Geranosaurus atavus BROOM 1911

GERANOSAURUS ATAVUS BROOM, 1911

Diagnosis: The type and only specimen of Geranosaurus atavus is poorly preserved and fragmentary, and is not diagnosable on the basis of unique characters, or on the basis of a unique combination of characters, and is regarded as a nomen dubium; the holotype and referred specimens are considered Heterodontosauridae indet.

Holotype: SAM-PK-1871 (Fig. 37A) a heavily eroded skull currently preserved in two parts, the first embedded in a block of sandstone includes a predentary and rostral ends of both dentaries (Broom, 1911: plate 17, fig. 24), whereas the second consists of a partial maxillary tooth row with crowns heavily damaged.

Referred specimen: SAM-PK-1857, a poorly preserved partial hindlimb, referred to Geranosaurus by Broom. Broom (1911: 306) also mentioned ‘a number of very imperfect vertebrae’. The whereabouts of all these specimens is unknown.

Horizon and locality: Holotype and referred specimens were collected from the Clarens Formation (‘ Cave Sandstone’ Early Jurassic), road-cutting near the summit of Barkly Pass (31°27′S, 27°51′E; Kitching & Raath, 1984: table 1), Farm Tulloch, Elliot, Eastern Cape Province, South Africa.

Discussion: The holotype specimen of Geranosaurus is poorly preserved (Fig. 37A). The crowns of all of the dentary teeth are missing and the crowns of the maxillary teeth are broken immediately above their bases. The dentaries are highly fractured. The specimen has evidently been damaged since its original description by Broom (1911): the enlarged dentary caniniform is now missing and the maxillary teeth also appear to have been damaged because Broom (1911: 307) described them as having ‘... flat, chiselshaped crowns with the outer face feebly ridged’; these features can no longer be confirmed.

The presence of a predentary that is wedge-like (lacking a well-defined median ventral process) and a dentary caniniform, and the apparent absence of replacement foramina, confirm the heterodontosaurid affinities first recognized by Crompton & Charig (1962). Crompton & Charig suggested that Geranosaurus differed from H. tucki in lacking a large diastema between the premaxilla and maxilla and the absence of an inset maxillary tooth row. Hopson (1980) suggested that the maxillary tooth row was inset, but poor preservation means the latter character is difficult to establish with any confidence. Geranosaurus is distinct from Abrictosaurus in possessing an enlarged dentary caniniform. However, neither autapomorphies nor a unique combination of characters must be recognized and so, taxonomically, Geranosaurus atavus can be considered to be a nomen dubium (Thulborn, 1974, 1978; Hopson, 1980; Weishampel & Witmer, 1990; Norman et al., 2004c).