Thursday, March 15, 2018

Lesser-Known Running Lizards (minor update at the end)

by Jaime Headden, from Wikipedia
In the last post, I briefly mentioned a group of dromaeosaurs called unenlagiines. This is still a rather obscure group, especially to the general public, so I wanted to give these South American dromaeosaurs some much-deserved time in the spotlight.

This is a small group—only three (but possibly four or five) genera have been named. They are atypical dromaeosaurs for a number of reasons, foremost among them the elongate, narrow snout packed with minute teeth which lack serrations. While most appear to have been small, one of them was one of the largest dromaeosaurs, approaching Achillobater and Utahraptor in terms of overall size.

Some pieces of Unenlagia comahuensis
As you might imagine from the group’s name, the first unenlagiine to be described was Unenlagia comahuensis (Novas & Puerta, 1997). Its remains are quite fragmentary—the pelvis, left scapula, right leg (missing the foot) and some vertebrae. Despite this, the authors marveled at how remarkably avian the scapula was with its laterally-facing glenoid. They suggest that Unenlagia is representative of a “proavian” that falls somewhere between Dromaeosauridae and Archaeopteryx. In 2004, a second species was described, Unenlagia paynemili, by Calvo, Porfiri& Kellner. Its remains are weirdly similar to U. comahuensis: elements of the pelvis, a humerus, a dorsal vertebra, and a toe (with claw). Their phylogenetic analysis recovered Unenlagia as either in a polytomy with other dromaeosaurs or as an outgroup of traditional Dromaeosauridae.

Some pieces of Neuquenraptor
Another Unenlagia-like theropod was described in 2005 by Novas & Pol: Neuquenraptor argentines. Sadly, it is even scrappier than Unenlagia, although an almost complete foot (including toes and claws) was found. Otherwise it consists of the proximal half of the left radius, right femur, distal tibia, and fragments of vertebrae and dorsal ribs. Novas & Pol find Neuquenraptor and Unenlagia in a polytomy with microraptorines and traditional Dromaeosauridae.

If only a better-preserved unenlagiiine could be found, right?

Well, we didn’t have to wait too long. Makovicky et al. described Buitreraptor gonzalezorum that same year. Buitreraptor finally gives us a good idea of what these South American dromaeosaurs looked like, and they are strange. Buitreraptor’s skull is long and narrow, with a multitude of tiny teeth which lack serrations, bringing to mind spinosaurs. Interestingly, the humerus is as long—proportionately—as basal birds and “some dromaeosaurids from Liaoning” (keep in mind that microraptorans were still new in 2005). The killing claw, so characteristic of dromaeosaurids, is small and offset medially, matching Neuquenraptor, primitive troodontids, and Microraptor. This was not an animal that was going after big game.

The skull of Buitreraptor, in right lateral view
Makovicky et al. formally defined the Unenlagiinae and suggested that Unenlagia and Neuquenraptor may be synonymous, but the lack of overlapping material makes the situation uncertain. They reference a “large unnamed deinonychosaurian from the uppermost Cretaceous of Argentina” which has similar teeth to Buitreraptor, which would make the Unenlagiinae quite a long-lived group (Buitreraptor is from the mid-Cenomanian). But hey, that giant deinonychosaurian sounds interesting, doesn’t it? What could that possibly be?

Some pieces of Austroraptor
Turns out it was Austroraptor cabazai, described by Novas et al. in 2009. While fragmentary, Austroraptor presents a dramatic example of how diverse the Unenlagiinae must have been. The authors estimate that this predator was five meters long. While incomplete, the skull is generally similar to that of Buitreraptor in being long and narrow, with a multitude of small, unserrated teeth. Most of the left hindlimb is preserved. The most surprising thing about Austroraptor apart from its size is its measly humerus. In stark contrast to Buitreraptor, it is quite short—less than half the length of the femur. Like Makovicky et al., the authors find a monophyletic Unenlagiinae that sits in a polytomy with microraptorines and traditional Dromaeosauridae.

A second specimen of Austroraptor was described in 2012 by Currie & Carabajal. Although even more fragmentary than the holotype, this second specimen preserves different bones, allowing a better overall picture of Austroraptor; for example, a complete right ulna and left radius and many foot and toe elements, including claws. Austroraptor still had a very short forelimb, but the proportions of the humerus to the forearm are more typical for dromaeosaurs. Austroraptor would have lived alongside derived abelisaurs, which is a neat image.

Andrea Cau has suggested that Austroraptor is not related to Buitreraptor but is, instead, a giant troodontid which I’d also be okay with.

Brisson Egli et al. (2017) tried to figure out whether Unenlagia and Neuquenraptor are synonymous or not by providing a detailed osteology of the latter. Sadly, the jury is still out. Of the few overlapping elements, the distal ends of their tibiae seem to differ in degree of gracility, and the proportions of the shared pedal phalanges are of slightly different proportions between the two.

The foot of Pamparaptor
Another potential unenlagiine, Pamparaptor, was described in 2011 by Porfiri, Calvo & dosSantos. Originally referred to Neuquenraptor, the holotype comprises only of a partial foot. The foot shows dromaeosaurid and troodontids characteristics, and the authors reject both the idea that Pamparaptor might be synonymous with Neuquenraptor and that Neuquenraptor and Unenlagia are synonymous. For reasons I’m not completely sure of, they conclude that this invalidates Turner et al’s (2012) Unenlagiinae and that South America’s dromaeosaurs are still in a polytomy.

Finally, you can’t spell “Unenlagiinae” with “Rahonavis.” This tiny Malagasy paravian, described in 1998 by Forster et al., has been kicked around the Paravian family tree ever since. Its partial skeleton presents a maddeningly ambiguous dinobird. Of the arm, only the ulna and radius were found, but they are proportionately quite long and the ulna bears quill knobs. The pelvis resembles that of both Archaeopteryx and Unenlagia, and the sacrum is composed of six co-ossified vertebrae (one more than Archaeopteryx). The femur resembles that of Archaeopteryx and more derived birds in that the fibula does not contact the calcaneum. A complete, articulated foot, missing only the claws of Digits III & IV, show a reversed hallux and a (proportionately) large sickle claw on Digit II, which was hyperextended when discovered.

Some pieces of Rahonavis
Forster et al. ran a phylogenetic  analysis and Rahonavis wound up in a monophyletic clade with Archaeopteryx and Unenlagia. However, the clade is not robustly supported. I should note here that, in 1998, Buitreraptor was unknown and alvarezsaurids were thought to be basal flightless birds.

Concerns have been raised that the Rahonavis holotype might represent a chimaera, as the arm bones (scapula, ulna & radius) are not in articulation with the rest of the specimen. Forester et al. recognized this possibility but were confident that all the bones belong to a single individual. Andrea Cau ran the arm bones alone through his matrix separately from the rest of the skeleton, and they wind up in completely different places: “Rahonavis Wing” is right next to enantiornithines, whereas the rest of it is next to Mahakala (which we now know is a halszkaraptorine).

Turner et al. (2012) found Rahonavis to be a member of the Unenlagiinae and a sister taxon to Unenlagia. Turner et al. (2007) got the same result in their description of Mahakala. As we saw in the anchiornithid post, though, more current phylogenies have moved Rahonavis back to a position among basal birds. I suspect that until we have more complete remains from an unambiguously single animal, its position will continue to jump around—especially if Andrea is right and Rahonavis is a chimaera. And so, that familiar refrain: we need more specimens to clarify what, exactly, Rahonavis is.

In 2011, Agnolin & Novas published an interesting phylogenetic analysis of Unenlagiidae in which they find the group as outside of Dromaeosauridae and, in fact, Deinonychosauria. Instead, Unenalagiines are the basalmost group of birds, forming an outgroup to Archaeopteryx + everyone else. Their analysis includes Rahonavis as an unenlagiine but they acknowledge that more detailed analyses may deposit it elsewhere in the phylogeny. Of course, this unique topology is in contrast to the standard tree, as might be exemplified by Turner, et al. (2012), in which Unenlagiinae is a basal group of dromaesaurs. Agnolin & Novas followed up their 2011 analysis with a much larger analysis in 2013. They again find unenlagiines to be essentially the outgroup of Avialae. They do remove Rahonavis from Unenlagiinae, though, and deposit it among basal birds, in a more derived position than Archaeopteryx.

More recently, Buitreraptor has received the monograph treatment in three parts: the skull, the tail, and the body. Turns out it has ridiculously long, spindly fingers. In general, Buitreraptor is similar to dromaeosaurids, troodontids, Archaeopteryx, and even Anchiornis in several features. This is not terribly surprising considering that unenlagiines are close to the origins of paravians generally. Still, it would be interesting to know how--if at all--this full description of Buitreraptor shakes up Agnolin & Novas' suggestion that unenlagiines are not dromaeosaurs.

From Novas et al. (2018): a very strange dromaeosaur, indeed.
I'm sure there's plenty more unenlagiine material left to be found in South America, and I'll be interested to see how this group fills out in the future.

I'm pushing this post out the door even though I'm not 100% happy with it because I really need to post something. Maybe placodonts next month! Cross your fingers.

UPDATE: A second paper on the postcranial osteology of Buitreraptor, based on both the holotype and referred specimens, was recently published by Gianichini et al. in PeerJ. Unlike Agnolin & Novas, above, they find that unenlagiines are perfectly good dromaeosaurs. Another interesting tidbit is that their phylogeny recovers Mahakala as an outgroup to the Unenlagiinae, which supports halszkaraptorines as basal dromaeosaurids, possibly allied closely with unenlagiines.


  1. Despite being 'pushed out the door', I thought this was a very informative article.

    I wasn't aware that Andrea had tested the chimeric nature of Rahonavis ('wing' versus 'the rest'). This raises the possibility (and I know I'm not the first to suggest this) that the 'Rahonavis wing' belongs to Vorona (which comes up in some analyses as an enantiornithine). The Rahonavis wing bones are actually quite advanced, and consistent with powered flight ability.

    I'm skeptical of the interpretation that Rahonavis had a reversed hallux. I know this appeared in the original description by Forster &c, but this seems to be based on the preserved articulation of the bones - which is very weak evidence. The first metatarsal lacks any anatomical characters one might associate with a reversed hallux.

    Finally, apparently Unquillosaurus has been regarded as a possible unenlagiine, but I can't pin down the source. Given the meager material for Unquillosaurus, any such interpretation is probably quite precarious.

  2. The "Austroraptor troodontid" post is 10 years old (...we are all getting older), and such result has rarely resulted again in my (now much larger) dataset.

    Gondwanan theropods are so "unusual" for our Laurasian-biased perspective, that Rahonavis may be a legit taxon and not a chimera. As I have learned from Halszkaraptor, just because something seems an unexpected mix of distinct clades does mean it is a chimera. ;-)

    1. Holy cow it IS ten years old. I didn't even notice that! O_O