Coelurus: a lightly built theropod of the Late Jurassic

This small Morrison Formation dinosaur gave Coelurosauria its name, but an incomplete skeleton leaves its skull and precise place in the group uncertain.

Coelurus reconstructed on a Morrison Formation floodplain
The elongated vertebrae and limbs are based on fossils; the head and any feather covering remain uncertain.

Coelurus fragilis was a small, lightly built theropod from the Late Jurassic of North America. Its reliable fossils come from Quarry 13 at Como Bluff in Wyoming, within the Morrison Formation. The deposits are generally assigned to the Kimmeridgian and are often placed in a broad interval around 155–150 million years ago, but the precise age of the bone-bearing horizon is not a single uncontested date.

The genus gave its name to Coelurosauria, a major theropod branch that includes tyrannosaurs, ornithomimosaurs, many feathered dinosaurs and birds. Yet Coelurus itself is represented mainly by an incomplete skeleton without a secure skull. The profile is part of our dinosaur catalogue, where it can be compared with better known Jurassic coelurosaurs without filling its missing anatomy by assumption.

Quick facts

Scientific nameCoelurus fragilis Marsh, 1879
GroupTheropoda, Tetanurae, Coelurosauria; finer placement remains debated
AgeLate Jurassic, probably Kimmeridgian, broadly about 155–150 million years ago
RegionQuarry 13, Como Bluff, Wyoming, USA
LengthAbout 2–2.5 m, including reconstructed missing parts
MassRoughly 10–20 kg; model-based estimate
DietProbably animal prey; no direct food evidence is known
FossilsIncomplete postcranial remains; skull assignment is uncertain
Evidence guide

What the fossils can tell us

A scattered type series

The name is based on vertebrae collected in separate lots from Quarry 13. Later study associated additional pelvic and limb bones with C. fragilis, but the material does not form one complete articulated skeleton.

Name and the history of the fossils

Coelurus combines Greek roots for “hollow” and “tail”, referring to the thin-walled, internally hollow vertebrae that impressed Othniel Charles Marsh. The species name fragilis means “fragile” and likewise reflects the light construction of the bones. It does not mean that the living animal was physically weak.

Marsh named Coelurus fragilis in 1879 from vertebrae collected at Quarry 13 near Como Bluff, Wyoming. The type material is divided among YPM 1991, YPM 1992 and YPM 1993. Fossils were recovered in separate collecting lots as quarry work expanded, making it difficult to determine which bones belonged to one individual and which belonged to different animals.

In 1881, Marsh described additional vertebrae and used the names Coeluria and Coeluridae for small theropods of this general kind. In 1884 he named Coelurus agilis for a pair of fused pubic bones from the same quarry, thought at the time to represent another, larger animal. John Ostrom re-examined the material in 1980 and concluded that the bones did not duplicate elements of C. fragilis, matched it in size and came from the same quarry level. Today C. agilis is usually treated as a junior synonym of C. fragilis, and the pubes may belong to the same skeleton.

Charles Gilmore combined Coelurus with Ornitholestes in 1920. Ostrom later demonstrated differences in vertebral, pelvic and limb proportions, and a 2005 redescription reinforced the genera’s separation. Coelurus has especially elongated vertebrae and a long, narrow foot. The case illustrates why small Jurassic theropods cannot be merged solely because their fossils are incomplete.

Classification and uncertain relatives

The secure broad placement is Theropoda, Tetanurae and Coelurosauria. Coelurosauria is a large branch that includes tyrannosauroids, ornithomimosaurs, maniraptorans and birds. The small Late Jurassic Ornitholestes is also a coelurosaur, but its skull and proportions differ from those of Coelurus. Similar body size does not establish close genus-level relationships.

The exact position of Coelurus within early coelurosaur evolution is unstable. Some analyses place it near the base of Coelurosauria, others associate it with Tanycolagreus, early tyrannosauroids or maniraptoromorphs. A 2025 analysis recovered it near the base of Ornithomimosauria together with the African genus Nqwebasaurus. That result does not make Coelurus a miniature Ornithomimus; its skull is unknown, and other character matrices produce different trees.

The historical family Coeluridae was a convenient grouping for small theropods before their relationships were well understood. It is not now a stable natural family containing every slender Jurassic predator. Because Coelurus lacks a known skull and several bones are incomplete, “early coelurosaur” is more defensible than a confident assignment to one later family.

Fossil record and limits

The best-supported material comes from Quarry 13 at Como Bluff. The locality yielded vertebrae and bones later associated with the pelvis and limbs. An isolated right lower-jaw fragment found in the same bone bed has fourteen tooth sockets but no teeth. Its thin construction raised the possibility that it belonged to another animal, so it cannot securely restore the head of Coelurus.

Two humeri from the Cleveland-Lloyd quarry in Utah have sometimes been referred to the genus, but that assignment is tentative. Reliable geographic claims should therefore centre on Como Bluff rather than presenting scattered, uncertain bones as a broad range. The known sample is not adequate to estimate how much the species varied among individuals or through growth.

No skin or feather impression, eggs, nest, trackway or associated prey is securely known. The fossil record preserves enough to reconstruct a lightly built biped, but not the colour, exact head shape, social behaviour or a detailed life history. The absence of evidence is particularly important for Coelurus because attractive restorations often make its face look more certain than it is.

Size, anatomy and movement

The full body length is usually estimated at about 2–2.5 m, while mass estimates commonly fall around 10–20 kg. These are reconstructions rather than measurements of a complete skeleton. A frequently repeated length near 2.4 m includes an extrapolated tail and missing body regions. Mass depends on trunk volume and soft tissue that are not preserved.

The neck and trunk were comparatively elongated. Some cervical vertebrae were more than four times as long as wide, with thin walls and internal cavities. The dorsal and tail vertebrae also had slender proportions. Hollow bone can reduce mass while retaining mechanical strength; it is not evidence that the skeleton was abnormally brittle in life.

The forelimbs were slender. The humerus was about 12 cm long and strongly curved, the ulna slightly bowed, and the fingers long and thin. A crescent-shaped wrist bone had clear joint surfaces, consistent with a mobile hand. That anatomy does not reveal the exact range of wrist movement or prove that it folded like a modern bird’s wing.

The femur was about 21 cm long and curved in two planes. The most striking element was the very long, narrow metatarsus. These proportions fit an active, fast-moving biped, but fossil bones alone do not yield a precise maximum speed. One femur preserves an abnormal area near its lower end; its cause is unknown. It could reflect injury or disease, but the specimen does not support a confident diagnosis.

Diet and Jurassic habitat

Coelurus is conventionally described as a carnivore because it was a theropod. No stomach contents, coprolites, characteristic tooth wear or prey bones with diagnostic bite marks are known. The possible jaw fragment has tooth sockets but cannot be assigned securely. Small vertebrates, larger invertebrates and scavenged remains are reasonable ecological possibilities, not a recovered menu. A specialised diet has not been demonstrated.

In the Late Jurassic, the Morrison Formation accumulated across river systems, floodplains, temporary ponds and drier interfluves under seasonal conditions. Quarry 13 is famous for bones of stegosaurs and Camptosaurus. The wider formation also preserves large theropods and sauropods, but its fossils span time and space; every named animal need not have occupied the same place at the same moment.

The long legs and elongated foot suggest active movement on land. No trackway can be securely linked to Coelurus, and there is no nest or group burial to show whether it was solitary or social. Its territory, courtship and parental care are unknown. The broader setting is described in our Jurassic Period overview.

Feathers, speed and common mistakes

One persistent mistake is treating Coelurus and Ornitholestes as the same animal. They were once combined, but their vertebral, pelvic and foot proportions differ, and the genera are now kept separate. Another is to draw Coelurus as a miniature tyrannosaur. A position near the base of Tyrannosauroidea is only one hypothesis; it does not imply the heavy skull, powerful jaws or shortened arms of later tyrannosaurids.

Long legs do not provide a stopwatch. Locomotor speed also depends on muscle, tendon, posture and gait, none of which is completely preserved. Estimates of extreme sprint speed are not measurements. Likewise, feathers are plausible by comparison with coelurosaur relatives, but no feathers or skin are known for this genus. Their density, distribution, shape and colour remain unknown.

Summary

Coelurus fragilis was a small Late Jurassic coelurosaur with a light skeleton, elongated body, mobile-looking hand and long foot. Its fossils helped give Coelurosauria its name, but the genus itself remains poorly known: the type is scattered among collecting lots, the skull is absent and its precise evolutionary position changes among analyses. The safest reconstruction preserves its distinctive limb and vertebral proportions while leaving diet, head, covering and behaviour open.

Frequently asked questions

Where and when did Coelurus live?

Its reliable fossils come from Quarry 13 at Como Bluff in Wyoming, in Late Jurassic Morrison Formation deposits, broadly around 155–150 million years old.

How many species of Coelurus are recognised?

One species, Coelurus fragilis, is generally recognised. C. agilis is usually treated as a synonym; C. gracilis is not securely referable to the genus.

Is the skull of Coelurus known?

No secure skull is known. A lower-jaw fragment from the same quarry may belong to another animal.

Was Coelurus feathered?

Feathers are plausible by comparison with coelurosaur relatives, but no skin or feather impression is known for Coelurus itself.