Available online at www.sciencedirect.com

Palaeoworld 17 (2008) 126–134

Research paper

New basal procolophonid reptile from the Katberg formation (Lower Triassic) of the South African Karoo Juan Carlos Cisneros ∗ Bernard Price Institute for Palaeontological Research, University of the Witwatersrand, Private Bag 3 WITS 2050, Johannesburg, South Africa Received 19 December 2007; received in revised form 1 May 2008; accepted 19 June 2008 Available online 27 June 2008

Abstract A new procolophonid reptile, Kitchingnathus untabeni n. gen. et n. sp., is described from the uppermost strata of the Lystrosaurus Assemblage Zone of the Karoo Basin, South Africa. The new taxon co-occurs with the well-known Procolophon trigoniceps. The most distinctive feature of the new taxon is the presence of numerous small bicuspid molariforms in both the maxilla and the dentary. A phylogenetic analysis indicates that Kitchingnathus occupies a basal position among procolophonids. Character optimisation suggests that bicuspid teeth were acquired independently by the new taxon, and originated twice in procolophonid evolution. © 2008 Nanjing Institute of Geology and Palaeontology, CAS. Published by Elsevier Ltd. All rights reserved. Keywords: Parareptiles; Procolophonids; Lystrosaurus Assemblage Zone; Triassic; South Africa

1. Introduction The Early Triassic Lystrosaurus Assemblage Zone (AZ) of the Karoo Basin is characterised by relatively low tetrapod diversity and the dominance of the dicynodont Lystrosaurus (Kitching, 1977). Collecting in the Lystrosaurus AZ (Fig. 1A) has traditionally been neglected due to the monotony of Lystrosaurus findings (Kitching, 1977), a genus that comprises up to 95% of the vertebrates in this horizon (Groenewald and Kitching, 1995). The procolophonoid Procolophon is also found in this biozone, occurring in isolated but usually large concentrations (Groenewald and Kitching, 1995). Procolophonoids are the only clade of parareptiles that survived the Permo-Triassic extinction event and constitute part of the Early Triassic recovery fauna of the Karoo Basin (Fig. 1B; Modesto et al., 2001; Smith and Botha, 2005; Botha and Smith, 2006; Botha et al., 2007). The group radiated throughout Pangaea, and its last members are known from Upper Triassic rocks in Brazil, Britain, Canada

and USA (Sues et al., 2000; Cisneros and Schultz, 2003; Fraser et al., 2005). In recent years, renewed attention has been given to the tetrapods of the Lystrosaurus AZ, resulting in the description of the new procolophonoids Owenetta kitchingorum (Reisz and Scott, 2002), Saurodektes rogersorum (Modesto et al., 2003), Coletta seca (Gow, 2000) and Sauropareion anoplus (Modesto et al., 2001), and probable new temnospondyl amphibians (Damiani et al., 2000; Damiani and Welman, 2001). Gow (1977) mentioned a procolophonid specimen from the Lystrosaurus AZ that somehow differed from the genus Procolophon in the dentition. However, Gow (1977) concluded that this specimen was a juvenile Procolophon and the differences with other specimens were due to ontogeny. Gow (2000) changed his view and stated the possibility that the fossil could represent a new taxon. Based on this specimen, a new genus and species of basal procolophonid is described herein. 1.1. Institutional abbreviations

∗

Present address: Departamento de Paleontologia e Estratigrafia, Universidade Federal do Rio Grande do Sul, Av. Bento Gonc¸alves 9500, Porto Alegre, CP 15001, 91540-000, Brazil. Tel.: +55 51 3308 6385; fax: +55 51 3308 7302. E-mail address: [email protected].

AMNH, American Museum of Natural History, New York; BP, Bernard Price Institute for Palaeontological Research, Johannesburg; NM, National Museum, Bloemfontein; SAM, Iziko, South African Museum, Cape Town.

1871-174X/$ – see front matter © 2008 Nanjing Institute of Geology and Palaeontology, CAS. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.palwor.2008.06.003

J.C. Cisneros / Palaeoworld 17 (2008) 126–134

127

Fig. 1. (A) Map of South Africa and Lesotho showing the location of the type locality (Hobbs Hill) of Kitchingnathus untabeni n. gen. et n. sp. The outcrops that yield Lystrosaurus Assemblage Zone fauna are represented in dark grey (after Groenewald and Kitching, 1995). (B) Stratigraphic chart showing the placement of South African procolophonoids (owenettids and procolophonids) in the geologic time. From left to right: periods, epochs, Karoo assemblage zones, Karoo formations, and taxa. Bars represent taxon ranges and solid circles indicate single specimen occurrences. The interval where the abundance of Procolophon trigoniceps is anomalously high (Neveling, 2004; Botha and Smith, 2006) is represented by a solid box. Owenettids are followed by an asterisk. Taxon ranges and occurrences after Neveling (2004), Botha and Smith (2006), Botha et al. (2007), Abdala et al. (2006) and Cisneros (2008b). Abbreviations: Chang., Changsingian; Ciste., Cistecephalus; Mid., Middleton Formation; P-Tr, Permian-Triassic Boundary; Wuchiaping., Wuchiapingian.

2. Systematic palaeontology Parareptilia Olson, 1947 Procolophonoidea Romer, 1956 Procolophonidae Cope, 1889 Kitchingnathus n. gen. Etymology: In honour of the late James W. Kitching, a prominent South African palaeontologist and collector of the specimen; and from the Greek gnathus, mandible. Diagnosis: As for the type and only known species. Type species: Kitchingnathus untabeni n. sp. Kitchingnathus untabeni n. sp. (Figs. 2 and 3) v. 1977 Procolophon trigoniceps Gow, p. 701, text-fig. 6. Etymology: An isiZulu term meaning “from the hill”, a reference to the locality where the fossil was collected. Types: Holotype, BP/1/1187. A partial skeleton, in the collection of the Bernard Price Institute for Palaeontological Research, University of the Witwatersrand, Johannesburg. Collected by James W. Kitching in October 1952. Locality and horizon: Hobbs Hill (Windvogelsberg), west of Cathcart, Eastern Cape Province, South Africa; middle-upper Katberg Formation, Beaufort Group, Karoo Supergroup, uppermost Lystrosaurus AZ (Kitching, 1977), late Early Triassic (Olenekian). Besides the holotype of Kitchingnathus untabeni, the locality Hobbs Hill has yielded several Procolophon trigo-

niceps specimens collected during field trips led by James W. Kitching in 1952 and 1966. Kitching (1977) assigned the fossiliferous horizons of this locality to the uppermost Lystrosaurus AZ sandstones. The matrix that holds BP/1/1187 is bright red sandstone, characteristic of the middle-upper Katberg Formation and lowermost Burgersdorp Formation (Johann Neveling, pers. comm., 2006). Kitchingnathus untabeni was most likely collected in the middle or upper horizons of the Katberg Formation , a stratigraphic assessment that is consistent with the presence of the index genus Procolophon at the locality. The occurrence of Procolophon in South Africa is restricted to the interval between the middle part of the Katberg Formation and the lowermost Burgersdorp Formation (Neveling, 2004; Smith and Botha, 2005; Botha and Smith, 2006). Diagnosis: Autapomorphies of the new taxon include: (1) At least nine upper and eight lower, chisel-like, small bicuspid molariforms, with labial cusps taller and thinner than lingual cusps; and (2) a long posterior process of the maxilla that extends along the rim of the subtemporal emargination as much as the quadratojugal. Remarks: The new taxon possesses the highest number of bicuspid teeth reported for a procolophonid. In Phaanthosaurus (Contritosaurus) from the Lower Triassic of Russia, a similar count of bicuspid teeth (up to eight) is recorded in the maxilla. However, this form can readily be distinguished from the new taxon in having a deeper snout and short upper incisiforms which do not surpass the height of the molariforms, and by the absence of lower bicuspid teeth. The new taxon differs from Sauropareion anoplus also by having a broader subtemporal

128

J.C. Cisneros / Palaeoworld 17 (2008) 126–134

emargination and a posterior median parietal projection. It is further distinguished from the co-occurring Procolophon trigoniceps by a number of characters (see below), including a longer snout and the absence of quadratojugal horns. 2.1. Description 2.1.1. Cranium The cranium of BP/1/1187 (Figs. 2 and 3A) is compressed laterally and fractured loosely along the midline, producing the separation of right and left sides before burial of the specimen. Most of the left side of the cranium is probably missing. Some unidentified flat bones that lay below the posterior margin of the right side of the cranium presumably belong to the left side. A tooth bearing fragment may represent a portion of the left maxilla exposed in medial view (Fig. 2B, ?lmx). A fracture separates the thin layer of matrix that contains most of the visible

part of the cranium and disarticulated mandibular elements of BP/1/1187 from the largest part of the block. This fracture was filled with non-reversible glue, probably in the field or during preparation for Gow’s (1977) study, making it difficult to search for additional remains of the left side of the cranium. Removal of the glue could cause serious damage to the prepared and already informative parts of the specimen, and for this reason, no attempt of further preparation was done below the exposed regions of the skull. Instead, some additional preparation was carried out on the exposed side of the specimen. Kitchingnathus possesses a gracile skull. The snout is elongated, and slightly deeper than in Coletta seca but comparable to that of Tichvinskia vjatkensis from Russia (see Ivakhnenko, 1973). The external surface of the snout is not well preserved, and it is very difficult to trace the sutures of the premaxilla, the nasal, the maxilla and the frontal in this area. A fracture has separated most of the right premaxilla and a small portion of the

Fig. 2. (A) Kitchingnathus untabeni n. gen. et n. sp., Lower Triassic, South Africa, BP/1/1187, holotype. Photograph of the skeleton. (B) Outline of the skeleton. Abbreviations: ac, anterior coracoid; ap, autopodium; c, clavicle; cr, cranium; g, gastralia; i, interclavicle; il, ilium; f, femur; lm, left mandibular ramus; lmx, left maxilla; of, orbitotemporal fenestra; r, rib; rm, right mandibular ramus; sc, sacrum; vt, vertebra. Digits are identified with Roman numerals ii–v. Postcranial bones that could not be identified are not labelled.

J.C. Cisneros / Palaeoworld 17 (2008) 126–134

129

Fig. 3. Kitchingnathus untabeni n. gen. et n. sp., Lower Triassic, South Africa, BP/1/1187, holotype. (A) Cranium, right view. (B) Right mandible in lateral view and selected teeth in occlusal view. (C) Left mandible in medial view and selected teeth in occlusal view. (D) Right mandible, lateral view. (E) Left mandible, dorsolateral view. (F) Left mandible, posterodorsal view. Abbreviations: amf, anterior maxillary foramen; d, dentary; cr, coronoid; f, frontal; j, jugal; l, lacrimal; mx, maxilla; n, nasal; op, opisthotic; p, parietal; pf, pineal foramen; pmx, premaxilla; pof, postfrontal; prf, prefrontal; pt, pterygoid; q, quadrate; qj, quadratojugal; sp, splenial; st, supratemporal; ect, ectopterygoid. Unshaded surfaces represent features preserved as natural casts. Roman numerals indicate tooth positions. Small-scale bar represents 5 mm for parts A, B and C, except for teeth in occlusal view where it represents 2 mm. Large-scale bar represents 5 mm for parts D, E and F.

maxilla from the rest of the skull. The external naris is subcircular. The anterior maxillary depression is damaged, and is located on the anterior portion of the maxilla, adjacent to the posterior margin of the external naris. This depression presumably comprises a small part of the nasal but it is not possible to locate the suture between the maxilla and the nasal with confidence to confirm this. No septomaxilla is preserved. Sixteen upper right marginal teeth are present. This is an elevated marginal tooth count for a procolophonid. Only Coletta seca from South Africa is known to have a comparable upper tooth count, 15 and 16, on the left and right side respectively

(Modesto et al., 2002). Owenettids, however, have considerably higher upper tooth counts (e.g., more than 30 in Owenetta rubidgei, Reisz and Scott, 2002). The upper marginal teeth are not inset from the maxillary surface, a feature that is common to basal procolophonids such as Coletta, Pintosaurus, Phaanthosaurus and Sauropareion (Ivakhnenko, 1979; Modesto et al., 2001, 2002; Pi˜neiro et al., 2004). The premaxilla–maxilla suture is not preserved but, as in most procolophonids, it is probably located below the anterior maxillary depression. A small tooth (Fig. 3A(iv), probably a replacement tooth) is located at this position, and it could represent the fourth premaxillary tooth

130

J.C. Cisneros / Palaeoworld 17 (2008) 126–134

or the first maxillary one. Therefore, a maximum of four premaxillary teeth is likely. Three premaxillary teeth are present in almost all procolophonids and four in Coletta seca (Modesto et al., 2002). All premaxillary teeth and the first few maxillary teeth are conical. The conical tooth region comprises seven teeth, and extends to the level of the anterior maxillary foramen. The first three are enlarged (ca. 50 per cent longer than other upper teeth) and their apices are slightly recurved. All the conical teeth of Kitchingnathus are thicker than the teeth of owenettids, but less bulbous than the conical teeth of Procolophon. The crown of the seventh tooth is preserved partially as a natural mould. There is a space between the seventh tooth and the eighth tooth. Although this space is large enough to hold another tooth, no socket or pulp cavity is present. The molariform region comprises nine teeth. The crowns of the eighth and the sixteenth teeth, the latter being the last maxillary tooth, are fully exposed. These teeth are molariform with labiolingually expanded bases, the maximum width of which occurs at the base. The crowns are bicuspid. The labial cusp is higher than the lingual cusp, and the cusps are connected by a labiolingual ridge. The rim of the orbitotemporal fenestrae is formed by the lachrymal, the prefrontal, the parietal, the postfrontal, the jugal, and presumably the postorbital. Only portions of the lachrymal and the prefrontal can be traced. As in other procolophonids, a small lateral extension of the frontal reaches the orbitotemporal fenestra, precluding the prefrontal and postfrontal from contacting each other in dorsal view. The postfrontal is not fused to the parietal, contrary to the condition present in Tichvinskia (Ivakhnenko, 1973), Kapes (Novikov and Sues, 2004), Hypsognathus (Sues et al., 2000) and other genera. The postfrontal is long and narrow, decreasing in width towards its anterior edge. The subtemporal emargination is broad, in contrast with the genus Sauropareion, in which the emargination is acute and narrow as in owenettids (Modesto et al., 2001). The subtemporal emargination is formed by the maxilla, the jugal and the quadratojugal. The posterior process of the maxilla is unusually long and thin, and contributes as much as the quadratojugal to the border of the subtemporal emargination. In other procolophonids, the posterior process of the maxilla is shorter and contributes less than the quadratojugal, or is completely excluded from the rim of the subtemporal emargination. A portion of the transverse flange of the pterygoid or ectopterygoid can be seen through the subtemporal emargination, but it is not possible to identify the suture between these elements. The quadratojugal appears to be higher than wide, although its complete width cannot be assessed because the otic notch lays covered with matrix. The quadratojugal lacks the lateral spine characteristic of other procolophonids. Part of the quadrate is visible below the quadratojugal; it is placed slightly below at the alveolar level of the maxilla. The squamosal and the postorbital are covered by the dislocated supratemporal. The right half of the pineal foramen is present. It is essentially circular as in most procolophonids and it appears to be placed in a shallow fossa. The parietal is broad and has a long contact with the supratemporal. The fronto-parietal suture is not visible. A smooth process extends from the posterior margin of the parietal which forms half of the median parietal projection present in

many procolophonids. The large supratemporal closely resembles that of Sauropareion, as depicted by Modesto et al. (2001). It is essentially rectangular, its anteroposterior length being larger than its mesolateral length. The supratemporal possesses a flat occipital surface (best shown in Fig. 2A) that is equivalent to approximately half of the dorsal surface. There is no evidence for postparietals. 2.1.2. Mandible The right mandibular ramus (Fig. 3B and D) is exposed in lateral view. The anterior and posterior portions of this ramus are missing, and its ventral surface is damaged. Posteriorly, a long element that lays partially below this ramus likely represents the dislocated quadrate process of the right pterygoid. Most of the dentary is present but its sutures with other elements in the mandible are poorly preserved. Parts of the splenial are visible through the damaged ventral portion of the mandible. Most of this bone was presumably covered in life by the dentary in lateral view. The articular is not preserved, but the position of the quadrate indicates that it was located at the alveolar level of the dentary teeth, not below it as in leptopleuronine procolophonids (Sues et al., 2000; Cisneros and Schultz, 2003, Fig. 2g). Thirteen teeth are preserved in the right dentary; the anteriormost incisiforms are missing (Fig. 3B and D). These teeth are not inset from the dentary surface. The first preserved tooth in the dentary is well exposed. It is monocuspid and conical, subcircular in basal cross-section, and its maximum width occurs at the cervix. The two following teeth are partially covered by matrix, but they are likely conical. Only one cusp is visible on each of them, positioned roughly at the centre of the crown. The fourth preserved tooth is labiolingually expanded and the only cusp preserved is positioned labially. The two following teeth are smaller in height and are probably replacement elements. Their crowns, and those of the following teeth, were prepared. These are molariform, and, as in the upper teeth, the labial cusps are higher and sharper than the lingual cusps and are connected by a labiolingual ridge. This ridge is sigmoidal in occlusal view but its sinuosity is less pronounced than in the lower molariforms of Procolophon (Carroll and Lindsay, 1985, Fig. 14). The crowns of the following teeth could not be prepared. Fifteen teeth were probably present in the right mandible and a maximum of 10 bicuspid molariforms is likely. The left ramus (Fig. 3C, E and F) is exposed in medial view. It is heavily weathered and only part of the dentary is present. The teeth are fractured at the base and their pulp cavities are exposed. It is not clear if the teeth possess real roots, but pulp cavities below the tooth cervices are shallow. Most crowns are well preserved. The dentition of this ramus shows 15 teeth. The three mesialmost teeth are preserved mainly as natural moulds and are conical, presumably monocuspid. These are followed by four more conical, monocuspid teeth. The last molariform is preserved as a mould. At least eight bicuspid molariforms are present (Fig. 3C, E and F). 2.1.3. Postcranium The partial postcranium (Fig. 2) is exposed in dorsal view, but is poorly preserved. The vertebral column was not preserved

J.C. Cisneros / Palaeoworld 17 (2008) 126–134

except for the sacrals, adjacent vertebrae, and a few isolated vertebrae. Some thoracic and sacral ribs are present, as well as gastralia. None of the thoracic ribs are preserved in full length, so it is not possible to see if the thoracic rib cage is expanded as in derived procolophonids. Part of the pectoral girdle is present mainly as a mould. This includes the interclavicle, a portion of the left clavicle, and paired coracoids. The right coracoids are better preserved than the left ones. Both anterior and posterior coracoids are essentially rounded. Coracoids are not known in most procolophonids, but in Procolophon (BP/1/962) and in SAM PK-K7711, a specimen from the Middle Triassic subzone B of the Cynognathus AZ erroneously referred to Procolophon by deBraga (2003, Figs. 3 and 5; see Cisneros, 2008a), the posterior coracoid is anteroposteriorly elongated. The only visible portion of the pelvic girdle lies adjacent to the pectoral girdle. This may be the result of the specimen being in a curled-up position before dying and/or postmortem disarticulation. At least two sacral vertebrae can be distinguished based on the presence of sacral ribs. Both ribs are preserved on the first sacral vertebrae and a right rib lies on the second sacral vertebrae. The distal portions of these ribs are expanded and overlapping. A count of three sacral vertebrae, however, is likely. Three sacrals are normal for procolophonids and owenettids (see Ivakhnenko, 1979; Reisz and Scott, 2002; deBraga, 2003). Sacral ribs may have become detached from the third sacral vertebra. The right ilium is partially preserved. It is articulated to two sacral ribs and extends posteriorly parallel to the third putative sacral vertebra. The right femur is preserved in articulation with the ilium. A autopodium is preserved between the right femur and the cranium, but it is not possible to confirm if it belongs to the forelimb or the hindlimb. Four digits have been tentatively identified in this autopodium.

131

explained by an early developmental stage within Procolophon, including: (1) marginal upper and lower dentitions that are not inset from the surfaces of the maxilla and the dentary, respectively; (2) an elongated, low snout (pre-orbital length, character 4 in Cisneros, 2008b); (3) a long posterior maxillary process; (4) a large supratemporal that is subrectangular in dorsal view; and (5) a slender quadratojugal. These anatomical features are in contrast not only with adult Procolophon individuals but with juveniles as well. Some Karoo specimens (Fig. 4; see also Carroll and Lindsay, 1985, Fig. 7) are much smaller than those considered by Gow (1977) and provide examples of individuals that can be readily referred to Procolophon trigoniceps but are morphologically incompatible with BP/1/1187. These specimens feature a reduced number (six or less) of very enlarged molariforms that are inset from the maxillary wall, plus a robust, deep snout, a short posterior maxillary process, a reduced supratemporal that is triangular in dorsal view and a robust quadratojugal with a small lateral process. It is very unlikely, judging from their enlarged molariforms, that any of these young Procolophon individuals could have fed on insects according to the model proposed by Gow (1977). It must be mentioned that when BP/1/1187 was first considered by Gow (1977) no other procolophonoids were known in the Lystrosaurus AZ besides Procolophon. Four more procolophonoids, however, are now recognized in this biozone: Coletta seca (Gow, 2000), ‘Owenetta’ kitchingorum (Reisz and Scott, 2002), Sauropareion anoplus (Modesto et al., 2001), and Saurodektes rogersorum (Modesto et al., 2003). These new procolophonoids are smaller and more gracile than Procolophon trigoniceps. In the description of Coletta seca, Gow (2000, p. 22) conceded that BP/1/1187 could represent a new taxon. 3.2. Phylogenetic relationships of the new taxon

3. Discussion 3.1. Comparison with Procolophon juveniles BP/1/1187 was originally referred to Procolophon trigoniceps by Gow (1977), although he openly admitted that this specimen fell outside the range of variation of the species (Gow, 1977, p. 701). The basis for his claim was that BP/1/1187 would represent an early ontogenetic stage, earlier than all other specimens of Procolophon trigoniceps he examined. Gow (1977) suggested that very young Procolophon individuals could have a more numerous marginal dentition, specialized for insectivory, that was completely replaced between the ontogenetic stages represented by BP/1/1187 and BP/1/959 (a small Procolophon specimen equivalent in size to BP/1/1187). There is no compelling evidence to allow determination of the ontogenetic stage of BP/1/1187 at the time of death. Nevertheless, the skeletal elements of BP/1/1187 appear to be well ossified, and the specimen is compatible in size with adults of other Early Triassic procolophonids, e.g. the Russian form Tichvinskia vjatkensis which is known from a series of individuals (Ivakhnenko, 1979). As noted in the description, BP/1/1187 differs from Procolophon trigoniceps not only in the number and morphology of its marginal teeth, but in several characters that could hardly be

In order to evaluate the phylogenetic position of Kitchingnathus untabeni, a cladistic analysis of procolophonid relationships was performed. The new taxon was added to the matrix of 23 taxa and 58 characters for procolophonids compiled by Cisneros (2008b). The analysis was performed with TNT (Goloboff et al., 2003) following the same settings used by Cisneros (2008b), using the Implicit Enumeration algorithm and Collapsing Rule 1 (minimum length = 0), with some “ordered” characters (0, 7, 17, 25, 30, 31) and equal weights. A preliminary run produced 14 most parsimonious trees (MPTs, tree length = 118, CI = 0.661 [excluding uninformative characters], RC = 0.527), placing Kitchingnathus untabeni in a polytomy with the basal forms Coletta, Sauropareion, Pintosaurus and Phaanthosaurus; and a clade that includes all other procolophonids considered in the analysis (clade C sensu Cisneros, 2008b). The analysis was run again with the same settings but excluding Pintosaurus, a taxon which could only be coded for 18 out of 58 characters. This analysis produced a single MPT (Fig. 5) and identifies Kitchingnathus untabeni as a basal form, being the sister taxon of a group formed by Phaanthosaurus and clade C (Cisneros, 2008b). Two non-ambiguous synapomorphies support the placement of Kitchingnathus in this group: a broadly excavated ventral margin of the temporal region of

132

J.C. Cisneros / Palaeoworld 17 (2008) 126–134

Fig. 4. Crania of juvenile Procolophon trigoniceps individuals from the South African Karoo. Parts (A and B) dorsal view and parts (E and F) left lateral view of AMNH 4980 (locality unknown); parts (C and D) left lateral view and parts (G and H) dorsal view of NM QR1447, from Bethulie, Free State Province. Abbreviations: a, angular; amf, anterior maxillary foramen; ar, articular; cr, coronoid; d, dentary; f, frontal; j, jugal; l, lacrimal; mx, maxilla; mc, meckelian canal; n, nasal; p, parietal; pf, pineal foramen; pmx, premaxilla; po, postorbital; pof, postfrontal; prf, prefrontal; q, quadrate; qf, quadrate foramen; qj, quadratojugal; sa, surangular; sm, septomaxilla; sq, squamosal; st, supratemporal.

the skull (character 12:2) and the loss of a distinct postparietal (character 16:1). This analysis produced a sister taxon relationship between the South African forms Coletta and Sauropareion, and the two taxa share the absence of teeth or denticles along the posterior medial suture of the vomer (character 35:1). This dichotomy has not been reported by previous workers on procolophonid relationships that evaluated both taxa, and could be a result of the presence of a new basal procolophonid in the analysis. Character optimisation (Fig. 5) suggests that bicuspid marginal dentition arose twice within Procolophonoidea, the only parareptile lineage that developed this tooth morphology. The analysis indicates that the bicuspid tooth condition was acquired by Kitchingathus and again later in a clade that comprises the genus Tichvinskia together with the groups Procolophoninae and Leptopleuroninae sensu Modesto et al. (2002). Most procolophonids are included in this clade. This hypothesis, however, should be considered with caution due to the low levels of support recorded in the analysis. 3.3. Palaeoecological and biostratigraphical considerations Bicuspid and labiolingually expanded teeth constitute a major acquisition that allowed the exploitation of durophagous and/or high-fibre herbivorous niches by amniotes (Colbert, 1946;

Hotton et al., 1997; Reisz and Sues, 2000). The feeding habits of Kitchingnathus were probably different from those of derived procolophonids, such as Procolophon or Hypsognathus, which are traditionally considered to be durophagous/herbivorous (Colbert, 1946; Gow, 1977; Reisz and Sues, 2000). Unlike Procolophon, the cusps in the molariforms of the new taxon are sharp and the crowns are not notably worn, a condition that is not compatible with the practice of durophagy or highfibre herbivory (Hotton et al., 1997). The presence of numerous conical teeth and small molariforms with sharp cusps instead suggests an insectivorous niche (Gow, 1977). The dentition of Kitchingnathus indeed resembles that of modern hedgehogs, which use their teeth to tear and puncture a variety of invertebrates. With the discovery of Kitchingnathus untabeni, there are now five procolophonoids known from the Katberg Formation of the Beaufort Group (Fig. 1B). The new taxon co-occurs at the type locality, Hobbs Hill, with the well-known and abundant form Procolophon trigoniceps. The three other procolophonoids reported for this formation, however, probably do not overlap the stratigraphic level of the new genus. These are the recently described Coletta seca (Gow, 2000), Sauropareion anoplus (Modesto et al., 2001) and Owenetta kitchingorum (Reisz and Scott, 2002). The only specimen of Coletta seems to come from the lower or middle part of the Katberg Formation (Botha et al., 2007). Owenetta kitchingorum and Sauropareion anoplus

J.C. Cisneros / Palaeoworld 17 (2008) 126–134

133

Fig. 5. Most parsimonious hypothesis of procolophonid relationships produced by TNT. Tree length = 115, CI = 0.678 (excluding uninformative characters); RC = 0.545. Symmetric resampling and decay index values are provided before each node (above and below, respectively) as measurements of support. Symmetric resampling was calculated from 5000 replicates (p = 33). The presence of bicuspid marginal teeth is indicated by asterisks. Sauropareion anoplus was recoded as polymorphic for character 49 (ectepicondylar foramen or groove on humerus present [0] or absent [1]) based on personal observation on specimen NM QR3544 (see also Botha et al., 2007, Fig. 1). Kitchingnathus untabeni was coded [12]1??0 10?00 [02]020? 110?? ???00 [12]0100 1200? ????1 0???? ????? ????? ??0.

were first reported for the underlying Palingkloof Member of the Balfour Formation (Damiani et al., 2003) but there are now records in the Katberg Formation below the stratigraphical range of Procolophon (Abdala et al., 2006; Botha et al., 2007).

author was recipient of a grant from the Palaeontological Scientific Trust (PAST) in South Africa and Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) in Brazil. References

Acknowledgements Special thanks to Adam Yates (BP, South Africa) for drawing my attention to the holotype at the collection of the Bernard Price Institute for Palaeontological Research. Johann Neveling (Council for Geoscience, South Africa) is acknowledged for his advice on geology, Ross Damiani (Staatliches Museum für Naturkunde, Germany) for comments on a preliminary version of the paper and Bruce Rubidge (BP, South Africa) for his support at the Bernard Price Institute for Palaeontological Research. Jennifer Botha and Elize Butler (NM, South Africa) are acknowledged for allowing the study of the fossils under their care. Thanks to D. Mbense (BP, South Africa) for his advice on isiZulu and to Charlton Dube (BP, South Africa) for additional preparation of the holotype. This contribution benefited from the constructive reviews by Andrew B. Heckert and Sean P. Modesto. The

Abdala, F., Cisneros, J.C., Smith, R.M.H., 2006. Faunal aggregation in the Early Triassic Karoo Basin: earliest evidence of shelter-sharing behaviour among tetrapods? Palaios 21, 507–512. Botha, J., Smith, R.M.H., 2006. Rapid vertebrate recuperation in the Karoo Basin of South Africa following the end-Permian extinction. Journal of African Earth Sciences 45, 502–514. Botha, J., Modesto, S.P., Smith, R.M.H., 2007. Extended procolophonoid reptile survivorship after the end-Permian extinction. South African Journal of Science 103, 54–56. Carroll, R.L., Lindsay, W., 1985. Cranial anatomy of the primitive reptile Procolophon. Canadian Journal of Earth Sciences 22, 1571–1587. Cisneros, J.C., 2008a. Taxonomic status of the reptile genus Procolophon from the Gondwanan Triassic. Palaeontologia Africana 43, 7–17. Cisneros, J.C., 2008b. Phylogenetic relationships of procolophonid parareptiles with remarks on their geological record. Journal of Systematic Palaeontology 22 pp., doi:10.1017/S1477201907002350. Cisneros, J.C., Schultz, C.L., 2003. Soturnia caliodon n. g. n. sp., a procolophonid reptile from the Upper Triassic of southern Brazil. Neues Jahrbuch für Geologie und Paläontologie, Abhandlungen 227, 365–380.

134

J.C. Cisneros / Palaeoworld 17 (2008) 126–134

Colbert, E.H., 1946. Hypsognathus, a Triassic reptile from New Jersey. Bulletin of the American Museum of Natural History 86, 225–274. Cope, E.D., 1889. Synopsis of the families of the Vertebrata. American Naturalist 23, 849–877. Damiani, R.J., Welman, J., 2001. A long-snouted trematosaurid amphibian from the Early Triassic of South Africa. South African Journal of Science 97, 318–320. Damiani, R., Neveling, J., Hancox, J., Rubidge, B., 2000. First trematosaurid temnospondyl from the Lystrosaurus Assemblage Zone of South Africa and its biostratigraphic implications. Geological Magazine 137, 659–665. Damiani, R., Neveling, J., Modesto, S., Yates, A., 2003. Barendskraal, a diverse amniote locality from the Lystrosaurus Assemblage Zone, Early Triassic of South Africa. Palaeontologia Africana 39, 53–62. deBraga, M., 2003. The postcranial skeleton, phylogenetic position, and probable lifestyle of the Triassic reptile Procolophon trigoniceps. Canadian Journal of Earth Sciences 40, 527–556. Fraser, N.C., Irmis, R.B., Elliot, D.K., 2005. A procolophonid (Parareptilia) from the Owl Rock Member, Chinle Formation of Utah, USA. Palaeontologia Electronica 8-1-13A, 7 pp. Goloboff, P.A., Farris, J.S., Nixon, K.C., 2003. TNT: Tree Analysis using New Technology (version 1.1 for Windows). Program and documentation available at http://www.zmuc.dk/public/phylogeny. Gow, C.E., 1977. Tooth function and succession in the Triassic reptile Procolophon trigoniceps. Palaeontology 20, 695–704. Gow, C.E., 2000. A new procolophonid (Parareptilia) from the Lystrosaurus Assemblage Zone, Beaufort Group, South Africa. Palaeontologia Africana 36, 21–23. Groenewald, G.H., Kitching, J.W., 1995. Biostratigraphy of the Lystrosaurus Assemblage Zone. In: Rubidge, B. (Ed.), Biostratigraphy of the Beaufort Group (Karoo Supergroup). Biostratigraphic Series, 1. South African Committee for Stratigraphy, Pretoria, pp. 35–39. Hotton III, N., Olson, E.C., Beerbower, R., 1997. Amniote origins and the discovery of herbivory. In: Sumida, S.S., Martin, K.L.M. (Eds.), Amniote Origins: Completing the Transition to Land. Academic Press, San Diego, pp. 206–264. Ivakhnenko, M.F., 1973. Skull structure in the Early Triassic procolophonian Tichvinskia vjatkensis. Paleontological Journal 4, 74–83. Ivakhnenko, M.F., 1979. Permian and Triassic procolophonians of the Russian Plataform. Trudy Paleontologicheskogo Instituta, Academiia Nauka SSSR 164, 1–80 (in Russian).

Kitching, J.W., 1977. The distribution of the Karroo vertebrate fauna. Memoir of the Bernard Price Institute for Palaeontological Research, University of the Witwatersrand 1, 1–131. Modesto, S., Sues, H.D., Damiani, R., 2001. A new Triassic procolophonoid reptile and its implications for procolophonoid survivorship during the Permo-Triassic extinction event. Proceedings of the Royal Society of London, Series B 268, 2047–2052. Modesto, S.P., Damiani, R.J., Sues, H.D., 2002. A reappraisal of Coletta seca, a basal procolophonoid reptile from the Lower Triassic of South Africa. Palaeontology 45, 883–895. Modesto, S.P., Damiani, R.J., Neveling, J., Yates, A.M., 2003. A new Triassic owenettid parareptile and the Mother of Mass Extinctions. Journal of Vertebrate Paleontology 23, 715–719. Neveling, J., 2004. Stratigraphic and sedimentological investigation of the contact between the Lystrosaurus and the Cynognathus assemblage zones (Beaufort Group: Karoo Supergroup). Bulletin of the Council for Geoscience, Pretoria 137, 1–165. Novikov, I.V., Sues, H.D., 2004. Cranial osteology of Kapes (Parareptilia: Procolophonidae) from the Lower Triassic of Orenburg Province, Russia. Neues Jahrbuch für Geologie und Paläontologie, Abhandlungen 232, 267–281. Olson, E.C., 1947. The family Diadectidae and its bearing on the classification of reptiles. Fieldiana: Geology 11, 1–53. Pi˜neiro, G., Rojas, A., Ubilla, M., 2004. A new procolophonoid (Reptilia, Parareptilia) from the upper Permian of Uruguay. Journal of Vertebrate Paleontology 24, 814–821. Reisz, R.R., Scott, D., 2002. Owenetta kitchingorum, sp. nov., a small parareptile (Procolophonia: Owenettidae) from the Lower Triassic of South Africa. Journal of Vertebrate Paleontology 22, 244–256. Reisz, R.R., Sues, H.D., 2000. Herbivory in late Paleozoic and Triassic terrestrial vertebrates. In: Sues, H.D. (Ed.), Evolution of Herbivory in Terrestrial Vertebrates: Perspectives from the Fossil Record. Cambridge University Press, Cambridge, pp. 9–41. Romer, A.S., 1956. Osteology of the Reptiles. University of Chicago Press, Chicago, 772 pp. Smith, R., Botha, J., 2005. The recovery of terrestrial vertebrate diversity in the South African Karoo Basin after the end-Permian extinction. Comptes Rendus Palevol 4, 555–568. Sues, H.D., Olsen, P.E., Scott, D.M., Spencer, P.S., 2000. Cranial osteology of Hypsognathus fenneri, a latest Triassic procolophonid reptile from the Newark Supergroup of eastern North America. Journal of Vertebrate Paleontology 20, 275–284.