Pentonyx gehafie Rüppell, 1835

Fritz, Uwe, Petzold, Alice, Kehlmaier, Christian, Kindler, Carolin, Campbell, Patrick, Hofmeyr, Margaretha D. & Branch, William R., 2014, Disentangling the Pelomedusa complex using type specimens and historical DNA (Testudines: Pelomedusidae), Zootaxa 3795 (5), pp. 501-522 : 510-513

Pentonyx gehafie Rüppell, 1835

Rüppell (1835: pp. 2–3) described Pentonyx gehafie based on terrapins which he found “häufig in allen fliessenden oder stagnirenden Gewässern auf dem östlichen Abhange der abyssinischen Küstengebirge” [frequently in all flowing and stagnant waters on the eastern slope of the Abyssinian coastal mountains]. Rüppell diagnosed Pentonyx gehafie from the Pentonyx species occurring at the Cape of Good Hope by the plastral scutation: The pectoral scutes of his new species are triangular and do not reach the median plastral seam, whereas in the Cape species the pectoral scutes are in contact. The character state of the Abyssinian specimens is nicely depicted in Rüppell’s plate 1. Obviously Rüppell collected quite a number of helmeted terrapins, which originally had all syntype status. Mertens (1937: p. 140) mentions that seven of these specimens are in the Senckenberg Museum, Frankfurt am Main. However, there are eight, and not seven, terrapins collected by Rüppell with identical data (SMF 7947–7949, SMF 7953, SMF 7960–7962, SMF 33054). Whilst SMF 7947–7949 and SMF 7960–7962 are alcohol-preserved, SMF 7953 and SMF 33054 are dry specimens ( Table 2 ). SMF 7947 was designated by Mertens (1937: p. 140) as lectotype, which is why the remaining terrapins bear today the status of paralectotypes. Further paralectotypes collected by Rüppell are in the Natural History Museum, London ( Gray 1856: p. 53; Boulenger 1889: p. 199) and the Muséum National d’Histoire naturelle, Paris (see below under Pelomedusa galeata var. disjuncta Vaillant & Grandidier, 1910 ). The lectotype is a subadult female of 11.42 cm straight carapacial length ( Fig. 5 , top).

Mertens (1937: p. 140) restricted the type locality of Pentonyx gehafie to “Massaua” [also known as Massawa, Missiwa or Mitsiwa, Eritrea], a town within the source region of the type specimens. However , according to Article 76.1 of the International Code of Zoological Nomenclature ( ICZN 1999 ), the type locality “is the geographical […] place of capture, collection or observation of the name-bearing type.” Thus, Rüppell’s original type locality has to remain unchanged and is to be identified with the eastern slope of coastal mountains, Eritrea .

Boulenger (1880: p. 151) synonymized Pelomedusa gehafie with Pelomedusa galeata (= Pelomedusa subrufa ) because he studied specimens with intermediate plastral characters from Madagascar, the Upper Nile Region, and Zanzibar, even though he acknowledged that separated triangular pectorals never occur in South African terrapins and that in Abyssinia the terrapins have consistently widely separated triangular pectorals. Based on similar observations, but drawing a completely different conclusion, Hewitt (1935: p. 325) recognized “ Pelomedusa gehafiae ” [sic] as a valid species and opposed it to his polytypic species Pelomedusa galeata . Hewitt noted that at Mt. Elgon (border region of Uganda and Kenya) helmeted terrapins with separated pectorals are found together with terrapins having the pectorals in contact. Finally, based on the occurrence of morphologically intermediate terrapins in some regions, Parker (1936: p. 609) and Mertens (1937: p. 140) recognized Pentonyx gehafie as a valid subspecies of the helmeted terrapin and used the names Pelomedusa galeata gehafie and Pelomedusa subrufa gehafie , respectively. In a similar vein, Loveridge (1941: p. 480) accepted the validity of a subspecies characterized by separated triangular pectoral scutes, but placed Pentonyx gehafie in the synonymy of Pelomedusa subrufa olivacea (Schweigger, 1812) . Summarizing his own observations and those of previous authors, Loveridge believed that this taxon occurs in “the drier regions of a belt extending from Senegal to Eritrea, intergrading with the typical form in the Anglo-Egyptian Sudan, Ethiopia, Somaliland, northern Kenya and Uganda ” (p. 481). Later authors identified Pelomedusa subrufa olivacea generally with the more northern populations of the helmeted terrapin, including the populations of the Arabian Peninsula (cf. Ernst & Barbour 1989; Iverson 1992; Ernst et al. 2000; Boycott & Bourquin 2008), until the usage of subspecies was abandoned after Gasperetti et al. (1993).

For the present investigation, we studied the lectotype and the nine paralectotypes of Pentonyx gehafie in Frankfurt am Main and London ; all specimens conform well to the description by Rüppell (1835) in having widely separated triangular pectoral scutes ( Fig. 5 ). Tissue for genetic examination was extracted from the eight Frankfurt specimens, but DNA sequences could be generated only for three terrapins, among them the lectotype ( Table 2 ) .

In phylogenetic analyses ( Fig. 1 ), the sequences of the type specimens constitute a deeply divergent clade being sister to Kenyan terrapins representing lineage V sensu Vargas-Ramírez et al. (2010). We examined 18 Pelomedusa from Kenya morphologically (7 live terrapins and 11 specimens in the collections of the natural history museums of Berlin, Bonn, London, and Vienna) and these differ consistently from the type series of Pentonyx gehafie in that their pectoral scutes are in wide contact in the midline, suggesting that Pentonyx gehafie could be a distinct taxon. At present, we refer to the genetic lineage of the type specimens of Pentonyx gehafie as lineage X, and do not relate it to any of the lineages identified by Vargas-Ramírez et al. (2010).

Figures

FIGURE 1. Maximum Likelihood tree for Pelomedusa, based on 1848 bp of mitochondrial DNA (12S, cyt b, ND4+tRNAs), rooted with Pelusios sinuatus. The tree topology completely matches the one inferred by Bayesian analyses. Numbers along nodes indicate bootstrap support and Bayesian posterior probabilities greater than 50 or 0.95, respectively (not shown for some terminal clades with short branches). Sample codes are voucher numbers; see Table S1 for explanation. Name-bearing types (red) and paralectotypes or topotypic samples (blue) highlighted. On the right, mitochondrial lineages of Vargas-Ramírez et al. (2010) indicated. No topotypic material of Pelomedusa galeata orangensis Hewitt, 1935 could be sequenced. However, the lost holotype is thought to come from the vicinity of Kimberley (South Africa) and all South African samples belong to lineage IX, so it can be concluded that Pelomedusa galeata orangensis represents the same lineage. Inset: Helmeted terrapin from the Addo Elephant National Park, South Africa. Photo: M. Gregorič.

FIGURE 5. Dorsal and ventral views of the lectotype of Pentonyx gehafie Rüppell, 1835 (SMF 7947, subadult female, straight carapacial length 11.42 cm; top) and of one of the paralectotypes in the Natural History Museum, London (BMNH 1947.3.5.78, adult male, straight carapacial length 16.70 cm; bottom). Scale bars, 3 cm. Photos: E. Morawa, P. Campbell.