Saturday, November 28, 2020

Is Falcatakely a bird?

So this week we got the description of the new Maevarano skull Falcatakely forsterae (O'Connor et al., 2020).  It's a pseudo-toucan, with a long, tall snout formed mostly by the maxilla unlike modern birds.  O'Connor et al. recover it as an enantiornithine using Brusatte et al.'s TWiG analysis and O'Connor's bird analysis.  Our problem is that we only have the anteroventral skull preserved, so no braincase, mandible or postcrania.  And being a Maastrichtian deposit in Africa, we don't have the most detailed coelurosaur record.  Add in the fact beaks are known to evolve fast on islands (as Madagascar was even back then) and we have a potential problem on our hands.


Follow your nose to the exciting topic of African coelurosaur diversity.  Reconstructed holotype skull of Falcatakely forsterae (UA 10015) (after O'Connor et al., 2020).

The Lori analysis places Falcatakely in two potential positions- a therizinosaurid or an omnivoropterygid.  The former doesn't make much sense biostratigraphically, but there is a Maevarano synsacrum of the correct size (FMNH PA 741) that was claimed to share characters with Sapeornis by O'Connor and Forster (2010).  So there's a possibility.  Forcing Falcatakely to be an enantiornithine as in O'Connor et al.'s analyses requires six more steps.  Most of the discordant characters relate to the beak, but the wide laterotemporal fenestra would be odd in an enantiornithine.  While I was writing this, Andrea Cau published a post on this topic and reported that he recovered Falcatakely as a noasaurid, which would be quite the phylogenetic jump, but it only takes three more steps in the Lori matrix, so is more parsimonious than the enantiornithine option.  It falls out as an elaphrosaurine, so could relate to e.g. Afromimus.  The non-beak contradictory characters here include a lack of antorbital fossa lacrimal foramen, long posterior lacrimal process and triradiate palatine, which seem more convincing to me.  Additional evidence against these latter two positions is the absence of small ceratosaur (Masiaksaurus is twice as big) or large enantiornithine (Maevarano elements are much smaller) postcrania.  Andrea reported (translated) "It takes 6 further steps to place it in Coelurosauria, and in that case it is a basal dromaeosaurid: interesting in that regard to note that Rahonavis , known from the same Formation, has also been hypothesized to be a basal dromaeosaurid. Can we rule out that Falcatakely is the (still unknown) skull of Rahonavis? The estimated dimensions of the two animals coincide."  Forcing Falcatakely to be Rahonavis only requires one more step, which is pretty impressive.  In their (amazing) osteology, Forster et al. (2020) refer an isolated dentary "found near the Rahonavis holotype (its precise location was not recorded during excavation)" which does not match Falcatakely's upper jaw, being upcurved and extensively toothed.  But it is similar to other unenlagiines like Buitreraptor and Austroraptor.  So much as we have two synsacrum types at this size, unenlagiine-like Rahonavis and Sapeornis-like, we have two cranial types, unenlagiine-like and Falcatakely which is Sapeornis-like in the combination of reduced maxillary dentition, triradiate palatine, modified/reduced antorbital fossa, anteriorly limited naris and strong postorbital-jugal articulation.

Referred dentary of Rahonavis ostromi (FMNH PA 740) as a transparent CT reconstruction (after Forster et al., 2020).

Thus my best guess is that Falcatakely is a basal avialan belonging to the same taxon as FMNH PA 741.  But this comes with a huge chunk of salt as it is so far removed temporally and geographically from potential comparable sister taxa.  Which is actually a common problem with this part of the tree, as shown by Balaur (= Elopteryx?), Hesperonychus, Imperobator and even Rahonavis itself.  We compare these Late Cretaceous taxa to our far more complete Early Cretaceous Jehol record and say Hesperonychus is sorta like Microraptor, Falcatakely is kind of like Sapeornis and Balaur is Jeholornis-grade, but North America, Africa and Europe had their own avialan fauna for 70 million years before them that we're basically unaware of.  If the only alvarezsauroid we had was Mononykus' holotype, could we place it correctly as a basal maniraptoran?  If the only oviraptorosaur we had was Citipati's skull, would we recover that correctly as the sister taxon of Paraves?  I think that's the position we find ourselves in with Falcatakely, and that future discoveries of African small theropods will lead to new interpretations.


References- O'Connor and Forster, 2010. A Late Cretaceous (Maastrichtian) avifauna from the Maevarano Formation, Madagascar. Journal of Vertebrate Paleontology. 30(4), 1178-1201.

Cau, 2020 online.

Forster, O'Connor, Chiappe and Turner, 2020. The osteology of the Late Cretaceous paravian Rahonavis ostromi from Madagascar. Palaeontologia Electronica. 23(2):a31.

O'Connor, Turner, Groenke, Felice, Rogers, Krause and Rahantarisoa, 2020. Late Cretaceous bird from Madagascar reveals unique development of beaks. Nature. DOI: 10.1038/s41586-020-2945-x