Amurosaurus: the hadrosaur from the Amur bonebed

Amurosaurus riabinini was a plant-eating lambeosaurine hadrosaurid from the Maastrichtian of eastern Asia. Its best-known fossils come from a bonebed at Blagoveshchensk, in Russia’s Amur Region. Hundreds of bones document many individuals, but no complete articulated skeleton has been found. The genus is therefore notable for a rich mixed assemblage rather than one intact animal.

Quick facts

Name and discovery

Amurosaurus means “Amur lizard”, combining the river’s name with the Greek-derived word for lizard. The species name riabinini honours Anatoly Riabinin, who studied dinosaurs from the Amur region in the first half of the twentieth century.

Russian geologists had worked near Jiayin, on the Chinese side of the Amur, in 1916–1917. Riabinin later described hadrosaur remains from that area. Bones on the Russian bank were reported in 1957 but were mistakenly thought to have been redeposited in younger sediments. A major change came in 1984, when Yuri Bolotsky and colleagues from the Amur Complex Integrated Research Institute found a bone-bearing layer in western Blagoveshchensk. By 1991, excavations covered about 200 square metres and yielded hundreds of bones.

Bolotsky and Sergei Kurzanov named Amurosaurus riabinini from part of this collection. The type, AEHM 1/12, consists of associated left maxilla and dentary bones. The first description was brief. In 2004, Pascal Godefroit, Yuri Bolotsky and Jimmy Van Itterbeeck redescribed a larger sample, identifying diagnostic skull and limb features that supported the genus as a distinct taxon.

The association of upper and lower jaw bones helps connect diagnostic features across the mouth. It does not preserve the missing front of the skull or the crest outline.

A 2022 revision compared material from the Ulaga locality in China. Most bones formerly assigned to Sahaliyania elunchunorum were judged indistinguishable from A. riabinini, and Sahaliyania was treated as a junior synonym. This would extend the genus into north-eastern China, but the conclusion depends on comparisons among mostly disarticulated bones and can be tested by future finds.

Classification and fossil assemblage

Amurosaurus belongs to Ornithopoda, Hadrosauridae and Lambeosaurinae, the hadrosaurid branch associated with hollow cranial crests. Its precise position within Lambeosaurinae has shifted among analyses. A 2004 study placed it among earlier branches; later datasets moved it closer to more derived forms, including Lambeosaurus. The family assignment is firmer than any single claim about its closest relative.

The Blagoveshchensk bonebed yielded hundreds of bones. More than 90 per cent of identifiable hadrosaur remains in the studied collection were lambeosaurines. Material includes braincases, jaws and teeth, skull-roof bones, vertebrae, ribs, shoulder blades, pelvic elements and limb bones. They represent many animals, not one composite skeleton. Another hadrosaurid, Kerberosaurus, is also present, so bones without diagnostic features cannot automatically be assigned to Amurosaurus.

Most bones are disarticulated and mixed. Apart from the type jaws, only a few vertebral sequences can confidently be linked to a single individual. A large femur, humerus and braincase may come from animals of different sizes, limiting estimates of body proportions. Juveniles and small subadults dominate, although large adult bones also occur.

Light bones are underrepresented, dental batteries are sometimes separated, and long bones are often broken. These patterns suggest that carcasses decomposed before flowing sediment moved and reburied their bones. Endocasts made from braincases show enlarged, rounded cerebral hemispheres and a relatively straight cavity, but they do not measure intelligence or reveal behaviour directly.

One right ulna, AEHM 1/1037, is about 32 centimetres long and preserves a healing fracture near its lower end. A callus and imperfect union show that the animal survived for a time after injury and may have had difficulty bearing weight. The cause is unknown, and the bone cannot establish that other animals cared for it.

Size and anatomy

Without a complete skeleton, body length must be estimated by comparing separate large bones with more complete lambeosaurines. A range of about 7–9 metres is reasonable for large individuals. Mass estimates illustrate how strongly the method matters: one study calculated about 2.03 tonnes with a bipedal formula and 4.79 tonnes using both humerus and femur in a quadrupedal model. These are alternative model outputs, not two directly measured masses; a broad range of roughly 2–5 tonnes communicates the uncertainty better.

The front of the mouth had a toothless keratinous beak, followed by dense dental batteries. The type dentary has about 37 vertical tooth columns, each with four or five replacement teeth; usually only two in a column contributed to the active surface. Constant replacement maintained a grinding edge for fibrous vegetation.

The frontal and prefrontal bones formed the base of a cranial crest. The premaxillae and nasals that would have shaped its outer profile are missing. A rounded helmet, tall dome or tube-like crest in artwork is consequently a reconstruction rather than preserved anatomy. This is an important limit: lambeosaurine identity supports a hollow crest, but does not supply a precise Amurosaurus silhouette.

The ulna and radius had a characteristic S-shaped profile, and the forelimbs were strong enough for regular support. Like other large hadrosaurids, Amurosaurus probably used all four limbs for much of its walking and feeding, with bipedal movement also possible. A long muscular tail balanced the trunk; ossified tendons reinforced parts of the spine.

Environment and feeding

The Russian fossils come from the Udurchukan Formation, deposited during the Maastrichtian. Correlating eastern Asian pollen zones with the global timescale is not exact. The Blagoveshchensk bonebed was once placed in the late Maastrichtian, so “the last several million years of the Cretaceous” is safer than assigning every bone one exact date.

The sediments formed in a continental river or alluvial system. Carcasses decomposed, and flows carrying sediment shifted bones before rapidly burying them when the current slowed. Fragile elements were probably not transported far, but the place where each animal died is unknown. Pollen and plant remains indicate vegetated plains in the Amur basin, yet no particular food plant is identified.

The beak could crop leaves and young shoots, and the dental batteries processed plant matter through complex jaw movements. No securely associated stomach contents or coprolites define an exact diet. Theropod teeth occur in the bonebed and possible tooth marks appear on fewer than two per cent of examined bones. This is consistent with occasional scavenging or feeding on carcasses, not proof that a named predator killed the hadrosaurs.

Bonebed, age profile and behaviour

The high proportion of late juveniles and small subadults prompted a hypothesis that the youngest animals lived apart and joined larger groups later. Mortality, transport, preservation and the excavated area could also shape the age profile. The bones had become disarticulated after decomposition and may have accumulated through multiple episodes, so the site is not direct evidence of a herd dying together.

Hadrosaurids may have been social, but the Blagoveshchensk assemblage does not reveal a fixed herd structure or prove that the animals were gregarious at the moment of death. Nor does the healed ulna demonstrate assistance from companions. These are biologically possible interpretations, not observations recorded by the bones.

What remains uncertain

The missing upper crest is the most visible gap in the reconstruction. The skull base and family placement support a hollow crest, but its height, outline and internal nasal passages are not preserved. A resemblance to Corythosaurus is a convenient artistic choice, not a fossil discovery.

One exact body length or mass also overstates the evidence because bones from different individuals are mixed and the sample is biased toward young animals. The changing placement within Lambeosaurinae and the proposed synonymy of Sahaliyania remain phylogenetic and taxonomic interpretations. Finally, an endocast records the shape of a brain cavity, not memory or intelligence; a healed fracture establishes survival, not group care.

Amurosaurus is a well-supported lambeosaurine known from an unusually rich bonebed. Its jaws, limbs, braincases and age range offer substantial evidence, while its adult proportions, crest outline, closest relatives and social life remain qualified. Compare its crest with the better-preserved Lambeosaurus, while keeping the two species’ anatomy separate. Explore it alongside other hadrosaurids in the dinosaur catalogue.