Thursday, September 9, 2010

Concavenator - feathers, Becklespinax and how not to delete taxa

So Concavenator is all the rage, and having just received the paper from Holtz (thanks!), I figured I'd share a few opinions.

First, the supposed feather quill knobs on the ulna.  I have no problem with feathered carnosaurs, or any sort of feathered dinosaur really.  But the knobs are in a line that while described as "posterolateral", is really more anterolateral based on the photo.  They extend from the distal tip of the brachial fossa area to the articular facet for the radius, which aren't areas particularly close to the skin.  I think its more likely an intermuscular line.

Right forearm of Concavenator corcovatus holotype MCCM-LH 6666 in lateral view.  Arrows point to the supposed quill knobs (after Ortega et al., 2010).

Second, I'm very surprised no one has yet mentioned Becklespinax in relation to Concavenator.  Both are Barremian carnosaurs from Europe.  More importantly, Becklespinax is known from three dorsal vertebrae, two with high spines in contact dorsally and the most anterior one with a naturally short spine.  That's diagnostic and just like Concavenator.  Seems like good evidence for a close relationship, though I would have to examine them in more detail to have an opinion on their synonymy.

Finally, a plea to perform phylogenetic analyses better.  The authors deleted the fragmentary taxa from the analysis before running it, but this is a poor way to do things.  If you must delete taxa to improve resolution (and studies have shown fragmentary taxa don't necessarily cause more polytomies as long as they have unique character combinations), do it AFTER the analysis is run.  That way, the taxa will still affect the topology.  As it is, the phylogeny in the paper is misleading because it places Concavenator as a basal carcharodontosaurid (in their usage) more derived than Eocarcharia, but this is only one of three different placements if Eocarcharia is included.  And who knows what the tree would look like if Xuanhanosaurus, Poekilopleuron, Piveteausaurus, Streptospondylus, Chilantaisaurus and Orkoraptor weren't deleted.

Ortega, Escaso and Sanz, 2010. A bizarre, humped Carcharodontosauria (Theropoda) from the Lower Cretaceous of Spain. Nature. 467, 203-206.


  1. The Becklespinax-Concavenator relationships is the theme of a post I'm preparing... (after entering the new taxon in my analysis to see if they form a clade).

  2. As you probably know by now Darren Naish is all over this.