Distribution of the dentary groove of theropod dinosaurs: Implications for theropod phylogeny and the validity of the genus Nanotyrannus Bakker et al., 1988 Author Joshua D. Schmerge Biodiversity Institute, University of Kansas, 1345 Jayhawk Blvd., Lawrence, KS, 66045, USA schmerge@ku.edu Author Bruce M. Rothschild Department of Vertebrate Paleontology, Carnegie Museum, 4400 Forbes Ave., Pittsburgh, PA, 44272, USA text Cretaceous Research 2016 2015-12-27 61 26 33 journal article 10.1016/j.cretres.2015.12.016 01816227-cd8c-4f5f-904f-e78a7c90c64c 3890523 A frequent point of contention in the literature for using skeletal characters to differentiate Nanotyrannus and Tyrannosaurus is the possibility of ontogenetic variation, the chance that characters become altered as an individual matures into adulthood. While the size, orientation, and depth of cranial fossa and such openings as the orbits in tyrannosaurs, as well as tooth count and morphology, have been interpreted to be ontogenetically variable ( Carr, 1999 ), these interpretations have been questioned ( Currie, 2003a ; Larson, 2013a ). The argument that the dentary groove is an ontogenetically variable character is unappealing for several reasons. First, if this feature corresponds to a system of nerves enervating the mandible, a dramatic change (e.g., metamorphosis) would need to be invoked to explain the loss of this feature through maturation. Second, there are other undisputed T. rex specimens, representing juvenile ( LACM 28471 , LACM 23845 ) and subadult ( LACM 150167 , KUVP 155809 ) stages, that lack this groove. The adult form (e.g., FMNH PR2081 ) lacks this groove. As all Tyrannosaurus rex specimens we investigated lack a dentary groove, even in juvenile and subadult individuals, and all Nanotyrannus specimens we investigated possess a dentary groove, this seems to suggest that the groove is not variable in a single species and that it does not change during ontogeny. The presence or absence of a groove may further be sufficient to diagnose one taxon from another taxon. The most reasonable explanation for this data is that N. lancenesis is a distinct taxon and not a juvenile form of T. rex . The dentary groove is such a phylogenetically conservative character (i.e., only five losses outside of Tyrannoraptora) that the presence or absence of the groove in tyrannosaurids seems a useful taxonomic character. Nanotyrannus has been suggested to be similar to Gorgosaurus , based on the presence of numerous cranial and dental characteristics ( Larson, 2013a ). Interestingly, CMNH 7531 was originally described as a species of Gorgosaurus ( Gilmore, 1946 ) . The presence of the dentary groove in Nanotyrannus seems to further confirm its affinity with Albertosaurinae rather than Tyrannosaurinae . Albertosaurinae is defined as the tyrannosaurs possessing: an antorbital cavity that reaches the nasomaxillary suture, lateral surface of nasal excluded from the antorbital vacity, and a dorsally-oriented, triangular corneal process of the lacrimal ( Holtz, 2004 ). Currie (2003b) defined Albertosaurinae as possessing, in contrast to Tyrannosaurinae , short and low skulls, shorter ilia, longer tibiae, longer metatarsals and longer toes, a definition Nanotyrannus meets. Larson (2013a) presented more than 30 skeletal characters that separate Nanotyrannus from Tyrannosaurus , with the following characters possibly uniting Nanotyrannus and the albertosaurines: greater dentary tooth counts, contact of the maxillary fenestra with the rostral margin of the antorbital fossa, a medial post-orbital fossa and an ectopterygoid pneumatic foramina bounded by a thick lip. Albertosaurines may also be regarded as having more circular than ovoid orbits. We propose that the presence of the dentary groove be added to the list of diagnostic albertosaurine characters, as it is a character that is lost in Tyrannosaurinae . We therefore propose that Nanotyrannus should be regarded as a member of Albertosaurinae . Fig. 3 shows three potential interpretations of the phylogeny of Tyrannosauroidea. Nanotyrannus has been previously suggested to be either (1) a valid taxon distinct from Tyrannosaurus ( Fig. 3A ; Larson, 2008 ; Larson, 2013a ), (2) the sister taxon to Tyrannosaurus , either as Tyrannosaurus lancensis ( Currie et al., 2005 ) or as Nanotyrannus ( Fig. 3B ; Currie et al., 2003 ), or (3) even a juvenile Tyrannosaurus ( Fig. 3C ; Brusatte et al., 2010 ). The most parsimonious distribution of the dentary groove found in this study ( Fig. 3A ) places Nanotyrannus as sister to the Albertosaurinae ( Gorgosaurus + Albertosaurus ). The Nanotyrannus + Tyrannosaurus hypothesis ( Fig. 3B ) requires five additional independent losses of the dentary groove, more than the tree which places Nanotyrannus closer to Albertosauriane. The hypothesis that Nanotyrannus is a juvenile Tyrannosaurus ( Fig. 3C ) requires four additional independent losses, as well as requiring an ontogenetic explanation for why juvenile T. rex possesses the character absent in adults. The fact that there are specimens interpreted as juvenile and subadult T. rex that lack the dentary groove should render that hypothesis completely unsupportable. 5. Conclusions The occurrence of the dentary groove in theropod dinosaurs serves as a useful feature for determining relationships and inferring the validity of the disputed taxon Nanotyrannus . The groove appears early in the theropod record and is a ubiquitous character for the duration of the Jurassic and Early Cretaceous. For the 92 theropod taxa investigated in this study, 48 possess this feature. As 80% of basal theropods possess this groove, it could be considered a defining character of theropod dinosaurs, lost in only a handful of lineages. While loss of the groove did not appear to result from the evolution of beaks in different theropod clades, it may have been one of many evolutionary changes that occurred which strengthened the jaws of tyrannosaurids. This feature has also proved useful for interpreting relationships among tyrannosauroids, which have a mosaic distribution of this character. Nanotyrannus clearly presents the dentary groove, whereas Tyrannosaurus rex lacks this feature. Without a modern analog that demonstrates the loss of such a groove through maturation, accepting that this feature is lost through ontogeny is speculative. “Jane”, and the other specimens referred to Nanotyrannus , therefore would not be examples of juvenile Tyrannosaurus , and stand alone as a distinct genus. Nanotyrannus should further be considered sister to the Albertosaurinae , rather than the Tyrannosaurinae , as they are the only other clade of large tyrannosaurs to possess this groove. This feature, in addition to many other cranial characters, makes an alignment of Nanotyrannus with the albertosaurines preferred.