Genus: Agnostiphys Fraser, Padian, Walkden, & Davis , 2002

Classification: Dinosauria - Saurischia - Theropoda - Herrerasauria

Etymology: Greek, unknown or uncertain, with reference to the position of the new form relative to the Dinosauria.

Synonyms: none

Type species: A. cromhallensis Fraser, Padian, Walkden, & Davis , 2002

Other Species: none

Diagnosis: as for type species

Species: Agnostiphys cromhallensis Fraser, Padian, Walkden, & Davis , 2002

Etymology: The specific name refers to the locality.

Synonyms: none

Holotype: VMNH 1745, left ilium.

Referred Specimens: VMNH 1751, left maxilla; 1748, left astragalus, 1749, right astragalus; and 1750, right humerus. AUP 11095A, isolated tooth.

Time: Upper Triassic fissure deposits.

Horizon: Cromhall Quarry

Location: Avon, England.

Total length: ~ 1m.

Mass: ?

Diagnosis: Dinosauromorph with a well-de®ned brevis fossa on the ilium; semi-perforate acetabulum; `kidney-shaped' antitrochanter; well-developed posterior portion of the iliac blade; two sacral vertebrae; subrectangular deltopectoral crest that is 33 per cent of the length of the humerus; astragalus with a distinct ascending process and a prominent depression immediately posterior to the ascending process; in dorsal aspect an acute anteromedial corner on the astragalus (Fraser, Padian, Walkden, & Davis , 2002).

Comments: Agnostiphys cromhallensis is described from the Triassic of south-west England. The description is based on isolated elements recovered from an assemblage of other dissociated tetrapod remains that include crocodylomorphs, rauisuchiforms and sphenodontians. The key elements in the new taxon are the ilium, astragalus, and the humerus, and these exhibit ®ve synapomorphies of Dinosauria. Three of these, namely a largely to fully perforate acetabulum, the presence of a brevis fossa, and a reduced astragalus with an ascending process, are considered to be particularly relevant. The de®nition and diagnosis of the Dinosauria are restated and the positions of the new form, herrerasaurs and Eoraptor relative to true dinosaurs are discussed (Fraser, Padian, Walkden, & Davis , 2002).

This dinosauriform from Cromhall cannot be unequivocally regarded as dinosaurian, but in many regards it is closer to true dinosaurs than Eoraptor and herrerasaurs appear to be. In our view, Eoraptor and
herrerasaurs are dinosauriforms, closer to true dinosaurs than other ornithodiran taxa, but are probably not true dinosaurs because they lack many synapomorphies that have been proposed to unite Saurischia and Ornithischia. To accept the view that they are true dinosaurs requires the reversal of several characteristics that appear difficult or unusual to reverse because the characters are often strongly involved in functional complexes, and reversals of such features are not known in other taxa. Of the 17 potential synapomorphies of dinosaurs listed above, based on uncontroversial saurischians and ornithischians, five
are present in Agnostiphys, and only one (three sacral vertebrae) does not seem to be present. Whether or not Agnostiphys is a true dinosaur, it is at least as close to dinosaurs as are Eoraptor and herrerasaurs. The test of the evolution of features that characterized the origin of dinosaurs must be made with reference to a node-based evolution of Dinosauria and a diagnosis that follows from this definition. Future discoveries will test the diagnosis of Dinosauria and the relationship of other taxa to dinosaurs, and will improve understanding of how dinosaurian adaptations were sequentially assembled. But the definition of Dinosauria, which is based on its ancestry, cannot change (Fraser, Padian, Walkden, & Davis , 2002).

Illustrations:
Agnosphitys cromhallensis, left maxilla, VMNH 1751, in A, lateral and B, medial views. Scale bar represents 5 mm (modified from Fraser, Padian, Walkden, & Davis , 2002).

References:
Fraser, N. C., K. Padian, G. M. Walkden, & A.L.M. Davis , 2002. Basal dinosauriform remains from Britain and the diagnosis of the Dinosauria. Palaeontology , 45 (1): 79-95

© Tetrapoda Database Roman Ulansky roman.ulansky@gmail.com or adios85@mail.ru