Tyrannosaurus rex Osborn, 1905

Christopher A. Brochu, 2003, Osteology of Tyrannosaurus rex: insights from a nearly complete skeleton and high-resolution computed tomographic analysis of the skull, Journal of Vertebrate Paleontology 22, pp. 1-138 : 3

Tyrannosaurus rex Osborn, 1905

- Tyrannosaurus rex is unquestionably a valid taxon. Osborn’s original work has been supplemented by more thorough descriptions (Osborn, 1906, 1912, 1917; Molnar, 1990), and the taxon is currently known from several partial skeletons. This appears to be the only tyrannosaurid known from the Late Maastrichtian Lance, Hell Creek, and Scollard Formations, and occurrences in the Late Maastrichtian elsewhere in western North America (e.g., the San Juan Basin and western Texas) probably also pertain to T. rex .

Osborn (1905) actually named two large Hell Creek tyrannosaurids, T. rex and Dynamosaurus imperiosus , both based on partial skeletons and skulls. Dynamosaurus was thought to have osteoderms, a feature that distinguished it from Tyrannosaurus . Doubts about association of the reported osteoderms with the type of Dynamosaurus ( AMNH 5866 ) were expressed, and Osborn later regarded Dynamosaurus imperiosus and Tyrannosaurus rex as synonymous (Osborn, 1906). Tyrannosaurus has priority, as it preceded Dynamosaurus in the description.

Manospondylus gigas Cope, 1892 probably represents the same species as Tyrannosaurus rex . The holotype ( AMNH 3982 ) is from a large theropod and comes from the Hell Creek Formation (see figures in Hatcher, 1907 and Osborn, 1917), where only a single large theropod ( T. rex ) is otherwise known. However, AMNH 3982 is a pair of badly-weathered vertebrae, and the taxon is considered to be a nomen vanum (Osborn, 1917). Recent press reports hold that more of this specimen was found at the type locality in South Dakota, and that because it can be positively identified as T. rex , this endangers the validity of Tyrannosaurus rex Osborn, 1905 . These claims are dubious on two grounds: first, one would have to demonstrate that the new material pertains to the same individual animal as the holotype of Manospondylus gigas , and given the poor state of preservation of AMNH 3982 , this is unlikely. The articular surfaces of the type vertebrae are poorly preserved, and it will be very difficult (if not impossible) to articulate a newly-collected fossil with them. Second, according to current taxonomic practice, a name is considered invalid if it has not been used in formal literature for fifty years and a junior synonym has gained widespread use (ICZN, 2000). Manospondylus has not, to my knowledge, been applied since its initial publication in 1892, and no one would seriously argue that the name Tyrannosaurus rex is not in widespread use.

Nearly all informative remains of T. rex have come from the Late Maastrichtian units of the Dakotas, Montana, Wyoming, Alberta, and Saskatchewan (Osborn, 1906, 1917; Molnar, 1990; Ryan and Russell, 2001; McIver, 2002). Much less complete material possibly referable to T. rex is known from the Maastrichtian of Texas ( Lawson, 1976) and New Mexico ( Gillette et al., 1986; Carr and Williamson, 2000). Carpenter (1990) argued that the Texas specimen ( TMM 41436-1 , an isolated left maxilla) lies outside the range of shape variation for T. rex , but the morphology of the promaxillary foramen and the number of teeth are both consistent with T. rex (pers. obs.). If it is a different species, it is a close relative.

Gilmore (1946) described a small skull from the Hell Creek Formation ( CMNH 7541 ) as Gorgosaurus lancensis. Russell (1970) regarded Gorgosaurus as a junior synonym of Albertosaurus , and he thus referred Gorgosaurus lancensis to Albertosaurus . Bakker et al. (1988) later erected the generic name Nanotyrannus for this fossil, arguing that it represented the adult of a small basal tyrannosaurid taxon. Carr (1999) has argued persuasively that the features distinguishing Nanotyrannus from T. rex represent ontogenetic variation, as the surficial bone texture within Nanotyrannus' antorbital fossa is characteristic of immature dinosaur bone. Moreover, most features diagnosing Nanotyrannus are also typical of immature Albertosaurus . I thus follow Carr (1999) and regard Nanotyrannus lancensis as a junior synonym of Tyrannosaurus rex .

One of the features allegedly separating Nanotyrannus from Tyrannosaurus was maxillary tooth morphology. The maxillary and dentary teeth of adult T. rex (and other mature tyrannosaurids) are stout in cross-section and not the slender blades found in most other theropods and, according to some authors, Nanotyrannus . Carr (1999) reports ontogenetic variation in this feature in growth series of Albertosaurus , with immature specimens having flattened teeth much like those of Nanotyrannus . One can reasonably hypothesize that tooth shape varies ontogenetically in tyrannosaurids. This has been recently challenged by the description of isolated teeth within the size range of Nanotyrannus , but with a morphology more like that of an adult T. rex (Larson, 1999) , but we cannot assume a linear relationship between tooth shape and size for tyrannosaurid populations.

In any case, the claim that Nanotyrannus has flatter maxillary teeth than an adult T. rex is not solid. Some of the maxillary teeth in the type of Nanotyrannus are restored in plaster, and most of the flattened maxillary teeth are toward the back of the toothrow (where they are generally flatter in all tyrannosaurids). Comparison of CT images through the skulls of FMNH PR2081 and the type of Nanotyrannus reveal very little difference in cross-section shape in more anterior maxillary teeth, and the maxillary alveoli of Nanotyrannus are circular, not elliptical.

The validity of Aublysodon Leidy, 1868 is highly questionable (Carr and Williamson, 2000). As currently applied, the name refers to small tyrannosaurids in which the premaxillary teeth lack serrations on the lingual carinae (Carpenter, 1982; Currie et al., 1990). Absence of serrations may be the result of wear in many specimens, including the lectotype (personal observation). Moreover, some immature tyrannosaurid maxillae (e.g., FMNH PR2211) bear serrated and unserrated teeth, and so the absence of serrations might also be an ontogenetic factor. A partial skeleton from the Hell Creek Formation has been referred to Aublysodon ( LACM 28471 ; Paul, 1988; Molnar and Carpenter, 1989), but the features distinguishing it from the larger Hell Creek tyrannosaurid are the same as those for Nanotyrannus , and I regard it as an immature specimen of T. rex . Similar suspicions have been voiced about nondental remains referred to Aublysodon from elsewhere in North America ( Lehman and Carpenter, 1990; Carr and Williamson, 2000).