392 



FOSSIL TURTLES OF NORTH AMERICA. 



Stylemys oregonensis Leidy. 



Stylemys oregonensis, LEIDY, Proc. Acad. Nat. Sci. Phila. 1871, p. 248. COPE, Bull. U. S. Geol. Surv. 



Terrs., v, 1879, p. 55. HAY, Bibliog. and Cat. Foss. Vert. N. A., 1902, p. 451. 

 TestuJo oregonensis, LEIDY, Contrib. Ext. Vert. Fauna West. Terrs., 18, pp. 226, 340, plate xv, fig. 10. 



In the year 1871, as cited above, Dr. Leidy proposed the name Stylemys oregonensis for 

 a species of turtle whose former existence he inferred from a single neural plate, supposed then 

 to be the eighth, sent to him from the John Day deposits of Oregon, the more exact locality 

 being given as Crooked River. In the later publication cited ( 1 873) the specimen is figured and 

 briefly described, being here regarded as the third neural. On page 340 of this work the sus- 

 picion is exprest that the species is identical with that from the White River deposits, Stylemys 

 nebrascensis. Professor E. D. Cope possest in his collection a number of well-preserved 

 specimens of a turtle from the John Day River region; and these he regarded as identical with 

 that described by Dr. Leidy and with S. nebrascensis. He notes various differences, but he did 

 not find them to be constant. 



Whatever Leidy's type neural may be it is not the third neural of any known Oregon 

 species of Stylemys, his type being hexagonal, that of the John Day species of Stylemys being 

 quadrangular. It is more probably the fifth neural, but not with any certainty that of Stylemys. 

 If a Stylemys, it may belong to either or neither of the two species recognized below by the 

 present writer. 



Stylemys capax sp. nov. 

 Plates 60, 61; teit-figs. 498, 499. 



Stylemys nebrascensis, in part, COPE, Vert. Tert. Form. West, 1884, p. 770, "No. 2." 



The specimen which furnishes the description of this supposed new species was collected 

 for Professor Cope from the John Day beds of Oregon. The exact locality is said to have been 

 the "Cove," in the region of the junction of the North Fork and South Fork of the John 

 Day River, in the western part of Grant County. The level is doubtless the Middle John Day 

 or the Diceratherium beds. This author regarded this individual, as well as the other John 

 Day specimens, as identical with S. nebrascensis, from the White River deposits. A brief 

 description and some measurements of this individual were furnisht by Cope as cited above. 

 The specimen now bears the American Museum's catalog number, 1357. 

 In form this species is of moderate width and height, and both the anterior and the pos- 

 terior borders are broadly rounded. The upper surface is convex in all directions. The 

 peripherals over the limbs are little, if any, flared. The anterior lobe of the plastron is long and 

 projects somewhat beyond the carapace. The hinder lobe is deeply notcht. 



The length of the carapace (plate 60; text-fig. 498) , taken in 

 a straight line, is 455 mm.; its greatest width is 360 mm. The 

 nuchal resembles that of S. conspecta. The neurals differ from 

 those of the species just named in being somewhat broader in 

 proportion to their length. There are not so great differences 

 among the costals as regards their widths as in 5. conspecta; but 

 this condition may be due to individual variation. The table 

 herewith gives the widths of each of the costals at its upper and 

 lower ends, excepting the first and the eighth. 



The peripherals are remarkable for their thickness and the 

 obtuseness of their free borders, being in these respects quite 

 unlike those of S. conspecta. The second peripheral is 35 mm. 

 thick, measured on the sulcus between the first and second dermal scutes. The border is obtuse, 

 with only a suggestion of an edge. In the type of S. conspecta, with shell nearly as long 

 (415 mm.), the second peripheral is only 25 mm. thick, and it comes down to an acute edge. 

 The hinder peripherals have a very obtuse border, nearly resembling the border of the fleshy 

 part of one's hand. The thickness of most of these hinder peripherals is 35 mm. The cor- 

 responding ones of S. conspecta are 25 mm. thick and are sharp-edged. 



The total length of the plastron (plate 61; text-fig. 499) reaches 415 mm. The anterior 

 lobe is 147 mm. long and 215 mm. wide at the base. The lip extends beyond the border of the 



