Fossil range: Late Olenekian (Early Triassic), 249.7–245 Ma
Life restoration of Hupehsuchus nanchangensis
Scientific classification














Hupehsuchia is an order of enigmatic diapsid marine reptiles belonging to the class Sauropsida. The order was short-lasting, with a temporal range restricted to the Spathian age of the late Olenekian, spanning only a few million years of the Early Triassic.[1] The order gets its name from Hubei Province, China, from which many specimens have been found. Of the entire order, Nanchangosauridae is the only recognized family. Hupehsuchians display an unusual combination of characteristics with other derived marine reptiles, including mosasaurs and ichthyosaurs, including polydactyly, narrow skull, and vivipary.

The order consists of two genera, Hupehsuchus and Nanchangosaurus, of which, the former is the type genus. Determining the exact classification and phylogeny of this group remains difficult, and may be attributed to the limited knowledge of the fossil record of diapsid reptiles in the Late Permian and Early Triassic, as well as the great amount of convergence exhibited by secondarily aquatic reptiles.


Hupehsuchians display an unusual combination of characteristics. The overall shape of the body is fusiform, with a long tail and large, paddle like limbs. The skull is elongate and the jaws are edentulous. The rostrum is flatteded with the premaxilla thought to form most of the dorsal and lateral surface while the maxilla is mostly restricted to the ventral surface beyond the base of the rostrum. An opening between the nasal and the prefrontal in one hupehsuchian specimen (known as IVPP V3232) was initially interpreted as an antorbital fenestra but is now thought to be an artifact caused by the damage of the surrounding bones during preservation. Its position is not indicative of a narial opening either. More likely the naris lies between the nasal and the maxilla in an area anterior to that of the supposed antorbital fenestra, although the preservation of this area in known specimens is too poor to prove definitively that it is the external naris and not an artifact of preservation as is the case for the fenestra.[2]

The neck is relatively elongated and the cervical ribs are short.[2] An unusual feature of the neural spines of the trunk region, from the 11 vertebra to the first caudal, is that each is divided into two distinct units by a suture line. One is proximinal to the rest of the vertebra and the other is distal to it. There is some anteroposterior displacement of these two units along the vertebral column, suggesting that they are not ossified to one another. In vertebrae 8-14, the distal ends of the distal portions of the neural spines are expanded posteriorly. An additional anterior expansion of the neural spine is seen in all vertebrae after the 14th. These expanded distal regions exhibit some sculpturing, and may have penetrated the dermis.[2] Another unusual characteristic of hupehsuchians is the presence of dermal plates over the cervical and trunkneural spines of the approximately 34 presacral vertebrae. A small dermal bone overlies each space between the posterior expansion of one distal neural spine and the anterior expansion of the one behind it. Above these dermal bones lie even larger dermal plates that directly overly even numbered neural spines.[2]

Gastralia are present in some specimens that form a type of ventral armor from the pectoral to the pelvic girdle. The medial row cosists of large, overlapping, V-shaped elements, and lateral rows consist of smaller, cylindrical bones that are widely spaced.[2]


Hupehsuchia was first defined as a suborder of "Thecodonta" in 1972 when it was first thought to be a group of early archosaurs.[3] It was also thought to be related to several other groups of Triassic reptiles previously thought to be clearly distinct. The presence of the supposed antorbital fenestra described above was seen as evidence for grouping hupesuchians within Archosauria, but the antorbital fenestra characteristic of archosaurs is surrounded by the maxilla and lacrimal, not the nasal and prefrontal. The dermal plates were also seen as evidence for terrestrial archosaur ancestors and comparisons were made with some early armored forms. However, a later study in 1976 could find no similarities between the dermal plates of hupehsuchians and the armor of early archosaurs.[4] It is more likely that the ancestors of hupehsuchians were earlier, more basal terrestrial diapsids, as suggested by several synapomorphies they share with such primitive ancestors. In 1991, Hupehsuchia was recognized as a distinct order after better methods of specimen preparation allowed new features to be revealed that distinguished Hupehsuchia from all other diapsid orders.[2]



Hupehsuchus nanchangensis on display at the Paleozoological Museum of China.

Only two named genera are known to belong to the order, and both are members of the same family, Nanchangosauridae. They are likely sister taxa. Both have been found from the Jialingjiang Formation in Hubei Province, China. Originally this formation was dated to the Anisian stage of the Middle Triassic but it is now thought to have formed slightly earlier during the late Olenekian stage of the Early Triassic. The first to be described was Nanchangosaurus, named in 1959.[5] Fossils of the genus have been found from the Daye Limestone in the Hsunjian District of Hubei Province from a single specimen.[5] Another genus, Hupehsuchus, was named in 1972 from a locality in the Xunjian Commune of Nanzhang County.[3] The two genera do not differ greatly, and were probably quite similar in appearance. Nanchangosaurus possesses a frontal that participates in the orbital margin and is quite long, more similar to what is seen in basal diapsids. Additionally, the dorsal plates lack sculpturing.[5] In Hupehsuchus, the zygapophyses of the trunk region are more modified,[3] while those of Nanchangosaurus resemble those of more primitive terrestrial reptiles.[5] Overall, Nanchangosaurus was smaller in size than Hupehsuchus, and although several characteristics of the skeleton single representative specimen of the former genus suggest it belonged to a juvenile individual, other differences between the two genera are clearly non-ontogeneic and dispel the possibility that the single specimen of Nanchangosaurus could be from an immature Hupehsuchus.[2] Different means of preservation can also be taken as evidence for different body forms in the two genera. It is important to note that all specimens of Hupehsuchus are preserved in lateral view[3] while the single specimen of Nanchangosaurus is preserved in dorsal view (except for the neural spines, which were probably too long to be preserved in this manner).[5] It is possible that Hupehsuchus had a more laterally compressed body than Nanchangosaurus, which, as a result, would favor preservation in lateral view.[2]

Unnamed generaEdit


Polydactylous hupehsuchian

Life restoration of a polydactylous hupehsuchian based on SSTM 5025.

One specimen, called V4070, has been found from the same locality as Hupehsuchus and is representative of a different, unnamed genus.[2] The pectoral girdle, neck, and base of the skull are present but are all badly weathered, with impressions present in the underlying matrix of the area that provide little more detail. An impression has been made by much of the postcranial skeleton but that too has experienced much weathering. V4070 is proportionally quite similar to Hupehsuchus, with the tail being around 93% the length of that of Hupehsuchus, but studies of the fore and hindlimbs reveal many differences between the two genera, with carpal and tarsal configuration differing greatly. The overall form of the limbs is more paddle-like than what is seen in Hupehsuchus, although the phalanges still retain a cylindrical configuration. Additional differences include shorter neural spines and neural and haemal arches in the dorsal and caudal vertebrae in V4070, although the cervical neural spines seem to be longer than those of Hupehsuchus. The dermal plates seem to extend down to the caudal vertebrae, unlike specimens of Hupehsuchus in which the dermal plates are absent from the caudal vertebrae. It is possible that caudal dermal plates were present in Hupehsuchus but have been lost during preservation.[2]

SSTM 5025Edit

In late 2003, a new specimen of hupehsuchian known as SSTM 5025, was found from the same area as Hupehsuchus and V4070,[2] was briefly mentioned in the journal Nature.[6] The remains of SSTM 5025 include the part and counterpart of an almost complete skeleton from which the anterior end of the snout and the tail tip are missing.[6] It was recovered from the Jialingjiang Formation near Xunjiansi, Nanzhang County, Hubei Province, China, and is referred to the only known order of Hupehsuchia, Nanchangosauridae. It is most notable for exhibiting polydactyly, in which there more than the usual maximum of five digits per limb as seen in most advanced tetrapods. Polydactyly is also seen in ichthyosaurs.[6] However, in ichthyosaurs, this condition occurs as either bilateral polydactyly in the case of ophthalmosaurids (extra digits anterior to digit I and posterior to digit V) or interdigital or postaxial phalangeal bifurcation as in non-ophthalmosaurids.[7] Preaxial polydactyly occurs in SSTM 5025, where extra digits only develop anteriorly to digit I. This condition is seen in earlier stem tetrapods from the Devonian period such as Ichthyostega and Acanthostega.[8] SSTM 5025 possessed seven digits on the forelimbs and six on the hindlimbs. The wide manus and pes of the specimen resemble the limb-like fins of extant frogfishes.[9][10]


Hupehsuchians were clearly well adapted to marine life, as they possessed limbs that were paddle-like in shape and had fusiform bodies. It is likely that the characteristically elongated neural spines were associated with well developed epaxial muscles (muscles lying above the transverse process of the vertebrae) that facilitated lateral undulation in an axial subundulatory mode.[11] The pattern of articulation in the vertebrae suggest that such undulation was concentrated posteriorly near the pelvic girdle and tail. It is likely that Hupehsuchus was better equipped for lateral undulation as a means of locomotion than Nanchangosaurus was. This is evidenced by the assumed greater degree of lateral compression in the body of the former genus as well as generally more elongate neural spines.[2]

The presence of polydactyly in SSTM 5025 may have been an adaptation to moving across underwater substrates in a similar manner to some early tetrapods of the Devonian such as Acanthostega.[6][8]

The flat, toothless rostrum may have supported an avian-like bill, or perhaps rows of baleen as seen in cetaceans of the suborder Mysticeti.[2] It seems that hupehsuchians were adapted to continuous ram feeding, a form of mobile suspension feeding.[12] The large skull and lack of a fixed symphysis with the lower jaw is indicative of such a feeding method, but the narrowness of the rostrum and the existence of marine vertebrates with larger skulls that were not completely edentulous (e. g. ichtyosaurs) seem to point against it.[2] The long necks of hupehsuchians would seem to inhibit continuous ram feeding at high speeds through the water, so it is more likely that they practiced intermittent ram feeding at slower speeds.[2]

The purpose of the dorsal dermal plates of hupehsuchians is unknown. These plates as well as the ventral gastralia and the tendency for pachyostosis would have added considerable weight to hupehsuchians, allowing them to have neutral buoyancy. However, the position of the dorsal plates are high above the center of gravity, and it seems that this would have made the bodies of hupehsuchians unstable. Hupehsuchians were among the largest marine animals of their time, so there would be no need for dorsal plates as a protective measure. The tendency for the dorsal plates to be more developed anteriorly may have allowed the anterior portion of the vertebral column to remain relatively rigid while the posterior portion could freely undulate.[2]

It is unsure whether the plates were acquired subsequent to an adaptation to the marine environment or were inherited from a terrestrial ancestor, in which such plates would have given rigidity to the spine and protection from predators. In any case, if Nanchangosaurus is seen as representative of an ancestral morphological pattern that led to the more derived Hupehsuchus, then the plates seem to have experienced further development in hupehsuchians and must have had some advantage.[2]

The general shape of hupehsuchians seems to have made any locomotion on land nearly impossible, and, although there is no direct evidence seen in known specimens, these animals were probably viviparous, giving live birth at sea rather than laying eggs on land.[2] Vivipary is also seen in ichthyosaurs.[13]

Taxonomy and phylogeneticsEdit

Relationship to ichthyosaursEdit

Temnodontosaurus plat1DB

Hupehsuchians bear a resemblance to some ichthyosaurs, such as the Temnodontosaurus platyodon pictured here.

The assumed position of the naris in hupehsuchians as mentioned above can be taken as evidence for possible ichthyosaurian affinities, as it is in the same general area as those of ichthyosaurs.[2] Hupehsuchians do resemble earlier ichthyosaurs in outward appearance with slightly fusiform bodies and long, straight, non-lunate tails. Other features shared with ichtyosaurs include a supraoccipital similar to what is seen in early forms, a relatively long antorbital region, and a short transverse process for the ribs. Many more differences exist between hupehsuchians and ichthyosaurs, however. In hupehsuchians, the surfaces of the vertebral centra that articulate with one another are distinctively flat, or acelous, while in ichtyosaurs they are noticeably heterocelous (it is also important to note that the surfaces of centra in the possible early diapsid ancestors of hupehsuchians were amphicelous).[2] Yet there were some early ichthyosaurs and ichthyosaur relatives such as Chaohusaurus and Utatsusaurus that possessed vertebrae that were not deeply heterocoelus and more closely resembled those of hupehsuchians.[3][14] Unlike more derived ichthyosaurs, the centra of these two genera are about as long as they are high.[2] In hupehsuchians, where the neural arches dominate the vertebral column, the height of the centra is reduced and the hight to length ratio of the centra is smaller, meaning that they are also about as long as they are high. This may suggest that hupehsuchians may have evolved from an ichthyosaur relative.[2] It has also been suggested that hupehsuchians were related to or members of Sauropterygia. Indeed, Nanchangosaurus was classified as a sauropterygian upon its initial description.[5]

Many of the features seen in known hupehsuchian specimens that are comparable to those of more well known diapsids may not necessarily be evidence of ancestry or relationship. The lower jaw and rostrum of hupehsuchians have been compared to many other secondarily aquatic tetrapods such as plesiosaurs, whales, and the early marine bird Hesperornis, all of which have developed a similar morphology independently in response to the need for adaptation to a marine environment.[2]


SSTM 5025 hupehsuchian

Specimen SSTM 5025

Classification of Hupehsuchia remains difficult because most of the derived characters exhibited in the clade that can be helpful in phylogenetic analyses are also present in other unrelated groups of secondarily aquatic reptiles, and the overall record of diapsids during the Late Permian-Early Triassic is relatively poor, making it difficult to find any closely related or ancestral taxa. Even higher level classification is difficult because many of the plesiomorphies that characterize such groups are absent in the highly derived, marine adapted hupehsuchians. For example, several characteristics suggest that Hupehsuchia belongs within Neodiapsida, but most of the derived characteristics that define the clade are absent or hard to distinguish in hupehsuchians, even though their ancestors may have possessed these characteristics at one point. Three derived characteristics of Neodiapsida are a reduced number of teeth on the pterygoid, an absence of teeth on the parasphenoid, and a lack of caniniform mixillary teeth, but none of these apply to hupehsuchians because they lack teeth altogether.[15] Many derived characteristics that define neodiapsids regard the limbs and girdles, but these characteristics are not seen in marine reptiles because the limbs and girdles are too highly modified. Many characteristics of the skull cannot be seen in hupehsuchians due to poor preservation of remains. Therefore, any placement of Hupehsuchia within Neodiapsida remains tentative until more specimens are found.

Due to the great number of derived characteristics in hupehsuchians that are a result of convergence, computational phylogenetic analyses using computer programs based on the method of maximum parsimony do not produce cladograms that can accurately establish a relationship between Hupehsuchians and other diapsids. Such programs do not recognize situations where sister group relations are not justified.[2]


  1. ^ Motani, R. (2009). The Evolution of Marine Reptiles. Evolution: Education and Outreach 2(2):224-235
  2. ^ a b c d e f g h i j k l m n o p q r s t u v w x Carrol, Robert L.; Dong, Z.-M. (1991). "Hupehsuchus, an enigmatic aquatic reptile from the Triassic of China, and the problem of establishing relationships". Philosophical Transactions: Biological Sciences 331 (1260): 131-153. 
  3. ^ a b c d e Young, C. C.; Dong, Z.-M. (1972). "On the aquatic reptiles of the Triassic in China". Vert. Paleont. Mem 9: 1-34. 
  4. ^ Charig, A. J., Krebs, B., Sues, H.-D. and Westphal, F. (1976). Encyclopedia of Paleoherpetology (ed. O. Kuhn), part 13. Stuttgart: Gustav Fischer Verlag.
  5. ^ a b c d e f Wang, K. (1959). Ueber eine neue fossile Reptilform von Provinz Hupeh, China. Acta Palaeontologica Sinica 7(5):373-378.
  6. ^ a b c d Wu, X.-C.; Li, Z.; Zhou, B.-C.; and Dong, Z.-M. (2003). "A polydactylous amniote from the Triassic period.". Nature 426: 516. 
  7. ^ Motani, R. (1999). On the evolution and homologies of ichthyopterygian forefins. Journal of Vertebrate Paleontology 19(1):28-41.
  8. ^ a b Coates, C. C.; and Clack, J. A. (1990). "Polydactyly in the earliest known tetrapod limbs". Nature 347: 66-99. 
  9. ^ Clack, J. A. (2002). Gaining Ground: the Origin and Evolution of Tetrapods. Indiana University Press, 369 pp.
  10. ^ Edwards, J. L. (1989). Two Perspectives on the Evolution of the Tetrapod Limb. American Zoologist 29(1):235-254.
  11. ^ Massare, J. A. (1988). Swimming capabilities of Mesozoic marine reptiles: implications for method of predation. Paleobiology 14:187-205.
  12. ^ Sanderson, S. L. and Wassersug, R. (1990). Suspension-feeding vertebrates. Scientific American 262(3):96-101.
  13. ^ William Antony S. Sarjeant and L. B. Halstead. Vertebrate fossils and the evolution of scientific concepts. Chapter: An ichthyosaur embryo from the lower Lias (Jurassic):Hettangian) of Somerset, England, with comments on the reproductive biology of ichthyosaurs. Routledge; 1 edition (February 1, 1996). ISBN-10: 2881249965 ISBN-13: 978-2881249969.
  14. ^ Shikama, T., Kamei, T. and Murata, M. (1978) Early Triassic ichthyosaurus, Utatsusaurus hataii gen. et. sp. nov., from the Kitakami Massif, Northeast Japan. Tohoku University, Science Reports, ser. 2 48(2):77-97.
  15. ^ Benton, M. J. (1985). Classification and phylogeny of the diapsid reptiles. Zoological Journal of the Linnean Society 84(2):97-164.

Ad blocker interference detected!

Wikia is a free-to-use site that makes money from advertising. We have a modified experience for viewers using ad blockers

Wikia is not accessible if you’ve made further modifications. Remove the custom ad blocker rule(s) and the page will load as expected.