Two issues are raised in this comment by Herne and colleagues. I have tried to address them below by providing my perspective, and a summary of what has already been said on these matters (see Benson et al. 2010a,b; Henre et al. 2010).

(1) Our research article suggests their previous work assigned NMV P186046 to Allosauroidea. But in fact, they assigned the specimen to an indeterminate neotetanuran (Herne et al. 2010).

I, and not my co-authors, am responsible for any mistakes in our research article. When we wrote the paper, I thought that Herne et al. (2010) were suggesting allosauroid affinities as an alternative to tyrannosauroid affinities, thus broadening the scope of possible identifications, and resulting in formal assignment to Neotetanurae (=Avetheropoda): the wider clade including both Allosauroidea and Tyrannosauroidea (and also other groups such as Ornithomimosauria and Dromaeosauridae).

I have explained this fully in a footnote below [1]. It is not my intention to refute the validity of the comment Herne et al., but to explain my interpretation of what they wrote in their original paper (Herne et al. 2010).

Thank you to Herne and colleagues for clarifying their intended position. Fortunately, this point does not affect anatomical support for our interpretation of NMV P186046, or the wider conclusions of our work on Cretaceous biogeography and Australian theropod diversity.

(2) Herne et al. claim to have refuted our identification of the specimen as a tyrannosauroid pubis. They also raise a new point that it has been suggested elsewhere to be an ornithomimosaur.

I have summarised the two points of refutation previously raised by Herne et al. (2010) below [1,2]. We have already addressed these points using quantitative measurements and clear photographic images (Benson et al. 2010a), as stated in our research article, and reiterated below for clarity [1,2].

Anatomical observations are the sole basis for inferring the affinities of fossil specimens. I feel that our observations represent objective, primary data that we have illustrated clearly and defended adequately in peer-reviewed work. Although we welcome new interpretations, our comprehensive literature review (including, but not limited to citations 83-189 in the article) and direct examination of specimens worldwide has not revealed any theropod clade other than Tyrannosauroidea, that possesses all the diagnostic features described in NMV P186046 (Benson et al. 2010a,b; and the recent article).

The suggestion of ornithomimosaur affinities is based on a statement made by Currie et al. (1996). Notably, this statement was intended only as a preliminary identification, and was not supported by evidence such as images or comparative data. The case of has not been fully explained or justified in the peer-reviewed literature. Importantly though, my co-authors and I have already stated ways in which NMV P186046 differs from all currently-known ornithomimosaurs (Benson et al. 2010b). Further specimen observations in North America and Asia, made since 2010, are consistent with our original statements, and have not supported ornithomimosaurian affinities.

Roger Benson, Department of Earth Sciences, University of Cambridge

References

Benson, R.B.J., Barrett, P.M., Rich, T.H., Vickers-Rich, P., Pickering, D., Holland, T. (2010a). Response to comment on “A southern tyrant reptile”. Science 329: 1013-d.

Benson, R.B.J., Barrett, P.M., Rich, T.H., Vickers-Rich, P. (2010b). A southern tyrant reptile. Science 327: 1613.

Currie, P.J., Vickers-Rich, P., Rich, T.H. (1996). Possible oviraptorosaur (Theropoda, Dinosauria) specimens from the Early Cretaceous Otway Group of Dinosaur Cove, Australia. Alcheringa 20: 73–79.

Herne, M.C., Nair, J.P., Salisbury, S.W. (2010). Comment on "A southern tyrant reptile". Science 329: 1013-c.

Footnotes

Herne et al. (2010) provided a discussion of NMV P186046, reiterating our comparative observations (Benson et al. 2010b), and supporting our conclusion that two aspects of the specimen might indicate tyrannosauroid affinities. However, they felt neither aspect was reliable. These are discussed in the footnotes below, which primarily convey information presented in peer-reviewed publications (Benson et al. 2010a,b; Herne et al. 2010).

[1] The distal expansion of the pubis (=pubic boot) of NMV P186046 is uniquely similar to tyrannosauroids. However, Herne et al (2010) correctly restated our observation that when the morphology of the boot is atomized into discrete observations, several morphological features are shared with “non-coelurosaurian neotetanurans” (i.e. with allosauroids, which by definition are the only non-coelurosaurian neotetanurans), and occasional features are shared with other coelurosaurian clades.

The pubic boot of NMV P186046 possesses a combination of: proportionally great anteroposterior length (shared with tyrannosauroids, allosauroids, and compsognathids); expansion is developed prominently both anteriorly and posteriorly (shared with tyrannosauroids, allosauroids, ornithomimosaurs and some other coelurosaurs); transversely narrow morphology (a unique synapomorphy of coelurosaurs; i.e. it is absent in allosauroids but present in tyrannosauroids). This last observation was questioned by Herne et al. (2010), but quantitatively confirmed by Benson et al. (2010a), and also in our research article.

Citing Herne’s summary of this directly: “…the broader distribution among neotetanurans of the other pubic boot traits identified in NMV P186046 indicates that a more inclusive neotetanuran placement would be more parsimonious.”

In other words, they say that although NMV P186046 shares many features with tyrannosauroids, the presence of some of these features in other clades, but especially in allosauroids, raises the possibility of a different identification. Although they invoked parsimony to support this hypothesis, parsimony actually supports an assignment to Tyrannosauroidea, the only clade in which all three pubic boot characters are present simultaneously.

This led to my impression that they were suggesting NMV P186046 might be an allosauroid.

[2] Herne et al. reiterated our observation that the ‘flange-like’, anterolaterally curving pubic peduncle we described in NMV P186046 is not known outside of Tyrannosauroidea and Dromaeosauridae (Benson et al. 2010b). The currently accepted framework of theropod phylogeny thus indicates the status of this feature as a synapomorphy of a node within Tyrannosauroidea (Benson et al. 2010b: also derived independently in some dromaeosaurids). However, because this structure is incompletely preserved in NMV P186046, they stated that our description was ‘speculative’.

We felt that the morphology of the more dorsal, broken portions of the tubercle was consistent with our description, though we admitted that this was based on incomplete data (Benson et al. 2010b). However, we later provided an image showing the well-preserved ventral portion of the tubercle, which is anterolaterally everted and ‘flange-like’ (Benson et al. 2010a), confirming our original description. This data is also visible in the published images of Herne et al. (2010).