354 



KOSSII. TURTLES OF NORTH AMERICA. 



This species is stated to be based on portions of 3 individuals, which were collected tor 

 Professor Cope by Mr. G. W. Marnock, in southwestern Texas. The beds are stated to 

 belong to the Pliocene; but now they are regarded as Pleistocene. 



In the collection of the Philadelphia Academy there is found a part of the anterior half of 

 a carapace which evidently belongs to this species and which is probably one of the three por- 

 tions referred to by Cope in his description. It is labeled in Cope's handwriting, and is stated 

 to have been collected by G .W. Marnock, from the Pliocene beds of Atascosa County, Texas. 

 Terra pene marnochii was described by Cope on the page following the description of the present 

 species and was said to come from the same formation. From the American Naturalist, 

 vol. xxni, 1889, page 1 6 1, we learn that the latter-named species was found in the Equus beds. 

 From these Equus beds we have therefore Testudo atascosa: Hay, Terrapene marnochii (Cope), 

 Trachemys bisornata (Cope) and T. trulla Hay. 



The specimen (plate 56, fig.i) in the Philadelphia Academy consists of the nuchal, the first 

 right peripheral, the first and second right costals, parts of the third and fourth right costals, 

 the first, second, fourth, and fifth neurals, and the proximal portions of the first and second 

 left costals. The third neural is missing, but its dimensions and form are easily determined. 

 The nuchal resembles that of T. scnpta. In front it is thin-edged and sharp. It is slightly 

 notcht at the ends of the sulci bounding the nuchal scute, and between the ends of these sulci 

 are 3 teeth. Its length along the midline is 49 mm.; its width anteriorly, 34 mm.; the max- 

 imum width, 55 mm. The table herewith gives the dimensions of the neurals represented. 



The thickness of the fourth neural is 7.5 mm. The first is 

 oval. The second is broadest in front and deeply excavated to 

 receive the first. The other neurals are hexagonal, with the broad 

 end in front. 



The second costal is 26 mm. wide near its neural; the third 

 is 28 mm. wide; the fourth, 23 mm. They are rather thick proxi- 

 mally, as is indicated by the thickness of the fourth neural. 



The nuchal scute is 19 mm. long and 10 mm. wide posteriorly. 

 The first vertebral is 46 mm. long, 33 mm. wide in front, 40 mm. 

 behind. The second is 52 .mm. long and 50 mm. wide. The third is 43 mm. long, 54 mm. 

 wide. The sulci are deeply imprest. 



Along the midline there appears a broad and shallow groove, beginning with the first 

 neural, showing most conspicuously on the second, and persisting on the fourth and fifth. 

 There is only a suggestion of a median carina. 



The ornamentation of the carapace has been described by Cope, and his description 

 applies to the specimen here considered. The whole surface of the carapace, as far as known, 

 is sculptured into ridges and grooves. The areas occupied by the vertebral scutes are smoothest, 

 but here too we find some sculpture. On those costal bones which are traverst by the inter- 

 costal sulci we find that the area in front of the sulcus presents sharp ridges, some of which 

 run at right angles with the sulcus, while others cross these parallel with the sulcus. Behind 

 the sulcus the ridges run only at right angles with the sulcus, and these ridges are broader than 

 those in front of the sulcus. Those costals which are not traverst by the intercostal sulci are 

 ornamented more like the anterior portion of the costal plates first described. The ornamenta- 

 tion may be otherwise described as follows: Along the front of each scutal area there is a band 

 of broad ridges. Most of the remainder of the area is occupied by ridges and grooves running 

 parallel with the length of the animal and crost by a system of less conspicuous ridges. The 

 lower border of the costal scutal area, that formed by the upper borders of the peripheral bones, 

 is occupied by coarse ridges at right angles with the length of the animal. 



The sculpture of the carapace of this tortoise has been compared by Cope to that found on 

 T. elegans. It appears to be still more like that of a large specimen of T . scnpta now in the 

 American Museum of Natural History. It can hardly be doubted that the fossil species is 

 closely related to, perhaps the immediate ancestor of, both T. elegans and T. scnpta. It has 

 differed from T. elegans in having a considerably thicker and more boldly sculptured carapace; 

 from T. scripta apparently in having a less well-defined dorsal carina. Remarks on the pos- 

 sible identity of this species with Trachemys petrolei will be found on page 353. 



