The main difference between Theropoda and Neotheropoda is that
neotheropods show the first big digital reduction: they have lost the fifth finger
entirely, shrunk the fourth finger and metacarpal (if present) to very little or do away
with it altogether, and the feet show only three truly functional digits, the fifth
reduced to a splint of metatarsal and the first drawn high up the metatarsals (except for
the therizinosaurians, which eventually revisited four
walking toes, relengthening the first toe [but then again, therizinosaurians always seemed
to be doing things their own way]).
Footprints of basal theropods hunkering down occasionally turn up with odd markings, like feathers or vegetation. It is not certain what the source is (and it certainly could be different in different cases), but if protofeathers were the source, it would push the beginning of feather evolution both farther back in time and further down the theropod family tree.
I'm going to retain a separate Coelophysoidea, although it may turn out to be paired with Ceratosauria. The relevant chapter of The Dinosauria II supports a combined Coelophysoidea+Ceratosauria, although it went to press before some of the more recent ceratosaurian taxa were published.
Coelophysoids and dilophosaurids are not all that different (if indeed they are distinct groups), still being fairly unspecialized. Coelophysoids are long and little, and dilophosaurids are long and big (Dilophosaurus). Both groups show a tendency for head ornamentation in the form of crests and also are often found in groups ranging from a couple to hundreds of individuals. This relative abundance of specimens makes it possible for two "morphs" to be distinguished; robust or gracile. Robust morphs have relatively longer necks and skulls, thicker limbs, and more developed muscle attachments, unlike the gracile morphs. Of course, it should be noted, using the inevitable caveat, that Dilophosauridae has not been universally supported; with the counter-caveat that analyses that have not found it were not set up to analyze that part of Theropoda. (Get your friends together! We can play Caveat/Counter-Caveat all night long!) As usual, time will tell. The obvious problem is that when you get to the basal part of a group, everything tends to run together. Basal tetanuran? Basal ceratosaurian? Coelophysoid? Spin the wheel.
The skull, particularly of Dilophosaurus, bears a kink between the premaxilla (upper jaw tip) and maxilla (main toothbearing bone of the upper jaw), resulting in an appearance somewhat like a crocodile's jaw, one of the reasons spinosaurids like Baryonyx, which also have such a kink, were for a time considered to be late coelophysoids. This kink contributed to the idea that coelophysoids could not attack live prey due to skull weakness, but this juncture appears to have been reinforced.
| `--"Syntarsus" kayentakatae
Neotheropoda i.s.: This covers those theropods that appear to be "coelophysoids" as historically understood (coelophysoids + dilophosaurids). You can substitute "Coelophysoidea i.s." for many of them if you prefer, but I tend to err on the conservative side with fragmentary material (especially when the higher-level taxonomy is unsettled, as it is here).
|Taxon or Taxa:||Time/Place:||Comments:|
|Camposaurus arizonensis (N.D.) Hunt, Lucas, Heckert, and Lockely, 1998 (?Coelophysis)||early Norian (LTr) of Arizona||Camposaurus (not to be confused with Camptosaurus, although I'd certainly understand) is based on partial hindlimb material, with some miscellaneous referred postcranial parts that may or may not belong to it. It's not all that distinctive compared to other historical coelophysoids (Coelophysis, we're looking at you, although it's a bit early to be C. bauri), but it's useful to keep in mind when discussing the spatial and temporal range of coelophysoids.|
|Dolichosuchus cristatus (N.D.) Huene, 1932||mid Norian (LTr) of Germany||This theropod is roughly contemporary with the much better known Liliensternus, and may be synonymous. It appears probably something akin to a historical coelophysoid.|
|Gojirasaurus quayi Carpenter, 1997||mid Norian (LTr) of New Mexico||Gojirasaurus, named after the Japanese name for Godzilla (Gojira), may be related to Liliensternus and Dilophosaurus. Based on a partial postcranial skeleton (although the association of material has been questioned), it may have been about 6 meters long.|
|Halticosaurus longotarsus (N.D.) Huene, 1908||mid Norian (LTr) of Germany||This has long been considered an early
theropod, an opinion based upon
remains now renamed Liliensternus. The
remains the type was based on are scrappy and some have suggested that some
of the remains could belong to a prosauropod.
The femur appears to belong to some sort of
Two other species were once assigned to it, but have been removed: H. liliensterni, based on two partial skeletons, now is the type of Liliensternus; and H. orbitoangulatus, based on a partial skull, is now considered to be a sphenosuchian crocodylomorph.
|Longosaurus longicollis (N.D.) Welles, 1984||early Norian (LTr) of New Mexico||This is not based on the same type as Coelurus longicollis, although it was intended to be a new genus name for that species. However, part of that type is assigned to Eucoelophysis baldwini. Why that one is not called Eucoelophysis longicollis is not clear, then, and frankly I don't want to think about it too much. Longosaurus is based on an ilium, C. longicollis was based on several bones (now whittled down to a cervical, which appears to pertain to an individual of Coelophysis bauri), and Eucoelophysis is based on some other of the original grab-bag "Coelophysis" type material.|
|Panguraptor lufengensis You H.-L., Azuma, Wang T., Wang Y.-M., and Dong Z.-M., 2014||Hettangian-Sinemurian (EJ) of China||Panguraptor is the second definitive theropod known from the Lufeng Formation, after Sinosaurus (below). It is based on the partial articulated skeleton of an immature individual, including the skull, presacrals, and parts of the girdles and limbs.|
|Podokesaurus holyokensis Talbot, 1911 (?N.D.) (?Coelophysis)||Hettangian-Sinemurian (EJ) of Massachusetts||Podokesaurus may well be the same as Coelophysis, but the type was destroyed in a fire, leaving behind only casts. If it does belong here, the family name should revert to Podokesauridae, and so on.|
|Pterospondylus trielbae (N.D.) Jaekel, 1913-14||late Norian-Rhaetian (LTr) of Germany||Although sometimes considered to be a synonym of Procompsognathus, the remains are much too large for this. The remains indicate that it is some sort of coelophysoid-type animal.|
|Sarcosaurus woodi (N.D.) Andrews, 1921||latest Rhaetian-Hettangian (EJ) of England||Based upon scrappy remains, including a femur, verts, and a partial hip, this animal is like Ceratosaurus in some ways, but much more like Dilophosaurus and Liliensternus, and may have been about the size of the latter (although the growth stage is uncertain). There may recently have been a skull found that belongs to this animal, but since this rumor hasn't figured in anything but a Dinosaur Mailing List message from several years ago, I'm beginning to suspect misidentification (or maybe it was confused with the skull named as part of the type of Dubreuillosaurus valesdunensis). A second partial skeleton cannot be confidently ascribed to this species, but is also a coelophysoid/dilophosaurid. Although it is best considered a dubious name at this point, it is probably distinct based on time and place.|
|"Sarcosaurus" andrewsi (N.D.) Huene, 1932||mid-late Hettangian (EJ) of England||This animal, based on a slender tibia, is apparently a largish coelophysoid/dilophosaurid. The name Magnosaurus woodwardi has also been applied to this material (in the same paper!), but usually today Sarcosaurus is given the honor of hosting this tibia.|
|?"Tanystropheus" willistoni (N.D.) Cope, 1887||early Norian (LTr) of New Mexico||The third in the grab bag of original Coelophysis species, this one was apparently based on an ilium. It should probably be at Theropoda i.s., but since it's probably just Coelophysis or Eucoelophysis or Eoeucoelophysoides or Prorioarribasaurus or something like that (note that the latter two are not real names), it might as well stay here.|
Coelophysoidea: The general form of a coelophysoid is a long,
low body combined with a long neck and skull, with short, effectively three-fingered hands
(coelophysoids sometimes had a splint of a fourth finger with a single phalanx on it).
Undescribed coelophysoid material is quite common, especially in North America. The best-known of these, an animal known informally as the "Shake 'n' Bake theropod" from the EJ, often pops up in studies. EJ material from the Isle of Skye, Scotland, is also known. An EJ form from Mexico known from a partial pelvis and sacrum appears to be closely related to "Syntarsus"; also found in the area was the partial skull of a more derived theropod. In addition, there may be more than one theropod taxa at Ghost Ranch; there are new remains like a proto-"Syntarsus" kayentakatae.
|Taxon or Taxa:||Time/Place:||Comments:|
|Liliensternus liliensterni Welles, 1984 (Huene, 1934 [originally Halticosaurus])||late Norian (LTr) of Germany||Based on two partial skeletons referred to Halticosaurus as a third species, this theropod is between Coelophysis and Dilophosaurus in both size and anatomy. Unfortunately, skull material is fragmentary, and the oft-illustrated dilophosaur-style crests speculative.|
|Lophostropheus airelensis Ezcurra and Cuny, 2007 (originally Lilensternus airelensis Cuny and Galton, 1993)||late Rhaetian-early Hettangian (LTr-EJ) of France||This genus is based on postcranial remains (five cervicals, two dorsals, four sacrals, much of the hip, and a tooth) originally referred to Halticosaurus sp., then to Liliensternus. The name refers to ridges on the cervicals.|
|Procompsognathus triassicus Fraas, 1913||mid Norian (LTr) of Germany||Along with Troodon,
this little theropod is one of dinosaur paleontology's ultimate survivors, weathering poor
preservation, challenges upon its validity and dinosaurhood, and losing a head, which
turned out not to belong with the partial skeleton that is its type, although it is
possible it belongs to this species. However, some cheeky person is
again trying to de-dinosaurify it; we'll see how this latest attempt turns
A pelvis from the Norian (LTr) of Wales may be referable.
Procompsognathus and the very distantly related Compsognathus found themselves mixed together for the various Jurassic Park media. In the first book, the "compies" are identified as Procompsognathus, either P. triassicus or the nonexistent P. amassicus, and included among their behavioral traits eating sauropod dung. However, Procompsognathus was not from the Jurassic, and anatomically is not particularly like Compsognathus, the name notwithstanding.
|Segisaurus halli Camp, 1936||?Pliensbachian (EJ) of Arizona||Segisaurus is based on a postcranial
skeleton that has good representation of most of the body parts. The
individual was not fully grown, possibly around five years old, and was
found in a position suggestive of crouching or even guarding a nest. For a long while, this animal was placed in its own family,
Segisauridae, based upon the belief that its long bones were solid, not
hollow (this even led to its dinosaurian nature being challenged). Also, for a long time it was unusual for being one of the few theropods for which a
good clavicle was known (now it turns out that instead of separated
clavicles, it had a full-blown furcula, with the two clavicles fused
together). Its hip bones have unusual holes and flattening. It is sometimes
considered to be close to Procompsognathus.
When first described, Charles Camp interpreted its "splint-like" cervical ribs as supporting flaps of skin like those of gliding lizards (except along the sides of the neck instead the torso), to assist the segisaurs when they were running around. This is no longer a popular interpretation. In Camp's defense, this was was one of the first early theropod skeletons, predated by Podokesaurus, Procompsognathus, and Liliensternus as Halticosaurus (Ghost Ranch was over a decade in the future), and the state of early dinosaur research was such that various prosauropods and "thecodonts" were still thought to be theropods; the year before Camp's publication, Major L. Brady described a partial prosauropod skeleton (one of the ambiguous Arizona Ammosaurus/Massospondylus skeletons) from the same formation as a theropod. Camp's idea becomes more interesting in hindsight because of the Jurassic Park Dilophosaurus embellishments.
|Coelophysis: Cope, 1889||C. bauri (type) (Cope, 1887 [originally Coelurus])||late Rhaetian (LTr) of New Mexico||This animal recently went through a severe taxonomic war
(see Eucoelophysis for details). It is best known from the hundreds of
specimens from the Ghost Ranch quarry in New Mexico.
It was long thought that Coelophysis was a cannibal, because some Ghost Ranch skeletons have juvenile-sized skeletons in the area of their stomachs. However, two such cases have turned out to be a juvenile's body trapped under an adult's, and a third case, while apparently truly gut contents, was of a nondinosaurian (possibly Hesperosuchus, a contemporary crocodylomorph, and remember that basal crocodylomorphs and basal dinosaurs look a lot alike).
|C. rhodesiensis (Raath, 1969 [originally Syntarsus])||Hettangian-Sinemurian (EJ) of Zimbabwe and South Africa||A contemporary of the common prosauropod Massospondylus, C. rhodesiensis is one of the first dinosaurs suggested to have borne a feathery coating. Its original name Syntarsus is preoccupied by a beetle (Fairmaire, 1869), and so the substitute name Megapnosaurus Ivie, Slipinski, and Wegrzynowicz, 2001 was proposed. However, almost no vertebrate paleontologists accepted this, and now it seems as if it doesn't matter because the evidence to synonymize Syntarsus with Coelophysis, an idea considered since the 1980s, appears to have won out.|
|"Syntarsus" kayentakatae Rowe, 1989 (?Coelophysis)||Sinemurian-mid Pliensbachian (EJ) of Arizona||"Syntarsus" kayentakatae, a contemporary of its larger cousin Dilophosaurus, is best known for the similar dual head-crests it possesses. In its case, though, they are smaller, and formed only by the nasals: if you imagine the animal as a toy, the crests are like a bit of plastic "flash" along the top of the skull in front of the eyes. It may represent its own genus.|
|?Kayentavenator elysiae Gay, 2010||Sinemurian-mid Pliensbachian (EJ) of Arizona||Kayentavenator is known from the partial skeleton of a juvenile theropod (hips, legs, and some verts). It was described as a basal tetanuran, but seems to be a coelophysoid instead.|
Dilophosauridae and friends:
|Taxon or Taxa:||Time/Place:||Comments:|
|Zupaysaurus rougieri Arcucci and Coria, 2003||Norian (LTr) of Argentina||This animal is based upon a superb skull, among other material. It apparently had a long neck. It was also thought to have twin thin nasal ridges, but this was an illusion based on crushing and displacement of the bones involved (the same thing may hold true for the crests of "Syntarsus" kayentakatae). It was first considered to be the most basal known tetanuran, but more recent study indicates a coelophysoid affinity.|
|Dracovenator regenti Yates, 2006||Hettangian-Sinemurian (EJ) of South Africa||Based on a partial skull, Dracovenator seems to be closest to Dilophosaurus and Zupaysaurus, although it has an odd mix of derived and basal features.|
|?Cryolophosaurus ellioti Hammer and Hickerson, 1994||Pliensbachian (EJ) of Antarctica||As its name implies, Cryolophosaurus is a crested
theropod from a cold land. Its discovery and publication early in the 1990s brought
additional proof that dinosaurs inhabited the now frozen continent of Antarctica.
It was originally described as a carnosaurian, which
was very interesting because it would have been the first known carnosaur.
More recent research puts it as a dilophosaurid, or as a basal tetanuran. Some researchers have suggested that the type material is
this is no longer thought possible.
Some have suggested that the type individual may have died after choking on the rib of a prosauropod, but this could just be a paleontological urban legend.
Its vertical crest shows convergence with the hairstyle of Elvis.
|Dilophosaurus wetherilli (Welles, 1954 [originally Megalosaurus]) (including D. breedorum Welles vide Welles and Pickering, 1999)||Sinemurian-mid Pliensbachian (EJ) of Arizona||Dilophosaurus has been the victim of
three important misunderstandings: that it is a species of Megalosaurus; that its
head crests and "weak" snout made it unable to hunt live prey; and the
monstrosity exhibited in Jurassic Park looked like the real thing. Adding
frog DNA changed a lot in this dinosaur's case, it would appear. This theropod is
best known for its double head crests and its undersized, oversimplified celluloid
representation shaking lizard frills and spitting poison in an otherwise decent
movie. The real thing may have been feathered, as suggested by an imprint of a
similarly-sized theropod sitting on the ends of its hips. This
imprint showed traces, possibly from feathers (although more recent
research indicates not), in the belly area.
Besides the head crests, the skull has another oddity: the teeth of Dilophosaurus are unusually long.
In the event that it appears that there are two species of Dilophosaurus, one crested and one not, the name D. breedorum could be used for the crested form, as the type species' crests are not apparent, either through breakage or the excellent reason of not having had them in the first place, but it is possible that perceived differences in skeletons are examples of the sexual dimorphism so strongly hinted at in other basal neotheropods. Juvenile individuals are now known.
|Tachiraptor admirabilis Langer, Rincón, Ramezani, Solórzano, and Rauhut, 2014||Hettangian (EJ) of Venezuela||Tachiraptor is a basal neotheropod based on a tibia, with a partial ischium from the same site thought to belong as well. It appears to been comparable in size to coelophysoids (1.5 m scale).|
triassicus Yang, 1948 (including Dilophosaurus sinensis
|Hettangian-Sinemurian (EJ) of China||Sinosaurus is based on a chunk of
theropod jaw with teeth. It had postcranial remains assigned to it which
may be a mix of elements from not one, but two different kinds of prosauropod,
namely plateosaurids and "melanorosaurids" (one specific
candidate is Jingshanosaurus). It
looked like a good bet to spend the rest of eternity in well-deserved
taxonomic obscurity, until Dilophosaurus sinensis came along.
Originally assigned as a second species of Dilophosaurus, the placement of the virtually complete skeleton and skull named D. sinensis was long questioned. Some workers suggested that it was actually a basal relative to the carnosaurians. To complicate things further, two distinct types of crested Dilophosaurus-like theropods are known from EJ China, going under the D. sinensis banner; they are apparently victims of at least one of the usual suspects (preservation, crushing, individual variation, age of the animals involved, etc.). Its crests are farther forward on the skull compared to Dilophosaurus wetherilli, and its scapula is expanded.
Adding the Dilophosaurus sinensis material once again makes Sinosaurus a going concern. This leads to another concern (see, I worked in two meanings of the same word): namely, where does it fit among theropods. "D." sinensis was interpreted as the most basal dilophosaurid when that family was quasi-formally introduced, but the animal may actually be more derived, a basal tetanuran.
|Taxon or Taxa:||Time/Place:||Comments:|
|Kakuru kujani Molnar and Pledge, 1980||Aptian (EK) of Australia||Based on a partial opalizied tibia (sold at auction, but now rediscovered), this species resembles Avimimus in its tibia's slenderness, but also bears some resemblance to abelisauroids.|
|"Thecospondylus" daviesi (N.D.) Seeley, 1888||Barremian (EK) of England||Also known as Thecocoelurus, this taxon, based on a cervical, does not pertain to the much larger Thecospondylus. Instead, it is probably either an abelisauroid or a caenagnathid/elmisaurid oviraptorosaurian; the former is more likely based on considerations of time and place. The problem is that the vertebra is interesting enough that people keep gnawing on it, but incomplete enough that you can interpret it several different ways.|
|Valdoraptor oweni Olshevsky, 1991 (originally Megalosaurus oweni Lydekker, 1889)||Berriasian-Valanginian (EK) of England||Valdoraptor is based on some odd metatarsi from a moderately-sized theropod. It's historically been dumped off as a megalosaur- or allosaur-type theropod, but is more likely to be either a noasaurid or ornithomimosaurian.|
Ceratosauria and Tetanurae: Until about the middle of the first decade
of the 2000s, the coelophysoids, dilophosaurids, and "neoceratosaurians" were
typically included in one group (Ceratosauria), but more recent analyses finds
the "neoceratosaurians" to be distinct (making up a more limited Ceratosauria)
and usually break up the coelophysoids by removing a distinct Dilophosauridae. True ceratosaurians are best known
from the Late Cretaceous of South America, India, and Madagascar.
The theropods belonging to Tetanurae include both birds and the most famous of classic theropods. Tetanurans lack the fourth digit of the hand, have all their upper jaw's teeth in front of their eyes, have a strap-like scapula (shoulder blade), an obturator process on the ischium (in most theropods, the only rearward pointing lower hip bone), and an amplified ascending astragalus process, among other characteristics. Tetanurans come in one of three major flavors: Megalosauroidea, Carnosauria, and Coelurosauria.
|Home Page||Alphabetical Dinosaur Index||Clado-Index|