the main river, must have been running at the height first of 100 feet, and afterwards at various lower levels above the present valley-plain, in those earlier times when the flint tools of the antique type were buried in successive river beds. I have said at various levels, because there are, here and there, patches of drift at heights intermediate between the higher and lower gravel, and also some deposits, showing that the river once flowed at elevations above as well as below the level of the platform of St. Acheul. As yet, however, no patch of gravel skirting the valley at heights exceeding 100 feet above the Somme has yielded flint tools or other signs of the former sojourn of Man in this region.

Possibly, in the earlier geographical condition of this country, the confluence of tributaries with the Somme afforded inducements to a hunting and fishing tribe to settle there, and some of the same natural advantages may have caused the first inhabitants of Amiens and Abbeville to fix on the same sites for their dwellings. If the early hunting and fishing tribes frequented the same spots for hundreds or thousands of years in succession, the number of the stone implements lost in the bed of the river need not surprise us. Ice-chisels, flint hatchets, and spear-heads may have slipped accidentally through holes kept constantly open, and the recovery of a lost treasure once sunk in the bed of the ice-bound stream, inevitably swept away with gravel on the breaking up of the ice in the spring, would be hopeless. During a long winter, in a country affording abundance of flint, the manufacture of tools would be continually in progress; and, if so, thousands of chips and flakes would be purposely thrown into the ice-hole, besides a great number of implements having flaws, or rejected as too unskilfully made to be worth preserving.

As to the fossil fauna of the drift, considered in relation to the climate, when, in 1859, I took a collection which I had made of all the more common species of land and freshwater shells from the Amiens and Abbeville drift, to my friend M. Deshayes at Paris, he declared them to be, without exception, the same as those now living in the basin of the Seine. This fact may seem at first sight to imply that the climate had not altered since the flint tools were fabricated; but it appears that all these species of molluscs now range as far north as Norway and Finland, and may therefore have flourished in the valley of the Somme when the river was frozen over annually in winter.* (* See Prestwich, Paper read to the Royal Society in 1862.)

In regard to the accompanying mammalia, some of them, like the mammoth and tichorhine rhinoceros, may have been able to endure the rigours of a northern winter as well as the reindeer, which we find fossil in the same gravel. It is a more difficult point to determine whether the climate of the lower gravels (those of Menchecourt, for example) was more genial than that of the higher ones. Mr. Prestwich inclines to this opinion. None of those contortions of the strata above described have as yet been observed in the lower drift. It contains large blocks of Tertiary sandstone and grit, which may have required the aid of ice to convey them to their present sites; but as such blocks already abounded in the older and higher alluvium, they may simply be monuments of its destruction, having been let down successively to lower and lower levels without making much seaward progress.

The Cyrena fluminalis of Menchecourt and the hippopotamus of St. Roch seem to be in favour of a less severe temperature in winter; but so many of the species of mammalia, as well as of the land and freshwater shells, are common to both formations, and our information respecting the entire fauna is still so imperfect, that it would be premature to pretend to settle this question in the present state of our knowledge. We must be content with the conclusion (and it is one of no small interest), that when Man first inhabited this part of Europe, at the time that the St. Acheul drift was formed, the climate as well as the physical geography of the country differed considerably from the state of things now established there.

Among the elephant remains from St. Acheul, in M. Garnier's collection, Dr. Falconer recognised a molar of the Elephas antiquus, Figure 19, the same species which has been already mentioned as having been found in the lower-level gravels of St. Roch. This species, therefore, endured while important changes took place in the geographical condition of the valley of the Somme. Assuming the lower-level gravel to be the newer, it follows that the Elephas antiquus and the hippopotamus of St. Roch continued to flourish long after the introduction of the mammoth, a well characterised tooth of which, as I before stated, was found at St. Acheul at the time of my visit in 1860.

As flint hatchets and knives have been discovered in the alluvial deposits both at high and low levels, we may safely affirm that Man was as old an inhabitant of this region as were any of the fossil quadrupeds above enumerated, a conclusion which is independent of any difference of opinion as to the relative age of the higher and lower gravels.

The disappearance of many large pachyderms and beasts of prey from Europe has often been attributed to the intervention of Man, and no doubt he played his part in hastening the era of their extinction; but there is good reason for suspecting that other causes co-operated to the same end. No naturalist would for a moment suppose that the extermination of the Cyrena fluminalis throughout the whole of Europe—a species which co-existed with our race in the valley of the Somme, and which was very abundant in the waters of the Thames at the time when the elephant, rhinoceros, and hippopotamus flourished on its banks—was accelerated by human agency. The great modification in climate and in other conditions of existence which affected this aquatic mollusc, may have mainly contributed to the gradual dying out of many of the large mammalia.

We have already seen that the peat of the valley of the Somme is a formation which, in all likelihood, took thousands of years for its growth. But no change of a marked character has occurred in the mammalian fauna since it began to accumulate. The contrast of the fauna of the ancient alluvium, whether at high or low levels, with the fauna of the oldest peat is almost as great as its contrast with the existing fauna, the memorials of Man being common to the whole series; hence we may infer that the interval of time which separated the era of the large extinct mammalia from that of the earliest peat, was of far longer duration than that of the entire growth of the peat. Yet we by no means need the evidence of the ancient fossil fauna to establish the antiquity of Man in this part of France. The mere volume of the drift at various heights would alone suffice to demonstrate a vast lapse of time during which such heaps of shingle, derived both from the Eocene and the Cretaceous rocks, were thrown down in a succession of river-channels. We observe thousands of rounded and half-rounded flints, and a vast number of angular ones, with rounded pieces of white Chalk of various sizes, testifying to a prodigious amount of mechanical action, accompanying the repeated widening and deepening of the valley, before it became the receptacle of peat; and the position of many of the flint tools leaves no doubt in the mind of the geologist that their fabrication preceded all this reiterated denudation.

ON THE ABSENCE OF HUMAN BONES IN THE ALLUVIUM OF THE SOMME.

It is naturally a matter of no small surprise that, after we have collected many hundred flint implements (including knives, many thousands), not a single human bone has yet been met with in the old alluvial sand and gravel of the Somme. This dearth of the mortal remains of our species holds true equally, as yet, in all other parts of Europe where the tool-bearing drift of the Pleistocene period has been investigated in valley deposits. Yet in these same formations there is no want of bones of mammalia belonging to extinct and living species. In the course of the last quarter of a century, thousands of them have been submitted to the examination of skilful osteologists, and they have been unable to detect among them one fragment of a human skeleton, not even a tooth. Yet Cuvier pointed out long ago, that the bones of Man found buried in ancient battle-fields were not more decayed than those of horses interred in the same graves. We have seen that in the Liege caverns, the skulls, jaws, and teeth, with other bones of the human race, were preserved in the same condition as those of the cave-bear, tiger, and mammoth.

That ere long, now that curiosity has been so much excited on this subject, some human remains will be detected in the older alluvium of European valleys, I confidently expect. In the meantime, the absence of all vestige of the bones which belonged to that population by which so many weapons were designed and executed, affords a most striking and instructive lesson in regard to the value of negative evidence, when adduced in proof of the non-existence of certain classes of terrestrial animals at given periods of the past. It is a new and emphatic illustration of the extreme imperfection of the geological record, of which even they who are constantly working in the field cannot easily form a just conception.

We must not forget that Dr. Schmerling, after finding extinct mammalia and FLINT TOOLS in forty-two Belgian caverns, was only rewarded by the discovery of human bones in three or four of those rich repositories of osseous remains. In like manner, it was not till the year 1855 that the first skull of the musk ox (Bubalus moschatus) was detected in the fossiliferous gravel of the Thames, and not till 1860, as will be seen in the next chapter, that the same quadruped was proved to have co-existed in France with the mammoth. The same theory which will explain the comparative rarity of such species would no doubt account for the still greater scarcity of human bones, as well as for our general ignorance of the Pleistocene terrestrial fauna, with the exception of that part of it which is revealed to us by cavern researches.

In valley drift we meet commonly with the bones of quadrupeds which graze on plains bordering rivers. Carnivorous beasts, attracted to the same ground in search of their prey, sometimes leave their remains in the same deposits, but more rarely. The whole assemblage of fossil quadrupeds at present obtained from the alluvium of Picardy is obviously a mere fraction of the entire fauna which flourished contemporaneously with the primitive people by whom the flint hatchets were made.

Instead of its being part of the plan of nature to store up enduring records of a large number of the individual plants and animals which have lived on the surface, it seems to be her chief care to provide the means of disencumbering the habitable areas lying above and below the waters of those myriads of solid skeletons of animals, and those massive trunks of trees, which would otherwise soon choke up every river, and fill every valley. To prevent this inconvenience she employs the heat and moisture of the sun and atmosphere, the dissolving power of carbonic and other acids, the grinding teeth and gastric juices of quadrupeds, birds, reptiles, and fish, and the agency of many of the invertebrata. We are all familiar with the efficacy of these and other causes on the land; and as to the bottoms of seas, we have only to read the published reports of Mr. MacAndrew, the late Edward Forbes, and other experienced dredgers, who, while they failed utterly in drawing up from the deep a single human bone, declared that they scarcely ever met with a work of art even after counting tens of thousands of shells and zoophytes, collected on a coast line of several hundred miles in extent, where they often approached within less than half a mile of a land peopled by millions of human beings.

LAKE OF HAARLEM.

It is not many years since the Government of Holland resolved to lay dry that great sheet of water formerly called the Lake of Haarlem, extending over 45,000 acres. They succeeded, in 1853, in turning it into dry land, by means of powerful pumps constantly worked by steam, which raised the water and discharged it into a canal running for 20 or 30 miles round the newly-gained land. This land was depressed 13 feet beneath the mean level of the ocean. I travelled, in 1859, over part of the bed of this old lake, and found it already converted into arable land, and peopled by an agricultural population of 5000 souls. Mr. Staring, who had been for some years employed by the Dutch Government in constructing a geological map of Holland, was my companion and guide. He informed me that he and his associates had searched in vain for human bones in the deposits which had constituted for three centuries the bed of the great lake.

There had been many a shipwreck, and many a naval fight in those waters, and hundreds of Dutch and Spanish soldiers and sailors had met there with a watery grave. The population which lived on the borders of this ancient sheet of water numbered between thirty and forty thousand souls. In digging the great canal, a fine section had been laid open, about 30 miles long, of the deposits which formed the ancient bottom of the lake. Trenches, also, innumerable, several feet deep, had been freshly dug on all the farms, and their united length must have amounted to thousands of miles. In some of the sandy soil recently thrown out of the trenches, I observed specimens of freshwater and brackish-water shells, such as Unio and Dreissena, of living species; and in clay brought up from below the sand, shells of Tellina, Lutraria, and Cardium, all of species now inhabiting the adjoining sea.

As the Dreissena is believed by conchologists to have been introduced into Western Europe in very modern times, brought with foreign timber in the holds of vessels from the rivers flowing into the Black Sea, the layer of sand containing it in the Haarlem lake is probably not more than a hundred years old.

One or two wrecked Spanish vessels, and arms of the same period, have rewarded the antiquaries who had been watching the draining operations in the hope of a richer harvest, and who were not a little disappointed at the result. In a peaty tract on the margin of one part of the lake a few coins were dug up; but if history had been silent, and if there had been a controversy whether Man was already a denizen of this planet at the time when the area of the Haarlem lake was under water, the archaeologist, in order to answer this question, must have appealed, as in the case of the valley of the Somme, not to fossil bones, but to works of art embedded in the superficial strata.

Mr. Staring, in his valuable memoir on the "Geological Map of Holland," has attributed the general scarcity of human bones in Dutch peat, notwithstanding the many works of art preserved in it, to the power of the humic and sulphuric acids to dissolve bones, the peat in question being plentifully impregnated with such acids. His theory may be correct, but it is not applicable to the gravel of the valley of the Somme, in which the bones of fossil mammalia are frequent, nor to the uppermost freshwater strata forming the bottom of a large part of the Haarlem Lake, in which it is not pretended that such acids occur.

The primitive inhabitants of the valley of the Somme may have been too wary and sagacious to be often surprised and drowned by floods, which swept away many an incautious elephant or rhinoceros, horse and ox. But even if those rude hunters had cherished a superstitious veneration for the Somme, and had regarded it as a sacred river (as the modern Hindoos revere the Ganges), and had been in the habit of committing the bodies of their dead or dying to its waters—even had such funeral rites prevailed, it by no means follows that the bones of many individuals would have been preserved to our time.

A corpse cast into the stream first sinks, and must then be almost immediately overspread with sediment of a certain weight, or it will rise again when distended with gases, and float perhaps to the sea before it sinks again. It may then be attacked by fish of marine species, some of which are capable of digesting bones. If, before being carried into the sea and devoured, it is enveloped with fluviatile mud and sand, the next flood, if it lie in mid-channel, may tear it out again, scatter all the bones, roll some of them into pebbles, and leave others exposed to destroying agencies; and this may be repeated annually, till all vestiges of the skeleton may disappear. On the other hand, a bone washed through a rent into a subterranean cavity, even though a rarer contingency, may have a greater chance of escaping destruction, especially if there be stalactite dropping from the roof of the cave or walls of a rent, and if the cave be not constantly traversed by too strong a current of engulfed water.

CHAPTER 9.

WORKS OF ART IN PLEISTOCENE ALLUVIUM OF FRANCE AND ENGLAND.

Flint Implements in ancient Alluvium of the Basin of the Seine. Bones of Man and of extinct Mammalia in the Cave of Arcy. Extinct Mammalia in the Valley of the Oise. Flint Implement in Gravel of same Valley. Works of Art in Pleistocene Drift in Valley of the Thames. Musk Ox. Meeting of northern and southern Fauna. Migrations of Quadrupeds. Mammals of Mongolia. Chronological Relation of the older Alluvium of the Thames to the Glacial Drift. Flint Implements of Pleistocene Period in Surrey, Middlesex, Kent, Bedfordshire, and Suffolk.

FLINT IMPLEMENTS IN PLEISTOCENE ALLUVIUM IN THE BASIN OF THE SEINE.

In the ancient alluvium of the valleys of the Seine and its principal tributaries, the same assemblage of fossil animals, which has been alluded to in the last chapter as characterising the gravel of Picardy, has long been known; but it was not till the year 1860, and when diligent search had been expressly made for them, that flint implements of the Amiens type were discovered in this part of France.

In the neighbourhood of Paris deposits of drift occur answering both to those of the higher and lower levels of the basin of the Somme before described.* (* Prestwich, "Proceedings of the Royal Society" 1862.) In both are found, mingled with the wreck of the Tertiary and Cretaceous rocks of the vicinity, a large quantity of granitic sand and pebbles, and occasionally large blocks of granite, from a few inches to a foot or more in diameter. These blocks are peculiarly abundant in the lower drift commonly called the "diluvium gris." The granitic materials are traceable to a chain of hills called the Morvan, where the head waters of the Yonne take their rise, 150 miles to the south-south-east of Paris.

It was in this lowest gravel that M. H.T. Gosse, of Geneva, found, in April 1860, in the suburbs of Paris, at La Motte Piquet, on the left bank of the Seine, one or two well-formed flint implements of the Amiens type, accompanied by a great number of ruder tools or attempts at tools. I visited the spot in 1861 with M. Hebert, and saw the stratum from which the worked flints had been extracted, 20 feet below the surface, and near the bottom of the "grey diluvium," a bed of gravel from which I have myself, in and near Paris, frequently collected the bones of the elephant, horse, and other mammalia.

More recently, M. Lartet has discovered at Clichy, in the environs of Paris, in the same lower gravel, a well-shaped flint implement of the Amiens type, together with remains both of Elephas primigenius and E. antiquus. No tools have yet been met with in any of the gravels occurring at the higher levels of the valley of the Seine; but no importance can be attached to this negative fact, as so little search has yet been made for them.

Mr. Prestwich has observed contortions indicative of ice-action, of the same kind as those near Amiens, in the higher-level drift of Charonne, near Paris; but as yet no similar derangement has been seen in the lower gravels—a fact, so far as it goes, in unison with the phenomena observed in Picardy.

In the cavern of Arcy-sur-Yonne a series of deposits have lately been investigated by the Marquis de Vibraye, who discovered human bones in the lowest of them, mixed with remains of quadrupeds of extinct and recent species. This cavern occurs in Jurassic limestone, at a slight elevation above the Cure, a small tributary of the Yonne, which last joins the Seine near Fontainebleau about 40 miles south of Paris. The lowest formation in the cavern resembles the "diluvium gris" of Paris, being composed of granitic materials, and like it derived chiefly from the waste of the crystalline rocks of the Morvan. In it have been found the two branches of a human lower jaw with teeth well-preserved, and the bones of the Elephas primigenius, Rhinoceros tichorhinus, Ursus spelaeus, Hyaena spelaea, and Cervus tarandus, all specifically determined by M. Lartet. I have been shown this collection of fossils by M. de Vibraye, and remarked that the human and other remains were in the same condition and of the same colour.

Above the grey gravel is a bed of red alluvium, made up of fragments of Jurassic limestone, in a red argillaceous matrix, in which were embedded several flint knives, with bones of the reindeer and horse, but no extinct mammalia. Over this, in a higher bed of alluvium, were several polished hatchets of the more modern type called "celts," and above all loam or cave-mud, in which were Gallo-Roman antiquities.* (* "Bulletin de la Societe Geologique de France" 1860.)

The French geologists have made as yet too little progress in identifying the age of the successive deposits of ancient alluvium of various parts of the basin of the Seine, to enable us to speculate with confidence as to the coincidence in date of the granitic gravel with human bones of the Grotte d'Arcy and the stone-hatchets buried in "grey diluvium" of La Motte Piquet, before mentioned; but as the associated extinct mammalia are of the same species in both localities, I feel strongly inclined to believe that the stone hatchets found by M. Gosse at Paris, and the human bones discovered by M. de Vibraye, may be referable to the same period.

VALLEY OF THE OISE.

A flint hatchet, of the old Abbeville and Amiens type, was found lately by M. Peigne Delacourt at Precy, near Creil, on the Oise, in gravel, resembling, in its geological position, the lower-level gravels of Montiers, near Amiens, already described. I visited these extensive gravel-pits in 1861, in company with Mr. Prestwich; but we remained there too short a time to entitle us to expect to find a flint implement, even if they had been as abundant as at St. Acheul.

In 1859, I examined, in a higher part of the same valley of the Oise, near Chauny and Noyon, some fine railway cuttings, which passed continuously through alluvium of the Pleistocene period for half a mile. All this alluvium was evidently of fluviatile origin, for, in the interstices between the pebbles, the Ancylus fluviatilis and other freshwater shells were abundant. My companion, the Abbe E. Lambert, had collected from the gravel a great many fossil bones, among which M. Lartet has recognised both Elephas primigenius and E. antiquus, besides a species of hippopotamus (H. major ?), also the reindeer, horse, and the musk ox (Bubalus moschatus). The latter seems never to have been seen before in the old alluvium of France.* (* Lartet, "Annales des Sciences Naturelles Zoologiques" tome 15 page 224.) Over the gravel above mentioned, near Chauny, are seen dense masses of loam like the loess of the Rhine, containing shells of the genera Helix and Succinea. We may suppose that the gravel containing the flint hatchet at Precy is of the same age as that of Chauny, with which it is continuous, and that both of them are coeval with the tool-bearing beds of Amiens, for the basins of the Oise and the Somme are only separated by a narrow water-shed, and the same fossil quadrupeds occur in both.

The alluvium of the Seine and its tributaries, like that of the Somme, contains no fragments of rocks brought from any other hydrographical basin; yet the shape of the land, or fall of the river, or the climate, or all these conditions, must have been very different when the grey alluvium in which the flint tools occur at Paris was formed. The great size of some of the blocks of granite, and the distance which they have travelled, imply a power in the river which it no longer possesses. We can hardly doubt that river-ice once played a much more active part than now in the transportation of such blocks, one of which may be seen in the Museum of the Ecole des Mines at Paris, 3 or 4 feet in diameter.

PLEISTOCENE ALLUVIUM OF ENGLAND, CONTAINING WORKS OF ART.

In the ancient alluvium of the basin of the Thames, at moderate heights above the main river and its tributaries, we find fossil bones of the same species of extinct and living mammalia, accompanied by recent species of land and freshwater shells, as we have shown to be characteristic of the basins of the Somme and the Seine. We can scarcely therefore doubt that these quadrupeds, during some part of the Pleistocene period, ranged freely from the continent of Europe to England, at a time when there was an uninterrupted communication by land between the two countries. The reader will not therefore be surprised to learn that flint implements of the same antique type as those of the valley of the Somme have been detected in British alluvium.

The most marked feature of this alluvium in the Thames valley is that great bed of ochreous gravel, composed chiefly of broken and slightly worn Chalk flints, on which a great part of London is built. It extends from above Maidenhead through the metropolis to the sea, a distance from west to east of 50 miles, having a width varying from 2 to 9 miles. Its thickness ranges commonly from 5 to 15 feet.* (* Prestwich, "Quarterly Journal of the Geological Society" volume 12 1856 page 131.) Interstratified with this gravel, in many places, are beds of sand, loam, and clay, the whole containing occasionally remains of the mammoth and other extinct quadrupeds. Fine sections have been exposed to view, at different periods, at Brentford and Kew Bridge, others in London itself, and below it at Erith in Kent, on the right bank of the Thames, and at Ilford and Gray's Thurrock in Essex, on the left bank. The united thickness of the beds of sand, gravel, and loam amounts sometimes to 40 or even 60 feet. They are for the most part elevated above, but in some cases they descend below, the present level of the overflowed plain of the Thames.

If the reader will refer to the section of the Pleistocene sands and gravels of Menchecourt, near Abbeville, given at page 96, he will perfectly understand the relations of the ancient Thames alluvium to the modern channel and plain of the river, and their relation, on the other hand, to the boundary formations of older date, whether Tertiary or Cretaceous.

So far as they are known, the fossil mollusca and mammalia of the two districts also agree very closely, the Cyrena fluminalis being common to both, and being the only extra-European shell, this and all the species of testacea being Recent. Of this agreement with the living fauna there is a fine illustration in Essex; for the determination of which we are indebted to the late Mr. John Brown, F.G.S., who collected at Copford, in Essex, from a deposit containing bones of the mammoth, a large bear (probably Ursus spelaeus), a beaver, stag, and aurochs, no less than sixty-nine species of land and freshwater shells. Forty-eight of these were terrestrial, and two of them, Helix incarnata and H. ruderata, no longer inhabit the British Isles, but are still living on the continent, ruderata in high northern latitudes.* (* "Quarterly Journal of the Geological Society" volume 8 1852 page 190. Mr. Brown calls them extinct species, which may mislead some readers, but he merely meant extinct in England. See also Jeffreys, "Brit. Conch." page 174.) The Cyrena fluminalis and the Unio littoralis, to which last I shall presently allude, were not among the number.

I long ago suggested the hypothesis, that in the basin of the Thames there are indications of a meeting in the Pleistocene period of a northern and southern fauna. To the northern group may have belonged the mammoth (Elephas primigenius) and the Rhinoceros tichorhinus, both of which Pallas found in Siberia, preserved with their flesh in the ice. With these are occasionally associated the reindeer. In 1855 the skull of the musk ox (Bubalus moschatus) was also found in the ochreous gravel of Maidenhead, by the Reverend C. Kingsley and Mr. Lubbock; the identification of this fossil with the living species being made by Professor Owen. A second fossil skull of the same arctic animal was afterwards found by Mr. Lubbock near Bromley, in the valley of a small tributary of the Thames; and two other skulls, those of a bull and a cow were dug up near Bath Easton from the gravel of the valley of the Avon by Mr. Charles Moore. Professor Owen has truly said, that "as this quadruped has a constitution fitting it at present to inhabit the high northern regions of America, we can hardly doubt that its former companions, the warmly-clad mammoth and the two-horned woolly rhinoceros (R. tichorhinus), were in like manner capable of supporting life in a cold climate."* (* "Quarterly Journal of the Geological Society" volume 12 1856 page 124.)

I have already alluded to the recent discovery of this same ox near Chauny, in the valley of the Oise, in France; and in 1856 I found a skull of it preserved in the museum at Berlin, which Professor Quenstedt, the curator, had correctly named so long ago as 1836, when the fossil was dug out of drift, in the hill called the Kreuzberg, in the southern suburbs of that city. By an account published at the time, we find that the mammalia which accompanied the musk ox were the mammoth and tichorhine rhinoceros, with the horse and ox;* (* "Leonhard and Bronn's Jahrbuch" 1836 page 215.) but I can find no record of the occurrence of a hippopotamus, nor of Elephas antiquus or Rhinoceros leptorhinus, in the drift of the north of Germany, bordering the Baltic.

On the other hand, in another locality in the same drift of North Germany, Dr. Hensel, of Berlin, detected, near Quedlinburg, the Norwegian Lemming (Myodes lemmus), and another species of the same family called by Pallas Myodes torquatus (by Hensel, Misothermus torquatus)—a still more arctic quadruped, found by Parry in latitude 82 degrees, and which never strays farther south than the northern borders of the woody region. Professor Beyrich also informs me that the remains of the Rhinoceros tichorhinus were obtained at the same place.* (* "Zeitschrift der Deutschen Geologischen Gesellschaft" volume 7 1855 page 497 etc.)

As an example of what may possibly have constituted a more southern fauna in the valley of the Thames, I may allude to the fossil remains found in the fluviatile alluvium of Gray's Thurrock, in Essex, situated on the left bank of the river, 21 miles below London. The strata of brick-earth, loam, and gravel exposed to view in artificial excavations in that spot, are precisely such as would be formed by the silting up of an old river channel. Among the mammalia are Elephas antiquus, Rhinoceros leptorhinus (R. megarhinus, Christol), Hippopotamus major, species of horse, bear, ox, stag, etc., and, among the accompanying shells, Cyrena fluminalis, which is extremely abundant, instead of being scarce, as at Abbeville. It is associated with Unio littoralis also in great numbers and with both valves united. This conspicuous freshwater mussel is no longer an inhabitant of the British Isles, but still lives in the Seine, and is still more abundant in the Loire. Another freshwater univalve (Paludina marginata, Michaud), not British, but common in the south of France, likewise occurs, and a peculiar variety of Cyclas amnica, which by some naturalists has been regarded as a distinct species. With these, moreover, is found a peculiar variety of Valvata piscinalis.

If we consult Dr. Von Schrenck's account of the living mammalia of Mongolia, lying between latitude 45 and 55 degrees north, we learn that, in that part of North-Eastern Asia recently annexed to the Russian empire, no less than thirty-four out of fifty-eight living quadrupeds are identical with European species, while some of those which do not extend their range to Europe are arctic, others tropical forms. The Bengal tiger ranges northwards occasionally to latitude 52 degrees north, where he chiefly subsists on the flesh of the reindeer, and the same tiger abounds in latitude 48 degrees, to which the small tailless hare or pika, a polar resident, sometimes wanders southwards.* (* Mammalia of Amoorland, "Natural History Review" volume 1 1861 page 12.) We may readily conceive that the countries now drained by the Thames, the Somme, and the Seine, were, in the Pleistocene period, on the borders of two distinct zoological provinces, one lying to the north, the other to the south, in which case many species belonging to each fauna endowed with migratory habits, like the living musk-ox or the Bengal tiger, may have been ready to take advantage of any, even the slightest, change in their favour to invade the neighbouring province, whether in the summer or winter months, or permanently for a series of years, or centuries. The Elephas antiquus and its associated Rhinoceros leptorhinus may have preceded the mammoth and tichorhine rhinoceros in the valley of the Thames, or both may have alternately prevailed in the same area in the Pleistocene period.

In attempting to settle the chronology of fluviatile deposits, it is almost equally difficult to avail ourselves of the evidence of organic remains and 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. I have seen an example of this at Ilford, where the Thames, or a tributary stream, has at some former period cut through sands containing Cyrena fluminalis, and again filled up the channel with argillaceous matter, evidently derived from the waste of the Tertiary London Clay. Such shiftings of the site of the main channel of the river, the frequent removal of gravel and sand previously deposited, and the throwing down of new alluvium, the flooding of tributaries, the rising and sinking of the land, fluctuations in the cold and heat of the climate—all these changes seem to have given rise to that complexity in the fluviatile deposits of the Thames, which accounts for the small progress we have hitherto made in determining their order of succession, and that of the imbedded groups of quadrupeds. It may happen, as at Brentford and Ilford, that sand-pits in two adjoining fields may each contain distinct species of elephant and rhinoceros; and the fossil remains in both cases may occur at the same depth from the surface, yet may be severally referable to different parts of the Pleistocene epoch, separated by thousands of years.

The relation of the glacial period to alluvial deposits, such as that of Gray's Thurrock, where the Cyrena fluminalis, Unio littoralis, and the hippopotamus seem rather to imply a warmer climate, has been a matter of long and animated discussion. Patches of the northern drift, at elevations of about 200 feet above the Thames, occur in the neighbourhood of London, as at Muswell Hill, near Highgate. In this drift, blocks of granite, syenite, greenstone, Coal-measure sandstone with its fossils, and other Palaeozoic rocks, and the wreck of Chalk and Oolite, occur confusedly mixed together. The same glacial formation is also found capping some of the Essex hills farther to the east, and extending some way down their southern slopes towards the valley of the Thames. Although no fragments washed out of these older and upland drifts have been found in the gravel of the Thames containing elephants' bones, it is fair to presume, as Mr. Prestwich has contended,* that the glacial formation is the older of the two. (* Prestwich, "Quarterly Journal of the Geological Society" volume 11 1855 page 110; ibid. volume 12 1856 page 133; ibid. volume 17 1861 page 446.) In short, we must suppose that the basin of the Thames and all its fluviatile deposits are post-glacial, in the modified sense of that term; i.e. that they were subsequent to the drift of the central and northern counties.

Having offered these general remarks on the alluvium of the Thames, I may now say something of the implements hitherto discovered in it. In the British Museum there is a flint weapon of the spear-headed form, such as is represented in Figure 8, which we are told was found with an elephant's tooth at Black Mary's, near Gray's Inn Lane, London. In a letter dated 1715, printed in Herne's edition of "Leland's Collectanea," volume 1 page 73, it is stated to have been found in the presence of Mr. Conyers, with the skeleton of an elephant.* (* Evans, "Archaeologia" 1860.) So many bones of the elephant, rhinoceros, and hippopotamus have been found in the gravel on which London stands, that there is no reason to doubt the statement as handed down to us. Fossil remains of all these three genera have been dug up on the site of Waterloo Place, St. James's Square, Charing Cross, the London Docks, Limehouse, Bethnal Green, and other places within the memory of persons now living. In the gravel and sand of Shacklewell, in the north-east district of London, I have myself collected specimens of the Cyrena fluminalis in great numbers (see Figure 17 c), with the bones of deer and other mammalia.

In the alluvium also of the Wey, near Guildford, in a place called Pease Marsh, a wedge-shaped flint implement, resembling one brought from St. Acheul by Mr. Prestwich, and compared by some antiquaries to a sling-stone, was obtained in 1836 by Mr. Whitburn, 4 feet deep in sand and gravel, in which the teeth and tusks of elephants had been found. The Wey flows through the gorge of the North Downs at Guildford to join the Thames. Mr. Austen has shown that this drift is so ancient that one part of it had been disturbed and tilted before another part was thrown down.* (* "Quarterly Journal of the Geological Society" volume 7 1851 page 278.)

Among other places where flint tools of the antique type have been met with in the course of the last three years, I may mention one of an oval form found by Mr. Whitaker in the valley of the Darent, in Kent, and another which Mr. Evans found lying on the shore at Swalecliff, near Whitstable, in the same county, where Mr. Prestwich had previously described a freshwater deposit, resting on the London Clay, and consisting chiefly of gravel, in which an elephant's tooth and the bones of a bear were embedded. The flint implement was deeply discoloured and of a peculiar bright light-brown colour, similar to that of the old fluviatile gravel in the cliff.

Another flint implement was found in 1860 by Mr. T. Leech, at the foot of the cliff between Herne Bay and the Reculvers, and on further search five other specimens of the spear-head pattern so common at Amiens. Messrs. Prestwich and Evans have since found three other similar tools on the beach, at the base of the same wasting cliff, which consists of sandy Eocene strata, covered by a gravelly deposit of freshwater origin, about 50 feet above the sea-level, from which the flint weapons must have been derived. Such old alluvial deposits now capping the cliffs of Kent seem to have been the river-beds of tributaries of the Thames before the sea encroached to its present position and widened its estuary. On following up one of these freshwater deposits westward of the Reculvers, Mr. Prestwich found in it, at Chislet, near Grove Ferry, the Cyrena fluminalis among other shells.

The changes which have taken place in the physical geography of this part of England during, or since, the Pleistocene period, have consisted partly of such encroachments of the sea on the coast as are now going on, and partly of a general subsidence of the land. Among the signs of the latter movement may be mentioned a freshwater formation at Faversham, below the level of the sea. The gravel there contains exclusively land and fluviatile shells of the same species as those of other localities of the Pleistocene alluvium before mentioned, and must have been formed when the river was at a higher level and when it extended farther east. At that era it was probably a tributary of the Rhine, as represented by Mr. Trimmer in his ideal restoration of the geography of the olden time.* (* "Quarterly Journal of the Geological Society" volume 9 1853 Plate 8 Number 4.) For England was then united to the continent, and what is now the North Sea was land. It is well known that in many places, especially near the coast of Holland, elephants' tusks and other bones are often dredged up from the bed of that shallow sea, and the reader will see in the map given in Chapter 13 how vast would be the conversion of sea into land by an upheaval of 600 feet. Vertical movements of much less than half that amount would account for the annexation of England to the continent, and the extension of the Thames and its valley far to the north-east, and the flowing of rivers from the easternmost parts of Kent and Essex into the Thames, instead of emptying themselves into its estuary.

More than a dozen flint weapons of the Amiens type have already been found in the basin of the Thames; but the geological position of no one of them has as yet been ascertained with the same accuracy as that of many of the tools dug up in the valley of the Somme.

FLINT IMPLEMENTS OF THE VALLEY OF THE OUSE, NEAR BEDFORD.

The ancient fluviatile gravel of the valley of the Ouse, around Bedford, has been noted for the last thirty years for yielding to collectors a rich harvest of the bones of extinct mammalia. By observations made in 1854 and 1858, Mr. Prestwich had ascertained that the valley was bounded on both sides by Oolitic strata, capped by boulder clay, and that the gravel Number 3, Figure 23, contained bones of the elephant, rhinoceros, hippopotamus, ox, horse, and deer, which animals he therefore inferred must have been posterior in date to the boulder clay, through which, as well as the subjacent Oolite, the valley had been excavated. Mr. Evans had found in the same gravel many land and freshwater shells, and these discoveries induced Mr. James Wyatt, of Bedford, to pay two visits to St. Acheul in order to compare the implement-bearing gravels of the Somme with the drift of the valley of the Ouse. After his return he resolved to watch carefully the excavation of the gravel-pits at Biddenham, 2 miles west-north-west of Bedford, in the hope of finding there similar works of art. With this view he paid almost daily visits for months in succession to those pits, and was at last rewarded by the discovery of two well-formed implements, one of the spear-head and the other of the oval shape, perfect counterparts of the two prevailing French types. Both specimens were thrown out by the workmen on the same day from the lowest bed of stratified gravel and sand, 13 feet thick, containing bones of the elephant, deer, and ox, and many freshwater shells. The two implements occurred at the depth of 13 feet from the surface of the soil, and rested immediately on solid beds of Oolitic limestone, as represented in the accompanying section (Figure 23).

Having been invited by Mr. Wyatt to verify these facts, I went to Biddenham within a fortnight of the date of his discovery (April 1861), and, for the first time, saw evidence which satisfied me of the chronological relations of those three phenomena, the antique tools, the extinct mammalia, and the glacial formation. On that occasion I examined the pits in company with Messrs. Prestwich, Evans, and Wyatt, and we collected ten species of shells from the stratified drift Number 3, or the beds overlying the lowest gravel from which the flint implements had been exhumed. They were all of common fluviatile and land species now living in the same part of England. Since our visit, Mr. Wyatt has added to them Paludina marginata, Michaud (Hydrobia of some authors), a species of the South of France no longer inhabiting the British Isles. The same geologist has also found, since we were at Biddenham, several other flint tools of corresponding type, both there and at other localities in the valley of the Ouse, near Bedford.

(FIGURE 23. SECTION ACROSS THE VALLEY OF THE OUSE, TWO MILES WEST-NORTH-WEST OF BEDFORD.* (* Prestwich, "Quarterly Journal of the Geological Society" volume 17 1861 page 364; and Wyatt, "Geologist" 1861 page 242.)

1. Oolitic strata. 2. Boulder clay, or marine northern drift, rising to about ninety feet above the Ouse. 3. Ancient gravel, with elephant bones, freshwater shells, and flint implements. 4. Modern alluvium of the Ouse. a. Biddenham gravel pits, at the bottom of which flint tools were found.)

The boulder clay Number 2 extends for miles in all directions, and was evidently once continuous from b to c before the valley was scooped out. It is a portion of the great marine glacial drift of the midland counties of England, and contains blocks, some of large size, not only of the Oolite of the neighbourhood, but of Chalk and other rocks transported from still greater distances, such as syenite, basalt, quartz, and New Red Sandstone. These erratic blocks of foreign origin are often polished and striated, having undergone what is called glaciation, of which more will be said by and by. Blocks of the same mineral character, embedded at Biddenham in the gravel Number 3, have lost all signs of this striation by the friction to which they were subjected in the old river bed.

The great width of the valley of the Ouse, which is sometimes 2 miles, has not been expressed in the diagram. It may have been shaped out by the joint action of the river and the tides when this part of England was emerging from the waters of the glacial sea, the boulder clay being first cut through, and then an equal thickness of underlying Oolite. After this denudation, which may have accompanied the emergence of the land, the country was inhabited by the primitive people who fashioned the flint tools. The old river, aided perhaps by the continued upheaval of the whole country, or by oscillations in its level, went on widening and deepening the valley, often shifting its channel, until at length a broad area was covered by a succession of the earliest and latest deposits, which may have corresponded in age to the higher and lower gravels of the valley of the Somme, already described.

At Biddenham, and elsewhere in the same gravel, remains of Elephas antiquus have been discovered, and Mr. Wyatt obtained, January 1863, a flint implement associated with bones and teeth of hippopotamus from gravel at Summerhouse hill, which lies east of Bedford, lower down the valley of the Ouse, and 4 miles from Biddenham.

One step at least we gain by the Bedford sections, which those of Amiens and Abbeville had not enabled us to make. They teach us that the fabricators of the antique tools, and the extinct mammalia coeval with them, were all post-glacial.

FLINT IMPLEMENTS IN A FRESHWATER DEPOSIT AT HOXNE IN SUFFOLK [NOTE 17].

So early as the first year of the nineteenth century, a remarkable paper was communicated to the Society of Antiquaries by Mr. John Frere, in which he gave a clear description of the discovery at Hoxne, near Diss, in Suffolk, of flint tools of the type since found at Amiens, adding at the same time good geological reasons for presuming that their antiquity was very great, or, as he expressed it, beyond that of the present world, meaning the actual state of the physical geography of that region. "The flints," he said, "were evidently weapons of war, fabricated and used by a people who had not the use of metals. They lay in great numbers at the depth of about 12 feet in a stratified soil which was dug into for the purpose of raising clay for bricks. Under a foot and a half of vegetable earth was clay 7 1/2 feet thick, and beneath this one foot of sand with shells, and under this 2 feet of gravel, in which the shaped flints were found generally at the rate of 5 or 6 in a square yard. In the sandy beds with shells were found the jawbone and teeth of an enormous unknown animal. The manner in which the flint weapons lay would lead to the persuasion that it was a place of their manufacture, and not of their accidental deposit. Their numbers were so great that the man who carried on the brick-work told me that before he was aware of their being objects of curiosity, he had emptied baskets full of them into the ruts of the adjoining road."

Mr. Frere then goes on to explain that the strata in which the flints occur are disposed horizontally, and do not lie at the foot of any higher ground, so that portions of them must have been removed when the adjoining valley was hollowed out. If the author had not mistaken the freshwater shells associated with the tools for marine species, there would have been nothing to correct in his account of the geology of the district, for he distinctly perceived that the strata in which the implements were embedded had, since that time, undergone very extensive denudation.* (* Frere, "Archaeologia" volume 13 1800 page 206.) Specimens of the flint spear-heads, sent to London by Mr. Frere, are still preserved in the British Museum, and others are in the collection of the Society of Antiquaries.

(FIGURE 24. SECTION SHOWING THE POSITION OF THE FLINT WEAPONS AT HOXNE, NEAR DISS, SUFFOLK. See Prestwich "Philosophical Transactions" Plate 11 1860.)

1. Gravel of Gold Brook, a tributary of the Waveney. 2. Higher-level gravel overlying the freshwater deposit. 3 and 4. Sand and gravel, with freshwater shells, and flint implements, and bones of mammalia. 5. Peaty and clayey beds, with same fossils. 6. Boulder clay or glacial drift. 7. Sand and gravel below boulder clay. 8. Chalk with flints.)

Mr. Prestwich's attention was called by Mr. Evans to these weapons, as well as to Mr. Frere's memoir after his return from Amiens in 1859, and he lost no time in visiting Hoxne, a village five miles eastward of Diss. It is not a little remarkable that he should have found, after a lapse of sixty years, that the extraction of clay was still going on in the same brick-pit. Only a few months before his arrival, two flint instruments had been dug out of the clay, one from a depth of 7 and the other of 10 feet from the surface. Others have since been disinterred from undisturbed beds of gravel in the same pit. Mr. Amyot of Diss has also obtained from the underlying freshwater strata the astragalus of an elephant, and bones of the deer and horse; but although many of the old implements have recently been discovered in situ in regular strata and preserved by Sir Edward Kerrison, no bones of extinct mammalia seem as yet to have been actually seen in the same stratum with one of the tools.

By reference to the annexed section, the geologist will see that the basin-shaped hollow a, b, c has been filled up gradually with the freshwater strata 3, 4, 5, after the same cavity a, b, c had been previously excavated out of the more ancient boulder clay Number 6. The relative position of these formations will be better understood when I have described in the twelfth chapter the structure of Norfolk and Suffolk as laid open in the sea-cliffs at Mundesley, about 30 miles distant from Hoxne, in a north-north-east direction.

I examined the deposits at Hoxne in 1860, when I had the advantage of being accompanied by the Reverend J. Gunn and the Reverend S.W. King. In the loamy beds 3 and 4, Figure 24, we observed the common river shell Valvata piscinalis in great numbers. With it, but much more rare, were Limnaea palustris, Planorbis albus, P. Spirorbis, Succinea putris, Bithynia tentaculata, Cyclas cornea; and Mr. Prestwich mentions Cyclas amnica and fragments of a Unio, besides several land shells. In the black peaty mass Number 5, fragments of wood of the oak, yew, and fir have been recognised. The flint weapons which I have seen from Hoxne are so much more perfect, and have their cutting edge so much sharper than those from the valley of the Somme, that they seem neither to have been used by Man, nor to have been rolled in the bed of a river. The opinion of Mr. Frere, therefore, that there may have been a manufactory of weapons on the spot, appears probable.

FLINT IMPLEMENTS AT ICKLINGHAM IN SUFFOLK.

In another part of Suffolk, at Icklingham, in the valley of the Lark, below Bury St. Edmund's, there is a bed of gravel, in which teeth of Elephas primigenius and several flint tools, chiefly of a lance-head form, have been found. I have twice visited the spot, which has been correctly described by Mr. Prestwich.* (* "Quarterly Journal of the Geological Society" volume 17 1861, page 364.)

The section of the Bedford tool-bearing alluvium, given in Figure 23, may serve to illustrate that of Icklingham, if we substitute Chalk for Oolite, and the river Lark for the Ouse. In both cases, the present bed of the river is about 30 feet below the level of the old gravel, and the Chalk hill, which bounds the valley of the Lark on the right side, is capped like the Oolite of Biddenham by boulder clay, which rises to the height of 100 feet above the Lark. About twelve years ago, a large erratic block, above 4 feet in diameter, was dug out of the boulder clay at Icklingham, which I found to consist of a hard siliceous schist, which must have come from a remote region. The tool-bearing gravel here, as in the case to which it has been compared near Bedford, is proved to be newer than the glacial drift, by containing pebbles of basalt and other rocks derived from that formation.

CHAPTER 10.

CAVERN DEPOSITS, AND PLACES OF SEPULTURE OF THE PLEISTOCENE PERIOD.

Flint Implements in Cave containing Hyaena and other extinct Mammalia in Somersetshire. Caves of the Gower Peninsula in South Wales. Rhinoceros hemitoechus. Ossiferous Caves near Palermo. Sicily once part of Africa. Rise of Bed of the Mediterranean to the Height of three hundred Feet in the Human Period in Sardinia. Burial-place of Pleistocene Date of Aurignac in the South of France. Rhinoceros tichorhinus eaten by Man. M. Lartet on extinct Mammalia and Works of Art found in the Aurignac Cave. Relative Antiquity of the same considered.

WORKS OF ART ASSOCIATED WITH EXTINCT MAMMALIA IN A CAVERN IN SOMERSETSHIRE.

The only British cave from which implements resembling those of Amiens have been obtained, since the attention of geologists has been awakened to the importance of minutely observing the position of such relics relatively to the associated fossil mammalia, is that recently opened near Wells in Somersetshire. It occurs near the cave of Wookey Hole, from the mouth of which the river Axe issues on the southern flanks of the Mendips. No one had suspected that on the left side of the ravine, through which the river flows after escaping from its subterranean channel, there were other caves and fissures concealed beneath the green sward of the steep sloping bank. About ten years ago, a canal was made, several hundred yards in length, for the purpose of leading the waters of the Axe to a paper-mill, now occupying the middle of the ravine. In carrying out this work, about 12 feet of the left bank was cut away, and a cavernous fissure, choked up to the roof with ossiferous loam, was then, for the first time, exposed to view. This great cavity, originally 9 feet high and 36 wide, traversed the Dolomitic Conglomerate; and fragments of that rock, some angular and others water-worn, were scattered through the red mud of the cave, in which fossil remains were abundant. For an account of them and the position they occupied we are indebted to Mr. Dawkins, F.G.S., who, in company with Mr. Williamson, explored the cavern in 1859, and obtained from it the bones of the Hyaena spelaea in such numbers as to lead him to conclude that the cavern had for a long time been a hyaena's den. Among the accompanying animals found fossil in the same bone-earth, were observed Elephas primigenius, Rhinoceros tichorhinus, Ursus spelaeus, Bos primigenius, Megaceros hibernicus, Cervus tarandus (and other species of Cervus), Felis spelaea, Canis lupus, Canis vulpes, and teeth and bones of the genus Equus in great numbers.

Intermixed with the above fossil bones were some arrowheads, made of bone, and many chipped flints, and chipped pieces of chert, a white or bleached flint weapon of the spearhead Amiens type, which was taken out of the undisturbed matrix by Mr. Williamson himself, together with a hyaena's tooth, showing that Man had either been contemporaneous with or had preceded the extinct fauna. After penetrating 34 feet from the entrance, Mr. Dawkins found the cave bifurcating into two branches, one of which was vertical. By this rent, perhaps, some part of the contents of the cave may have been introduced.* (* Boyd Dawkins, "Proceedings of the Geological Society" January 1862.)

When I examined the spot in 1860, after I had been shown some remains of the hyaena collected there, I felt convinced that a complete revolution must have taken place in the topography of the district since the time of the extinct quadrupeds. I was not aware at the time that flint tools had been met with in the same bone-deposit.

CAVES OF GOWER IN GLAMORGANSHIRE, SOUTH WALES.

The ossiferous caves of the peninsula of Gower in Glamorganshire have been diligently explored of late years by Dr. Falconer and Lieutenant-Colonel E.R. Wood, who have thoroughly investigated the contents of many which were previously unknown. Among these Dr. Falconer's skilled eye has recognised the remains of almost every quadruped which he had elsewhere found fossil in British caves: in some places the Elephas primigenius, accompanied by its usual companion, the Rhinoceros tichorhinus, in others Elephas antiquus, associated with Rhinoceros hemitoechus, Falconer; the extinct animals being often embedded, as in the Belgian caves, in the same matrix with species now living in Europe, such as the common badger (Meles taxus), the common wolf, and the fox.

In a cavernous fissure called the Raven's Cliff, teeth of several individuals of Hippopotamus major, both young and old, were found; and this in a district where there is now scarce a rill of running water, much less a river in which such quadrupeds could swim. In one of the caves, called Spritsail Tor, bones of the elephants above named were observed, with a great many other quadrupeds of Recent and extinct species.

From one fissure, called Bosco's Den, no less than one thousand antlers of the reindeer, chiefly of the variety called Cervus Guettardi, were extracted by the persevering exertions of Colonel Wood, who estimated that several hundred more still remained in the bone-earth of the same rent.

They were mostly shed horns, and of young animals; and had been washed into the rent with other bones, and with angular fragments of limestone, and all enveloped in the same ochreous mud. Among the other bones, which were not numerous, were those of the cave-bear, wolf, fox, ox, stag, and field-mouse.

But the discovery of most importance, as bearing on the subject of the present work, is the occurrence in a newly-discovered cave, called Long Hole, by Colonel Wood, in 1861, of the remains of two species of rhinoceros, R. tichorhinus and R. hemitoechus, Falconer, in an undisturbed deposit, in the lower part of which were some well-shaped flint knives, evidently of human workmanship. It is clear from their position that Man was coeval with these two species. We have elsewhere independent proofs of his co-existence with every other species of the cave-fauna of Glamorganshire; but this is the first well-authenticated example of the occurrence of R. hemitoechus in connection with human implements.

In the fossil fauna of the valley of the Thames, Rhinoceros leptorhinus was mentioned as occurring at Gray's Thurrock with Elephas antiquus. Dr. Falconer, in a memoir which he is now preparing for the press on the European Pliocene and Pleistocene species of the genus Rhinoceros, has shown that, under the above name of R. leptorhinus, three distinct species have been confounded by Cuvier, Owen, and other palaeontologists:—

1. R. megarhinus, Christol, being the original and typical R. leptorhinus of Cuvier, founded on Cortesi's Monte Zago cranium, and the ONLY Pliocene, or Pleistocene European species, that had not a nasal septum.—Gray's Thurrock, etc.

2. R. hemitoechus, Falconer, in which the ossification of the septum dividing the nostrils is incomplete in the middle, besides other cranial and dental characters distinguishing it from R. tichorhinus, accompanies Elephas antiquus in most of the oldest British bone-caves, such as Kirkdale, Cefn, Durdham Down, Minchin Hole, and other Gower caverns—also found at Clacton, in Essex, and in Northamptonshire.

3. R. etruscus, Falconer, a comparatively slight and slender form, also with an incomplete bony septum,* (* Falconer, "Quarterly Journal of the Geological Society" volume 15 1859 page 602.) occurs deep in the Val d'Arno deposits, and in the "Forest bed," and superimposed blue clays, with lignite, of the Norfolk coast, but nowhere as yet found in the ossiferous caves in Britain.

Dr. Falconer announced in 1860 his opinion that the filling up of the Gower caves in South Wales took place after the deposition of the marine boulder clay,* (* Ibid. volume 16 1860 page 491.) an opinion in harmony with what we have since learnt from the section of the gravels near Bedford, given above (Figure 23), where a fauna corresponding to that of the Welsh caves characterises the ancient alluvium, and is shown to be clearly post-glacial, in the sense of being posterior in date to the boulder-clay of the midland counties. In the same sense the late Edward Forbes declared, in 1846, his conviction that not only the Cervus megaceros, but also the mammoth and other extinct pachyderms and carnivora, had lived in Britain in post-glacial times.* (* "Memoir of the Geological Survey" pages 394 to 397.) The Gower caves in general have their floors strewed over with sand, containing marine shells, all of living species; and there are raised beaches on the adjoining coast, and other geological signs of great alteration in the relative level of land and sea, since that country was inhabited by the extinct mammalia, some of which, as we have seen, were certainly coeval with Man.

OSSIFEROUS CAVES IN THE NORTH OF SICILY.

Geologists have long been familiar with the fact that on the northern coast of Sicily, between Termini on the east, and Trapani on the west, there are several caves containing the bones of extinct animals. These caves are situated in rocks of Hippurite limestone, a member of the Cretaceous series, and some of them may be seen on both sides of the Bay of Palermo. If in the neighbourhood of that city we proceed from the sea inland, ascending a sloping terrace, composed of the marine Newer Pliocene strata, we reach about a mile from the shore, and at the height of about 180 feet above it a precipice of limestone, at the base of which appear the entrances of several caves. In that of San Ciro, on the east side of the bay, we find at the bottom sand with marine shells, forty species of which have been examined, and found almost all to agree specifically with mollusca now inhabiting the Mediterranean. Higher in position, and resting on the sand, is a breccia, composed of pieces of limestone, quartz, and schist in a matrix of brown marl, through which land shells are dispersed, together with bones of two species of hippopotamus, as determined by Dr. Falconer. Certain bones of the skeleton were counted in such numbers as to prove that they must have belonged to several hundred individuals. With these were associated the remains of Elephas antiquus, and bones of the genera Bos, Cervus, Sus, Ursus, Canis, and a large Felis. Some of these bones have been rolled as if partially subjected to the action of water, and may have been introduced by streams through rents in the Hippurite limestone; but there is now no running water in the neighbourhood, no river such as the hippopotamus might frequent, not even a small brook, so that the physical geography of the district must have been altogether changed since the time when such remains were swept into fissures, or into the channels of engulfed rivers.

No proofs seem yet to have been found of the existence of Man at the period when the hippopotamus and Elephas antiquus flourished at San Ciro. But there is another cave called the Grotto di Maccagnone, which much resembles it in geological position, on the opposite or west side of the Bay of Palermo, near Carini. In the bottom of this cave a bone deposit like that of San Ciro occurs, and above it other materials reaching to the roof, and evidently washed in from above, through crevices in the limestone. In this upper and newer breccia Dr. Falconer discovered flint knives, bone splinters, bits of charcoal, burnt clay, and other objects indicating human intervention, mingled with entire land shells, teeth of horses, coprolites of hyaenas, and other bones, the whole agglutinated to one another and to the roof by the infiltration of water holding lime in solution. The perfect condition of the large fragile helices (Helix vermiculata) afforded satisfactory evidence, says Dr. Falconer, that the various articles were carried into the cave by the tranquil agency of water, and not by any tumultuous action. At a subsequent period other geographical changes took place, so that the cave, after it had been filled, was washed out again, or emptied of its contents with the exception of those patches of breccia which, being cemented together by stalactite, still adhere to the roof.* (* "Quarterly Journal of the Geological Society" volume 16 1860 page 105.)

Baron Anca, following up these investigations, explored, in 1859, another cave at Mondello, west of Palermo, and north of Mount Gallo, where he discovered molars of the living African elephant, and afterwards additional specimens of the same species in the neighbouring grotto of Olivella. In reference to this elephant, Dr. Falconer has reminded us that the distance between the nearest part of Sicily and the coast of Africa, between Marsala and Cape Bon, is not more than 80 miles, and Admiral Smyth, in his Memoir on the Mediterranean, states (page 499) that there is a subaqueous plateau, named by him Adventure Bank, uniting Sicily to Africa by a succession of ridges which are not more than from 40 to 50 fathoms under water.* (* Cited by Horner, "Presidential Address to the Geological Society" 1861 page 42.) Sicily therefore might be re-united to Africa by movements of upheaval not greater than those which are already known to have taken place within the human period on the borders of the Mediterranean, of which I shall now proceed to cite a well-authenticated example, observed in Sardinia.

RISE OF THE BED OF THE SEA TO THE HEIGHT OF 300 FEET, IN THE HUMAN PERIOD, IN SARDINIA.

Count Albert de la Marmora, in his description of the geology of Sardinia,* (* "Partie Geologique" volume 1 pages 382 and 387.) has shown that on the southern coast of that island, at Cagliari and in the neighbourhood, an ancient bed of the sea, containing marine shells of living species, and numerous fragments of antique pottery, has been elevated to the height of from 230 to 324 feet above the present level of the Mediterranean. Oysters and other shells, of which a careful list has been published, including the common mussel (Mytilus edulis), many of them having both valves united, occur, embedded in a breccia in which fragments of limestone abound. The mussels are often in such numbers as to impart, when they have decomposed, a violet colour to the marine stratum. Besides pieces of coarse pottery, a flattened ball of baked earthenware, with a hole through its axis, was found in the midst of the marine shells. It is supposed to have been used for weighting a fishing net. Of this and of one of the fragments of ancient pottery Count de la Marmora has given figures.

The upraised bed of the sea probably belongs in this instance to the Pleistocene period, for in a bone breccia, filling fissures in the rocks around Cagliari, the remains of extinct mammalia have been detected; among which is a new genus of carnivorous quadruped, named Cynotherium by M. Studiati, and figured by Count de la Marmora in his Atlas (Plate 7), also an extinct species of Lagomys, determined by Cuvier in 1825. Embedded in the same bone-breccia, and enveloped with red earth like the mammalian remains, were detected shells of the Mytilus edulis before mentioned, implying that the marine formation containing shells and pottery had been already upheaved and exposed to denudation before the remains of quadrupeds were washed into these rents and included in the red earth. In the vegetable soil covering the upraised marine stratum, fragments of Roman pottery occur.

If we assume the average rate of upheaval to have been, as before hinted, 2 1/2 feet in a century, 300 feet would give an antiquity of 12,000 years to the Cagliari pottery, even if we simply confine our estimate to the upheaval above the sea-level, without allowing for the original depth of water in which the mollusca lived. Even then our calculation would merely embrace the period during which the upward movement was going on; and we can form at present no conjecture as to the probable era of its commencement or termination.

I learn from Captain Spratt, R.N., that the island of Crete or Candia, about 135 miles in length, has been raised at its western extremity about 25 feet; so that ancient ports are now high and dry above the sea, while at its eastern end it has sunk so much that the ruins of old towns are seen under water. Revolutions like these in the physical geography of the countries bordering the Mediterranean, may well help us to understand the phenomena of the Palermo caves, and the presence in Sicily of African species of mammalia.

CLIMATE AND HABITS OF THE HIPPOPOTAMUS.

As I have alluded more than once in this chapter to the occurrence of the remains of the hippopotamus in places where there are now no rivers, not even a rill of water, and as other bones of the same genus have been met with in the lower-level gravels of the Somme where large blocks of sandstone seem to imply that ice once played a part in their transportation, it may be well to consider, before proceeding farther, what geographical and climatal conditions are indicated by the presence of these fossil pachyderms.

It is now very generally conceded that the mammoth and tichorhine rhinoceros were fitted to inhabit northern regions, and it is therefore natural to begin by asking whether the extinct hippopotamus may not in like manner have flourished in a cold climate. In answer to this inquiry, it has been remarked that the living hippopotami, anatomically speaking so closely allied to the extinct species, are so aquatic and fluviatile in their habits as to make it difficult to conceive that their congeners could have thriven all the year round in regions where, during winter, the rivers were frozen over for months. Moreover, I have been unable to learn that, in any instance, bones of the hippopotamus have been found in the drift of northern Germany associated with the remains of the mammoth, tichorhine rhinoceros, musk-ox, reindeer, lemming, and other arctic quadrupeds before alluded to; yet, though not proved to have ever made a part of such a fauna, the presence of the fossil hippopotamus north of the fiftieth parallel of latitude naturally tempts us to speculate on the migratory powers and instincts of some of the extinct species of the genus. They may have resembled, in this respect, the living musk-ox, herds of which pass for hundreds of miles over the ice to the rich pastures of Melville Island, and then return again to southern latitudes before the ice breaks up.

We are indebted to Sir Andrew Smith,* (* "Illustrations of the Zoology of South Africa": article "Hippopotamus.") an experienced zoologist, for having given us an account of the migratory habits of the living hippopotamus of Southern Africa (H. amphibius, Linn. ).

He states that, when the Dutch first colonised the Cape of Good Hope, this animal abounded in all the great rivers, as far south as the land extends; whereas, in 1849, they had all disappeared, scarcely one remaining even within a moderate distance of the colony. He also tells us that this species evinces great sagacity in changing its quarters whenever danger threatens, quitting every district invaded by settlers bearing fire-arms. Bulky as they are, they can travel speedily for miles over land from one pool of a dried-up river to another; but it is by water that their powers of locomotion are surpassingly great, not only in rivers, but in the sea, for they are far from confining themselves to fresh water. Indeed, Sir A. Smith finds it "difficult to decide whether, during the daytime and when not feeding, they prefer the pools of rivers or the waters of the ocean for their abode." In districts where they have been disturbed by Man, they feed almost entirely in the night, chiefly on certain kinds of grass, but also on brushwood. Sir A. Smith relates that, in an expedition which he made north of Port Natal, he found them swarming in all the rivers about the tropic of Capricorn. Here they were often seen to have left their footprints on the sands, entering or coming out of the salt water; and on one occasion Smith's party tried in vain to intercept a female with her young as she was making her way to the sea. Another female, which they had wounded on her precipitate retreat to the sea, was afterwards shot in that element.

The geologist, therefore, may freely speculate on the time when herds of hippopotami issued from North African rivers, such as the Nile, and swam northwards in summer along the coasts of the Mediterranean, or even occasionally visited islands near the shore. Here and there they may have landed to graze or browse, tarrying awhile and afterwards continuing their course northwards. Others may have swum in a few summer days from rivers in the south of Spain or France to the Somme, Thames, or Severn, making timely retreat to the south before the snow and ice set in.

BURIAL-PLACE AT AURIGNAC, IN THE SOUTH OF FRANCE, OF PLEISTOCENE DATE.

I have alluded in the beginning of the fourth chapter to a custom prevalent among rude nations of consigning to the tomb works of art, once the property of the dead, or objects of their affection, and even of storing up, in many cases, animal food destined for the manes of the defunct in a future life. I also cited M. Desnoyers' comments on the absence among the bones of wild and domestic animals found in old Gaulish tombs of all intermixture of extinct species of quadrupeds, as proving that the oldest sepulchral monuments then known in France (1845) had no claims to high antiquity founded on palaeontological data.

M. Lartet, however, has recently published a circumstantial account of what seems clearly to have been a sepulchral vault of the Pleistocene period, near Aurignac, not far from the foot of the Pyrenees. I have had the advantage of inspecting the fossil bones and works of art obtained by him from that grotto, and of conversing and corresponding with him on the subject, and can see no grounds for doubting the soundness of his conclusions.* (* See Lartet, "Annales des Sci. Nat." 4mo. Ser. Zoologie volume 15 page 177 translated in "Natural History Review" London January 1862.)

(FIGURE 25. SECTION OF PART OF THE HILL OF FAJOLES PASSING THROUGH THE SEPULCHRAL GROTTO OF AURIGNAC (E. Lartet).

a. Part of the vault in which the remains of seventeen human skeletons were found. b. Layer of made ground, two feet thick, inside the grotto in which a few human bones, with entire bones of extinct and living species of animals, and many works of art were embedded. c. Layers of ashes and charcoal, six inches thick, with broken, burnt, and gnawed bones of extinct and Recent mammalia; also hearth-stones and works of art; no human bones. d. Deposit with similar contents and a few scattered cinders. e. Talus of rubbish washed down from the hill above. f, g. Slab of rock which closed the vault, not ascertained whether it extended to h. f i. Rabbit burrow which led to the discovery of the grotto. h, k. Original terrace on which the grotto opened. N. Nummulitic limestone of hill of Fajoles.)

The town of Aurignac is situated in the department of the Haute-Garonne, near a spur of the Pyrenees; adjoining it is the small flat-topped hill of Fajoles, about 60 feet above the brook called Rodes, which flows at its foot on one side. It consists of Nummulitic limestone, presenting a steep escarpment towards the north-west, on which side in the face of the rock, about 45 feet above the brook, is now visible the entrance of a grotto a, Figure 25, which opened originally on the terrace h, c, k, which slopes gently towards the valley.

Until the year 1852, the opening into this grotto was masked by a talus of small fragments of limestone and earthy matter e, such as the rain may have washed down the slope of the hill. In that year a labourer named Bonnemaison, employed in repairing the roads, observed that rabbits, when hotly pursued by the sportsman, ran into a hole which they had burrowed in the talus, at i f, Figure 25. On reaching as far into the opening as the length of his arm, he drew out, to his surprise, one of the long bones of the human skeleton; and his curiosity being excited, and having a suspicion that the hole communicated with a subterranean cavity, he commenced digging a trench through the middle of the talus, and in a few hours found himself opposite a large heavy slab of rock f h, placed vertically against the entrance. Having removed this, he discovered on the other side of it an arched cavity a, 7 or 8 feet in its greatest height, 10 in width, and 7 in horizontal depth. It was almost filled with bones, among which were two entire skulls, which he recognised at once as human. The people of Aurignac, astonished to hear of the occurrence of so many human relics in so lonely a spot, flocked to the cave, and Dr. Amiel, the Mayor, ordered all the bones to be taken out and reinterred in the parish cemetery. But before this was done, having as a medical man a knowledge of anatomy, he ascertained by counting the homologous bones that they must have formed parts of no less than seventeen skeletons of both sexes, and all ages; some so young that the ossification of some of the bones was incomplete. Unfortunately the skulls were injured in the transfer; and what is worse, after the lapse of eight years, when M. Lartet visited Aurignac, the village sexton was unable to tell him in what exact place the trench was dug, into which the skeletons had been thrown, so that this rich harvest of ethnological knowledge seems for ever lost to the antiquary and geologist.

M. Lartet having been shown, in 1860, the remains of some extinct animals and works of art, found in digging the original trench made by Bonnemaison through the bed d under the talus, and some others brought out from the interior of the grotto, determined to investigate systematically what remained intact of the deposits outside and inside the vault, those inside, underlying the human skeletons, being supposed to consist entirely of made ground. Having obtained the assistance of some intelligent workmen, he personally superintended their labours, and found outside the grotto, resting on the sloping terrace h k, the layer of ashes and charcoal c, about 6 inches thick, extending over an area of 6 or 7 square yards, and going as far as the entrance of the grotto and no farther, there being no cinders or charcoal in the interior. Among the cinders outside the vault were fragments of fissile sandstone, reddened by heat, which were observed to rest on a levelled surface of Nummulitic limestone and to have formed a hearth. The nearest place from whence such slabs of sandstone could have been brought was the opposite side of the valley.

Among the ashes, and in some overlying earthy layers, d, separating the ashes from the talus e, were a great variety of bones and implements; amongst the latter not fewer than a hundred flint articles—knives, projectiles, sling stones, and chips, and among them one of those siliceous cores or nuclei with numerous facets, from which flint flakes or knives had been struck off, seeming to prove that some instruments were occasionally manufactured on the very spot.

Among other articles outside the entrance was found a stone of a circular form, and flattened on two sides, with a central depression, composed of a tough rock which does not belong to that region of the Pyrenees. This instrument is supposed by the Danish antiquaries to have been used for removing by skilful blows the edges of flint knives, the fingers and thumb being placed in the two opposite depressions during the operation. Among the bone instruments were arrows without barbs, and other tools made of reindeer horn, and a bodkin formed out of the more compact horn of the roedeer. This instrument was well shaped, and sharply pointed, and in so good a state of preservation that it might still be used for piercing the tough skins of animals.

Scattered through the same ashes and earth were the bones of the various species of animals enumerated in the subjoined lists, with the exception of two, marked with an asterisk, which only occurred in the interior of the grotto:—

TABLE 10/1. NUMBERS OF INDIVIDUALS, BONES OF WHICH WERE FOUND IN THE AURIGNAC CAVE.

COLUMN 1 : NAME OF SPECIES.

COLUMN 2 : NUMBER OF INDIVIDUALS.

1. CARNIVORA

1. Ursus spelaeus (cave-bear) : 5 to 6. 2. Ursus arctos ? (brown bear) : 1. 3. Meles taxus (badger) : 1 to 2. 4. Putorius vulgaris (polecat) : 1. 5. *Felis spelaea (cave-lion) : 1. 6. Felis catus ferus (wild cat) : 1. 7. Hyaena spelaea (cave-hyaena) : 5 to 6. 8. Canis lupus (wolf) : 3. 9. Canis vulpes (fox) : 18 to 20.

2. HERBIVORA.

1. Elephas primigenius (mammoth, two molars). 2. Rhinoceros tichorhinus (Siberian rhinoceros) : 1. 3. Equus caballus (horse) : 12 to 15. 4. Equus asinus (?) (ass) : 1. 5. *Sus scrofa (pig, two incisors). 6. Cervus elaphus (stag) : 1. 7. Megaceros hibernicus (gigantic Irish deer) : 1. 8. C. capreolus (roebuck) : 3 to 4. 9. C. tarandus (reindeer) : 10 to 12. 10. Bison europaeus (aurochs) : 12 to 15.

The bones of the herbivora were the most numerous, and all those on the outside of the grotto which had contained marrow were invariably split open, as if for its extraction, many of them being also burnt. The spongy parts, moreover, were wanting, having been eaten off and gnawed after they were broken, the work, according to M. Lartet, of hyaenas, the bones and coprolites of which were mixed with the cinders, and dispersed through the overlying soil d. These beasts of prey are supposed to have prowled about the spot and fed on such relics of the funeral feasts as remained after the retreat of the human visitors, or during the intervals between successive funeral ceremonies which accompanied the interment of the corpses within the sepulchre. Many of the bones were also streaked, as if the flesh had been scraped off by a flint instrument.

Among the various proofs that the bones were fresh when brought to the spot, it is remarked that those of the herbivora not only bore the marks of having had the marrow extracted and having afterwards been gnawed and in part devoured as if by carnivorous beasts, but that they had also been acted upon by fire (and this was especially noticed in one case of a cave-bear's bone), in such a manner as to show that they retained in them at the time all their animal matter.

Among other quadrupeds which appear to have been eaten at the funeral feasts, and of which the bones occurred among the ashes, were those of a young Rhinoceros tichorhinus, the bones of which had been, like those of the accompanying herbivora, broken and gnawed by a beast of prey at both extremities.