508 FOSSIL TURTLES OF NORTH AMERICA. 
This species needs to be compared with A spideretes ellipticus and A. guttatus. The form of 
the carapace is about the same in the three. It is broader than in the type of 4. ellrpticus, but 
slightly narrower than in another specimen. The pre- 
neural of the species last named is, in all the known 
specimens, considerably longer and broader than in the 
type of 4. grangert, while the neurals are broader. The 
proximal ends of the costals of the first pair are narrower 
in A. grangert than in A. ellrpticus, their width being 
contained in the length of the carapace 6.33 times in the 
former and only 5.5 times in the latter. The nuchal is 
longer, its length being contained in the length of the 
carapace two times; in A. ellt ptic us twoand a hale times. 
The sculpture of the carapace is coarser in 4. ellipticus 
than in 4. grangeri, there being in the former 4 or 5 
pits in a line 20 mm. long while in the latter there are 
from 5 to 7. In making comparisons between the pits of 
the two species, measurements must be taken on corre- 
sponding parts of the shell; 36 measurements taken on 
all parts of the carapace of 2 specimens of 4. ellipticus 
show that there are, on an average, 4.44 pits in a 20 mm. 
line; 18 measurements made on corresponding parts of 
the carapace of A. grangeri indicate an ayerage of 5.83 
pits in the distance mamede that 1s, the pits of he former 
p ; species have an average diameter of 4.5 mm. while those 
1G OO}. “As pidenetes grangert, of the latter have an average diameter of mm., a 
Carapace of type. 4. : . : 5 343s 
2 difference of a little more than 1 mm. In 4. ellipticus 
the small pits are found almost wholly on the nuchal and close to the border of the shell toward 
the front. The writer regards the finer sculpture of the specimen here described as an indi- 
cation of a distinct species. 
The sculpture of the carapace of this supposed species is apparently identical with that 
of 4. guttatus and it is possible that the specimen ought to be referred to the latter species. 
However, the type of 4. granger: has a well-developt preneural bone and no fontanel on each 
side of it; whereas, the specimen of A. guttatus has large fontanels and only a rudimentary 
preneural. 4. grangert is a strongly archt species. It is possible that the flatness of all the 
specimens of 4. guttatus is due to distortion during fossilization. 
é 
, 
‘ 
SIN 
“ 
‘ 
wr a 4 
Genus AXESTEMYS Hay. 
Axestus, Copr, Proc. Amer. Philos. Soc., x11, 1872, p. 462. (Preoccupied in entomology.) 
Axestemys, Hay, Amer. Geologist, xxiv, 1899, p. 348. 
Plastron, so far as known, without the usual pits and ridges. Nuchal bone partly separated 
from the first costals by fontanels. Probably no preneural. 
Type: A xestus byssinus Cope. 
Professor Cope based the genus 4 xestus on the species 4. byssinus, which was represented 
by the left xiphiplastron, most of the pelvis, the larger portions of both femora, both humeri 
with distal ends missing, the right scapula and eeracoil complete, a cervical vertebra, two 
sacrals, a few caudals, and some phalanges. This type belongs to the U. S. National Museum, 
but the caudal vertebra and the phalanges appear to have been lost. 
The character on which Cope relied for separating this genus from other genera of Trio- 
nychidz is found in the absence of superficial sculpture on the lower side of the plastron, and 
the presence there of a layer which “‘is marked with decussating lines of osseous deposits, as in 
woven linen.” The other characters given by him in his generic descriptions are probably 
common to most of the Trionychide. 
This layer of decussating bony fibers (plate 104, fig. 4) is also common to various fossil 
and living species of Trionychide, and may be seen in bach the carapace and the plastron of 
Platypeltis spinifera near the borders of the bones where the sculptured layer thins out. 
As Cope says, such a layer is found in the dermal scutes of the Crocodilia; and Marsh has 
