(FIGURE 136. Dinotherium giganteum, Kaup.)

At some points, as at Louans, south of Tours, the shells are stained of a ferruginous colour, not unlike that of the Red Crag of Suffolk. The species are, for the most part, marine, but a few of them belong to land and fluviatile genera. Among the former, Helix turonensis (Figure 38, Chapter 3) is the most abundant. Remains of terrestrial quadrupeds are here and there intermixed, belonging to the genera Dinotherium (Figure 136), Mastodon, Rhinoceros, Hippopotamus, Chaeropotamus, Dichobune, Deer, and others, and these are accompanied by cetacea, such as the Lamantin, Morse, Sea-calf, and Dolphin, all of extinct species.

The fossil testacea of the faluns of the Loire imply, according to the late Edward Forbes, that the beds were formed partly on the shore itself at the level of low water, and partly at very moderate depths, not exceeding ten fathoms below that level. The molluscan fauna is, on the whole, much more littoral than that of the Pliocene Red and Coralline Crag of Suffolk, and implies a shallower sea. It is, moreover, contrasted with the Suffolk Crag by the indications it affords of an extra-European climate. Thus it contains seven species of Cypraea, some larger than any existing cowry of the Mediterranean, several species of Oliva, Ancillaria, Mitra, Terebra, Pyrula, Fasciolaria, and Conus. Of the cones there are no less than eight species, some very large, whereas the only European cone now living is of diminutive size. The genus Nerita, and many others, are also represented by individuals of a type now characteristic of equatorial seas, and wholly unlike any Mediterranean forms. These proofs of a more elevated temperature seem to imply the higher antiquity of the faluns as compared with the Suffolk Crag, and are in perfect accordance with the fact of the smaller proportion of testacea of recent species found in the faluns.

Out of 290 species of shells, collected by myself in 1840 at Pontlevoy, Louans, Bossee, and other villages twenty miles south of Tours, and at Savigne, about fifteen miles north-west of that place, seventy-two only could be identified with recent species, which is in the proportion of twenty-five per cent. A large number of the 290 species are common to all the localities, those peculiar to each not being more numerous than we might expect to find in different bays of the same sea.

The total number of species of testaceous mollusca from the faluns in my possession is 302, of which forty-five only, or fourteen per cent, were found by Mr. Wood to be common to the Suffolk Crag. The number of corals, including bryozoa and zoantharia, obtained by me at Doue and other localities before adverted to, amounts to forty-three, as determined by Mr. Lonsdale, of which seven (one of them a zoantharian) agree specifically with those of the Suffolk Crag. Some of the genera occurring fossil in Touraine, as the corals Astrea and Dendrophyllia, and the bryozoan Lunulites, have not been found in European seas north of the Mediterranean; nevertheless, the zoantharia of the faluns do not seem to indicate, on the whole, so warm a climate as would be inferred from the shells.

It was stated that, on comparing about 300 species of Touraine shells with about 450 from the Suffolk Crag, forty-five only were found to be common to both, which is in the proportion of only fifteen per cent. The same small amount of agreement is found in the corals also. I formerly endeavoured to reconcile this marked difference in species with the supposed co-existence of the two faunas, by imagining them to have severally belonged to distinct zoological provinces or two seas, the one opening to the north and the other to the south, with a barrier of land between them, like the Isthmus of Suez, now separating the Red Sea and the Mediterranean. But I now abandon that idea for several reasons; among others, because I succeeded in 1841 in tracing the Crag fauna southward in Normandy to within seventy miles of the Falunian type, near Dinan, yet found that both assemblages of fossils retained their distinctive characters, showing no signs of any blending of species or transition of climate.

The principal grounds, however, for referring the English Crag to the older Pliocene and the French faluns to the Upper Miocene epochs, consist in the predominance of fossil shells in the British strata identifiable with species not only still living, but which are now inhabitants of neighbouring seas, while the accompanying extinct species are of genera such as characterise Europe. In the faluns, on the contrary, the recent species are in a decided minority; and most of them are now inhabitants of the Mediterranean, the coast of Africa, and the Indian Ocean; in a word, less northern in character, and pointing to the prevalence of a warmer climate. They indicate a state of things receding farther from the present condition of Central Europe in physical geography and climate, and doubtless, therefore, receding farther from our era in time.

(FIGURE 137. Voluta Lamberti, Sowerby. Variety characteristic of Faluns of Touraine. Miocene.)

Among the conspicuous fossils common to the faluns of the Loire and the Suffolk Crag is a variety of the Voluta Lamberti, a shell already alluded to (Figure 123). The specimens of this shell which I have myself collected in Touraine, or have seen in museums, are thicker and heavier than British individuals of the same species, and shorter in proportion to their width, and have the folds on the columella less oblique, as represented in Figure 137.

UPPER MIOCENE OF BORDEAUX AND THE SOUTH OF FRANCE.

A great extent of country between the Pyrenees and the Gironde is overspread by tertiary deposits of various ages, and chiefly of Miocene date. Some of these, near Bordeaux, coincide in age with the faluns of Touraine, already mentioned, but many of the species of shells are peculiar to the south. The succession of beds in the basin of the Gironde implies several oscillations of level by which the same wide area was alternately converted into sea and land and into brackish-water lagoons, and finally into fresh-water ponds and lakes.

Among the fresh-water strata of this age near the base of the Pyrenees are marls, limestones and sands, in which the eminent comparative anatomist, M. Lartet, has obtained a great number of fossil mammalia common to the faluns of the Loire and the Upper Miocene beds of Switzerland, such as Dinotherium giganteum and Mastodon angustidens; also the bones of quadrumana, or of the ape and monkey tribe, which were discovered in 1837, the first of that order of quadrupeds detected in Europe. They were found near Auch, in the Department of Gers, in latitude 43 degrees 39' N. About forty miles west of Toulouse. They were referred by MM. Lartet and Blainville to a genus closely allied to the Gibbon, to which they gave the name of Pliopithecus. Subsequently, in 1856, M. Lartet described another species of the same family of long-armed apes (Hylobates), which he obtained from strata of the same age at Saint-Gaudens, in the Haute Garonne. The fossil remains of this animal consisted of a portion of a lower jaw with teeth and the shaft of a humerus. It is supposed to have been a tree-climbing frugivorous ape, equalling man in stature. As the trunks of oaks are common in the lignite beds in which it lay, it has received the generic name of Dryopithecus. The angle formed by the ascending ramus of the jaw and the alveolar border is less open, and therefore more like the human subject, than in the Chimpanzee, and what is still more remarkable, the fossil, a young but adult individual, had all its milk teeth replaced by the second set, while its last true molar (or wisdom-tooth) was still undeveloped, or only existed as a germ in the jaw-bone. In the mode, therefore, of the succession of its teeth (which, as in all the old-World apes, exactly agree in number with those in man) it differed from the Gorilla and Chimpanzee, and corresponded with the human species.

UPPER MIOCENE BEDS OF OENINGEN, IN SWITZERLAND.

The faluns of the Loire first served, as already stated, as the type of the Miocene formations in Europe. They yielded a plentiful harvest of marine fossil shells and corals, but were entirely barren of plants and insects. In Switzerland, on the other hand, deposits of the same age have been discovered, remarkable for their botanical and entomological treasures. We are indebted to Professor Heer, of Zurich, for the description, restoration, and classification of several hundred species and varieties of these fossil plants, the whole of which he has illustrated by excellent figures in his "Flora Tertiaria Helvetiae." This great work, and those of Adolphe Brongniart, Unger, Goppert and others, show that this class of fossils is beginning to play the same important part in the classification of the tertiary strata containing lignite or brown coal as an older flora has long played in enabling us to understand the ancient coal or carboniferous formation. No small skepticism has always prevailed among botanists as to whether the leaves alone and the wood of plants could ever afford sufficient data for determining even genera and families in the vegetable kingdom. In truth, before such remains could be rendered available a new science had to be created. It was necessary to study the outlines, nervation, and microscopic structure of the leaves, with a degree of care which had never been called for in the classification of living plants, where the flower and fruit afforded characters so much more definite and satisfactory. As geologists, we can not be too grateful to those who, instead of despairing when so difficult a task was presented to them, or being discouraged when men of the highest scientific attainments treated the fossil leaves as worthless, entered with full faith and enthusiasm into this new and unexplored field. That they should frequently have fallen into errors was unavoidable, but it is remarkable, especially if we inquire into the history of Professor Heer's researches, how often early conjectures as to the genus and family founded on the leaves alone were afterwards confirmed when fuller information was obtained. As examples to be found on comparing Heer's earlier and later works, I may instance the chestnut, elm, maple, cinnamon, magnolia, buckbean or Menyanthes, vine, buckthorn (Rhamnus), Andromeda and Myrica, and among the conifers Sequoia and Taxodium. In all these cases the plants were first recognised by their leaves, and the accuracy of the determination was afterwards confirmed when the fruit, and in some instances both fruit and flower, were found attached to the same stem as the leaves.

But let us suppose that no fruit, seed, or flower had ever been met with in a fossil state, we should still have been indebted to the persevering labours of botanical palaeontologists for one of the grandest scientific discoveries for which the present century is remarkable— namely, the proofs now established of the prevalence of a mild climate and a rich arborescent flora in the arctic regions in that Miocene epoch on the history of which we are now entering. It may be useful if I endeavour to give the reader in a few words some idea of the nature of the evidence of these important conclusions, to show how far they may be safely based on fossil leaves alone. When we begin by studying the fossils of the Newer Pliocene deposits, such as those of the Upper Val d'Arno, before alluded to, we perceive that the fossil foliage agrees almost entirely with the trees and shrubs of a modern European forest. In the plants of the Older Pliocene strata of the same region we observe a larger proportion of species and genera which, although they may agree with well-known Asiatic or other foreign types, are at present wanting in Italy. If we then examine the Miocene formations of the same country, exotic forms become more abundant, especially the palms, whether they belong to the European or American fan-palms, Chamaerops and Sabal, or to the more tropical family of the date-palms or Phoenicites, which last are conspicuous in the Lower Miocene beds of Central Europe. Although we have not found the fruit or flower of these palms in a fossil state, the leaves are so characteristic that no one doubts the family to which they belong, or hesitates to accept them as indications of a warm and sub-tropical climate.

When the Miocene formations are traced to the northward of the 50th degree of latitude, the fossil palms fail us, but the greater proportion of the leaves, whether identical with those of existing European trees or of forms now unknown in Europe, which had accompanied the Miocene palms, still continue to characterise rocks of the same age, until we meet with them not only in Iceland, but in Greenland, in latitude 70 degrees N., and in Spitzbergen, latitude 78 degrees 56', or within about 11 degrees of the pole, and under circumstances which clearly show them to have been indigenous in those regions, and not to have been drifted from the south (see Chapter 15). Not only, therefore, has the botanist afforded the geologist much palaeontological assistance in identifying distinct tertiary formations in distant places by his power of accurately discriminating the forms, veining, and microscopic structure of leaves or wood, but, independently of that exact knowledge derivable from the organs of fructification, we are indebted to him for one of the most novel, unexpected results of modern scientific inquiry.

The Miocene formations of Switzerland have been called MOLASSE, a term derived from the French MOL, and applied to a SOFT, incoherent, greenish sandstone, occupying the country between the Alps and the Jura. This molasse comprises three divisions, of which the middle one is marine, and being closely related by its shells to the faluns of Touraine, may be classed as Upper Miocene. The two others are fresh-water, the upper of which may be also grouped with the faluns, while the lower must be referred to the Lower Miocene, as defined in the next chapter.

UPPER FRESH-WATER MOLASSE.

This formation is best seen at Oeningen, in the valley of the Rhine, between Constance and Schaffhausen, a locality celebrated for having produced in the year 1700 the supposed human skeleton called by Scheuchzer "homo diluvii testis," a fossil afterwards demonstrated by Cuvier to be a reptile, or aquatic salamander, of larger dimensions than even its great living representative, the salamander of Japan.

The Oeningen strata consist of a series of marls and limestones, many of them thinly laminated, and which appear to have slowly accumulated in a lake probably fed by springs holding carbonate of lime in solution. The elliptical area over which this fresh-water formation has been traced extends, according to Sir Roderick Murchison, for a distance of ten miles east and west from Berlingen, on the right bank of the river to Wangen, and to Oeningen, near Stein, on the left bank. The organic remains have been chiefly derived from two quarries, the lower of which is about 550 feet above the level of the Lake of Constance, while the upper quarry is 150 feet higher. In this last, a section thirty feet deep displays a great succession of beds, most of them splitting into slabs and some into very thin laminae. Twenty-one beds are enumerated by Professor Heer, the uppermost a bluish-grey marl seven feet thick, with organic remains, resting on a limestone with fossil plants, including leaves of poplar, cinnamon, and pond- weed (Potamogeton), together with some insects; while in the bed No. 4, below, is a bituminous rock, in which the Mastodon tapiroides, a characteristic Upper Miocene quadruped, has been met with. The 5th bed, two or three inches thick, contains fossil fish, e.g., Leuciscus (roach), and the larvae of dragon-flies, with plants such as the elm (Ulmus), and the aquatic Chara. Below this are other plant-beds; and then, in No. 9, the stone in which the great salamander (Andrias Scheuchzeri) and some fish were found. Below this other strata occur with fish, tortoises, the great salamander before alluded to, fresh-water mussels, and plants. In No. 16 the fossil fox of Oeningen, galecynus Oeningensis, Owen, was obtained by Sir R. Murchison. To this succeed other beds with mammalia (Lagomys), reptiles, (Emys), fish, and plants, such as walnut, maple, and poplar. In the 19th bed are numerous fish, insects, and plants, below which are marls of a blue indigo colour.

In the lower quarry eleven beds are mentioned, in which, as in the upper, both land and fresh-water plants and many insects occur. In the 6th, reckoning from the top, many plants have been obtained, such as Liquidambar, Daphnogene, Podogonium, and Ulmus, together with tortoises, besides the bones and teeth of a ruminant quadruped, named by H. von Meyer Palaeomeryx eminens. No. 9 is called the insect-bed, a layer only a few inches thick, which, when exposed to the frost, splits into leaves as thin as paper. In these thin laminae plants such as Liquidambar, Daphnogene, and Glyptostrobus, occur, with innumerable insects in a wonderful state of preservation, usually found singly. Below this is an indigo- blue marl, like that at the bottom of the higher quarry, resting on yellow marl ascertained to be at least thirty feet thick.

(FIGURE 138. Cinnamomum polymorphum, Ad. Brong. Upper and Lower Miocene. a. Leaf. b. Flower, natural size; Heer Plate 93 Figure 28. c. Ripe fruit of Cinnamomum polymorphum, from Oeningen; Heer, Plate 94 Figure 14. d. Fruit of recent Cinnamomum camphorum of Japan; Heer, Plate 152 Figure 18.)

All the above fossil-bearing strata were evidently formed with extreme slowness. Although the fossiliferous beds are, in the aggregate, no more than a few yards in thickness, and have only been examined in the small area comprised in the two quarries just alluded to, they give us an insight into the state of animal and vegetable life in part of the Upper Miocene period, such as no other region in the world has elsewhere supplied. In the year 1859, Professor Heer had already determined no less than 475 species of plants and more than 800 insects from these Oeningen beds. He supposes that a river entering a lake floated into it some of the leaves and land insects, together with the carcasses of quadrupeds, among others a great Mastodon. Occasionally, during tempests, twigs and even boughs of trees with their leaves were torn off and carried for some distance so as to reach the lake. Springs, containing carbonate of lime, seem at some points to have supplied calcareous matter in solution, giving origin locally to a kind of travertin, in which organic bodies sinking to the bottom became hermetically sealed up. The laminae, says Heer, which immediately succeed each other were not all formed at the same season, for it can be shown that, when some of them originated, certain plants were in flower, whereas, when the next of these layers was produced, the same plants had ripened their fruit. This inference is confirmed by independent proofs derived from insects. The principal insect-bed is rarely two inches thick, and is composed, says Heer, of about 250 leaf-like laminae, some of which were deposited in the spring, when the Cinnamomum polymorphum (Figure 138) was in flower, others in summer, when winged ants were numerous, and when the poplar and willow had matured their seed; others, again, in autumn, when the same Cinnamomum polymorphum (Figure 138) was in fruit, as well as the liquidambar, oak, clematis, and many other plants. The ancient lake seems to have had a belt of poplars and willows round its borders, countless leaves of which were imbedded in mud, and together with them, at some points, a species of reed, Arundo, which was very common.

One of the most characteristic shrubs is a papilionaceous and leguminous plant of an extinct genus, called by Heer Podogonium, of which two species are known. Entire twigs have been found with flowers, and always without leaves, as the flowers evidently came out, as in the poplar and willow tribe, before any leaves made their appearance. Other specimens have been obtained with ripe fruits accompanied by leaves, which resemble those of the tamarind, to which it was evidently allied, being of the family Caesalpineae, now proper to warmer regions.

(FIGURE 139. Acer trilobatum, normal form; Heer, Flora Tert. Helv. Plate 114 Figure 2. Size 1/2 diam. (Part only of the long stalk of the original fossil specimen is here given ). Upper Miocene, Oeningen; also found in Lower Miocene of Switzerland.)

(FIGURE 140. Acer trilobatum. a. Abnormal variety of leaf; Heer, Plate 110 Figure 16. b. Flower and bracts, normal form; Heer, Plate 111 Figure 21. c. Half a seed-vessel; Heer, Plate 111 Figure 5.)

(FIGURE 141. Platanus aceroides, Gopp.; Heer, Plate 88 Figures 5-8. Size 2/3 diam. Upper Miocene, Oeningen. a. Leaf. b. The core of a bundle of pericarps. c. Single fruit or pericarp, natural size.)

The Upper Miocene flora of Oeningen is peculiarly important, in consequence of the number of genera of which not merely the leaves, but, as in the case of the Podogonium just mentioned, the fruit also and even the flower are known. Thus there are nineteen species of maple, ten of which have already been found with fruit. Although in no one region of the globe do so many maples now flourish, we need not suspect Professor Heer of having made too many species in this genus when we consider the manner in which he has dealt with one of them, Acer trilobatum, Figures 139 and 140. Of this plant the number of marked varieties figured and named is very great, and no less than three of them had been considered as distinct species by other botanists, while six of the others might have laid claim, with nearly equal propriety, to a like distinction. The common form, called Acer trilobatum, Figure 139, may be taken as a normal representative of the Oeningen fossil, and Figure 140, as one of the most divergent varieties, having almost four lobes in the leaf instead of three.

(FIGURE 142. Smilax sagittifera; Heer, Plate 30 Figure 7. Size 1/2 diameter. a. Leaf. b. Flower magnified, one of the six petals wanting at d. Upper Miocene, Oeningen. c. Smilax obtusifolia; Heer, Plate 30 Figure 9; natural size. Upper Miocene, Oeningen.)

(FIGURE 143. Fruit of the fossil and recent species of Hakea, a genus of Proteaceae. a. Leaf of fossil species, Hakea salicina. Upper Miocene, Oeningen; Heer Plate 97 Figure 29. 1/3 diameter. b. Impression of woody fruit of same, showing thick stalk. 2/3 diameter. c. Seed of same, natural size. d. Fruit of living Australian species, Hakea saligna, R. Brown. 1/2 diameter. e. Seed of same, natural size.)

Among the conspicuous genera which abounded in the Miocene period in Europe is the plane-tree, Platanus, the fossil species being considered by Heer to come nearer to the American P. occidentalis than to P. orientalis of Greece and Asia Minor. In some of the fossil specimens the male flowers are preserved. Among other points of resemblance with the living plane-trees, as we see them in the parks and squares of London, fossil fragments of the trunk are met with, having pieces of their bark peeling off.

The vine of Oeningen, Vitis teutonica, Ad. Brong, is of a North American type. Both the leaves and seeds have been found at Oeningen, and bunches of compressed grapes of the same species have been met with in the brown coal of Wetteravia in Germany. No less than eight species of smilax, a monocotyledonous genus, occur at Oeningen and in other Upper Miocene localities, the flowers of some of them, as well as the leaves, being preserved; as in the case of the very common fossil, S. sagittifera, Figure 142, a.

Leaves of plants supposed to belong to the order Proteaceae have been obtained partly from Oeningen and partly from the lacustrine formation of the same age at Locle in the Jura. They have been referred to the genera Banksia, Grevillea, Hakea, and Persoonia. Of Hakea there is the impression of a supposed seed- vessel, with its characteristic thick stalk and seeds, but as the fruit is without structure, and has not yet been found attached to the same stem as the leaf, the proof is incomplete.

To whatever family the foliage hitherto regarded as proteaceous by many able palaeontologists may eventually be shown to belong, we must be careful not to question their affinity to that order of plants on those geographical considerations which have influenced some botanists. The nearest living Proteaceae now feel the in Abyssinia in latitude 20 degrees N., but the greatest number are confined to the Cape and Australia. The ancestors, however, of the Oeningen fossils ought not to be looked for in such distant regions, but from that European land which in Lower Miocene times bore trees with similar foliage, and these had doubtless an Eocene source, for cones admitted by all botanists to be proteaceous have been met with in one division of that older Tertiary group (see Figure 206 Chapter 16). The source of these last, again, must not be sought in the antipodes, for in the white chalk of Aix-la-Chapelle leaves like those of Grevillea and other proteaceous genera have been found in abundance, and, as we shall see in Chapter 17, in a most perfect state of preservation. All geologists agree that the distribution of the Cretaceous land and sea had scarcely any connection with the present geography of the globe.

(FIGURE 144. Glyptostrobus Europaeus. Branch with ripe fruit; Heer, Plate 20 Figure 1. Upper Miocene, Oeningen.)

In the same beds with the supposed Proteaceae there occurs at Locle a fan-palm of the American type Sabal (for genus see Figure 151), a genus which ranges throughout the low country near the sea from the Carolinas to Florida and Louisiana. Among the Coniferae of Upper Miocene age is found a deciduous cypress nearly allied to the Taxodium distichum of North America, and a Glyptostrobus (Figure 144), very like the Japanese G. heterophyllus, now common in our shrubberies.

Before the appearance of Heer's work on the Miocene Flora of Switzerland, Unger and Goppert had already pointed out the large proportion of living North American genera which distinguished the vegetation of the Miocene period in Central Europe. Next in number, says Heer, to these American forms at Oeningen the European genera preponderate, the Asiatic ranking in the third, the African in the fourth, and the Australian in the fifth degree. The American forms are more numerous than in the Italian Pliocene flora, and the whole vegetation indicates a warmer climate than the Pliocene, though not so high a temperature as that of the older or Lower Miocene period.

The conclusions drawn from the insects are for the most part in perfect harmony with those derived from the plants, but they have a somewhat less tropical and less American aspect, the South European types being more numerous. On the whole, the insect fauna is richer than that now inhabiting any part of Europe. No less than 844 species are reckoned by Heer from the Oeningen beds alone, the number of specimens which he has examined being 5080. The entire list of Swiss species from the Upper and Lower Miocene together amount to 1322. Almost all the living families of Coleoptera are represented, but, as we might have anticipated from the preponderance of arborescent and ligneous plants, the wood-eating beetles play the most conspicuous part, the Buprestidae and other long-horned beetles being particularly abundant.

(FIGURE 145. Harpactor maculipes, Heer. Upper Miocene, Oeningen.)

The patterns and some remains of the colours both of Coleoptera and Hemiptera are preserved at Oeningen, as, for example in Harpactor (Figure 145), in which the antennae, one of the eyes, and the legs and wings are retained. The characters, indeed, of many of the insects are so well defined as to incline us to believe that if this class of the invertebrata were not so rare and local, they might be more useful than even the plants and shells in settling chronological points in geology.

MIDDLE OR MARINE MOLASSE (UPPER MIOCENE) OF SWITZERLAND.

It was before stated that the Miocene formation of Switzerland consisted of, first, the upper fresh-water molasse, comprising the lacustrine marls of Oeningen; secondly, the marine molasse, corresponding in age to the faluns of Touraine; and thirdly, the lower fresh-water molasse. Some of the beds of the marine or middle series reach a height of 2470 feet above the sea. A large number of the shells are common to the faluns of Touraine, the Vienna basin, and other Upper Miocene localities. The terrestrial plants play a subordinate part in the fossiliferous beds, yet more than ninety of them are enumerated by Heer as belonging to this falunian division, and of these more than half are common to subjacent Lower Miocene beds, while a proportion of about forty-five in one hundred are common to the overlying Oeningen flora. Twenty-six of the ninety-two species are peculiar.

UPPER MIOCENE OF THE BOLDERBERG, IN BELGIUM.

(FIGURE 146. Oliva Dufresnii, Bast. Bolderberg, Belgium; natural size. a. Front view. b. Back view.)

In a small hill or ridge called the Bolderberg, which I visited in 1851, situated near Hasselt, about forty miles E.N.E. of Brussels, strata of sand and gravel occur, to which M. Dumont first called attention as appearing to constitute a northern representative of the faluns of Touraine. On the whole, they are very distinct in their fossils from the two upper divisions of the Antwerp Crag before mentioned (Chapter 13), and contain shells of the genera Oliva, Conus, Ancillaria, Pleurotoma, and Cancellaria in abundance. The most common shell is an Olive (Figure 146), called by Nyst Oliva Dufresnii; and constituting, as M. Bosquet observes, a smaller and shorter variety of the Bordeaux species.

So far as the shells of the Bolderberg are known, the proportion of recent species agrees with that in the faluns of Touraine, and the climate must have been warmer than that of the Coralline Crag of England.

UPPER MIOCENE BEDS OF THE VIENNA BASIN.

In South Germany the general resemblance of the shells of the Vienna tertiary basin with those of the faluns of Touraine has long been acknowledged. In the late Dr. Hornes's excellent work on the fossil mollusca of that formation, we see accurate figures of many shells, clearly of the same species as those found in the falunian sands of Touraine.

According to Professor Suess, the most ancient and purely marine of the Miocene strata in this basin consist of sands, conglomerates, limestones, and clays, and they are inclined inward, or from the borders of the trough towards the centre, their outcropping edges rising much higher than the newer beds, whether Miocene or Pliocene, which overlie them, and which occupy a smaller area at an inferior elevation above the sea. M. Hornes has described no less than 500 species of gasteropods, of which he identifies one-fifth with living species of the Mediterranean, Indian, or African seas, but the proportion of existing species among the lamellibranchiate bivalves exceeds this average. Among many univalves agreeing with those of Africa on the eastern side of the Atlantic are Cypraea sanguinolenta, Buccinum lyratum, and Oliva flammulata. In the lowest marine beds of the Vienna basin the remains of several mammalia have been found, and among them a species of Dinotherium, a Mastodon of the Trilophodon family, a Rhinoceros (allied to R. megarhinus, Christol), also an animal of the hog tribe, Listriodon, von Meyer, and a carnivorous animal of the canine family. The Helix turonensis (Figure 38 Chapter 3), the most common land shell of the French faluns, accompanies the above land animals. In a higher member of the Vienna Miocene series are found Dinotherium giganteum (Figure 136 Chapter 14), Mastodon longirostris, Rhinoceros Schleiermacheri, Acerotherium incisivum, and Hippotherium gracile, all of them equally characteristic of an Upper Miocene deposit occurring at Eppelsheim, in Hesse Darmstadt; a locality also remarkable as having furnished in latitude 49 degrees 50 north the bone of a large ape of the Gibbon kind, the most northerly example yet discovered of a quadrumanous animal.

(FIGURE 147. Amphistegina Hauerina, d'Orbigny. Upper Miocene strata, Vienna.)

M. Alcide d'Orbigny has shown that the foraminifera of the Vienna basin differ alike from the Eocene and Pliocene species, and agree with those of the faluns, so far as the latter are known. Among the Vienna foraminifera, the genus Amphistegina (Figure 147) is very characteristic, and is supposed by d'Archiac to take the same place among the Rhizopods of the Upper Miocene era which the Nummulites occupy in the Eocene period.

The flora of the Vienna basin exhibits some species which have a general range through the whole Miocene period, such as Cinnamomum polymorphum (Figure 138 Chapter 14), and C. Scheuchzeri, also Planera Richardi, Mich., Liquidambar europaeum (Figure 135 Chapter 13) Juglans bilinica, Cassia ambigua, and C. lignitum. Among the plants common to the Upper Miocene beds of Oeningen, in Switzerland, are Platanus aceroides (Figure 141 Chapter 14), Myrica vindobonensis, and others.

UPPER MIOCENE STRATA OF ITALY.

We are indebted to Signor Michelotti for a valuable work on the Miocene shells of Northern Italy. Those found in the hill called the Superga, near Turin, have long been known to correspond in age with the faluns of Touraine, and they contain so many species common to the Upper Miocene strata of Bordeaux as to lead to the conclusion that there was a free communication between the northern part of the Mediterranean and the Bay of Biscay in the Upper Miocene period.

UPPER MIOCENE FORMATIONS OF GREECE.

At Pikerme, near Athens, MM. Wagner and Roth have described a deposit in which they found the remains of the genera Mastodon, Dinotherium, Hipparion, two species of Giraffe, Antelope, and others, some living and some extinct. With them were also associated fossil bones of the Semnopithecus, showing that here, as in the south of France, the quadrumana were characteristic of this period. The whole fauna attests the former extension of a vast expanse of grassy plains where we have now the broken and mountainous country of Greece; plains, which were probably united with Asia Minor, spreading over the area where the deep Aegean Sea and its numerous islands are now situated. We are indebted to M. Gaudry, who visited Pikerme, for a treatise on these fossil bones, showing how many data they contribute to the theory of a transition from the mammalia of the Upper Miocene through the Pliocene and Post-pliocene forms to those of living genera and species.

UPPER MIOCENE OF INDIA. SIWALIK HILLS.

The Siwalik Hills lie at the southern foot of the Himalayan chain, rising to the height of 2000 and 3000 feet. Between the Jumna and the Ganges they consist of inclined strata of sandstone, shingle, clay, and marl. We are indebted to the indefatigable researches of Dr. Falconer and Sir Proby Cautley, continued for fifteen years, for the discovery in these marls and sandstones of a great variety of fossil mammalia and reptiles, together with many fresh-water shells. Out of fifteen species of shells of the genera Paludina, Melania, Ampullaria, and Unio, all are extinct or unknown species with the exception of four, which are still inhabitants of Indian rivers. Such a proportion of living to extinct mollusca agrees well with the usual character of an Upper Miocene or Falunian fauna, as observed in Touraine, or in the basin of Vienna and elsewhere.

The genera of mammalia point in the same direction. One of them, of the genus Chalicotherium (or Anisodon of Lartet), is a pachyderm intermediate between the Rhinoceros and Anoplothere, and characteristic of the Upper Miocene strata of Eppelsheim, and of the south of France. With it occurs also an extinct form of Hippopotamus, called Hexaprotodon, and a species of Hippotherium and pig, also two species of Mastodon, two of elephant, and three other elephantine proboscidians; none of them agreeing with any fossil forms of Europe, and being intermediate between the genera Elephas and Mastodon, constituting the sub-genus Stegodon of Falconer. With these are associated a monkey, allied to the Semnopithecus entellus, now living in the Himalaya, and many ruminants. Among these last, besides the giraffe, camel, antelope, stag, and others, we find a remarkable new type, the Sivatherium, like a gigantic four-horned deer. There are also new forms of carnivora, both feline and canine, the Machairodus among the former, also hyaenas, and a subursine form called the Hyaenarctos, and a genus allied to the otter (Enhydriodon), of formidable size.

The giraffe, camel, and a large ostrich may be cited as proofs that there were formerly extensive plains where now a steep chain of hills, with deep ravines, runs for many hundred miles east and west. Among the accompanying reptiles are several crocodiles, some of huge dimensions, and one not distinguishable, says Dr. Falconer, from a species now living in the Ganges (C. Gangeticus); and there is still another saurian which the same anatomist has identified with a species now inhabiting India. There was also an extinct species of tortoise of gigantic proportions (Colossochelys Atlas), the curved shell of which was twelve feet three inches long and eight feet in diameter, the entire length of the animal being estimated at eighteen feet, and its probable height seven feet.

Numerous fossils of the Siwalik type have also been found in Perim Island, in the Gulf of Cambay, and among these a species of Dinotherium, a genus so characteristic of the Upper Miocene period in Europe.

OLDER PLIOCENE AND MIOCENE FORMATIONS IN THE UNITED STATES.

Between the Alleghany Mountains, formed of older rocks, and the Atlantic, there intervenes, in the United States, a low region occupied principally by beds of marl, clay, and sand, consisting of the cretaceous and tertiary formations, and chiefly of the latter. The general elevation of this plain bordering the Atlantic does not exceed 100 feet, although it is sometimes several hundred feet high. Its width in the middle and southern states is very commonly from 100 to 150 miles. It consists, in the South, as in Georgia, Alabama, and South Carolina, almost exclusively of Eocene deposits; but in North Carolina, Maryland, Virginia, Delaware, more modern strata predominate, of the age of the English Crag and faluns of Touraine. (Proceedings of the Geological Society volume 4 part 3 1845 page 547.)

(FIGURE 148. Fulgur canaliculatus. Maryland.)

(FIGURE 149. Fusus quadricostatus, Say. Maryland.)

In the Virginian sands, we find in great abundance a species of Astarte (A. undulata, Conrad), which resembles closely, and may possibly be a variety of, one of the commonest fossils of the Suffolk Crag (A. Omalii); the other shells also, of the genera Natica, Fissurella, Artemis, Lucina, Chama, Pectunculus, and Pecten, are analagous to shells both of the English Crag and French faluns, although the species are almost all distinct. Out of 147 of these American fossils I could only find thirteen species common to Europe, and these occur partly in the Suffolk Crag, and partly in the faluns of Touraine; but it is an important characteristic of the American group, that it not only contains many peculiar extinct forms, such as Fusus quadricostatus, Say (see Figure 149), and Venus tridacnoides, abundant in these same formations, but also some shells which, like Fulgur carica of Say and F. canaliculatus (see Figure 148), Calyptraea costata, Venus mercenaria, Lam., Modiola glandula, Totten, and Pecten magellanicus, Lam., are recent species, yet of forms now confined to the western side of the Atlantic— a fact implying that some traces of the beginning of the present geographical distribution of mollusca date back to a period as remote as that of the Miocene strata.

Of ten species of corals which I procured on the banks of the James River, one agrees generically with a coral now living on the coast of the United States. Mr. Lonsdale regarded these corals as indicating a temperature exceeding that of the Mediterranean, and the shells would lead to similar conclusions. Those occurring on the James River are in the 37th degree of N. latitude, while the French faluns are in the 47th; yet the forms of the American fossils would scarcely imply so warm a climate as must have prevailed in France when the Miocene strata of Touraine originated.

(FIGURE 150. Astrangia lineata, Lonsdale. Syn. Anthophyllum lineatum. Williamsburg, Virginia.)

Among the remains of fish in these post-eocene strata of the United States are several large teeth of the shark family, not distinguishable specifically from fossils of the faluns of Touraine.

CHAPTER XV.

LOWER MIOCENE (OLIGOCENE OF BEYRICH).

Lower Miocene Strata of France. Line between Miocene and Eocene. Lacustrine Strata of Auvergne. Fossil mammalia of the Limagne d'Auvergne. Lower Molasse of Switzerland. Dense Conglomerates and Proofs of Subsidence. Flora of the Lower Molasse. American Character of the Flora. Theory of a Miocene Atlantis. Lower Miocene of Belgium. Rupelian Clay of Hermsdorf near Berlin. Mayence Basin. Lower Miocene of Croatia. Oligocene Strata of Beyrich. Lower Miocene of Italy. Lower Miocene of England. Hempstead Beds. Bovey Tracey Lignites in Devonshire. Isle of Mull Leaf-Beds. Arctic Miocene Flora. Disco Island. Lower Miocene of United States. Fossils of Nebraska.

LINE BETWEEN MIOCENE AND EOCENE FORMATIONS.

The marine faluns of the valley of the Loire have been already described as resting in some places on a fresh-water tertiary limestone, fragments of which have been broken off and rolled on the shores and in the bed of the Miocene sea. Such pebbles are frequent at Pontlevoy on the Cher, with hollows drilled in them in which the perforating marine shells of the Falunian period still remain. Such a mode of superposition implies an interval of time between the origin of the fresh-water limestone and its submergence beneath the waters of the Upper Miocene sea. The limestone in question forms a part of the formation called the Calcaire de la Beauce, which constitutes a large table-land between the basins of the Loire and the Seine. It is associated with marls and other deposits, such as may have been formed in marshes and shallow lakes in the newest part of a great delta. Beds of flint, continuous or in nodules, accumulated in these lakes, and aquatic plants called Charae, left their stems and seed-vessels imbedded both in the marl and flint, together with fresh-water and land shells. Some of the siliceous rocks of this formation are used extensively for mill- stones. The flat summits or platforms of the hills round Paris, and large areas in the forest of Fontainebleau, as well as the Plateau de la Beauce, already alluded to, are chiefly composed of these fresh-water strata. Next to these in the descending order are marine sands and sandstone, commonly called the Gres de Fontainebleau, from which a considerable number of shells, very distinct from those of the faluns, have been obtained at Etampes, south of Paris, and at Montmartre and other hills in Paris itself, or in its suburbs. At the bottom of these sands a green clay occurs, containing a small oyster, Ostrea cyathula, Lam., which, although of slight thickness, is spread over a wide area. This clay rests immediately on the Paris gypsum, or that series of beds of gypsum and gypseous marl from which Cuvier first obtained several species of Palaeotherium and other extinct mammalia. (Bulletin 1856 Journal volume 12 page 768.)

At this junction of the clay and the gypsum the majority of French geologists have always drawn the line between the Middle and Lower Tertiary, or between the Miocene and Eocene formations, regarding the Fontainebleau sands and the Ostrea cyathula clay as the base of the Miocene, and the gypsum, with its mammalia, as the top of the Eocene group. I formerly dissented from this division, but I now find that I must admit it to be the only one which will agree with the distribution of the Miocene mammalia, while even the mollusca of the Fontainebleau sands, which were formerly supposed to present at preponderance of affinities to an Eocene fauna, have since been shown to agree more closely with the fossils of certain deposits always regarded as Middle Tertiary at Mayence and in Belgium. In fact, we are now arriving at that stage of progress when the line, wherever it be drawn between Miocene and Eocene, will be an arbitrary one, or one of mere convenience, as I shall have an opportunity of showing when the Upper Eocene formations in the Isle of Wight are described in the sixteenth chapter.

LOWER MIOCENE OF CENTRAL FRANCE.

Lacustrine strata, belonging, for the most part, to the same Miocene system as the Calcaire de la Beauce, are again met with farther south in Auvergne, Cantal, and Velay. They appear to be the monuments of ancient lakes, which, like some of those now existing in Switzerland, once occupied the depressions in a mountainous region, and have been each fed by one or more rivers and torrents. The country where they occur is almost entirely composed of granite and different varieties of granitic schist, with here and there a few patches of Secondary strata, much dislocated, and which have suffered great denudation. There are also some vast piles of volcanic matter, the greater part of which is newer than the fresh-water strata, and is sometimes seen to rest upon them, while a small part has evidently been of contemporaneous origin. Of these igneous rocks I shall treat more particularly in the sequel.

The study of these regions possesses a peculiar interest very distinct in kind from that derivable from the investigation either of the Parisian or English Tertiary areas. For we are presented in Auvergne with the evidence of a series of events of astonishing magnitude and grandeur, by which the original form and features of the country have been greatly changed, yet never so far obliterated but that they may still, in part at least, be restored in imagination. Great lakes have disappeared— lofty mountains have been formed, by the reiterated emission of lava, preceded and followed by showers of sand and scoriae— deep valleys have been subsequently furrowed out through masses of lacustrine and volcanic origin— at a still later date, new cones have been thrown up in these valleys— new lakes have been formed by the damming up of rivers— and more than one assemblage of quadrupeds, birds, and plants, Eocene, Miocene, and Pliocene, have followed in succession; yet the region has preserved from first to last its geographical identity; and we can still recall to our thoughts its external condition and physical structure before these wonderful vicissitudes began, or while a part only of the whole had been completed. There was first a period when the spacious lakes, of which we still may trace the boundaries, lay at the foot of mountains of moderate elevation, unbroken by the bold peaks and precipices of Mont Dor, and unadorned by the picturesque outline of the Puy de Dome, or of the volcanic cones and craters now covering the granitic platform. During this earlier scene of repose deltas were slowly formed; beds of marl and sand, several hundred feet thick, deposited; siliceous and calcareous rocks precipitated from the waters of mineral springs; shells and insects imbedded, together with the remains of the crocodile and tortoise, the eggs and bones of water-birds, and the skeletons of quadrupeds, most of them of genera and species characteristic of the Miocene period. To this tranquil condition of the surface succeeded the era of volcanic eruptions, when the lakes were drained, and when the fertility of the mountainous district was probably enhanced by the igneous matter ejected from below, and poured down upon the more sterile granite. During these eruptions, which appear to have taken place towards the close of the Miocene epoch, and which continued during the Pliocene, various assemblages of quadrupeds successively inhabited the district, among which are found the genera mastodon, rhinoceros, elephant, tapir, hippopotamus, together with the ox, various kinds of deer, the bear, hyaena, and many beasts of prey which ranged the forest or pastured on the plain, and were occasionally overtaken by a fall of burning cinders, or buried in flows of mud, such as accompany volcanic eruptions. Lastly, these quadrupeds became extinct, and gave place in their turn to the species now existing. There are no signs, during the whole time required for this series of events, of the sea having intervened, nor of any denudation which may not have been accomplished by currents in the different lakes, or by rivers and floods accompanying repeated earthquakes, or subterranean movements, during which the levels of the district have in some places been materially modified, and perhaps the whole upraised relatively to the surrounding parts of France.

AUVERGNE.

The most northern of the fresh-water groups is situated in the valley-plain of the Allier, which lies within the department of the Puy de Dome, being the tract which went formerly by the name of the Limagne d'Auvergne. The average breadth of this tract is about twenty miles; and it is for the most part composed of nearly horizontal strata of sand, sandstone, calcareous marl, clay, and limestone, none of which observe a fixed and invariable order of superposition. The ancient borders of the lake wherein the fresh-water strata were accumulated may generally be traced with precision, the granite and other ancient rocks rising up boldly from the level country. The actual junction, however, of the lacustrine beds and the granite is rarely seen, as a small valley usually intervenes between them. The fresh-water strata may sometimes be seen to retain their horizontality within a very slight distance of the border-rocks, while in some places they are inclined, and in few instances vertical. The principal divisions into which the lacustrine series may be separated are the following:— first, Sandstone, grit, and conglomerate, including red marl and red sandstone; secondly, Green and white foliated marls; thirdly, Limestone, or travertin, often oolitic in structure; fourthly, Gypseous marls.

The relations of these different groups can not be learnt by the study of any one section; and the geologist who sets out with the expectation of finding a fixed order of succession may perhaps complain that the different parts of the basin give contradictory results. The arenaceous division, the marls, and the limestone may all be seen in some places to alternate with each other; yet it can by no means be affirmed that there is no order of arrangement. The sands, sandstone, and conglomerate constitute in general a littoral group; the foliated white and green marl, a contemporaneous central deposit more than 700 feet thick, and thinly foliated, a character which often arises from the innumerable thin shells or carapace valves shed by the small crustacean called Cypris in the ancient lakes of Auvergne; and lastly the limestone is for the most part subordinate to the newer portions of both the above formations.

It seems that, when the ancient lake of the Limagne first began to be filled with sediment, no volcanic action had yet produced lava and scoriae on any part of the surface of Auvergne. No pebbles, therefore, of lava were transported into the lake— no fragments of volcanic rocks imbedded in the conglomerate. But at a later period, when a considerable thickness of sandstone and marl had accumulated, eruptions broke out, and lava and tuff were deposited, at some spots, alternately with the lacustrine strata. It is not improbable that cold and thermal springs, holding different mineral ingredients in solution, became more numerous during the successive convulsions attending this development of volcanic agency, and thus deposits of carbonate and sulphate of lime, silex, and other minerals were produced. Hence these minerals predominate in the uppermost strata. The subterranean movements may then have continued until they altered the relative levels of the country, and caused the waters of the lakes to be drained off, and the further accumulation of regular fresh-water strata to cease.

LOWER MIOCENE MAMMALIA OF THE LIMAGNE.

It is scarcely possible to determine the age of the oldest part of the fresh- water series of the Limagne, large masses both of the sandy and marly strata being devoid of fossils. Some of the lowest beds may be of Upper Eocene date, although, according to M. Pomel, only one bone of a Palaeotherium has been discovered in Auvergne. But in Velay, in strata containing some species of fossil mammalia common to the Limagne, no less than four species of Palaeothere have been found by M. Aymard, and one of these is generally supposed to be identical with Palaeotherium magnum, an undoubted Upper Eocene fossil, of the Paris gypsum, the other three being peculiar.

Not a few of the other mammalia of the Limagne belong undoubtedly to genera and species elsewhere proper to the Lower Miocene. Thus, for example, the Cainotherium of Bravard, a genus not far removed from the Anoplotherium, is represented by several species, one of which, as I learn from Mr. Waterhouse, agrees with Microtherium Renggeri of the Mayence basin. In like manner, the Amphitragulus elegans of Pomel, an Auvergne fossil, is identified by Waterhouse with Dorcatherium nanum of Kaup, a Rhenish species from Weissenau, near Mayence. A small species, also, of rodent, of the genus Titanomys of H. von Meyer, is common to the Lower Miocene of Mayence and the Limagne d'Auvergne, and there are many other points of agreement which the discordance of nomenclature tends to conceal. A remarkable carnivorous genus, the Hyaenodon of Laizer, is represented by more than one species. The same genus has also been found in the Upper Eocene marls of Hordwell Cliff, Hampshire, just below the level of the Bembridge Limestone, and therefore a formation older than the Gypsum of Paris. Several species of opossum (Didelphis) are met with in the same strata of the Limagne. The total number of mammalia enumerated by M. Pomel as appertaining to the Lower Miocene fauna of the Limagne and Velay falls little short of a hundred, and with them are associated some large crocodiles and tortoises, and some Ophidian and Batrachian reptiles.

LOWER MOLASSE OF SWITZERLAND.

The two upper divisions of the Swiss Molasse— the one fresh-water, the other marine— have already been described in the preceding chapter. I shall now proceed to treat of the third division, which is of Lower Miocene age. Nearly the whole of this Lower Molasse is fresh-water, yet some of the inferior beds contain a mixture of marine and fluviatile shells, the Cerithium margaritaceum, a well-known Lower Miocene fossil, being one of the marine species. Notwithstanding, therefore, that some of these Lower Miocene strata consist of old shingle-beds several thousand feet in thickness, as in the Rigi, near Lucerne, and in the Speer, near Wesen, mountains 5000 and 7000 feet above the sea, the deposition of the whole series must have begun at or below the sea- level.

The conglomerates, as might be expected, are often very unequal in thickness, in closely adjoining districts, since in a littoral formation accumulations of pebbles would swell out in certain places where rivers entered the sea, and would thin out to comparatively small dimensions where no streams or only small ones came down to the coast. For ages, in spite of a gradual depression of the land and adjacent sea-bottom, the rivers continued to cover the sinking area with their deltas; until finally, the subsidence being in excess, the sea of the Middle Molasse gained upon the land, and marine beds were thrown down over the dense mass of fresh-water and brackish-water deposit, called the Lower Molasse, which had previously accumulated.

FLORA OF THE LOWER MOLASSE.

In part of the Swiss Molasse, which belongs exclusively to the Lower Miocene period, the number of plants has been estimated at more than 500 species, somewhat exceeding those which were before enumerated as occurring in the two upper divisions. The Swiss Lower Miocene may best be studied on the northern borders of the Lake of Geneva, between Lausanne and Vevay, where the contiguous villages of Monod and Rivaz are situated. The strata there, which I have myself examined, consist of alternations of conglomerate, sandstone, and finely laminated marls with fossil plants. A small stream falls in a succession of cascades over the harder beds of pudding-stone, which resist, while the sandstone and plant-bearing shales and marls give way. From the latter no less than 193 species of plants have been obtained by the exertions of MM. Heer and Gaudin, and they are considered to afford a true type of the vegetation of the Lower Miocene formations of Switzerland— a vegetation departing farther in its character from that now flourishing in Europe than any of the higher members of the series before alluded to, and yet displaying so much affinity to the flora of Oeningen as to make it natural for the botanist to refer the whole to one and the same Miocene period. There are, indeed, no less than 81 species of these Older Miocene plants which pass up into the flora of Oeningen.

This fact is important as bearing on the propriety of classing the Lower Molasse of Switzerland as belonging to the Miocene rather than to the latter part of the Eocene period. There are, indeed, so many types among the fossils, both specific and generic, which have a wide range through the whole of the Molasse, that a unity of character is thereby stamped on the whole flora, in spite of the contrast between the plants of the uppermost and lowest formations, or between Oeningen and Monod. The proofs of a warmer climate, and the excess of arborescent over herbaceous plants, and of evergreen trees over deciduous species, are characters common to the whole flora, but which are intensified as we descend to the inferior deposits.

(FIGURE 151. Sabal major, Unger sp. Vevay. Lower Miocene; Heer, Plate 41.)

Nearly all the plants at Monod are contained in three layers of marl separated by two of soft sandstone. The thickness of the marls is ten feet, and vegetable matter predominates so much in some layers as to form an imperfect lignite. One bed is filled with large leaves of a species of fig (Ficus populina), and of a hornbeam (Carpinus grandis), the strength of the wind having probably been great when they were blown into the lake; whereas another contiguous layer contains almost exclusively smaller leaves, indicating, apparently, a diminished strength in the wind. Some of the upper beds at Monod abound in leaves of Proteaceae, Cyperaceae, and ferns, while in some of the lower ones Sequoia, Cinnamomum, and Sparganium are common. In one bed of sandstone the trunk of a large palm-tree was found unaccompanied by other fossils, and near Vevay, in the same series of Lower Miocene strata, the leaves of a palm of the genus Sabal (Figure 151), a genus now proper to America, were obtained.

Among other genera of the same class is a Flabellaria occurring near Lausanne, and a magnificent Phoenicites allied to the date palm. When these plants flourished the climate must have been much hotter than now. The Alps were no doubt much lower, and the palms now found fossil in strata elevated 2000 feet above the sea grew nearly at the sea-level, as is demonstrated by the brackish- water character of some of the beds into which they were carried by winds or rivers from the adjoining coast.

(FIGURE 152. Banksia. a. Fruit of fossil Banksia. b. Leaf of Banksia Deekiana.)

In the same plant-bearing deposits of the Lower Molasse in Switzerland leaves have been found which have been ascribed to the order Proteaceae already spoken of as well represented in the Oeningen beds (see Chapter 14). The Proteas and other plants of this family now flourish at the Cape of Good Hope; while the Banksias, and a set of genera distinct from those of Africa, grow most luxuriantly in the southern and temperate parts of Australia. They were probably inhabitants, says Heer, of dry hilly ground, and the stiff leathery character of their leaves must have been favourable to their preservation, allowing them to float on a river for great distances without being injured, and then to sink, when water-logged, to the bottom. It has been objected that the fruit of the Proteaceae is of so tough and enduring a texture that it ought to have been more commonly met with; but in the first place we must not forget the numerous cones found in the Eocene strata of Sheppey, which all admit to be proteaceous and to belong to at least two species (see Chapter 14). Secondly, besides the fruit of Hakea before mentioned (Chapter 14), Heer found associated with fossil leaves, having the exact form and nervation of Banksia, fruit precisely such as may have come from a cone of that plant, and lately he has received another similar fruit from the Lower Miocene strata of Lucerne. They may have fallen out of a decayed cone in the same way as often happens to the seeds of the spruce fir, Pinus abies, found scattered over the ground in our woods. It is a known fact that among the living Proteaceae the cones are very firmly attached to the branches, so that the seeds drop out without the cone itself falling to the ground, and this may perhaps be the reason why, in some instances in which fossil seeds have been found, no traces of the cone have been observed.

(FIGURE 153. Sequoia Langsdorfii. Ad. Brong., 1/3 natural size. Rivaz, near Lausanne; Heer, Plate 21 Figure 4. Upper and Lower Miocene and Lower Pliocene, Val d'Arno. a. Branch with leaves. b. Young cone.)

Among the Coniferae the Sequoia here figured is common at Rivaz, and is one of the most universal plants in the Lowest Miocene of Switzerland, while it also characterises the Miocene Brown Coals of Germany and certain beds of the Val d'Arno, which I have called Older Pliocene, Chapter 13.

(FIGURE 154. Lastraea stiriaca, Unger; Heer's Flora, Plate 143 Figure 8. Natural size. Lower and Upper Miocene, Switzerland. a. Specimen from Monod, showing the position of the sori on the middle of the tertiary nerves. b. More common appearance, where the sori remain and the nerves are obliterated.)

Among the ferns met with in profusion at Monod is the Lastraea stiriaca, Unger, which has a wide range in the Miocene period from strata of the age of Oeningen to the lowest part of the Swiss Molasse. In some specimens, as shown in Figure 154, the fructification is distinctly seen.

(FIGURE 155. Cinnamomum Rossmassleri, Heer. Daphnogene cinnamomifolia, Unger. Upper and Lower Miocene, Switzerland and Germany.)

Among the laurels several species of Cinnamomum are very conspicuous. Besides the C. polymorphum, before figured, Chapter 14, another species also ranges from the Lower to the Upper Molasse of Switzerland, and is very characteristic of different deposits of Brown Coal in Germany. It has been called Cinnamomum Rossmassleri by Heer (see Figure 155). The leaves are easily recognised as having two side veins, which run up uninterruptedly to their point.

AMERICAN CHARACTER OF THE FLORA.

If we consider not merely the number of species but those plants which constitute the mass of the Lower Miocene vegetation, we find the European part of the fossil flora very much less prominent than in the Oeningen beds, while the foreground is occupied by American forms, by evergreen oaks, maples, poplars, planes, Liquidambar, Robinia, Sequoia, Taxodium, and ternate-leaved pines. There is also a much greater fusion of the characters now belonging to distinct botanical provinces than in the Upper Miocene flora, and we shall find this fusion still more strikingly exemplified as we go back to the antecedent Eocene and Cretaceous periods.

Professor Heer has advocated the doctrine, first advanced by Unger to explain the large number of American genera in the Miocene flora of Europe, that the present basin of the Atlantic was occupied by land over which the Miocene flora could pass freely. But other able botanists have shown that it is far more probable that the American plants came from the east and not from the west, and instead of reaching Europe by the shortest route over an imaginary Atlantis, migrated in an opposite direction, crossing the whole of Asia.

ARCTIC MIOCENE FLORA.

But when we indulge in speculations as to the geographical origin of the Miocene plants of Central Europe, we must take into account the discoveries recently made of a rich terrestrial flora having flourished in the Arctic Regions in the Miocene period from which many species may have migrated from a common centre so as to reach the present continents of Europe, Asia, and America. Professor Heer has examined the various collections of fossil plants that have been obtained in North Greenland (latitude 70 degrees), Iceland, Spitzbergen, and other parts of the Arctic regions, and has determined that they are of Miocene age and indicate a temperate climate. (Heer "Miocene baltische Flora" and "Fossil-flora von Alaska" 1869.) Including the collections recently brought from Greenland by Mr. Whymper, the Arctic Miocene flora now comprises 194 species, and that of Greenland 137 species, of which 46, or exactly one-third, are identical with plants found in the Miocene beds of Central Europe. Considerably more than half the number are trees, which is the more remarkable since, at the present day, trees do not exist in any part of Greenland even 10 degrees farther south.

More than thirty species of Coniferae have been found, including several Sequoias (allied to the gigantic Wellingtonia of California), with species of Thujopsis and Salisburia now peculiar to Japan. There are also beeches, oaks, planes, poplars, maples, walnuts, limes, and even a magnolia, two cones of which have recently been obtained, proving that this splendid evergreen not only lived but ripened its fruit within the Arctic circle. Many of the limes, planes, and oaks were large-leaved species, and both flowers and fruit, besides immense quantities of leaves, are in many cases preserved. Among the shrubs were many evergreens, as Andromeda, and two extinct genera, Daphnogene and M'Clintockia, with fine leathery leaves, together with hazel, blackthorn, holly, logwood, and hawthorn. A species of Zamia (Zamites) grew in the swamps, with Potamogeton, Sparganium, and Menyanthes, while ivy and vines twined around the forest trees and broad-leaved ferns grew beneath their shade. Even in Spitzbergen, as far north as latitude 78 degrees 56', no less than ninety-five species of fossil plants have been obtained, including Taxodium of two species, hazel, poplar, alder, beech, plane-tree, and lime. Such a vigorous growth of trees within 12 degrees of the pole, where now a dwarf willow and a few herbaceous plants form the only vegetation, and where the ground is covered with almost perpetual snow and ice, is truly remarkable.

The identity of so many of the fossils with Miocene species of Central Europe and Italy not only proves that the climate of Greenland was much warmer than it is now, but also renders it probable that a much more uniform climate prevailed over the entire northern hemisphere. This is also indicated by the whole character of the Upper Miocene flora of Central Europe, which does not necessitate a mean temperature very much greater than exists at present, if we suppose such absence of winter cold as is proper to insular climates. Professor Heer believes that the mean temperature of North Greenland must have been at least 30 degrees higher than at present, while an addition of 10 degrees to the mean temperature of Central Europe would probably be as much as was required. The chief locality where this wonderful flora is preserved is at Atanekerdluk in North Greenland (latitude 70 degrees), on a hill at an elevation of about 1200 feet above the sea. There is here a considerable succession of sedimentary strata pierced by volcanic rocks. Fossil plants occur in all the beds, and the erect trunks as thick as a man's body which are sometimes found, together with the abundance of specimens of flowers and fruit in good preservation, sufficiently prove that the plants grew where they are now found. At Disco island and other localities on the same part of the coast, good coal is abundant, interstratified with beds of sandstone, in some of which fossil plants have also been found, similar to those at Atanekerdluk.

LOWER MIOCENE, BELGIUM.

(FIGURE 156. Leda (Nucula) Deshayesiana, Nyst.)

(FIGURE 157. Vanessa pluto; natural size. Lower Miocene, Radaboj, Croatia.)

The Upper Miocene Bolderberg beds, mentioned in Chapter 14, rest on a Lower Miocene formation called the Rupelian of Dumont. This formation is best seen at the villages of Rupelmonde and Boom, ten miles south of Antwerp, on the banks of the Scheldt and near the junction with it of a small stream called the Rupel. A stiff clay abounding in fossils is extensively worked at the above localities for making tiles. It attains a thickness of about 100 feet, and though very different in age, much resembles in mineral character the "London clay," containing, like it, septaria or concretions of argillaceous limestone traversed by cracks in the interior, which are filled with calc-spar. The shells, referable to about forty species, have been described by MM. Nyst and De Koninck. Among them Leda (or Nucula) Deshayesiana (see Figure 156) is by far the most abundant; a fossil unknown as yet in the English tertiary strata, but when young much resembling Leda amygdaloides of the London Clay proper (see Figure 213 Chapter 16). Among other characteristic shells are Pecten Hoeninghausii, and a species of Cassidaria, and several of the genus Pleurotoma. Not a few of these testacea agree with English Eocene species, such as Actaeon simulatus, Sowb, Cancellaria evulsa, Brander, Corbula pisum (Figure 157), and Nautilus (Aturia) ziczac. They are accompanied by many teeth of sharks, as Lamna contortidens, Ag., Oxyrhinaxiphodon, Ag., Carcharodon angustidens (see Figure 196 Chapter 16), Ag., and other fish, some of them common to the Middle Eocene strata.

KLEYN SPAWEN BEDS.

The succession of the Lower Miocene strata of Belgium can be best studied in the environs of Kleyn Spawen, a village situated about seven miles west of Maestricht, in the old province of Limburg in Belgium. In that region, about 200 species of testacea, marine and fresh-water, have been obtained, with many foraminifera and remains of fish. In none of the Belgian Lower Miocene strata could I find any nummulites; and M. d'Archiac had previously observed that these foraminifera characterise his "Lower Tertiary Series," as contrasted with the Middle, and they therefore serve as a good test of age between Eocene and Miocene, at least in Belgium and the North of France. (D'Archiac Monograph pages 79, 100.) Between the Bolderberg beds and the Rupelian clay there is a great gap in Belgium, which seems, according to M. Beyrich, to be filled up in the North of Germany by what he calls the Sternberg beds, and which, had Dumont found them in Belgium, he might probably have termed Upper Rupelian.

LOWER MIOCENE OF GERMANY.

RUPELIAN CLAY OF HERMSDORF, NEAR BERLIN.

Professor Beyrich has described a mass of clay, used for making tiles, within seven miles of the gates of Berlin, near the village of Hermsdorf, rising up from beneath the sands with which that country is chiefly overspread. This clay is more than forty feet thick, of a dark bluish-grey colour, and, like that of Rupelmonde, contains septaria. Among other shells, the Leda Deshayesiana, before mentioned (Figure 156), abounds, together with many species of Pleurotoma, Voluta, etc., a certain proportion of the fossils being identical in species with those of Rupelmonde.

MAYENCE BASIN.

An elaborate description has been published by Dr. F. Sandberger of the Mayence tertiary area, which occupies a tract from five to twelve miles in breadth, extending for a great distance along the left bank of the Rhine from Mayence to the neighbourhood of Manheim, and which is also found to the east, north, and south-west of Frankfort. M. De Koninck, of Liege, first pointed out to me that the purely marine portion of the deposit contained many species of shells common to the Kleyn Spawen beds, and to the clay of Rupelmonde, near Antwerp. Among these he mentioned Cassidaria depressa, Tritonium argutum, Brander (T. flandricum, De Koninck), Tornatella simulata, Aporrhais Sowbyi, Leda Deshayesiana (Figure 156), Corbula pisum, (Figure 158) and others.

LOWER MIOCENE BEDS OF CROATIA.

The Brown Coal of Radaboj, near Angram in Croatia, not far from the borders of Styria, is covered, says Von Buch, by beds containing the marine shells of the Vienna basin, or, in other words, by Upper Miocene or Falunian strata. They appear to correspond in age to the Mayence basin, or to the Rupelian strata of Belgium. They have yielded more than 200 species of fossil plants, described by the late Professor Unger. These plants are well preserved in a hard marlstone, and contain several palms; among them the Sabal, Figure 151, and another genus allied to the date-palm Phoenicites spectabilis. The only abundant plant among the Radaboj fossils which is characteristic of the Upper Miocene period is the Populus mutabilis, whereas no less than fifty of the Radaboj species are common to the more ancient flora of the Lower Molasse of Switzerland.

The insect fauna is very rich, and, like the plants, indicates a more tropical climate than do the fossils of Oeningen presently to be mentioned. There are ten species of Termites, or white ants, some of gigantic size, and large dragon- flies with speckled wings, like those of the Southern States in North America; there are also grasshoppers of considerable size, and even the Lepidoptera are not unrepresented. In one instance, the pattern of a butterfly's wing has escaped obliteration in the marl-stone of Radaboj; and when we reflect on the remoteness of the time from which it has been faithfully transmitted to us, this fact may inspire the reader with some confidence as to the reliable nature of the characters which other insects of a more durable texture, such as the beetles, may afford for specific determination. The Vanessa above figured retains, says Heer, some of its colours, and corresponds with Vanessa Hadena of India.

Professor Beyrich has made known to us the existence of a long succession of marine strata in North Germany, which lead by an almost gradual transition from beds of Upper Miocene age to others of the age of the base of the Lower Miocene. Although some of the German lignites called Brown Coal belong to the upper parts of this series, the most important of them are of Lower Miocene date, as, for example, those of the Siebengebirge, near Bonn, which are associated with volcanic rocks.

Professor Beyrich confines the term "Miocene" to those strata which agree in age with the faluns of Touraine, and he has proposed the term "Oligocene" for those older formations called Lower Miocene in this work.

LOWER MIOCENE OF ITALY.

In the hills of which the Superga forms a part there is a great series of Tertiary strata which pass downward into the Lower Miocene. Even in the Superga itself there are some fossil plants which, according to Heer, have never been found in Switzerland so high as the marine Molasse, such as Banksia longifolia, and Carpinus grandis. In several parts of the Ligurian Apennines, as at Dego and Carcare, the Lower Miocene appears, containing some nummulites, and at Cadibona, north of Savona, fresh-water strata of the same age occur, with dense beds of lignite inclosing remains of the Anthracotherium magnum and Anthracotherium minimum, besides other mammalia enumerated by Gastaldi. In these beds a great number of the Lower Miocene plants of Switzerland have been discovered.

LOWER MIOCENE OF ENGLAND— HEMPSTEAD BEDS.

We have already stated that the Upper Miocene formation is nowhere represented in the British Isles; but strata referable to the Lower Miocene period are found both in England, Scotland, and Ireland. In the Hampshire basin these occupy a very small superficial area, having been discovered by the late Edward Forbes at Hempstead near Yarmouth, in the northern part of the Isle of Wight, where they are 170 feet thick, and rich in characteristic marine shells. They overlie the uppermost of an extensive series of Eocene deposits of marine, brackish, and fresh-water formations, which rest on the Chalk and terminate upward in strata corresponding in age to the Paris gypsum, and containing the same extinct genera of quadrupeds, Palaeotherium, Anoplotherium, and others which Cuvier first described. The following is the succession of these Lower Miocene strata, most of them exposed in a cliff east of Yarmouth:

(FIGURE 158. Corbula pisum. Hempstead Beds, Isle of Wight.)

(FIGURE 159. Cyrena semistriata. Hempstead Beds.)

1. The uppermost or Corbula beds, consisting of marine sands and clays, contain Voluta Rathieri, a characteristic Lower Miocene shell; Corbula pisum (Figure 158), a species common to the Upper Eocene clay of Barton; Cyrena semistriata (Figure 159), several Cerithia, and other shells peculiar to this series.

(FIGURE 160. Cerithium plicatum, Lam., Hempstead.)

(FIGURE 161. Cerithium elegans. Hempstead.)

(FIGURE 162. Rissoa Chastelii, Nyst, sp. Hempstead, Isle of Wight.)

(FIGURE 163. Paludina lenta. Hempstead Bed.)

2. Next below are fresh-water and estuary marls and carbonaceous clays in the brackish-water portion of which are found abundantly Cerithium plicatum, Lam. (Figure 160), Cerithium elegans (Figure 161), and Cerithium tricinctum; also Rissoa Chastelii (Figure 162), a very common Kleyn Spawen shell, and which occurs in each of the four subdivisions of the Hempstead series down to its base, where it passes into the Bembridge beds. In the fresh-water portion of the same beds Paludina lenta (Figure 163) occurs; a shell identified by some conchologists with a species now living, Paludina unicolor; also several species of Lymneus, Planorbis, and Unio.

3. The next series, or middle fresh-water and estuary marls, are distinguished by the presence of Melania fasciata, Paludina lenta, and clays with Cypris; the lowest bed contains Cyrena semistriata (Figure 159), mingled with Cerithia and a panopaea.

4. The lower fresh-water and estuary marls contain Melania costata, Sowerby, Melanopsis, etc. The bottom bed is carbonaceous, and called the "Black band," in which Rissoa Chastelii (Figure 162), before alluded to, is common. This bed contains a mixture of Hempstead shells with those of the underlying Upper Eocene or Bembridge series. The mammalia, among which is Hyopotamus bovinus, differ, so far as they are known, from those of the Bembridge beds. Among the plants, Professor Heer has recognised four species common to the lignite of Bovey Tracey, a Lower Miocene formation presently to be described: namely, Sequoia Couttsiae, Heer; Andromeda reticulata, Ettings.; Nelumbium (Nymphoea) doris, Heer; and Carpolithes Websteri, Brong. (Pengelly, preface to The Lignite Formation of Bovey Tracey page 17. London 1863.) The seed-vessels of Chara medicaginula, Brong, and Chara helicteres are characteristic of the Hempstead beds generally.

The Hyopotamus belongs to the hog tribe, or the same family as the Anthracotherium, of which seven species, varying in size from the hippopotamus to the wild boar, have been found in Italy and other part of Europe associated with the lignites of the Lower Miocene period.

LIGNITES AND CLAYS OF BOVEY TRACEY, DEVONSHIRE.

Surrounded by the granite and other rocks of the Dartmoor hills in Devonshire, is a formation of clay, sand, and lignite, long known to geologists as the Bovey Coal formation, respecting the age of which, until the year 1861, opinions were very unsettled. This deposit is situated at Bovey Tracey, a village distant eleven miles from Exeter in a south-west, and about as far from Torquay in a north-west direction. The strata extend over a plain nine miles long, and they consist of the materials of decomposed and worn-down granite and vegetable matter, and have evidently filled up an ancient hollow or lake-like expansion of the valleys of the Bovey and Teign.

The lignite is of bad quality for economical purposes, as there is a great admixture in it of iron pyrites, and it emits a sulphurous odour, but it has been successfully applied to the baking of pottery, for which some of the fine clays are well adapted. Mr. Pengelly has confirmed Sir H. De la Beche's opinion that much of the upper portion of this old lacustrine formation has been removed by denudation. (Philosophical Transactions 1863. Paper by W. Pengelly F.R.S. and Dr. Oswald Heer.)

At the surface is a dense covering of clay and gravel with angular stones probably of the Post-pliocene period, for in the clay are three species of willow and the dwarf birch, Betula nana, indicating a climate colder than that of Devonshire at the present day.

Below this are Lower Miocene strata about 300 feet in thickness, in the upper part of which are twenty-six beds of lignite, clay, and sand, and at their base a ferruginous quartzose sand, varying in thickness from two to twenty-seven feet. Below this sand are forty-five beds of alternating lignite and clay. No shells or bones of mammalia, and no insect, with the exception of one fragment of a beetle (Buprestis); in a word, no organic remains, except plants, have as yet been found. These plants occur in fourteen of the beds— namely, in two of the clays, and the rest in the lignites. One of the beds is a perfect mat of the debris of a coniferous tree, called by Heer Sequoia Couttsiae, intermixed with leaves of ferns. The same Sequoia (before mentioned as a Hempstead fossil) is spread through all parts of the formation, its cones, and seeds, and branches of every age being preserved. It is a species supplying a link between Sequoia Langsdorfii (see Figure 153) and Sequoia Sternbergi, the widely spread fossil representatives of the two living trees Sequoia sempervirens and Sequoia gigantea (or Wellingtonia), both now confined to California. Another bed is full of the large rhizomes of ferns, while two others are rich in dicotyledonous leaves. In all, Professor Heer enumerates forty-nine species of plants, twenty of which are common to the Miocene beds of the Continent, a majority of them being characteristic of the Lower Miocene. The new species, also of Bovey, are allied to plants of the older Miocene deposits of Switzerland, Germany, and other Continental countries. The grape-stones of two species of vine occur in the clays, and leaves of the fig and seeds of a water-lily. The oak and laurel have supplied many leaves. Of the triple-nerved laurels several are referred to Cinnamomum. There are leaves also of a palm of which the genus is not determined. Leaves also of proteaceous forms, like some of the Continental fossils before mentioned, occur, and ferns like the well-known Lastraea stiriaca (Figure 154), displaying at Bovey, as in Switzerland, its fructification.

The croziers of some of the young ferns are very perfect, and were at first mistaken by collectors for shells of the genus Planorbis. On the whole, the vegetation of Bovey implies the existence of a sub-tropical climate in Devonshire, in the Lower Miocene period.

SCOTLAND: ISLE OF MULL.

In the sea-cliffs forming the headland of Ardtun, on the west coast of Mull, in the Hebrides, several bands of tertiary strata containing leaves of dicotyledonous plants were discovered in 1851 by the Duke of Argyll. (Quarterly Geological Journal 1851 page 19.) From his description it appears that there are three leaf-beds, varying in thickness from 1 1/2 to 5 1/2 feet, which are interstratified with volcanic tuff and trap, the whole mass being about 130 feet in thickness. A sheet of basalt 40 feet thick covers the whole; and another columnar bed of the same rock, ten feet thick, is exposed at the bottom of the cliff. One of the leaf-beds consists of a compressed mass of leaves unaccompanied by any stems, as if they had been blown into a marsh where a species of Equisetum grew, of which the remains are plentifully imbedded in clay.