Discovery and naming
In the autumn of 1834 surgeon Henry Riley and curator of the Bristol Institution Samuel Stutchbury began to excavate "saurian remains" at the quarry of Durdham Down, at Clifton, presently a part of Bristol. In 1834 and 1835 they briefly reported on the finds. They provided their initial description in 1836, naming a new genus: Thecodontosaurus. The name is derived from Greek thekè, "socket", and odous, "tooth", a reference to the fact that the roots of the teeth were not fused with the jaw bone, as in present lizards, but positioned in separate tooth sockets. Thecodontosaurus was the fifth dinosaur named, after Megalosaurus, Iguanadon, Straeptospondylus
and Hylaeosaurus, though Riley and Stutchbury were not aware of this, the very concept Dinosauria only being created in 1842. In 1843 John Morris in his catalogue of British fossils provides a complete species name: Thecodontosaurus antiquus. The specific epithet, "antiquus", means "ancient" in Latin. The original type specimen or holotype of Thecodontosaurus, BCM 1, a lower jaw, fell victim to heavy World War II bombings by the Germans. Many remains of this dinosaur and other material related to it were destroyed in November 1940 during the Bristol Blitz. However, most bones were salvaged: 184 are today part of the collection of the Bristol City Museum and Art Gallery. Also, later some more remains were found near Bristol at Tytherington. At present in total about 245 fragmentary specimens are known, representing numerous individuals. In 1985 Peter Galton designated another lower jaw, a right dentary, as the neotype, BCM 2. The remains have been found in chalkstone infillings, breccia deposited in fissures in older rocks. The age of these deposits was once estimated as old as the late Carnian but recent studies indicate that they date from the Rhaetian.
Apart from the original type species, Thecodontosaurus antiquus, seventeen other species would be named in the genus.
Riley and Stutchbury also found some teeth of carnivorous phytosaurians that they named Paleosaurus cylindrodon and P. platyodon. In the late nineteenth century the theory became popular that such remains would have belonged to carnivorous prosauropods: animals with the body of Thecodontosaurus but with heads armed with slicing teeth. In 1890 Arthur Smith Woodward accordingly named a Thecodontosaurus platyodon, in 1908 Friedrich von Huene a Thecodontosaurus cylindrodon. Though still defended by Michael Cooper in 1981, the hypothesis that such creatures existed has now been totally discredited. On one occasion material of Thecodontosaurus was by mistake described as a separate genus. In 1891 Harry Govier Seeley named Agrosaurus macgillivrayi, assuming the remains had been collected in 1844 by the crew of HMS Fly on the northeast coast of Australia. It was long considered the first dinosaur found in Australia, but in 1999 it transpired that the bones probably belonged to a lot sent by Riley and Stutchbury to the British Museum of Natural History and then mislabelled. Already in 1906 von Huene had noted the close resemblance and renamed the species Thecodontosaurus macgillivrayi. It is thus a junior synonym of Thecodontosaurus antiquus. It also happened that remains of completely different animal groups were referred to Thecodontosaurus. Thecodontosaurus latespinatus and Thecodontosaurus primus, both named by von Huene in 1905, were based on material from Tanystropheus. In 1908 von Huene tentatively renamed Plateosaurus elizae Sauvage 1907 into Thecodontosaurus? elizae; it is a nomen dubium based on theropod remains. The proterosuchid Ankistrodon Huxley 1865 was renamed Thecodontosaurus indicus. A larger category consists of related basal sauropodomorphs assigned to Thecodontosaurus. These include in chronological order: Thecodontosaurus gibbidens Cope 1878; Thecodontosaurus skirtopodus Seeley 1894; Thecodontosaurus polyzelus (Hitchcock 1865) von Huene 1906; Thecodontosaurus hermannianus von Huene 1908; Thecodontosaurus diagnosticus Fraas 1912; Thecodontosaurus minor Haughton 1918 (= "Thecodontosaurus minimus"); Thecodontosaurus dubius Haughton 1924; Thecodontosaurus browni (Seeley 1895) von Huene 1932 and Thecodontosaurus alophos Haughton 1932. All these are today referred to other genera or seen as nomina dubia. The last of these is Thecodontosaurus caducus named by Adam Yates in 2003 for a juvenile specimen found in Wales; in 2007 this was made the separate genus Pantydraco.
Presently, the only valid species is thus T. antiquus.
From the fragmentary remains of Thecodontosaurus most of the skeleton can be reconstructed, the main exception being the front of the skull. Thecodontosaurus had a rather short neck supporting a fairly large skull with large eyes. Its jaws contained many small- to medium-sized, serrated, leaf-shaped teeth. This dinosaur's hands and feet each had five digits, and the hands were long and rather narrow with an extended claw on each. This dinosaur's front limbs were much shorter than the legs, and its tail was much longer than the head, neck and body put together. On average, it was 1.2 metres (3.9 ft) long, 30 centimetres (12 in) tall, and weighed 11 kilograms (24 lb). The largest individuals had an estimated length of 2.5 metres (8.2 ft). Michael Benton in 2000 noted the existence of a robust morph in the population, seen by him as a possible second species or, more likely, an instance of sexual dimorphism. Benton also indicated some unique derived traits, or autapomorphies for the species: a long basipterygoid process on the braincase; a dentary that is short in relation tot the total length of the lower jaw; an ilium that has a back end that is subquadrate instead of rounded. The small size has been explained as an instance of insular dwarfism.
Riley and Stutchbury originally saw Thecondontosaurus as a member of the Squamata, the group containing lizards and snakes. This did not change when Richard Owen in 1842 coined the Dinosauria, because Owen did not recognise Thecodontosaurus as a dinosaur; in 1865 he assigned it to the Thecodontia. Only in 1870 was Thomas Huxley the first to understand it was a dinosaur, though referring it incorrectly to the Scelidosauridae. Later it would normally be placed in either the Anchisauridae or its own Thecodontosauridae. Modern exact cladistic analyses have not been conclusive. Although not actually the earliest member of the group, Thecodontosaurus is sometimes placed in a very basal position among the sauropodomorph dinosaurs. It was earlier included under the Prosauropoda but more recently it has been suggested that Thecodontosaurus and its relatives were prior to the prosauropod-sauropod split. New reconstructions show that its neck is proportionally shorter than in more advanced early sauropodomorphs.