Why I Use Pseudosuchia Instead of Crurotarsi for the Crocodile-line Archosaurs

This issue was briefly mentioned on the Vertebrate Paleontology List-server about a week ago and I feel like discussing this a bit further. A quick perusal of archosaurian literature will show that there is a dichotomy in the nomenclature when referring to the lineage of Archosauria that leads to the crocodylians. Some workers will refer to this clade as Pseudosuchia and others as Crurotarsi, with the latter seeming to be the most popular. However, I along with some of my closest Triassic collaborators prefer Pseudosuchia over Crurotarsi, and often when I review papers I will suggest that workers using Crurotarsi switch. Why? Well for me it comes down to two issues, priority and stability.


This problem has been reviewed by several workers, such as Brochu (1997) and most recently Senter (2005). Gauthier and Padian (1985) defined Pseudosuchia as "crocodiles and all archosaurs closer to crocodiles than to birds"; however, Pseudosuchia has a much longer history dating back to the late 1800s, and although it has always had the intent of containing non-dinosaurian archosaurs, the membership of this clade has changed through the years and at some points even included some of what are now considered to be non-dinosaurian ornithodirans (the bird-line clade).

Several authors disliked Gauthier and Padian's (1985) redefining of what they considered to be an "ill-defined and misused" name. Furthermore, the name is in a sense contradictory as Pseudosuchia means "false-crocodiles" yet includes crocodiles as members. Accordingly Benton and Clark (1988) suggested a new name, "Crocodylotarsi", and although they did not explicitly define this clade they inferred that it was the same as Pseudosuchia. Sereno and Arcucci (1990) proposed a third name, Crurotarsi, which Sereno (1991) defined as "Parasuchia, Ornithosuchidae, Prestosuchus, Suchia, and all decendents of their common ancestor".

However, just because a group once contained members that have since been recognized as belonging to other groups does not warrant abandonment. Indeed if this were the case very few names would be valid, including Dinosauria. Likewise, contradictory names also are not grounds for dismissal, for example the name phytosaur means 'plant-reptile' although they surely ate everything but. Despite this we are still stuck with the contradictory name. Furthermore, there is no confusion among modern workers as to the meaning of Pseudosuchia, so statements to the contrary are moot. When I say something is a pseudosuchian, those familiar with archosaurs clearly understand what I mean.

Crocodylotarsi has been used by some workers but has since fallen out of usage with most modern workers using either Pseudosuchia or Crurotarsi. As all three groups currently have the same membership it has been argued by some that Pseudosuchia should be used as it has precedence. I agree in principle that the first defined name should have priority and this is the reason that I use Pseudosuchia instead of Crocodylotarsi. However, as noted by Brochu (1997) and Senter (2005) the definitions of Pseudosuchia and Crurotarsi are not the same.

Pseudosuchia is stem-based and thus is flexible to future changes, as any archosaur that is not an ornithodiran is included in this group. Crurotarsi, however, is defined as a node-based taxon and thus has an explicit membership, most notably phytosaurs (parasuchians), ornithosuchids, Prestosuchus, aetosaurs, "rauisuchians", and crocodylomorphs. This definition is much less flexible. In fact let's just suppose that the basal most group of crurotarsans, the phytosaurs, fell outside of the crown-clade Archosauria, and were instead considered to be derived archosauriforms. What would happen to Pseudosuchia and Crurotarsi? The content of Pseudosuchia would be pretty much the same except that phytosaurs would no longer be constituents. In contrast, because the base of Crurotarsi is specified by phytosaurs, Crurotarsi would now include phytosaurs plus all of Archosauria. Thus dinosaurs (including birds) would be Crurotarsans by definition. As you can see this definition is much less stable, another reason why I prefer and highly recommend that all workers use Pseudosuchia over Crurotarsi. Admit it, having to say "non-phytosaurian and non-ornithiodiran crurotarsan" is pretty clunky!

Actually, if this ever did happen ;), in my eyes Crurotarsi might actually now be a useful name when discussing phytosaurs, as you could now simply say that phytosaurs are the basalmost crurotarsans and still be correct. This is probably just slightly more explicit than simply saying they are derived archosauriforms.

For much more detail on this issue read Brochu (1997) and Senter (2005). You can also check out this page for a different opinion.

REFERENCES

Benton, M.J., and J.M. Clark. 1988. Archosaur phylogeny and the relationships of the Crocodylia. Pp. 295–338 in M.J. Benton (ed.). The Phylogeny and Classifi cation of the Tetrapods, Volume 1: Amphibians, Reptiles, Birds. Clarendon Press, Oxford.

Brochu, C.J. 1997. Synonymy, Redundancy, and the name of the crocodile stem group. Journal of Vertebrate Paleontology 17:448-449.

Gauthier, J., and K. Padian. 1985. Phylogenetic, functional, and aerodynamic analyses of the origin of birds and their flight. Pp. 185–197 in M.K. Hecht, J.H. Ostrom, G. Viohl, and P.
Wellnhofer (eds.). The Beginnings of Birds. Freunde des Jura- Museums, Eichstätt.

Senter, P. 2005. Phylogenetic taxonomy and the names of the major archosaurian (Reptilia) clades. PaleoBios 25:1–7.
Sereno, P.C. 1991. Basal archosaurs: phylogenetic relationships and functional implications. Society of Vertebrate Paleontology Memoir 2:1–53.

Sereno, P.C., and A.B. Arcucci. 1990. The monophyly of crurotarsal archosaurs and the origin of bird and crocodile ankle joints. Neues Jahrbuch für Geologie und Paläontologie Abhandlungen
180:21–52.

24 comments:

  1. I want it on a fuckin T-shirt.
    As Bill circumspectly alluded to, the definition differences between Pseudosuchia and Crurotarsi are about to get very important, and the latter term is about to change its understood meaning to the point of becoming completely useless.

    ReplyDelete
  2. As I noted on the VERTPALEO List, the total group may be named "Pan-Crocodylia" under the PhyloCode, anyway (as part of the general convention of naming total groups "Pan-[name of corresponding crown group]"). This name is not mandated, though, so one of the others could be used instead. But even if something else is chosen, "pan-Crocodylia" (unitalicized, lower-case "p") would remain useable as an informal name.

    Personally, I think this is one case where the Pan- convention works quite well, but I guess we will see.

    ReplyDelete
  3. Actually Bill, I think the more likely situation for Crurotarsi to be valid with that definition is that there would be taxa basal to Crurotarsi within Pseudosuchia... no?

    Cheers,
    Nick

    ReplyDelete
  4. So effectively both names are probably not representing the same clade, just maybe most analyses don't have good enough taxon sampling or we don't yet have taxa in between yet known...? :P

    Sorry for the double post... I hit post too quickly before I'd finished my thoughts...

    ReplyDelete
  5. Crurotarsi and Pseudosuchia currently have the same content because all of the specifiers for Crurotarsi currently fall out after the split wih Ornithodira. If one of these specifiers falls outside of the crown clade then the content of Crurotarsi would change significantly.

    I would use the redefined Crurotarsi if another name (Pseudosuchia) did not exist. However, Pseudosuchia has priority, is stable, and is not confusing. Why should we use Crurotarsi?

    ReplyDelete
  6. Well Bill, there have to be taxa between the clade that has the stem-based definition and the clade that has the node-based definition, we just either don't recognize them currently or we haven't found them yet.

    That's what I'm getting at.

    Best Wishes,
    Nick

    ReplyDelete
  7. Interesting problem. I do support using names with temporal priority and I don't care at all about etymology. My issue with Gauthier's definition is that it uses crocodiles as the internal specifier instead of Dyoplax, Aetosaurus or Typothorax (Zittel's original pseudosuchians). And yes, I have the same issue with definitions of Dinosauria, Saurischia, Theropoda, Coelurosauria, etc. that include birds. In fact, Pseudosuchia originally excluded not only mesoeucrocodylians, but also parasuchians. So I'd support Pseudosuchia for some (Aetosaurus <- Parasuchus, Crocodylus) clade, but not for pan-crocodylians. And yes, I know Dinosauria etc. didn't originally include birds, but the latter ended up being deeply nested in it whereas pseudosuchians ended up being thrown every which way compared to parasuchians and crocs.

    As for Crurotarsi, Sereno's definition is terrible, so par for the course there. I cringe at redefining clades a decade after the fact though, so I might very well support Crocodylotarsi as the lesser of three evils when it comes to pan-crocs. Or would I overcome my hatred of pan-names and embrace Pan-Crocodylia? Luckily I work on theropods so don't have to come to a conclusion any time soon. ;)

    Incidentally, I agree with Nick that the currently crappy topologies in that part of the tree leave open the possibility of non-crurotarsan pseudosuchians (sensu Gauthier and Padian).

    ReplyDelete
  8. FWIW, the PhyloCode, which both Gauthier and Padian are involved with, would disfavor Gauthier and Padian's definition for the very reason Mickey states. See Recommendation 11A (with Dinosauria as a specific example).

    I'm not sure about Padian, but Gauthier now supports the Pan- convention for names of total groups. Arguing for using "Pseudosuchia" for the total group may put you in the odd position of disagreeing with the original definitional authors! Granted, there's no good published alternative for now, but hopefully "Pan-Crocodylia" will be published along with the PhyloCode or shortly thereafter.

    ReplyDelete
  9. Nick,

    The definition of Crurotarsi is based in part on where phytosaurs fall out (I'm picking on them because they are currently accepted as the basal-most taxon). So say a future analysis finds Revueltosaurus outside of Phytosauria, then yes Revueltosaurus (or some future unknown taxon in the same position) would be a non-crurotarsan Pseudosuchian.

    Pan-Crocodylia....ugh... I don't know if I can do that yet, although I'll admit that its meaning would be clear. Still, it would just be a replacement name for Pseudosuchia, so what is the point except to make crown clade names uniform (BTW.. what happens if all modern crocodylians happen to go extinct?). I really don't care about the historic usage of the word Pseudosuchia, just its current meaning. It's not too hard to block out that past usage if you try.

    ReplyDelete
  10. Mickey/Mike,

    Recommendation 11A is a good one for FUTURE use as this will help eliminate potential confusion; however, I do not see why we need to discard all of the older defined names just because this recommendation was not in place in 1984. Bringing back Crocodylotarsi doesn't help anything.

    It seems to me that there is just a general dislike among VPs for the name Pseudosuchia. Am I imagining this? Is it really that confusing?

    ReplyDelete
  11. Sometimes the conversion of a paraphyletic group to a clade takes (Dinosauria, Synapsida) and sometimes it doesn't (Reptilia, Thecodontia). This is one where I think it's not that clear.

    Still, note that Recommendation 11A might allow a definition like Clade(Aetosaurus ← [insert dinosaur here]) or something. So there's a legitimate case to be made for naming the total group "Pseudosuchia" under the PhyloCode. If that happens, though, I'll probably prefer the informal alternative "pan-Crocodylia" because it fits in better with the general scheme of "panclade names". (Vertebrate paleontology is one of the few fields--or the only one?--where total groups have been explicitly named. Few other taxa have stem groups that are as well-known and studied.)

    I like the "Pan-" convention. It's clear, and to can be applied to any crown group name. There's only one place where I think it falls apart, and that's for apes. (Think about it....)

    ReplyDelete
  12. "BTW.. what happens if all modern crocodylians happen to go extinct?"

    Then we'll have other problems to worry about, since, if this happens in our lifetimes, it will probably mean that the planet's entire ecosystem is imploding. :)

    Seriously, though ... PhyloCode definitions that rely on the concept of "extant" require it to have an objective, unchanging meaning in the context of the definition. Either the author(s) can select one, or, if they do not, then the implicit meaning is "extant on [the definition's] publication date" (Art. 9.5). In cases where this is not clear, the definition can be clarified through an unrestricted emendation (Art. 15.11–15.15).

    In short, they'd still be crocodylians, and everything sharing closer ancestry with them than with anything else extant at the time of definition would be pan-crocodylians.

    ReplyDelete
  13. I really hate the idea of using "Pan-" as a prefix for all total clades. Not only is it unimaginative, but it really screws things up for stuff like databases. Imagine having to sort through a drop-down menu of clade names, and half of them start with Pan-. Or even sorting through a simple list! If you are going to standardize clade names, I would much prefer a standard suffix, like -morpha or -formes.

    ReplyDelete
  14. Um, doesn't it make it easier to find names of total clades? A suffix would ensure that they are scattered to the four winds. (Although they would be paired with their corresponding crown clade names, which I suppose would be nice. In either case, a standardized affix helps.)

    As for "unimaginative", utility trumps that. If I know there's a clade called "Goomboppiformes" and later on I see a clade called "Pan-Goomboppiformes", then I know exactly what it is. If I see "Ploogenjoamorpha" then I have no idea.

    Even if we did use a suffix, "-formes" would be a bad choice, since a great number of crown group names end in that suffix. We'd have stuff like Passeriformes-formes. (I guess "-morpha might be all right, though. And it has been tried.)

    Still, though -- "Pan" means "total" and "morpha" means "forms". If you want to establish a convention for a type of taxon, doesn't it make more sense to pick an affix that refers directly to that type of taxon? These are total groups, not form taxa.

    ReplyDelete
  15. I'm curious as to what people think the name of the sister total group, i.e., the avian total group, should be. Sereno discussed this and opted for "Avemetatarsalia". For the same reasons outlined above, I prefer "pan-Aves" (or "Pan-Aves" once it's published). Other options include "Ornithodira", "Ornithosuchia", "Panaves", "Aves", "Ornithotarsi", and the rather poetic "Dracones".

    ReplyDelete
  16. "I really don't care about the historic usage of the word Pseudosuchia, just its current meaning. It's not too hard to block out that past usage if you try."

    To me it seems improper to change the meaning of names so much that they lose all resemblence to their original usage.

    "Recommendation 11A is a good one for FUTURE use as this will help eliminate potential confusion; however, I do not see why we need to discard all of the older defined names just because this recommendation was not in place in 1984. Bringing back Crocodylotarsi doesn't help anything."

    Sure it does, since Crocodylotarsi has only ever had one concept associated with it. With Pseudosuchia, I'd have to specify sensu Gauthier and Padian, sensu Zittel, sensu Romer, etc..

    "It seems to me that there is just a general dislike among VPs for the name Pseudosuchia. Am I imagining this? Is it really that confusing?"

    I can assure you I have no personal bias against the name. I'm the one who's all for maintaining Reptilia, Megalosauroidea, and other archaic names after all (go Podokesauroidea!). I just think it was improperly defined by Gauthier and Padian, just like I won't use Sereno's (1998, 1999) definitions of Ornithomimidae, despite them being the first published.

    "I'm curious as to what people think the name of the sister total group, i.e., the avian total group, should be. Sereno discussed this and opted for "Avemetatarsalia". For the same reasons outlined above, I prefer "pan-Aves" (or "Pan-Aves" once it's published). Other options include "Ornithodira", "Ornithosuchia", "Panaves", "Aves", "Ornithotarsi", and the rather poetic "Dracones"."

    Ornithodira and Ornithotarsi are less inclusive, Ornithosuchia fails due to not including Ornithosuchus, Aves shouldn't be nearly so inclusive, Dracones has never been defined, and Panaves was defined after Avemetatarsalia. So I go with Avemetatarsalia.

    ReplyDelete
  17. I always forget that when Phylocode is implimented we will pretty much be starting over with names.

    This has been a real interesting discussion. Thanks guys. Believe me this discussion will have relevance soon.

    BTW...for the other clade I tend to use Ornithodira because as Mickey says Ornithosuchia is out because it includes Ornithosuchus. Guess that sort of makes me a hypocrite as Ornithosuchia is the first defined name. ;)

    ReplyDelete
  18. Mike - yeah, I find it much better to have the total and crown group names paired together rather than separated. And maybe its just me, but I find it less intuitive and more of a pain to sort through hundreds of "Pan-" names, when if a suffix is implemented, I only have to look through two names that start the same way.

    Bill - I agree with you about the "starting over" part. I am very much in favor of Phylocode, but I find it dissapointing that so many perfectly well-defined clade names and definitions will be thrown away. I hope in their wisdom, the authors of the chapters for the Phylonyms companion volume will try and preserve as many of these commonly used names as possible. I think that priority of definition is an important thing, and I'd hate to see all that thrown away.

    One thing that bothers me specifically about Archosauriformes is that none of the workers who are real active among basal archosaurs have been asked to co-author the chapter for Phylonyms, as far as I'm aware. Shouldn't the active users of the definitions have a say in their construction?

    ReplyDelete
  19. Phylonyms could actually use some more authors in these areas, for sure. (And other areas as well -- last I heard we have NOTHING for Insecta at all.) If you're interested in working on this volume, it may not be too late. (Although it may be -- I'm not sure.) Contact the editors: Jacques Gauthier, Kevin de Queiroz, and Philip Cantino. And even if definitions for these names don't get in to Phylonyms, there's still time to prepare manuscripts for when the PhyloCode is enacted.

    On a related note, I'm currently a member of the Committee on Phylogenetic Nomenclature (as is Kevin Padian, previously mentioned), and we are currently considering some revisions to the code, especially to do with the term "extant". (The term "Recent" will likely be stricken from the the next draft.) If anyone has concerns, I'd be happy to bring them up.

    ReplyDelete
  20. One more thought. Mike Keesey wrote- "Still, note that Recommendation 11A might allow a definition like Clade(Aetosaurus ← [insert dinosaur here]) or something. So there's a legitimate case to be made for naming the total group "Pseudosuchia" under the PhyloCode."

    The problem I have with that is that dinosaurs were not originally paired against Pseudosuchia. Zitell named it as a taxon of Crocodilia, opposing Parasuchia, Mesosuchia and Eusuchia. If anything, his version of Crurotarsi was Crocodilia, since that included aetosaurs, paruchians, sphenosuchians, more derived crocs and later ornithosuchids. Nor is the classical Pseudosuchia of Romer paired against dinosaurs, with it being a subgroup of Thecodontia paired against aetosaurs, parasuchians and proterosuchians. As it included basal avemetatarsalians (e.g. Lagosuchus, Scleromochlus, Lagerpeton), it would be equivalent to crown Archosauria if anything.

    ReplyDelete
  21. But there is a way we can use both Crurotarsi and Pseudosuchia... :
    Pseudosuchia is the clade containing Crurotarsi and the problematic (obscure and possiblu Nomen nudum) Teleocrateridae
    What do you think about it?

    ReplyDelete
  22. Sure we can use both names and we tend to use them interchangeably. The point is that they are defined differently, and if certain specifiers within Crurotarsi are found the fall outside of Archosauria, then what is included within Crurotarsi can change radically (e.g., dinosaurs would be crurotarsans).

    I'm not familiar with Teleocrateridae which is presumably based on Teleocrater, an isolated vertebra from the Middle Triassic of Tanzania and currently a nomen nudum. Presumably this is a Family created for a group of "rauisuchids"?

    ReplyDelete

Markup Key:
- <b>bold</b> = bold
- <i>italic</i> = italic
- <a href="http://www.fieldofscience.com/">FoS</a> = FoS