Search

Author(s): Fallon, Bailey R.; Boessenecker, Robert W. | Abstract: The leatherback sea turtle family Dermochelyidae has an extensive evolutionary history, though it is represented by only one living species today, Dermochelys coriacea. Dermochelyid fossils occur worldwide from upper Cretaceous to Pliocene marine strata. Herein described is the first occurrence of a sea turtle from the lowermost Pliocene Purisima Formation of northern California, a single carapacial non-ridge ossicle. The ossicle exhibits external morphological and internal structural characteristics (ossicle thickness, internal layering, serrate margins) that are comparable to both the extinct genus Psephophorus and to the extant genus Dermochelys. Identification of the ossicle as cf. Psephophorus is based on examination of its thickness, internal structure, surface textures and geochronological age. This paper reports the third occurrence of leatherback sea turtle fossils from the western coast of the United States.

Psephophorus polygonusMeyer, 1847, the first fossil leatherback turtle to be named, was described on the basis of shell ossicles from the middle Miocene (MN6–7/8?) of Slovakia. The whereabouts of this material is uncertain but a slab on display at the Naturhistorisches Museum Wien is considered the neotype. We rediscovered further type locality ossicles in four European institutions, re-evaluated their gross morphology and described for the first time their microstructure by comparing them withDermochelys coriacea, the only living dermochelyid turtle. The gross morphology is congruent with that already described forP. polygonus, but with two significant exceptions: the ridged ossicles ofP. polygonusmay have a distinctly concave ventral surface as well as a tectiform shape in cross-section. They do not develop the external keel typical of many ossicles ofD. coriacea. Both ridged and non-ridged ossicles ofP. polygonusare characterized by compact diploe structures with an internal cortex consisting of a coarse fibrous meshwork, whereas the proportionately thinner ossicles ofD. coriaceatend to lose the internal cortex, and thus their diploe, during ontogeny. The ossicles of bothP. polygonusandD. coriaceadiffer from those of other lineages of amniotes whose carapace is composed of polygonal ossicles or platelets, in having growth centres situated at the plate centres just interior to the external bone surface and not within the cancellous core or closer to the internal compact layer. The new diagnosis ofP. polygonusallows us to preliminarily re-evaluate the taxonomy of some of thePsephophorus-like species. Despite some macro- and micromorphological differences, it seems likely thatPsephophoruswas as cosmopolitan as extantDermochelysand had a broadly similar ecology, with a possible difference concerning the dive depth.

The partial skeletons of two marine turtles are reported from Late Eocene strata of Northland, northern New Zealand. One, from the autochthonous Ruatangata Sandstone near Whangarei, is identified as a new species of the previously monotypic cheloniid genus Eochelone Dollo, 1903. It is larger than the European type species, E. brabantica Dollo, 1903. The new species, E. monstigris, represents a geographic range extension for Eochelone and for the group of ‘Eocene stem cheloniines’ to which it belongs, a long-distance interchange that the configuration of Late Eocene–Early Oligocene marine currents and land-sea distribution must have permitted. The other specimen, from the allochthonous Pahi Greensand of northeast Kaipara Harbour, was previously reported as a cheloniid but we believe it more likely to belong to the Toxochelyidae. Given previous records of two dermochelyids (Psephophorus and Maoriochelys), the Middle to Late Eocene of New Zealand hosted at least four species of marine turtle which a...

Publisher Summary This chapter discusses the paleogeography of marine and coastal turtles of the North Atlantic and Trans-Saharan regions. Among the chelonioid species, the small eochelyines and osteopygids, exhibit a strong provincialism, whereas much larger forms such as Allopleuron and the leathery turtle Eosphargis gigas have a cosmopolitan distribution similar to that of the extant Dermochelys coriacea. The abundance of several genera and species, such as Rhinochelys, Allopleuron hoffmani, Puppigerus camperi, and Psephophorus, are probably associated with location of paleo-feeding and paleo-nesting sites. This may be simply an artifact of preservation and discovery, and it may also reflect a greater diversity and abundance of chelonioids during the Cretaceous-Tertiary Periods. Provincialism is also evident among the large pelomedusids of this period, although Taphrophys appears to be more widespread, occupying coastal niches on both sides of the opening Atlantic Ocean and in the Mediterranean area of Tethys. The chapter also discusses the taxonomic relationships and distribution of both chelonioids and pelomedusids.

Four genera of cheloniid (Caretta, Chelonia, Eretmochelys, Lepidochelys) and one genus of dermochelyid (Psephophorus) sea turtles are recorded from the Bone Valley Formation of central Florida. The fossils derive from phosphate deposits of early Pliocene (late Hemphillian) age, 4.5-5 Ma. Although similar in age to fossils from the Yorktown Formation (Lee Creek Mine) in North Carolina, the Bone Valley sea turtle fauna appears to lack two genera found at Lee Creek (Syllomus, Procolpochelys) and contains one genus that is not found at Lee Creek (Eretmochelys). The Caretta from the two areas may not be conspecific. The fossil Chelonia, Eretmochelys, and Lepidochelys cannot be distinguished confidently from modern species, but conversely only the Lepidochelys appears obviously related to a particular modern species, L. kempi. This paper provides the first report of Eretmochelys in Pliocene deposits and the first association of Psephophorus with all living cheloniid genera except the Australian Natator. We suggest that shallow seas in the Bone Valley region provided rich feeding habitat for sea turtles in the early Pliocene in the same way that shallow water habitats in Florida and the Bahamas support a similar assemblage today. The Bone Valley phosphate mining region of central Florida long has been known as a rich source of terrestrial and marine vertebrate fossils (Sellards, 1916; Allen, 1921; Simpson, 1930; Kellogg, 1944; Brodkorb, 1955). Phosphatic sediments of the Bone Valley Formation in this region have produced a sequence of vertebrate faunas dating from the middle Miocene (about 16 Ma) to the early Pliocene (4.5 Ma) (Webb and Hulbert, 1986). The best known Bone Valley fauna is early Pliocene in age and has been referred to the late Hemphillian North American Land Mammal Age (NALMA; Berta and Morgan, 1985; Webb and Hulbert, 1986; Tedford et al., 1987). Following Webb and Hulbert (1986), we use the term Palmetto Fauna to distinguish the late Hemphillian vertebrate assemblage from older faunas present in the Bone Valley region. The Palmetto Fauna contains a fossil assemblage representing marine, estuarine, freshwater, and terrestrial components, and appears to have accumulated at a time when sea levels were about 50 m higher than present (Webb and Tessman, 1968; MacFadden and Webb, 1982). The marine and estuarine components consist of sharks, rays, bony fish, sea turtles, pelagic birds, and many marine mammals, including a walrus, three species of seals, a sea otter, a sea cow, and numerous cetaceans (Berta and Morgan, 1985). The presence of sea turtles in late Hemphillian age deposits in the Bone Valley region was noted briefly by Berta and M rgan (1985) and Dodd (1990), but these fossils have not been previously described. Tertiary sea turtle fossils have been described from a number of localities on the Atlantic Coastal Plain, including the Paleocene Hornerstown Sand and Vincetown Formation and Eocene Manasquan Formation of New Jersey (Weems, 1988), the Paleocene Brightseat and Aquia Formations of Maryland and Virginia (Weems, 1988), the Oligocene Ashley Formation of South Carolina (Hay, 1923), the Miocene Kirkwood Formation of New Jersey (Zangerl and Turnbull, 1955; Weems, 1974), the Miocene Calvert Formation of Maryland and Virginia (Weems, 1974, 1980), and the Pliocene Yorktown Formation in North Carolina (Zug, in press). The very few published records of Tertiary sea turtles from the Gulf Coastal Plain include a questionable Eocene dermochelyid from Alabama (Thurmond and Jones, 1981) and a middle Eocene cheloniid from Florida (P. Meylan, 1984). There also are some unreported cheloniid fossils from Florida, mostly consisting of shell fragments, from the middle Eocene Avon Park Limestone and Inglis Formation and the Oligocene Suwannee Limestone (G. Morgan, unpubl. obs.). A few Florida Pleistocene fragments belonging to modern genera (Chelonia, Caretta) were mentioned by Hay (1916, 1917), but no comprehensive summary or description of these fossils is available. One of the largest collections of Tertiary sea turtle fossils in North America is from the Lee This content downloaded from 157.55.39.224 on Sat, 27 Aug 2016 04:05:15 UTC All use subject to http://about.jstor.org/terms C. K. DODD, JR. AND G. S. MORGAN

When I was in Vienna in the spring of 1879, Franz Ritter von Hauer, the Director of the Imperial Geological Survey, requested me to examine and described the remarkable organism on which Von Meyer had founded the genus Psephophorus , which, although noticed by himself, by Von Meyer, and more recently, by Dr. Fuchs, has never been figured. Its nature was at first sight so problematical that opinion loaned to the conclusion that it was the dermal covering of an Edentate closely allied to the Armadilloes. The dorsal surface of the fossil was perhaps insufficient to settle this question without a good deal of comparative work; but Von Hauer courteously allowed me to partially develop some fragments of bones, which are imperfectly preserved on the underside of the intractable sandstone matrix of the slab; and these fragments of procœlous vertebrae proved to be altogether reptilian; and though differing from the vertebrae of known reptiles, yet, by forward projection of the zygapophyses, they indicate the animal to be Chelonian, and therefore show the fossil to be more nearly allied to Sphargis than to any other type in the Chelonian order. When Von Meyer first gave a name to this genus (‘Jahrbuch,’ 1846, p. 472), it was only known to him by isolated dermal plates ; but subsequently a drawing was sent to him by Partsch, and on that he made a further note in the ‘Jahrbuch’ for 1847 (p. 579). This specimen, then in Pressburg, he describes as a fragment of a