Aurillac from the west, we pass over great heathy plains, where the sterile mica-schist is barely covered with vegetation. Near Ytrac, and between La Capelle and Viscamp, we find the surface strewed over with loose broken flints, some of them black in the interior, but with a white external coating; others stained with tints of yellow and red, and in appearance precisely like the flint gravel of our chalk districts. When heaps of this gravel have thus announced our approach to a new formation, we arrive at length at the escarpment of the lacustrine beds. At the bottom of the hill which rises before us, we see strata of clay and sand resting on micaschist; and above, in the quarries of Belbet, Leybros, and Bruel, a white limestone, in horizontal strata, the surface of which has been hollowed out into irregular furrows, since filled up with broken flint, marl, and dark vegetable mould. In these cavities we recognise an exact counterpart to those which are so numerous on the furrowed surface of our own white chalk. Advancing from these quarries, along a road made of the white limestone, which reflects as glaring a light in the sun, as do our roads composed of chalk, we reach, at length, in the neighbourhood of Aurillac, hills of limestone and calcareous marl, in horizontal strata, separated in some places by regular layers of flint in nodules, the coating of each nodule being of an opaque white colour, like the exterior of the flinty nodules of our chalk. This hard white substance has been ascertained in England to consist, in some instances, wholly of siliceous matter, and sometimes to contain a small admixture of carbonate of lime,* and_the analysis of the similar rocks in the Cantal would probably give the same results. The Aurillac flints have precisely the appearance of having separated from their matrix after the siliceous and calcareous matter had been blended together. The calcareous marl sometimes occupies small sinuous cavities in the flint; and the siliceous nodule, when detached, is often as irregular in form as those found in our chalk.

By what means, then, can the geologist at once decide that the limestone and silex of Aurillac are referable to an epoch entirely distinct from that of the English chalk? It is not by reference to position; for we can merely say of the lacustrine beds, as we should have been able to declare of the true chalk had it been present, that they overlie the granite rocks of this part of France. It is from the organic remains only that we are able to pronounce the formation to belong to the Eocene tertiary period. Instead of the marine Alcyonia of our cretaceous system, the silicified seed-vessels of the Chara, a plant which grows at the bottom of lakes, abound in the flints of Aurillac, both in those which are in situ and those forming the gravel. Instead of the Echini and marine testacea of the chalk, we find in these marls and limestones the shells of the Planorbis, and other lacustrine testacea, all of them, like the gyrogonites, agreeing specifically with species of the Eocene type.

*

Phillips, Geol. Trans., First Series, vol. v. p. 22.-Outlines of Geology, p. 95.

Proofs of the gradual deposition of marl.-Some sections of the foliated marls in the valley of the Cer, near Aurillac, attest, in the most unequivocal manner, the extreme slowness with which the materials of the lacustrine series were amassed. In the hill of Barrat, for example, we find an assemblage of calcareous and siliceous marls, in which, for a depth of at least sixty feet, the layers are so thin that thirty are sometimes contained in the thickness of an inch; and when they are separated we see preserved in every one of them the flattened stems of Charæ, or other plants, or sometimes myriads of small paludina and other freshwater shells. These minute foliations of the marl resemble precisely some of the recent laminated beds of the Scotch marl lakes, and may be compared to the pages of a book, each containing a history of a certain period of the past. The different layers may be grouped together in beds from a foot to a foot and a half in thickness, which are distinguished by differences of composition and colour, the tints being white, green, and brown. Occasionally there is a parting layer of pure flint, or of black carbonaceous vegetable matter, about an inch thick, or of white pulverulent marl. We find several hills in the neighbourhood of Aurillac composed of such materials for the height of more than 200 feet from their base, the whole sometimes covered by rocky currents of trachytic or basaltic lava.*

Thus wonderfully minute are the separate parts of which some of the most massive geological monuments are made up! When we desire to classify, it is necessary to contemplate entire groups of strata in the aggregate; but if we wish to understand the mode of their formation, and to explain their origin, we must think only of the minute subdivisions of which each mass is composed. We must bear in mind how many thin leaf-like seams of matter, each containing the remains of myriads of testacea and plants, frequently enter into the composition of a single stratum, and how vast a succession of these strata unite to form a single group! We must remember also, that volcanos like the Plomb du Cantal, which rises in the immediate neighbourhood of Aurillac, are themselves equally the result of successive accumulation, consisting of reiterated flows of lava and showers of scoriæ; and I have shown, when treating of the high antiquity of Etna, how many distinct lava-currents and heaps of ejected substances are required to make up one of the numerous conical envelopes whereof a volcano is composed.-Lastly, we must not forget that continents and mountain-chains, colossal as are their dimensions, are nothing more than an assemblage of many such igneous and aqueous groups, formed in succession during an indefinite lapse of ages, and superimposed upon each other.

* Lyell and Murchison, sur les Dépôts Lacust. Tertiaires du Cantal, &c. Ann. des Sci. Nat., Oct. 1829.

VOL. II.-2 W

CHAPTER XVIII.

EOCENE FORMATIONS-PARIS BASIN.

Marine Eocene strata-Paris basin how far analogous to deposits of Central France -Connexion of Auvergne and Paris basins-Groups in Paris basin-Observations of M. C. Prevost-Contemporaneous marine and freshwater strata-Abundance of Cerithia (p. 373.)-Upper marine formation-All the Parisian groups EoceneMicroscopic shells (p. 377.)-Bones of quadrupeds in gypsum-Strata with and without organic remains alternating-Extent of our knowledge of the physical geography, fauna, and flora of the Eocene period-Concluding remarks.

THE geologist who has studied the lacustrine formations described in the last chapter cannot enter the tract usually termed "the Paris Basin" without immediately recognising a great variety of rocks with which his eye has already become familiar. The green and white marls of Auvergne, Cantal, and Velay, again present themselves, together with limestones and quartzose grits, siliceous and gypseous marls, nodules and layers of flint, and saccharoid gypsum; lastly, in addition to all this identity of mineral character, he finds an assemblage of the same species of fossil animals and plants.

When we consider the geographical proximity of the two districts, we are the more prepared to ascribe this correspondence in the mineral composition of these groups to a combination of similar circumstances at the same era. From the Map (Fig. 144., ante, p. 359) in the last chapter, it will be seen that the united waters of the Allier and Loire, after descending from the valleys occupied by the freshwater formations of Central France, flow on till they reach the southern extremity of what is called the Paris basin. M. Omalius d'Halloy long ago suggested the very natural idea that there existed formerly a chain of lakes, reaching from the highest part of the central mountain-group of France, and terminating in the basin of Paris, which he supposes was at that time an arm of the sea.

Notwithstanding the great changes which the physical geography of this part of France must since have undergone, we may easily conceive that many of the principal features in the configuration of the country may have remained unchanged, or but slightly modified. Hills of volcanic matter have indeed been formed since the Eocene formations were accumulated, and the levels of large tracts have been altered in relation to the sea; lakes have been drained, and a gulf of the sea turned into dry land, but many of the reciprocal relations of the different parts of the surface may still remain the same. The waters which flowed from the granitic heights into the Eocene lakes may now descend in the same manner through valleys once the basins of those lakes. Let us, for illustration,

suppose the great Canadian lakes, and the gulf into which their waters are discharged, to be elevated and laid dry by subterranean movements. The whole hydrographical basin of the St. Lawrence might be upraised during these convulsions, yet that river might continue, even after so extraordinary a revolution, to drain the same elevated regions, and might still convey its waters in the same direction from the interior of the continent to the Atlantic. Instead of traversing the lakes, it would hold its course through deposits of lacustrine sand and shelly marl, such as we know to be now forming in Lakes Superior and Erie; and these freshwater strata would occupy the site, and bear testimony to the pristine existence of the lakes. Marine strata may also be brought into view in the space where an inlet of the sea, like the estuary of the St. Lawrence, had once received the continental waters; and in such formations we might discover shells of lacustrine and fluviatile species intermingled with marine testacea and zoophytes.

Subdivisions of strata in the Paris basin.—The area which has been called the Paris basin is about 180 miles in its greatest length from northeast to south-west, and about ninety miles from east to west. This space

may be described as a depression in the chalk (see Fig. 82., ante, p. 202), which has been filled up by alternating groups of marine and freshwater strata. MM. Cuvier and Brongniart attempted in 1811 to distinguish five different formations, and to arrange them in the following order, beginning with the lowest :

These formations were supposed to have been deposited in succession upon the chalk; and it was imagined that the waters of the ocean had been by turns admitted into and excluded from the same region. But the subsequent investigations of several geologists, especially of M. Constant Prevost,* have led to great modifications in the theoretical views entertained respecting the order in which the several groups were formed; and it now appears that the formations Nos. 1, 2, and 3, of the table of MM.

* Bulletin des Sci. de la Soc. Philom., May, 1835, p. 74.

Cuvier and Brongniart, instead of having originated one after the other, are divisible into four nearly contemporaneous groups.

Superposition of different formations in the Paris basin.-A comparison of the two accompanying diagrams will show at a glance the different relations which the several sets of strata bear to each other, according to the original as well as the more modern classification. I shall now proceed to lay before the reader a brief sketch of the several sets of strata referred to in the above systems.

Immediately upon the chalk a layer of broken chalk flints, often cemented into a breccia by siliceous sand, is very commonly found. These flints probably indicate the action of the sea upon reefs of chalk when a portion of that rock had emerged, and before the regular tertiary beds were superimposed. To this partial layer no reference is made in the annexed sections.

Plastic clay and sand.-Upon this flinty stratum, or, if it be wanting upon the chalk itself, rests frequently a deposit of clay and lignite (No. 1. of the above tables). It includes the remains of freshwater shells and drift-wood, and was, at first, regarded as a proof that the Paris basin had originally been filled with fresh water. But it has since been shown that this group is not only of very partial extent, but is by no means restricted to a fixed place in the series; for it alternates with the marine calcaire grossier (No. 2. of the tables), and is repeated in the very middle of that limestone at Veaugirard, Bagneux, and other places, where the same Planorbes, Paludinæ and Limneæ occur.* M. Desnoyers pointed out to me a section in the suburbs of Paris, laid open in 1829, where a similar intercalation was seen in a still higher part of the calcaire grossier.

* Prevost, Sur les Submersions Itératives, &c. Mém. de la Soc. d'Hist. Nat. de Paris, tome iv. p. 74.