Ten unusual dinosaurs

Some dinosaurs seem strange because their fossils preserve structures unlike those familiar from living animals. Therizinosaurus carried enormous hand claws, Deinocheirus combined a broad bill with a high-backed body, and Yi qi had a membrane supported by an elongated wrist bone. These traits are directly observable; what they did is a separate question.

This selection focuses on ten named dinosaurs with distinctive anatomy rather than on invented monsters. “Unusual” is a human judgement, not a scientific category. Each example is included because a fossil feature is notable, and each interpretation is limited to what the evidence can support.

1. Therizinosaurus: giant hand claws

Therizinosaurus cheloniformis is best known for enormous, curved claw bones from the forelimbs. The claws were among the longest recorded in any land animal, although the keratin sheath that covered each bone is not preserved. The animal was a large therizinosaur, a branch of theropods whose members shifted toward plant-heavy diets.

The claws may have helped with display, defence or pulling vegetation within reach. Mechanical studies make forceful digging or stabbing less convincing, but no fossil preserves the animal using them. A spectacular tool does not reveal one certain behaviour.

2. Deinocheirus: arms that preceded the rest

For nearly half a century, Deinocheirus mirificus was represented chiefly by two forelimbs about 2.4 metres long. Later discoveries from Mongolia revealed a large ornithomimosaur with a broad duck-like bill, tall spines along the back, a deep torso and relatively short legs. Its body was far heavier than the sleek ostrich-like outline once imagined from the arms alone.

One specimen preserves fish remains and more than a thousand gastroliths in the body region. These finds support an omnivorous diet that included aquatic food, but they do not prove a single feeding routine. The name means “terrible hand”, an apt reflection of the first fossils rather than a description of the complete animal.

3. Yi qi: a membrane-bearing theropod

Yi qi was a small Jurassic scansoriopterygid from China. Its fossils preserve a long, rod-like wrist bone, the styliform element, alongside impressions interpreted as a skin membrane. Together with feathers, these structures indicate an unusual flight surface unlike the feathered wing of a modern bird.

Whether Yi qi could make powered flights, controlled glides or short parachuting descents remains debated. The membrane itself is evidence; its exact outline and flight performance are reconstructed from a compressed and incomplete fossil.

4. Stegouros: a flat tail weapon

Stegouros elengassen comes from Late Cretaceous rocks in southern Chile. This small armoured dinosaur had seven pairs of broad bony plates at the end of its tail, forming a flattened structure unlike the club of an ankylosaur or the spikes of a stegosaur. The original authors interpreted it as a distinctive tail weapon. Its shape is preserved; the way it was used cannot be observed directly.

5. Psittacosaurus: bristles and a reconstructed colour pattern

Psittacosaurus was an early relative of horned dinosaurs. One exceptionally preserved specimen retains long bristle-like structures above the tail and a detailed outline of the skin. Pigment-based analysis of that individual has been used to reconstruct a darker back and lighter underside, a countershading pattern that may have reduced the appearance of body depth. The fossil evidence constrains the pattern, but exact hues remain a model, not a colour photograph.

6. Microraptor: feathers on four limbs

Microraptor carried long flight feathers on both arms and legs, creating several aerodynamic surfaces together with the tail fan. Exceptional fossils also preserve dark, iridescent plumage in one specimen, while different individuals contain remains of a bird, fish or lizard. These finds document a varied diet and a feather arrangement; how the animal launched and controlled its glide is still tested with models.

7. Carnotaurus: horns and unusually short arms

Carnotaurus sastrei is recognisable by a pair of horns above the eyes and extremely short forelimbs. The arms had four fingers, but their mobility and practical function were limited. Skin impressions beside the skeleton preserve small scales and larger conical bumps. This specimen therefore records the outer skin better than many large theropod fossils; it provides no evidence of feathers on this individual.

8. Amargasaurus: a double row of long neck spines

Amargasaurus from Early Cretaceous Argentina had unusually tall paired neural spines along the neck. The bone bases are preserved, while any keratin covering, skin or sail is unknown. Different reconstructions show the spines as exposed structures or support for soft tissue. The fossils do not decide between every proposed silhouette.

9. Oviraptor: a “thief” that may have guarded its own nest

Oviraptor was named from a skeleton found over a nest of eggs and was initially accused of raiding them. Later embryos within similar eggs were identified as oviraptorosaurs, and adults were found brooding their own clutches. The original interpretation was misleading: association with eggs was not proof of theft. This is a reminder that fossils can overturn a memorable story.

10. Pachycephalosaurus: a dome with injury evidence

Pachycephalosaurus had a massive bony dome on the roof of its skull. Injuries in adults of related pachycephalosaurids cluster near the upper part of the dome, a pattern consistent with head contact. It does not preserve a complete fight or settle whether animals butted heads directly, pushed, or struck from the side. The dome may also have signalled identity, so one function need not explain every use.

When unusual does not mean understood

New skeletons can transform an animal's reconstruction, as happened when more of Deinocheirus became known. Names can also fail when a fossil cannot be diagnosed or its supposed dinosaur identity is not supported. The Russian overview excludes “Behtialosaurus” and “Shunisaurus” because no reliably described dinosaur genera are established under those names. A scientific list should rely on identifiable fossils rather than memorable but unsupported labels.

These dinosaurs show how much anatomy can be learned from bones, skin impressions, feathers, eggs and gut contents, and how much behaviour remains inferential. A claw can be measured without knowing every task it performed. An egg clutch can reveal care without recording a daily routine. A membrane or tail weapon invites functional tests but does not answer every question by itself.

For additional examples and fuller evidence-led profiles, browse the dinosaur catalogue. The most reliable reconstructions keep observed anatomy separate from proposed function, colour and behaviour.