Welcome to the third, and last, of the rhynchosaur articles. The other two are mandatory reading: part I is a general intro, part II is on jaws and teeth. This time round, we look at the form and function of the postcranial anatomy (well, predominantly at the limbs actually), and also at rhynchosaur phylogeny and at their place in the grand scheme of things [life restoration of Hyperodapedon shown above, from Benton (1983)].
We begin with the forelimb. The rhynchosaur humerus is stout, with large crests for muscle attachment and a wide, flaring distal end. The rest of the forelimb appears fairly typical for a diapsid reptile. Many sources state that the rhynchosaur hindlimb appears well suited for scratch-digging (this is where curved claws are used, in a raking motion, to shift sediment), but few elaborate on it. An excellent evaluation of rhynchosaur hindlimb function was provided by Benton (1983), and most of the following is taken from that discussion. While illustrations usually depict rhynchosaurs as ground-hugging animals with short, fully sprawling limbs, the details of their limbs and limb girdles show that their limbs were semi-erect and that their bodies were typically held well up off the ground.
In the shoulder girdle, the glenoid (= socket) is located far posteriorly and faces backwards and sideways (rather than being located closer to the middle of the girdle and facing fully sideways). These features and others indicate that a semi-erect forelimb gait could be adopted. In the hindlimb, the acetabulum (= hip socket) is broad and shallow, and faces downwards and backwards as well as sideways. When the in-turned head of the femur is articulated with the acetabulum, a semi-erect (rather than fully sprawling) posture seems to work best. The location of the distal condyles on the distal end of the femur (rather than on its ventral surface) indicate that the femur normally functioned in a near-vertical (rather than near-horizontal) pose, and the straight, column-like tibia also indicates a semi-erect stance. Rhynchosaur hands and feet were not close together and directly underneath the body, as they are in erect-limbed tetrapods, but they were closer together than those of truly sprawling tetrapods [diagram below, from Benton (1983), shows Hyperodapedon hindlimb in lateral and anterior view. The two lateral views show some of the inferred musculature].
The hindlimb bones are thin anteroposteriorly, but broad mediolaterally. Large attachment areas were available for well developed flexors and extensors of the foot. Combined with the near-parasagittal (parasagittal = parallel to the long axis of the body) movement of the femur, this shows that the long, strongly clawed feet could have functioned well in digging.
The foot claws of rhynchosaurids were not long and pointed, but very deep and narrow, and borne on digits that allowed a wide range of flexion and extension. As you can see from this diagram [below] from Benton (1983), these claws were similar in shape to those of some extant scratch-diggers, like golden moles and marsupial moles [in diagram below, from Benton (1983), rhynchosaur foot claws and a single hand claw (all at left) are compared to those of (from top to bottom) marsupial mole, pangolin, long-nosed armadillo, pygmy anteater, aardvark and golden mole]. It is inferred from these details that rhynchosaurids used their feet to break up and shift sediment, perhaps while they were foraging for roots, tubers and other plant structures. Longer, slimmer unguals were present in Mesosuchus, so this basal rhynchosaur may not have indulged in the same behaviour.
The three ages of rhynchosaur
Rhynchosaurs can be imagined to fall into three rough groups, two of which are grades, not clades. Basal rhynchosaurs - namely Mesosuchus, Howesia and Noteosuchus from the Early Triassic - share with all later forms a united naris, down-turned premaxillae and the presence of several maxillary tooth rows, but they were less 'extreme' in morphology than the more derived rhynchosaurs, all of which are united in Rhynchosauridae.
Basal rhynchosaurids form the second group. These animals had deeper, shorter skulls than basal rhynchosaurs, a more robust body, stockier limbs that were more similar in length than those of basal rhynchosaurs, and a much shorter tail. The several Rhynchosaurus species belong here, as does the African Stenaulorhynchus and the English Fodonyx. Traditionally, these taxa were classified together as the Rhynchosaurinae (Chatterjee 1969, 1974). However, this group is evidently paraphyletic and has now been abandoned [skeleton of basal rhynchosaur Mesosuchus shown below, from Dilkes (1998)].
Finally, hyperodapedontines form the third group. These are the 'most extreme' rhynchosaurs: the biggest ones, and the ones with the broadest, shortest, deepest skulls and the most complicated tooth fields. Their fore- and hindlimbs were nearly equal in length, and their bodies were bulkier than those of other rhynchosaurs. Langer & Schultz (2000) defined Hyperodapedontinae as a branch-based taxon that includes all rhynchosaurs closer to Hyperodapedon than to Fodonyx. The branch-based definition of Hyperodapedontinae means that Ammorhynchus and 'Scaphonyx' sulcognathus are both part of the clade if we follow the topology recovered by Hone & Benton (2008). Several hyperodapedontine species previously regarded as representing distinct genera, including Scaphonyx from Brazil and Paradapedon from India, were found by Langer & Schultz (2000) to be nested within Hyperodapedon. Isalorhynchus from Madagascar has also been shown to be referable to Hyperodapedon (Langer et al. 2000).
What are rhynchosaurs?
Like so many other Triassic reptile groups, rhynchosaurs have been moved all about the tetrapod family tree since their recognition in 1842. On naming Rhynchosaurus articeps in that year, Richard Owen initially described some rhynchosaur material as that of a 'labyrinthodont' (= temnospondyl). He later realised this mistake and came to regard rhynchosaurs as close relatives of dicynodonts, as the tusk-like caniniforms of the latter were wrongly thought to be homologous with the paired premaxillary beaks of rhynchosaurs*. An affinity with Sphenodon (tuatara) was also suggested, as the beak-like premaxillae of the tuatara was, again, thought homologous with the beaks of rhynchosaurs. Thomas Huxley also supported an affinity between rhynchosaurs and Sphenodon when he described Hyperodapedon in 1869, and he proposed the name Sphenodontina for this group [Sphenodon shown here, from wikipedia].
* Incidentally, I've just noticed that Owen referred to big teiids as 'New World monitors'. How cool.
While numerous different classification schemes were proposed during the course of the 20th century, it was mostly thought by the 1960s and 70s that tuatara and rhynchosaurs were close relatives, that they formed a group termed Rhynchocephalia, and that Rhynchocephalia was part of Lepidosauria (the group that also includes lizards and snakes). Yes yes, I know that things were actually much more complicated, but that'll do as a review: for more complete retellings see Benton (1985, 1990) and Dilkes (1998).
Numerous studies published since the 1980s have shown that rhynchosaurs and the members of the tuatara clade were not really closely related (R. Burckhardt had argued as early as 1900 that the supposedly similar premaxillary beaks of tuatara and rhynchosaurs were actually completely different). Tuatara and kin - now termed the Sphenodontia* - are lepidosaurs (and hence close to squamates), while rhynchosaurs are archosauromorphs: part of the same group as archosaurs and their relatives. Archosauromorpha has been defined as a branch-based taxon that includes all taxa closer to Protorosaurus than to Lepidosauromorpha** (Dilkes 1998, p. 528) and autapomorphies of the clade include slender cervical ribs, a notch on the leading margin of the interclavicle, and an ilium with a small anterior and large posterior process (Dilkes 1998).
* Sphenodontians are closely related to Gephyrosaurus from the Early Jurassic and both form a clade that needs a name. Gauthier et al. (1988) decided to co-opt Rhynchocephalia for this purpose. Given the confusing history of this name (in addition to sphenodontians and rhynchosaurs, 'Rhynchocephalia' of old authors might also include choristoderes, thalattosaurs and/or younginiforms), I think this is a really dumb idea and would have preferred a new, more 'sphenodontian-themed' name, like Sphenodontiformes or Sphenodontomorpha. Anyway, too late to change this now.
** A definition using an archosaur specifier would have been more sensible in my opinion.
Within Archosauromorpha, the distribution of characters indicate that rhynchosaurs are closest to the Prolacerta + archosaur clade, with trilophosaurs and protorosaurs being successively more distant. Not all studies agree on this topology: Prolacerta was recovered as the sister-taxon to a rhynchosaur + archosaur clade by Müller (2003). In the diagram above (it depicts a very simplified 'consensus' cladogram), note that Lepidosauria and Archosauromorpha are both branch-based clades that, together, form the node-based clade Sauria. Rhynchosauria and Archosauria are both node-based (I've used Archosauria here in the crown-group sense, in which case the more inclusive clade - corresponding to Archosauria of tradition - is Archosauriformes. Archosauriformes is also node-based, so even this name does not correspond to the archosaur total group. How dumb). So far as I can tell, there is no published phylogenetic definition of Protorosauria.
The basal members of all archosauromorph clades seem to have been long-tailed quadrupeds, superficially similar to living monitor lizards, and generally somewhere round about 50 cm in length. After appearing in the Late Permian, these basal forms had produced a weird and wonderful assemblage by the end of the Early Triassic. In fact, the development of so many 'extreme' morphologies among Triassic archosauromorphs suggests that some pretty heavy selection pressure was acting on these Permo-Triassic generalists. Look at rhynchosaurid rhynchosaurs, trilophosaurs, Teraterpeton and Tanystropheus*. You see, this is why we need time travel to be invented.
* Drepanosaurids and Longisquama might be in there too, but this was contested by Senter (2004). I'm currently pretty sceptical of his argument.
So, that's rhynchosaurs done at last. Only trilophosaurs, protorosaurs, thalattosaurs, choristoderes and all those Triassic archosaurs to go... Oh yeah, and all the rest of those temnospondyls.
Refs - -
Benton, M. J. 1983. The Triassic reptile Hyperodapedon from Elgin: functional morphology and relationships. Philosophical Transactions of the Royal Society of London B 302, 605-718.
- . 1985. Classification and phylogeny of the diapsid reptiles. Zoological Journal of the Linnean Society 84, 97-164.
- . 1990. The species of Rhynchosaurus, a rhynchosaur (Reptilia, Diapsida) from the Middle Triassic of England. Philosophical Transactions of the Royal Society of London B 328, 213-306.
Chatterjee, S. 1969. Rhynchosaurs in time and space. Proceedings of the Geological Society of London 1658, 203-208.
- . 1974. A rhynchosaur from the Upper Triassic Maleri Formation of India. Philosophical Transactions of the Royal Society of London B 276, 209-261.
Dilkes, D. W. 1998. The Early Triassic rhynchosaur Mesosuchus browni and the interrelationships of basal archosauromorph reptiles. Philosophical Transactions of the Royal Society of London B 353, 501-541.
Gauthier, J. A., Estes, R. & de Queiroz, K. 1988. A phylogenetic analysis of the Lepidosauromorpha. In Estes, R. & Pregill, G. (eds) Phylogenetic Relationships of the Lizard Families: Essays Commemorating Charles L. Camp. Stanford University Press (Stanford), pp. 15-98.
Hone, D. W. E. & Benton, M. J. 2007. A new genus of rhynchosaur from the Middle Triassic of south-west England. Palaeontology 51, 95-115.
Langer, M. C., Boniface, M., Cuny, G. & Barbieri, L. 2000. The phylogenetic position of Isalorhynchus genovefae, a Late Triassic rhynchosaur from Madagascar. Annales de Paléontologie 86, 101-127.
- . & Schultz, C. L. 2000. A new species of the Late Triassic rhynchosaur Hyperodapedon from the Santa Maria Formation of south Brazil. Palaeontology 43, 633-652.
Müller, J. 2003. Early loss and multiple return of the lower temporal arcade in diapsid reptiles. Naturwissenschaften 90, 473-476.
Senter, P. 2004. Phylogeny of Drepanosauridae (Reptilia: Diapsida). Journal of Systematic Palaeontology 2, 257-268.
- Log in to post comments
More like this
Great stuff Darren, it's nice to have the dalt with properly online so I can direct people to it and furutre and indeed use it myself. At the very least it saves me having to do it! ;-)
Thanks Dave. Mike Benton himself has been good enough to visit, he had this to say...
Darren,
Thanks for making such a fuss about rhynchosaurs - nobody has ever cared much about them. I've cross-linked those blogs, and will add your third, on my reprints web site.
You should definitely do trilophosaurs next. They are so misunderstood, and so little known about them is in the public sphere. I saw one of their skeletons at a museum in Texas once, and I mistook it for a Permian synapsid. They're actually that large!
On a similar but unrelated note, did you know that some paleontologists consider trilophosaurs to be very basal rhyncosaurs. I personally doubt this, but its still weird.
"Anyway, too late to change this now."
Not at all, since the PhyloCode hasn't been enacted yet. In fact, I would not be surprised if the total group were called Pan-Sphenodon in the Companion Volume. It would then just remain to define Rhynchocephalia as something like, "the most inclusive clade containing Sphenodon punctatus and Rhynchosaurus articeps but not Lacerta agilis, Crocodylus niloticus or Vultur gryphus," and then we can nullify the name forever!
"... note that Lepidosauria and Archosauromorpha are both branch-based clades that ..."
Psst .. Lepidosauromorpha (or "pan-Lepidosauria").
Ok, trilophosaurs one day. But protorosaurs first.
Re: Lepidosauromorpha, I deliberately didn't use it because there are conflicting definitions. Gauthier et al. (1988) defined Lepidosauromorpha as the clade including younginiforms, Palaeagama, Paliguana, Saurosternon, Kuehneosauridae, Rhynchocephalia and Squamata.
But Laurin (1991) later found younginiforms to be outside of the lepidosaur + archosaur clade, in which case Lepidosauromorpha sensu Gauthier et al. now has the same content as Sauria. This is obviously contrary to the spirit of the name, so Laurin (1991) provided a new definition for Lepidosauromorpha: 'the most recent common ancestor of Palaeagama, Saurosternon, Paliguana, Kuehneosaurus, Sphenodon and squamates and all its descendants' (p. 89). That makes sense, but it ignores the priority established by the Gauthier et al. (1988) definition.
I decided to avoid discussing this by just ignoring Lepidosauromorpha altogether. Cases like this show why we shouldn't adhere to strict rules of priority: a matter recently discussed in Jack Conrad's AMNH bulletin on squamate phylogeny, if I remember correctly.
Refs - -
Gauthier, J. A., Estes, R. & de Queiroz, K. 1988. A phylogenetic analysis of the Lepidosauromorpha. In Estes, R. & Pregill, G. (eds) Phylogenetic Relationships of the Lizard Families: Essays Commemorating Charles L. Camp. Stanford University Press (Stanford), pp. 15-98.
Laurin, M. 1991. The osteology of a Lower Permian eosuchian from texas and a review of diapsid phylogeny. Zoological Journal of the Linnean Society 101, 59-95.
Incidentally, given that Mike has mentioned the PhyloCode volume, anyone got any news on its planned appearance?
I had assumed rhynchosaurs were affiliated with dicynodonts, because they looked so much like them. Were they like them in another way too, flourishing during the Triassic, and then being wiped out at the end of that age (though I think some isolated pockets of dicynodonts survived even into the Cretaceous.) Your description of their gait sounds like the crocodilian "high walk", not fully erect, but not sprawling either.
Hmm. All of the few sources I've seen had R. articeps. But R. atriceps would make at least as much sense: it could refer to the big cavities in the head (the huge temporal fenestrae), while... "art-headed"... Has someone got the original paper and can check?
Article 11 of the PhyloCode contains several recommendations in that direction, but no rule (Art. 11.7 does not apply, because Archosauromorpha isn't derived from Archosaurus). What do you think? Should I propose one? I mean... I'm in the Committee on Phylogenetic Nomenclature now, and not a single e-mail has been exchanged since I got elected... :-)
I thought Lepidosauria is node-based? I also think a branch-based definition for Lepidosauromorpha exists... or at least I hope so; the name was coined back when some people thought branch-based definitions were to be avoided because they thought they were somehow automatically unstable.
But Archosauromorpha does. Archosauromorpha is precisely the archosaur total group.
Unsurprising, because the content/monophyly of that group (respectively Prolacertiformes) is unclear. No phylogenetic analysis of the problem with reasonable taxon sampling exists yet.
Oho! May I ask why?
Not necessarily: all except the last two could be stem-diapsids and have been found as such by Müller (2003, 2004) and/or Senter (2004), IIRC.
Well, we should, but we should ignore the infancy of phylogenetic nomenclature. That's the idea behind the companion volume.
Yes: keep waiting. One contribution has been accepted so far.
Incidentally, concerning Laurin 1991, there's a surprise on Apsisaurus in the pipeline.
The $10,000 question is whether choristoderes (including champsosaurs) are archosauromorphs, lepidosauromorphs, or basal diapsids...
Choristores are the new turtles?
Thanks David for comments. It is indeed Rhynchosaurus articeps, my bad.
Definition of Archosauromorpha: while I think we can be confident that Protorosaurus is closer to archosaurs than to lepidosaurs, we would guarantee an Archosauromorpha that includes Archosauria if we use an archosaur specifier. So, yeah, propose one (after, that is, you consult with others).
Lepidosauria: I now can't recall (or find) where I got the idea of a branch-based definition from. I may have become confused with the concept of a branch-based Lepidosauromorpha (see below). Most authors seem to have a node-based definition for Lepidosauria in mind: in Gauthier's thesis, for example, it's simply stated to be the clade that includes Rhynchocephalia and Squamata.
Lepidosauromorpha: Gauthier et al.'s definition was branch-based, Laurin's node-based. The best definition would be a branch-based one that corresponds to the lepidosaur total group, and both Gauthier et al.'s and Laurin's definitions fall short. Note, incidentally, that 13.2.3 of the PhyloCode draft uses Lepidosauromorpha to illustrate homonymy.
On Archosauriformes not corresponding to the archosaur total group...
Oh yeah. I was referring instead to the group that includes all archosaur-branch taxa that go back to the common ancestor with rhynchosaurs. There is no name for this clade so far as I can tell.
On lack of definition for Protorosauria: several studies have found Prolacerta to group closer to archosaurs than to Protorosaurus (and see Modesto & Sues 2004), and Protorosaurus to group with tanystropheids and other taxa conventionally classified as 'prolacertiforms' or 'protorosaurs'. So, this 'Protorosaurus clade' should be called Protorosauria. You're right that an extensive study of protorosaur and other archosauromorph taxa is still wanting, but I still expected there to be a defition kicking around out there (e.g., all taxa closer to Protorosaurus spenceri than to... insert trilophosaur, rhynchosaur and archosaur specifiers). I lost my copy of Rieppel et al. (2003) while writing the rhynchosaur articles. I found it in the end (it was in a folder of drepanosaur literature, and not with the protorosaur s. s. stuff): no definition therein.
Sceptical about Senter (2004): well, let's see more testing. To discuss Senter's data at length would require a paper in itself. Maybe later :)
Finally... Apsisaurus: don't tell me... stem-synapsid?
Ref - -
Modesto, S. P. & Sues, H.-D. 2004. The skull of the Early Triassic archosauromorph reptile Prolacerta broomi and its phylogenetic significance. Zoological Journal of the Linnean Society 140, 335-351.
Rieppel, O., Fraser, N. C. & Nosotti, S. 2003. The monophyly of Protorosauria (Reptilia, Archosauromorpha): a preliminary analysis. Atti Soc. it. Sci. nat. Museo civ. Stor. nat. Milano 144, 359-382.
You're willing to pay ten kilobucks for my PhD thesis? Really? :-)
I've heard that the evidence shows that they are most likely a sister group to archosauromorpha, or else one of the more basal groups within it (alongside prolacertiforms and rhyncosaurs)
Re: naming clades. I ***wish*** there were a convention to the effect that, when X-iformes and X-omorpha are both used, the endings allowed you to tell which is more inclusive. I gather there isn't (and, given that the two names might be coined by different researchers at different times, the convention might not be workable anyway).
---
With that off my chest... very, very nice article (in three parts), Darren!
Further complicating matters is the status of the vermiformes, members of which have been identified as lizards, serpents, turtles, and in the case of the terraverminids, mammals. Recent anatomical and genetic studies have shown that the group is derived reptilian, more closely related to the crocodilians than any other.
Of course, if you happen to live in the real world, none of this actually applies.
Those are some mighty dainty-looking hands and forearms. I could imagine it mincing around like a raccoon, except aren't raccoon feet pretty central under the body? Maybe they just weren't equipped to run.
Wow, Mike reading blogs, he is taking is sabbatical seriously. Again though, from me at least (and certainly Mike too it seems), well done mate.
Thank you darling. Oh, and see you in a few weeks.
Wait a minute. I'll check the paper, because if "the clade including" really is how it's worded, this is not a definition at all! There isn't just a single clade that includes all of these, there's a practically infinite number, for example the clade that includes all known life!
Here we go again. <headdesk> There are several clades that include all life which happens to be known right now, for example the crown-group and the total group.
It is if you follow the fine recommendation of Taylor (2007:2):
In a spirit of generosity, and ... in the absence of a formal code governing phylogenetic definitions, it is probably better to interpret malformed definitions, where the intention is clear, as meaning what the author intended rather than to discard them entirely. This latitude may be allowed because pre-PhyloCode phylogenetic definitions are best understood as the output of an ongoing learning process as the biological community gradually becomes accustomed to using phylogenetic nomenclature. Such license should certainly not be extended to future malformed definitions published under the governance of the PhyloCode, but until then it is better not to exclude such definitions from the corpus.
(PDF available from my publications page.)
Would that be Fodonyx Hone and Benton 2007?
I think you see the problem here :-)
Whoops, I had thought Lepidosauromorpha was branch-defined. Well, it should be! (Or replaced with "Pan-Lepidosauria, whichever.)
Wait, who's on sabbatical? (Man, I hate having this name sometimes. I think there should be a 50-year moratorium on naming any children "Michael".)
Exactly.
Why don't you use your first one, actually...?
Is there any reason to suppose every Precambrian acritarch is inside the crown group?
"Timothy" is my father; I've never been called by my first name and it would feel weird to start now. (Many people do just use my last name, though.)
I'd like to add my thanks for taking the time to put all this information together. I can only imagine how much time and effort it takes, but I wouldn't know where to start looking to find all this out independently.
I have Reign of the Reptiles - I picked it up in one of those cheapy bookshops for a couple of quid. One of my favourite books on my Science / Nature shelves because there has been such a dearth of non-dino reptile information available that I've seen. I trawl this blog and others you link to for more book recommendations all the time - my Amazon list grows exponentially I think :)
Thanks again!
Darren, a triumph! (tri=three, umph=ooomph=resoundingimpactfulness). Your 3 articles on rhynchosaurs I mean.
Re gait, in the images: is Mesosuchus doing a swinging-from-the-hips thing (turning its femur in a horizontal arc, from forward-and-outward to backward-and-outward, like Dimetrodon is often shown doing), in contrast to the Hyperodapedon walk? Is this a real difference or conjecture or rash reconstruction? (In your opinion... time travel not yet being availble to check ;-) ) What tale do the ankles tell you?
Oh and Hyperodapedon (in the top image, right) has a MUCH less cute face here in iii than in part ii. Downright fiercey-looking!
Better remember to smile and look cute if it wants adopted or soft-toyified.
On the other hand... maybe it's looking also to the "children-who-want-their-extinct-critters-red-in-tooth-and-claw" market.
PS Did I miss it, or did you not actually tell us wht you think they ate with all that weird (but no doubt highly suitable for whatever it was they ate) dentition? Do tell!!
(Time travel caveat applies of course).
One last PS... (!) Do we know the orientation of any semicircular canals in the skull? and does that (with axial skeleton morphology etc) provide useful info on rhynchosaurs' habitual head posture? and does that throw any light on eating habits? (Grazing, browsing, fruit-grabbing-and-dragging-down, etc).
How mobile was the neck, and could the head roll enough to apply bite laterally - or only sagittally? (I am thinking of piercing or gnawing vertical stems/trunks to lap juice, or any crazy such thing).
er... I mean I am thinking of rhynchosaurs doing that... not myself.
Perhaps rhynchosaurs were the Triassic equivalent of the larger hystricomorph rodents and castorimorph rodents, using their large teeth to crop vegetation, possibly thick tubers and roots.
By the way, how likely is it for rhyncosaurs to colonize offshore islands, especially compared to dicynodonts? I ask this because it seems that they were adaptable little critters, and some seem to be found in areas that might allow them to get in a sweepstakes distribution to an island. Once there, freed from the competition with the larger dicynodonts, they could grow to larger sizes. Who knows, there could be a cow-sized rhyncosaur out there somewhere!