DRIFT CARRIED BY ICEBERGS.

An account was given so long ago as the year 1822, by Scoresby, of icebergs seen by him in the Arctic seas drifting along in latitudes 69 and 70 degrees north, which rose above the surface from 100 to 200 feet, and some of which measured a mile in circumference. Many of them were loaded with beds of earth and rock, of such thickness that the weight was conjectured to be from 50,000 to 100,000 tons. A similar transportation of rocks is known to be in progress in the southern hemisphere, where boulders included in ice are far more frequent than in the north. One of these icebergs was encountered in 1839, in mid-ocean, in the antarctic regions, many hundred miles from any known land, sailing northward, with a large erratic block firmly frozen into it. Many of them, carefully measured by the officers of the French exploring expedition of the Astrolabe, were between 100 and 225 feet high above water, and from two to five miles in length. Captain d'Urville ascertained one of them which he saw floating in the Southern Ocean to be 13 miles long and 100 feet high, with walls perfectly vertical. The submerged portions of such islands must, according to the weight of ice relatively to sea-water, be from six to eight times more considerable than the part which is visible, so that when they are once fairly set in motion, the mechanical force which they might exert against any obstacle standing in their way would be prodigious.

We learn, therefore, from a study both of the arctic and antarctic regions, that a great extent of land may be entirely covered throughout the whole year by snow and ice, from the summits of the loftiest mountains to the sea-coast, and may yet send down angular erratics to the ocean. We may also conclude that such land will become in the course of ages almost everywhere scored and polished like the rocks which underlie a glacier. The discharge of ice into the surrounding sea will take place principally through the main valleys, although these are hidden from our sight. Erratic blocks and moraine matter will be dispersed somewhat irregularly after reaching the sea, for not only will prevailing winds and marine currents govern the distribution of the drift, but the shape of the submerged area will have its influence; inasmuch as floating ice, laden with stones, will pass freely through deep water, while it will run a ground where there are reefs and shallows. Some icebergs in Baffin's Bay have been seen stranded on a bottom 1000 or even 1500 feet deep. In the course of ages such a sea-bed may become densely covered with transported matter, from which some of the adjoining greater depths may be free. If, as in West Greenland, the land is slowly sinking, a large extent of the bottom of the ocean will consist of rock polished and striated by land-ice, and then overspread by mud and boulders detached from melting bergs.

The mud, sand, and boulders thus let fall in still water must be exactly like the moraines of terrestrial glaciers, devoid of stratification and organic remains. But occasionally, on the outer side of such packs of stranded bergs, the waves and currents may cause the detached earthy and stony materials to be sorted according to size and weight before they reach the bottom, and to acquire a stratified arrangement.

I have already alluded to the large quantity of ice, containing great blocks of stone, which is sometimes seen floating far from land, in the southern or Antarctic seas. After the emergence, therefore, of such a submarine area, the superficial detritus will have no necessary relation to the hills, valleys, and river-plains over which it will be scattered. Many a water-shed may intervene between the starting-point of each erratic or pebble and its final resting- place, and the only means of discovering the country from which it took its departure will consist in a careful comparison of its mineral or fossil contents with those of the parent rocks.

CHAPTER XII.

POST-PLIOCENE PERIOD, CONTINUED.— GLACIAL CONDITIONS, CONCLUDED.

Glaciation of Scandinavia and Russia. Glaciation of Scotland. Mammoth in Scotch Till. Marine Shells in Scotch Glacial Drift. Their Arctic Character. Rarity of Organic Remains in Glacial Deposits. Contorted Strata in Drift. Glaciation of Wales, England, and Ireland. Marine Shells of Moel Tryfaen. Erratics near Chichester. Glacial Formations of North America. Many Species of Testacea and Quadrupeds survived the Glacial Cold. Connection of the Predominance of Lakes with Glacial Action. Action of Ice in preventing the silting up of Lake-basins. Absence of Lakes in the Caucasus. Equatorial Lakes of Africa.

GLACIATION OF SCANDINAVIA AND RUSSIA.

In large tracts of Norway and Sweden, where there have been no glaciers in historical times, the signs of ice-action have been traced as high as 6000 feet above the level of the sea. These signs consist chiefly of polished and furrowed rock-surfaces, of moraines and erratic blocks. The direction of the erratics, like that of the furrows, has usually been conformable to the course of the principal valleys; but the lines of both sometimes radiate outward in all directions from the highest land, in a manner which is only explicable by the hypothesis above alluded to of a general envelope of continental ice, like that of Greenland (Chapter 11.) Some of the far-transported blocks have been carried from the central parts of Scandinavia towards the Polar regions; others southward to Denmark; some south-westward, to the coast of Norfolk in England; others south-eastward, to Germany, Poland, and Russia.

In the immediate neighbourhood of Upsala, in Sweden, I had observed, in 1834, a ridge of stratified sand and gravel, in the midst of which occurs a layer of marl, evidently formed originally at the bottom of the Baltic, by the slow growth of the mussel, cockle, and other marine shells of living species, intermixed with some proper to fresh water. The marine shells are all of dwarfish size, like those now inhabiting the brackish waters of the Baltic; and the marl, in which many of them are imbedded, is now raised more than 100 feet above the level of the Gulf of Bothnia. Upon the top of this ridge repose several huge erratics, consisting of gneiss for the most part unrounded, from nine to sixteen feet in diameter, and which must have been brought into their present position since the time when the neighbouring gulf was already characterised by its peculiar fauna. Here, therefore, we have proof that the transport of erratics continued to take place, not merely when the sea was inhabited by the existing testacea, but when the north of Europe had already assumed that remarkable feature of its physical geography which separates the Baltic from the North Sea, and causes the Gulf of Bothnia to have only one- fourth of the saltness belonging to the ocean. In Denmark, also, recent shells have been found in stratified beds, closely associated with the boulder clay.

GLACIATION OF SCOTLAND.

Mr. T.F. Jamieson, in 1858, adduced a great body of facts to prove that the Grampians once sent down glaciers from the central regions in all directions towards the sea. "The glacial grooves," he observed, "radiate outward from the central heights towards all points of the compass, though they do not always strictly conform to the actual shape and contour of the minor valleys and ridges."

These facts and other characteristics of the Scotch drift lead us to the inference that when the glacial cold first set in, Scotland stood higher above the sea than at present, and was covered for the most part with snow and ice, as Greenland is now. This sheet of land-ice sliding down to lower levels, ground down and polished the subjacent rocks, sweeping off nearly all superficial deposits of older date, and leaving only till and boulders in their place. To this continental state succeeded a period of depression and partial submergence. The sea advanced over the lower lands, and Scotland was converted into an archipelago, some marine sand with shells being spread over the bottom of the sea. On this sand a great mass of boulder clay usually quite devoid of fossils was accumulated. Lastly, the land re-emerged from the water, and, reaching a level somewhat above its present height, became connected with the continent of Europe, glaciers being formed once more in the higher regions, though the ice probably never regained its former extension. (Jamieson Quarterly Geological Journal 1860 volume 16 page 370.) After all these changes, there were some minor oscillations in the level of the land, on which, although they have had important geographical consequences, separating Ireland from England, for example, and England from the Continent, we need not here enlarge.

MAMMOTH IN SCOTCH TILL.

Almost all remains of the terrestrial fauna of the Continent which preceded the period of submergence have been lost; but a few patches of estuarine and fresh- water formations escaped denudation by submergence. To these belong the peaty clay from which several mammoths' tusks and horns of reindeer were obtained at Kilmaurs, in Ayrshire as long ago as 1816. Mr. Bryce in 1865 ascertained that the fresh-water formation containing these fossils rests on carboniferous sandstone, and is covered, first by a bed of marine sand with arctic shells, and then with a great mass of till with glaciated boulders. (Bryce Quarterly Geological Journal volume 21 page 217 1865.) Still more recent explorations in the neighbourhood of Kilmaurs have shown that the fresh-water formation contains the seed of the pond-weed Potamogeton and the aquatic Ranunculus; and Mr. Young of the Glasgow Museum washed the mud adhering to the reindeer horns of Kilmaurs and that which filled the cracks of the associated elephants' tusks, and detected in these fossils (which had been in the Glasgow Museum for half a century) abundance of the same seeds.

All doubts, therefore, as to the true position of the remains of the mammoth, a fossil so rare in Scotland, have been set at rest, and it serves to prove that part of the ancient continent sank beneath the sea at a period of great cold, as the shells of the overlying sand attest. The incumbent till or boulder clay is about 40 feet thick, but it often attains much greater thickness in the same part of Scotland.

MARINE SHELLS OF SCOTCH DRIFT.

(FIGURE 107. Astarte borealis, Chem.; (A. arctica, Moll. A. compressa, Mont.)

(FIGURE 108. Leda lanceolata (oblonga), Sowerby.)

(FIGURE 109. Saxicava rugosa, Penn.)

(FIGURE 110. Pecten islandicus, Moll. Northern shell common in the drift of the Clyde, in Scotland. )

(FIGURE 111. Natica clausa, Bred. Northern shell common in the drift of the Clyde, in Scotland.)

(FIGURE 112. Trophon clathratum, Linne. Northern shell common in the drift of the Clyde, in Scotland.)

(FIGURE 113. Leda truncata. a. Exterior of left valve. b. Interior of same.)

(FIGURE 114. Tellina calcarea, Chem. (Tellina proxima, Brown.) a. Outside of left valve. b. Interior of same.)

The greatest height to which marine shells have yet been traced in this boulder clay is at Airdie, in Lanarkshire, ten miles east of Glasgow, 524 feet above the level of the sea. At that spot they were found imbedded in stratified clays with till above and below them. There appears no doubt that the overlying deposit was true glacial till, as some boulders of granite were observed in it, which must have come from distances of sixty miles at the least.

The shells figured in Figures 107 to 112 are only a few out of a large assemblage of living species, which, taken as a whole, bear testimony to conditions far more arctic than those now prevailing in the Scottish seas. But a group of marine shells, indicating a still greater excess of cold, has been brought to light since 1860 by the Reverend Thomas Brown, from glacial drift or clay on the borders of the estuaries of the Forth and Tay. This clay occurs at Elie, in Fife, and at Errol, in Perthshire; and has already afforded about 35 shells, all of living species, and now inhabitants of arctic regions, such as Leda truncata, Tellina proxima (see Figures 113 and 114), Pecten Groenlandicus, Crenella laevigata, Crenella nigra, and others, some of them first brought by Captain Sir E. Parry from the coast of Melville Island, latitude 76 degrees north. These were all identified in 1863 by Dr. Torell, who had just returned from a survey of the seas around Spitzbergen, where he had collected no less than 150 species of mollusca, living chiefly on a bottom of fine mud derived from the moraines of melting glaciers which there protrude into the sea. He informed me that the fossil fauna of this Scotch glacial deposit exhibits not only the species but also the peculiar varieties of mollusca now characteristic of very high latitudes. Their large size implies that they formerly enjoyed a colder, or, what was to them a more genial climate, than that now prevailing in the latitude where the fossils occur. Marine shells have also been found in the glacial drift of Caithness and Aberdeenshire at heights of 250 feet, and in Banff of 350 feet, and stratified drift continuous with the above ascends to heights of 500 feet. Already 75 species are enumerated from Caithness, and the same number from Aberdeenshire and Banff, and in both cases all but six are arctic species.

I formerly suggested that the absence of all signs of organic life in the Scotch drift might be connected with the severity of the cold, and also in some places with the depth of the sea during the period of extreme submergence; but my faith in such an hypothesis has been shaken by modern investigations, an exuberance of life having been observed both in arctic and antarctic seas of great depth, and where floating ice abounds. The difficulty, moreover, of accounting for the entire dearth of marine shells in till is removed when once we have adopted the theory of this boulder clay being the product of land-ice. For glaciers coming down from a continental ice-sheet like that which covers Greenland may fill friths many hundred feet below the sea-level, and even invade parts of a bay a thousand feet deep, before they find water enough to float off their terminal portions in the form of icebergs. In such a case till without marine shells may first accumulate, and then, if the climate becomes warmer and the ice melts, a marine deposit may be superimposed on the till without any change of level being required.

Another curious phenomenon bearing on this subject was styled by the late Hugh Miller the "striated pavements" of the boulder clay. Where portions of the till have been removed by the sea on the shores of the Forth, or in the interior by railway cuttings, the boulders imbedded in what remains of the drift are seen to have been all subjected to a process of abrasion and striation, the striae and furrows being parallel and persistent across them all, exactly as if a glacier or iceberg had passed over them and scored them in a manner similar to that so often undergone by the solid rocks below the glacial drift. It is possible, as Mr. Geikie conjectures, that this second striation of the boulders may be referable to floating ice. (Geikie Transactions of the Geological Society of Glasgow volume 1 part 2 page 68 1863.)

CONTORTED STRATA IN DRIFT.

(FIGURE 115. Section of contorted drift overlying till, seen on left bank of South Esk, near Cortachie, in 1840. Height of section seen, from a to d, about 50 feet. a, b. Gravel and sand. f, g. Contorted drift. Till.)

In Scotland the till is often covered with stratified gravel, sand, and clay, the beds of which are sometimes horizontal and sometimes contorted for a thickness of several feet. Such contortions are not uncommon in Forfarshire, where I observed them, among other places, in a vertical cutting made in 1840 near the left bank of the South Esk, east of the bridge of Cortachie. The convolutions of the beds of fine and coarse sand, gravel, and loam, extend through a thickness of no less than 25 feet vertical, or from b to c, Figure 115, the horizontal stratification being resumed very abruptly at a short distance, as to the right of f, g. The overlying coarse gravel and sand, a, is in some places horizontal, in others it exhibits cross bedding, and does not partake of the disturbances which the strata b, c, have undergone. The underlying till is exposed for a depth of about 20 feet; and we may infer from sections in the neighbourhood that it is considerably thicker.

In some cases I have seen fragments of stratified clays and sands, bent in like manner, in the middle of a great mass of till. Mr. Trimmer has suggested, in explanation of such phenomena, the intercalation in the glacial period of large irregular masses of snow or ice between layers of sand and gravel. Some of the cliffs near Behring's Straits, in which the remains of elephants occur, consist of ice mixed with mud and stones; and Middendorf describes the occurrence in Siberia of masses of ice, found at various depths from the surface after digging through drift. Whenever the intercalation of snow and ice with drift, whether stratified or unstratified, has taken place, the melting of the ice will cause such a failure of support as may give rise to flexures, and sometimes to the most complicated foldings. But in many cases the strata may have been bent and deranged by the mechanical pressure of an advancing glacier, or by the sideway thrust of huge islands of ice running aground against sandbanks; in which case, the position of the beds forming the foundation of the banks may not be at all disturbed by the shock.

There are indeed many signs in Scotland of the action of floating ice, as might have been expected where proofs of submergence in the Glacial Period are not wanting. Among these are the occurrence of large erratic blocks, frequently in clusters at or near the tops of hills or ridges, places which may have formed islets or shallows in the sea where floating ice would mostly ground and discharge its cargo on melting. Glaciers or land-ice would, on the contrary, chiefly discharge their cargoes at the bottom of valleys. Traces of an earlier and independent glaciation have also been observed in some regions where the striation, apparently produced by ice proceeding from the north-west, is not explicable by the radiation of land-ice from a central mountainous region. (Milne Home Transactions of the Royal Society Edinburgh volume 25 1868-9.)

GLACIATION OF WALES AND ENGLAND.

The mountains of North Wales were recognised, in 1842, by Dr. Buckland, as having been an independent centre of the dispersion of erratics— great glaciers, long since extinct, having radiated from the Snowdonian heights in Carnarvonshire, through seven principal valleys towards as many points of the compass, carrying with them large stony fragments, and grooving the subjacent rocks in as many directions.

Besides this evidence of land-glaciers, Mr. Trimmer had previously, in 1831, detected the signs of a great submergence in Wales in the Post-pliocene period. He had observed stratified drift, from which he obtained about a dozen species of marine shells, near the summit of Moel Tryfaen, a hill 1400 feet high, on the south side of the Menai Straits. I had an opportunity of examining in the summer of 1863, together with the Reverend W.S. Symonds, a long and deep cutting made through this drift by the Alexandra Mining Company in search of slates. At the top of the hill above-mentioned we saw a stratified mass of incoherent sand and gravel 35 feet thick, from which no less than 54 species of mollusca, besides three characteristic arctic varieties— in all 57 forms— have been obtained by Mr. Darbishire. They belong without exception to species still living in British or more northern seas; eleven of them being exclusively arctic, four common to the arctic and British seas, and a large proportion of the remainder having a northward range, or, if found at all in the southern seas of Britain, being comparatively less abundant. In the lowest beds of the drift were large heavy boulders of far-transported rocks, glacially polished and scratched on more than one side. Underneath the whole we saw the edges of vertical slates exposed to view, which here, like the rocks in other parts of Wales, both at greater and less elevations, exhibit beneath the drift unequivocal marks of prolonged glaciation. The whole deposit has much the appearance of an accumulation in shallow water or on a beach, and it probably acquired its thickness during the gradual subsidence of the coast— an hypothesis which would require us to ascribe to it a high antiquity, since we must allow time, first for its sinking, and then for its re-elevation.

The height reached by these fossil shells on Moel Tryfaen is no less than 1300 feet— a most important fact when we consider how very few instances we have on record beyond the limits of Wales, whether in Europe or North America, of marine shells having been found in glacial drift at half the height above indicated. A marine molluscous fauna, however, agreeing in character with that of Moel Tryfaen, and comprising as many species, has been found in drift at Macclesfield and other places in central England, sometimes reaching an elevation of 1200 feet.

Professor Ramsay estimated the probable amount of submergence during some part of the glacial period at about 2300 feet; for he was unable to distinguish the superficial sands and gravel which reached that high elevation from the drift which, at Moel Tryfaen and at lower points, contains shells of living species. The evidence of the marine origin of the highest drift is no doubt inconclusive in the absence of shells, so great is the resemblance of the gravel and sand of a sea beach and of a river's bed, when organic remains are wanting; but, on the other hand, when we consider the general rarity of shells in drift which we know to be of marine origin, we can not suppose that, in the shelly sands of Moel Tryfaen, we have hit upon the exact uppermost limit of marine deposition, or, in other words, a precise measure of the submergence of the land beneath the sea since the glacial period.

We are gradually obtaining proofs of the larger part of England, north of a line drawn from the mouth of the Thames to the Bristol Channel, having been under the sea and traversed by floating ice since the commencement of the glacial epoch. Among recent observations illustrative of this point, I may allude to the discovery, by Mr. J.F. Bateman, near Blackpool, in Lancashire, fifty miles from the sea, and at the height of 568 feet above its level, of till containing rounded and angular stones and marine shells, such as Turritella communis, Purpura lapillus, Cardium edule, and others, among which Trophon clathratum (=Fusus Bamffius), though still surviving in North British seas, indicates a cold climate.

ERRATICS NEAR CHICHESTER.

The most southern memorials of ice-action and of a Post-pliocene fauna in Great Britain is on the coast of the county of Sussex, about 25 miles west of Brighton, and 15 south of Chichester. A marine deposit exposed between high and low tide occurs on both sides of the promontory called Selsea Bill, in which Mr. Godwin-Austen found thirty-eight species of shells, and the number has since been raised to seventy.

This assemblage is interesting because on the whole, while all the species are recent, they have a somewhat more southern aspect than those of the present British Channel. It is true that about forty of them range from British to high northern latitudes; but several of them, as, for example, Lutraria rugosa and Pecten polymorphous, which are abundant, are not known at present to range farther north than the coast of Portugal, and seem to indicate a warmer temperature than now prevails on the coast where we find them fossil. What renders this curious is the fact that the sandy loam in which they occur is overlaid by yellow clayey gravel with large erratic blocks which must have been drifted into their present position by ice when the climate had become much colder. These transported fragments of granite, syenite, and greenstone, as well as of Devonian and Silurian rocks, may have come from the coast of Normandy and Brittany, and are many of them of such large size that we must suppose them to have been drifted into their present site by coast-ice. I measured one of granite, at Pagham, 21 feet in circumference. In the gravel of this drift with erratics are a few littoral shells of living species, indicating an ancient coast-line.

GLACIAL FORMATIONS OF NORTH AMERICA.

In the western hemisphere, both in Canada and as far south as the 40th and even 38th parallel of latitude in the United States, we meet with a repetition of all the peculiarities which distinguish the European boulder formation. Fragments of rock have travelled for great distances, especially from north to south: the surface of the subjacent rock is smoothed, striated, and fluted; unstratified mud or TILL containing boulders is associated with strata of loam, sand, and clay, usually devoid of fossils. Where shells are present, they are of species still living in northern seas, and not a few of them identical with those belonging to European drift, including most of those already given in Figures 107 to 112. The fauna also of the glacial epoch in North America is less rich in species than that now inhabiting the adjacent sea, whether in the Gulf of St. Lawrence, or off the shores of Maine, or in the Bay of Massachusetts.

The extension on the American continent of the range of erratics during the Post-pliocene period to lower latitudes than they reached in Europe, agrees well with the present southward deflection of the isothermal lines, or rather the lines of equal winter temperature. It seems that formerly, as now, a more extreme climate and a more abundant supply of ice prevailed on the western side of the Atlantic. Another resemblance between the distribution of the drift fossils in Europe and North America has yet to be pointed out. In Canada and the United States, as in Europe, the marine shells are generally confined to very moderate elevations above the sea (between 100 and 700 feet), while the erratic blocks and the grooved and polished surfaces of rock extend to elevations of several thousand feet.

I have already mentioned that in Europe several quadrupeds of living, as well as extinct, species were common to pre-glacial and post-glacial times. In like manner there is reason to suppose that in North America much of the ancient mammalian fauna, together with nearly all the invertebrata, lived through the ages of intense cold. That in the United States the Mastodon giganteus was very abundant after the drift period, is evident from the fact that entire skeletons of this animal are met with in bogs and lacustrine deposits occupying hollows in the glacial drift. They sometimes occur in the bottom even of small ponds recently drained by the agriculturist for the sake of the shell-marl. In 1845 no less than six skeletons of the same species of Mastodon were found in Warren county, New Jersey, six feet below the surface, by a farmer who was digging out the rich mud from a small pond which he had drained. Five of these skeletons were lying together, and a large part of the bones crumbled to pieces as soon as they were exposed to the air.

It would be rash, however, to infer from such data that these quadrupeds were mired in MODERN times, unless we use that term strictly in a geological sense. I have shown that there is a fluviatile deposit in the valley of the Niagara, containing shells of the genera Melania, Lymnea, Planorbis, Velvata, Cyclaz, Unio, Helix, etc., all of recent species, from which the bones of the great Mastodon have been taken in a very perfect state. Yet the whole excavation of the ravine, for many miles below the Falls, has been slowly effected since that fluviatile deposit was thrown down. Other extinct animals accompany the Mastodon giganteus in the post-glacial deposits of the United States, and this, taken with the fact that so few of the mollusca, even of the commencement of the cold period, differ from species now living is important, as refuting the hypothesis, for which some have contended, that the intensity of the glacial cold annihilated all the species in temperate and arctic latitudes.

CONNECTION OF THE PREDOMINANCE OF LAKES WITH GLACIAL ACTION.

It was first pointed out by Professor Ramsay in 1862, that lakes are exceedingly numerous in those countries where erratics, striated blocks, and other signs of ice-action abound; and that they are comparatively rare in tropical and sub- tropical regions. Generally in countries where the winter cold is intense, such as Canada, Scandinavia, and Finland, even the plains and lowlands are thickly strewn with innumerable ponds and small lakes, together with some others of a larger size; while in more temperate regions, such as Great Britain, Central and Southern Europe, the United States, and New Zealand, lake districts occur in all such mountainous tracts as can be proved to have been glaciated in times comparatively modern or since the geographical configuration of the surface bore a considerable resemblance to that now prevailing. In the same countries, beyond the glaciated regions, lakes abruptly cease, and in warmer and tropical countries are either entirely absent, or consist, as in equatorial Africa, of large sheets of water unaccompanied so far as we yet know by numerous smaller ponds and tarns.

The southern limits of the lake districts of the Northern Hemisphere are found at about 40 degrees N. latitude on the American continent, and about 50 degrees in Europe, or where the Alps intervene four degrees farther south. A large proportion of the smaller lakes are dammed up by barriers of unstratified drift, having the exact character of the moraines of glaciers, and are termed by geologists "morainic," but some of them are true rock-basins, and would hold water even if all the loose drift now resting on their margins were removed.

In a paper read before the Geological Society of London in 1862, Professor Ramsay maintained that the first formation of most existing lakes took place during the glacial epoch, and was due, not to elevation or subsidence, but to actual erosion of their basins by glaciers. M. Mortillet in the same year advanced the theory that after the Alpine lake-basins had been filled up with loose fluviatile deposits, they were re-excavated by the great glaciers which passed down the valleys at the time of the greatest cold, a doctrine which would attribute to moving ice almost as great a capacity of erosion as that which assumed that the original basins were scooped out of solid rock by glaciers. It is impossible to deny that the mere geographical distribution of lakes points to the intimate connection of their origin with the abundance of ice during a former excess of cold, but how far the erosive action of moving ice has been the sole or even the principal cause of lake-basins, is a question still open to discussion.

The lakes of Switzerland and the north of Italy are some of them twenty and thirty miles in length, and so deep that their bottoms are in some cases from 1000 to 2000 feet beneath the level of the sea. It is admitted on all hands that they were once filled with ice, and as the existing glaciers polish and grind down, as before stated, the surface of the rocks, we are prepared to find that every lake-basin in countries once covered by ice should bear the marks of superficial glaciation, and also that the ice during its advance and retreat should have left behind it much transported matter as well as some evidence of its having enlarged the pre-existing cavity. But much more than this is demanded by the advocates of glacial erosion. They suggest that as the old extinct glaciers were several thousand feet thick, they were able in some places gradually to scoop out of the solid rock cavities twenty or thirty miles in length, and as in the case of Lago Maggiore from a thousand to two thousand six hundred feet below the previous level of the river-channel, and also that the ice had the power to remove from the cavity formed by its grinding action all the materials of the missing rocks. A constant supply, it is argued, of fine mud issues from the termination of every glacier in the stream which is produced by the melting of the ice, and this result of friction is exhibited both during winter and summer, affording evidence of the continual deepening and widening of the valleys through which glaciers pass. As the fine mud is carried away by a river from the deep pool which is formed from the base of every cataract, so it seems to be imagined that lake-basins may be gradually emptied of the mud formed by abrasion during the glacial period.

I am by no means disposed to object to this theory on the ground of the insufficiency of the time during which the extreme cold endured, but we must carefully consider whether that same time is not so vast as to make it probable that other forces, besides the motion of glaciers, must have cooperated in converting some parts of the ancient valley courses into lake-basins. They who have formed the most exalted conceptions of the erosive energy of moving ice do not deny that during the period termed "Glacial" there have been movements of the earth's crust sufficient to produce oscillations of level in Europe amounting to 1000 feet or more in both directions. M. Charpentier, indeed, attributed some of the principal changes of climate in Switzerland, during the glacial period, to a depression of the central Alps to the extent of 3000 feet, and Swiss geologists have long been accustomed to attribute their lake basins, in part, to those convulsions by which the shape and course of the valleys may have been modified. Our experience, in the lifetime of the present generation, of the changes of level witnessed in New Zealand during great earthquakes is entirely opposed to the notion that the movements, whether upward or downward, are uniform in amount or direction throughout areas of indefinite extent. On the contrary, the land has been permanently raised in one region several feet or yards, and the rise has been found gradually to die out, so as to be imperceptible at a distance of twenty miles, and in some areas is even exchanged for a simultaneous downward movement of several feet.

But, it is asked, if such inequality of movement can have contributed towards the production of lake basins, does it not leave unexplained the comparative rarity of lakes in tropical and subtropical countries. In reply to this question it may be observed that in our endeavour to estimate the effects of subterranean movements in modifying the superficial geography of a country we must remember that each convulsion effects a very slight change. If it interferes with the drainage, whether by raising the lower or sinking the higher portion of a hydrographical basin, the upheaval or depression will only amount to a few feet at a time, and there may be an interval of years or centuries before any further movement takes place in the same region. In the mean time an incipient lake if produced may be filled up with sediment, and the recently-formed barrier will then be cut through by the river, whereas in a country where glacial conditions prevail no such obliteration of the temporary lake-basin would take place; for however deep it became by repeated sinking of the upper or rising of the lower extremity, being always filled with ice it might remain, throughout the greater part of its extent, free from sediment or drift until the ice melted at the close of the glacial period.

One of the most serious objections to the exclusive origin by ice-erosion of wide and deep lake-basins arises from their capricious distribution, as for example in Piedmont, both to the eastward and westward of Turin, where great lakes are wanting (Antiquity of Man page 313.), although some of the largest extinct glaciers descending from Mont Blanc and Monte Rosa came down from the Alps, leaving their gigantic moraines in the low country. Here, therefore, we might have expected to find lakes of the first magnitude rivalling the contiguous Lago Maggiore in importance.

A still more striking illustration of the same absence of lakes where large glaciers abound is afforded by the Caucasus, a chain more than 300 miles long, and the loftiest peaks of which attain heights from 16,000 to 18,000 feet. This greatest altitude is reached by Elbruz, a mountain in latitude 43 degrees north three degrees south of Mont Blanc, but on the other hand 3000 feet higher. The present Caucasian glaciers are equal or superior in dimensions to those of Switzerland, and like them give rise occasionally to temporary lakes by obstructing the course of rivers, and causing great floods when the icy barriers give way. Mr. Freshfield, a careful observer, writing in 1869, says: "A total absence of lakes on both sides of the chains is the most marked feature. Not only are there no great subalpine sheets of water, like Como or Geneva, but mountain tarns, such as the Dauben See on the Gemmi, or the Klonthal See near Glarus, are equally wanting." (Travels in Central Caucasus 1869 page 452.) The same author states on the authority of the eminent Swiss geologist, Mons. E. Favre, who also explored the Caucasus in 1868, that moraines of great height and huge erratics of granite and other rocks "justify the assertion that the present glaciers of the Caucasus, like those of the Alps, are only the shadows of their former selves."

It seems safe to assume that the chain of lakes, of which the Albert Nyanza forms one in equatorial Africa, was due to causes other than glacial. Yet if we could imagine a glacial period to visit that region filling the lakes with ice and scoring the rocks which form their sides and bottoms, we should be unable to decide how much the capacity of the basins had been enlarged and the surface modified by glacial erosion. The same may be true of the Lago Maggiore and Lake Superior, although the present basins of both of them afford abundant superficial markings due to ice-action.

But to whatever combination of causes we attribute the great Alpine lakes one thing is clear, namely, that they are, geologically speaking, of modern origin. Every one must admit that the upper valley of the Rhone has been chiefly caused by fluviatile denudation, and it is obvious that the quantity of matter removed from that valley previous to the glacial period would have been amply sufficient to fill up with sediment the basin of the Lake of Geneva, supposing it to have been in existence, even if its capacity had been many times greater than it is now. (See Principles volume 1 page 420 10th edition 1867.)

On the whole, it appears to me, in accordance with the views of Professor Ramsay, M. Mortillet, Mr. Geikie, and others, that the abrading action of ice has formed some mountain tarns and many morainic lakes; but when it is a question of the origin of larger and deeper lakes, like those of Switzerland or the north of Italy, or inland fresh-water seas, like those of Canada, it will probably be found that ice has played a subordinate part in comparison with those movements by which changes of level in the earth's crust are gradually brought about.

TERTIARY OR CAINOZOIC PERIOD.

CHAPTER XIII.

PLIOCENE PERIOD.

Glacial Formations of Pliocene Age. Bridlington Beds. Glacial Drifts of Ireland. Drift of Norfolk Cliffs. Cromer Forest-bed. Aldeby and Chillesford Beds. Norwich Crag. Older Pliocene Strata. Red Crag of Suffolk. Coprolitic Bed of Red Crag. White or Coralline Crag. Relative Age, Origin, and Climate of the Crag Deposits. Antwerp Crag. Newer Pliocene Strata of Sicily. Newer Pliocene Strata of the Upper Val d'Arno. Older Pliocene of Italy. Subapennine Strata. Older Pliocene Flora of Italy.

It will be seen in the description given in the last chapter of the Post- pliocene formations of the British Isles that they comprise a large proportion of those commonly termed glacial, characterised by shells which, although referable to living species, usually indicate a colder climate than that now belonging to the latitudes where they occur fossil. But in parts of England, more especially in Yorkshire, Norfolk, and Suffolk, there are superficial formations of clay with glaciated boulders, and of sand and pebbles, containing occasional, though rare, patches of shells, in which the marine fauna begins to depart from that now inhabiting the neighbouring sea, and comprises some species of mollusca not yet known as living, as well as extinct varieties of others, entitling us to class them as Newer Pliocene, although belonging to the close of that period and chronologically on the verge of the later or Post-pliocene epoch.

BRIDLINGTON DRIFT.

To this era belongs the well-known locality of Bridlington, near the mouth of the Humber, in Yorkshire, where about seventy species or well-marked varieties of shells have been found on the coast, near the sea-level, in a bed of sand several feet thick resting on glacial clay with much chalk debris, and covered by a deposit of purple clay with glaciated boulders. More than a third of the species in this drift are now inhabitants of arctic regions, none of them extending southward to the British seas; which is the more remarkable as Bridlington is situated in latitude 54 degrees north. Fifteen species are British and Arctic, a very few belong to those species which range south of our British seas. Five species or well-marked varieties are not known living, namely, the variety of Astarte borealis (called A. Withami); A. mutabilis; the sinistral form of Tritonium carinatum, Cardita analis, and Tellina obliqua, Figure 120. Mr. Searles Wood also inclines to consider Nucula Cobboldiae, Figure 119, now absent from the European seas and the Atlantic, as specifically distinct from a closely-allied shell now living in the seas surrounding Vancouver's Island, which some conchologists regard as a variety. Tellina obliqua also approaches very near to a shell now living in Japan.

GLACIAL DRIFT OF IRELAND.

Marine drift containing the last-mentioned Nucula and other glacial shells reaches a height of from 1000 to 1200 feet in the county of Wexford, south of Dublin. More than eighty species have already been obtained from this formation, of which two, Conovulus pyramidalis and Nassa monensis, are not known as living; while Turritella incrassata and Cypraea lucida no longer inhabit the British seas, but occur in the Mediterranean. The great elevation of these shells, and the still greater height to which the surface of the rocks in the mountainous regions of Ireland have been smoothed and striated by ice-action, has led geologists to the opinion that that island, like the greater part of England and Scotland, after having been united with the continent of Europe, from whence it received the plants and animals now inhabiting it, was in great part submerged. The conversion of this and other parts of Great Britain into an archipelago was followed by a re-elevation of land and a second continental period. After all these changes the final separation of Ireland from Great Britain took place, and this event has been supposed to have preceded the opening of the straits of Dover. (See Antiquity of Man chapter 14.)

DRIFT OF NORFOLK CLIFFS.

(FIGURE 116. Tellina balthica (T. solidula).)

There are deposits of boulder clay and till in the Norfolk cliffs principally made up of the waste of white chalk and flints which, in the opinion of Mr. Searles Wood, jun., and others, are older than the Bridlington drift, and contain a larger proportion of shells common to the Norwich and Red Crag, including a certain number of extinct forms, but also abounding in Tellina balthica (T. solidula, Figure 116), which is found fossil at Bridlington, and living in our British seas, but wanting in all the formations, even the newest, afterwards to be described as Crag. As the greater part of these drifts are barren of organic remains, their classification is at present a matter of great uncertainty.

They can nowhere be so advantageously studied as on the coast between Happisburgh and Cromer. Here we may see vertical cliffs, sometimes 300 feet and more in height, exposed for a distance of fifty miles, at the base of which the chalk with flints crops out in nearly horizontal strata. Beds of gravel and sand repose on this undisturbed chalk. They are often strangely contorted, and envelop huge masses or erratics of chalk with layers of vertical flint. I measured one of these fragments in 1839 at Sherringham, and found it to be eighty feet in its longest diameter. It has been since entirely removed by the waves of the sea. In the floor of the chalk beneath it the layers of flint were horizontal. Such erratics have evidently been moved bodily from their original site, probably by the same glacial action which has polished and striated some of the accompanying granitic and other boulders, occasionally six feet in diameter, which are imbedded in the drift.

CROMER FOREST-BED.

Intervening between these glacial formations and the subjacent chalk lies what has been called the Cromer Forest-bed. This buried forest has been traced from Cromer to near Kessingland, a distance of more than forty miles, being exposed at certain seasons between high and low water mark. It is the remains of an old land and estuarine deposit, containing the submerged stumps of trees standing erect with their roots in the ancient soil. Associated with the stumps and overlying them, are lignite beds with fresh-water shells of recent species, and laminated clay without fossils. Through the lignite and forest-bed are scattered cones of the Scotch and spruce firs with the seeds of recent plants, and the bones of at least twenty species of terrestrial mammalia. Among these are two species of elephant, E. meridionalis, Nesti, and E. antiquus, the former found in the Newer Pliocene beds of the Val d'Arno, near Florence. In the same bed occur Hippopotamus major, Rhinoceros etruscus, both of them also Val d'Arno species, many species of deer considered by Mr. Boyd Dawkins to be characteristic of warmer countries, and also a horse, beaver, and field-mouse. Half of these mammalia are extinct, and the rest still survive in Europe. The vegetation taken alone does not imply a temperature higher than that now prevailing in the British Isles. There must have been a subsidence of the forest to the amount of 400 or 500 feet, and a re-elevation of the same to an equal extent in order to allow the ancient surface of the chalk or covering of soil, on which the forest grew, to be first covered with several hundred feet of drift, and then upheaved so that the trees should reach their present level. Although the relative antiquity of the forest-bed to the overlying glacial till is clear, there is some difference of opinion as to its relation to the crag presently to be described.

CHILLESFORD AND ALDEBY BEDS.

(FIGURE 117. Natica helicoides, Johnson.)

It is in the counties of Norfolk, Suffolk, and Essex, that we obtain our most valuable information respecting the British Pliocene strata, whether newer or older. They have obtained in those counties the provincial name of "Crag," applied particularly to masses of shelly sand which have long been used in agriculture to fertilise soils deficient in calcareous matter. At Chillesford, between Woodbridge and Aldborough in Suffolk, and Aldeby, near Beccles, in the same county, there occur stratified deposits, apparently older than any of the preceding drifts of Yorkshire, Norfolk, and Suffolk. They are composed at Chillesford of yellow sands and clays, with much mica, forming horizontal beds about twenty feet thick. Messrs. Prestwich and Searles Wood, senior, who first described these beds, point out that the shells indicate on the whole a colder climate than the Red Crag; two-thirds of them being characteristic of high latitudes. Among these are Cardium Groenlandicum, Leda limatula, Tritonium carinatum, and Scalaria Groenlandica. In the upper part of the laminated clays a skeleton of a whale was found associated with casts of the characteristic shells, Nucula Cobboldiae and Tellina obliqua, already referred to as no longer inhabiting our seas, and as being extinct varieties if not species. The same shells occur in a perfect state in the lower part of the formation. Natica helicoides (Figure 117) is an example of a species formerly known only as fossil, but which has now been found living in our seas.

At Aldeby, where beds occur decidedly similar in mineral character as well as fossil remains, Messrs. Crowfoot and Dowson have now obtained sixty-six species of mollusca, comprising the Chillesford species and some others. Of these about nine-tenths are recent. They are in a perfect state, clearly indicating a cold climate; as two-thirds of them are now met with in arctic regions. As a rule, the lamellibranchiate molluscs have both valves united, and many of them, such as Mya arenaria, stand with the siphonal end upward, as when in a living state. Tellina balthica, before mentioned (Figure 116) as so characteristic of the glacial beds, including the drift of Bridlington, has not yet been found in deposits of Chillesford and Aldeby age, whether at Sudbourn, East Bavent, Horstead, Coltishall, Burgh, or in the highest beds overlying the Norwich Crag proper at Bramerton and Thorpe.

NORWICH OR FLUVIO-MARINE CRAG.

(FIGURE 118. Mastodon arvernensis, third milk molar, left side, upper jaw; grinding surface, natural size. Norwich Crag, Postwick, also found in Red Crag, see below.)

The beds above alluded to ought, perhaps, to be regarded as beds of passage between the glacial formations and those called from a provincial name "Crag," the newest member of which has been commonly called the "Norwich Crag." It is chiefly seen in the neighbourhood of Norwich, and consists of beds of incoherent sand, loam, and gravel, which are exposed to view on both banks of the Yare, as at Bramerton and Thorpe. As they contain a mixture of marine, land, and fresh- water shells, with bones of fish and mammalia, it is clear that these beds have been accumulated at the bottom of a sea near the mouth of a river. They form patches rarely exceeding twenty feet in thickness, resting on white chalk. At their junction with the chalk there invariably intervenes a bed called the "Stone-bed," composed of unrolled chalk-flints, commonly of large size, mingled with the remains of a land fauna comprising Mastodon arvernensis, Elephas meridionalis, and an extinct species of deer. The mastodon, which is a species characteristic of the Pliocene strata of Italy and France, is the most abundant fossil, and one not found in the Cromer forest before mentioned. When these flints, probably long exposed in the atmosphere, became submerged, they were covered with barnacles, and the surface of the chalk became perforated by the Pholas crispata, each fossil shell still remaining at the bottom of its cylindrical cavity, now filled up with loose sand from the incumbent crag. This species of Pholas still exists, and drills the rocks between high and low water on the British coast. The name of "Fluvio-marine" has often been given to this formation, as no less than twenty species of land and fresh-water shells have been found in it. They are all of living species; at least only one univalve, Paludina lenta, has any, and that a very doubtful, claim to be regarded as extinct.

(FIGURE 119. Nucula Cobboldiae.)

(FIGURE 120. Tellina obliqua.)

Of the marine shells, 124 in number, about 18 per cent are extinct, according to the latest estimate given me by Mr. Searles Wood; but, for reasons presently to be mentioned, this percentage must be only regarded as provisional. It must also be borne in mind that the proportion of recent shells would be augmented if the uppermost beds at Bramerton, near Norwich, which belong to the most modern or Chillesford division of the Crag, had been included, as they were formerly, by Mr. Woodward and myself, in the Norwich series. Arctic shells, which formed so large a proportion in the Chillesford and Aldeby beds, are more rare in the Norwich Crag, though many northern species— such as Rhynchonella psittacea, Scalaria Groenlandica, Astarte borealis, Panopaea Norvegia, and others— still occur. The Nucula Cobboldiae and Tellina obliqua, Figures 119 and 120, before mentioned, are frequent in these beds, as are also Littorina littorea, Cardium edule, and Turritella communis, of our seas, proving the littoral origin of the beds.

OLDER PLIOCENE STRATA.

RED CRAG.

(FIGURE 121. Section through (left) sea, Red Crag, London Clay and Chalk (right).)

Among the English Pliocene beds the next in antiquity is the Red Crag, which often rests immediately on the London Clay, as in the county of Essex, illustrated in Figure 121.

It is chiefly in the county of Suffolk that it is found, rarely exceeding twenty feet in thickness, and sometimes overlying another Pliocene deposit, the Coralline Crag, to be mentioned in the sequel. It has yielded— exclusive of 25 species regarded by Mr. Wood as derivative— 256 species of mollusca, of which 65, or 25 per cent, are extinct. Thus, apart from its order of superposition, its greater antiquity than the Norwich and glacial beds, already described, is proved by the greater departure from the fauna of our seas. It may also be observed that in most of the deposits of this Red Crag, the northern forms of the Norwich Crag, and of such glacial formations as Bridlington, are less numerous, while those having a more southern aspect begin to make their appearance. Both the quartzose sand, of which it chiefly consists, and the included shells, are most commonly distinguished by a deep ferruginous or ochreous colour, whence its name. The shells are often rolled, sometimes comminuted, and the beds have much the appearance of having been shifting sand- banks, like those now forming on the Dogger-bank, in the sea, sixty miles east of the coast of Northumberland. Cross stratification is almost always present, the planes of the strata being sometimes directed towards one point of the compass, sometimes to the opposite, in beds immediately overlying. That such a structure is not deceptive or due to any subsequent concretionary rearrangement of particles, or to mere bands of colour produced by the iron, is proved by each bed being made up of flat pieces of shell which lie parallel to the planes of the smaller strata.

(FIGURE 122. Purpura tetragona, Sowerby; natural size.)

(FIGURE 123. Voluta Lamberti, Sowerby. Variety characteristic of Suffolk Crag. Pliocene.)

(FIGURE 124. Voluta Lamberti, young individual, Cor. and Red Crag.)

It has long been suspected that the different patches of Red Crag are not all of the same age, although their chronological relation can not be decided by superposition. Separate masses are characterised by shells specifically distinct or greatly varying in relative abundance, in a manner implying that the deposits containing them were separated by intervals of time. At Butley, Tunstall, Sudbourn, and in the Red Crag of Chillesford, the mollusca appear to assume their most modern aspect when the climate was colder than when the earliest deposits of the same period were formed. At Butley, Nucula Cobboldiae, so common in the Norwich and certain glacial beds, is found, and Purpura tetragona (Figure 122) is very abundant. On the other hand, at Walton-on-the-Naze, in Essex, we seem to have an exhibition of the oldest phase of the Red Crag; and a warmer climate seems indicated, not only by the absence of many northern forms, but also by the abundance of some now living in the British seas and the Mediterranean. Voluta Lamberti (see Figures 123 and 124), an extinct form, which seems to have flourished chiefly in the antecedent Coralline Crag period, is still represented here by individuals of every age.

(FIGURE 125. Trophon antiquum, Muller. (Fusus contrarius) half natural size.)

The reversed whelk (Figure 125) is common at Walton, where the dextral form of that shell is unknown. Here also we find most frequently specimens of lamellibranchiate molluscs, with both the valves united, showing that they belonged to this sea of the Upper Crag, and were not washed in from an older bed, such as the Coralline, in which case the ligament would not have held together the valves in strata so often showing signs of the boisterous action of the waves. No less than forty species of lamellibranchiate molluscs, with double valves, have been collected by Mr. Bell from the various localities of the Red Crag.

At and near the base of the Red Crag is a loose bed of brown nodules, first noticed by Professor Henslow as containing a large percentage of earthy phosphates. This bed of coprolites (as it is called, because they were originally supposed to be the faeces of animals) does not always occur at one level, but is generally in largest quantity at the junction of the Crag and the underlying formation. In thickness it usually varies from six to eighteen inches, and in some rare cases amounts to many feet. It has been much used in agriculture for manure, as not only the nodules, but many of the separate bones associated with them, are largely impregnated with phosphate of lime, of which there is sometimes as much as sixty per cent. They are not unfrequently covered with barnacles, showing that they were not formed as concretions in the stratum where they now lie buried, but had been previously consolidated. The phosphatic nodules often collect fossil crabs and fishes from the London Clay, together with the teeth of gigantic sharks. In the same bed have been found many ear- bones of whales, and the teeth of Mastodon arvernensis, Rhinoceros Schleiermacheri, Tapirus priscus, and Hipparion (a quadruped of the horse family), and antlers of a stag, Cervus anoceros. Organic remains also of the older chalk and Lias are met with, showing how great was the denudation of previous formations during the Pliocene period. As the older White Crag, presently to be mentioned, contains similar phosphatic nodules near its base, those of the Red Crag may be partly derived from this source.

WHITE OR CORALLINE CRAG.

The lower or Coralline Crag is of very limited extent, ranging over an area about twenty miles in length, and three or four in breadth, between the rivers Stour and Alde, in Suffolk. It is generally calcareous and marly— often a mass of comminuted shells, and the remains of bryozoa (or polyzoa), passing occasionally into a soft building-stone. (Ehrenberg proposed in 1831 the term Bryozoum, or "Moss-animal," for the molluscous or ascidian form of polyp, characterised by having two openings to the digestive sack, as in Eschara, Flustra, Retepora, and other zoophytes popularly included in the corals, but now classed by naturalists as mollusca. The term Polyzoum, synonymous with Bryozoum, was, it seems, proposed in 1830, or the year before, by Mr. J.O. Thompson.) At Sudbourn and Gedgrave, near Orford, this building-stone has been largely quarried. At some places in the neighbourhood the softer mass is divided by thin flags of hard limestone, and bryozoa placed in the upright position in which they grew. From the abundance of these coralloid mollusca the lowest or White Crag obtained its popular name, but true corals, as now defined, or zoantharia, are very rare in this formation.

The Coralline Crag rarely, if ever, attains a thickness of thirty feet in any one section. Mr. Prestwich imagines that if the beds found at different localities were united in the probable order of their succession, they might exceed eighty feet in thickness, but Mr. Searles Wood does not believe in the possibility of establishing such a chronological succession by aid of the organic remains, and questions whether proof could be obtained of more than forty feet. I was unable to come to any satisfactory opinion on the subject, although at Orford, especially at Gedgrave, in the neighbourhood of that place, I saw many sections in pits, where this crag is cut through. These pits are so unconnected, and of such limited extent, that no continuous section of any length can be obtained, so that speculations as to the thickness of the whole deposit must be very vague. At the base of the formation at Sutton a bed of phosphatic nodules, very similar to that before alluded to in the Red Crag, with remains of mammalia, has been met with.

(FIGURE 126. Section near Woodbridge, in Suffolk. Through Sutton (left), Shottisham Creek, Ramsholt (right) and R. Deben. a. Red Crag. b. Coralline Crag. c. London Clay.)

Whenever the Red and Coralline Crag occur in the same district, the Red Crag lies uppermost; and in some cases, as in the section represented in Figure 126, which I had an opportunity of seeing exposed to view in 1839, it is clear that the older deposit, or Coralline Crag, b, had suffered denudation, before the newer formation, a, was thrown down upon it. At D there was not only seen a distinct cliff, eight or ten feet high, of Coralline Crag, running in a direction N.E. and S.W., against which the Red Crag abuts with its horizontal layers, but this cliff occasionally overhangs. The rock composing it is drilled everywhere by Pholades, the holes which they perforated having been afterwards filled with sand, and covered over when the newer beds were thrown down. The older formation is shown by its fossils to have accumulated in a deeper sea, and contains none of those littoral forms such as the limpet, Patella, found in the Red Crag. So great an amount of denudation could scarcely take place, in such incoherent materials, without some of the fossils of the inferior beds becoming mixed up with the overlying crag, so that considerable difficulty must be occasionally experienced by the palaeontologist in deciding which species belong severally to each group.

(FIGURE 127. Fascicularia aurantium, Milne Edwards. Family, Tubuliporidae, of same author. Bryozoan of extinct genus, from the inferior or Coralline Crag, Suffolk. a. Exterior. b. Vertical section of interior. c. Portion of exterior magnified. d. Portion of interior magnified, showing that it is made up of long, thin, straight tubes, united in conical bundles.)

(FIGURE 128. Astarte omalii, laj.; species common to Upper and lower crag.)

Mr. Searles Wood estimates the total number of marine testaceous mollusca of the Coralline Crag at 350, of which 110 are not known as living, being in the proportion of thirty-one per cent extinct. No less than 130 species of bryozoa have been found in the Coralline Crag, and some belong to genera unknown in the living creation, and of a very peculiar structure; as, for example, that represented in Figure 127, which is one of several species having a globular form. Among the testacea the genus Astarte (see Figure 128) is largely represented, no less than fourteen species being known, and many of these being rich in individuals. There is an absence of genera peculiar to hot climates, such as Conus, Oliva, Fasciolaria, Crassatella, and others. The absence also of large cowries (Cyprea), those found belonging exclusively to the section Trivia, is remarkable. The large volute, called Voluta Lamberti (Figure 123), may seem an exception; but it differs in form from the volutes of the torrid zone, and, like the living Voluta Magellanica, must have been fitted for an extra-tropical climate.

(FIGURE 129. Lingula Dumortieri, Nyst; Suffolk and Antwerp Crag.)

(FIGURE 130. Pyrula reticulata, Lam.; Coralline Crag, Ramsholt.)

(FIGURE 131. Temnechinus excavatus, Forbes; Temnopleurus excavatus, Wood; Coralline Crag, Ramsholt.)

The occurrence of a species of Lingula at Sutton (see Figure 129) is worthy of remark, as these Brachiopoda seem now confined to more equatorial latitudes; and the same may be said still more decidedly of a species of Pyrula, supposed by Mr. Wood to be identical with P. reticulata (Figure 130), now living in the Indian Ocean. A genus also of echinoderms, called by Professor Forbes Temnechinus (Figure 131), occurs in the Red and Coralline Crag of Suffolk, and until lately was unknown in a living state, but it has been brought to light as an existing form by the deep-sea dredgings, both of the United States survey, off Florida, at a depth of from 180 to 480 feet, and more recently (1869), in the British seas, during the explorations of the "Porcupine."

CLIMATE OF THE CRAG DEPOSITS.

One of the most interesting conclusions deduced from a careful comparison of the shells of the British Pliocene strata and the fauna of our present seas has been pointed out by Professor E. Forbes. It appears that, during the Glacial period, a period intermediate, as we have seen, between that of the Crag and our own time, many shells, previously established in the temperate zone, retreated southward to avoid an uncongenial climate, and they have been found fossil in the Newer Pliocene strata of Sicily, Southern Italy, and the Grecian Archipelago, where they may have enjoyed, during the era of floating icebergs, a climate resembling that now prevailing in higher European latitudes. (E. Forbes Mem. Geological Survey of Great Britain volume 1 page 386.) The Professor gave a list of fifty shells which inhabited the British seas while the Coralline and Red Crag were forming, and which, though now living in our seas, were wanting, as far as was then known, in the glacial deposits. Some few of these species have subsequently been found in the glacial drift, but the general conclusion of Forbes remains unshaken.

The transport of blocks by ice, when the Red Crag was being deposited, appears to me evident from the large size of some huge, irregular, quite unrounded chalk flints, retaining their white coating, and 2 feet long by 18 inches broad, in beds worked for phosphatic nodules at Foxhall, four miles south-east of Ipswich. These must have been tranquilly drifted to the spot by floating ice. Mr. Prestwich also mentions the occurrence of a large block of porphyry in the base of the Coralline Crag at Sutton, which would imply that the ice-action had begun in our seas even in this older period. The cold seems to have gone on increasing from the time of the Coralline to that of the Norwich Crag, and became more and more severe, not perhaps without some oscillations of temperature, until it reached its maximum in what has been called the Glacial period, or at the close of the Newer Pliocene, and in the Post-pliocene periods.

RELATION OF THE FAUNA OF THE CRAG TO THAT OF THE RECENT SEAS.

By far the greater number of the recent marine species occurring in the several Crag formations are still inhabitants of the British seas; but even these differ considerably in their relative abundance, some of the commonest of the Crag shells being now extremely scarce— as, for example, Buccinum Dalei— while others, rarely met with in a fossil state, are now very common, as Murex erinaceus and Cardium echinatum. Some of the species also, the identity of which with the living would not be disputed by any conchologist, are nevertheless distinguishable as varieties, whether by slight deviations in form or a difference in average dimensions. Since Mr. Searles Wood first described the marine testacea of the Crags, the additions made to that fossil fauna have not been considerable, whereas we have made in the same period immense progress in our knowledge of the living testacea of the British and arctic seas, and of the Mediterranean. By this means the naturalist has been enabled to identify with existing species many forms previously supposed to be extinct.

In the forthcoming supplement to the invaluable monograph communicated by Mr. Wood to the Palaeontographical Society, in which he has completed his figures and descriptions of the British crag shells of every age, list will be found of all the fossil shells, of which a summary is given in the table below.

TABLE OF NUMBER OF KNOWN SPECIES OF MARINE TESTACEA IN THE CRAG.

COLUMN 1: KNOWN SPECIES. COLUMN 2: TOTAL NUMBER OF KNOWN SPECIES. COLUMN 3: NUMBER OF SPECIES NOT KNOWN AS LIVING.

CHILLESFORD AND ALDEBY BEDS: Bivalves: 61 : 4. Univalves: 33 : 5. Brachiopods: 0 : 0. PERCENTAGE OF SHELLS NOT KNOWN AS LIVING : 9.5.

NORWICH OR FLUVIO-MARINE CRAG: Bivalves: 61 : 10. Univalves: 64 : 12. Brachiopods: 1 : 0. PERCENTAGE OF SHELLS NOT KNOWN AS LIVING : 17.5.

RED CRAG (Exclusive of many derivative shells): Bivalves: 128 : 31. Univalves: 127 : 33. Brachiopods: 1 : 1. PERCENTAGE OF SHELLS NOT KNOWN AS LIVING : 25.0.

CORALLINE CRAG: Bivalves: 161 : 47. Univalves: 184 : 60 Brachiopods: 5 : 3 PERCENTAGE OF SHELLS NOT KNOWN AS LIVING : 31.5

To begin with the uppermost or Chillesford beds, it will be seen that about 9 per cent only are extinct, or not known as living, whereas in the Norwich, which succeeds in the descending order, seventeen in a hundred are extinct. Formerly, when the Norwich or Fluvio-marine Crag was spoken of, both these formations were included under the same head, for both at Bramerton and Thorpe, the chief localities where the Norwich Crag was studied, an overlying deposit occurs referable to the Chillesford age. If now the two were fused together as of old, their shells would, according to Mr. Wood, yield a percentage of fifteen in a hundred of species extinct or not known as living.

To come next to the Red Crag, the reader will observe that a percentage of 25 is given of shells unknown as living, and this increases to 31 in the antecedent Coralline Crag. But the gap between these two stages of our Pliocene deposits is really wider than these numbers would indicate, for several reasons. In the first place, the Coralline Crag is more strictly the product of a single period, the Red Crag, as we have seen, consisting of separate and independent patches, slightly varying in age, of which the newest is probably not much anterior to the Norwich Crag. Secondly, there was a great change of conditions, both as to the depth of the sea and climate, between the periods of the Coralline and Red Crag, causing the fauna in each to differ far more widely than would appear from the above numerical results.

The value of the analysis given in the above table of the shells of the Red and Coralline Crags is in no small degree enhanced by the fact that they were all either collected by Mr. Wood himself, or obtained by him direct from their discoverers, so that he was enabled in each case to test their authenticity, and as far as possible to avoid those errors which arise from confounding together shells belonging to the sea of a newer deposit, and those washed into it from a formation of older date. The danger of this confusion may be conceived when we remember that the number of species rejected from the Red Crag as derivative by Mr. Wood is no less than 25. Some geologists have held that on the same grounds it is necessary to exclude as spurious some of the species found in the Norwich Crag proper; but Mr. Wood does not entertain this view, believing that the spurious shells which have sometimes found their way into the lists of this crag have been introduced by want of care from strata of Red Crag.

There can be no doubt, on the other hand, that conchologists have occasionally rejected from the Red and Norwich Crags, as derivative, shells which really belonged to the seas of those periods, because they were extinct or unknown as living, which in their eyes afforded sufficient ground for suspecting them to be intruders. The derivative origin of a species may sometimes be indicated by the extreme scarcity of the individuals, their colour, and worn condition; whereas an opposite conclusion may be arrived at by the integrity of the shells, especially when they are of delicate and tender structure, or their abundance, and, in the case of the lamellibranchiata, by their being held together by the ligament, which often happens when the shells have been so broken that little more than the hinges of the two valves are preserved. As to the univalves, I have seen from a pit of Red Crag, near Woodbridge, a large individual of the extinct Voluta Lamberti, seven inches in length, of which the lip, then perfect, had in former stages of its growth been frequently broken, and as often repaired. It had evidently lived in the sea of the Red Crag, where it had been exposed to rough usage, and sustained injuries like those which the reversed whelk, Trophon antiquum, so characteristic of the same formation, often exhibits. Additional proofs, however, have lately been obtained by Mr. Searles Wood that this shell had not died out in the era of the Red Crag by the discovery of the same fossil near Southwold, in beds of the later Norwich Crag.

ANTWERP CRAG.

Strata of the same age as the Red and Coralline Crag of Suffolk have been long known in the country round Antwerp, and on the banks of the Scheldt, below that city; and the lowest division, or Black Crag, there found, is shown by the shells to be somewhat more ancient than any of our British series, and probably forms the first links of a downward passage from the strata of the Pliocene to those of the Upper Miocene period.

NEWER PLIOCENE STRATA OF SICILY.

(FIGURE 132. Murex vaginatus, Phil.)

At several points north of Catania, on the eastern sea-coast of Sicily— as at Aci-Castello, for example, Trezza, and Nizzeti— marine strata, associated with volcanic tuffs and basaltic lavas, are seen, which belong to a period when the first igneous eruptions of Mount Etna were taking place in a shallow bay of the Mediterranean. They contain numerous fossil shells, and out of 142 species that have been collected all but eleven are identical with species now living. Some few of these eleven shells may possibly still linger in the depths of the Mediterranean, like Murex vaginatus, see Figure 132. The last-mentioned shell had already become rare when the associated marine and volcanic strata above alluded to were formed. On the whole, the modern character of the testaceous fauna under consideration is expressed not only by the small proportion of extinct species, but by the relative number of individuals by which most of the other species are represented, for the proportion agrees with that observed in the present fauna of the Mediterranean. The rarity of individuals in the extinct species is such as to imply that they were already on the point of dying out, having flourished chiefly in the earlier Pliocene times, when the Subapennine strata were in progress.

Yet since the accumulation of these Newer Pliocene sands and clays, the whole cone of Etna, 11,000 feet in height and about 90 miles in circumference at its base, has been slowly built up; an operation requiring many tens of thousands of years for its accomplishment, and to estimate the magnitude of which it is necessary to study in detail the internal structure of the mountain, and to see the proofs of its double axis, or the evidence of the lavas of the present great centre of eruption having gradually overwhelmed and enveloped a more ancient cone, situated 3 1/2 miles to the east of the present one. (See a Memoir on the Lavas and Mode of Origin of Mount Etna by the Author in Philosophical Transactions 1858.)

It appears that while Etna was increasing in bulk by a series of eruptions, its whole mass, comprising the foundations of subaqueous origin above alluded to, was undergoing a slow upheaval, by which those marine strata were raised to the height of 1200 feet above the sea, as seen at Catera, and perhaps to greater heights, for we can not trace their extension westward, owing to the dense and continuous covering of modern lava under which they are buried. During the gradual rise of these Newer Pliocene formations (consisting of clays, sands, and basalts) other strata of Post-pliocene date, marine as well as fluviatile, accumulated round the base of the mountain, and these, in their turn, partook of the upward movement, so that several inland cliffs and terraces at low levels, due partly to the action of the sea and partly to the river Simeto, originated in succession. Fossil remains of the elephant, and other extinct quadrupeds, have been found in these Post-Pliocene strata, associated with recent shells.

There is probably no part of Europe where the Newer Pliocene formations enter so largely into the structure of the earth's crust, or rise to such heights above the level of the sea, as Sicily. They cover nearly half the island, and near its centre, at Castrogiovanni, reach an elevation of 3000 feet. They consist principally of two divisions, the upper calcareous and the lower argillaceous, both of which may be seen at Syracuse, Girgenti, and Castrogiovanni. According to Philippi, to whom we are indebted for the best account of the tertiary shells of this island, thirty-five species out of one hundred and twenty-four obtained from the beds in central Sicily are extinct.

A geologist, accustomed to see nearly all the Newer Pliocene formations in the north of Europe occupying low grounds and very incoherent in texture, is naturally surprised to behold formations of the same age so solid and stony, of such thickness, and attaining so great an elevation above the level of the sea. The upper or calcareous member of this group in Sicily consists in some places of a yellowish-white stone, like the Calcaire Grossier of Paris; in others, of a rock nearly as compact as marble. Its aggregate thickness amounts sometimes to 700 or 800 feet. It usually occurs in regular horizontal beds, and is occasionally intersected by deep valleys, such as those of Sortino and Pentalica, in which are numerous caverns. The fossils are in every stage of preservation, from shells retaining portions of their animal matter and colour to others which are mere casts. The limestone passes downward into a sandstone and conglomerate, below which is clay and blue marl, from which perfect shells and corals may be disengaged. The clay sometimes alternates with yellow sand.

South of the plain of Catania is a region in which the tertiary beds are intermixed with volcanic matter, which has been for the most part the product of submarine eruptions. It appears that, while the clay, sand, and yellow limestone before mentioned were in course of deposition at the bottom of the sea, volcanoes burst out beneath the waters, like that of Graham Island, in 1831, and these explosions recurred again and again at distant intervals of time. Volcanic ashes and sand were showered down and spread by the waves and currents so as to form strata of tuff, which are found intercalated between beds of limestone and clay containing marine shells, the thickness of the whole mass exceeding 2000 feet. The fissures through which the lava rose may be seen in many places, forming what are called DIKES.

(FIGURE 133. Pecten jacobaeus; half natural size.)

No shell is more conspicuous in these Sicilian strata than the great scallop, Pecten jacobaeus (Figure 133), now so common in the neighbouring seas. The more we reflect on the preponderating number of this and other recent shells, the more we are surprised at the great thickness, solidity, and height above the sea of the rocky masses in which they are entombed, and the vast amount of geographical change which has taken place since their origin. It must be remembered that, before they began to emerge, the uppermost strata of the whole must have been deposited under water. In order, therefore, to form a just conception of their antiquity, we must first examine singly the innumerable minute parts of which the whole is made up, the successive beds of shells, corals, volcanic ashes, conglomerates, and sheets of lava; and we must afterwards contemplate the time required for the gradual upheaval of the rocks, and the excavation of the valleys. The historical period seems scarcely to form an appreciable unit in this computation, for we find ancient Greek temples, like those of Girgenti (Agrigentum), built of the modern limestone of which we are speaking, and resting on a hill composed of the same; the site having remained to all appearances unaltered since the Greeks first colonised the island.

It follows, from the modern geological date of these rocks, that the fauna and flora of a large part of Sicily are of higher antiquity than the country itself. The greater part of the island has been raised above the sea since the epoch of existing species, and the animals and plants now inhabiting it must have migrated from adjacent countries, with whose productions the species are now identical. The average duration of species would seem to be so great that they are destined to outlive many important changes in the configuration of the earth's surface, and hence the necessity for those innumerable contrivances by which they are enabled to extend their range to new lands as they are formed, and to escape from those which sink beneath the sea.

NEWER PLIOCENE STRATA OF THE UPPER VAL D'ARNO.

When we ascend the Arno for about ten miles above Florence, we arrive at a deep narrow valley called the Upper Val d'Arno, which appears once to have been a lake, at a time when the valley below Florence was an arm of the sea. The horizontal lacustrine strata of this upper basin are twelve miles long and two broad. The depression which they fill has been excavated out of Eocene and Cretaceous rocks, which form everywhere the sides of the valley in highly inclined stratification. The thickness of the more modern and unconformable beds is about 750 feet, of which the upper 200 feet consist of Newer Pliocene strata, while the lower are Older Pliocene. The newer series are made up of sands and a conglomerate called "sansino." Among the imbedded fossil mammalia are Mastodon arvernensis, Elephas meridionalis, Rhinoceros etruscus, Hippopotamus major, and remains of the genera bear, hyaena, and felis, nearly all of which occur in the Cromer forest-bed (see Chapter 13).

In the same upper strata are found, according to M. Gaudin, the leaves and cones of Glyptostrobus europaeus, a plant closely allied to G. heterophyllus, now inhabiting the north of China and Japan. This conifer had a wide range in time, having been traced back to the Lower Miocene strata of Switzerland, and being common at Oeningen in the Upper Miocene, as we shall see in the sequel (Chapter 14.)

OLDER PLIOCENE OF ITALY.— SUBAPENNINE STRATA.

The Apennines, it is well-known, are composed chiefly of Secondary or Mesozoic rocks, forming a chain which branches off from the Ligurian Alps and passes down the middle of the Italian peninsula. At the foot of these mountains, on the side both of the Adriatic and the Mediterranean, are found a series of tertiary strata, which form, for the most part, a line of low hills occupying the space between the older chain and the sea. Brocchi was the first Italian geologist who described this newer group in detail, giving it the name of the Subapennine. Though chiefly composed of Older Pliocene strata, it belongs, nevertheless, in part, both to older and newer members of the tertiary series. The strata, for example, of the Superga, near Turin, are Miocene; those of Asti and Parma Older Pliocene, as is the blue marl of Sienna; while the shells of the incumbent yellow sand of the same territory approach more nearly to the recent fauna of the Mediterranean, and may be Newer Pliocene.

We have seen that most of the fossil shells of the Older Pliocene strata of Suffolk which are of recent species are identical with testacea now living in British seas, yet some of them belong to Mediterranean species, and a few even of the genera are those of warmer climates. We might therefore expect, in studying the fossils of corresponding age in countries bordering the Mediterranean, to find among them some species and genera of warmer latitudes. Accordingly, in the marls belonging to this period at Asti, Parma, Sienna, and parts of the Tuscan and Roman territories, we observe the genera Conus, Cypraea, Strombus, Pyrula, Mitra, Fasciolaria, Sigaretus, Delphinula, Ancillaria, Oliva, Terebellum, Terebra, Perna, Plicatula, and Corbis, some characteristic of tropical seas, others represented by species more numerous or of larger size than those now proper to the Mediterranean.

OLDER PLIOCENE FLORA OF ITALY.

(FIGURE 134. Oreodaphne Heerii. Leaf half natural size. (Feuilles fossiles de la Toscane.))

I have already alluded to the Newer Pliocene deposits of the Upper Val d'Arno above Florence, and stated that below those sands and conglomerates, containing the remains of the Elephas meridionalis and other associated quadrupeds, lie an older horizontal and conformable series of beds, which may be classed as Older Pliocene. They consist of blue clays with some subordinate layers of lignite, and exhibit a richer flora than the overlying Newer Pliocene beds, and one receding farther from the existing vegetation of Europe. They also comprise more species common to the antecedent Miocene period. Among the genera of flowering plants, M. Gaudin enumerates pine, oak, evergreen oak, plum, plane, alder, elm, fig, laurel, maple, walnut, birch, buckthorn, hickory, sumach, sarsaparilla, sassafras, cinnamon, Glyptostrobus, Taxodium, Sequoia, Persea, Oreodaphne (Figure 134), Cassia, and Psoralea, and some others. This assemblage of plants indicates a warm climate, but not so subtropical an one as that of the Upper Miocene period, which will presently be considered.

(FIGURE 135. Liquidambar europaeum, var. trilobatum, A. Br. (sometimes four- lobed, and more commonly five-lobed). a. Leaf, half natural size. b. Part of same, natural size. c. Fruit, natural size. d. Seed, natural size. Oeningen.)

M. Gaudin, jointly with the Marquis Strozzi, has thrown much light on the botany of beds of the same age in another part of Tuscany, at a place called Montajone, between the rivers Elsa and Evola, where, among other plants, is found the Oreodaphne Heerii, Gaud. (See Figure 134), which is probably only a variety of Oreodaphne foetens, or the laurel called the Til in Madeira, where, as in the Canaries, it constitutes a large portion of the native woods, but can not now endure the climate of Europe. In the fossil specimens the same glands or protuberances are preserved (see Figure 134) as those which are seen in the axils of the primary veins of the leaves in the recent Til. (Contributions a la Flore fossile Italienne. Gaudin and Strozzi. Plate 11 Figure 3. Gaudin page 22.) Another plant also indicating a warmer climate is the Liquidambar europaeum, Brong. (see Figure 135), a species nearly allied to L. styracifluum, L., which flourishes in most places in the Southern States of North America, on the borders of the Gulf of Mexico.

CHAPTER XIV.

MIOCENE PERIOD.— UPPER MIOCENE.

Upper Miocene Strata of France.— faluns of Touraine. Tropical Climate implied by Testacea. Proportion of recent Species of Shells. faluns more ancient than the Suffolk Crag. Upper Miocene of Bordeaux and the South of France. Upper Miocene of Oeningen, in Switzerland. Plants of the Upper Fresh-water Molasse. Fossil Fruit and Flowers as well as Leaves. Insects of the Upper Molasse. Middle or Marine Molasse of Switzerland. Upper Miocene Beds of the Bolderberg, in Belgium. Vienna Basin. Upper Miocene of Italy and Greece. Upper Miocene of India; Siwalik Hills. Older Pliocene and Miocene of the United States.

UPPER MIOCENE STRATA OF FRANCE.— FALUNS OF TOURAINE.

The strata which we meet with next in the descending order are those called by many geologists "Middle Tertiary," for which in 1833 I proposed the name of Miocene, selecting the "faluns" of the valley of the Loire, in France, as my example or type. I shall now call these falunian deposits Upper Miocene, to distinguish them from others to which the name of Lower Miocene will be given.

No British strata have a distinct claim to be regarded as Upper Miocene, and as the Lower Miocene are also but feebly represented in the British Isles, we must refer to foreign examples in illustration of this important period in the earth's history. The term "faluns" is given provincially by French agriculturists to shelly sand and marl spread over the land in Touraine, just as similar shelly deposits were formerly much used in Suffolk to fertilise the soil, before the coprolitic or phosphatic nodules came into use. Isolated masses of such faluns occur from near the mouth of the Loire, in the neighbourhood of Nantes, to as far inland as a district south of Tours. They are also found at Pontlevoy, on the Cher, about seventy miles above the junction of that river with the Loire, and thirty miles south-east of Tours. Deposits of the same age also appear under new mineral conditions near the towns of Dinan and Rennes, in Brittany. I have visited all the localities above enumerated, and found the beds on the Loire to consist principally of sand and marl, in which are shells and corals, some entire, some rolled, and others in minute fragments. In certain districts, as at Doue, in the Department of Maine and Loire, ten miles south- west of Saumur, they form a soft building-stone, chiefly composed of an aggregate of broken shells, bryozoa, corals, and echinoderms, united by a calcareous cement; the whole mass being very like the Coralline Crag near Aldborough, and Sudbourn in Suffolk. The scattered patches of faluns are of slight thickness, rarely exceeding fifty feet; and between the district called Sologne and the sea they repose on a great variety of older rocks; being seen to rest successively upon gneiss, clay-slate, various secondary formations, including the chalk; and, lastly, upon the upper fresh-water limestone of the Parisian tertiary series, which, as before mentioned (Chapter 9), stretches continuously from the basin of the Seine to that of the Loire.