Erketu: the long-necked sauropod of Mongolia

The bones confirm exceptional elongation within the neck, but the rest of the skeleton is largely absent. This profile separates measurable vertebrae from the much less certain whole-animal reconstruction.

Illustrative reconstruction of Erketu in a Cretaceous Mongolian floodplain
Illustrative reconstruction based on an incomplete cervical series and hind limb; head, trunk, tail and soft tissues are inferred.

Erketu ellisoni is a sauropod from the eastern Gobi of Mongolia, best known for a series of very long neck vertebrae. The bones make cervical elongation an observed feature, not merely an artistic impression. They do not preserve most of the animal, however, so images of a complete giant sauropod include substantial comparative reconstruction.

Quick facts

SpeciesErketu ellisoni Ksepka & Norell, 2006
AgeEarly Late Cretaceous; exact age of the type bed uncertain
LocalityBor Guve, Dornogovi, Mongolia
Type specimenIGM 100/1803, partial neck and right hind limb
PlacementTitanosauriformes, Somphospondyli; finer placement debated
DietPlant eating inferred from sauropod relationships

Evidence guide

The preserved vertebrae document an unusually elongated neck

The type series includes cervical vertebrae through the ninth position. Their elongation can be measured directly, but the total vertebral count and complete neck length remain unknown.

Discovery at Bor Guve

An American Museum of Natural History and Mongolian Academy of Sciences expedition found the type locality in 2002 in Dornogovi Province. Daniel Ksepka and Mark Norell named Erketu in 2006. The genus name refers to Erketu Tengri, a creator deity in Mongolian tradition; the species honours museum paleoartist Mick Ellison. The holotype is IGM 100/1803, held in the Mongolian Institute of Paleontology collection.

The original specimen includes an articulated anterior cervical series, a right sternal plate and an articulated right lower hind limb. A return expedition recovered three more middle-to-posterior neck vertebrae in 2003. These extend the same individual’s series through cervical 9, although cervical 6 is incomplete. The skull, teeth, dorsal column, sacrum, tail, pelvis, forelimbs and femur are not known for the holotype.

A neck made of unusually long bones

Several preserved cervical centra are much longer than tall. Their proportions can be calculated from the bones and are among the most distinctive features of Erketu. The vertebrae also have deeply divided neural spines, long cervical ribs and extensive internal chambers. CT imaging showed camellate pneumatic spaces, a pattern that lightens the skeleton while retaining internal struts.

These features help explain how a long neck could be supported without making every bone solid and excessively heavy. They do not reveal a habitual pose. A raised neck, a more horizontal posture and a flexible range of positions are biomechanical possibilities; the fossils do not preserve muscles, ligaments or a complete neck series that would settle the question. The total cervical count is also unknown, so the absolute neck length cannot be read directly from the surviving vertebrae.

What the rest of the skeleton cannot tell us

Reconstructions often show Erketu as a 12–15 metre animal weighing several tonnes. Those values are extrapolations from relatives because the trunk, tail and major limb bones needed for robust whole-body estimates are missing. The preserved tibia and fibula show a substantial sauropod hind limb, but not the volume of the torso or the length of the tail. Exact mass claims would therefore imply a precision the specimen cannot supply.

The animal is securely a titanosauriform within Somphospondyli. More specific relationships have changed between analyses. In 2010, new neck material helped revise comparisons with other Asian sauropods and weakened earlier arguments that some were brachiosaurids. Placement within Euhelopodidae or close to individual genera is less stable than the broader titanosauriform assignment.

Age and environment

Bor Guve consists of siltstone and channel sandstone associated with the Bayn Shire Formation in the eastern Gobi. The 2006 description treated the deposits as late Early Cretaceous; later correlations place the area in the early part of the Late Cretaceous. Formation-level dates constrain the wider succession but are not a direct radiometric date on the type bed. The precise age of IGM 100/1803 remains less certain than its broad Cretaceous assignment.

Other sauropod vertebrae found kilometres from the holotype were not assigned to Erketu. They cannot be combined into its skeleton merely because they came from the same region. Plant eating follows from its sauropod relationships, while its exact menu, feeding height, group life, migration, colour and reproduction are not recorded. See other long-necked dinosaurs in the A–Z dinosaur catalogue.

Frequently asked questions

What is unusual about Erketu?

Several preserved neck vertebrae are exceptionally elongated. Their proportions are direct measurements, but the total neck length and body proportions are not known.

How complete is the Erketu skeleton?

The type individual preserves the front of the neck, a sternal plate and parts of one hind limb. The skull, trunk, tail, pelvis and much of the limbs are missing.

How long was Erketu?

Reconstructions around 12–15 metres are approximate and rely on related sauropods because the trunk and tail are absent. The fossils do not support one exact total length or mass.

Where does Erketu fit among sauropods?

It is a titanosauriform somphospondylan. Its more precise position, including proposed family relationships, varies among phylogenetic analyses.