The "flightless pterodactyl" that never was: Ornithopsis hulkei

Prior to and during the Victoria era, British fossil hunters came upon huge or peculiar bones of reptiles from Middle Jurassic to Early Cretaceous deposits in England that they interpreted as belonging to huge crocodile-like archosaurs, namely those from the Oolite Group of the Midlands, the Kimmeridge Clay of southern and eastern England, and the Wealden Supergroup of Sussex and the Isle of Wight. For instance, the type specimens of the basal eusauropod Cetiosaurus and the basal titanosauriform  Pelorosaurus were initially thought to have represented gigantic sea-going crocodiles, until more complete finds in the 1870s showed that they were actually dinosaurs and not crocodiles. However, most paleontology gurus overlook the fact that one Early Cretaceous titanosauriform sauropod from the UK, Ornithopsis, was misinterpreted by its describer as belonging not to a huge crocodile-like reptile, but instead as a flightless pterosaur!

Anterior view of the lectotype of Ornithopsis hulkei (NHMUK R28632) (from Owen 1875)

The story of the discovery and naming of Ornithopsis begins in the early 1850s, when an anterior dorsal vertebra was found in the Early Cretaceous (Barremian) Wessex Formation of the Isle of Wight along the English Channel coast of southern England and kept by Gideon Mantell (describer of Pelorosaurus) in his personal fossil collection, before being acquired by the British Museum in 1853 (a year after Mantell's death) and assigned the catalogue number BMNH R28632 (now NHMUK R28632). The dorsal vertebra, however, was not published in the scientific literature until Seeley (1870) erected the name Ornithopsis hulkei for NHMUK R28632 as well as NHMUK R2239, a dorsal vertebra found in the Early Cretaceous (late Valanginian) Tunbridge Wells Sand Formation of West Sussex in the 1820s by Gideon Mantell and misidentified by Owen (1854) as a quadrate of Iguanodon. He noted that NHMUK R28632 and NHMUK R2239 had cavities for air sacs seen in the bones of birds and pterosaurs, and thus surmised that Ornithopsis could be a missing link between pterosaurs and birds, but also possibly allied with dinosaurs, hence the name Ornithopsis meaning "bird face" in Greek.

Sir Richard Owen (1804-1892), who correctly determined that Ornithopsis was a sauropod dinosaur and not a flightless pterosaur

The identification of Ornithopsis as potentially being a flightless pterosaur would not hold water for very long, however. Owen (1875) agreed with Seeley (1870) that NHMUK R2239 was a dorsal vertebra rather than a quadrate, but he rejected Seeley's interpretation of Ornithopsis as a close relative of birds and pterosaurs and instead considered NHMUK R2239 and R28632 to be congeneric with his new sauropod genus Bothriospondylus from the Late Jurassic (Kimmerdgian) Kimmeridge Clay Formation of Wiltshire. The hypodigm for Ornithopsis hulkei was split into two species, with NHMUK R28632 receiving the new name Bothriospondylus magnus and NHMUK R2239 being made the holotype of the new species Bothriospondylus elongatus. For one thing, Richard Owen was a creationist and not a fan of Darwinist thought, so his lack of enthusiasm for Charles Darwin's theory of evolution endeared him to recognize that Ornithopsis belonged to a sauropod and not a pterosaur-like archosaur. Owen also must have been aware that because the Bothriospondylus elongatus holotype is from an older horizon than NHMUK R28632, the two vertebrae were most likely not conspecific. As a matter of fact, a year after he assigned the Ornithopsis hulkei material to Bothriospondylus, Owen (1876) changed his mind about B. magnus being congeneric with the Bothriospondylus type species (B. suffosus) and referred it to the sauropod genus Chondrosteosaurus from the same geologic horizon and location as NHMUK R28632. Forthwith, Ornithopsis would now be recognized not as a flightless pterodactyl, but instead as a member of Dinosauria --- although Richard Owen had been familiar with extinct and extant flightless birds, no one ever found a genuine pterosaur fossil with flightless abilities. 

In the 1860s and 1870s additional sauropod material was uncovered from the Wessex Formation of the Isle of Wight by Reverend William Fox and John Whitaker Hulke, including some vertebrae that would become the type specimens of the titanosauriforms Chondrosteosaurus magnus and Eucamerotus foxi. In his description of the new titanosauriform remains from the Isle of Wight, Hulke (1879) disputed Owen's opinion about the generic name Ornithopsis being misleading by pointing out that the syntypes of O. hulkei were lightly constructed regardless of the reclassification of Ornithopsis as a sauropod. He designated NHMUK R28632 as the lectotype of O. hulkei, making Bothriospondylus magnus a junior objective synonym of Ornithopsis, and the genera Chondrosteosaurus and Eucamerotus (the latter also described from the Isle of Wight) were synonymized with Ornithopsis. Although Hulke (1882, p. 375) treated the holotype of Bothriospondylus elongatus as the O. hulkei lectotype and referred NHMUK R28632 and the type material of Eucamerotus foxi the Wessex Formation of the Isle of Wight to his new species Ornithopsis eucamerotus, the earlier lectotype designation for O. hulkei by Hulke (1879) stands, as pointed out by Lydekker (1888). 

Although Ornithopsis had been wrongly interpreted as a flightless pterosaur when first named in 1870, it was nonetheless one of the first sauropod taxa to be described from the Isle of Wight and the fourth sauropod taxon described from the Early Cretaceous of Europe (after Pelorosaurus, Haestasaurus, and Oplosaurus). On a few occasions, Ornithopsis was synonymized with Pelorosaurus by von Huene (1909), Romer (1956), and Steel (1970) but Blows (1995) noted that the O. hulkei lectotype does not overlap with the holotype of Pelorosaurus conybeari and found Ornihopsis to be a distinct and valid genus of basal titanosauriform (followed by Upchurch et al. 2011).

References:

Blows, W.T., 1995. The Early Cretaceous brachiosaurid dinosaurs Ornithopsis and Eucamerotus from the Isle of Wight, England. Palaeontology 38 (1): 187–197.

Huene, F. v., 1909. Skizze zu einer Systematik und Stammesgeschichte der Dinosaurier. Centralblatt für Mineralogie, Geologie und Paläontologie 1909:12-22.

Hulke, J.W., 1879. Note (3rd) on (Eucamerotus, Hulke) Ornithopsis, H. G. Seeley, = Bothrospondylus magnus, Owen, = Chondrosteous magnus, Owen. Quarterly Journal of the Geological Society 35 (1–4): 752–762.

Hulke, J.W., 1882. Note on the Os Pubis and Ischium of Ornithopsis eucamerotus. Quarterly Journal of the Geological Society 38 (1–4): 372–376.

Lydekker, R. (1888). Catalogue of the Fossil Reptilia and Amphibia in the British Museum (Natural History). Part I. Containing the Orders Ornithosauria, Crocodilia, Dinosauria, Squamata, Rhynchocephalia, and Proterosauria. British Museum (Natural History). Department of Geology. 309 pp.

Owen, R., 1854. Monograph on the Fossil Reptilia of the Wealden Formations. Part II. Dinosauria (Iguanodon).  Monographs of the Palaeontographical Society 8 (27): 1–54.

Owen, R., 1875. Monographs on the British Fossil Reptilia of the Mesozoic Formations. Part II. (Genera Bothriospondylus, Cetiosaurus, Omosaurus). Monographs of the Palaeontographical Society 29 (133): 15–93.

Owen, R., 1876. Monograph on the Fossil Reptilia of the Wealden and Purbeck Formations. Supplement No. VII. Crocodilia (Poikilopleuron) and Dinosauria? (Chondrosteosaurus). Monographs of the Palaeontographical Society 30 (136): 1–7.

Romer, A.S., 1956. Osteology of the Reptiles. University of Chicago Press: Chicago: IL 772 pp. 

Seeley, H.G., 1870. Ornithopsis, a gigantic animal of the Pterodacyle kind from the Wealden. Annals and Magazine of Natural History 5 (4): 305–318.

Steel, R., 1970. Part 14. Saurischia. Handbuch der Paläoherpetologie. Gustav Fischer Verlag: Stuttgart, 87 pp.

Upchurch, P., Mannion, P.D., and Barrett, P.M., 2011. Sauropod dinosaurs. pp. 476–525. In: Batten, D.J. (ed.). English Wealden Fossils. The Palaeontological Association.

Personal thoughts on Amphicoelias paper by Mannion et al. (2021)

During the Bone Wars in the late 1800s, Edward Drinker Cope (1840-1897) and Othniel Charles Marsh (1831-1899) described several sauropod taxa from the Morrison Formation of western North America, with Marsh erecting the most sauropod species from the Morrison. Although several sauropod genera erected by Marsh have stood the taxonomic test of time, like Apatosaurus, Barosaurus, Brontosaurus, and Diplodocus, the only sauropod genus from the Morrison Formation named by Cope whose validity has been upheld is Camarasaurus, while Caulodon has been synonymized with Camarasaurus (but see here). One sauropod genus described from the Morrison Formation by E.D. Cope whose taxonomic status has fluctuated over time, however, is Amphicoelias Cope, 1877. Although Amphicoelias is poorly known in terms of the holotype of its type species, A. altus, being representing by a few elements, one nominal species of Amphicoelias, A. fragillimus, enjoyed conjectural fame as a super-giant diplodocid until Carpenter (2018) drastically reduced the size estimates for this taxon to 102 feet (31 meters) and reclassified it as a rebbachisaurid, erecting the new genus Maraapunisaurus for it. On the other hand, the validity and precise systematic position of Amphicoelias has been debated, with some studies placing it as a basal diplodocoid and others recovering it as diplodocid. Recently, a new paper on the anatomy and systematics of Amphicoelias was published by Mannion et al. (2021), and while it reaffirms the validity of Amphicoelias as upheld by several authors, I have taken the liberty of expressing some thoughts about the Mannion et al. paper regarding Amphicoelias with respect to diagnostic characters, phylogenetic position, and the bearing of studies on Morrison diplodocoid ontogeny upon Morrison sauropod diversity. 

Selected elements of the holotype of Amphicoelias altus (AMNH 5764): posterior dorsal vertebra (top) and right femur (bottom) (from Mannion et al. 2021)

In the section of their paper in which they redescribe the holotype of Amphicoelias altus (AMNH 5764), Mannion et al. list the "femoral shaft with subcircular cross-section" as one of three autapomorphies for Amphicoelias in the revised diagnosis for this taxon, noting that the femur of AMNH 5764 differs from described Morrison diplodocoid taxa in having a ratio of the mediolateral to anteroposterior diameter of the femur being 0.99 to 1.1 (despite a few signs of taphonomic crushing). However, they also note that the dicraeosaurid specimen MOR 592 found in Montana also has a femur whose cross-section is subcircular at the midshaft; in fact, the subcircular cross-section of the femur was used by Wilson and Smith (1996) to justify referring MOR 592 to Amphicoelias and conclude that Amphicoelias was a basal diplodocoid based on cladistic results that were never published. However, Whitlock (2011) assigned MOR 592 to the family Dicraeosauridae due to the presence of a sharp supraoccipital crest and a symphyseal tuberosity on the dentary, although Woodruff & Fowler (2012) and Woodruff et al. (2017) regarded MOR 592 as an immature diplodocine specimen, but nevertheless recovered Amphicoelias as a basal diplodocoid more derived than Haplocanthosaurus and Amazonsaurus. According to Mannion et al. (2021), the ratio of the mediolateral to anteroposterior diameter of the femur of MOR 592 is approximately 1.3, slightly greater than that for Amphicoelias altus, and MOR 592 has a femur with a slightly beveled distal end in contrast to the more pronounced beveling of the distal femur of AMNH 5764. On the other hand, Tschopp et al. (2015) note that the holotype of Brontosaurus parvus (CM 566) also has a subcircular femoral cross-section, while Wilhite (2005) reports that the subcircular femoral cross-section observed in Amphicoelias, the Brontosaurus parvus holotype, and MOR 592 also occurs in a few diplodocid femora from the Dry Mesa Quarry in Colorado. Since Amphicoelias is recovered as either a basal diplodocid by Tschopp et al. (2015) or an apatosaurine diplodocid by Tschopp and Mateus (2017), whereas Amphicoelias is variously recovered as a stem diplodocoid more derived than Haplocanthosaurus or a diplodocid by Mannion et al. (2021), the subcircular femoral cross-section described for Amphicoelias most likely evolved convergently among a few taxa within Flagellicaudata because Amphicoelias altus is distinguished by Mannion et al. (2021) from all other diplodocoids in having the apex of the posterior dorsal neural spine with rounded, non-tapered lateral projections resulting from the expansion of spinodiapophyseal laminae and little material is preserved in AMNH 5764.

Stratigraphic chart of dinosaur localities in the Morrison Formation (from Turner and Peterson 1999). Despite the opinion of some that the diversity of diplodocoids in the Morrison Formation has been inflated, the type localities of Haplocanthosaurus delfsi (CO-5) and Brontosaurus yahnanpin (WY-44) are stratigraphically low in the Morrison Formation, and type locality of Amphicoelias altus (CO-71) is situated near the top of the Brushy Basin Member of the Morrison Formation, being stratigraphically higher than the type localities of Apatosaurus ajax, A. louisae, Brontosaurus excelsus, and B. parvus. Moreover, three different groups of the diplodocoids (haplocanthosaurids, diplodocids, and dicraeosaurids) have been found at the Felch Quarry 1 (CO-3) in Garden Park, Colorado.  

When addressing the question of whether or not some Morrison diplodocoid species are growth stages of well-known taxa as hinted by Woodruff (2019), Mannion et al. stress that the basal position of the genus Haplocanthosaurus within Diplodocoidea is not attributable to ontogeny given that known specimens of H. priscus and H. delfsi are of the adult/near-adult stage. When taking into account the cladistic diversity and stratigraphic distribution of sauropods within the Morrison Formation, it should be noted that Brontosaurus (=Eobrontosaurus) yahnahpin and Haplocanthosaurus delfsi hail from the lower half of the upper part of the Salt Wash Member of the Morrison Formation whereas Amphicoelias altus was found near the top of the Brushy Basin Member (Turner and Peterson 1999, fig. 7) and that no members of Turiasauria or Mamenchisauridae have yet been reported from the Morrison Formation, although the Lourinhã and Tendaguru Formations have yielded members of Diplodocoidea, Macronaria and Turiasauria. Additionally, given that Whitlock and Wilson Mantilla (2020) note that the juvenile diplodocine specimens CM 3452 and CM 11255 (the latter probably Barosaurus; Melstrom et al. 2016) differ from Kaatedocus, Smitanosaurus, Suuwassea, and MOR 592 in lacking a postparietal foramen despite being juveniles, although the adult apatosaurine specimen BYU 17096 has this feature, it is not hard to imagine four dicraeosaurid taxa existing in the Morrison Formation because known specimens of Kaatedocus, Smitanosaurus, and MOR 592 were found in the upper part of the Salt Wash Member and lowermost part of the Brushy Basin Member, whereas Suuwassea probably was found high in the Morrison Formation (Harris and Dodson 2004; Turner and Peterson 1999). While I agree with Mannion et al. (2021) that ontogeny is an important factor to take into account when determining whether small or primitive sauropod specimens from the Morrison Formation are juveniles of existing species or more basal than well-known diplodocids, the assignment of Suuwassea and MOR 592 to Dicraeosauridae by Whitlock (2011) took into account the possibility that the sub-adult status of the Suuwassea holotype was why Suuwassea defied precise classification within Diplodocoidea when first described by Harris and Dodson (2004), while bearing in mind the fact that some characteristics used to refer MOR 592 to Amphicoelias by Wilson and Smith (1996) were likely to be found in other diplodocoid taxa. Moreover, since Brontosaurus yahnanpin was found lower in the Morrison Formation than Amphicoelias or other species of Brontosaurus, it is possible that it is actually more basal than either B. excelsus, B. parvus, or Apatosaurus because the holotype of Amphicoelias altus contains a few skeletal elements and was found in the uppermost layer of the Brushy Basin Member.     

References:

Carpenter, K., 2018. Maraapunisaurus fragillimus, N.G. (formerly Amphicoelias fragillimus), a basal Rebbachisaurid from the Morrison Formation (Upper Jurassic) of Colorado. Geology of the Intermountain West 5: 227–244.

Harris, J.D. and Dodson, P., 2004. A new diplodocoid sauropod dinosaur from the Upper Jurassic Morrison Formation of Montana, USA. Acta Palaeontologica Polonica 49 (2): 197–210.

Mannion P.D., Tschopp E., and Whitlock, J.A. 2021. Anatomy and systematics of the diplodocoid  Amphicoelias altus supports high sauropod dinosaur diversity in the Upper Jurassic Morrison Formation of the USA. Royal Society Open Science 8 (6): Article ID 210377.  doi:10.1098/rsos.210377        

Melstrom, K.M., D’Emic, M.D., Chure, D.J., and Wilson, J.A., 2016. A juvenile sauropod dinosaur from the Late Jurassic of Utah, USA, presents further evidence of an avian style air-sac system. Journal of Vertebrate Paleontology e1111898.

Tschopp, E., Mateus, O., and Benson, R.B.J., 2015. A specimen-level phylogenetic analysis and taxonomic revision of Diplodocidae (Dinosauria, Sauropoda).  PeerJ  3:e857  https://doi.org/10.7717/peerj.857  

Tschopp, E., and Mateus, O., 2017. Osteology of Galeamopus pabsti sp. nov. (Sauropoda: Diplodocidae), with implications for neurocentral closure timing, and the cervico-dorsal transition in diplodocids. PeerJ 5:e3179 https://doi.org/10.7717/peerj.3179

Turner, C.E. and Peterson, F., 1999. Biostratigraphy of dinosaurs in the Upper Jurassic Morrison Formation of the Western Interior, U.S.A. pp. 77–114. In: Gillette, D.D. (ed.), Vertebrate Paleontology in Utah. Utah Geological Survey Miscellaneous Publication 99-1.

Wilhite, D.R. 2005. Variation in the appendicular skeleton of North American sauropod dinosaurs: taxonomic implications. pp. 268-301. In: Tidwell, V., and Carpenter, K. (eds.), Thunder-lizards: the Sauropodomorph dinosaurs. Indiana University Press, Bloomington.

Wilson, J.A., and Smith, M., 1996. New remains of Amphicoelias Cope (Dinosauria: Sauropoda) from the Upper Jurassic of Montana and diplodocoid phylogeny. Journal of Vertebrate Paleontology 16 (supp. to volume 3): 73A.

Whitlock, J. A. 2011. A phylogenetic analysis of Diplodocoidea (Saurischia: Sauropoda). Zoological Journal of the Linnean Society 161: 872–915.

Whitlock, C., and Wilson Mantilla, J., 2020. The Late Jurassic sauropod dinosaur  'Morosaurus’  agilis  Marsh, 1889 reexamined and reinterpreted as a dicraeosaurid. Journal of Vertebrate Paleontology 40  (6) DOI: 10.1080/02724634.2020.1780600

Woodruff, C., and Fowler, D. W. 2012., Ontogenetic influence on neural spine bifurcation in Diplodocoidea (Dinosauria: Sauropoda): A critical phylogenetic character. Journal of Morphology 273: 754–764. 

 

Woodruff, D. C., Fowler, D. W. and Horner, J. R., 2017. A new multi-faceted framework for deciphering diplodocid ontogeny. Palaeontologia Electronica 20.3.43A: 1–53.