TESTUDINID&. 449 
about 75 mm. The limbs and the region about the tail present the armor that has been already 
described. Another shell collected by Mr. Sternberg and now in the American Museum has 
a total length of 800 mm. 
Among the materials collected by Mr. C. H. Sternberg and his party, in 1877, in the Loup 
Fork beds of Decatur County, Kansas, are some portions of a large tortoise which appears to 
be specifically different from 7. orthopygia. The portions represented by determinable bones 
are the greater portion of the shoulder-girdle, the right half of the pelvis, and the femur. The 
number of the specimen is 2416 of the American Museum of Natural History. 
The length of the coracoid, from the suture in the glenoid fossa to the inner hinder angle, 
is 135 mm., the breadth of the inner border, 110 mm., and the breadth across the neck of the 
bone 37 mm. The procoracoid process is a stout bone, without distinctive characters. Only 
a portion of the scapula remains. 
The width of the pelvis (fig. 607), from the pubic border of one acetabulum to that of the 
other, was 226 mm.; the length along the midline, from the hinder border of the ischia to the 
notch in the pubes, 135 mm. The least width of the ischium mesiad of the ischial tuberosity 
is 43 mm.; laterad of the tuberosity, 28 mm. The least width of the pubis mesiad of the 
lateral ee is 32 mm.; posterior to the lateral process, 38 mm. The symphysis of the 
ischia is very thick, 22 mm.; that of the symphysis of the pubes, only 10 mm. The chief 
feature which distinguishes the pelvis from that of TZ. orthopygia is found in the lateral process 
of the pubis. In 7. orthopygia this is relatively short, but broad and comprest, with the upper 
surface convex, the lower concave, the distal end broadly rounded, and the horizontal diameter 
nearly twice the perpendicular. In the pelvis here described the process is nearly twice as 
long as in T. orthopygia, tapering, convex above, with a prominent ridge below, and the 
horizontal diameter less than the perpendicular. 
The right femur (fig. 608) is nearly complete, the head only being somewhat damaged. It 
isa Senet sleaderent element than the corresponding bone of 7. orthopygia. The table pre- 
sents the measurements of this bone 
Diameter. ' compared with those of the femur 
Seen Length _ a ~ of T. orthopygia, No. 1325, American 
of femur. | Thru Shaft Thru M N ar : 
trochanters. ° 4 tuberosities. . WEST ge atura : Istory. i 
= While the articular head of the 
sete an Moe vith femur of the form here described is 
Testudo orthopygia. 143. | 63 24.5 60 damaged so that its exact dimensions 
Undetermined form. 155 53 21 64 
can not be determined, it appears not 
— : to have been so large as that of the 
species with which it is here compared. On the other hand, the digital fossa is much larger 
than in T. orthopygia, having nearly twice the diameter. It is to be observed that the diameter 
thru the distal tuberosities of the present form is greater than in the specimen of T. orthopygza. 
The type of the latter species has the distal end ‘of the femur 63 mm. wide. 
It is, of course, possible that the materials here described belong to some already named 
species of Testudo, but this can be determined only by future research 
Testudo gilberti Hay. 
Plate 80, figs. 2-5. 
Xerobates? undata, GILBERT, Kansas Uniy. Quart., vil, ser. a, 1898, p. 143, figs. 1-4. 
Testudo gilbertit, Hay, Amer. Geologist, xxiv, 1899, p. 349; Bibliog. and Cat. Foss. Vert. N. A., 
1902, p. 451. 
The only part of this animal yet known is the skull, including the lower jaw. This was 
found in the Loup Fork deposits of Phillips County, Kansas, and was described by Mr. J. Z. 
Gilbert, who referred it with some doubt to Cope’s New Mexican species, Nerobates undata. 
The latter was based wholly on portions of the carapace, and there is no evidence that the skull 
belongs to the same species, except that the two occur in the same formation. The skull has 
suffered some injuries. ‘The squamosal and a part of the quadrate on the left side are gone, 
thus exposing the tympanic cavity. The pedicels of both quadrates are broken away, as well as 
the occipital condyle. 
29) 
