Potamochelys Bergounioux and Crouzel, 1968

GAFFNEY, EUGENE S, TONG, HAIYAN & MEYLAN, PETER A, 2006, EVOLUTION OF THE SIDE-NECKED TURTLES: THE FAMILIES BOTHREMYDIDAE, EURAXEMYDIDAE, AND ARARIPEMYDIDAE, Bulletin of the American Museum of Natural History 300 (300), pp. 1-698 : 106-109

Potamochelys Bergounioux and Crouzel, 1968 .

TYPE AND ONLY INCLUDED SPECIES: Potamochelys gigantea Bergounioux and Crouzel, 1968 .

DISTRIBUTION: Maastrichtian of Niger,?Eocene of Mali.

ETYMOLOGY: Presumably in allusion to the country of Niger.

REVISED DIAGNOSIS: A bothremydid pleurodire of the subtribe Nigeremydina ; differs from Arenila in these characters: smaller orbits; suborbital maxillary plate deeper; labial ridge blunt.

PREVIOUS WORK: See species Previous Work.

DISCUSSION: See table 16 for a comparison of the genera in the tribe Taphrosphyini .

Nigeremys gigantea ( Bergounioux and

Crouzel, 1968)

TYPE SPECIMEN: MNHN (P) NIR 1 (label), a nearly complete skull (figs. 221– 223) lacking lower jaws with its surface badly eroded. However, Bergounioux and Crouzel (1968: 183) gave its number as ‘‘Muséum de Paris … 1964-27’’.

TYPE LOCALITY: ‘‘Niger’’ ( Bergounioux and Crouzel, 1968: 179). ‘‘East of Ibeceten, Niger’’ ( Lapparent de Broin and Werner, 1998: 145) (fig. 14).

HORIZON: ‘‘Maastrichtian’’ ( Bergounioux and Crouzel, 1968: 183).

DEPOSITIONAL ENVIRONMENT: Specimen found with: ‘‘ Mosasaurus nigeriensis Swinton , Stratodus apcalis Cope , Onchosaurus pharao Dames …’’ ( Bergounioux and Crouzel, 1968: 183), suggesting a near-shore marine burial environment.

DIAGNOSIS: As for genus.

ETYMOLOGY: Unknown, presumably referring to the large size of the skull.

REFERRED MATERIAL: BMNH R 10927, articulated right and left premaxillae and maxillae, partial quadrate, ‘‘loc 3, In Fargas, Samit limestone’’ (label), Eocene, Mali. The specimen as preserved is almost identical with Nigeremys , but it is too incomplete to be certain, so we extend the range of Nigeremys with doubt.

PREVIOUS WORK: Nigeremys Broin, 1977 began life as Potamochelys Bergounioux and Crouzel, 1968 . Bergounioux and Crouzel (1968) named what they thought was a new genus for a large, battered skull from Niger. They thought that in the absence of a carapace it was difficult to place their ‘‘ Potamochelys ’’ taxonomically, but they concluded that it was a cryptodire, possibly related to chelydrids. F. M. Bergounioux, a Catholic priest in Lyon, was infamous for his other work on fossil turtles (see section on Dubious Taxa). Bergounioux and Crouzel (1968) gave a brief description with dorsal and ventral photographs of the skull ( Bergounioux and Crouzel, 1968: figs. 1, 2) and a dorsal line drawing (fig. 1b). The dorsal line drawing is in error; there are no nasals and we have interpreted other sutures differently, although sutures are very difficult to see on this specimen. Later, Broin (1977: 83) recognized that Potamochelys Bergounioux and Crouzel, 1968 was preoccupied by Potamochelys Fitzinger, 1843 (a synonym of Trionyx according to Romer, 1956: 514), and created the new genus, Nigeremys , for the species gigantea .

Bergounioux and Crouzel (1968) had referred their ‘‘ Potamochelys ’’ to the Chelydridae with doubt, and Broin (1977) was the first to recognize its pleurodiran affinities. Antunes and Broin (1988) identified Nigeremys as a bothremydid, but unfortunately allied it with the undiagnosable ‘‘ Sokotochelys ’’ Halstead, 1979b in a ‘‘Groupe Nigeremys ’’, which was in a larger group containing Bothremys and Rosasia , but not Taphrosphys . In the description and discussion sections, there was new information about Nigeremys .

Lapparent de Broin and Werner (1998) provided the most complete previous description of Nigeremys . They argued for the close relationship of Arenila with Nigeremys , a conclusion supported by us, and provided a comparative description for the two taxa, line drawings for Nigeremys (fig. 4b, f) and measurements of the type skull, MNHN(P) NIR 1. Lapparent de Broin and Werner (1998) showed the ‘‘phyletic Group Nigeremys ’’ containing Nigeremys , ‘‘ Sokotochelys ’’, and Arenila , along with more fragmentary material.

DISCUSSION: Nigeremys is based on a nearly complete, but poorly preserved and badly prepared, skull. Painted sutures and repair materials obscure important areas. Although we disagree with some of the Lapparent de Broin and Werner (1998) sutures, we agree with their conclusion that Nigeremys is closely related to Arenila . In fact, our reconstruction of Arenila shows it to be very similar to Nigeremys . Both could be placed in the same genus as separate species, as they are a strictly monophyletic group, but we keep the genera separate here for old times’ sake.

Arenila Lapparent de Broin and Werner, 1998

TYPE AND ONLY INCLUDED SPECIES: Arenila krebsi Lapparent de Broin and Werner, 1998 .

DISTRIBUTION: Late Cretaceous, Egypt.

ETYMOLOGY: Arena, Latin for sand ( Lapparent de Broin and Werner, 1998).

DIAGNOSIS: A bothremydid pleurodire of the subtribe Nigeremydina ; differs from Nigeremys in these characters: larger orbits; suborbital maxillary plate shallower; labial ridge acute.

DISCUSSION: See table 16 for a comparison of the genera in the tribe Taphrosphyini .

Arenila krebsi Lapparent de Broin and

Werner, 1998

TYPE SPECIMEN: TUB Vb-641, a partial skull (figs. 226–230).

TYPE LOCALITY: ‘‘Ammonite Hills, interdunal channel 28, loc. 291080/2’’ ( Lapparent de Broin and Werner, 1998: 174) (fig. 14).

HORIZON: ‘‘Dakla Formation, Ammonite Hill Member, Maastrichtian’’ ( Lapparent de Broin and Werner, 1998: 174).

DIAGNOSIS: Same as for genus.

ETYMOLOGY: In honor of Dr. Bernard Krebs ( Lapparent de Broin and Werner, 1998: 174).

REFERRED MATERIAL: Lapparent de Broin and Werner (1998) identified a partial carapace, TUB Vb-648, lacking peripherals, nuchal, and pygal, as ‘‘? cf. Arenila krebsi ’’. The ‘‘cf.’’ is an abbreviation of the Latin, confere, meaning to compare, but there is no shell associated with the type skull to be used for comparison. The reasons for identifying this shell as Arenila are not specifically stated, but it occurs in the same unit as the type skull, is of an appropriate size, and is excluded from their ‘‘ Bothremys Group’’ and ‘‘ Taphrosphys Group’’ by Lapparent de Broin and Werner (1998). It is too incomplete to be included in the dataset presented here, and is considered Pelomedusoides incertae sedis.

PREVIOUS WORK: Arenila has been described only in Lapparent de Broin and Werner (1998). The photographs, plates VI and VII ( Lapparent de Broin and Werner, 1998), are of good quality but lack line drawings identifying elements, and only partial restorations of the palate and lateral views are shown ( Lapparent de Broin and Werner, 1998: fig. 12). Here we provide line drawings for these photographs (figs. 227, 229) as well as new photographs in the same orientation.

DISCUSSION: Lapparent de Broin and Werner (1998) identified Arenila as a bothremydid and a member of the ‘‘ Nigeremys Group’’. They identified a separate ‘‘ Taphrosphys Group’’. Here, we argue that Arenila is closely related to Nigeremys and agree with Lapparent de Broin and Werner.

Examinations of the type skull and associated material in the Technische Universität Berlin revealed that two skull elements, not identified by Lapparent de Broin and Werner , are part of the type skull, TUB Vb-641. These elements are included in the line drawings. They are: a partial right maxilla, premaxilla, and vomer ; and a partial right pterygoid consisting of the processus trochlearis pterygoidei. These elements articulate with the type skull directly, so there is little doubt of the association.

Our restoration of the skull of Arenila (fig. 224) differs from that of Lapparent de Broin and Werner (1998: fig. 12) in some details in ventral view and in overall shape in lateral view. These differences result from the newly added elements, different interpretation of some sutures, and a different interpretation of postmortem crushing. Considering the poor state of preservation of the type skull, the new restoration does not differ greatly from the original one.

DUBIOUS TAXA

A congenital disease of paleontology is the naming of poorly preserved specimens as new taxa. No doubt many of these are actually organic remains, although the published descriptions often do not demonstrate this unequivocally. Nonetheless, many names in the literature are based on inadequate specimens, and much effort is spent repeating them, usually to no positive effect. We have essentially divided the names into those with hope for future workers (incertae sedis) and those without hope of further identifications (nomina dubia). We have chosen these designations, which are commonly used to categorize poorly known taxa. We use ‘‘incertae sedis’’ (5 ‘‘of uncertain position’’) to designate taxa that preserve enough characters to be usefully diagnosed at the alpha level, but do not have enough characters to test their wider relationships at the present time. In our opinion, these taxa are complete enough so that there is hope for future material to be found and indentified with the type specimens. For incertae sedis the degree of uncertainty is always indicated.

The incertae sedis designation is not used for taxa that cannot be diagnosed adequately on the basis of the type specimen. Rather, the designation ‘‘nomen dubium’’ is used in the sense of Mayr (1969: 347), ‘‘an available name which cannot be assigned to a definite taxon owing to shortcomings in the original diagnosis or the type material.’’ The decision on what are an adequate diagnosis and type specimen is subjective, but we are of the opinion that a more restricted view serves systematists and other researchers better than a more lax one. Certainly many of these fragmentary specimens may represent unique taxa, but once they are named and become referred to in the literature, they are often used as the basis for studies in biogeography, diversity, and evolution, when the original material is completely inadequate for such work.

TAXA INCERTAE SEDIS

Apodichelys lucianoi Price, 1954

TYPE SPECIMEN: DNPM 418-R, a steinkern.

DISCUSSION: This taxon is from the Late Cretaceous Apodi Formation of Rio Grande del Norte, Brazil, and is described by Price (1954). It is clearly a pleurodire, showing a sutured pelvis, and study of the type supports the presence of laterally placed mesoplastra. Antunes and Broin (1988) placed it in the Bothremydidae because the internal mold shows ‘‘une morphologie et des rapports de dimensions conformes a ceux de Bothremys ’’ ( Antunes and Broin, 1988: 179). Presumably this refers to the short, broad, anterior plastral lobe seen in both Apodichelys and ‘‘ Bothremys ’’ (now Chedighaii ) barberi . However, while this taxon may be a bothremydid, a wide anterior lobe is not sufficient to objectively identify Apodichelys as such. The podocnemidid, Bairdemys venezuelensis (fig. 275) also has a short, wide anterior lobe. Nonetheless, Apodichelys is diagnosable; the wide, nearly horizontal epiplastra are distinct from other Pelomedusoides. However, it has too many missing data to be analyzed in our dataset.

CURRENT STATUS: Pelomedusoides incertae sedis.

‘‘ Chrysemys ’’ montolivensis Roman, 1897

TYPE SPECIMEN: University of Lyon 92839.

DISCUSSION: This record consists of a shell with carapace and plastron from the Oligocene of Montoulieu, France. First described by Roman (1897) as an emydid, it was questionably identified as the podocnemidid Neochelys by Broin (1977). Roman has photographs ( Roman, 1897) and Broin has line drawings showing sutures, sulci, and a dorsal view of the xiphiplastron ( Broin, 1977: fig. 66). Later, Lapparent de Broin and Werner (1998) and Lapparent de Broin (2001) identified this taxon as a bothremydid, but without giving any reasons. The shell is nearly complete, but unfortunately lacks the anteromedial region, so that the nuchal bone, entoplastron, and most of the epiplastra, all areas with useful characters, are missing. The relatively short anterior plastral lobe is typical of bothremydids, but also occurs in podocnemidids (i.e., Bairdemys ). The pectoral scale barely reaches the mesoplastron, also found in both bothremydids and podocnemidids. The surface ornamentation appears to be smooth. There are seven neurals, with the seventh and eighth costals meeting on the midline, again found in both podocnemidids and bothremydids. The pubic articulation scar is small, also not unique to either family. At present, this form, while probably a member of the Pelomedusoides, could be either a bothremydid or a podocnemidid. It is inadequate to extend the range of the Bothremydidae into the Oligocene, as claimed by Lapparent de Broin and Werner (1998). While we accept it as diagnosable, it is only marginally so.

CURRENT STATUS: Pelomedusoides incertae sedis.