M. E. Sismonda referred all these groups to the Eocene, considering the middle one as the equivalent of the Eocene of the Paris Basin. M. Pareto*, however, clearly showed a separation between the Flysch and the "Calcaires à Fucoides" on the one part, and the upper Nummulitic zone on the other; there is also an incontestable relation between the true Miocene and this upper zone, which overlies, at Sassello and at Carcare, lignites with Cyrena Brongniarti, Bast. (C. convexa, Brong.), and Cerithium margaritaceum, Brocc., and which he considers the equivalent of the lignites of Cadibona.

From this it results that the great Nummulitic system of the Alps and Northern Italy presents three groups of different ages, namely, the Nummulitic formations (1) of Nice, (2) of the High Alps, and (3) of Bormida, of which the two lower groups are Eocene, and the upper one Miocene.

Two or three years ago the author procured a series of fossils from the Vicentin, and while grouping them according to their localities, was struck with the difference and the independence of the fauna so separated. From an examination of these fossils he has concluded that in the Vicentin there are different beds corresponding to the different stages of the Tertiary series of Paris, namely:

1. The beds of Priabona (Valle di Boro) are the equivalents of the Biarritz, that is of the Lower, Eocene. To this level also belong, on the one side Bolca, and on the other Brandola, and some other localities of Monte Berici, and probably also Val Rovina, and Monteglosso, near Bassano, to the north-east of Salcedo.

2. San Giovanni Ilarione and San Pietro Mussolino are synchronous with the "Calcaire Grossier Inférieur."

3. Villagrande, near Ronca, is contemporaneous with the "calcaire grossier supérieur," including, perhaps, the "sables de Beauchamp."

4. Castel Gomberto (Monte Grumi, San Valentino), Montecchio maggiore (la Trinità), Monte Carloto near Monteviale, Monte Postale, and, further north, Salcedo to the north of Bregauze, and Sangonini near Monte Sumano, and to the north-east Schio, correspond exactly to the horizon of Gaas and the lower part of the "Sables de Fontainebleau."

There is between Ronca and Castel-Gomberto a considerable gap, which is filled up in the Alps by the limestones with Nummulites striata and N. contorta, the Flysch, and the "Calcaire à Fucoides." These deposits, then, become synchronous with the gypsum, and represent the Upper Eocene. The system of the valley of Bormida presents a remarkable admixture of the fossils of the Lower Miocene of Castel-Gomberto, and of the "Sables de Fontainebleau" with the middle Miocene fossils of Touraine and of the Superga. This system is, then, posterior to that of Castel-Gomberto, and as it is anterior to that of the Superga, it must be placed upon the level of the "Calcaire de Beauce," of which it constitutes the marine equivalent.

The author then gives it as his opinion that the mountain masses to the north, in the Vicentin and the Véronais, and to the south, in the Ligurian Apennines, which form the great basin of the Po, have been during the Tertiary period subjected to an oscillatory movement, which alternately elevated the northern and depressed the * Bull. Soc. Géol. France, 2e Sér., tome xii. p. 370.

southern part, and vice versá, during which the deposition of the several beds was taking place.

In the long succession of fauna and physical phenomena described by M. Hébert, it is between the Nummulitic fauna of the High Alps and that of Castel-Gomberto that the greatest difference is manifested, and it is there where M. Pareto, incorrectly as the author thinks, places the limit between the Miocene and the Eocene.

An intimate relation exists between the beds of Castel-Gomberto and Salcedo, and the Upper Nummulitic beds of the Bormida, and between these latter and the true Miocene of the Superga, and of Léognan, or of Touraine.

An altogether analogous succession unites together, in Aquitaine, Gaas and Léognan, and M. Hébert considers that at present it does not appear necessary to add to the three great divisions of the Tertiary a fourth-the Oligocene,-which is included in his Lower Miocene.

If in the north of Europe this Oligocene presents great differences from the Miocene (Middle Miocene), it only shows that the upper part of this group is not generally represented by a freshwater formation; but the continuity is found again in the south, and there it is not possible to admit the Oligocene. There is not then sufficient reason to change the ternary division of the Tertiary strata, and the best line of demarcation between the middle and lower group will remain that which, in a general manner, M. Élie de Beaumont has indicated, namely, below the Fontainebleau sands, and which the author has attempted to define more clearly by fixing it between the freshwater marls above the gypsum, and the marls with Cyrena convexa and Cerithium plicatum, &c., below the Calcaire de Brie.

A paper by M. Beyrich, published in 1858, in the "Berichte" of the Academy of Berlin*, is then noticed by M. Hébert, who says that the author has, in some instances, given to his statements a meaning which was not exactly intended. For instance, his opinion that although the sea of the Faluns of Touraine differed as much from the sea of the Fontainebleau sands, as these from the sea of the "Calcaire grossier," it was preferable to adhere to the ternary division, has been rendered altogether differently in the German text.

M. Beyrich has placed the Fontainebleau sands as the equivalent of his Middle Oligocene, and the gypsum of Montmartre as synchronous with the Lower Oligocene, which would be represented in Belgium by the lower beds of Limbourg. This classification, M. Hébert thinks, cannot be justified by facts. First of all, in the Paris basin the gypsum belongs certainly to the lower part :

(1) By the marine bands intercalated with its lower beds, of which the fauna is identical with that from the marls with Pholadomya Ludensis, which belong to the Beauchamp sands; (2) by the freshwater beds above the gypsum, the fauna of which approaches nearer to that of the Saint-Ouen limestone, situated below the gypsum, and the marine marls with P. Ludensis, than of that of the Brie limestone; (3) by the contained Mammifers, which it would be difficult to associate with those from the Beauce limestone, this being the consequence of classing the latter as Upper Oligocene, and the gypsum as Lower Oligocene.

* See also Quart. Journ. Geol. Soc. vol. xx. Part 2, Miscell. p. 5.

In the second place, the fauna of the Lower Oligocene, as far as the author was able to judge in his journey to Latdorf, seemed to be singularly analogous to the fauna, not of the first marine beds of the Fontainebleau sands, but to those of Morigny, near Étampes, superior to the marine marls of Montmartre. From Latdorf about forty species were obtained, and the author has not been able to establish the identity of one of them with Eocene forms. While admitting that the fauna of the Lower Oligocene may be anterior to that of Morigny, he considers that there is still enough margin between this zone and the gypsum in which to class it, without separating it from the Lower Miocene; for in the Paris basin, between the horizon of Morigny and the freshwater marls above the gypsum, which M. Hébert considers as the upper limit of the Eocene, there are, (1) sands and marls with Natica crassatina, Deshayesia cochlearea, Brong., &c. (2) Freshwater limestones and " meulières" of Brie. (3) Green and yellow marls with Cyrena convexa, Brong. sp., Psammobia plana, Brong. sp., Cerithium plicatum, Lam., C. trochleare, Lam., Bithynia plicata, d'Arch. and de Verneuil, sp.

There is not sufficient reason to place the Lower Oligocene below the last of these.

He

The lignites upon which the marine beds of Latdorf rest, and generally the Lower Oligocene, may, the author admits, possibly correspond to the gypsum, although it is impossible to assign a definite place to it until after the discovery in it of organic remains. also thinks that the beds of Limbourg may be ranged with the Middle and Lower Oligocene of M. Beyrich, but considers that too much haste has been shown in identifying the Lower Oligocene of Belgium and Germany with the Upper Eocene of England, especially with the Barton clay.

The idea of a general catastrophe putting an end to the Eocene fauna, which had been attributed to the author by M. Beyrich, is denied, and a more gradual operation, such as accompanies all general movements of the surface, is maintained to have been the cause. The author then gives a Table showing the correlation of the beds belonging to the Lower Miocene in France, Germany, and Belgium.

Lower
Miocene.

Upper bed

...

Beauce limestone. Upper Oligocene

Middle bed... Sands of Étampes. Middle Oligocene. Upper Limbourg.

Lower bed

Upper Eocene

Middle Eocene (up- Beauchamp sands Missing per bed)

Missing.

M. Hébert attempts to show that it is more probable that the Middle Oligocene of the basin of the Rhine, and of Switzerland is related to the beds of Tongres and of Maestricht than, as M. Bey

rich considers, to those in the environs of Orléans, and he denies that he, in conjunction with M. Renevier, ever asserted that the Nummulitic strata of the High Alps are contemporaneous with the "Sables de Fontainebleau;" but, on the contrary, that he had refererd them to the gypsum.

Lastly, M. Hébert says he is still of opinion that the Barton clay is synchronous with the lower part of the Beauchamp sands, and the Headon sands with the upper part. The upper part of the beds of Colwell Bay belong to the Fontainebleau sands, and the fossils of the gypsum are found between these two horizons in the freshwater formation of Hordwell. No new reason has been shown for classing the gypsum in the Lower Oligocene.

The conclusions at which the author has arrived are these:-that the Lower and Middle Tertiary series, regarded in the north and south of Europe, present their maximum of difference, both palæontological and stratigraphical, between the gypsum and the base of the Fontainebleau sands on the one part, and between the Flysch and the beds of Castel-Gomberto on the other; and that it is there where the limit between the Eocene and the Lower Miocene, or Oligocene, should be placed. This last is allied much more with the Miocene, properly so called, or Middle Miocene, than with the most recent Eocene beds.

At the epoch of the gypsum the sea retired from the north, where, in the opinion of the author, there does not exist any marine equivalent of the beds with Palæotherium, and advanced to the south, where it penetrated to the Alps, and occupied a part of Switzerland, but without communicating with the depression already existing in the valley of the Rhine. Afterwards, when at the commencement of the Miocene epoch, the sea extended to the north of Germany, it withdrew from the Alps to the Vicentin gulf. On both sides, it seems well proved that during the period which this study embraces, it was at this time that occurred the greatest differences in the general distribution of sea and land in Europe. [A. S.]

On the OCCURRENCE of EozooN in the PRIMARY ROCKS of EASTERN BAVARIA. By Prof. GÜMBEL,

[Ueber das Vorkommen von Eozoon in dem ostbayerischen Urgebirge. Von Herrn Gümbel. Sitzungsberichte der königl. bayer Akademie der Wis

senschaften zu München. 1866, I. Heft 1. pp. 25-70. 3 plates.]

Prof. Gümbel introduces the special subject of his memoir by a concise exposition of the facts relating to the discovery and determination of Eozoon, including its structure, mineral condition, and geological position. He then describes the geological features and relations of the older rocks of Bavaria, which he arranges in descending order as follows :—

1. Hercynian clay-slate.
2. Hercynian mica-schist.
3. Hercynian gneiss.

4. Bojic gneiss.

The Hercynian gneiss is considered to be equivalent to that of the Danube, and both are stated to abound in layers of graphite, which

the author thinks is important as proving the existence of organic life at the time of their deposition. The Hercynian and Bojic gneissformations together are held to represent the Lower Laurentian system of Canada; the Hercynian mica-schist, the Upper Laurentian or Labrador series; and the Hercynian clay-slate, the Huronian of Canada and the Cambrian of England. Prof. Gümbel then describes the mineralogical characters of the Hercynian gneiss, and afterwards shows that a stratified ophicalcite occurring near Steinhag, exhibits structures corresponding with those of the Canadian Eozoon in addition to other appearances which he also believes to be of organic origin, and some of which he compares to the sections of certain Bryozoa.

The occurrence of a second species of Eozoon, to which he gives the name Eozoon Bavaricum, has been discovered by him in a rock consisting of a granular aggregation of calcite, serpentine, and a white hornblendic mineral, arranged in flakes or stripes. The rock belongs to the Hercynian Clay-slate formation, supposed to be of Huronian or Cambrian age, and the specimens examined were obtained from near Wunseidel and Thiersheim, and between Hohenberg and the Steinberg, especially the last-named locality. It exhibits:(1) A thin band almost entirely calcareous, and traversed by a network of straight lines, or, when treated with acid, divided by band-like ribs into irregular cell-like spaces, the calcite filling which is seen to be granular. (2) Thicker calcareous portions abounding in tufts of fine tubes, exactly as in Eozoon; these tubes end at the serpentinous portions (3), which have generally the same form as in the Eozoon from Steinhag before described, but are much smaller. In decalcified examples they may be seen to possess the same vaulted margins as Eozoon; their breadth averages 1 millim. and the diameter of the tubes 01 millim. Generally these serpentine bands pass into an adjoining portion (4), of one-half the width, or less, made up of very much twisted lamellæ, consisting of serpentine or a whitish mineral, and possessing highly vaulted and deeply channelled outlines. Prof. Gümbel considers that on the whole these characters undoubtedly prove the affinity of this more recent and very much smaller form to the group Eozoon; but as the last-mentioned structure (4) differs from what has been observed in Eozoon Canadense, he gives it the distinctive name of Eozoon Bavaricum*.

In conclusion Prof. Gümbel describes certain structures which appear to indicate with greater or less probability the presence of Eozoon in the pargasite of Pargas in Finland, in the coccolite-limestone of New York, in ophicalcite from Tunaberg, in a granular limestone containing chondrodite, hornblende, and garnet, from Boden in Saxony, in a blackish serpentinous limestone from Hodrisch in Hungary, and in ophicalcite from Reichenbach in Silesia.

[H. M. J.]

* Dr. Carpenter has lately recognized the similarity of the Connemara Eczoonal structures to those of Eozoon Bavaricum (supra, Part 1, p. 228). This observation is of considerable interest when viewed in connection with the relative geological age of certain Eozoonal rocks, more especially the probable Cambrian date of Eozoon Bavaricum, the Lower Silurian position of the Connemara serpentine, and the Laurentian age of Eozoon Canadense.-EDIT.