Gwyneddosaurus is Tanytrachelos- misusing nomen dubium part II

Remember that post on Agnolin et al.'s paper where they used the term nomen dubium incorrectly for Rapator?  Well, it seems Olsen has a similar issue with Gwyneddosaurus where he thinks it's clearly synonymous with a taxon he named later, and even includes a diagnosic element, but wants to make it a nomen dubium so that his more complete specimen can be the name-bearer.  Read on for more details of this ex-theropod.

Gwyneddosaurus Bock, 1945
= Tanytrachelos Olsen, 1979
G. erici Bock, 1945
= Tanytrachelos ahynis Olsen, 1979
Late Carnian, Late Triassic
Gwynedd or North Wales Member of Lockatong Formation, Pennsylvania, US

Holotype- (ANSP 15072) braincase or palatal element?, axis (8 mm), two anterior cervical neural arches (~8 mm), three posterior cervicals (7.5 mm), cervical vertebra, cervical ribs, dorsal vertebra, at least 11 dorsal ribs, gastralia, distal caudal vertebra (4 mm), three chevrons, heterotopic bone?, neural arch, scapulocoracoids, clavicles, interclavicle, radius (9 mm), proximal ulna (~9 mm), incomplete ilium (10 mm), pubis?, ischium?, femur (23 mm), proximal fibula, three phalanges (2.5, 4 mm), fragments, coelacanth fragments?
Paratype- ?(ANSP coll.) femur (36.5 mm), tibia (30 mm)

Holotype of Gwyneddosaurus erici after preparation damage with elements numbered as in Bock (1945), and taken from that paper.

Comments- Gwyneddosaurus erici was described by Bock (1945) based on a disarticulated specimen, in which he labeled the bones he identified with numbers 1-51. Bock provisionally placed the reptile in Theropoda and furthermore in Podokesauridae within Coelurosauria. This was based on the supposedly short forelimbs and gracile (perhaps hollow) postcrania, with the Triassic age probably responsible for the familial identification. The procoelous vertebrae, distally expanded presacral neural spines, and differently shaped pectoral girdle were all seen as different from Podokesaurus though. Bock also referred to "several leg bones, similar to those of our specimen" from the same locality though he did not describe these.

Gwyneddosaurus holotype elements after Huene (1948), all at same scale.  The radius and maybe ulna may be metatarsals instead.

Huene (1948) examined photographs of the material from before it was prepared by Bock, leading him to reidentify almost every element. The supposed pterygoid (16), and two other cranial elements were reidentified as a possible squamosal, basisphenoid or parasphenoid (15), and cervical vertebra (13) respectively. The supposed dorsal vertebrae (3 and 4) were reidentified as posterior cervicals, the cervical ribs (39) as chevrons, and one of the dorsal ribs (45) as an ulna. The scapula (18) and coracoid (17) were each actually a complete scapulocoracoid, and the bone Bock identified as a fused radius and ulna (22) was not mentioned but is far too short to be such. One of the ilia is actualy a probable ischium (2), and a supposed tarsal (25, perhaps meant as metatarsal) is a radius. Huene also described and figured a femur and tibia larger than the holotype which is seemingly the additional specimen noted by Bock. He felt these were possibly referrable to Gwyneddosaurus, as the femur "may be structurally identical." Huene assigned the taxon to Protorosauria based on cervicals similar to Macrocnemus and Microcnemus, holocephalous dorsal ribs, elongate and only slightly curved gastralia as in Macrocnemus and Tanystropheus, extremely short scapula with one broad and one rounded corner as in Adelosaurus, clavicle similar to Macrocnemus, ilium with small preacetabular and large postacetabular process as in Protorosaurus, and distally narrowing paratype femur as in Macrocnemus. This was followed by Camp et al. (1953) and Romer (1966), who placed it in Protorosauridae and ?Prolacertidae respectively. However, the cervical and clavicle similarities were never specified, the ilial proportions are plesiomorphic for amniotes, Adelosaurus has been more recently recognized as being of Youngina-grade, holocephalous dorsals are found in numerous taxa, gastralial morphology has not been shown to be diagnostic for tanystropheids, and the holotype femur seems not to distally narrow. Huene later (1956) placed Gwyneddosaurus in Askeptosauridae, but Askeptosaurus has shorter amphicoelous presacrals, a shorter scapula, rod-like interclavicle, no preactabular process, a longer postacetabular process, and much more robust limb elements with poorly developed features. Perhaps it was a typo.

Steel (1970) followed Bock in placing Gwyneddosaurus in Theropoda, and Welles (1984) placed it in Theropoda incertae sedis (though stating "it certainly is not even remotely related to Dilophosaurus"). While Huene never explicitly stated reasons to exclude the taxon from Theropoda, they include- procoelous cervical centra, no cervical pleurocoels, holocephalous dorsal ribs, short scapular blade, unfused clavicles, ossified interclavicle, short preacetabular process, femoral head not inturned, internal trochanter and intertrochanteric fossa present. Furthermore, the supposedly short forelimb elements were incorrectly identified, and numerous reptiles have gracile postcrania which can appear hollowed.

Olsen first (1979) stated that Gwyneddosaurus had an emargnate scapulocoracoid (though this is untrue), so could be a lizard, but was not closely related to his new tanystropheid Tanytrachelos. He did state it was nearly identical to some elements of Rhabdopelix though, and would be redescribed in the future. Olsen later (1980) thought it and Rhabdopelix could be senior synonyms of Tanytrachelos, as some elements could not be distinguished from it. As other elements differ between the taxa, Olsen thought Gwyneddosaurus should be a nomen dubium. Yet chimaerical status does not make a taxon a nomen dubium. Olsen and Baird (1986) considered Gwyneddosaurus to be a chimaera of coelacanth and possible Tanytrachelos bones. Similarly, Olsen and Flynn (1989) thought it to be a chimaera of Tanytrachelos and possible coelacanth fragments. Specifically, they stated "the large, quadranglular and keeled interclavicle characteristic of Tanytrachelos is present as are procoelous vertebrae." Instead of sinking Tanytrachelos into Gwyneddosaurus, Olsen and Flynn incorrectly consider the latter a nomen dubium because "there is significant room for doubt- certainly at the species level it is indeterminate." This statement is meaningless though as Tanytrachelos only has one recognized species, and the historical recognization of genus- and species-level characters for monospecific taxa is outdated. Indeed, when "at least part of Gwyneddosaurus more assuredly is Tanytrachelos", the former cannot be a nomen dubium. An ICZN petition would be necessary to sink Gwyneddosaurus, and since this has not been made yet it is more proper to make Gwyneddosaurus a senior synonym of Tanytrachelos. The identity of Rhabdopelix deserves further scrutiny, though it does include procoelous vertebrae. Whether the referred hindlimb is correctly referred is uncertain, as there are no proposed hindlimb apomorphies of Gwyneddosaurus, but is does seem to be tanystropheid based on the distal narrowing and is similar in size to large Tanytrachelos specimens. Spamer et al. (1995) placed Gwyneddosaurus in Tanystropheidae.

Unfortunately, comparison between the two taxa is difficult, as the pre-preparation photos of Gwyneddosaurus remain unpublished (though they are archived at the ANSP), while Tanytrachelos has only received a preliminary description. Bone 2 is not mentioned by either Bock or Huene but may be another posterior cervical. I think Huene switched bones 5 and 6 in his description, with 5 being the axis and 6 being a neural arch in dorsal view. Element 14 was believed to be an atlas or cranial bone by Bock, but is certainly too large for the former. Olsen and Baird's description of the interclavicle and comparison with Tanytrachelos suggests it is element 15, identified as a cranial bone by Bock and a para- or basisphenoid by Huene. Element 16 was identified as a possible squamosal by Huene, but does not match that element in Tanytrachelos and may be from the braincase or palate instead. Bock's supposed fused radius and ulna (22), may be an anterior dorsal rib if element 22 refers to the elongate curved bone and not the posterior cervical(?) the distal end lies over. Elements 23 and 24, identified as ischia by Bock, seem to be posterior dorsal ribs. Based on the identification as congeneric with Tanytrachelos, the two metatarsals or phalanges identified by Huene (31 and another unlabeled) are definitely the latter. The radius (25) and perhaps the ulna (45) as identified by Huene may actually be metatarsals then. A supposed (meta?)tarsal identified by Bock (27) seems more likely to be a posterior dorsal rib. Interestingly, curved triangular element 29 (called a rib by Bock), may be a heterotopic bone as found in some Tanytrachelos individuals. All of the "foot bones (32-38) which probably are toes" are anterior cervical ribs or gastralia as is at least one of the supposed manus bones (41). The other proposed manual elements (40 and 42) are very small and could be distal gastralia or fish bones. Notably, despite Olsen's statements Gwyneddosaurus contains coelacanth material (presumably of the common and contemporaneous Osteopleurus newarki), which material this refers to has never been specified nor has the referral been justified.

References- Bock, 1945. A new small reptile from the Triassic of Pennsylvania. Notulae Naturae of the Academy of Natural Sciences of Philadelphia. 154, 1-8.

Huene, 1948. Notes on Gwyneddosaurus. American Journal of Science. 246, 208-213.

Camp, Welles and Green, 1953. Bibliography of Fossil Vertebrates 1934-1938. The Geological Society of America, GSA Special Paper 42.

Huene, 1956. Palaeontologie und Phylogenie der Niederen Tetrapoden. Jena. 716 pp.

Romer, 1966. Vertebrate Paleontology. Chicago University Press. 468 pp.

Steel, 1970. Part 14. Saurischia. Handbuch der Paläoherpetologie. Gustav Fischer Verlag, Stuttgart. 1-87.

Olsen, 1979. New aquatic eosuchian from the Newark Supergroup (Late Triassic-Early Jurassic) of North Carolina and Virginia. Postilla. 176, 14 pp.

Olsen, 1980. Comparison of the vertebrate assemblages from the Newark and Hartford basins (Early Mesozoic, Newark Supergroup) of eastern North America. in Jacobs (ed.). Aspects of Vertebrate History. Flagstaff, Museum of Northern Arizona Press. 35-53.

Welles, 1984. Dilophosaurus wetherilli (Dinosauria, Theropoda): Osteology and comparisons. Palaeontographica Abteilung A. 185, 85-180.

Olsen and Baird, 1986. The ichnogenus Atreipus and its significance for Triassic biostratigraphy. in Padian (ed.). In the Beginning of the Age of Dinosaurs: Faunal Change Across the Triassic-Jurassic Boundary. Cambridge: Cambridge University Press. 61-87.

Olsen and Flynn, 1989. Field guide to the vertebrate paleontology of Late Triassic rocks in the southwestern Newark Basin (Newark Supergroup, New Jersey and Pennsylvania). The Mosasaur. 4, 1-35.

Spamer, Daeschler and Vostreys-Shapiro, 1995. A study of fossil vertebrate types in the Academy of Natural Sciences of Philadelphia: Taxonomic, systematic and historical perspectives. The Academy of Natural Sciences of Philadelphia. Special Publication 16, 435 pp.