Chapter XXV: Antiquity of the River-Drift (2)

Turning now to the existing valley of the Little Ouse, we find, at Brandon Down, the gravel occupying the summit of a high ridge of land almost at right angles to the present course of the |682| river. It is difficult to account for its occurring in this position, unless we are to suppose that at an early period before the complete denudation of the Fen country, and while the Boulder Clay still covered the surface of the Chalk, and the level of saturation was higher in the latter than at present, a tributary stream, possibly the old representative of the Lark, flowed into the Little Ouse near this spot, and the gravel was deposited on the tongue of land near the confluence. The country drained by the Little Ouse seems at one time to have been almost covered by Glacial deposits, including beds of shingle, composed for the greater part of quartzite pebbles. The beds at Brandon Down are nearer the sea than any analogous beds towards the source of the stream, and occupy a higher position relatively to the existing river, being 90 feet above it. If they resulted from river-action, they would, in accordance with the hypothesis, be among the oldest of the river-deposits; and would, as indeed they do, consequently contain a far larger proportion of the quartzite pebbles than those of somewhat later age and farther up the valley.

At Bromehill, where the drift is but a few feet higher than the present level of the stream, and would, in accordance with the hypothesis, belong to a later period, there are but few of these quartzite pebbles, but the gravel contains a very large proportion of rolled fragments of chalk, which, so far as I have observed, are absent in the probably older beds, at Brandon Down; the implements also are frequently much rolled and water-worn. This fact is also in accordance with the hypothesis, for the river at the time of the formation of these lower beds would, in the lower part of its course, have completely cut through the Glacial deposits above the Chalk, and would have been attacking the Chalk itself. There is also an abundance of rolled chalk in the Shrub Hill beds, which seem to be of much the same age. In the valley of the Lark, the rolled chalk pebbles occur in gravels at a somewhat greater elevation. Higher up the Little Ouse, the gravel at Santon Downham occupies the slope of a hill on the inner side of a great sweep of the river, while at Thetford, the beds form a long terrace by the side of the stream, with a rather abrupt slope towards it. Here also, land and freshwater shells have been found in the gravel, but neither these nor implements have as yet been observed in the gravels of the valley of the Little Ouse, or of its tributaries, above Thetford.

Tracing the main stream back to its source, we find that both |683| the Little Ouse and the Waveney, the one flowing westward, and the other eastward, take their rise in the same valley, and within a few hundred yards of each other, at Lopham Ford. With regard to the elevation of this spot above the sea-level, there has been some diversity of opinion. On the Greenough map, published by the Geological Society, it is erroneously stated at 15 feet; and Mr. Flower,[2772] in arguing in favour of his views, that the beds at Brandon are not connected with any river-action, assigns it a height of only 23 feet above high-water mark. That this also is erroneous can be readily shown, for Sir Joseph Prestwich[2773] has recorded the level of the Waveney at Moor Bridge, near Hoxne, ten miles below its source, as being 59 feet 9 inches above high-water mark at Yarmouth. Mr. Alger, of Diss, who has surveyed the district, informs me that the level at Lopham Ford is 75 feet 3 inches above high-water mark; and as by actual survey he found the fall, from the head of the Waveney to Hoxne Mill, to be upwards of 15 feet, there can be little doubt of this level being approximately correct. Still, the gravel beds at Brandon being upwards of 90 feet above high-water mark, there can be no doubt of their being at an elevation actually above the source of the present stream; and at first sight, this fact appears difficult of reconciliation with the view that they are due to fluviatile action. Without, however, calling to aid any possible oscillations in the level of the land, varying in amount at different parts of the course of the stream, an examination of the local geological conditions suffices to throw light on the causes, why the erosion of the land at the sources of the Little Ouse and Waveney has been abnormally great; so that not only have the streams excavated back the heads of their respective valleys until they have met, but their inclination at the upper part of their course, instead of being as usual in chalk countries at the rate of 12 to 18 feet in a mile, is only about 18 inches.

The general level of the country for some distance around Lopham Ford is at least 100 feet above it, and the Chalk and the superimposed beds are for the most part covered with a deposit of impervious Boulder Clay, through which the valleys of the Little Ouse and of the Waveney have been cut. But, at the time of the last emergence of this district of country from beneath the sea, this clay must have been continuous across the tract since |684| excavated, so that at that time the sources of the streams flowing in either direction must have been at least 100 feet above their present level, and 80 feet above the gravels at Brandon Down, and probably at some distance apart. That the heads of the two streams should have cut back their valleys, and at last have met, appears to be due to the fact that, previously to the covering of Boulder Clay being deposited, there existed an old depression in the Chalk, which had been filled with laminated sandy clays, either Glacial or belonging to what is known by geologists as the Chillesford series. These being more easily acted on than the chalk by running water, led the streams to follow the course of the old depression which they filled, and it is to their presence that the small inclination of the upper part of the valley of the Waveney appears to be mainly due. Another cause is to be found in the country near Lopham Ford being coated with clay, so that the streams, even at the present day, exhibit the remarkable phenomenon of being liable to floods at their source. An isolated hill, about 30 feet high, formed of the laminated beds, and with a slight capping of gravel, still remains in the valley of the Waveney, near Redgrave, to show the nature of the beds which have been removed.

The only spot in the valley of the Waveney, where as yet Palæolithic implements have been found, is at Hoxne, where the summit of the beds is about 111 feet above high-water mark at Yarmouth, and though at a higher level than the existing source of the Waveney, probably much below the level of its earlier source. Since the beds were deposited, the surface of the ground in the neighbourhood has been completely remodelled by subaërial denudation, and they now lie in a trough on the summit of a hill,[2774] both sides of which slope down to small streams which are tributary to the Waveney, and are still at work cutting out their valleys in the Boulder Clay. The beds in which the implements occur are beyond all doubt of freshwater origin, being full of freshwater shells. The trough in which they lie, has much the appearance of the deserted bed of a river, silted up under more lacustrine conditions. Such a change in the position of a river-bed, and its subsequent infilling, is quite in accordance with the hypothetical case of river-action, especially when, as here, its eventual valley had not been distinctly carved out.

The phenomena at Hoxne have lately been more fully examined |685| by Mr. Clement Reid,[2775] by means of grants from the British Association and the Royal Society; and the views that I expressed in 1872 have been in the main corroborated. The deposits are proved to be distinctly more recent than the Chalky Boulder Clay of the district, and there is evidence of oscillations in climate since the valley was formed in which the lacustrine beds were laid down, and before any Palæolithic implements or the brick-earth containing them had been deposited.

The beds at High Lodge, near Mildenhall, are of somewhat similar character to those at Hoxne, though occupying a depression on the slope of a hill, instead of a trough on the summit; and were probably deposited under nearly the same circumstances, though as yet no testaceous remains have been found in them.

Turning south, to the valley of the Thames, we find the gravel-beds at Acton and Ealing, though occasionally at a higher level, forming a terrace 80 or 90 feet above Ordnance Datum, along the side of the broad valley, at a height of some 50 feet above the general surface of the valley. In the bottom of this are spread out other beds of gravel, sand, and brick-earth, exactly as might be expected on the river-hypothesis; while at Highbury New Park, and Hackney Down, we have beds of the same character, which contain land and freshwater shells and flint implements, at a height, in some cases, of 100 feet above Ordnance Datum. The presence of these beds in such a position, consisting, as they do at Highbury, of sand and brick-earth, such as can only have been deposited in comparatively tranquil water, involves the necessity either of a large lake having existed at the spot, or of its having been within access of the flood-waters of the river. But either of these conditions is impossible, unless we are to suppose that the lower part of valley of the Thames, in which London now stands, was at that time non-existent. It must, therefore, have been subsequently excavated. But again, at lower levels at Hackney Down, and in Gray’s Inn Lane, we have gravels of a more distinctly fluviatile character, and also containing palæolithic implements. The existence, character, and position of all these beds is, therefore, perfectly in accordance with the theory of the excavation of the valley by the river, and it is extremely difficult, if not impossible, to account for them satisfactorily in any other manner.

At Hitchin beds of much the same character occur, which there also are newer than the Boulder Clay of the district. |686|

At Caddington the discoveries are quite consistent with the hypothesis, but point to a period when the excavations of the existent valleys had made but little progress.

Higher up the Thames valley at Reading and at Oxford the phenomena are all in accordance with the hypothesis; at the former place the river has deepened its valley to the extent of at least 100 feet.

The discoveries in the gravels capping the North Downs and those made near Ightham and Limpsfield in the transverse valley at the foot of the Downs, seem at first sight difficult to reconcile with any river-theory. But assuming that the beds capping the hills were at one time continuous with others in the Wealden area, and that the transverse valley was produced by denudation at a later date, the difficulties disappear, though the time requisite to effect such superficial changes may seem to be immense.

Passing by other localities where implements have been found in the valley of the Thames, such as Swanscombe and Northfleet, though it may be observed that the gravels in which they have occurred are, on the river-theory, exactly where they might have been expected to be present, we come to the beds near Reculver, where they have been found in large numbers. Looking, however, at the enormous encroachment of the sea, even within the last few centuries, upon the soft cliffs of sand and clay at that spot, it is difficult to form any satisfactory idea of the conditions under which a river may have flowed near the spot at a remote period, or of the position of the coast at the time. Where, however, as is here the case, a large tract of land has been washed away, which must of necessity have had its system of superficial drainage by streams, and may possibly have had rivers passing through it, which now, owing to the altered conditions, find their way into the sea at a point much nearer their source than formerly, we should expect to find on the top of the cliffs traces of the former state of things; and where any portion of the slope of an old valley remained, to see its gravels, though now so close to the sea, at a height far above its level. Still, it is hard to say whether the implement-bearing beds at Reculver are connected with the old valley of the Thames, or with that of some other stream which has now disappeared, but of which the upper portion is to be traced in the Swale, which now separates the Isle of Sheppey from Kent, and which appears to afford, in its junction with the West Swale and Long Reach, an instance of two valleys being gradually eroded inland until they met. The beds may even be connected with the |687| valley of the Stour; for it is by no means impossible that the present second and northward mouth of that stream may run along the valley of an old river, which originally flowed southward past Reculver, and joined the old representative of the Stour, somewhere to the south of where is now the village of Sarre.

The great tract of gravel which at some little distance inland fringes the East Essex coast, between Shoeburyness[2776] and the Blackwater estuary, may also be connected with some old river; but as yet no well-defined implements or freshwater shells have been found in it, though Mr. Whitaker has discovered shells near Southend. The fluvio-marine deposits at a lower level at Clacton, just north of the Blackwater, like those at Chislet, in Kent, seem to belong to a somewhat later period, when the rivers had so far deepened their beds as to have become tidal.

Though no land or freshwater shells have as yet been found in the gravel beds near Canterbury, yet their position is quite in accordance with the theory of the excavation of the valley by river-action; and here as elsewhere the implements from the lower beds are often much water-worn.

The superficial deposits of the south of Hampshire and the Isle of Wight, and in a lesser degree those of the neighbouring counties, have been fully discussed in an able paper by Mr. T. Codrington, F.G.S.,[2777] though since it was published a large number of implements has been found near Bournemouth, Barton, and Hordwell. He has pointed out that the whole of the New Forest, between Poole and Southampton Water, appears at one time to have been an extensive plain, with a gradual slope to the south, very generally covered with gravel and brick-earth. This has since been in great part cut up, and over large areas entirely removed by the action of the streams and rivers, which latter flow in well-defined valleys.

The formation of this table-land and the overlying deposit of gravel which, in places far inland, is found at a height of more than 420 feet above the present sea-level, appears to be due to marine action, though as yet no marine remains have been discovered in it. Sea-shells have, however, been found by Sir Joseph Prestwich[2778] in an old sea-beach at Waterbeach, near Goodwood, and similar beds, at Avisford Bridge, near Arundel, occur at a height of 80 or 100 feet above the sea. We seem, then, here to have evidence |688| of a considerable elevation of the land from beneath the sea; and as the gravel in places overlies late Tertiary beds, this must have taken place at a comparatively late geological epoch. When rivers run through a tract of country covered with a marine gravel of this kind, itself apparently deposited in a somewhat contracted area, it is, in the absence of organic remains, difficult to distinguish the reconstructed gravels resulting from fluviatile action, from the older beds. Any one, however, who is acquainted with the country, or who will examine Mr. Codrington’s map, will see what an enormous denudation has been effected in this great sheet of gravel, by rivers and streams, and by subäerial action. When once the protecting gravel has been cut through, and the soft Tertiary beds of sand and clay below have been reached, the process seems to go on with great rapidity. A large tract of land west of Southampton appears to have been in this way almost cleared of its gravel, of which but patches are left. Even the principal portion of the old table-land which has survived, that to the east and south-east of Fordingbridge, is deeply cut into by numerous valleys, many of a depth of 200 feet. The existence of these valleys is clearly in accordance with the river theory.

Let us now examine the discoveries in the valleys of the Test and of the Itchen from this point of view. Looking at the numerous instances of the finding of flint implements in gravels containing terrestrial and freshwater remains, and looking at the improbability of their occurring in a purely marine deposit, I venture to regard them as being equally characteristic of freshwater deposits as any organic fossils, and to claim the beds in which they occur as being of freshwater origin.

At Southampton several implements have been found in the pits upon the Common at heights ranging from 80 to 150 feet above the sea-level. The gravel there slopes at a considerably greater inclination than that of the table-land nearer Chilworth, with which it is continuous, and from which it would appear to have been in part derived. It occupies a tongue of land between the valley of the Itchen and that of the Test, now widened out by tidal action. It is in places covered by brick-earth, and its position and character are quite in accordance with a fluviatile origin. If, from their proximity to the apparently marine gravels, we assume these beds to belong to an early period in the history of the excavation of the valley, their high position above the present tidal stream is such as, according to the hypothesis, was to be expected. |689|

The gravels found lower down the course of the river, at Hill Head, Brown Down, and Lee on the Solent, appear to belong to a somewhat later period; and to bear much the same relation to those of Southampton Common, as do the beds at Shrub Hill to those of Brandon Down. As I pointed out long ago, “There can be but little doubt that these gravel beds are merely an extension of the valley-gravels of the rivers Test, Itchen, Hamble, and other streams, which at the time they were deposited, flowed at this spot in one united broad stream, at an elevation of some forty feet above the existing level of their outfall, over a country which has since, by erosive action, been in part converted into the Southampton Water.”[2779] We shall shortly have to revert to this circumstance; but before returning to the coast, we must take a short glance at the features of the discoveries near Salisbury.

In the neighbourhood of this city there can be no doubt of the deposits being thoroughly in accordance with the river theory. The Fisherton and Milford Hill beds occupy points or spurs of land, in the forks above the junction of streams, or precisely those spots in which their presence was to be expected. There are the usual beds of gravel, sand, and clay, the usual bones of the Quaternary fauna, some representing what are now Arctic species, and therefore presumably indicative of a severer climate than at present; and the usual land and freshwater shells. Though the valleys, being confluent, are excavated to the same depth, yet, on examination, their sectional areas will be found to be approximately proportional to the extent of country drained by the rivers still flowing through them. At Milford Hill, the deposit is cut off from the main spur of land by a kind of transverse valley, about thirty feet in depth, besides having on either side a valley some 100 feet deep. On any hypothesis of the beds having been deposited by aqueous action—and no other can for a moment be entertained—these valleys must have been mainly excavated since the deposition of the gravels. For had the valleys at that time existed, we can conceive of no conditions under which a body of water sufficient to fill the valleys to their summit, and able to carry along detrital matter with it, would leave its heavy contents at the top of the hills instead of at the bottom. The old fluviatile beds occur also at various levels on the slopes, in complete accordance with the theory of gradual excavation; and farther down the valley, at |690| Fordingbridge, we find them again occurring with remains of _Elephas primigenius_ at about forty feet above the river.

The circumstances of the discoveries at Bournemouth seem at first sight almost irreconcilable with any river-hypothesis; as it is difficult to conceive how gravels capping the cliffs along the sea-shore for miles, and at an elevation of from 130 to 90 feet above its level, can have been deposited in such a position by the agency of a stream. And yet on a closer examination of the case, all such difficulties vanish, and the ancient existence of a river at such an elevation, and running in such a direction that it would leave these gravels to testify to its former course, seems absolutely demonstrable. Without being aware of the results at which others had arrived, I came, after due consideration of the facts of the case, to the conclusion that, as has already been mentioned in an earlier page, there must in ancient times have existed a river draining an extensive tract of country along the southern coast, and flowing in an easterly direction; and that of this river a portion still survives in an altered and enlarged condition as the Solent Sea, which separates the Isle of Wight from the mainland. Mr. Codrington, whose paper I have already so often quoted, arrived on independent grounds at substantially the same conclusion. But at an earlier epoch still—in 1862—before any flint implements had been found at Bournemouth, or indeed in any of the gravels of the South of England, the late Rev. W. Fox,[2780] of Brixton, in the Isle of Wight, published nearly similar views as to the origin of the Solent. As his opinions cannot by any possibility be supposed to have been influenced by preconceived views as to the antiquity of man, I prefer stating the case, in the first instance, in his words rather than in my own:—“The severance of this island (the Isle of Wight) from the mainland, it appears to me, was effected under very unusual circumstances, and at a very distant period. The present channel of the Solent, being pretty nearly equally deep and equally broad throughout its entire length of twelve or fourteen miles, proves at once that it was not formed in the usual way of island-severing channels, that is, by gradual encroachments of the sea on the two opposite sides of a narrow neck of land” . . . “it is to be accounted for, therefore, not by the excavations of a gradually approaching sea, but, as I shall hereafter have to attempt to show, by its being originally the trunk or outlet of a very considerable river.” . . . “Whoever, as a geologist, |691| examines the vertical strata of the Chalk at the Needles, nay, and throughout the whole length of the Isle of Wight, and the strata of the same rock in exactly the same unusual position on the bold white cliff on the Dorsetshire coast some twenty miles westward of the Needles, will not doubt but that the two promontories were once united, forming a rocky neck of land from Dorset to the Needles. This chain of chalk might, or might not, be so cleft in twain as to allow the rivers of Dorset and Wilts to find a passage through them to the main ocean. My opinion, however, is that they had no such outlet, but that at that far distant period, the entire drainage of more than two counties, embracing the rivers that join the sea at Poole and Christchurch, flowed through what is now called Christchurch Bay, down the Solent, and joined the sea at Spithead.”

“According to this theory, the Solent was at that time an estuary somewhat like the Southampton Water, having but one opening to the British Channel, but of so much more importance than the latter as it was fed by a vastly greater flow of fresh water.” “Of course, according to this view, the sea would lose its original condition as an estuary at the time when the British Channel had so far made a breach through the chain of rocks connecting the Isle of Wight with Dorsetshire as to give an opening into itself for the Dorsetshire rivers, somewhere opposite to the town of Christchurch. From that time forth the Solent would become what it is at present, losing its character as an estuary, and assuming that of a long narrow sea.” . . . “The distant period at which such changes took place it would be hopeless to guess at, amid the dimness of the data on which calculations could be founded. It could not be less, however, than many thousands of years, seeing that since that time, the British Channel has not only made a broad breach of twenty miles through a chain of slowly yielding rocks, but has also pushed its way gradually across the broad extent of the Poole and Christchurch Bays.”

Such is the theory of Mr. Fox, which places the probable course of events fully and fairly before our view. I see in it but little on which to comment, except that it does not appear to have sufficiently taken into account the widening of the Solent subsequently to the time of its becoming a channel of the sea; and that in a passage, which I have not quoted, Mr. Fox estimates the drainage area of the ancient river as but little inferior to that of the Thames or Humber. Taking the basins of all the streams discharging |692| into the sea between Ballard Down, near Poole, on the west, and Calshot Castle and the Medina on the east, but not including the latter river, I find that, according to the Ordnance Map,[2781] the present land area which would have drained into an ancient river such as that supposed, is 1,617 square miles. To this may be added another 100 square miles, representing the area included between the present coast and an extension of the chalk downs from Ballard Down to the Needles, the whole of which has been washed away; though within this large area, the present depth of the sea attains in but very few places to ten fathoms. The drainage area of the ancient river Solent can therefore have been but about one-third of that of the Thames and its affluents, unless we are to suppose that, as is the case in the neighbourhood of Carisbrooke Castle and with the Medina, a portion of land to the south of the old chalk downs drained northward through some gap in the range of hills. That such land existed seems probable, from the occurrence of gravels with elephant remains along the south-west coast of the Isle of Wight at an elevation of 80 feet and upwards above the sea, which, Mr. Codrington has suggested, may have been deposited by tributary streams of a river flowing northwards through the chalk range to the Solent. But even with any such addition the area drained by the old River Solent can hardly have been half that of the basin of the Thames.

With regard, however, to the former existence of this range of chalk hills and the land to the north of them, Mr. Codrington has shown, in the paper already so often quoted, that the spreading out of the marine gravel, and the levelling of the table-lands was probably effected in an inlet of the sea, shut in on the southern side by land which connected the Isle of Wight with the mainland, and opening to the eastward. Assuming, then, the existence of this ridge of high land, it is evident, as Mr. Fox has pointed out, that the only outlet for the rivers now represented by the Frome, the Trent or Piddle, the Stour, Blackwater, Avon, and other streams now discharging into the sea, must have been by an eastward channel, in fact, a continuation of the rivers now discharging through Poole Harbour. The course of such a river would naturally be guided, in the first instance, by the configuration of the surface of the old marine gravels of the sloping table-land. This, as has been shown, slopes upward from the present coast northward, and attains its highest level inland; but traces of the same gravel occur also in the Isle of Wight, though it there |693| slopes upward in a southerly direction, attaining a height of 368 feet at St. George’s Down, but being only from 100 to 160 feet above the sea in various places along the northern shore of the island, at a distance of about a mile inland. It appears, therefore, that there must originally have been a valley running east and west in the old marine gravel, forming a natural course for the drainage of the country, and probably finding its way towards the sea, somewhere within the space now occupied by the Solent and Spithead, though not actually discharging into the sea until it had attained some distance eastward.

Evidence as to the highest level at which freshwater action removed and re-deposited the marine gravel on the southern slope of the valley is at present wanting; but, judging from a section across the Isle of Wight from St. George’s Down to Norris Castle, given by Mr. Codrington, the declivity is so much more rapid below the 160 feet level than above it, that the ancient river may have commenced its action at about that level. How far eastward the Isle of Wight may have extended at that time it is difficult to say; but from the enormous denudation of land to the west, and the range of the ten-fathom line, there may probably have been land at all events as far east as opposite to Selsey, the extensive estuarine beds at which place, containing remains of _Elephas primigenius_,[2782] are possibly connected with this old river.

The precise manner in which the Foreland gravels and brick-earth, in which Mr. Codrington found a palæolithic implement at 85 feet above the sea-level, were connected with the old river-deposits, is difficult to determine. Mr. Codrington is inclined to think that a rise of land to the extent of 70 or 80 feet must have taken place since the deposition of the brick-earth in which the flint implement was embedded, but this to me seems unnecessary. It is, however, unsafe to speculate on a single specimen found in such a position. The implements found at Seaview and Bembridge may have been washed out of gravel-beds at a lower level than those of the Foreland, or even have been transported for some distance by marine currents.

Turning to Bournemouth, where so many more have been found, the highest and most westerly point at which implements have occurred appears to be about 130 feet above the sea.[2783] Farther |694| east, near Boscombe, the level is about 120 feet; midway between that spot and Hengistbury Head, the height of the gravel is 90 feet; at High Cliff, 84 feet; at Hordwell, where implements abound, a short distance inland, 60 feet; and about midway along the northern shore of the Solent, 50 feet. The surface of the ground is, of course, much cut up by the numerous streams coming in from the north; but the general fall of the gravel from west to east is perfectly in accordance with its having been deposited in the valley of an ancient river running in this direction, the whole of the southern side of which has since been carried away by the sea. Whether the old river had become tidal so far west as Hurst Castle, when first it was intercepted by the sea to the south, does not appear to me to be a matter of importance, inasmuch as no doubt a valley was already formed, along the course of which the encroachments of the sea would be more rapid than where the cliffs were higher, and more solid matter had to be removed. That the valley, in which is now Southampton Water, was also originally, for the most part, scooped out by the rivers coming from the north, which in remote times flowed into the old River Solent, is, I think, beyond all reasonable doubt. The increased volume of the ancient river, after receiving so important an affluent, is evinced by the widening of the channel, from Calshot Castle eastward by Spithead, to a full third more than it is to the west, along what is now the Solent Sea.

As to the character of the gravels at Bournemouth, it is, as already observed, hard to distinguish those presumably of fluviatile origin from the older and probably marine beds. In the railway-cutting between Bournemouth and Christchurch, I thought, however, that in places I could trace the superposition of the one upon the other. The more recent deposits contain water-worn fragments of quartz, granite, and porphyry, as was noticed long ago by Mr. Godwin-Austen,[2784] who, from this circumstance, saw reason for connecting them with the gravels capping the tabular hills of Devon and Dorset to the west.

It is, of course, evident that at the period when the river ran at this high level, past the spot where now is Bournemouth, all the land to the immediate west must have been far higher than it is at present, and that Poole Harbour could not have existed. In attempting to reconstruct the map of a country, the shores of which have been much wasted by the sea, in order to show what |695| must have been at some remote period the old coast-line, the task is rendered difficult and within certain limits impossible by the absence of any evidence as to the elevation above the sea of the land removed, and as to the channels along which the sea could work. In this case, however, there is a strong presumption as to the unbroken continuity of the chalk-range, and of its elevation having been much the same throughout, as it now is at both ends of the breach. The general character of the beds above the Chalk, so far as their power of resistance to water-action goes, seems also much the same at either extremity; though perhaps the beds at the Isle of Wight end of the breach are somewhat the harder. Assuming nearly equal conditions, and looking at the form of the present coast-line, which is indented by two distinct broad bays, it seems probable that the old course of the river may have been intercepted by the sea at two several points, the one nearer Poole and the other nearer Lymington. Directly this closer communication with the sea was formed for the Dorsetshire rivers, they would, of course, owing to the more rapid fall, excavate their valleys with greater speed at their mouth, and directly they became tidal, the sea would make rapid inroads on the soft sand and clay exposed to its action. So effective is this action, that at Hordwell Cliff the waste of the shore is said to be now going on at the rate of about a yard per annum,[2785] or upwards of half a mile every thousand years, though perhaps this is somewhat exaggerated.

In discussing this question, I have purposely avoided complicating the subject with the effects of any general lowering of the surface of the ground by erosion either chemical or mechanical; or of upheavals and depressions of the land during the period of the formation of the valleys, though no doubt this also has taken place, especially along the southern coast of Britain. I must, however, mention the existence of a submerged forest, occasionally visible at low water, at the foot of the cliffs at Bournemouth, which seems to show that there as elsewhere a depression of a former land surface has taken place. The late Mr. Albert Way, F. S. A., who had the opportunity of examining some of the stumps of trees exposed at rare intervals at low water, informed me that they appeared to be those of the true Scotch fir; and also that local tradition speaks of an impassable morass having, so late as the commencement of the present century, |696| intervened between the line of cliffs and the sea. On the occasion of one of my visits to Bournemouth, some of these stumps were fortunately visible, and were pointed out to me by Mr. Way at a spot but a few yards to the west of the pier, and between high and low water-mark. They appear to be of no very remote antiquity, geologically speaking, and to be connected rather with the present valley of the Bourne than with the valley of the old river Solent, as the trees, some of which were fully a hundred years old, grew on the surface of a thick bed of hard peat. Under any circumstances, however, the presence of such remains at the foot of the cliff does not tend to diminish our estimate of the antiquity of the freshwater beds containing the works of man, which we find occupying their summit.

In passing the deposits containing flint implements in different parts of this country under review, enough has, I think, now been said to show that in position, in character, and in the nature of their organic contents, they are perfectly in accordance with what might have been expected from river-action under certain circumstances. The case might indeed have been made much stronger had deposits in other places, in all respects similar, except that the presence of flint implements has not as yet been observed in them, been brought into account; and it must not be forgotten that this might, with perfect propriety, have been done, as there can be no possible doubt that a certain series of gravels, sands, and clays, containing organic remains and flint implements in extremely variable quantity, all belong to one geological period, and owe their existence to similar causes.

But though on no other hypothesis than that of river-action can the phenomena be accounted for, yet, as has already been seen, it is necessary, in order that river-action should have produced such effects, that the streams, during some portion of the year at all events, should have been more torrential in character than they are at the present day. If, however, we see satisfactory grounds for attributing these beds containing land and freshwater shells and remains of terrestrial animals, to rivers formerly flowing at much higher levels than at present, which have since excavated their valleys—and it seems impossible to do otherwise—then we must also accept as a fact that the climatal conditions were such as would enable the rivers to perform the work. It is, as Sir Joseph Prestwich[2786] has shown, quite out of the question |697| to suppose that with the valleys excavated to the present depth, any meteorological causes could fill them to their summits; or even if they could and did, that they would leave such deposits as we find at high elevations on their slopes, or even on detached eminences. It will, however, be well to examine briefly any corroborative evidence that may be forthcoming, as to the probability either of a severer climate involving a greater accumulation of winter snows, or of a greater rainfall, or of both. The one, indeed, seems hardly probable without the other, as a cold land surface “presented to vapour-laden sea-winds, as in the mountainous districts of the north-west of Spain, in our own lake districts, and in Scandinavia,”[2787] involves of necessity a heavy rainfall.

With regard to climate, we may take into account that which prevailed at a somewhat earlier date; for there appears no doubt that the flint implement-bearing gravels are all of later date than the Chalky Boulder Clay of the Eastern Counties, a deposit which belongs to the so-called Glacial Period, during a portion of which a great part of England and Scotland was submerged beneath the sea, and became coated with masses of Boulder Clay and other deposits, derived for the most part from the moraines of glaciers, sometimes at no great distance, and possibly in the main transported and dropped in their present positions by means of icebergs and coast ice. That they are of later date is proved by more than one of the implement-bearing beds reposing in valleys either in, or cut through, this Chalky Boulder Clay; and at Hoxne the interval between the Glacial deposits and the Palæolithic beds is marked by two sets of lacustrine strata, the lower and earlier with a flora characteristic of a mild climate, and the upper by one which points to the recurrence of Arctic conditions. Prof. Boyd Dawkins[2788] has suggested the probability of the higher ground of North Wales and the northern part of England having been still enveloped in an ice-mantle at the time that the mammoth, reindeer, and other post-glacial mammals were living in the lower and less inclement districts. But this view is to some extent founded on negative evidence, and on the assumption that palæolithic implements do not exist in this northern area. I have already commented[2789] on the possibility of implements being eventually found in it.

The crumpling and contortion of some of the beds of |698| River-drift, especially at high levels, has been regarded by Sir Joseph Prestwich[2790] as possibly resulting from the lateral pressure produced by packing and jamming together of blocks of ice, such as may now be witnessed in rivers like the Danube and the Rhine. The “trail and warp” of Mr. Trimmer, those superficial deposits so common over a large portion of this country, which, indeed, constitute so large a portion of the arable soil, seem also, as the Rev. Osmond Fisher[2791] has pointed out, to be significant of a severer climate than at present prevails. The “Palæolithic floors,” both near London and at Caddington, are buried under a considerable thickness of this “trail.” There is moreover a high probability that, at the time of the deposit of the gravels, Britain was still united to the continent; so that, apart from other causes, there was a tendency for the climate to partake more of a continental character than at present, and to induce greater cold in winter and greater heat in summer.

That the existence of enormous glaciers is as indicative of the action of heat, in order to convert the water of the ocean into vapour, as of cold to condense it, has been insisted on by Professor Tyndall,[2792] and even more strongly by Professor Frankland. If at the time of the rivers flowing at the high level, Britain was still connected with the continent, it is by no means impossible that the temperature of the seas on either side of the connecting isthmus may have been different. That connected more immediately with the Southern Ocean would have been the warmer of the two, from which a copious supply of vapour would be carried by the southerly winds, and be condensed as rain in its passage northward.

Mr. Alfred Tylor, F.G.S.,[2793] in his profusely illustrated papers on the Amiens gravel, and on Quaternary gravels, contends for the existence of a “Pluvial period” subsequent to the Glacial, in which the rainfall was far greater than at present, and such a view has much to commend it for acceptance. But when he proceeds to assert that the surface of the Chalk in the valley of the Somme, and in all other valleys of the same character, had assumed its present form prior to the deposition of any of the gravel or loess now to be seen there, and to argue that the whole of the gravels at all levels on the slopes are of one age, and due |699| to floods extending to a height of at least 80 feet above the level of the rivers, we may well hesitate before we give in our adhesion to such views. In the first place, it is, to say the least of it, unphilosophical to rely too much on a single example, such as that of the valley of the Somme; and to account for its phenomena by causes which are evidently incapable of producing the effects observable in other localities, as, for instance, at Southampton, close to the sea, and 160 feet above its level. But what shall we say to floods raising the levels of rivers upwards of 80 feet, yet having no erosive power, and the waters of which, regardless of the laws of gravity, tranquilly deposited their solid contents evenly over the slopes, or often in the greatest thickness on their higher part, and in some cases on almost isolated hills, instead of principally on the bottom of the river-valley? Whence all the materials for the gravels are to be derived, how they are to be reduced to a subangular condition by water-wear, especially in the case of the flint implements occurring in the gravels, are points on which further information will have to be supplied, before any such views can be seriously entertained.

I have up to this point almost left out of view any distinctive differences between the deposits at a high level and those at a low level in the river-valleys. That such, however, exist has been pointed out by Sir Joseph Prestwich;[2794] and judging from the northern range of the group of shells found in the high-level beds, the absence of southern species, the character of the mammalian and vegetable remains, the transport of large blocks such as could only be effected by ice and the other physical features of the case, he is inclined to assign a winter temperature to the period of their deposit from 19° to 29° Fahr. below that which now obtains in these regions. From a consideration of the features of the low-level deposits he considers that at the time of their deposit, the climate was rather less severe, by about 5°. The presence of the mammoth and woolly-haired rhinoceros, animals specially adapted for cold climates; of the musk-ox, the reindeer, the lemming, and marmot, corroborates the same view; while the hippopotamus, which seems characteristic of the low-level deposits, is suggestive of a somewhat warmer climate. Like the mammoth and rhinoceros, its structure may, however, have been somewhat modified, so as to enable it to occupy colder regions than at present, or it may have been merely a summer visitor ranging northwards before the |700| separation of Britain from the continent. Under any circumstances its presence seems to indicate that the volume of the rivers was probably in excess of what it is at the present time. But whatever may have been the degree of winter cold, or the amount of the snow and rainfall, the one was not so extreme as to prevent there being an abundance of animal life, nor the other so great as to interfere with the growth of a sufficient supply of vegetable food on which it might subsist.

It has, indeed, been supposed by some that the remains of the early mammals occurring in the gravels are derived from older beds, and that their presence in association with flint implements no more proves the contemporaneity of the men who made those implements with the old Quaternary fauna, than their association with Chalk fossils proves that mankind were originally inhabitants of the bed of the Cretaceous ocean. Did the gravels only occur at such levels as are within reach of existing streams, there might be some reason in such a view, which is, moreover, in certain cases and within certain limits, probably correct. For we have seen how in the course of the excavation of a valley, the beds deposited at one time are liable to be disturbed at another, and re-deposited in a fresh place; which could hardly happen without an admixture of fresh materials, some probably of a more recent date. In the process of transport, however, not only the implements but the still softer bones are liable to wear and abrasion of the angles, and it is impossible to conceive that, assuming the Quaternary fauna to have disappeared from this region before the valleys were excavated, and the implement-bearing beds deposited, their bones could still exist in such numbers, and so often in an unrolled condition in the low-level beds.

Had this older fauna disappeared, it is evident that man could not have subsisted here alone, unaccompanied by other animals to furnish him with food; and if these animals belonged to the later or “prehistoric” fauna, where, as Sir John Lubbock pertinently asks, are their bones? If, however, we acknowledge that the pleistocene mammals still occupied this country at the time of the low-level beds being formed, and if we find their remains also in those at a high level, and at all intermediate heights, it is evident that they must have persisted here during the whole period of the excavation of the valleys; while, if we find also flint implements in an unrolled and unworn condition at all heights, it is evident that those who made them must also have been |701| co-occupants of the region during the same period. If, indeed, as appears to be in some valleys the case, the unworn implements occur only in the high-level deposits, while in the lower they are either absent or in a much worn condition, the inference is, that in those particular valleys the occupation by man, though for some time contemporaneous with that of the mammoth and his congeners, ceased before the extinction or emigration of the old fauna. In some cases, however, as at Fisherton,[2795] the worked flints have been found below the remains of mammoth; while in the beds at Menchecourt, near Abbeville,[2796] in which the implements occur, were found the bones of a hind leg of rhinoceros still in their natural position, so that they must have retained their ligaments when deposited, and could not since have been disturbed. With regard to the amelioration of climatal conditions which led to the cessation of the excavation of the valleys, it may not impossibly have been connected with the insulation of the country, when the isthmus connecting it with the continent was cut through by the sea. But this is hardly the place for such speculations. If, however, we may regard the estuarine deposits at Selsey, in which almost entire skeletons of mammoth occur, as belonging to the period when the deposit of the low-level gravels was ceasing, it would appear from the associated molluscan forms, as interpreted by Mr. Godwin-Austen, that the temperature of the waters of the English Channel was at that time such as may now be met with twelve degrees farther south.

If there was a difference in the climatal conditions of the high and low-level deposits, it might have produced some effect on the method of living, and on the implements of the men of the two periods. At one time I thought it probable that a marked distinction might eventually be drawn between the high-and low-level implements, but so far as Britain is concerned, this can hardly be done. Still the _facies_ of a collection from two different spots is rarely quite the same, and I think there is generally a preponderance of the ruder pointed implements in the high-level gravels, and of the flat ovate sharp-rimmed implements in the low-level. In the valley of the Somme, the broad polygonal flakes are certainly most abundant in the lower beds, as at Montiers, near Amiens.

I would, however, deprecate the introduction of such terms as |702| “Eolithic” and “Mesolithic” in order to distinguish two phases in the Stone Period as being both unfounded and misleading. We know not where or when the dawn of human civilization arose, but it was probably long before the date of our earliest River-Gravels and in some part of the world more favoured by climate than Britain. Why then should we speak of British implements as Eolithic? And how can we apply the term Mesolithic to a period intervening between the Palæolithic and Neolithic Ages, when we know neither when the one ended nor when the other began?