PROTOSTEGID)E, a family of mostly gigantic marine Testudinata, or uturtles,s now wholly extinct, from the mid-Cretaceous Nio brara Chalk beds of western Kansas, and the overlying Fort Pierre formation of South Dakota. These turtles lived in the great but shallow supra-continental or fepeinci Nio brara and Pierre seas. They were shell fish eaters, well able to navigate the open ocean and among the most formidable of marine reptiles. The group is of extreme osteologic interest.
The first specimens were discovered along the Smoky Hill River by Cope in 1871, and called by him Protostega gigas, being recog nized as a new family type. The great Yale specimens of a second genus Archelon were secured by Wieland in 1895 to the east of the Black Hills along the Cheyenne River in the uppermost Pierre shales near the overlap of Bad Lands (Tertiary). Ten years later the veteran fossil hunter Charles ,H. Sternberg, to whom American paleontologists owe many unique ((finds) again collected Protostegas along the Smoirg Hill River which aided greatly in completing t e details of the skull and limb organization. One of these, made the subject of a restoration in the Carnegie Museum at Pittsburgh, is truly remarkable, the flippers hav ing nearly all their elements in place.
The Yale Archelon from the Cheyenne River is only less complete, but much larger and an even finer fossil. It represents a huge turtle nearly 12 feet long with front flippers of an indicated 15-foot span. (See Plates). As a unit of size, it may be mentioned that the humerus as fossilized, and mounted, weighs It should be recalled that Dermochelys dif fers from all other turtles in the absence of a dorsal bony carapace of neurals, pleurals and marginals interlocked. This protecting roof Is instead replaced by what may be termed a pseudo-caraitace of innumerable interlocking polygonal plates, all osteodermal in character and marked by seven ridges or dorsal keels. (See Fig. 1). Under this shield the ribs are slightly expanded proximally but neither unite with each other or with neurals, which wholly fail; though oddly enough there is a large thin detached nuchal.
The osteodermal mosaic of plates also con tinues around on the nether or plastral side presenting five more of the keels. But here the
plastron is found in the underlying normal, that is subdermal position, except for the fur ther anomaly of a missing entoplastron. That this element should lack, while an unsupported nuchal persists is singular. The plastral bones, however, are slender and thin, enclosing a vast fontanel, and having every appearance of ex tensive reduction. Now, the sum of Dermoche lan characters, despite certain cranial resemb lances to other sea turtles and more remotely to Chelydra, might lead to the view that of all the Testudinata, Dermochelys is nearest to a primitive form with the ancient keels still present. But if evidence of osteodermal keels is found in fossil forms, and if additional skeletal resemblances are there recognizable, then Dermochelys is far more closely related to modern forms than at first appears to be the fact,—is in a word not the least, but the most specialized of turtles.
The latter alternative appears nearest the truth. Protostega Copei of the Niobrara has a shell as heavy as that of marine species in gen eral, but there is some plastral reduction with digitation of the separate elements. In the contemporaneous Protostega gigas rib and plastral reduction are seen to have gone still further; while in Archelon of the considerably later Pierre, carapacial reduction has proceeded very far. The ribs (see Plate) only show a small thin proximal expansion, while the neurals are nearly eliminated and ridden by a median keel of elements as truly osteodermal as those of the Dermochelan carapace. There arc moreover highly digitate marginals, with pretty clear evidence of a supramarginals row. Bearing in mind that the marginals of all Tes tudinates are dermal, this accounts for the chief of the Dermochelan keels. With (1) segmenta tion of larger into lesser interlocking dermal or epithecal elements; (2) gradual thickening of these into a heavier shell; (3) a corres pondent elimination of rib expansion and neurals; and (4) marked plastral reduction, the chief features of Dermochelys would be evolved in the late Cretaceous and Tertiary; though the actual line of descent might be a discrete one far back in geologic time.