edge of the ectopterygoids in Pariasaurus and Pariotichus, but not in Elginia or Empedias. A character of the American genera is the weakness of the attachment of the basioccipital to the sphenoid. The basioccipital is lost from the only known specimen of Elginia, and the sphenoid projects freely below it in Pariasaurus, The roof of the mouth in this order is a good deal like that of the Lacertilia, lacking the palatine foramen.

The order Cotylosauria was defined by me in the AMERICAN NATURALIST for 1880, p. 304, and in 1889 (October). In 1889 (Transac. Roy. Soc. London, p. 292), Prof. Seeley gave it the name Pariasauria. In my Syllabus of Lectures on Vertebrate Paleontology (1891, p. 38), I arranged the group as a suborder of the Theromora. In 1892 (Trans Amer. Philos. Soc., p. 13, Pl. I), I again regarded the Cotylosauria as an order, and described the characters of the skull in three of the genera, and gave figures of them.

Seeley has objected to the reference of the genera Pariasaurus and Empedias to the same order, on the ground that the elements connecting the supraoccipital and the quadrate rest on the occipital elements in the latter, while they are elevated above them in the former. This character would not, however, define orders, as both conditions are found in Lacertilia; but might distinguish families within an order. However, Seeley's description and figure of the occipital region in Pariasaurus bainii show that the structure only differs from that of the Diadectidae in the presence of a large foramen between the supraoccipital and exoccipital bones on each side.

The known species of the Cotylosauria range in dimensions from that of the South American Caimans (Chilonyx, Pariasaurus sp.) to that of the smaller Lacertilia, e. g., Eumeces quinquelineatus (Isodectes and Pariotichus sp.). They range from the Coal Measures to the Trias, inclusive, and have been found in South Africa, North America and Scotland. A single genus has been found in the Coal Measures of Ohio, which is represented by a species which I called Tuditanus punctulatus. It is of small size, and as the maxillary teeth are of equal length, I cannot distinguish it from Isodectes, which belongs to the Pariotichidæ. The other species which were referred to Tuditanus are Stegocephalia. This is the first identification of a true reptile in the Coat Measures.

Philos. Transac. Roy. Soc. 1892, p. 326, Pl. XVIII, Fig. 2.

6Transac. Amer. Philosoph. Society, April, 1874, separate p. 11. Report Geol. Survey of Ohio, 1875, Paleontology, p. 302, Plate XXIV, fig. 1 (erroneously named in explanation Tuditanus longipes).

'Proceeds. Amer. Philos. Soc., 1871, p. 177.

This order embraces, at present, four families, comprising 24 species distributed among 12 genera, as follows: Elginiidae, 1 genus, 1 species; Pariasaurida, 3 genera, 7 species; Diadectidæ, 5 genera, 9 species; Pariotichida 6 genera (of which 3 are new, viz.: Isodectes, Captorhinus and Hypopnous), and 12 species, of which 5 are new. Total, 29 species, 15 genera.-E. D. COPE.

EXPLANATION OF PLATE VIIa.

Pariotichus aguti Cope. From the Proceeding Amer. Philos. Society, November, 1895. Fig. 1, Skull, from side. Fig. 2, Skull, with angular parts of mandible adherent, cervical vertebrae and scapular arch, from below. Fig. 3, Skull, from above, with cervical vertebrae. Fig. 4, Anterior two-thirds of mandibular arch, with adherent premaxillary bones, from above. Fig. 5, Humerus. N., Nasal bone; F., Frontal; Pef., Prefrontal; Pof., Postfrontal; P., Parietal; Pmx., Premaxillary; Mx., Maxillary; J., Jugal; Qj., Quadratojugal; St., Supratemporal; Sm., Supramastoid; Tab., Tabulare; So., Supraoccipital; V., Vomer; Pa., Palatine; Par., Paroccipital; Ecp., Ectopterygoid; Ps., Pterygoid; Q., Quadrate; Ce. Clavicle; Ep., Episternum; H., Humerus.

The Puget Group.-Sir Wm. Dawson confirms the opinion advanced by Dr. G. M. Dawson in 1890 that the formation in the north-western part of the United States to which the name Puget group has been given, extends into British Columbia as far as Burrard's Inlet. This great estuarine deposit extends southward as far as the Columbia River and from the coast line to the Cascade range, within which its beds rise to a height estimated at from 800 to 5000 feet above the level of the sea. They overlie the Cretaceous Chico series in the United States, and its equivalent the Nanaimo formation in Canada. The latest views of paleobotanists and geologists of the United States seem to be that these beds are of Eocene age and that the fossil plants may be best compared with those of the Upper Laramie of the interior plains. In so far as Canada is concerned it has been established that the Upper Laramie beds underlie a formation containing animal fossils of the White River Miocene period, so there can be no doubt as to their Eocene age, and consequently of the Eocene age of the Puget group in Canada.

A further confirmation as to this view of the age of the formation in question is found in a collection of fossil plants from the vicinity of Burrard Inlet. These were referred to Sir Wm. Dawson for identification who sums up the results of his study as follows:

"A comparison with the flora of the Upper Cretaceous Nanaimo series shows that the Burrard Inlet species are distinct and of more modern aspect. On the other hand, they are also distinct from those of older Miocene deposits of the Similkamen district and other parts of the interior of British Columbia. Between these they occupy an intermediate position; in this resect corresponding with the Laramie of the interior plains east of the Rocky Mountains. They also resemble this formation in the general facies of the flora, which is not dissimilar from that of the Upper Laramie or Fort Union group."

"We may thus refer these plants to the Paleocene or Eocene, and regard them as corresponding with those of the Atanékerdluk beds in Greenland, the lignitic series of the McKenzie River, and the beds holding similar plants in Alaska."

"This flora thus serves to fill the gap in our western series of fossil plants, namely, that between the Cretaceous and the Lower Miocene." (Trans. Roy. Soc. Can. (2), Vol. I, 1895-'96.)

The Geological Structure of Florida is according to Prof. E. T. Cox, remarkable for its simplicity. The underlying rock is a soft limestone of Upper Eocene age; resting on this are beds of phosphate of lime; and covering the phosphate and limestone is a bed of sand that varies from a few inches to 20 feet and more in depth.

The Eocene limestone is filled with fossil marine shells. It shows no evidence of disturbance and is without a trace of stratification. It has an amorphous structure and is of unknown thickness. The phosphate of lime occurs in detached masses scattered over an area about 20 miles wide, and exending in a belt, follows in general way the trend of the Gulf coast from the northern limits of the state and beyond, to the western edge of the Everglades on the south. The author believes the phosphate to be the result of the mineralization of guano. The covering of sand is found all over the Peninsula. blown by the winds from the gulf and ocean beaches. Mixed with the sand is clay in the form of fine dust. In several localities the associated clay has been separated from the sand by running water and deposited as kaolin. This kaolin has been tested and found to be of superior quality for the manufacture of the finest porcelain.

It has been

Florida is not a level plain. A ridge from 30 to 50 miles wide extends from the northern part of the state to the Everglades, having an elevation of more than 230 feet in some places. From this ridge the land slopes to the Atlantic on the east and the Gulf on the west.

The elevation of the Peninsula was due to that continental force, extended over a vast period of time, which brought the tops of the Rocky