Tanystrosuchus- history and relationships

Everyone's talking about Sanjuansaurus, but the interesting Triassic maybe-theropod is surely Tanystrosuchus posthumus.  Generally passed off as an indeterminate theropod or halticosaur, Kuhn might have been right to give it a -suchus suffix after all.

Tanystrosuchus Kuhn, 1963
T. posthumus (Huene, 1908) Kuhn, 1963
= Tanystropheus posthumus Huene, 1908
= Thecodontosaurus posthumus (Huene, 1908) Fraas, 1913
= Halticosaurus posthumus (Huene, 1908) Huene, 1932
= Coelophysis posthumus (Huene, 1908) Olshevsky, 1991

Norian, Late Triassic
Middle Stubensandstein, Germany
Holotype- (SMNS 4385) distal caudal vertebra (39 mm)

Diagnosis- (after Huene, 1908) no notch between distal caudal prezygapophysis and centrum.
Other diagnoses- Huene (1908) also distinguished posthumus from Tanystropheus conspicuus and T. antiquus based on its supposedly enlarged and elongated postzygapophyses, while the prezygapophyses were said to be rudimentary. Yet he had the vertebra oriented backwards, as it is actually the prezygapophyses which are elongate (65% of central length). Comparably elongate prezygapophyses are known in other taxa such as Effigia, herrerasaurids and many neotheropods. Additionally, the neural spine was said to be more reduced than other Tanystropheus species, and the ventral surface has a median groove instead of being flat and/or keeled. The former is typical of theropod distal caudals while the latter is also found in Effigia and most avepods.
Huene (1932) states a unique character is the lack of elongation, but this varies continuously in most archosaur caudal series.

Tanystrosuchus posthumus holotype distal caudal vertebra SMNS 4385 in right lateral (A) and dorsal (B) views (after Rauhut and Hungerbuhler, 2000). Scale = 10 mm.

Comments- Collected in the 1860's, this was first described and figured by Meyer (1865) as the caudal vertebra of an unknown reptile. Huene (1908) named it Tanystropheus (consistantly misspelled Tanystrophaeus) posthumus, as the elongate cervicals of Tanystropheus were thought to be theropod caudals at the time. SMNS 4385 was noted to be similar to Tanystropheus cervicals in being elongate, lacking transverse processes and having a reduced neural spine. These characters led Huene to refer Tanystropheus to Coeluridae, though they are now recognized as typical features for theropod caudal vertebrae. Today, coelurids and other related basal coelurosaurs are known to be restricted to the Jurassic and Cretaceous and have shorter prezygapophyses than Tanystrosuchus. Both Meyer and Huene had the vertebra backwards, interpreting its elongate prezygapophyses as postzygapophyses.

Fraas (1913) thought much of the Pfaffenhofen quarry material was probably referrable to a small thecodontosaur, which he provisionally used the combination Thecodontosaurus (misspelled Thekodontosaurus) posthumus for. However, posthumus is from the Heslach quarry, not Pfaffenhofen. Fraas did not provide evidence for his reassignment, and Tanystrosuchus differs supposed Thecodontosaurus distal caudals (e.g. YPM 56736) in having a ventral groove, smaller more medially placed postzygapophyses, and larger, longer prezygapophyses.

By 1932, Huene had oriented the vertebra correctly and removed posthumus from Tanystropheus, since new remains of the latter had shown it was a nondinosaurian reptile whose elongated vertebrae were cervicals. He instead called it "Tanystropheus" (gen. indet.) posthumus and assigned it to Coelurosauria, although in one table it is listed as Halticosaurus posthumus. This may have been mistakenly retained from an earlier version of the paper, since he does say posthumus could belong to Halticosaurus or Dolichosuchus. Steel (1970) also suggested posthumus might be referrable to Halticosaurus, while Olshevsky (1991) listed it as possibly being Halticosaurus or Liliensternus (which was placed in Halticosaurus until 1984).

While Halticosaurus and Dolichosuchus are known from the same formation (but different quarries), they do not preserve distal caudal vertebrae so cannot be compared. Liliensternus loses its transverse processes on caudal 22 and its neural spines at about caudal 30. This fits with Tanystrosuchus' holotype being most similar in proportions to caudal 32 of those illustrated for Liliensternus. It differs from Liliensternus in having longer prezygapophyses (65% of central length compared to 17-22%), less dorsally projected zygapophyses, and lacking a notch between the prezygapophysis and centrum. These seem to be true in all illustrated caudals similar in position to Tanystrosuchus' (20, 24, 32, 38), so probably indicate it is not synonymous.

Kuhn (1963) created the new genus Tanystrosuchus for posthumus, as he also decided it was not referrable to Tanystropheus. Wild (1973) also rejected the synonymy of posthumus with Tanystropheus conspicuus. Tanystrosuchus differs from Tanystropheus in having longer prezygapophyses and lacks Tanystropheus' tall bladelike neural spine. In 1965, Kuhn stated Tanystrosuchus might be protorosaurid instead of dinosaurian. Yet Protorosaurus distal caudals differ in having short prezygapophyses, longer postzygapophyses, knob-like transverse processes and a tall bifurcated neural spine.

Norman (1990) noted the long prezygapophyses were suggestive of theropod relationship, but considered it a nomen dubium. Olshevsky (1991) placed Tanystrosuchus in Halticosauridae and listed the combination Coelophysis posthumus as a prior synonym, though I have yet to locate this in an earlier published work. Glut (1997) merely listed the genus as a nondinosaurian reptile. Most recently, Rauhut and Hungerbuhler (2000) briefly redescribed and illustrated the material, listing it as Tanystrophaeus posthumus. Their conclusions match Norman's- that it is an indeterminate theropod based on the prezygapophyseal length.

Tanystrosuchus is similar to coelophysoids in general form- elongate amphicoelous centrum (3.25 times longer than tall) with median ventral groove, neural spine reduced to a slight ridge over the postzygapophyses, transverse process reduced to a longitudinal ridge, elongate prezygopophysis and short postzygapophysis. However, the prezygapophysis is actually longer than in coelophysids, 65% of central length compared to 35% in Megapnosaurus, 25% in Coelophysis, 22% in Liliensternus and 37% in Dilophosaurus. In this respect Tanystrosuchus is more similar to herrerasaurids and neotheropods (e.g. Elaphrosaurus). It differs from herrerasaurids in having the ventral groove and from known neotheropods in being Triassic in age, though if coelophysoids are monophyletic we would expect Norian neotheropods. Another perhaps more plausible identification is as a shuvosaurine, which are common in the Late Triassic. Effigia has distal caudals which exhibit the same characters noted above for coelophysoids (including the ventral groove), except its prezygapophyses are longer (at least 58% of central length). The preserved caudals have neural spines, albeit low ones, but these are all proximal to the thirtieth caudal and it is probable that more distal ones lacked neural spines as in theropods and Tanystrosuchus. It differs from all examined theropods and Effigia in lacking a notch between the prezygapophysis and centrum.

While the anatomy favors a shuvosaurine or neotheropod identification, stratigraphy favors the former only to the extent that Triassic shuvosaurines are certainly known while Triassic neotheropods are unknown but possible depending on theropod topology. Also a factor is that variation along the tail is poorly described for most taxa discussed here, so while the published evidence suggests coelophysoid prezygapophyses never exceed half of centrum length and herrerasaurids lack ventral grooves, this is not as well established as it could be. For now, I recommend Tanystrosuchus posthumus be referred to Archosauria incertae sedis.

References- Meyer, 1865. Reptilien aus dem Stubensandstein des oberen Keupers (Dritte Folge). Palaeontographica. 14, 99-124.

Huene, 1908. Die Dinosaurier der europäischen Triasformation mit Berücksichtiging der aussereuropäischen Vorkommnisse [The dinosaurs of the European Triassic Formation, with consideration of non-European occurrences]. Geologische und Paläontologische Abhandlungen Supplement-Band. 1, 419 pp.

Fraas, 1913. Die neuesten Dinosaurierfunde in der schwäbischen Trias [The newest dinosaur finds in the Swabian Trias]. Naturwissenschaften. 1(45), 1097-1100.

Huene, 1932. Die fossile Reptil-Ordnung Saurischia, ihre entwicklung und geschichte. Monographien zur Geologia und Palaeontologie. 1, 1-362.

Kuhn, 1963. Sauria (Supplementum I). In Fossilium Catalogus I. Animalia. 104. 87 pp.

Kuhn, 1965. Saurischia (Supplementum 1). In Fossilium Catalogus 1. Animalia. 109, 94 pp.

Steel, 1970. Part 14. Saurischia. Handbuch der Paläoherpetologie. Gustav Fischer Verlag, Stuttgart. 1-87.

Wild, 1973. Die Triasfauna der Tessiner Kalkalpen. XXIII. Tanystropheus longobardicus (Bassani) (Neue Ergebnisse). Schweizerische Palaontologische Abhandlungen. 95, 1-162.

Norman, 1990. Problematic Theropoda: "Coelurosaurs". in Weishampel, Dodson and Osmolska (eds). The Dinosauria. University of California Press: Berkeley. 280-305.

Olshevsky, 1991. A Revision of the Parainfraclass Archosauria Cope, 1869, Excluding the Advanced Crocodylia. Mesozoic Meanderings. 2, 196 pp.

Glut, 1997. Dinosaurs - The Encyclopedia. McFarland Press, Jefferson, NC. 1076 pp.

Rauhut and Hungerbuhler, 2000. A review of European Triassic theropods. Gaia. 15, 75-88.