English : 21 : 13 3/4 : 12 1/2 : 4 4/10 : 7 7/8 : 5 1/3.

"In making the preceding statement, it must be clearly understood that I neither desire to affirm that the Engis and Neanderthal skulls belong to the Australian race, nor to assert even that the ancient skulls belong to one and the same race, so far as race is measured by language, colour of skin, or character of hair. Against the conclusion that they are of the same race as the Australians various minor anatomical differences of the ancient skulls, such as the great development of the frontal sinuses, might be urged; while against the supposition of either the identity, or the diversity, of race of the two arises the known independence of the variation of cranium on the one hand, and of hair, colour, and language on the other.

"But the amount of variation of the Borreby skulls, and the fact that the skulls of one of the purest and most homogeneous of existing races of men can be proved to differ from one another in the same characters, though perhaps not quite to the same extent, as the Engis and Neanderthal skulls, seem to me to prohibit any cautious reasoner from affirming the latter to have been necessarily of distinct races.

(FIGURE 6. OUTLINES OF THE SKULL FROM THE NEANDERTHAL, OF AN AUSTRALIAN SKULL FROM PORT ADELAIDE, AND OF THE SKULL FROM THE CAVE OF ENGIS, DRAWN TO THE SAME ABSOLUTE LENGTH, IN ORDER THE BETTER TO CONTRAST THEIR PROPORTIONS.

a. The glabella. b. The occipital protuberance, or the point on the exterior of each skull which corresponds roughly with the attachment of the tentorium, or with the inferior boundary of the posterior cerebral lobes. e. The position of the auditory foramen of the Engis skull.)

"The marked resemblances between the ancient skulls and their modern Australian analogues, however, have a profound interest, when it is recollected that the stone axe is as much the weapon and the implement of the modern as of the ancient savage; that the former turns the bones of the kangaroo and of the emu to the same account as the latter did the bones of the deer and the urus; that the Australian heaps up the shells of devoured shellfish in mounds which represent the "refuse-heaps" or "Kjokkenmodding," of Denmark; and, finally, that, on the other side of Torres Straits, a race akin to the Australians are among the few people who now build their houses on pile-works, like those of the ancient Swiss lakes.

"That this amount of resemblance in habit and in the conditions of existence is accompanied by as close a resemblance in cranial configuration, illustrates on a great scale that what Cuvier demonstrated of the animals of the Nile valley is no less true of men; circumstances remaining similar, the savage varies little more, it would seem, than the ibis or the crocodile, especially if we take into account the enormous extent of the time over which our knowledge of man now extends, as compared with that measured by the duration of the sepulchres of Egypt.

"Finally, the comparatively large cranial capacity of the Neanderthal skull, overlaid though it may be by pithecoid bony walls, and the completely human proportions of the accompanying limb-bones, together with the very fair development of the Engis skull, clearly indicate that the first traces of the primordial stock whence Man has proceeded need no longer be sought, by those who entertain any form of the doctrine of progressive development, in the newest Tertiaries; but that they may be looked for in an epoch more distant from the age of the Elephas primigenius than that is from us."

The two skulls which form the subject of the preceding comments and illustrations have given rise to nearly an equal amount of surprise for opposite reasons; that of Engis because being so unequivocally ancient, it approached so near to the highest or Caucasian type; that of the Neanderthal, because, having no such decided claims to antiquity, it departs so widely from the normal standard of humanity. Professor Huxley's observation regarding the wide range of variation, both as to shape and capacity, in the skulls of so pure a race as the native Australian, removes to no small extent this supposed anomaly, assuming what though not proved is very probable, that both varieties co-existed in the Pleistocene period in Western Europe.

As to the Engis skull, we must remember that although associated with the elephant, rhinoceros, bear, tiger, and hyaena, all of extinct species, it nevertheless is also accompanied by a bear, stag, wolf, fox, beaver, and many other quadrupeds of species still living. Indeed many eminent palaeontologists, and among them Professor Pictet, think that, numerically considered, the larger portion of the mammalian fauna agrees specifically with that of our own period, so that we are scarcely entitled to feel surprised if we find human races of the Pleistocene epoch undistinguishable from some living ones. It would merely tend to show that Man has been as constant in his osteological characters as many other mammalia now his contemporaries. The expectation of always meeting with a lower type of human skull, the older the formation in which it occurs, is based on the theory of progressive development, and it may prove to be sound; nevertheless we must remember that as yet we have no distinct geological evidence that the appearance of what are called the inferior races of mankind has always preceded in chronological order that of the higher races.

It is now admitted that the differences between the brain of the highest races of Man and that of the lowest,* (* "Natural History Review" 1861 page 8.) though less in degree, are of the same order as those which separate the human from the simian brain; and the same rule holds good in regard to the shape of the skull. The average Negro skull differs from that of the European in having a more receding forehead, more prominent superciliary ridges, and more largely developed prominences and furrows for the attachment of muscles; the face also, and its lines, are larger proportionally. The brain is somewhat less voluminous on the average in the lower races of mankind, its convolutions rather less complicated, and those of the two hemispheres more symmetrical, in all which points an approach is made to the simian type. It will also be seen, by reference to the late Dr. Morton's works, and by the foregoing statements of Professor Huxley, that the range of size or capacity between the highest and lowest human brain is greater than that between the highest simian and lowest human brain; but the Neanderthal skull, although in several respects it is more ape-like than any human skull previously discovered, is, in regard to volume, by no means contemptible.

Eminent anatomists have shown that in the average proportions of some of the bones the Negro differs from the European, and that in most of these characters, he makes a slightly nearer approach to the anthropoid quadrumana;* but Professor Schaaffhausen has pointed out that in these proportions the Neanderthal skeleton does not differ from the ordinary standard, so that the skeleton by no means indicates a transition between Homo and Pithecus. (* "The inferior races of mankind exhibit proportions which are in many respects intermediate between the higher, or European, orders, and the monkeys. In the Negro, for instance, the stature is less than in the European. The cranium, as is well known, bears a small proportion to the face. Of the extremities the upper are proportionately longer, and there is, in both upper and lower, a less marked preponderance of the proximal over the distal segments. For instance, in the Negro, the thigh and arm are rather shorter than in the European; the leg is actually of equal length in both races, and is therefore, relatively, a little longer in the Negro; the fore-arm in the latter is actually, as well as relatively, a little longer; the foot is an eighth, and the hand a twelfth longer than in the European. It is well known that the foot is less well formed in the Negro than in the European. The arch of the instep, the perfect conformation of which is essential to steadiness and ease of gait, is less elevated in the former than in the latter. The foot is thereby rendered flatter as well as longer, more nearly resembling the monkey's, between which and the European there is a marked difference in this particular."—From "A Treatise on the Human Skeleton" by Dr. Humphry, Lecturer on Surgery and Anatomy in the Cambridge University Medical School, page 91.)

There is doubtless, as shown in the diagram Figure 4, a nearer resemblance in the outline of the Neanderthal skull to that of a chimpanzee than had ever been observed before in any human cranium; and Professor Huxley's description of the occipital region shows that the resemblance is not confined to the mere excessive prominence of the superciliary ridges.

The direct bearing of the ape-like character of the Neanderthal skull on Lamarck's doctrine of progressive development and transmutation, or on that modification of it which has of late been so ably advocated by Mr. Darwin, consists in this, that the newly observed deviation from a normal standard of human structure is not in a casual or random direction, but just what might have been anticipated if the laws of variation were such as the transmutationists require. For if we conceive the cranium to be very ancient, it exemplifies a less advanced stage of progressive development and improvement. If it be a comparatively modern race, owing its peculiarities of conformation to degeneracy, it is an illustration of what botanists call "atavism," or the tendency of varieties to revert to an ancestral type, which type, in proportion to its antiquity, would be of lower grade. To this hypothesis, of a genealogical connection between Man and the lower animals, I shall again allude in the concluding chapters. [Note 11.]

CHAPTER 6.

PLEISTOCENE ALLUVIUM AND CAVE DEPOSITS WITH FLINT IMPLEMENTS.

General Position of Drift with extinct Mammalia in Valleys. Discoveries of M. Boucher de Perthes at Abbeville. Flint Implements found also at St. Acheul, near Amiens. Curiosity awakened by the systematic Exploration of the Brixham Cave. Flint Knives in same, with Bones of extinct Mammalia. Superposition of Deposits in the Cave. Visits of English and French Geologists to Abbeville and Amiens.

PLEISTOCENE ALLUVIUM CONTAINING FLINT IMPLEMENTS IN THE VALLEY OF THE SOMME.

Throughout a large part of Europe we find at moderate elevations above the present river-channels, usually at a height of less than 40 feet, but sometimes much higher, beds of gravel, sand, and loam containing bones of the elephant, rhinoceros, horse, ox, and other quadrupeds, some of extinct, others of living, species, belonging for the most part to the fauna already alluded to in the fourth chapter as characteristic of the interior of caverns. The greater part of these deposits contain fluviatile shells, and have undoubtedly been accumulated in ancient river-beds. These old channels have long since been dry, the streams which once flowed in them having shifted their position, deepening the valleys, and often widening them on one side.

It has naturally been asked, if Man co-existed with the extinct species of the caves, why were his remains and the works of his hands never embedded outside the caves in ancient river-gravel containing the same fossil fauna? Why should it be necessary for the geologist to resort for evidence of the antiquity of our race to the dark recesses of underground vaults and tunnels which may have served as places of refuge or sepulture to a succession of human beings and wild animals, and where floods may have confounded together in one breccia the memorials of the fauna of more than one epoch? Why do we not meet with a similar assemblage of the relics of Man, and of living and extinct quadrupeds, in places where the strata can be thoroughly scrutinised in the light of day?

Recent researches have at length demonstrated that such memorials, so long sought for in vain, do in fact exist, and their recognition is the chief cause of the more favourable reception now given to the conclusions which MM. Tournal, Christol, Schmerling, and others, arrived at thirty years ago respecting the fossil contents of caverns. [Note 12.]

A very important step in this new direction was made thirteen years after the publication of Schmerling's researches, by M. Boucher de Perthes, who found in ancient alluvium at Abbeville, in Picardy, some flint implements, the relative antiquity of which was attested by their geological position. The antiquarian knowledge of their discoverer enabled him to recognise in their rude and peculiar type a character distinct from that of the polished stone weapons of a later period, usually called "celts." In the first volume of his "Antiquites Celtiques," published in 1847, M. Boucher de Perthes styled these older tools "antediluvian," because they came from the lowest beds of a series of ancient alluvial strata bordering the valley of the Somme, which geologists had termed "diluvium." He had begun to collect these implements in 1841. From that time they had been annually dug out of the drift or deposits of gravel and sand, of which fine sections were laid open from 20 to 35 feet in depth, whenever excavations were made in repairing the fortifications of Abbeville; or as often as flints were wanted for the roads, or loam for making bricks. For years previously bones of quadrupeds of the genera elephant, rhinoceros, bear, hyaena, stag, ox, horse, and others, had been collected there, and sent from time to time to Paris to be examined and named by Cuvier, who had described them in his Ossements Fossiles. A correct account of the associated flint tools and of their position was given in 1847 by M. Boucher de Perthes in his work above cited, and they were stated to occur at various depths, often 20 or 30 feet from the surface, in sand and gravel, especially in those strata which were nearly in contact with the subjacent white Chalk. But the scientific world had no faith in the statement that works of art, however rude, had been met with in undisturbed beds of such antiquity. Few geologists visited Abbeville in winter, when the sand-pits were open, and when they might have opportunities of verifying the sections, and judging whether the instruments had really been embedded by natural causes in the same strata with the bones of the mammoth, rhinoceros, and other extinct mammalia. Some of the tools figured in the "Antiquites Celtiques" were so rudely shaped, that many imagined them to have owed their peculiar forms to accidental fracture in a river's bed; others suspected frauds on the part of the workmen, who might have fabricated them for sale, or that the gravel had been disturbed, and that the worked flints had got mingled with the bones of the mammoth long after that animal and its associates had disappeared from the earth.

No one was more sceptical than the late eminent physician of Amiens, Dr. Rigollot, who had long before (in the year 1819) written a memoir on the fossil mammalia of the valley of the Somme. He was at length induced to visit Abbeville, and, having inspected the collection of M. Boucher de Perthes, returned home resolved to look for himself for flint tools in the gravel-pits near Amiens. There, accordingly, at a distance of about 30 miles from Abbeville, he immediately found abundance of similar flint implements, precisely the same in the rudeness of their make, and the same in their geological position; some of them in gravel nearly on a level with the Somme, others in similar deposits resting on Chalk at a height of about 90 feet above the river.

Dr. Rigollot having in the course of four years obtained several hundred specimens of these tools, most of them from St. Acheul in the south-east suburbs of Amiens, lost no time in communicating an account of them to the scientific world, in a memoir illustrated by good figures of the worked flints and careful sections of the beds. These sections were executed by M. Buteux, an engineer well qualified for the task, who had written a good description of the geology of Picardy. Dr. Rigollot, in this memoir, pointed out most clearly that it was not in the vegetable soil, nor in the brick-earth with land and freshwater shells next below, but in the lower beds of coarse flint-gravel, usually 12, 20, or 25 feet below the surface, that the implements were met with, just as they had been previously stated by M. Boucher de Perthes to occur at Abbeville. The conclusion, therefore, which was legitimately deduced from all the facts, was that the flint tools and their fabricators were coeval with the extinct mammalia embedded in the same strata.

BRIXHAM CAVE, NEAR TORQUAY, DEVONSHIRE.

Four years after the appearance of Dr. Rigollot's paper, a sudden change of opinion was brought about in England respecting the probable co-existence, at a former period, of Man and many extinct mammalia, in consequence of the results obtained from a careful exploration of a cave at Brixham, near Torquay, in Devonshire. As the new views very generally adopted by English geologists had no small influence on the subsequent progress of opinion in France, I shall interrupt my account of the researches made in the valley of the Somme, by a brief notice of those which were carried on in 1858 in Devonshire with more than usual care and scientific method. Dr. Buckland, in his celebrated work, entitled "Reliquiae Diluvianae," published in 1823, in which he treated of the organic remains contained in caves, fissures, and "diluvial gravel" in England, had given a clear statement of the results of his own original observations, and had declared that none of the human bones or stone implements met with by him in any of the caverns could be considered to be as old as the mammoth and other extinct quadrupeds. Opinions in harmony with this conclusion continued until very lately to be generally in vogue in England; although about the time that Schmerling was exploring the Liege caves, the Reverend Mr. McEnery, a Catholic priest, residing near Torquay, had found in a cave one mile east of that town, called "Kent's Hole," in red loam covered with stalagmite, not only bones of the mammoth, tichorhine rhinoceros, hippopotamus, cave-bear, and other mammalia, but several remarkable flint tools, some of which he supposed to be of great antiquity, while there were also remains of Man in the same cave of a later date.* (* The manuscript and plates prepared for a joint memoir on Kent's Hole, by Mr. McEnery and Dr. Buckland, have recently been published by Mr. Vivian of Torquay, from which, as well as from some of the unprinted manuscript, I infer that Mr. McEnery only refrained out of deference to Dr. Buckland from declaring his belief in the contemporaneousness of certain flint implements of an antique type and the bones of extinct animals. Two of these implements from Kent's Hole, figured in Plate 12 of the posthumous work above alluded to, approach very closely in form and size to the common Abbeville implements.)

About ten years afterwards, in a "Memoir on the Geology of South Devon," published in 1842 by the Geological Society of London,* (* "Transactions of the Geological Society" 2nd series volume 6 page 444.) an able geologist, Mr. Godwin-Austen, declared that he had obtained in the same cave (Kent's Hole) works of Man from undisturbed loam or clay, under stalagmite, mingled with the remains of extinct animals, and that all these must have been introduced "before the stalagmite flooring had been formed." He maintained that such facts could not be explained away by the hypothesis of sepulture, as in Dr. Buckland's well-known case of the human skeleton of Paviland, because in the Devon cave the flint implements were widely distributed through the loam, and lay beneath the stalagmite.

As the osseous and other contents of Kent's Hole had, by repeated diggings, been thrown into much confusion, it was thought desirable in 1858, when a new and intact bone-cave was discovered at Brixham, about four miles south of Torquay, to have a thorough and systematic examination made of it. The Royal Society, chiefly at the instance of Dr. Falconer, made two grants towards defraying the expenses, and Miss Burdett-Coutts contributed liberally towards the same object. A committee of geologists was charged with the investigations, among whom Dr. Falconer and Mr. Prestwich took a prominent part, visiting Torquay while the excavations were in progress. Mr. Pengelly, another member of the committee, well qualified for the task by nearly twenty years' previous experience in cave explorations, zealously directed and superintended the work. By him, in 1859, I was conducted through the subterranean galleries after they had been cleared out; and Dr. Falconer, who was also at Torquay, showed me the numerous fossils which had been discovered, and which he was then studying, all numbered and labelled, with reference to a journal in which the geological position of each specimen was recorded with scrupulous care.

The discovery of the existence of this suite of caverns near the sea at Brixham was made accidentally by the roof of one of them being broken through in quarrying. None of the four external openings now exposed to view in steep cliffs or in the sloping side of a valley were visible before the breccia and earthy matter which blocked them up were removed during the late exploration. According to a ground-plan drawn up by Professor Ramsay, it appears that some of the passages which run nearly north and south are fissures connected with the vertical dislocation of the rocks, while another set, running nearly east and west, are tunnels, which have the appearance of having been to a great extent hollowed out by the action of running water. The central or main entrance, leading to what is called the "reindeer gallery," because a perfect antler of that animal was found sticking in the stalagmitic floor, is 95 feet above the level of the sea, being also 78 above the bottom of the adjoining valley. The united length of the galleries which were cleared out amounted to several hundred feet. Their width never exceeded 8 feet. They were sometimes filled up to the roof with mud, but occasionally there was a considerable space between the roof and floor. The latter, in the case of the fissure-caves, was covered with stalagmite, but in the tunnels it was usually free from any such incrustation. The following was the general succession of the deposits forming the contents of the underground passages and channels:—

First. At the top, a layer of stalagmite varying in thickness from 1 to 15 inches, which sometimes contained bones, such as the reindeer's horn, already mentioned, and an entire humerus of the cave-bear.

Secondly. Next below, loam or bone-earth, of an ochreous red colour, with angular stones and some pebbles, from 2 to 13 feet in thickness.

Thirdly. At the bottom of all, gravel with many rounded pebbles in it. This was everywhere removed so long as the tunnels which narrowed downwards were wide enough to be worked. It proved to be almost entirely barren of fossils.

The mammalia obtained from the bone-earth consisted of Elephas primigenius, or mammoth; Rhinoceros tichorhinus; Ursus spelaeus; Hyaena spelaea; Felis spelaea, or the cave-lion; Cervus tarandus, or the reindeer; a species of horse, ox, and several rodents, and others not yet determined.

No human bones were obtained anywhere during these excavations, but many flint knives, chiefly from the lowest part of the bone-earth; and one of the most perfect lay at the depth of 13 feet from the surface, and was covered with bone-earth of that thickness. Neglecting the less perfect specimens, some of which were met with even in the lowest gravel, about fifteen knives, recognised as artificially formed by the most experienced antiquaries, were taken from the bone-earth, and usually from near the bottom. Such knives, considered apart from the associated mammalia, afford in themselves no safe criterion of antiquity, as they might belong to any part of the age of stone, similar tools being sometimes met with in tumuli posterior in date to the era of the introduction of bronze. But the contemporaneity of those at Brixham with the extinct animals is demonstrated not only by the occurrence at one point in overlying stalagmite of the bone of a cave-bear, but also by the discovery at the same level in the bone-earth, and in close proximity to a very perfect flint tool, of the entire left hind-leg of a cave-bear. This specimen, which was shown me by Dr. Falconer and Mr. Pengelly, was exhumed from the earthy deposit in the reindeer gallery, near its junction with the flint-knife gallery, at the distance of about sixty-five feet from the main entrance. The mass of earth containing it was removed entire, and the matrix cleared away carefully by Dr. Falconer in the presence of Mr. Pengelly. Every bone was in its natural place, the femur, tibia, fibula, ankle-bone, or astragalus, all in juxtaposition. Even the patella or detached bone of the knee-pan was searched for, and not in vain. Here, therefore, we have evidence of an entire limb not having been washed in a fossil state out of an older alluvium, and then swept afterwards into a cave, so as to be mingled with flint implements, but having been introduced when clothed with its flesh, or at least when it had the separate bones bound together by their natural ligaments, and in that state buried in mud.

If they were not all of contemporary date, it is clear from this case, and from the humerus of the Ursus spelaeus, before cited, as found in a floor of stalagmite, that the bear lived after the flint tools were manufactured, or in other words, that Man in this district preceded the cave-bear.

A glance at the position of Windmill Hill, in which the caverns are situated, and a brief survey of the valleys which bound it on three sides, are enough to satisfy a geologist that the drainage and geographical features of this region have undergone great changes since the gravel and bone-earth were carried by streams into the subterranean cavities above described. Some worn pebbles of haematite, in particular, can only have come from their nearest parent rock, at a period when the valleys immediately adjoining the caves were much shallower than they now are. The reddish loam in which the bones are embedded is such as may be seen on the surface of limestone in the neighbourhood, but the currents which were formerly charged with such mud must have run at a level 78 feet above that of the stream now flowing in the same valley. It was remarked by Mr. Pengelly that the stones and bones in the loam had their longest axes parallel to the direction of the tunnels and fissures, showing that they were deposited by the action of a stream.* (* Pengelly, "Geologist" volume 4 1861 page 153.)

It appears that so long as the flowing water had force enough to propel stony fragments, no layer of fine mud could accumulate, and so long as there was a regular current capable of carrying in fine mud and bones, no superficial crust of stalagmite. In some passages, as before stated, stalagmite was wanting, while in one place seven or eight alternations of stalagmite and loam were observed, seeming to indicate a prevalence of more rainy seasons, succeeded by others, when the water was for a time too low to flood the area where the calcareous incrustation accumulated.

If the regular sequence of the three deposits of pebbles, mud, and stalagmite was the result of the causes above explained, the order of superposition would be constant, yet we could not be sure that the gravel in one passage might not sometimes be coeval with the bone-earth or stalagmite in another.

If therefore the flint knives had not been very widely dispersed, and if one of them had not been at the bottom of the bone-earth, close to the leg of the bear above described, their antiquity relatively to the extinct mammalia might have been questioned. No coprolites were found in the Brixham excavations, and very few gnawed bones. These few may have been brought from some distance before they reached their place of rest. Upon the whole, the same conclusion which Dr. Schmerling came to, respecting the filling up of the caverns near Liege, seems applicable to the caves of Brixham.

Dr. Falconer, after aiding in the investigations above alluded to near Torquay, stopped at Abbeville on his way to Sicily, in the autumn of 1858, and saw there the collection of M. Boucher de Perthes. Being at once satisfied that the flints called hatchets had really been fashioned by the hand of Man, he urged Mr. Prestwich, by letter, thoroughly to explore the geology of the valley of the Somme. This he accordingly accomplished, in company with Mr. John Evans [Note 13], of the Society of Antiquaries, and, before his return that same year, succeeded in dissipating all doubts from the minds of his geological friends by extracting, with his own hands, from a bed of undisturbed gravel, at St. Acheul, a well-shaped flint hatchet. This implement was buried in the gravel at a depth of 17 feet from the surface, and was lying on its flat side. There were no signs of vertical rents in the enveloping matrix, nor in the overlying beds of sand and loam, in which were many land and freshwater shells; so that it was impossible to imagine that the tool had gradually worked its way downwards, as some had suggested, through the incumbent soil, into an older formation.* (* Prestwich, "Proceedings of the Royal Society" 1859 and "Philosophical Transactions" 1860.)

There was no one in England whose authority deserved to have so much weight in overcoming incredulity in regard to the antiquity of the implements in question. For Mr. Prestwich, besides having published a series of important memoirs on the Tertiary formations of Europe, had devoted many years specially to the study of the drift and its organic remains. His report, therefore, to the Royal Society, accompanied by a photograph showing the position of the flint tool in situ before it was removed from its matrix, not only satisfied many inquirers, but induced others to visit Abbeville and Amiens; and one of these, Mr. Flower, who accompanied Mr. Prestwich on his second excursion to St. Acheul, in June 1859, succeeded, by digging into the bank of gravel, in disinterring, at the depth of 22 feet from the surface, a fine, symmetrically-shaped weapon of an oval form, lying in and beneath strata which were observed by many witnesses to be perfectly undisturbed.* (* "Quarterly Journal of the Geological Society" volume 16 1860 page 190.)

Shortly afterwards, in the year 1859, I visited the same pits, and obtained seventy flint tools, one of which was taken out while I was present, though I did not see it before it had fallen from the matrix. I expressed my opinion in favour of the antiquity of the flint tools to the meeting of the British Association at Aberdeen, in the same year.* (* See "Report of British Association" for 1859. ) On my way through Rouen, I stated my convictions on this subject to M. George Pouchet, who immediately betook himself to St. Acheul, commissioned by the municipality of Rouen, and did not quit the pits till he had seen one of the hatchets extracted from gravel in its natural position.* (* "Actes du Musee d'Histoire Naturelle de Rouen" 1860 page 33.)

M. Gaudry also gave the following account of his researches in the same year to the Royal Academy of Sciences at Paris. "The great point was not to leave the workmen for a single instant, and to satisfy oneself by actual inspection whether the hatchets were found in situ. I caused a deep excavation to be made, and found nine hatchets, most distinctly in situ in the diluvium, associated with teeth of Equus fossilis and a species of Bos, different from any now living, and similar to that of the diluvium and of caverns."* (* "Comptes rendus" September 26 and October 3, 1859.) In 1859, M. Hebert, an original observer of the highest authority, declared to the Geological Society of France that he had, in 1854, or four years before Mr. Prestwich's visit to St. Acheul, seen the sections at Abbeville and Amiens, and had come to the opinion that the hatchets were imbedded in the "lower diluvium," and that their origin was as ancient as that of the mammoth and the rhinoceros. M. Desnoyers also made excavations after M. Gaudry, at St. Acheul, in 1859, with the same results.* (* "Bulletin" volume 17 page 18.)

After a lively discussion on the subject in England and France, it was remembered, not only that there were numerous recorded cases leading to similar conclusions in regard to cavern deposits, but, also, that Mr. Frere had, so long ago as 1797, found flint weapons, of the same type as those of Amiens, in a freshwater formation in Suffolk, in conjunction with elephant remains; and nearly a hundred years earlier (1715), another tool of the same kind had been exhumed from the gravel of London, together with bones of an elephant; to all which examples I shall allude more fully in the sequel.

I may conclude this chapter by quoting a saying of Professor Agassiz, "that whenever a new and startling fact is brought to light in science, people first say, 'it is not true,' then that 'it is contrary to religion,' and lastly, 'that everybody knew it before.'"

If I were considering merely the cultivators of geology, I should say that the doctrine of the former co-existence of Man with many extinct mammalia had already gone through these three phases in the progress of every scientific truth towards acceptance. But the grounds of this belief have not yet been fully laid before the general public, so as to enable them fairly to weigh and appreciate the evidence. I shall therefore do my best in the next three chapters to accomplish this task.

CHAPTER 7.

PEAT AND PLEISTOCENE ALLUVIUM OF THE VALLEY OF THE SOMME.

Geological Structure of the Valley of the Somme and of the surrounding Country. Position of Alluvium of different Ages. Peat near Abbeville. Its animal and vegetable Contents. Works of Art in Peat. Probable Antiquity of the Peat, and Changes of Level since its Growth began. Flint Implements of antique Type in older Alluvium. Their various Forms and great Numbers.

GEOLOGICAL STRUCTURE OF THE SOMME VALLEY.

The valley of the Somme in Picardy, alluded to in the last chapter, is situated geologically in a region of white Chalk with flints, the strata of which are nearly horizontal. The Chalk hills which bound the valley are almost everywhere between 200 and 300 feet in height. On ascending to that elevation, we find ourselves on an extensive table-land, in which there are slight elevations and depressions. The white Chalk itself is scarcely ever exposed at the surface on this plateau, although seen on the slopes of the hills, as at b and c (Figure 7). The general surface of the upland region is covered continuously for miles in every direction by loam or brick-earth (Number 4), about 5 feet thick, devoid of fossils. To the wide extent of this loam the soil of Picardy chiefly owes its great fertility. Here and there we also observe, on the Chalk, outlying patches of Tertiary sand and clay (Number 5, Figure 7), with Eocene fossils, the remnants of a formation once more extensive, and which probably once spread in one continuous mass over the Chalk, before the present system of valleys had begun to be shaped out. It is necessary to allude to these relics of Tertiary strata, of which the larger part is missing, because their denudation has contributed largely to furnish the materials of gravels in which the flint implements and bones of extinct mammalia are entombed. From this source have been derived not only the regular-formed egg-shaped pebbles, so common in the old fluviatile alluvium at all levels, but those huge masses of hard sandstone, several feet in diameter, to which I shall allude in the sequel. The upland loam also (Number 4) has often, in no slight degree, been formed at the expense of the same Tertiary sands and clays, as is attested by its becoming more or less sandy or argillaceous, according to the nature of the nearest Eocene outlier in the neighbourhood.

The average width of the valley of the Somme between Amiens and Abbeville is one mile. The height, therefore, of the hills, in relation to the river-plain, could not be correctly represented in the annexed diagram (Figure 7), as they would have to be reduced in altitude; or if not, it would be necessary to make the space between c and b four times as great. The dimensions also of the masses, of drift or alluvium, 2 and 3, have been exaggerated, in order to render them sufficiently conspicuous; for, all important as we shall find them to be as geological monuments of the Pleistocene period, they form a truly insignificant feature in the general structure of the country, so much so, that they might easily be overlooked in a cursory survey of the district, and are usually unnoticed in geological maps not specially devoted to the superficial formations.

(FIGURE 7. SECTION ACROSS THE VALLEY OF THE SOMME IN PICARDY.

1. Peat, 20 to 30 feet thick, resting on gravel, a. 2. Lower level gravel with elephants' bones and flint tools, covered with fluviatile loam, 20 to 40 feet thick. 3. Higher level gravel with similar fossils, and with overlying loam, in all 30 feet thick. 4. Upland loam without shells (Limon des plateaux), 5 or 6 feet thick. 5. Eocene strata, resting on the Chalk in patches.)

It will be seen by the description given of the section (Figure 7) that Number 2 indicates the lower level gravels, and Number 3 the higher ones, or those rising to elevations of 80 or 100 feet above the river. Newer than these is the peat Number 1, which is from 10 to 30 feet in thickness, and which is not only of later date than the alluvium, 2 and 3, but is also posterior to the denudation of those gravels, or to the time when the valley was excavated through them. Underneath the peat is a bed of gravel, a, from 3 to 14 feet thick, which rests on undisturbed Chalk. This gravel was probably formed, in part at least, when the valley was scooped out to its present depth, since which time no geological change has taken place, except the growth of the peat, and certain oscillations in the general level of the country, to which we shall allude by and by. A thin layer of impervious clay separates the gravel a from the peat Number 1, and seems to have been a necessary preliminary to the growth of the peat.

PEAT OF THE VALLEY OF THE SOMME.

As hitherto, in our retrospective survey, we have been obliged, for the sake of proceeding from the known to the less known, to reverse the natural order of history, and to treat of the newer before the older formations, I shall begin my account of the geological monuments of the valley of the Somme by saying something of the most modern of all of them, the peat. This substance occupies the lower parts of the valley far above Amiens, and below Abbeville as far as the sea. It has already been stated to be in some places 30 feet thick, and is even occasionally more than 30 feet, corresponding in that respect to the Danish mosses before described (Chapter 2). Like them, it belongs to the Recent period; all the embedded mammalia, as well as the shells, being of the same species as those now inhabiting Europe. The bones of quadrupeds are very numerous, as I can bear witness, having seen them brought up from a considerable depth near Abbeville, almost as often as the dredging instrument was used. Besides remains of the beaver, I was shown, in the collection of M. Boucher de Perthes, two perfect lower jaws with teeth of the bear, Ursus arctos; and in the Paris Museum there is another specimen, also from the Abbeville peat.

The list of mammalia already comprises a large proportion of those proper to the Swiss lake-dwellings, and to the shell-mounds and peat of Denmark; but unfortunately as yet no special study has been made of the French fauna, like that by which the Danish and Swiss zoologists and botanists have enabled us to compare the wild and tame animals and the vegetation of the age of stone with that of the age of iron.

Notwithstanding the abundance of mammalian bones in the peat, and the frequency of stone implements of the Celtic and Gallo-Roman periods, M. Boucher de Perthes has only met with three or four fragments of human skeletons.

At some depth in certain places in the valley near Abbeville, the trunks of alders have been found standing erect as they grew, with their roots fixed in an ancient soil, afterwards covered with peat. Stems of the hazel, and nuts of the same, abound; trunks, also, of the oak and walnut. The peat extends to the coast, and is there seen passing under the sand-dunes and below the sea-level. At the mouth of the river Canche, which joins the sea near the embouchure of the Somme, yew trees, firs, oaks, and hazels have been dug out of peat, which is there worked for fuel, and is about three feet thick.* (* D'Archiac, "Histoire des Progres" volume 2 page 154.) During great storms, large masses of compact peat, enclosing trunks of flattened trees, have been thrown up on the coast at the mouth of the Somme; seeming to indicate that there has been a subsidence of the land and a consequent submergence of what was once a westward continuation of the valley of the Somme into what is now a part of the English Channel.

Whether the vegetation of the lowest layers of peat differed as to the geographical distribution of some of the trees from the middle, and this from the uppermost peat, as in Denmark, has not yet been ascertained; nor have careful observations been made with a view of calculating the minimum of time which the accumulation of so dense a mass of vegetable matter must have taken. A foot in thickness of highly compressed peat, such as is sometimes reached in the bottom of the bogs, is obviously the equivalent in time of a much greater thickness of peat of spongy and loose texture, found near the surface. The workmen who cut peat, or dredge it up from the bottom of swamps and ponds, declare that in the course of their lives none of the hollows which they have found, or caused by extracting peat, have ever been refilled, even to a small extent. They deny, therefore, that the peat grows. This, as M. Boucher de Perthes observes, is a mistake; but it implies that the increase in one generation is not very appreciable by the unscientific.

The antiquary finds near the surface Gallo-Roman remains, and still deeper Celtic weapons of the stone period. [Note 14.] But the depth at which Roman works of art occur varies in different places, and is no sure test of age; because in some parts of the swamps, especially near the river, the peat is often so fluid that heavy substances may sink through it, carried down by their own gravity. In one case, however, M. Boucher de Perthes observed several large flat dishes of Roman pottery, lying in a horizontal position in the peat, the shape of which must have prevented them from sinking or penetrating through the underlying peat. Allowing about fourteen centuries for the growth of the superincumbent vegetable matter, he calculated that the thickness gained in a hundred years would be no more than three centimetres.* (* "Antiquites Celtiques" volume 2 page 134.) This rate of increase would demand so many thousands of years for the formation of the entire thickness of 30 feet that we must hesitate before adopting it as a chronometric scale. Yet, by multiplying observations of this kind, and bringing one to bear upon and check another, we may eventually succeed in obtaining data for estimating the age of the peaty deposit. [Note 15.]

The rate of increase in Denmark may not be applicable to France; because differences in the humidity of the climate, or in the intensity and duration of summer's heat and winter's cold, as well as diversity in the species of plants which most abound, would cause the peat to grow more or less rapidly, not only when we compare two distinct countries in Europe, but the same country at two successive periods.

I have already alluded to some facts which favour the idea that there has been a change of level on the coast since the peat began to grow. This conclusion seems confirmed by the mere thickness of peat at Abbeville, and the occurrence of alder and hazel-wood near the bottom of it. If 30 feet of peat were now removed, the sea would flow up and fill the valley for miles above Abbeville. Yet this vegetable matter is all of supra-marine origin, for where shells occur in it they are all of terrestrial or fluviatile kinds, so that it must have grown above the sea-level when the land was more elevated than now. We have already seen what changes in the relative level of sea and land have occurred in Scotland subsequently to the time of the Romans, and are therefore prepared to meet with proofs of similar movements in Picardy. In that country they have probably not been confined simply to subsidence, but have comprised oscillations in the level of the land, by which marine shells of the Pleistocene period have been raised some 10 feet or more above the level of the sea.

Small as is the progress hitherto made in interpreting the pages of the peaty record, their importance in the valley of the Somme is enhanced by the reflection that, whatever be the number of centuries to which they relate, they belong to times posterior to the ancient implement-bearing beds, which we are next to consider, and are even separated from them, as we shall see, by an interval far greater than that which divides the earliest strata of the peat from the latest.

FLINT IMPLEMENTS OF THE PLEISTOCENE PERIOD IN THE VALLEY OF THE SOMME.

The alluvium of the valley of the Somme exhibits nothing extraordinary or exceptional in its position or external appearance, nor in the arrangement or composition of its materials, nor in its organic remains; in all these characters it might be matched by the drift of a hundred other valleys in France or England. Its claim to our peculiar attention is derived from the wonderful number of flint tools, of a very antique type, which, as stated in the last chapter, occur in undisturbed strata, associated with the bones of extinct quadrupeds.

As much doubt has been cast on the question, whether the so-called flint hatchets have really been shaped by the hands of Man, it will be desirable to begin by satisfying the reader's mind on that point, before inviting him to study the details of sections of successive beds of mud, sand, and gravel, which vary considerably even in contiguous localities.

Since the spring of 1859, I have paid three visits to the Valley of the Somme, and examined all the principal localities of these flint tools. In my excursions around Abbeville, I was accompanied by M. Boucher de Perthes, and during one of my explorations in the Amiens district, by Mr. Prestwitch. The first time I entered the pits at St. Acheul, I obtained seventy flint instruments, all of them collected from the drift in the course of the preceding five or six weeks. The two prevailing forms of these tools are represented in the annexed Figures 8 and 9, each of which are half the size of the originals; the first being the spear-headed form, varying in length from six to eight inches; the second, the oval form, which is not unlike some stone implements, used to this day as hatchets and tomahawks by natives of Australia, but with this difference, that the edge in the Australian weapons (as in the case of those called celts in Europe) has been produced by friction, whereas the cutting edge in the old tools of the valley of the Somme was always gained by the simple fracture of the flint, and by the repetition of many dexterous blows.

The oval-shaped Australian weapons, however, differ in being sharpened at one end only. The other, though reduced by fracture to the same general form, is left rough, in which state it is fixed into a cleft stick, which serves as a handle. To this it is firmly bound by thin straps of opossum's hide. One of these tools, now in my possession, was given me by Mr. Farquharson of Haughton, who saw a native using it in 1854 on the Auburn river, in Burnet district, North Australia.

Out of more than a hundred flint implements which I obtained at St. Acheul, not a few had their edges more or less fractured or worn, either by use as instruments before they were buried in gravel, or by being rolled in the river's bed.

Some of these tools were probably used as weapons, both of war and of the chase, others to grub up roots, cut down trees, and scoop out canoes. Some of them may have served, as Mr. Prestwich has suggested, for cutting holes in the ice both for fishing and for obtaining water, as will be explained in the eighth chapter when we consider the arguments in favour of the higher level drift having belonged to a period when the rivers were frozen over for several months every winter.

(FIGURE 8. FLINT IMPLEMENT FROM ST. ACHEUL, NEAR AMIENS, OF THE SPEAR-HEAD SHAPE (half the size of the original, which is 7 1/2 inches long).

a. Side view. b. Same seen edgewise.

These spear-headed implements have been found in greater number, proportionally to the oval ones, in the upper level gravel at St. Acheul, than in any of the lower gravels in the valley of the Somme. In these last the oval form predominates, especially at Abbeville.)

When the natural form of a Chalk-flint presented a suitable handle at one end, as in the specimen, Figure 10, that part was left as found. The portion, for example, between b and c has probably not been altered; the protuberances which are fractured having been broken off by river action before the flint was chipped artificially. The other extremity, a, has been worked till it acquired a proper shape and cutting edge.

(FIGURES 9 AND 10. FLINT IMPLEMENTS FROM THE PLEISTOCENE DRIFT OF ABBEVILLE AND AMIENS.

FIGURE 9. a. OVAL-SHAPED FLINT HATCHET FROM MAUTORT, NEAR ABBEVILLE,

half size of original, which is 5 1/2 inches long, from a bed of gravel underlying the fluvio-marine stratum. b. Same seen edgewise. c. Shows a recent fracture of the edge of the same at the point a, or near the top. This portion of the tool, c, is drawn of the natural size, the black central part being the unaltered flint, the white outer coating, the layer which has been formed by discoloration or bleaching since the tool was first made. The entire surface of Number 9 must have been black when first shaped, and the bleaching to such a depth must have been the work of time, whether produced by exposure to the sun and air before it was embedded, or afterwards when it lay deep in the soil.

FIGURE 10. FLINT TOOL FROM ST. ACHEUL, seen edgewise; original 6 1/2 inches long, and 3 inches wide.

b, c. Portion not artificially shaped. a, b. Part chipped into shape, and having a cutting edge at a.)

Many of the hatchets are stained of an ochreous-yellow colour, when they have been buried in yellow gravel, others have acquired white or brown tints, according to the matrix in which they have been enclosed.

This accordance in the colouring of the flint tools with the character of the bed from which they have come, indicates, says Mr. Prestwich, not only a real derivation from such strata, but also a sojourn therein of equal duration to that of the naturally broken flints forming part of the same beds.* (* "Philosophical Transactions" 1861 page 297.)

(FIGURES 11, 12 AND 13. DENDRITES ON SURFACES OF FLINT HATCHETS IN THE DRIFT OF ST. ACHEUL, NEAR AMIENS.

FIGURE 11. a. Natural size.

FIGURE 12. b. Natural size. c. Magnified.

FIGURE 13. d. Natural size. e. Magnified.)

The surface of many of the tools is encrusted with a film of carbonate of lime, while others are adorned by those ramifying crystallisations called dendrites (see Figures 11, 12 and 13), usually consisting of the mixed oxides of iron and manganese, forming extremely delicate blackish brown sprigs, resembling the smaller kinds of sea weed. They are a useful test of antiquity when suspicions are entertained of the workmen having forged the hatchets which they offer for sale. The most general test, however, of the genuineness of the implements obtained by purchase is their superficial varnish-like or vitreous gloss, as contrasted with the dull aspect of freshly fractured flints. I also remarked, during each of my three visits to Amiens, that there were some extensive gravel-pits, such as those of Montiers and St. Roch, agreeing in their geological character with those of St. Acheul, and only a mile or two distant, where the workmen, although familiar with the forms, and knowing the marketable value of the articles above described, assured me that they had never been able to find a single implement.

Respecting the authenticity of the tools as works of art, Professor Ramsay, than whom no one could be a more competent judge, observes: "For more than twenty years, like others of my craft, I have daily handled stones, whether fashioned by nature or art; and the flint hatchets of Amiens and Abbeville seem to me as clearly works of art as any Sheffield whittle."* (* "Athenaeum" July 16, 1859.)

Mr. Evans classifies the implements under three heads, two of which, the spear heads and the oval or almond-shaped kinds, have already been described. The third form (Figure 14) consists of flakes, apparently intended for knives or some of the smaller ones for arrow heads.

(FIGURE 14. FLINT KNIFE OR FLAKE FROM BELOW THE SAND CONTAINING CYRENA FLUMINALIS. MENCHECOURT, ABBEVILLE. d. Transverse section along the line of fracture, b, c. Size, two-thirds of the original. )

In regard to their origin, Mr. Evans observes that there is a uniformity of shape, a correctness of outline, and a sharpness about the cutting edges and points, which cannot be due to anything but design.* (* "Archaeologia" volume 38.)

Of these knives and flakes, I obtained several specimens from a pit which I caused to be dug at Abbeville, in sand in contact with the Chalk, and below certain fluvio-marine beds, which will be alluded to in the next chapter.

Between the spear-head and oval shapes, there are various intermediate gradations, and there are also a vast variety of very rude implements, many of which may have been rejected as failures, and others struck off as chips in the course of manufacturing the more perfect ones. Some of these chips can only be recognised by an experienced eye as bearing marks of human workmanship.

It has often been asked, how, without the use of metallic hammers, so many of these oval and spear-headed tools could have been wrought into so uniform a shape. Mr. Evans, in order experimentally to illustrate the process, constructed a stone hammer, by mounting a pebble in a wooden handle, and with this tool struck off flakes from the edge on both sides of a Chalk flint, till it acquired precisely the same shape as the oval tool, Figure 9.

If I were invited to estimate the probable number of the more perfect tools found in the valley of the Somme since 1842, rejecting all the knives, and all that might be suspected of being spurious or forged, I should conjecture that they far exceeded a thousand. Yet it would be a great mistake to imagine that an antiquary or geologist, who should devote a few weeks to the exploration of such a valley as that of the Somme, would himself be able to detect a single specimen. But few tools were lying on the surface. The rest have been exposed to view by the removal of such a volume of sand, clay, and gravel, that the price of the discovery of one of them could only be estimated by knowing how many hundred labourers have toiled at the fortifications of Abbeville, or in the sand and gravel pits near that city, and around Amiens, for road materials and other economic purposes, during the last twenty years.

(FIGURE 15. FOSSILS OF THE WHITE CHALK.

a, b. Coscinopora globularis, D'Orbigny. Orbitolina concava, Parker and Jones. c. Part of same magnified.)

In the gravel pits of St. Acheul, and in some others near Amiens, small round bodies, having a tubular cavity in the centre, occur. They are well known as fossils of the White Chalk. Dr. Rigollot suggested that they might have been strung together as beads, and he supposed the hole in the middle to have been artificial. Some of these round bodies are found entire in the Chalk and in the gravel, others have naturally a hole passing through them, and sometimes one or two holes penetrating some way in from the surface, but not extending to the other side. Others, like b, Figure 15, have a large cavity, which has a very artificial aspect. It is impossible to decide whether they have or have not served as personal ornaments, recommended by their globular form, lightness, and by being less destructible than ordinary Chalk. Granting that there were natural cavities in the axis of some of them, it does not follow that these may not have been taken advantage of for stringing them as beads, while others may have been artificially bored through. Dr. Rigollot's argument in favour of their having been used as necklaces or bracelets, appears to me a sound one. He says he often found small heaps or groups of them in one place, all perforated, just as if, when swept into the river's bed by a flood, the bond which had united them together remained unbroken.* (* Rigollot, "Memoire sur des Instruments en Silex" etc., Amiens 1854 page 16.)

CHAPTER 8.

PLEISTOCENE ALLUVIUM WITH FLINT IMPLEMENTS OF THE VALLEY OF THE SOMME—CONCLUDED.

Fluvio-marine Strata, with Flint Implements, near Abbeville. Marine Shells in same. Cyrena fluminalis. Mammalia. Entire Skeleton of Rhinoceros. Flint Implements, why found low down in Fluviatile Deposits. Rivers shifting their Channels. Relative Ages of higher and lower-level Gravels. Section of Alluvium of St. Acheul. Two Species of Elephant and Hippopotamus coexisting with Man in France. Volume of Drift, proving Antiquity of Flint Implements. Absence of Human Bones in tool-bearing Alluvium, how explained. Value of certain Kinds of negative Evidence tested thereby. Human Bones not found in drained Lake of Haarlem.

In the section of the valley of the Somme given in Figure 7, the successive formations newer than the Chalk are numbered in chronological order, beginning with the most modern, or the peat, which is marked Number 1, and which has been treated of in the last chapter. Next in the order of antiquity are the lower-level gravels, Number 2, which we have now to describe; after which the alluvium, Number 3, found at higher levels, or about 80 and 100 feet above the river-plain, will remain to be considered.

I have selected, as illustrating the old alluvium of the Somme occurring at levels slightly elevated above the present river, the sand and gravel-pits of Menchecourt, in the northwest suburbs of Abbeville, to which, as before stated, attention was first drawn by M. Boucher de Perthes, in his work on Celtic antiquities. Here, although in every adjoining pit some minor variations in the nature and thickness of the superimposed deposits may be seen, there is yet a general approach to uniformity in the series. The only stratum of which the relative age is somewhat doubtful, is the gravel marked a, underlying the peat, and resting on the Chalk. It is only known by borings, and some of it may be of the same age as Number 3; but I believe it to be for the most part of more modern origin, consisting of the wreck of all the older gravel, including Number 3, and formed during the last hollowing out and deepening of the valley immediately before the commencement of the growth of peat.

The greater number of flint implements have been dug out of Number 3, often near the bottom, and twenty-five, thirty, or even more than thirty feet below the surface of Number 1.

A geologist will perceive by a glance at the section that the valley of the Somme must have been excavated nearly to its present depth and width when the strata of Number 3 were thrown down, and that after the deposits Numbers 3, 2, and 1 had been formed in succession, the present valley was scooped out, patches only of Numbers 3 and 2 being left. For these deposits cannot originally have ended abruptly as they now do, but must have once been continuous farther towards the centre of the valley.

(FIGURE 16. SECTION OF FLUVIO-MARINE STRATA, CONTAINING FLINT IMPLEMENTS AND BONES OF EXTINCT MAMMALIA, AT MENCHECOURT, ABBEVILLE.* (* For detailed sections and maps of this district, see Prestwich, "Philosophical Transactions" 1860 page 277.)

1. Brown clay with angular flints, and occasionally Chalk rubble, unstratified, following the slope of the hill, probably of subaerial origin, of very varying thickness, from 2 to 5 feet and upwards. 2. Calcareous loam, buff-coloured, resembling loess, for the most part unstratified, in some places with slight traces of stratification, containing freshwater and land shells, with bones of elephants, etc.; thickness about 15 feet. 3. Alternations of beds of gravel, marl, and sand, with freshwater and land shells, and, in some of the lower sands, a mixture of marine shells; also bones of elephant, rhinoceros, etc., and flint implements; thickness about 12 feet. a. Gravel underlying peat, age undetermined. b. Layer of impervious clay, separating the gravel from the peat.)

To begin with the oldest, Number 3, it is made up of a succession of beds, chiefly of freshwater origin, but occasionally a mixture of marine and fluviatile shells is observed in it, proving that the sea sometimes gained upon the river, whether at high tides or when the fresh water was less in quantity during the dry season, and sometimes perhaps when the land was slightly depressed in level. All these accidents might occur again and again at the mouth of any river, and give rise to alternations of fluviatile and marine strata, such as are seen at Menchecourt.

In the lowest beds of gravel and sand in contact with the Chalk, flint hatchets, some perfect, others much rolled, have been found; and in a sandy bed in this position some workmen, whom I employed to sink a pit, found four flint knives. Above this sand and gravel occur beds of white and siliceous sand, containing shells of the genera Planorbis, Limnea, Paludina, Valvata, Cyclas, Cyrena, Helix, and others, all now natives of the same part of France, except Cyrena fluminalis (Figure 17), which no longer lives in Europe, but inhabits the Nile, and many parts of Asia, including Cashmere, where it abounds. No species of Cyrena is now met with in a living state in Europe. Mr. Prestwich first observed it fossil at Menchecourt, and it has since been found in two or three contiguous sand-pits, always in the fluvio-marine bed. [Note 16.]

(FIGURE 17. Cyrena fluminalis, O.F. Muller, sp.* (* For synonyms, see S. Woodward "Tibet Shells" "Proceedings of the Zoological Society" July 8, 1856.)

a. Interior of left valve, from Gray's Thurrock, Essex. b. Hinge of the same magnified. c. Interior of right valve of a small specimen, from Shacklewell, London. d. Outer surface of right valve, from Erith, Kent.)

TABLE 8/1. DATES OF SPECIFIC NAMES.

COLUMN 1 : SPECIES.

COLUMN 2 : DATE.

LIVING:

Tellina fluminalis, O.F. Muller : 1774. Venus fluminalis Euphratis, Chemnitz : 1782. Cyclas Euphratica, Lam. : 1806. Cyrena cor, Lam. (Nile): 1818. Cyrena consobrina, Caillaud (Nile) : 1823. Cyrena Cashmiriensis, Desh. : Corbicuia fluminalis, Muhlfeldt. : 1811.

FOSSIL:

Cyrena trigonula, S. Woodward : 1834. Cyrena Gemmellarii, Philippi : 1836. Cyrena Duchastelii, Nyst : 1838.

The following marine shells occur mixed with the freshwater species above enumerated:—Buccinum undatum, Littorina littorea, Nassa reticulata, Purpura lapillus, Tellina solidula, Cardium edule, and fragments of some others. Several of these I have myself collected entire, though in a state of great decomposition, lying in the white sand called "sable aigre" by the workmen. They are all littoral species now proper to the contiguous coast of France. Their occurrence in a fossil state associated with freshwater shells at Menchecourt had been noticed as long ago as 1836 by MM. Ravin and Baillon, before M. Boucher de Perthes commenced the researches which have since made the locality so celebrated.* (* D'Archiac, "Histoire des Progres" etc. volume 2 page 154.) The numbers since collected preclude all idea of their having been brought inland as eatable shells by the fabricators of the flint hatchets found at the bottom of the fluvio-marine sands. From the same beds, and in marls alternating with the sands, remains of the elephant, rhinoceros, and other mammalia have been exhumed.

Above the fluvio-marine strata are those designated Number 2 in the section (Figure 16), which are almost devoid of stratification, and probably formed of mud or sediment thrown down by the waters of the river when they overflowed the ancient alluvial plain of that day. Some land shells, a few river shells, and bones of mammalia, some of them extinct, occur in Number 2. Its upper surface has been deeply furrowed and cut into by the action of water, at the time when the earthy matter of Number 1 was superimposed. The materials of this uppermost deposit are arranged as if they had been the result of land floods, taking place after the formations 2 and 3 had been raised, or had become exposed to denudation.

The fluvio-marine strata and overlying loam of Menchecourt recur on the opposite or left bank of the alluvial plain of the Somme, at a distance of 2 or 3 miles. They are found at Mautort, among other places, and I obtained there the flint hatchet shown in Figure 9, of an oval form. It was extracted from gravel, above which were strata containing a mixture of marine and freshwater shells, precisely like those of Menchecourt. In the alluvium of all parts of the valley, both at high and low levels, rolled bones are sometimes met with in the gravel. Some of the flint tools in the gravel of Abbeville have their angles very perfect, others have been much triturated, as if in the bed of the main river or some of its tributaries.

The mammalia most frequently cited as having been found in the deposits Numbers 2 and 3 at Menchecourt, are the following:—

Elephas primigenius. Rhinoceros tichorhinus. Equus fossilis, Owen. Bos primigenius. Cervus somonensis, Cuvier. C. tarandus priscus, Cuvier. Felis spelaea. Hyaena spelaea.

The Ursus spelaeus has also been mentioned by some writers; but M. Lartet says he has sought in vain for it among the osteological treasures sent from Abbeville to Cuvier at Paris, and in other collections. The same palaeontologist, after a close scrutiny of the bones sent formerly to the Paris Museum from the valley of the Somme, observed that some of them bore the evident marks of an instrument, agreeing well with incisions such as a rude flint-saw would produce. Among other bones mentioned as having been thus artificially cut, are those of a Rhinoceros tichorhinus, and the antlers of Cervus somonensis.* (* "Quarterly Journal of the Geological Society" volume 16 1860 page 471.)

The evidence obtained by naturalists that some of the extinct mammalia of Menchecourt really lived and died in this part of France, at the time of the embedding of the flint tools in fluviatile strata, is most satisfactory; and not the less so for having been put on record long before any suspicion was entertained that works of art would ever be detected in the same beds. Thus M. Baillon, writing in 1834 to M. Ravin, says: "They begin to meet with fossil bones at the depth of 10 or 12 feet in the Menchecourt sand-pits, but they find a much greater quantity at the depth of 18 and 20 feet. Some of them were evidently broken before they were embedded, others are rounded, having, without doubt, been rolled by running water. It is at the bottom of the sand-pits that the most entire bones occur. Here they lie without having undergone fracture or friction, and seem to have been articulated together at the time when they were covered up. I found in one place a whole hind limb of a rhinoceros, the bones of which were still in their true relative position. They must have been joined together by ligaments, and even surrounded by muscles at the time of their interment. The entire skeleton of the same species was lying at a short distance from the spot."* (* "Societe Roy. d'Emulation d'Abbeville" 1834 page 197.)

If we suppose that the greater number of the flint implements occurring in the neighbourhood of Abbeville and Amiens were brought by river action into their present position, we can at once explain why so large a proportion of them are found at considerable depths from the surface, for they would naturally be buried in gravel and not in fine sediment, or what may be termed "inundation mud," such as Number 2 (Figure 16), a deposit from tranquil water, or where the stream had not sufficient force or velocity to sweep along Chalk flints, whether wrought or unwrought. Hence we have almost always to pass down through a mass of incumbent loam with land shells, or through fine sand with freshwater molluscs, before we get into the beds of gravel containing hatchets. Occasionally a weapon used as a projectile may have fallen into quiet water, or may have dropped from a canoe to the bottom of the river, or may have been floated by ice, as are some stones occasionally by the Thames in severe winters, and carried over the meadows bordering its banks; but such cases are exceptional, though helping to explain how isolated flint tools or pebbles and angular stones are now and then to be seen in the midst of the finest loams.

The endless variety in the sections of the alluvium of the valley of the Somme, may be ascribed to the frequent silting up of the main stream and its tributaries during different stages of the excavation of the valley, probably also during changes in the level of the land. As a rule, when a river attacks and undermines one bank, it throws down gravel and sand on the opposite side of its channel, which is growing somewhere shallower, and is soon destined to be raised so high as to form an addition to the alluvial plain, and to be only occasionally inundated. In this way, after much encroachment on cliff or meadow at certain points, we find at the end of centuries that the width of the channel has not been enlarged, for the new made ground is raised after a time to the average height of the older alluvial tract. Sometimes an island is formed in midstream, the current flowing for a while on both sides of it, and at length scooping out a deeper channel on one side so as to leave the other to be gradually filled up during freshets and afterwards elevated by inundation mud, or "brick-earth." During the levelling up of these old channels, a flood sometimes cuts into and partially removes portions of the previously stratified matter, causing those repeated signs of furrowing and filling up of cavities, those memorials of doing and undoing, of which the tool-bearing sands and gravels of Abbeville and Amiens afford such reiterated illustrations, and of which a parallel is furnished by the ancient alluvium of the Thames valley, where similar bones of extinct mammalia and shells, including Cyrena fluminalis, are found.

Professor Noeggerath, of Bonn, informs me that, about the year 1845, when the bed of the Rhine was deepened artificially by the blasting and removal of rock in the narrows at Bingerloch, not far from Bingen, several flint hatchets and an extraordinary number of iron weapons of the Roman period were brought up by the dredge from the bed of the great river. The decomposition of the iron had caused much of the gravel to be cemented together into a conglomerate. In such a case we have only to suppose the Rhine to deviate slightly from its course, changing its position, as it has often done in various parts of its plain in historical times, and then tools of the stone and iron periods would be found in gravel at the bottom with a great thickness of sand and overlying loam deposited above them.

Changes in a river plain, such as those above alluded to, give rise frequently to ponds, swamps, and marshes, marking the course of old beds or branches of the river not yet filled up, and in these depressions shells proper both to running and stagnant water may be preserved, and quadrupeds may be mired. The latest and uppermost deposit of the series will be loam or brick-earth, with land and amphibious shells (Helix and Succinea), while below will follow strata containing freshwater shells, implying continuous submergence; and lowest of all in most sections will be the coarse gravel accumulated by a current of considerable strength and velocity.

When the St. Katharine docks were excavated at London, and similar works executed on the banks of the Mersey, old ships were dug out, as I have elsewhere noticed,* (* "Principles of Geology" 10th edition volume 2 page 547.) showing how the Thames and Mersey have in modern times been shifting their channels. Recently, an old silted-up bed of the Thames has been discovered by boring at Shoeburyness at the mouth of the river opposite Sheerness, as I learn from Mr. Mylne. The old deserted branch is separated from the new or present channel of the Thames, by a mass of London Clay which has escaped denudation. The depth of the old branch, or the thickness of fluviatile strata with which it has been filled up, is 75 feet. The actual channel in the neighbourhood is now 60 feet deep, but there is probably 10 or 15 feet of stratified sand and gravel at the bottom; so that, should the river deviate again from its course, its present bed might be the receptacle of a fluvio-marine formation 75 feet thick, equal to the former one of Shoeburyness, and more considerable than that of Abbeville. It would consist both of freshwater and marine strata, as the salt water is carried by the tide far up above Sheerness; but in order that such deposits should resemble, in geological position, the Menchecourt beds, they must be raised 10 or 15 feet above their present level, and be partially eroded. Such erosion they would not fail to suffer during the process of upheaval, because the Thames would scour out its bed, and not alter its position relatively to the sea, while the land was gradually rising.

Before the canal was made at Abbeville, the tide was perceptible in the Somme for some distance above that city. It would only require, therefore, a slight subsidence to allow the salt water to reach Menchecourt, as it did in the Pleistocene period. As a stratum containing exclusively land and freshwater shells usually underlies the fluvio-marine sands at Menchecourt, it seems that the river first prevailed there, after which the land subsided; and then there was an upheaval which raised the country to a greater height than that at which it now stands, after which there was a second sinking, indicated by the position of the peat, as already explained. All these changes happened since Man first inhabited this region.

At several places in the environs of Abbeville there are fluviatile deposits at a higher level by 50 feet than the uppermost beds at Menchecourt, resting in like manner on the Chalk. One of these occurs in the suburbs of the city at Moulin Quignon, 100 feet above the Somme and on the same side of the valley as Menchecourt, and containing flint implements of the same antique type and the bones of elephants; but no marine shells have been found there, nor in any gravel or sand at higher elevations than the Menchecourt marine shells.

It has been a matter of discussion among geologists whether the higher or the lower sands and gravels of the Somme valley are the more ancient. As a general rule, when there are alluvial formations of different ages in the same valley, those which occupy a more elevated position above the river plain are the oldest. In Auvergne and Velay, in Central France, where the bones of fossil quadrupeds occur at all heights above the present rivers from 10 to 1000 feet, we observe the terrestrial fauna to depart in character from that now living in proportion as we ascend to higher terraces and platforms. We pass from the lower alluvium, containing the mammoth, tichorhine rhinoceros, and reindeer, to various older groups of fossils, till, on a tableland 1000 feet high (near Le Puy, for example), the abrupt termination of which overlooks the present valley, we discover an old extinct river-bed covered by a current of ancient lava, showing where the lowest level was once situated. In that elevated alluvium the remains of a Tertiary mastodon and other quadrupeds of like antiquity are embedded.

If the Menchecourt beds had been first formed, and the valley, after being nearly as deep and wide as it is now, had subsided, the sea must have advanced inland, causing small delta-like accumulations at successive heights, wherever the main river and its tributaries met the sea. Such a movement, especially if it were intermittent, and interrupted occasionally by long pauses, would very well account for the accumulation of stratified debris which we encounter at certain points in the valley, especially around Abbeville and Amiens. But we are precluded from adopting this theory by the entire absence of marine shells, and the presence of freshwater and land species, and mammalian bones, in considerable abundance, in the drift both of higher and lower levels above Abbeville. Had there been a total absence of all organic remains, we might have imagined the former presence of the sea, and the destruction of such remains might have been ascribed to carbonic acid or other decomposing causes; but the Pleistocene and implement-bearing strata can be shown by their fossils to be of fluviatile origin.

FLINT IMPLEMENTS IN GRAVEL NEAR AMIENS. GRAVEL OF ST. ACHEUL.

When we ascend the valley of the Somme, from Abbeville to Amiens, a distance of about 25 miles, we observe a repetition of all the same alluvial phenomena which we have seen exhibited at Menchecourt and its neighbourhood, with the single exception of the absence of marine shells and of Cyrena fluminalis. We find lower-level gravel, such as Number 2, Figure 7, and higher-level alluvium, such as Number 3, the latter rising to 100 feet above the plain, which at Amiens is about 50 feet above the level of the river at Abbeville. In both the upper and lower gravels, as Dr. Rigollot stated in 1854, flint tools and the bones of extinct animals, together with river shells and land shells of living species, abound.

(FIGURE 18.* Elephas primigenius. Penultimate molar, lower jaw, right side, one-third of natural size, Pleistocene. Co-existed with Man.)

(FIGURE 19.* Elephas antiquus, Falconer. Penultimate molar, lower jaw, right side, one-third of natural size, Pleistocene and Newer Pliocene. Co-existed with Man.)

(FIGURE 20.* Elephas meridionalis, Nesti. Penultimate molar, lower jaw, right side, one-third of natural size, Newer Pliocene, Saint Prest, near Chartres, and Norwich Crag. Not yet proved to have coexisted with Man.)

(* For Figure 20 I am indebted to M. Lartet, and Figure 18 will be found in his paper in "Bulletin de la Societe Geologique de France" March 1859. Figure 19 is from Falconer and Cautley "Fauna Sivalensis.")

Immediately below Amiens, a great mass of stratified gravel, slightly elevated above the alluvial plain of the Somme, is seen at St. Roch, and half a mile farther down the valley at Montiers. Between these two places a small tributary stream, called the Celle, joins the Somme. In the gravel at Montiers, Mr. Prestwich and I found some flint knives, one of them flat on one side, but the other carefully worked, and exhibiting many fractures, clearly produced by blows skilfully applied. Some of these knives were taken from so low a level as to satisfy us that this great bed of gravel at Montiers, as well as that of the contiguous quarries of St. Roch, which seems to be a continuation of the same deposit, may be referred to the human period. Dr. Rigollot had already mentioned flint hatchets as obtained by him from St. Roch, but as none have been found there of late years, his statement was thought to require confirmation. The discovery, therefore, of these flint knives in gravel of the same age was interesting, especially as many tusks of a hippopotamus have been obtained from the gravel of St. Roch—some of these recently by Mr. Prestwich; while M. Garnier of Amiens has procured a fine elephant's molar from the same pits, which Dr. Falconer refers to Elephas antiquus, see Figure 19. Hence I infer that both these animals co-existed with Man.

The alluvial formations of Montiers are very instructive in another point of view. If, leaving the lower gravel of that place, which is topped with loam or brick-earth (of which the upper portion is about 30 feet above the level of the Somme), we ascend the Chalk slope to the height of about 80 feet, another deposit of gravel and sand, with fluviatile shells in a perfect condition, occurs, indicating most clearly an ancient river-bed, the waters of which ran habitually at that higher level before the valley had been scooped out to its present depth. This superior deposit is on the same side of the Somme, and about as high, as the lowest part of the celebrated formation of St. Acheul, 2 or 3 miles distant, to which I shall now allude.

The terrace of St. Acheul may be described as a gently sloping ledge of Chalk, covered with gravel, topped as usual with loam or fine sediment, the surface of the loam being 100 feet above the Somme, and about 150 above the sea.

Many stone coffins of the Gallo-Roman period have been dug out of the upper portion of this alluvial mass. The trenches made for burying them sometimes penetrate to the depth of 8 or 9 feet from the surface, entering the upper part of Number 3 of the sections Figures 21 and 22. They prove that when the Romans were in Gaul they found this terrace in the same condition as it is now, or rather as it was before the removal of so much gravel, sand, clay, and loam, for repairing roads, and for making bricks and pottery.

(FIGURE 21. SECTION OF GRAVEL PIT CONTAINING FLINT IMPLEMENTS AT ST. ACHEUL, NEAR AMIENS, OBSERVED IN JULY 1860.

1. Vegetable soil and made ground, 2 to 3 feet thick. 2. Brown loam with some angular flints, in parts passing into ochreous gravel, filling up indentations on the surface of Number 3, 3 feet thick. 3. White siliceous sand with layers of chalky marl, and included fragments of Chalk, for the most part unstratified, 9 feet. 4. Flint-gravel, and whitish chalky sand, flints subangular, average size of fragments, 3 inches diameter, but with some large unbroken Chalk flints intermixed, cross stratification in parts. Bones of mammalia, grinder of elephant at b, and flint implement at c, 10 to 14 feet. 5. Chalk with flints. a. Part of elephant's molar, 11 feet from the surface. b. Entire molar of Elephas primigenius, 17 feet from the surface. c. Position of flint hatchet, 18 feet from the surface.)

In the annexed section (Figure 21), which I observed during my last visit in 1860, it will be seen that a fragment of an elephant's tooth is noticed as having been dug out of unstratified sandy loam at the point a, 11 feet from the surface. This was found at the time of my visit; and at a lower point, at b, 18 feet from the surface, a large nearly entire and unrolled molar of the same species was obtained, which is now in my possession. It has been pronounced by Dr. Falconer to belong to Elephas primigenius.

A stone hatchet of an oval form, like that represented at Figure 9, was discovered at the same time, about one foot lower down, at c, in densely compressed gravel. The surface of the fundamental Chalk is uneven in this pit, and slopes towards the valley-plain of the Somme. In a horizontal distance of 20 feet, I found a difference in vertical height of 7 feet. In the chalky sand, sometimes occurring in interstices between the separate fragments of flint, constituting the coarse gravel Number 4, entire as well as broken freshwater shells are often met with. To some it may appear enigmatical how such fragile objects could have escaped annihilation in a river-bed, when flint tools and much gravel were shoved along the bottom; but I have seen the dredging instrument employed in the Thames, above and below London Bridge, to deepen the river, and worked by steam power, scoop up gravel and sand from the bottom, and then pour the contents pell-mell into the boat, and still many specimens of Limnaea, Planorbis, Paludina, Cyclas, and other shells might be taken out uninjured from the gravel.

It will be observed that the gravel Number 4 is obliquely stratified, and that its surface had undergone denudation before the white sandy loam Number 3 was superimposed. The materials of the gravel at d must have been cemented or frozen together into a somewhat coherent mass to allow the projecting ridge, d, to stand up 5 feet above the general surface, the sides being in some places perpendicular. In Number 3 we probably behold an example of a passage from river-silt to inundation mud. In some parts of it, land shells occur.

It has been ascertained by MM. Buteux, Ravin, and other observers conversant with the geology of this part of France, that in none of the alluvial deposits, ancient or modern, are there any fragments of rocks foreign to the basin of the Somme—no erratics which could only be explained by supposing them to have been brought by ice, during a general submergence of the country, from some other hydrographical basin.

But in some of the pits at St. Acheul there are seen in the beds Number 4, Figure 21, not only well-rounded Tertiary pebbles, but great blocks of hard sandstone, of the kind called in the south of England "greywethers," some of which are 3 or 4 feet and upwards in diameter. They are usually angular, and when spherical owe their shape generally to an original concretionary structure, and not to trituration in a river's bed. These large fragments of stone abound both in the higher and lower level gravels round Amiens and at the higher level at Abbeville. They have also been traced far up the valley above Amiens, wherever patches of the old alluvium occur. They have all been derived from the Tertiary strata which once covered the Chalk. Their dimensions are such that it is impossible to imagine a river like the present Somme, flowing through a flat country, with a gentle fall towards the sea, to have carried them for miles down its channel unless ice co-operated as a transporting power. Their angularity also favours the supposition of their having been floated by ice, or rendered so buoyant by it as to have escaped much of the wear and tear which blocks propelled along the bottom of a river channel would otherwise suffer. We must remember that the present mildness of the winters in Picardy and the northwest of Europe generally is exceptional in the northern hemisphere, and that large fragments of granite, sandstone, and limestone are now carried annually by ice down the Canadian rivers in latitudes farther south than Paris.* (* "Principles of Geology" 9th edition page 220.)

(FIGURE 22. CONTORTED FLUVIATILE STRATA AT ST. ACHEUL (Prestwich, "Philosophical Transactions" 1861, page 299).

1. Surface soil. 2. Brown loam as in Figure 21, thickness, 6 feet. 3. White sand with bent and folded layers of marl, thickness, 6 feet. 4. Gravel, as in Figure 21, with bones of mammalia and flint implements. A. Graves filled with made ground and human bones. b and c. Seams of laminated marl often bent round upon themselves. d. Beds of gravel with sharp curves.)