Baryonychidae Charig and Milner, 1986

Rauhut, Oliver W. M., 2003, The interrelationships and evolution of basal theropod dinosaurs, Special papers in palaeontology 69, pp. 1-213 : 34-36

Baryonychidae Charig and Milner, 1986

Included taxa. Angaturama limai Kellner and Campos, 1996 ; Baryonyx walkeri Charig and Milner, 1986 ; Irritator challenged Martill, Cruickshank, Frey, Small and Clark, 1996 ; Spinosaurus aegyptiacus Stromer, 1915 , partim; Suchomimus tenerensis Sereno, Beck, Dutheil, Gado, Larsson, Lyon, Marcot, Rauhut, Sadleir, Sidor, Varricchio, Wilson and Wilson, 1998 .

Temporal range. Barremian-Cenomanian.

Occurrence. Romualdo Member of the Santana Formation, Ceara, Brazil; Upper Weald Clay, Surrey, England; Baharia Formation, Marsa Matruh, Egypt; Elrhaz Formation, Agadez, Niger; Kem Kem beds, Morocco.

Diagnosis. Dentary with strongly developed anterior expansion; anterior dentary teeth much larger than the relatively small and closely spaced posterior teeth; medial alveolar border is as high as the lateral border and formed by a sheet of bone of the denary rather than by separately ossified interdental plates; teeth almost round in basal cross section and only slightly recurved; very long premaxillae, forming a rostral rosette; seven premaxillary teeth; ventral margin of premaxilla strongly concave; anterior ramus of maxilla strongly elongated; angle between anterior and ventral ramus of the lacrimal less than 45 degrees; dorsal vertebrae with several small vertical laminae connecting the transverse process with the neural spine dorsally; humerus extremely robust, with strongly expanded internal tuberosity and distal condyles; ulna with a broad and very strongly developed olecranon process; ischium with a long and low obturator flange.

Remarks. The family Baryonychidae ( Text-fig. 6b ) was proposed by Charig and Milner (1986) to include a single species, Baryonyx walkeri . Buffetaut (1989, 1992) noted similarities between the jaws of Baryonyx and the slightly younger Spinosaurus from the Cenomanian of Egypt, and concluded that both animals probably belong to the family Spinosauridae . This view has since been accepted by several authors (e.g. Sereno et al. 1994, 1996, 1998). Although Charig and Milner (1997) retained Baryonyx in its own family, they acknowledged the similarities between this genus and Spinosaurus and united the Baryonychidae and the Spinosauridae in the superfamily Spinosauroidea.

The matter is complicated by the problematic status of Spinosaurus aegyptiacus . The holotype of this genus was found in the Baharia Oasis of Egypt at the beginning of the twentieth century and was subsequently described by Stromer (1915). The material consisted of two dentaries, a maxillary fragment, and several vertebrae. The only other accounts of the original specimen were by Stromer (1934 b, 1936), before all the material was destroyed during a bombing raid in 1944.

In the introduction to his original description Stromer (1915, p. 3) commented on the association and preservation of the specimen:

‘Many of the bones had already been broken and deformed in the sediment; furthermore, the remains were completely jumbled. The skull seems to have originally been present, but due to its exposed position had been almost completely eroded away. Especially the maxillary fragment shows signs of erosion of an exposed bone; cracks in the posterior ends of the lower jaws and the especially anteriorly eroded cervical vertebrae also indicate a superficial position of the fossils.

No parts of the appendicular skeleton were found, indicating that the skeleton was already incomplete when it was covered by sediment. The soft parts had surely decayed before burial, thus allowing the dislocation of the bones. Several teeth, some of them already loosened by erupting replacement teeth, fell out of the jaws, complete with their roots, and all the bones were jumbled, perhaps due to scavengers or water flow. However, given the preservation of fine details and the lack of abrasion, prolonged transport of the bones can be excluded. In the sediment, the fossils suffered from pressure, probably caused by gypsum- and salt-flow rather than by tectonic events.

Given their positions in the sediment and their preservation, the remains represent one individual. Only the size of the sacral and caudal vertebrae are problematic in this respect, which will be discussed in the respective descriptions.’ (my translation).

Stromer later (1934 b, p. 21) stated that: ‘It remains questionable, if... especially the oversized anterior caudal vertebra n... belongs to Spinosaurus at all.’ (my translation). Judging from Stromer’s illustration (1915, pl. 1, fig. 1a -b) the vertebra in question looks more like an o rn ithischian caudal than a theropodan element. Given this probable mix of taxa in the original type material, it cannot be excluded that the rest of the material also represents more than one taxon. The following points may indicate that the holotype of S. aegyptiacus actually represents a mixture of remains of different theropods:

1. The dentary figured by Stromer (1915, pl. 1, figs 6, 12) agrees with dentaries of Baryonyx and material referred to Suchomimus in almost every detail and shows all the baryonychid synapomorphies listed above. 2. The dorsal vertebrae of the type of Spinosaurus aegyptiacus lack the strong pneumatisation and additional laminae seen in both Baryonyx and Suchomimus . Their principal characters are comparable to those seen in allosauroids, with the exception of the strongly elongated neural spines. Elongated neural spines are present in the carcharodontosaur Acrocanthosaurus , though to a lesser degree than in S. aegyptiacus (Stovall and Langston 1950) . Since large carcharodontosaurs are present in Baharia, the vertebrae might represent these animals.

3. The cervical vertebrae figured by Stromer (1915, pl. 2, figs 1-2) differ significantly from the dorsal vertebrae in respect of the height of their neural spines. This difference is so marked that it might be questionable that they belong to the same taxon, although an abrupt change in spine height is present in other high-spined dinosaurs (e.g. Ouranosaurus ; Taquet 1976).

Given this uncertainty of the association of the holotype material of Spinosaurus aegyptiacus , it seems at present better to use the family name Baryonychidae than the name Spinosauridae until new material becomes available to clarify the status of Spinosaurus . However, the dentary of the type of Spinosaurus can at least be referred to the Baryonychidae .

Suchomimus tenerensis from the Aptian of Niger is based on a partial postcranial skeleton (MNN GDF 500) and several referred specimens, including a partial skull (MNN GDF 501; see Sereno et al. 1998). All the material is very similar to Baryonyx , and can thus be referred to the Baryonychidae with certainty.

Two baryonychids from South America were described in 1996: Irritator challengeri , based on a skull lacking the anterior part of the snout (SMNS 58022), by Martill et al., and Angaturama limai , based on the tip of a snout, by Kellner and Campos. Irritator (SMNS 58022) shows the diagnostic angle in the lacrimal, as well as a braincase that is very similar to that of Baryonyx and can, therefore, be referred to the Baryonychidae . Angaturama can also be referred to this clade, based on the presence of a rostral rosette and a premaxilla with seven tooth positions ( Kellner 1996; Kellner and Campos 1996). Since both specimens come from the same horizon, they might represent the same taxon (in which case the name Irritator challengeri would be the senior synonym), but more material is needed to confirm this (see Kellner 1996).

The recently described species Spinosaurus marocannus Russell, 1996 , and Cristatusaurus lapparenti Taquet and Russell, 1998 , are regarded as nomina dubia.

Figures

text-fig. 6. Skull reconstructions of representatives of Cretaceous OTUs in left lateral view, a, abelisaurid Camotaurus sastrei, Late Cretaceous (Campanian), Gorro Frigio Formation, Argentina; based on Bonaparte et al. (1990) and MACN CH 894. b, composite skull of a generalized baryonychid; mainly based on Suchomimus tenerensis (MNN GDF 501, 503-506) and Irritator challenged (SMNS 58022), with some elements reconstructed after Baryonyx walkeri (BMNH R 9951). c, carcharodontosaurid Carcharodontosaurus sahariens, Late Cretaceous (Cenomanian), Kem Kem beds, Morocco; based on Sereno et al. (1996) and SGM-Din 1, unpreserved elements shaded. D, dromaeosaurid Velociraptor mongoliensis, Late Cretaceous (Campanian), Djadokhta Formation, Mongolia; based on Barsbold (1983), Barsbold and Osmólska (1999), and AMNH 6515. e, ornithomimosaur Dromiceiomimus brevitertius, Late Cretaceous (Maastrichtian), Horseshoe Canyon Formation, Alberta, Canada; redrawn from Russell (1972). F, oviraptorosaur Oviraptor philoceratops, Late Cretaceous (Campanian), Djadokhta Formation, Mongolia; redrawn from Barsbold et al. (1990). G, therizinosauroid Erlikosaurus andrewsi, Late Cretaceous (Cenomanian- Turonian), Baynshiren Svita, Mongolia; based on Clark et al. (1994). H, tyrannosaurid Gorgosaurus libratus, Late Cretaceous (Campanian), Judith River Formation, Alberta, Canada, redrawn from Molnar et al. (1990). Abbreviations as in Text-figures 4 and 5, and: acc. op., accessory opening of unknown function in the premaxilla of Oviraptor (Barsbold et al. 1990); amf, anterior maxillary foramen; pn, pneumatic openings. Scale bars represent 50 mm (d-g) and 100 mm (a-c, h).