Friday, December 25, 2015

Drepanosaurs: a Primer

Another exciting entry in the "Primer" series (which started with hupehsuchians)! Again, we're tackling a Triassic oddity. There are lots to choose from. Indeed, I sense a theme!

The Late Triassic was an exciting time in the history of life on Earth. Ecosystems were finally in full recovery after the brutal Permio-Triassic Extinction in which life nearly died, and the ancestors of modern groups were becoming established—this is where you find the great-great granddaddies of birds, mammals, and crocodilians. However, even though the world’s food webs and the roles within it were similar to today, the actual composition of those roles was much different. Mother Nature was going through a period of divine inspiration: the predator guild was ruled by large, vicious (distant) relatives of crocodiles—Postosuchus, Carnufex, and Teratosaurus—preyed on their own herbivorous relatives, the armored aetosaurs. These critters looked like a cross between an armadillo and a crocodile, but grubbed around on the ground for plants. The other big role in the herbivore guild went to the dicynodonts, therapsid holdovers from the Permian with tusked, toothless, beaked jaws, some of which grew to be the size of cows.

The Earth’s marine waterways were also going through a revolution—as I described previously—but freshwater environments were also going through big changes. Dangerous crocodile-like phytosaurs ambushed animals coming to drink and/or fed on fish in the manner of modern gharials. With their noses set high atop their skulls, phytosaurs were more strongly water-adapted than modern crocodilians. Other strange reptiles, like Vancleavea, also ventured into freshwater habitats. Man, I could do a whole damn post about Vancleavea, because damn.

And let's not forget that dinosaurs originated in South America and quickly spread around the world. Their closest relatives, the herbivorous silosaurids, were also wandering around. Basal theropods entered the carnivore guild with toothy predators like Herrerasaurus, Daemonosaurus, and Coelophysis. Basal sauropodomorphs diversified quickly, perhaps in the absence of competition, and achieved large size quickly. Ornithischians were slower to diversify, but Pisanosaurus (who might actually be a silesaurid) and Eocursor prove that they were present, if incredibly rare for reasons which remain unclear. Ornithischians don’t diversify until after the Triassic-Jurassic extinction event which freed up many niches for takeover by dinosaurian actors.

Unlike the ground and the water, however, the treetops remained fairly uninhabited. Rhamphorhychoid pterosaurs, of course, were fluttering around in the Late Triassic, but one wonders exactly how arboreal they were. As far as I know, mammals hadn’t gotten into the treetops yet—they were still in the shadow of their reptilian overlords, hiding in burrows and surviving by moonlight. Synapsids used to have a player in the arboreal game: Suminia getmanovi was a Muppet-faced anomodont (related to dicynodonts) with big eyes, long fingers and grasping hands—making it something of a Permian primate. However, Suminia died out at the end of the Permian along with almost everyone else, leaving the trees free of vertebrate interlopers for a very long time.

Perhaps surprisingly, Suminia's closest Triassic relatives are the big, fat dicynodonts.

Well…until the Late Triassic. Finally, we get to the subject of this essay: drepanosauromorphs. Once known as “monkey-lizards,” these were an unholy combination of chameleon, bird, praying mantis, and (in one case) silky anteater. They’re actually a fairly new group to science: the OG drepanosaur, Drepanosaurus, was named by Pinna in 1980, the same year Calzavara et al. named Megalancosaurus. Pinna would spend the next several years describing his find, but it wasn’t until 1986 that Olsen & Sues actually erected the family, Drepanosauridae.

Over the next three decades, three more genera and various bits and pieces would be described by various authors. In 2010, Renesto et al. summarized the history of drepanosaur research in an excellent monograph which will largely be the focus of this post (there's a link at the bottom).

Exciting Notice: All specimen drawings and skeletals (except one) are taken from Renesto, et al. (2010). All life restorations are by Matt Celeskey, as is that remaining skeletal (you'll know the one). Matt did that skeletal comparison prior to 2010, so the skeletons are not as up-to-the-minute as in Renesto, et al., but the size comparison is accurate, as is the skeletal of Dolabrosaurus--which isn't even in Renesto, et al.

Put simply, drepanosaurs are a small family of geographically disparate (Italy, New Jersey, and New Mexico), but clearly related, arboreal, lizard-like archosauromorphs. They vary in terms of body size and degree of arboreal adaptation. Beyond that, it’s difficult to generalize across all five genera; each is surprisingly unique. I’ll go over the highlights here, but I highly recommend Renesto et al. (2010) for its detail, completeness, and the fact that it’s open access. Actually, all the monographs at the New Mexico Museum of Natural History Bulletins are OA and jaw-dropping.

The original, and only, specimen of Drepanosaurus. Those arms be freaky, yo.

Drepanosaurus unguicaudatus is the original drepanosaur and, in many ways, the most bizarre. Unlike its close cousin Megalancosaurus, Drepanosaurus lacks the specialized hand and foot morphology that suggest improved branch-maneuvering abilities. Rather, its forelimb seems designed specifically to confound future paleontologists.

The humerus is largely unremarkable, but beyond that, things get, well, weird.

Please explain what that fan-shaped bone is in the image above. Notice that it articulates to the humerus in a manner that suggests it is the ulna.

But wait—notice also that it is a separate bone from the more traditional ulna-shaped bone below it, the one that actually looks like two bones overlapping each other. If that fan-shaped bone really is the ulna that means the ulna-shaped bone is a heavily modified wrist bone.

Let that sink in for a minute. And I’m still not sure whether that ulna-shaped thing is one or two separate bones!

Thankfully, the radius is perfectly normal, if a little robust, probably to make up for that...ulna situation. Well that’s it, folks, nothing more to see—oh wait, the hand.

The finger bones are all narrow and spool-shaped and perfectly normal. Digits I and III-V have small, slightly-recurved claws. That’s all fine. But the second digit (NOT the first digit) has a claw that’s more massive than that fan-shaped “ulna!” Yes, that’s a hypertrophied claw on a finger that doesn’t seem to have any obvious adaptations for carrying a claw that’s larger than any other bone in the forearm.

Drepanosaurus' arm compared to a human arm, so you can see how nuts it is. Illustration by Matt Celeskey.

So there’s that.

Questions arise immediately: how, exactly, do the soft-tissue connections work when dealing with a comically large claw on an otherwise unremarkable hand? Does this claw have something to do with the bizarre “ulna” of the lower arm? What in Arceus’ name is Drepanosaurus using this single, ridiculously large claw for? (it turns out there’s a perfectly good answer for that last question, but we’ll get to it later.)

Notice the “hump” on the back, in the skeletal below, which is made up of elongated, sometimes fused, neural spines with T-shaped ends. That’s weird too, and it’s a feature shared by most drepanosaurs. Can’t help but notice that giant fan-shaped thing which seems to articulate with the first couple neural spines! This is not, in fact, your imagination. That’s a real thing, shared by derived drepanosaurs. It's called the supraneural bone, and we’ll talk about it way later.

Take note that of supraneural bone--it shows up repeatedly.

Further back, Drepanosaurus has a grasping foot in which all five toes curl the same way. This will be an important point of difference between Drepanosaurus and its closest relatives.

The tail has hypertrophied haemal spines which arch beneath subsequent caudals—this prevents lateral movement, but the tail was perfectly capable of downward prehensility (though not to the degree of a chameleon’s spiral curl). Notice one other classic bizarre drepanosaur feature at the end of the tail: a thing that looks for all intents and purposes like a damn claw but is made up of modified caudals.

Still, let’s call it what it is: a freaking tail claw.

Drepanosaurus is only known from one specimen—a beautiful but kind of crushed skeleton from northern Italy. It lacks a skull, so we don’t know whether its divergent forelimb anatomy corresponds with a change in skull morphology. One would think it would, but who freaking knows.

I need a drink.

Megalancosaurus endennae, by Matt Celeskey.

Okay. Let’s talk about Megalancosaurus preonensis, another drepanosaur from northern Italy. It’s actually known from several specimens, so we have a complete picture of its skeleton and skull. Despite looking a little more “normal” than Drepanosaurus, it’s still freaky in many important ways.
Importantly, the skull is shockingly birdlike. It is so birdlike, kids, that Feduccia & Wild couldn’t help but link Megalancosaurus to Longisquama and Triassic birds (remember Protoavis?) because of course they did. I'd be worried if they hadn't. In fact, the skull is avian by convergence, but you knew that already because you're a rational person (I assume).

The orbits are enormous, as are the nares, but the upper and lower temporal fenestrae are relatively small. The mandible is low and narrow. The teeth are small and conical—the ones at the front of the snout are larger than those in the maxilla.

The chameleon-like (sort of) skeleton of Megalancosaurus.

The cervical vertebrae are elongated and, like the skull, surprisingly birdlike. Megalancosaurus has the usual drepanosaur hunch-back, increased in height by a strange “supraneural” bone. But let’s get to where the action is: the forearm.

The forearm is long and slender. The ulna is definitely a normal ulna, although it is joined to the wrist by a pair of elongated wrist bones—one wonders if we’re seeing the beginning of the Drepanosaurus condition here. And yes, of course, the hand is a definitely strange, but not in an alien way—rather, it’s essentially similar to the hand of chameleons, with two fingers opposing the other three. 

The feet differ between M. preonensis and M. endennae.

Megalancosaurus was a branch-grasper. The foot, unlike Drepanosaurus, is also modified for advanced grasping—the first two opposes the other four and consists of a sort of hook-like bone and a small, square nubbin. Now, this differs in the two named species of Megalancosaurus: the opposable digit is only present in M. endennae, but M. preonensis appears to have a perfectly normal, un-opposed first digit more like Drepanosaurus.

The tail has elongate haemal arches, a series of which are pierced by largish holes (I mean, why not). Like Drepanosaurus, the arches sweep back beneath several caudals, meaning the tail was not particularly flexible from side to side, but could be curled downward. Also like Drepanosaurus, Megalancosaurus has a tail spike.

Megalancosaurus: not as weird as it could be, folks, but it's tryin' real hard!

Adorable Vallesaurus zorzinensis, by Matt Celesky.

The third Italian drepanosaur is little Vallesaurus—about half the length of Megalancosaurus, this tiny reptile (only 5.9 inches long) otherwise features many of the same adaptations. Of note, it is the only other drepanosaur to preserve a skull—a crushed skull, but a skull nonetheless. It is roughly similar to Megalancosaurus, but short-snouted and deeper in profile. While Vallesaurus does have a small “supraneural” bone, the neural spines of its first few dorsals are not fused together as they are in Drepanosaurus and Megalancosaurus. Its haemal arches lack the holes that characterize Megalancosaurus and the terminal tail spike, present in Megalancosaurus and Drepanosaurus, is MIA.

The pretty great type specimen of Vallesaurus cenensis.

Its hands are not chameleon-like and lack any opposable digits. The foot, however, is a different story. Like Megalancosaurus, two species of Vallesaurus have been named—V. cenensis in 2006, and V. zorzinensis in 2010. Also like Megalancosaurus, they differ based on the structure of the foot. 

In the former, the first toe is ridiculous:

The bone that articulates with the tarsals is actually an enlarged metatarsal bone, and that Y-shaped thing is the one and only phalange. Exactly how it could move in relation to the metatarsal boggles the mind. Does it have a range of motion? Or is it just kind of stuck there, forming an immobile hook? Was it covered in skin or keratin? It looks kind of like a big claw, doesn't it?

Like Megalancosaurus, the feet of Vallesaurus differ between species.

In V. zorzinensis, the first toe is preserved at a more-or-less right angle to the rest of the foot. It is locked in place by the 1st distal tarsal. The metatarsal is similar to V. cenensis, but the preserved phalange is just big, and pinched in the middle. As the distal end of the phalange has a big articular surface, so I assume a second phalange is present but unpreserved. Were these phalanges flexible against the immobile metatarsal?

Notice that the supraneural area: it's quite a bit different than Megalancosaurus and Drepanosaurus.

What bothers me more is the possibility that V. cenensis and V. zorzinensis are two sexes of the same species. They are both known from the same quarry in the same location, in the same time period. The only difference concerns the hallux morphology. Renesto et al. (2010) do not discuss sexual dimorphism in Vallesaurus (but they do entertain the possibility for Megalancosaurus).

Dolabrosaurus, by Matt Celeskey.

We’re not going to talk about New Mexico’s own Dolabrosaurus aquatilis very much because it’s very fragmentary, very small, and broadly similar to Vallesaurus. It seems to have Vallesaurus zorzinensis' opposable toe. But not a whole lot else is known about it.

Otherwise, hey, it’s a drepanosaur in New Mexico!

AND THAT’S ALL FOLKS—wait, what? We forgot one?

Aw, dammit. We have to talk about Hyperonector, don’t we?

Wait a second—I’m gonna get another stiff one.

Hypuronector, with its freaky leaf-tail, by Matt Celeskey

Hypuronector limnaios is known from one partial skeleton and several referred specimens, all from America’s least glamorous Real Housewives state: New Jersey. It’s the freaky drepanosaur that doesn’t look anything like other drepanosaurs. For awhile there, it was considered aquatic, which is DEFINITELY not a drepanosaur characteristic.

Here’s the skeletal, from Renesto, et al. (2010). Drink it in, kids.

This thing doesn't even look like a drepanosaur.

Hypuronector doesn’t have the hunch-back of its relatives or a supraneural bone. Because its hands and feet are poorly preserved, it’s unclear how arboreal it actually was. The forelimbs are longer than the hindlimbs, which is new. Like its relatives, the tail of Hypuronector angles upward from the pelvis, but rather than forming a chameleon-like curling tail, Hypuronector went to town on its haemal arches (might better be termed haemal “rods” at this point), forming a broad, leaf-shaped tail that was only mobile at the base.

Colbert & Olsen, who described this weirdo in 2001, thought the tail was adapted for swimming and that Hypuronector used its leaf-tail as a sculling organ. They also didn’t think that the tail was tilted upward at its base, so that—on the ground—Hypuronector would literally be dragging the bottom of its giant tail through the dirt as it walked.

These are not unreasonable suggestions; after all, Hypuronector is the black sheep in an entire family of very dark-colored sheep.

Here's the holotype of Hypuronector. For the most part, the feet are kind of missing.

However, as Renesto et al. pointed out in 2010, the tail of Hypuronector, like other drepanosaurs, both tilted upward at its base (thus negating the “tail dragging” argument) and the extreme elongation of the haemal rods, each of which underlapped many, many subsequent caudals, meant that the tail was also totally inflexible side-to-side, and wasn’t able to undulate like a proper swimming tail. It probably wasn’t all that mobile up-and-down, either. Renesto et al. (2010) speculate that the giant tail was used for balance in the treetops—assuming that Hypuronector was as arboreal as its cousins—and that it could have many uses, among them parachuting and righting the body during a fall. And while they do not come right out and say it, Renesto et al. do write that:

"The appendicular skeleton of Hypuronector is consistent with that of a patagial glider, including forelimbs and hindlimbs that are approximately the same length, with the forelimbs being even longer than the hindlimbs in some cases, well suited to keep a membrane stretched."

Perhaps Hypuronector was the Triassic equivalent of a flying squirrel or gliding gecko! Eat your heart out, Drepanosaurus.

So now that you’ve met the family, let’s talk about what the heck they were doing all day.
Clearly, these were strongly arboreal animals, and in the case of Hypuronector, possibly scansorial. Talking specifically about Megalancosaurus and Vallesaurus, chameleons seem to be the most obvious modern analogue. For Drepanosaurus, the silky (pygmy) anteater seems like the best comparison. However, the “best” analogue is rarely perfect, and these animals all differ strongly from their modern equivalents.

They probably weren't doing a whole lotta this.

For Megalancosaurus and Vallesaurus, the most obvious non-chameleon feature is the fact that they have long necks, avian skulls, and no adaptations suggesting a chameleon’s specialized tongue. Instead, it’s more likely that these two drepanosaurs normally kept their neck in a raised, bird-like position, darting the mouth forward quickly to capture prey. In Megalancosaurus and Drepanosaurus at least, the wrist and ankle both allowed rotation of the hands and feet. In all known drepanosaurs except Hypuronector, the tail could probably achieve a loose coil and would have acted as a fifth limb to grasp branches and stabilize the body while wandering through the canopy.

It's official: this is the cutest animal on the planet.

Now, Drepanosaurus diverges by lacking any opposable digits at all. Like the pygmy anteater (inexplicably named Cyclopes, by the way), Drepanosaurus has that ridiculously oversized claw on its hand, a prehensile tail for grasping branches, and “barrel-like trunk,” according to Renesto et al. (2010). The lack of opposable digits is unusual, but not unprecedented: remember that Vallesaurus doesn’t have any opposable fingers, and Megalancosaurus preonensis doesn’t have any opposable toes. Drepanosaurus probably occupied its time scratch-digging through bark (I’d like to know what kinds of insects were bark-burrowers in the Triassic), and I wonder, again, if its head and neck morphology differed from Megalancosaurus because of this lifestyle difference.

A very recent paper by Castiello et al. (2015), reconstructing the myology of Megalancosaurus’ forelimbs, suggests that it would’ve been perfectly capable of grabbing and raking in addition to grasping branches. Thus, prey capture may have been an important part of the forelimbs’ role. Renesto et al. (2010) made the same supposition offhand, comparing it to a reptilian praying mantis, but Castiello et al. really did the myology legwork.

Since these guys weren't around in the Triassic (right?!), somebody had to do it.

Next topic: what, exactly, ARE drepanosaurs?

The answer has been hard to pin down. Pinna (1984) and Berman & Reisz (1992) suggested that drepanosaurs are lepidosauromorphs, but Renesto has repeatedly (1994, 2000) argued for an archosauromorph relationship. Benton & Allen (1997) and Dilkes (1998) agreed, adding that drepanosaurs were close to prolacteriforms. In 2004, Senter came to a wildly different conclusion: drepanosaurs weren’t even neodiapsids. He combined drepanosaurs, Longisquama and Coelurosauravidae in a large clade he named “Avicephala,” and he essentially renamed the Drepanosauridae the “Simiosauria” (monkey lizards).

Renesto & Binelli (2006) disagreed with Senter’s methods and (oddly) found a sister group relationship between pterosaurs and drepanosaurs, and this pairing formed a sister group to the whole of Archosauromorpha.

In 2010, Renesto et al. revisited the issue of drepanosaur relationships, revising characters and revisiting old matrices. They advocate the abandonment of “Avacephala” and “Simiosauria.” Drepanosaurs came out as the sister group of Tanystropheidae within Protosauria. Thus, they are archosauromorphs! But man, that’s still a bizarre relationship to think about. Be tee dubs, have you read Mark Witton’s double post on Tanystropheus? Really good stuff, kids.

Matt Celeskey's excellent size comparison. Notice they're not called monkey-lizards anymore, although I've always liked that name.

So one more brain tickler before we leave: what’s up with those hunch-backs and “supraneural” bones that cap the hunches of Megalancosaurus, Drepanosaurus, and Vallesaurus?

Renesto et al. (2010) suppose that they have to do with the neck musculature, perhaps allowing for “sudden extension of the neck, projecting the head toward prey.” Instead of a chameleon tongue, perhaps drepanosaurs were “shooting” their entire heads towards their invertebrate quarry.

With analogues among chameleons, pygmy anteaters, and praying mantises, drepanosaurs were clearly a strange group of reptiles. Their reign did not last long: they first appear at the end of the Carnian age and seem to disappear towards the close of the Norian. The group was wide-ranging, from New Mexico to New Jersey and Italy, and even some fragmentary material from the United Kingdom. I wouldn’t be surprised if drepanosaurs turn up in other Late Triassic localities. They were a flash in the pan, so to speak, but a memorable one nonetheless. 

By the way, as promised, HERE is a link to Renesto et al. (2010). It is excellent, as are all the monographs in that collection.

The arrows point to supraneural bones, all of which are weird. These were weird animals.


  1. Great post.

    Pisanosaurus was recently shown to be a silesaurid (Agnolin, 2015).

    I think you dropped the ball by saying Megalancosaurus' cervicals are "otherwise normal" besides their elongation. They are heterocoelous like derived birds (front end is saddle-shaped) and have huge hypapophyses projecting backwards (ventral midline processes). No doubt this was for their head-lunging habit.

    I agree re: the sexual dimorphism thing, btw.

    1. I've already found some edits I need to make, so I will add that cervical info--thanks!

  2. I’d like to know what kinds of insects were bark-burrowers in the Triassic

    It looks like no modern wood-boring insect families were around in the Triassic (siricoid wood-wasps, for instance, turn up in the Jurassic). There is preserved evidence of borings in wood from that time, but these were likely to have been made in dead, decaying wood and may have been made by non-insects such as oribatid mites. There are a number of insect groups in the Triassic that were likely to have laid their eggs in gymnosperm cones, such as miomopterans (which may have been basal holometabolous insects) and obrieniids (which were kind of Triassic pseudo-weevils), and I suppose drepanosaurids may have been breaking open cones to get at larvae within.

    Mantids, offhand, don't turn up until the early Cretaceous.

  3. Nice to see another devotee of the bizarre drepanosaurs. FYI. there are several additional and significant records for drepanosaurs in North America. You may have missed them as they are only in the form of published abstracts over the last 4 years or so for the SVP annual meetings. A couple involve multiple 3D preserved skeletons from the eolian Nugget Sandstone in NE Utah (Chure and colleagues), one on a partial skull (Pritchard and colleagues, and one on hundreds of isolated drepanosaur bones from the Late Triassic in Texas (Chatteerjee and colleagues I believe). There is also a reputed primitive form from Siberia (partial skeleton) but the jury is still out as to whether or not that creature actually belongs to this group.

  4. And then there is the new Saint and Sinners depranosaur that Chure et al talked about at the SVP.

  5. Hi Zach - I just wanted to leave another compliment on the blog, from another humanities grad with a life-long interest in paleo.

    You're doing something really interesting with your writing: not JUST the smart syntheses about obscure, fascinating groups, making the scientific literature accessible through great framing narratives and authorial voice - which reminds me of the stars at Tet Zoo or SV-POW - but also how you bring your personal story into it a bit as well. You might be right not to make it TOO much about yourself, but the hints about your medical saga, life in Alaska, etc., are something I haven't seen much of yet around the paleo blogosphere. If you took it to a logical extreme, it might look something like New Journalism, but for paleo outreach. Long story short: I'm a fan.

    It's great to see you attracting so many heavy hitters in your comments, and hopefully even more people will discover you in 2016. Also - since you teased it in your very first post, and keep dropping mentions like it's the secret, overarching story of the whole blog - is this the year you reveal the Secret Project??

    1. Thanks for the kind words, Nathan! And yes, I hope to be both more productive in 2016 and attract more readers. As for the Secret Project, I'll probably have more to share in the coming months. I don't want to show my hand too soon, is all. :-)

  6. Please keep in mind that Drepanosaurus head is so far unknown, thus maybe its diet may well have been different, Possibly the new US drepanosaurs of North America will shed new light on this aspect, and other ones as well.
    Looking forward to see them published.

    Silvio Renesto