Author's note: these pages were written some years ago. I am not planning to update them. For a more current coverage, see the link to palaeos com Palaeos website (to which many links on these pages point to anyway. More info here

Superfamily Coelophysoidea

Liliensternus (subfamily Dilophosaurinae
Liliensternus liliensterni, artwork copyright © M. Shiraishi, reproduced with permission.

Coelophysis bauri - skull The Coelophysoidea represent the earliest theropod radiation, flourishing during the late Triassic and early Jurassic. These graceful animals were characterised by a long, low body, a long neck, and short three-fingered hands for clutching prey. The long fairly lightly built skull (left) bears a distinctive kink between the premaxilla (upper jaw tip) and maxilla (in main upper jaw). It was thought that this kink meant that coelophysids could not attack live prey due to skull weakness, but it appears to have been reinforced. Gregory Paul argued that Spinosaurid Megalosaurs like Baryonyx, which also have such a kink, were be late coelophysoids, but this is now not considered the case.

Coelophysoids probably died out during the Toarcian turnover, perhaps due to the extinction of the prosauropod dinosaurs (their main food source perhaps) at the same time. It is possible that some stragglers (e.g., Elaphrosaurus) continued on to the late Jurassic. It can be assumed that the Coelophysoidea were the ancestors of all later theropod dinosaurs.

drawing of Coelophysis skull courtesy of Professor Paul Olsen

Together with Ceratosauridae and the Abelisauroidae, the Coelophysoidea are included in the taxa Ceratosauria, ( external link  link to palaeos com strict definition here), generally considered a monophyletic clade (no descendents). It is I feel more likely that earlier and unspecialised Coelophysoids gave rise to more advanced theropods, which would then have lost their original ceratosaurian characteristics. It is not even certain that the three component groups of Ceratosaurs even belong together.

The following is a very tentative and speculative cladogram, which differs a little from other cladograms for this group (mainly in speculating on links to later taxa):

<<-o COELOPHYSIODEA [?=Podokesauroidea]
   `--o Procompsognathidae
      |-+- Procompsognathus triassicus
      |  `-- Segisaurus halli 
      | ?--Elaphrosaurus bambergi
      | ??- ABELISAURIA 
      `+-o-Coelophysidae [= Podokesauridae]
       `--o Dilophosaurinae [= Halticosaurinae]
          |--+ Gojirasaurus quayi
          |  |- Dilophosaurus wetherilli
          |  `?-+- Dilophosaurus sinensis
          |      `?- MEGALOSAURIA
          |  ?-- Liliensternus airelensis
          `--+-? Liliensternus orbitoangulatus 
             `-+- Liliensternus liliensterni
               `-o Coelophysinae
                  `-+?- Eucoelophysis baldwini
                    |-?- Coelophysis holyokensis [=Podokesaurus]  
                    |-- Coelophysis bauri 
                    `--+--Syntarsus rhodesiensis 
                       `-- Syntarsus kayentakatae 

family Procompsognathidae (=Segisauridae)

The Procompsognathids and Segisaurids, previously considered seperate families, are now sometimes included under the Coelophysidae. These are both small lightly built forms, not unlike Coelophysis. Very little is known of the procompsognathids. There are currently two species, each based on a single poorly preserved partial postcranial skeleton.

possible Procompsognathus skull
Procompsognathus skull from Rodney Steele, Saurischia

Procompsognathus triassicus Fraas, 1913

Horizon: Middle Keuper of Württemberg, Germany
Age: middle link to palaeos com Norian
Place: middle-north link to palaeos com Pangea
Remains: partial skeleton. A skull may also belong to this species
Length: 1.1 meter
Weight: 1 kg

Comments: Procompsognathus possess relatively broad pubes (hip-bones), a primitive feature that distinguishes it from all other Theropods. The previously associated skull does not to belong with the type specimen (the postcrania), although it is possible it belongs to this species. It is low and elongate, about 8 cm long, and appears to lack the distinctiove kink. In all other respects it resembles Coelophysis. The teeth are relatively large.

Segisaurus halli Camp, 1936

Horizon: Navajo Sandstone of Arizona
Age: link to palaeos com Toarcian
Place: middle-northwest link to palaeos com Pangea
Remains: partial skeleton
Length: about 1.3 meters
Weight: 5 kg?

Comments: Previously placed in its own family, the Segisauridae, this is a small, very slender animal. Unusual for the fact that some of its limb bones are not hollow, like all other small theropods, but solid. I wonder whether this might not be an artifact of preservation, or a pathological specimen, not hollow. Also, for a long time it was unusual for being one of the few theropods for which a good clavicle was known. Its hip bones have unusual holes and flattening. According to Gregory Paul, it also possesses some advanced features.

Family Coelophysidae


These early theropod dinosaurs are generally divided into two groups, the smaller Coelophysines and the larger Dilophosaurines (also called Halticosaurines). Althiough it would logically seem that the small Coelophysines evolved from the big Dilophosaurines, in fact the Dilophosaurs are more primitive, and a recent revision of external link Triassic stratigraphy by S. G. Lucas Coelophysis bauri to be late Norian or Rhaetic (latest Triassic) in age, rather than late Carnian as origionally believed. The implication then is that the Dilophosaurs appeared first, and after some fifteen million years or so gave rise to the smaller Coelophysines. In any case, both lineages existed alongside each other for some 20 million years or more. Both were graceful, active predators, both lineages continued for several tens of millions of years, indicating that these animals were very successful in their environment. Coelophysids show a tendency for head ornamentation in the form of crests, almost certainly for intraspecific display

subfamily Dilophosaurinae

(= Halticosaurinae)

Dilophosaurus sketch by Gregory S. Paul

The Dilophosaurs, or Haltcosaurs, include some of the earliest theropod dinosaurs. They were also the first of the great dinosaurian carnivores. These were large graceful bipedal preditors with lengths of upto 6 to 7 metres. Evolving, one would assume, from a small Procompsognathid ancestor, these animals very quickly attained an optimal size. Even the earliest forms, such as Liliensternus liliensterni and Gojirasaurus quayi equalled in size later and more specialised types like the early Jurassic Dilophosaurus wetherilli; all attaining estimated adult weights of 300 to 500 kg. Dilophosaurus is also distinctive for having some of the largest teeth relative to head size of all theropods. These clearly succesful animal sontinued with little change for some 40 million years

Gojirasaurus quayi Carpenter, 1997

Horizon: Dockum formation of New Mexico
Age: ?late link to palaeos com Carnian-early link to palaeos com Norian
Place: middle-north-west link to palaeos com Pangea
Remains: partial skeleton
Length: about 6 meters
Weight: about 200 kg

Comments: One of the largest and also one of the earliest theropods of the Triassic, Gojirasaurus has the honour of being named after the Japanese movie monster Godzilla (Gojira).


Dilophosaurus wetherilli

Dilophosaurus skeleton and image at head (below left) fromexternal linkDINOSAURS AND THE HISTORY OF LIFE - GEOLOGY V1001x site - Professor Paul Eric Olsen

Dilophosaurus wetherilli (Welles, 1954)

Megalosaurus wetherilli Welles, 1954
Horizon: Kayenta Formation, Arizona
Age: late link to palaeos com Sinemurian / early link to palaeos com Pliensbachian
Place: middle-north-west link to palaeos com Pangea
Remains: 2 subadult skeletons, two skulls, additional partial skeleton and fragmentary remains
Length: length 6 to 7 metres
Weight: weight 300 to 450 kg

DilophosaurusComments: The dominant preditor of the early Jurassic, this animal is distinguished by its large low build, unusually long teeth, and distinctive twin crests (left). Apart from its distinctive head-crest, Dilophosaurus is one of the most unspecialsed of the Coelophysids. The fact that this animal lived much later than otehr more advanced types shows that there were actually a number of coelophysid lines evolving in parallel over many millions of years.

This handsome animal has had the misfortune of being represented in the movie Jurassic Park as a sort of cross between a gremlin and a frill-necked lizard. Not only is the size in the movie version wrong (too small), but the lower jaw is the wrong shape as well. The fallacy that the delicate head crests and "weak" snout (actually now known to be reinforced) only capable of made it unable to tackle live prey may also have led to the most unkely assumption (which first appears in Michael Chrichton's novel) that it spat poison (no archosaur is known to). D. breedorum Welles vide Welles and Pickering, 1999 is probably a synonym. It has been suggested on basis of an imprint of a similarly-sized theropod sitting down that these animals may have been feathered.

Dilophosaurus sinensis Hu, 1993

Horizon: Lower Lufeng Series - Yunnan, China
Age: early Jurassic ( link to palaeos com Hettangian/ link to palaeos com Sinemurian/ link to palaeos com Pliensbachian)
Place: east link to palaeos com Pangea
Remains: skull and almost complete skeleton
Length: about 5 or 6 meters

Comments: same size as D. wetherilli but with taller, more robust crest. The skull is somewhat similar to that of Ceratosaurus. Recently, the placement of the species in this genus has been questioned. It has been proposed that it is actually a primitive relative of the Allosauria. Its crests are farther forward on the skull, and its scapula is expanded. It may well be that this species represents a transitional form between Dilophosaurs and higher theropods

Liliensternus liliensterni skull
Liliensternus liliensterni skull from Rodney Steele, Saurischia

Liliensternus liliensterni (von Huene, 1934)

synonym: Halticosaurus liliensterni Huene 1934
Horizon: upper Keuper of Thuringia, Germany
Age: late link to palaeos com Norian
Place: middle-north link to palaeos com Pangea
Remains: two subadult partial skeletons
Length: 5.15 meters (subadult)
Weight: 127 kg (adult to 500 kg?)

Comments: intermediate between the Coelophysines and Dilophosaurus in anatomy; both known specimens are juveniles, so adults may have attained about 7 meters.

Liliensternus airelensis Curry & Galton, 1993

Horizon: Rhaetian of France
Age: link to palaeos com Rhaetian
Place: middle-north link to palaeos com Pangea
Remains: tooth, vertebrae, partial pelvic girdle

Comments: originally just "Halticosaurus sp."

Liliensternus orbitoangulatus skull
Liliensternus orbitoangulatus skull from Rodney Steele, Saurischia

?Liliensternus orbitoangulatus (von Huene, 1932)

synonym: Halticosaurus orbitoangulatus Huene, 1932
Horizon: Stubensandstein of Württemberg, Germany
Age: mid or late link to palaeos com Norian
Place: middle-north link to palaeos com Pangea
Remains: imperfect skull and lower jaw

Comments: This poorly known animal would seem to be an earlier version of Dilophosaurus, with a smaller crest. It is not clear if it belongs in teh genus Liliensternus (or even if it really is a coelophysid)

Sinosaurus triassica Young 1948

Horizon: Dull Purplish Beds, Lower Lufeng Series of Yunnan, China
Age: link to palaeos com Hettangian?
Place: east link to palaeos com Pangea
Remains: Jaws and teeth
Length: Length: 6 metres?

Comments: The jaws and postcrania previously included under the taxon "Sinosaurus triassica" are now known to come from two totally different animals; the jaws and teeth (Sinosaurus triassica proper) from a large carnivore; either a Rauisuchian [Galton] (which would then make these beds Rhaetian in age) or a large theropod, probably a Dilophosaurid [Bakker, Sci Am], while the post-crania are from a large prosauropod dinosaur.

subfamily Coelophysinae

Coelophysis bauri  -  Coelophysinae
sketch by Greg S. Paul

The Coelophysines probably evolved from early Dilophosaurs. They were an important component of latest Triassic and early Jurassic vertebrate faunas. The best known early type is Coelophysis bauri (see above sketch), a small (about the size of an average dog) and very graceful creature known from the late Triassic Chinle formation of Arizona. Remains of a very similar form are known from early Jurassic rocks in South Africa. Since at this time all the land-masses were joined into a single supercontinent, link to palaeos com Pangea, it is likely that the Coelophysids had a world-wide distribution over many millions of years.

The better known Coelophysines consist of several closely related species, with very little to separate them. As far as the build and proportions of the skeleton go they are very is to those of the late Jurassic Compsognathus and proto-bird Archaeopteryx. In life these creatures probably had a very bird-like appearance (it is not known whether or not they had feathers, although an impression left by a Dilophosaurus-sized theropod may indicate belly feathers). These animals may have hunted in packs, like modern jackals.

Coelophysines are sometimes found in groups ranging from a couple to hundreds of individuals. This relative abundance of specimens makes it possible for two types to be distinguished; robust or gracile. These most likely correspond to sexual differences. Robust forms have relatively longer necks and skulls, thicker limbs, and more developed muscle attachments.

Coelophysis bauri

image courtesy Professor Paul Olsen

Coelophysis bauri(Cope, 1887)

synonyms: Coelurus bauri Cope 1887; Coelophysis bauri Cope, 1889c; Rioarribasaurus colberti, Camposaurus arizonensis Hunt, Lucas, Heckert, and Lockely, 1998
Horizon: Chinle Formation, Arizona, New Mexico, and Utah
Age: late link to palaeos com Norian or link to palaeos com Rhaetian (latest Triassic) [ref Lucas 1998, p.374]
Place: middle-north-west link to palaeos com Pangea
Remains: several hundred skeletons (including juveniles and adults, males and females)
Length: 2.5 to 3 meters
Weight: 15 to 30 kg

Comments: This was a pretty typical early unspecialised Coelophysine. It may have evolved from an animal like Liliensternus. It is best known from the hundreds of specimens from the Ghost Ranch quarry in New Mexico. It is believed that this animal practiced cannibalism (not uncommon among crocodilians), because some Ghost Ranch skeletons have juvenile-sized skeletons in the area of their stomachs. Eucoelophysis baldwini Sullivan and Lucas, 1999, based on a rediscription of Cope's original type specimen of "Coelurus" bauri (= Coelophysis bauri) appears to be a distinct, if closley related, species.

Syntarsus rhodesiensis Raath, 1969

Horizon: Upper Elliot Formation, Cape Province, South Africa; and Forest Sandstone of Matabeleland, Zimbabwe,
Age: link to palaeos com Hettangian/Sinemurian
Place: south link to palaeos com Pangea
Remains: At least 30 partially articulated skeletons, juvenile to adult.
Length: 2 to 3 metres;
Weight: 10 to 25 kg

Comments: A descendent of the Triassic Coelophysis, this active little hunter shows a number of more advanced features, such as a much larger "window" (the antorbital fenestra) in the front side of the snout, possibly for housing a gland of some kind, and teeth that are entirely forward of the eye. There is also the fusion of some of the upper foot-bones (metatarsals), a bird-like characteristic that would increase the strength of the leg. Gregory Paul suggests (Predatory Dinosaurs of the World) that this species should be included under Coelophysis, but in view of its more advanced characteristics it should perhaps be reatined a sa seperate genus

The behaviour appears to be the same as Coelophysis, for the remains of large numbers of this species, in various ages, are found concentrated together in a single small quarry area [Paul, p.262]. Like present-day birds, Syntarsus seems to have lived in large groups or flocks of individuals.

"Syntarsus" kayentakatae Rowe, 1989

Horizon: Kayenta Formation, Arizona
late link to palaeos com Sinemurian or early link to palaeos com Pliensbachian
Place: middle-north-west link to palaeos com Pangea
Remains: 16 individuals, juvenile to adult.
Length: 2 m

Comments: This species possesses two small crests, similar to those of it's larger cousin Dilophosaurus. Some have suggetse dthat because of this this species represents a new genus.

Coelophysis holyokensis Talbot, 1911 synonym: Podokesaurus holyokensis
Horizon: Unnamed unit (= ?Portland Formation), Massachusetts
Age: late? link to palaeos com Pliensbachian or link to palaeos com Toarcian
Place: north-west link to palaeos com Pangea
Remains: incomplete skeleton lacking skull
Length: about 1 meter

Comments: Known only from a single, poorly-preserved skeleton, the original fossil of which was destroyed in a fire some years ago, it is not known whether this small theropod belongs to either Coelophysis or a closley related genus. It differs from C. bauri in possessing less elongate neural spines and a differnt shaped ischia. Clearly, this was one of a number of small bipedal carnivores that were around at the time. In older books this dinosaur is called "Podokesaurus".

?Family Elaphrosauridae

Elaphrosaurus bambergi
Elaphrosaurus drawing by Greg Paul, Predatory Dinosaurs of the World

Elaphrosaurus bambergi Janensch, 1920

Horizon: Tendaguru beds of Tanzania
Age: link to palaeos com Kimmeridgian
Place: Central link to palaeos com Gondwana
Remains: partial skeleton lacking skull
Length: 6.2 meters
Weight: 210 kg

Comments: Elaphrosaurus may or may not be a late surviving coelophysoid. In build it is one of the slenderest and longest (in proprtion to height) of the theropods. It has been variously considered a typical "coelurosaur" (in the old non-cladistic sense), a coelophysid, and an ancestral ornithomimid. The last-named option, which for a while was the standard position, is now considered more unlikely in view of the difference in hip structure. In the absense of a skull, more cannot be determined. I am going along with Greg Paul and considering ita coelophysid.

Elaphrosaurus philtippetorum Pickering, 1995

Horizon: Morrison Formation of Colorado
Age: link to palaeos com Kimmeridgian
Place: link to palaeos com Laurasia
Remains: humerus

Comments: isolated remains indicate contemporary French and North American species which may tentatively be assigned to Elaphrosaurus

top of page

some printed references some Links and References Web links

UCMP pagephotosaudio Dilophosaurus - A Narrated Exhibition - a guided tour of Dilophosaurus - narrated by the discoverer of Dilophosaurus, Sam Welles, formerly a professor at the University of California Museum of Paleontology (UCMP). Dr Welles has also done important work in the study of the giant Cretaceous Elasmosaurid plesiosaurs

cladogram link to palaeos com Ceratosauria - Palaeos

Great Triassic Assemblages Pt 1 - The Chinle and Newark by Professor Paul Eric Olsen - part of DINOSAURS AND THE HISTORY OF LIFE - GEOLOGY V1001x

cladogram Neotheropoda - includes coverage of the Coelophysoidea - Thescelosaurus!

Dinosauricon Coelophysoidea - Dinosauricon

cladogram Ceratosauria Mikko's phylogeny site

Jurassic Gallery Jurassic Gallery - fantastic artwork by M. Shiraishi. By Japanese and English

graphicsDilophosaurus - Artistexternal linkCandi Marshall

printed reference Gregory S. Paul, Predatory Dinosaurs of the World Simon and Schuster, 1988

printed reference Rodney Steele, "Saurischia", Part 14, Encyclopedia of Paleoherpetology, Gutsav Fischer Verlag, Stuttgart and New York, 1970

printed reference Rowe, T., & Gauthier, J., "Ceratosauria", pp. 151-168, in Weishampel, D. B., Dodson, P., & Osmólska, H. (eds.), The Dinosauria University of California Press, Berkley, Los Angeles, Oxford, 1990

printed reference Lucas, S.G., 1998. Global Triassic tetrapod biostratigraphy and biochronology: Palaeogeogr. Palaeoclimat. Palaeoecol., 143: 345-382

internal link cautionary note (please read before using this page as reference material!)

up one node (internal link) back to Theropoda

Dinosaur main page
main page

contact me

page history

page uploaded 21 November 1998
page renamed & modified 6 May, & modified again 24 & 30 July and 7 Oct 1999
and on 18 Feb 2000
converted to css format and a large amount of additional material added 2 Feb 2001
a few further modifications 3 Feb

content and html by M.Alan Kazlev