mains of this age have been with more or less probability regarded as being Ornithic. The Upper Cretaceous rocks of the United States have, however, yielded the remains of undoubted birds (Laornis, Telmatornis, Scolopax, and Palæotringa). More remarkable than any of the above are the recently-discovered Hesperornis and Ichthyornis, of which the former is a large Diver-like bird, whilst the latter exhibits the singular peculiarity that the vertebræ are biconcave. Mammals have not hitherto been detected in any Cretaceous deposit.

CHAPTER LIV.

EOCENE PERIOD.

BEFORE Commencing the study of the subdivisions of the Kainozoic series, there are some general considerations to be noted. In the first place, there is a complete and entire physical break between the rocks of the Mesozoic and Kainozoic periods. In no instance are Tertiary strata to be found resting conformably upon any Secondary rock. The Chalk has invariably suffered much erosion and denudation before the lowest Tertiary strata were deposited upon it. This is shown by the fact that the actually eroded surface of the Chalk can often be seen, or, failing this, that we can point to the presence of the chalk-flints in the Tertiary strata. This last, of course, affords unquestionable proof that the Chalk must have been subjected to enormous denudation prior to the formation of the Tertiary beds, all the chalk itself having been removed, and nothing left but the flints, while these are all rolled and rounded.

In the second place, there is a complete break in the life of the Mesozoic and Kainozoic periods. With the exception of a few Foraminifera, and one Brachiopod (the latter doubtful), no Cretaceous species is known to have survived the Cretaceous period; while several characteristic families, such as the Ammonitida and Hippuritida, died out entirely with the close of the Cretaceous rocks. In the Tertiary rocks, on the other hand, not only are all the animals and plants more or less like existing types, but we meet with a constantly-increasing number of living species as we pass from the bottom of the Kainozoic series to the top. Upon this last fact is founded

the modern classification of the Kainozoic rocks, propounded by Sir Charles Lyell.

It follows from the constant want of conformity between the Cretaceous and Tertiary rocks, and still more from the entire difference in life, that the Cretaceous and Tertiary periods are separated by an enormous lapse of unrepresented time. How long this interval may have been, we have no means of judging exactly, but it very possibly was as long as the whole Kainozoic epoch itself. Some day we shall doubtless find, at some part of the earth's surface, strata which were deposited during this period, and which will contain fossils intermediate in character between the organic remains which respectively characterise the Secondary and Tertiary periods. At present, we have only slight traces of such deposits, as, for instance, the Maestricht beds.

CLASSIFICATION OF THE TERTIARY ROCKS.-The classification of the Tertiary rocks is a matter of unusual difficulty, in consequence of their occurring in disconnected basins, forming a series of detached areas, which hold no relations of superposition to one another. The order, therefore, of the

Tertiaries in point of time, can only be determined by an appeal to fossils; and in such determination Sir Charles Lyell proposed to take as the basis of classification the proportion of living or existing species of Mollusca which occurs in each stratum or group of strata. Acting upon this principle, Sir Charles Lyell divides the Tertiary series into four groups :

1. The Eocene formation (Gr. eos, dawn; kainos, new), containing the smallest proportion of existing species, and being, therefore, the oldest division. In this classification only the Mollusca are taken into account; and it was found that of these about three and a half per cent were identical with existing species.

II. The Miocene formation (Gr. meion, less; kainos, new), with more recent species than the Eocene, but less than the succeeding formation, and less than one-half the total number in the formation. As before, only the Mollusca are taken into account, and about 17 per cent of these agree with existing species.

III. The Pliocene formation (Gr. pleion, more; kainos, new), with more than half the species of shells identical with existing species; the proportion of these varying from 35 to 50 per cent in the lower beds of this division, up to 90 or 95 per cent in its higher portion.

IV. The Post-Tertiary Formations, in which all the shells belong to existing species. This, in turn, is divided into two

minor groups-the Post-Pliocene and Recent Formations. In the Post-Pliocene formations, while all the Mollusca belong to existing species, most of the Mammals belong to extinct species. In the Recent period, the quadrupeds, as well as the shells, belong to living species.

The above, with some modifications, was the original classification proposed by Sir Charles Lyell for the Tertiary rocks, and now universally accepted. More recent researches, it is true, have somewhat altered the proportions of existing species to extinct, as stated above. The general principle, however, of an increase in the number of living species, still holds good: and this is as yet the only satisfactory basis upon which it has been proposed to arrange the Tertiary deposits.

EOCENE FORMATION.

The Eocene rocks are the lowest of the Tertiary series, and comprise all those Tertiary deposits in which there is only a small proportion of existing Mollusca-from three and a half to five per cent. The Eocene rocks occur in several basins in Britain, France, the Netherlands, and other parts of Europe, and in the United States. The subdivisions which have been established are extremely numerous, and it is often impossible to parallel those of one basin with those of another. It will be sufficient, therefore, to accept the division of the Eocene formation into three great groups-Lower, Middle, and Upper Eocene-and to consider some of the more important beds comprised under these heads in Europe and in North America. I. LOWER EOCENE. The base of the Eocene series in Britain is constituted by about 90 feet of light-coloured, sometimes argillaceous sands (Thanet Sands), which are of marine origin. Above these, or forming the base of the formation where these are wanting, come mottled clays and sands with lignite (Woolwich and Reading series), which are estuarine or fluvio-marine in origin. The highest member of the Lower Eocene of Britain is the "London Clay," consisting of a great mass of dark-brown or blue clay, sometimes with sandy beds, or with layers of "septaria," the whole attaining a thickness of from 200 to as much as 500 feet. The London Clay is a purely marine deposit, containing many marine fossils, with the remains of terrestrial animals and plants; all of which indicate a high temperature of the sea and tropical or subtropical conditions of the land.

II. MIDDLE EOCENE.-The inferior portion of the Middle Eocene of Britain consists of marine beds, chiefly consisting

of sands, clays, and gravels, and attaining a very considerable thickness (Bagshot and Bracklesham beds). The superior portion of the Middle Eocene of Britain, on the other hand, consists of deposits which are almost exclusively fresh-water or brackish-water in origin (Headon and Osborne series).

The chief Continental formations of Middle Eocene age are the "Calcaire grossier" of the Paris basin, and the "Nummulitic Limestone" of the Alps.

III. UPPER EOCENE.-If the Headon and Osborne beds of the Isle of Wight be placed in the Middle Eocene, the only British representatives of the Upper Eocene are the Bembridge and Hempstead beds-though the latter are regarded by Sir Charles Lyell as being of Lower Miocene age. These strata consist of limestones, clays, and marls, which have for the most part been deposited in fresh or brackish water.

IV. EOCENE BEDS OF THE PARIS BASIN. - The Eocene strata are very well developed in the neighbourhood of Paris, where they occupy a large area or basin scooped out of the Chalk. The beds of this area are partly marine, partly freshwater in origin; and the following table (after Sir Charles Lyell) shows their subdivisions and their parallelism with the English series:

GENERAL TABLE OF FRENCH EOCENE STRATA.

V. EOCENE STRATA OF THE UNITED STATES.-In North America, Lower Eocene Rocks are extensively developed at

Claiborne, Alabama, and consist of clays, lignites, marls, and impure limestones. The fossils of the Claiborne series are much the same in their characters as those of the London clay, and the lignites contain numerous plant-remains.

The Middle Eocene group is represented in North America by lignitic clays and marls which occur at Jackson, Mississippi. Amongst the more remarkable fossils of the Jackson beds are the teeth and bones of Cetaceans of the genus Zeuglodon.

Rocks of Upper Eocene age occur in North America at Vicksburg, Mississippi, and consist of lignites, clays, marls, and limestones. On the White River they are about 1000 feet thick, and consist of clays, sandstones, and limestones, of fresh-water origin. Among their most remarkable fossils are the remains of Mammals, of which about forty species have been already determined.

LIFE OF THE EOCENE PERIOD.

Little need be added here as to the life of the Eocene period, fossils being so abundant as to render it impossible to do more than indicate some general considerations. Upon the whole, the plants and animals of the Eocene period closely resemble those now in existence upon the globe; not, however, necessarily in the exact localities in which they are now found. Thus, the modern representatives of the plants and animals of the Eocene Rocks of Europe are not to be found in Europe itself, but in some tropical or sub-tropical region. The climatic conditions of Europe in the Eocene period were very different to those at present subsisting, and the animals and plants were correspondingly different. Still, there are few Eocene fossils which have not their modern representatives in warm countries.

The Protozoa are represented in Eocene times chiefly by Foraminifera, which are often extraordinarily abundant, and the shells of which may in some cases be said without exaggeration to compose whole mountain-masses. The great and widespread formation of the Nummulitic Limestone is largely made up of the shells of Nummulites and Orbitoides, some of the former occasionally reaching a size of more than an inch. in diameter. One of the limestones of the Paris basin is largely composed of the shells of a species of Miliola. Numerous other forms occur; and the genus Nummulites is exclusively confined to this formation.

Corals are not particularly abundant in the Eocene rocks, but they are mostly of types identical with, or nearly allied to,