basicranial area is not known to me to occur so distinctly in any species of Cyclopidius where the skull characters have been described. The palatonarial border is opposite the posterior lobe of M3. In Cyclopidius emydinus it is nearly halfway between the glenoid articular surface and M3. The palate is slightly concave.

The transverse diameter of the posterior lobe of M3 is but a trifle greater than that of P3. The tooth rows are not parallel, as in Oreodon and most of the genera of this family, but converge anteriorly. The canine is situated inwardly of the line of the tooth row and is nearly circular in section, being of somewhat greater diameter at the internal face. As stated above, the incisor border is missing, but the amount of space available for incisors was not more than would accommodate two unless they were extremely small. However, I am inclined to believe that it had but two of these teeth.

The infra-orbital foramen is single and above the interval between P3 and P. The external auditory meatus is directed more posteriorly than outwardly, and situated posterior to a line above the paramastoid process.

Measurements of Holotype.

mm.

Skull, length, oceip. condyles to canine inc., approx.

93

Bizygomatic diam.

82

Cope, E. D. 1878A. New artiodactyles of the upper Tertiary. Amer. Nat., 12, 58.

1878B. Descriptions of new Vertebrata from the upper Tertiary formations of the West. Proc. Amer. Philos. Soc., 17, 219-231. 1884. Synopsis of the species of Oreodontidæ. Ibid., 21, 503-572. Leidy, J.

1856A. Notices of remains of extinct Mammalia, discovered by Dr. F. V. Hayden in Nebraska Territory. Proc. Acad. Nat. Sei. Phila., 8, 88-90.

- 1856B. Notice of some remains of extinct vertebrated animals. Ibid., 8, 163-165.

Leidy, J.

1858.

Notice of remains of extinct Vertebrata, from the valley

of the Niobrara River. Ibid., 10, 20-29.

1869. The extinct mammalian fauna of Dakota and Nebraska. Jour.

Acad. Nat. Sci. Phila. (2), 7, 1-472.

Loomis, F. B. 1920. On Ticholeptus rusticus and the genera of Oreodontidæ. This Journal (4), 50, 281-292.

Matthew, W. D. 1899. A provisional classification of the fresh-water Tertiary of the West. Bull. Amer. Mus. Nat. Hist., vol. 12, 19-75.

1909. Faunal lists of the Tertiary Mammalia of the West. Survey, Bull. 361, 91-120.

U. S. Geol.

Scott, W. B. 1890. Beiträge zur Kenntniss der Oreodontidæ. Morpholog.

Jahrbuch, 16, 319-395.

1893. The mammals of the Deep River beds. Amer. Nat., 27, 659-662. 1895. The Mammalia of the Deep River beds. Trans. Amer. Philos.

Soc., 18, 55-185.

1899. The selenodont artiodactyls of the Uinta Eocene. Trans. Wagner Free Inst. Sci., Phila., 6, 1-121.

Sinclair, W. J. 1910. The restored skeleton of Leptauchenia decora. Proc. Amer. Philos. Soc., 49, 196-199.

ART. XXVIII.-A Potamogeton from the Upper Creta ceous; by EDWARD W. BERRY.

Since in some cases it is not practicable to determine fossil foliar remains conclusively, those cases in which this is possible merit emphasis lest they be buried in large systematic works on paleobotany and thus run the risk of escaping the attention of botanists and those authors who speculate on the existing distribution of plants while ignoring that of their ancestors.

The conclusive characters of the Potamogeton described in the present note merit calling attention to it in advance of my account of the associated flora, which may be long delayed. Relics of aquatic vegetation are generally much less durable than those of terrestrial forms and hence there is a relative paucity of such types in the geological record. Circumstantial evidence, as for example the delicate dental armature of Trachodon and other herbivorous dinosaurs, seemingly demands a soft plant food such as is furnished by aquatic plants, but the paleobotanist has but little to offer to the zoologist in answer to questions of this sort.

There is then a special interest attaching to aquatic fossil plants. In the very large American Upper Cretaceous floras the bulk of the fossils represent conifers and dicotyledonous leaves. Monocotyledons, aside from palms, are exceedingly rare, and this is usually considered as due to the imperfection of the record rather than as an actual portrayal of the true facts.

That the secondarily acquired aquatic adaptation in the angiosperms had already progressed a considerable distance before the close of the Upper Cretaceous is indicated by a number of rare forms found in deposits of estuary or lagoonal muds toward the upper part of the Mississippi embayment in Ripley time. These comprise the species of Potamogeton described below: A second species of Potamogeton, as yet undescribed since it may represent the submerged leaves of the former: a form referred to the genus Alismaphyllum: another that suggests the existing genus Hydrilla of the family Hydrocharitaceae: and a fifth of unknown identity, which is believed to have had the form and habit of Vallisneria, and to belong also to the Hydrocharitaceæ.

The new species of Potamogeton may be described as follows:

Potamogeton perryi sp. nov.

Leaves of relatively large size, elliptical to oblanceolate in general outline; with a rounded or emarginate apex and an abruptly or gradually narrowed base, decurrent on the broad flat petiole. Leaf substance thin, but firm and not membranous. Length ranging from 7 cm. to 10 cm. Maximum width ranging from 1.8 cm. to 2.6 cm.

FIGS. 1-3.-Potamogeton perryi. Natural size.

Petiole broad and thin, 1.5 cm. in length and about 3 mm. in width, slightly curved, made up of a row of parallel vascular bundles. Midvein similar to the petiole, i. e. broad and flat, not prominent, becoming much attenuated distad by the divergence of vascular bundles. Secondaries thin, immersed in the leaf substance, three or more acrodrome pairs, connected by thin oblique veinlets.

Named for the collector, Mr. E. S. Perry, who obtained. it at the Perry Place in Henry County, Tennessee, from beds of Ripley age.

This is an exceedingly well-marked species and so modern in its facies as to be distinguished with difficulty from numerous existing species, such as the European Potamogeton rufescens, or the North American Potamogeton nuttallii, alpinus, lonchitis, and lucens. Two of these latter are also Old World forms, namely P. alpinus Balbis, and P. lucens Linné, and it is these two that are most similar to the fossil, particularly in the shortening of the petiole. Both are prevailingly pond rather than stream forms, and range on this continent from the Atlantic to the Pacific, and from Canada to Florida and Mexico in the case of P. lucens, although P. alpinus does not go so far to the southward.

Characteristic Cretaceous forms of Potamogeton are rare, and it would seem to be something more than a coincidence that several well-marked species appear in the geological record at about the same time in rather widely separated regions. These are, in addition to the present form, Potamogeton cretaceous Heer1 from the Patoot beds of western Greenland-a form very similar to P. perryi; Potamogeton middendorfensis Berry from the Middendorf arkose member of the Black Creek formation in South Carolina; and Potamogeton ripleyensis Berry from the Ripley formation in western Tennessee.

Strangely enough the genus has not been authentically determined from Europe in strata earlier than the Oligocene, although allied forms occur in the Eocene of France. Nor are there any pre-Miocene records from Asia, the last due no doubt to the scarcity of known earlier plant beds on the latter continent.

There are upward of two score described fossil species, which are not uncommon throughout the Tertiary of the Northern Hemisphere. Many still existing species, represented by both leaves and fruits, appear in the Pleistocene records.

The existing species number more than three score, and they are present in both tropical and temperate regions, but are more varied in the latter. Most of the species

1

1 Heer, O., Fl. Foss. Aret., vol. 7, p. 19, pl. 55, figs. 23, 24, 1883.

Berry, E. W., U. S. Geol. Survey Prof. Paper 84, p. 27, pl. 4, fig. 6, 1914, 3 In MS.