A NEAELY COMPLETE CARAPACE OF A FOSSIL TUR 1 LF,AMYDA VIRGINIANA (CLARK)
By W. Gardner LynnOf the Johns Hopkins University
The species Amyda virginiana was first described under the nameTrionyx virginianus by Dr. William Bullock Clark in 1895 fromfragments found at Aquia Creek, Va., in Eocene deposits of the AquiaCreek stage. On a recent trip to this same locality a nearly com-plete carapace referable to this species was obtained; and since thisis, so far as can be ascertained, the most complete specimen of thislarge fossil turtle yet discovered it seems worthy of description.The type fragments of the species were described by Clark asfollows
:
Fragments of costals with tuberculated surfaces characteristic of the genusTrionyx. The longitudinal ridges are prominent, at times irregular and in-osculate ; relatively remote and separated by intervals about twice their width,generally entirely disappear near the margins of the plates.A number of fragments of the plates of this large species were found in thevicinity of Aquia Creek, Va. This species shows some points of similarity withT. cariosus (Cope), from the Eocene of New Mexico, but is undoubtedly adifferent form.Dimensions.?Length of largest fragment, 130mm. ; width, 45mm. ; thick-ness, 18mm,Hay, in his Fossil Turtles of North America, gives more detaileddescriptions and measurements of the 2-type fragments which, hebelieves, indicate a possibilitj^ that the two represent distinct species.This idea is based upon the fact that the sculpturing of the two frag-ments differs. The first fragment (the distal portion of a costal)(pi. 2, center) shows rather regular sculpturing, which consists ofridges and grooves, five of which are found in a line 22 mm. long.The sculpturing of the second fragment (the proximal portion oi: acostal) (pi. 2, lower) is more irregular and the pits are somewhatlarger, five being contained in a line 25 mm. long. Moreover, Haynotes resemblances to Amyda pennata (Cope) of the Eocene of NewJersey, although he does not seem to believe that the two are synony-mous. However, it is clear that the position and even the authen-ticity of this little-known species is much in question, and it isNo. 2823.?Proceedings U. S. National Museum. Vol. 76, Art. 26] ;.74736?29
PROCEEDIN-QS OF THE NATIONAL MUSEUM VOL. 7Gbelieved that the information obtained from the specimen here de-scribed will be of advantage in clearing up these points.The present specimen (Cat. No. 11944, U.S.N.M.) consists of acarapace which is complete, except for the distal portions of thefourth, fifth, sixth, and seventh costals of the left side and small por-tions of the neurals. The carapace is broad, rovmded in front, andsomewhat truncate behind. At the free margins of the costals theupper layers of bone project somewhat beyond the lower layers, caas-ing a longitudinal grooving of the carapace; beyond this the marginsdrop off gently to a thin edge. This was an exceedingly large turtle
;
the length of the carapace measured in a straight line is 735 mm., itsgreatest width G40 mm., not including the extension of the ribs be-yond the margin of the shell. It is composed of a nuchal plate, 7neurals, and 8 pairs of costals, the 2 posterior pairs meeting in themid line.The nuchal measures 380 mm. across and 87 mm. in an antero-posterior direction in the mid line, narrowing toward the outer ends.It is 27 mm. thick in the central portions and tapers off to a thin edgeat the margins. Little ornamentation is apparent on this bone.The measurements of the costals of the right side (where all arecomplete) are given in the following table
:
Costal No.
ART. 26 A FOSSIL TUETLE LYNN S
angles to the sutures, the grooves sometimes broken up into pitsby cross-ridges. In the best-preserved portions of the carapace thissculpturing is continued quite to the beveled margin of the shell.The width of ridges and grooves is variable, being in general greatertoward the posterior end. Thus on the first costal plate five ridgesand five grooves are contained in a line 18 mm. long; on the fourthcostal the same number is contained in a line 22 mm. long; and onthe eighth costal a line 28 mm. long is required. Moreover, theridges run more irregularly and are more broken up on the anteriorcostals than on the posterior ones; and in all the costals the sculp-turing is in general much more regular toward the distal ends.Plate 2, upper, shows the sculpturing on a large fragment fromthe distal end of the fifth costal of the right side. The ridges andpits on the neurals are extremely irregular in arrangement, produc-ing a reticulate apjDearance. These facts indicate that the typefragments do belong to a single species, the difference in sculptur-ing described by Hay being attributable to the differences normallypresent between distal and proximal j)ortions of the plates.Moreover, little doubt remains as to the authenticitj^ of the species.Hay's description of Amyda {Triom/x) cariosa (Cope) shows thatit differs considerably not only in length and thickness but also inthe sculpturing, for in Amyda cariosa the ornamentation consistschiefly of irregularly arranged pits, whereas in Amyda virginianuslong longitudinal grooves predominate. The chief difference betweenAmyda {Tr'ionyx) fennata (Cope) and the specimen under con-sideration is, as Hay remarks, that in the former the pits " arearranged in rows that run from the sutural edges toward the middleof the bone and at the same time toward the distal end." This isquite different from the condition in Amyda mrgimanus^ where theridges show no tendency to run toward the distal ends of the bone.Moreover, the fragments of Amyda pennata indicate that it was amuch smaller turtle than was the one represented by the presentspecimen. However, the known fragments of Amyda fennata areso small and so few that it is impossible to clear up this point withabsolute finality, although evidence thus far available seems to indi-cate that this species also is distinct from Amyda virginianus.BIBLIOGRAPHYClark, William B.Johns Hopkinp ^jniversity Circular, vol. 15, No. 4, p. 4, 1895.U. S. Geol. Survey Bull. 141, p. 59, pi. 8, figs, la and 16, 1896.Hay, Olives: P.U. S. Geol. Survey Bull. 179, p. 455, 1902.Fossil Turtles of North America, p. 515, pi. 96, figs. 7 and 8, text figures,670 and 671, 1908.
4 PROCEEDINGS OF THE NATIONAL MUSEUM voi i6EXPLANATION OF PLATESPlate 1
Leift: Ventral view of the carapace described above, X %.Right: Dorsal view, X Ys. A foot rule is shown at the side.Plate 2Upper: A fragment forming the distal end of the fifth costal of the rightside to show the sculpturing, X %.Center: One of the type specimens. Distal portion of a costal plate, X 1.Museum Wagner Free Institute of Science, Philadelphia.Lower: Type specimen. Proximal portion of a costal plate, X 1. WagnerFree Institute of Science, Philadelphia.
U. S. GOVERNMENT PRINTING OFFICE: I92?

U. S. NATIONAL MUSEUM PROCEEDINGS, VOL, 76. ART. 26 PL. 2
'"^^i:f.
'If fly -iiW ' n fHWiif fin IP -
Fragments of Costal plate of Amyda virginiana (Clark)For explanation of plate see page 4