I spent some time a little bit ago rehashing an old premise of mine, that of the growing “primitive” tyrannosauroids and the grades they make. Of these, there are “true” tyrannosauroids (tyrants, roughly corresponding to Tyrannosauridae), “middle” tyrannosauroids (despots, mid-sized to small Tyrannosauroidea), and the precursor forms which seem to stem from very small, gracile, and compsognathid-like coelurosaurs (peons, which may or may not fall into Tyrannosauroidea). It should be noted that under the currently accepted forms for the usage of the term, Tyrannosauroidea contains all theropods that are closer to Tyrannosaurus rex than to, say, a bird, a dromaeosaurid, a troodontid, an allosaur, etc. The explicit definition is flexible, but it basically means if it’s on the lineage toward a tyrannosaurid like T. rex, then it IS a tyrannosauroid.
Enter an oddball.
Earlier this year, another peculiar fossil from Australia [Don’t ALL peculiar fossils come from Australia? I’m almost certain that they do….] was described: NMV P186046 . This specimen adds to a list of Australian taxa that seem to fit into odd places in the dinosaurian trees, or in some cases, all around Diapsida (with reference to a possible Cretaceous dicynodont jaw fragment ).
One of the first things to note about the paper by Benson, Barrett, Rich & Vickers-Rich, is that the pubis comes from the Early Cretaceous of Australia, during a time when each of the continents was well-separated from one another, although there might still have been intromittent contact between some of them, and Australia would have been conjoined still with Antarctica, yet isoltated from the remainder of Gondwana. During this time period, various despots abounded, most of them in Asia and Europe, and none of them south of the Equator. But, based on various features (a very narrow pubic boot, which is fairly elongated craniocaudally, and a “flangelike” form of the pubic tubercle, a process that is linked to the origin for the m. ambiens) the authors supposed the specimen derived from a tyrannosauroid; but not just any tyrannosauroid, a form closely allied to tyrannosaurid taxa. They argue, therefore, it was a type of Australian tyrant.
Jump to present day.
Herne, Nair and Salisbury have recently published a review of the paper , arguing against the conclusions. They do this based on various concerns over the level of preservation of the specimen, including the fact that much breakage has occurred in the proximal end of the pubis, and the distal end (at the boot) is eroded to a degree that the true proportions of the boot are obscured. The authors fail to find a flange for the ambiens process (or pubic tubercle), much less a “moundlike” shape as would be inferred in other coelurosaurs, due only to the fact that the material is too incomplete here to demonstrate this feature. In addition, the authors argue, the pubic boot is incomplete fore, aft, and ventrally, as well as laterally, preventing an accurate account of the true proportions of the boot to be determined, which can alter the statistical variables used by Benson et al.  in demnonstrating tyrannosaurid-like anatomy.
As is common in commentary submitted to most journals, a reply is present. Benson, Barrett, Rich, Vickers-Rich, Pickering & Holland  (after tacking on two additional authors) respond by noting that certain features of the pubis can only lead to a tyrannosauroid identity. This includes a patch of rugose-textured bone around a depression caudal to the raised ambiens process, which Herne et al.  could not find.
The two commentaries spar at trying to find a reasonable answer for the material that is NMV P186046. On the one hand, some physical features (such as the relative proportions of the boot) imply a close tie to several coelurosaurian clades, among them ornithomimosaurs, some oviraptorosaurs, and tyrannosauroids; on the other hand, some of the proximal morphologies appear to be questionable due to breakage and distortion of the material, while a long trough is present on the lateral surface, identified as a “depression” [4:fig.1A] but originally this same region was noted as “rugose” [1:fig.1B] without allusion to a depression save in the text; no tyrannosauroid has an elongated trough posterior to the anterior margin of the pubic shaft, and this region of the shaft (the anterior rim of the trough) is the region that Benson et al. [1,4] point to as the remnants of a “flangelike” ambiens process.
We can add to this several additional observations, observations I held in reserve for if (and when) the previous replies were provided.
1. Biogeography. Benson et al. [1,4] argue that due to the similarity of this specimen to tyrannosauroids, and specifically to tyrannosaurids (all of which are Campanian or Maastrichtian and restricted to Asia and North America) it should be a tyrannosauroid. The pubis derives from the Eumeralla Formation of Victoria, the same formation that produced the “hypsilophodonts” Leaellynasaura amicagraphica and Atlascopcosaurus loadsi, and the “ornithomimosaur” Timimus hermani. The latter was recently argued to be a dromaeosaurid , which has an older history than currently recognized for ornithomimosaurs, while the “hypsies” are being revised and may represent true relicts of older taxa preserved into the Cretaceous, as is assumed by Australia’s lost of contact with other hypsilophodont or dromaeosaurid-producing regions of Gondwana and Laurasia. The biogeography of Australia is undoubtedly odd, but it gets no less confusing the longer the material remains incomplete enough to prevent adequant phylogenetic reconstruction. Nonetheless, the biogeography casts substantive doubt on the placement of this form within a clade that otherwise appears to be consistently derived within the latest ages of the Late Cretaceous of Laurasia, especially since large-bodied tyrannosauroids (“despots”) are consistently arrayed ion Laurasian sediments much, much younger than NMV P186046.
2. Anatomy. Despite the allusion to the slender form of the pubis, Herne et al.  remain correct on the issue that without adequate reconstruction of the missing material, the features of the pubis are problematic, and do not lend themselves to ready explanations for the morphology of missing material.
The proximal pubis is very, very narrow, unlike tyrannosauroids which have broad contact between ilium and pubis. This raises a slight concern only because the pubic shaft distal remains narrow, until the expansion into a boot. The pubic apron is not fully preserved, but appears to be very thin, narrow, and inset from the anterior surface of the shaft. And finally, unlike nearly all coelurosaurs, the pubis boots lateral margins are parallel; they are almost certainly always slightly triangular in distal view in all other coelurosaurs, and especially in tyrannosauroids (it was, in fact, a feature that was used to connect tyrannosauroids to allosauroids in the classic Carnosauria).
So the morphology of the pubis is unusual. While the description of the pubis as a tyrannosauroid is very nice, it tends to require particularly effective anatomical features, rather than a broad similarity of the boot, and a premise that the morphology of the ambiens process is indicative of tyrannosauroids; this latter feature is important to note, because it is the authors unpublished assertion this is so, without analytical backup of this statement, and this assertion is made in the original paper  as well as in defense  of Herne et al.’s contradiction  of it.
I eagerly await a full description of this specimen, with a full analysis of it compared to various recent and forthcoming phylogenetic analyses. It is almost certainly a new, unique animal, although potentially represented by other material from the Eumeralla, but the material appears to be inadequate (in my opinion) to represent the claim made for it by its descriptive authors [1,4].
 Benson, R. B. J., Barrett, P. M., Rich, T. H. & Vickers-Rich, P. 2010. A southern tyrant reptile. Science 327:1613.
 Thulborn, T. & Turner, S. 2003. The last dicynodont: An Australian Cretaceous relict. Proceedings of the Royal Society of London, B 270:985-993.
 Herne, M. C., Nair, J. P. & Salisbury, S. W. 2010. Comment on “A southern tyrant reptile.” Science 329:1013c.
 Benson, R. B. J., Barrett, P. M., Rich, T. H., Vickers-Rich, P., Pickering, P. & Holland, T. 2010. Response to comment on “A southern tyrant reptile.” Science 329:1013d.
 Agnolin, F. L., Eczurra, M. D., Pais, D. F. & Salisbury, S. W. 2010. A reappraisal of the Cretaceous non-avian dinosaur faunas from Australia and New Zealand: evidence for their Gondwanan affinities. Journal of Systematic Paleontology 8(2):257-300.