canine, the Machairodus among the former, also hyænas, and a suburs ine form called the Hyænarctos, 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.

That some of the reptiles should, as well as many of the shells, have survived from the Upper Miocene to the human epoch, need scarcely excite surprise, for we have no reason to assume that the mean temperature of India in the Miocene period differed materially from that which now prevails; although the climate must have been greatly modified by the revolution which has since occurred in the physical geography of the district. The heat may be as great now, if not greater, than when the Sivatherium and Chalicotherium flourished.

Numerous fossils of the Siwâlik 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.

Atlantic Islands.-Something will be said of the Upper Miocene formations of marine origin in Madeira, the Canary Islands, and the Azores, when I speak, in the thirty-first chapter, of the volcanic rocks of those countries.

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, which, after examining them in 1842, I supposed to be of the age of the English Crag and faluns of Touraine.* If, chronologically speaking,

*Proceed. of the Geol. Soc., vol. iv. Pt. 3, 1845, p. 547.

they can be truly said to be the representatives of these two European formations, they may range in age from the Older Pliocene to the Miocene epoch, according to the classification of European strata adopted in this chapter.

The proportion of fossil shells agreeing with recent, out of 147 species collected by me, amounted to about 17 per cent., or one-sixth of the whole; but as the fossils so assimilated were almost always the same as species now living in the neighboring Atlantic, the number may hereafter be augmented, when the recent fauna of that ocean is better known. In different localities, also, the proportion of recent species varied considerably.

On the banks of the James River, in Virginia, about twenty miles below Richmond, in a cliff about 30 feet high, I observed yellow and white sands overlying an Eocene marl, just as the yellow sands of the crag lie on the blue London clay in Suffolk and Essex in England. In the Virginian sands, we find a profusion of an Astarte (4. undulata, Conrad), which resembles closely, and may possibly be a variety of, one of the commonest fossils of the Suffolk Crag (A. bipartita); the other shells also, of the genera Natica, Fissurella, Artemis, Lucina, Fig. 207.

Fig. 208.

Chama, Pectunculus, and Pecten, are analogous 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 13 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 fig. 208), and Venus tridacnoides, abundant in these same formations, but also some shells which, like Fulgur carica of Say and F. canaliculatus (see fig. 207), Calyptræa costata, Venus mercenaria, Lam., Modiola alandula, 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 geographi

cal distribution of mollusca date back to a period as remote as that of the Miocene strata.

Fig. 209.

Of ten species of zoophytes which I procured on the banks of the James River, one was formerly supposed by Mr. Lonsdale to be identical with a fossil from the faluns of Touraine, but this species (see fig. 209) proves on reëxamination to be different, and to agree generically with a coral now living on the coast of the United States. With respect to climate, Mr. Lonsdale regards 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.

Astrangea 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.

LOWER MIOCENE, UNITED STATES.

Nebraska. In the territory of Nebraska, on the Upper Missouri, near the Platte River, lat. 42° N., a tertiary formation occurs, consisting of white limestone, marls, and siliceous clay, described by Dr. D. Dale Owen,* in which many bones of extinct quadrupeds, and of chelonians of land or freshwater forms, are met with. Among these, Dr. Leidy describes a gigantic quadruped, called by him Titanotherium, nearly allied to the Palæotherium, but larger than any of the species found in the Paris gypsum. With these are several species of the genus Oreodon, Leidy, uniting the characters of pachyderms and ruminants also; Eucrotaphus, another new genus of the same mixed character; two species of rhinoceros of the sub-genus Acerotherium, a Lower Miocene form of Europe before mentioned; two species of Archaeotherium, a pachyderm allied to Charopotamus and Hyracotherium; also Pabrotherium, an extinct ruminant allied to Dorcatherium, Kaup; also Agriochagus of Leidy, a ruminant allied to Merycopotamus of Falconer and Cautley; and, lastly, a large carnivorous animal of the genus Machairodus, the most ancient example of which in Europe occurs in the Lower Miocene strata of Auvergne, but of which some species are found in Pliocene deposits. The tur

* David Dale Owen, Geol. Survey of Wisconsin, &c.; Philad. 1852.

tles are referred to the genus Testudo, but have some affinity to Emys. On the whole, the Nebraska formation is probably newer than the Paris gypsum, and referable to the Lower Miocene period, as above defined.

CHAPTER XVI.

EOCENE FORMATIONS.

Upper Eocene strata of England-Fluvio-marine series in the Isle of Wight and Hampshire Successive groups of Eocene mammalia-Boundary-line between Lower Miocene and Eocene-Fossils of Barton Clay-British Middle EoceneShells, nummulites, fishes, and reptiles of the Bagshot and Bracklesham bedsVegetation of Middle Eocene period-Lower Eocene strata of England-Fossil plants and shells of the London Clay proper-Strata of Kyson in Suffolk-Plastic clays and sands-Thanet sands-Eocene formations of France-Gypseous series of Montmartre and extinct quadrupeds-Fossil footprints-Calcaire grossier-Miliolites-Lower Eocene in France-Nummulitic formations of Europe, Africa, and Asia -Their wide extent--Referable to the Middle Eocene periodEocene strata in the United States-Section at Claiborne, Alabama-Colossal cetacean Orbitoidal limestone-Burr stone.

THE strata next in order in the descending series are those which I term Upper Eocene. In the accompanying map, the position of several Eocene areas is pointed out, such as the basin of the Thames,

N. B. The space left blank is occupied by secondary formations from the Devonian or old red sandstone to the chalk inclusive.

part of Hampshire, part of the Netherlands, and the country round Paris. The three last-mentioned areas contain some marine and

freshwater formations, which have been already spoken of as Lower Miocene, but their superficial extent is insignificant, except in the Paris basin between the Seine and the Loire.

UPPER EOCENE FORMATIONS, ENGLAND.

The following table will show the order of succession of the strata found in the Tertiary areas, commonly called the London and Hampshire basins. (See also Table, p. 103, et seq.)

A. 1. Bembridge Series--North coast of Isle of Wight,
A. 2. Osborne or St. Helen's Series-ibid.,

A. 3. Headon Series--Isle of Wight, and Hordwell Cliff, Hants,
A. 4. Barton Clay, Isle of Wight, and Barton Cliff, Hants, -

MIDDLE EOCENE.

B. Bagshot and Bracklesham Sands and Clays-London and Hants basins,

LOWER EOCENE.

700

C. 1. London Clay proper and Bognor beds-London and Hants basins,

C. 2. Plastic and Mottled Clays and Sands-London and Hants basins,

350 to 500

100

C. 3. Thanet Sands--Reculvers, Kent, and Eastern part of London basin,

90

The true relative position of the Hempstead beds and of the Bembridge, A. 1, and the Osborne or St. Helen's series, A. 2, were not made out in a satisfactory manner till Professor Forbes studied them in detail in 1852. The true place of the Bagshot sands, B., and of the Thanet sands, C. 3, was first accurately ascertained by Mr. Prest wich in 1847 and 1852.

UPPER EOCENE, ENGLAND.

Bembridge series, A. 1.—These beds are about 120 feet thick, and, as before stated (p. 239), are conformable with the Hempstead beds, near Yarmouth, in the Isle of Wight. They consist of marls, clays, and limestones of freshwater, brackish, and marine origin. Some of the most abundant shells, as Cyrena semistriata var., and Paludina lenta, fig. 176, p. 240, are common to this and to the overlying Hempstead series; but the majority of the species are distinct. The following are the subdivisions described by Professor Forbes :