Poposaur palates

The palates of poposaurs are poorly known Some have not been described or reconstructed (Fig.1). Others have been wrongly reconstructed or partially reconstructed (Fig. 4). Here (Fig. 1) are two poposaurs, Effigia and Shuvosaurus next to Daemonosaurus (Sues et al. 2011, also largely guessed at from broken pieces) and Thecodontosaurus, which provides more certitude. Most unfortunately, the palate of Lotosaurus has not been described or illustrated despite the presence of several specimens and museum casts. The little question is: On Daemonosaurus, which way do the ectopterygoids go? Long side against the pterygoid, as in rauisuchids? Or short side, as in Effigia and other dinosaurs?

Figure 2. Effigia palate in situ (left) and reconstructed by reassembling colored elements (at right).

Figure 2. Effigia palate in situ (left) and reconstructed by reassembling colored elements (at right).

On rauisuchians, as in ornithosuchians (Fig. 2), the ectopterygoid has a larger contact area with the lateral pterygoid and it produces a small “stem” to contact the jugal (as in Saurosuchus) or the maxilla (as in Riojasuchus). If you flip the ectopterygoid of Daemonosaurus, you get the rauisuchian type of ectopterygoid. Left as is (Fig. 1), however, you get the dinosaurian type,  and that is the preferred reconstruction here based on phylogenetic bracketing.

Click to enlarge. Euparkeriid, ornithosuchian, rauisuchian, aetosaurian, and basal archosaur palates.

Figure 2. Click to enlarge. Euparkeriid, ornithosuchian, rauisuchian, aetosaurian, and basal archosaur palates. Here are Euparkeria and Osmolskina, both euparkeriids. Ornithosuchus and Riojasuchus are ornithosuchids. Saurosuchus and Postosuchus are both rauisuchians. Stagonolepis is an aetosaur. Pseudhesperosuchus is close to the basal archosaur pattern with a much smaller ectopterygoid and smaller ectopterygoid/pterygoid contact. The original configuration is shown on the right side. A possible alternative is shown on the left. Not sure how it was preserved. I’d like to know if you have this data. If the left is correct in figure 2 (Pseudohesperosuchus), and Shuvosaurus is also correct in figure 1, these suggest that Daemonosaurus is correctly drawn in figure 1.

Silesaurus palate with missing elements restored on the right.

Figure 4. Silesaurus palate with missing elements restored on the right. Illustration (without color) from Dzik 2003 who illustrated missing elements on the left.

Silesaurus Palate The missing ectopterygoid and palatine were not illustrated for Silesaurus. Given the palates of related taxa (Fig.1), I have added the missing elements on the right here (Fig. 4) to match them. Thus these restorations are guesses that appear to make sense in context. When better data come along, we’ll make improvements.

This has been a first attempt at reconstructing the palates of several poposaurs at once based on similar morphologies in close kin. The palates should remain somewhat similar. If anyone has good data on the palates of other rauisuchians and basal dinosaurs, please forward them on.

As always, I encourage readers to see specimens, make observations and come to your own conclusions. Test. Test. And test again.

Evidence and support in the form of nexus, pdf and jpeg files will be sent to all who request additional data.

Figure 3 is absent from this post now. Apologies. I had it in my files for several years and thought it had been published by now. It had not. 


References Bonaparte JF 1969. Dos nuevas “Faunas” de reptiles Triasicos de Argentina: I. Gondwana Symp., IVGS: 283-306.
Borsuk-Bialynicka M and Evans SE 2009. Cranial and mandibular osteology of the Early Triassic archosauriform Osmolskina czatkowicensis from Poland. Palaeontologia Polonica 65, 235–281.
Brusatte SL, Benton MJ, Desojo JB and Langer MC 2010. The higher-level phylogeny of Archosauria (Tetrapoda: Diapsida), Journal of Systematic Palaeontology, 8:1, 3-47.
Chatterjee S 1985. Postosuchus, a new Thecodontian reptile from the Triassic of Texas and the origin of Tyrannosaurs. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 309 (1139): 395–460. doi:10.1098/rstb.1985.0092.
Chatterjee S 1991. An unusual toothless archosaur from the Triassic of Texas: the world’s oldest ostrich dinosaur? Abstract, Journal of Vertebrate Paleontology, 8(3): 11A.
Chatterjee S 1993. Shuvosaurus, a new theropod: an unusual theropod dinosaur from the Triassic of Texas. National Geographic Research and Exploration 9 (3): 274–285.
Dzik J 2003. A beaked herbivorous archosaur with dinosaur affinities from the early Late Triassic of Poland. Journal of Vertebrate Paleontology 23: 556-574.
Ewer RF 1965. The Anatomy of the Thecodont Reptile Euparkeria capensis Broom Philosophical Transactions of the Royal Society London B 248 379-435. doi: 10.1098/rstb.1965.0003
Rauhut OWM 1997. On the cranial anatomy of Shuvosaurus inexpectatus (Dinosauria: Theropoda). In: Sachs, S., Rauhut, O. W. M. & Weigert, A. (eds) 1. Treffen der deutschsprachigen Palaeoherpetologen, Düsseldorf, 21.-23.02.1997; Extended Abstracts. Terra Nostra 7/97, pp. 17-21.
Long R and Murry P 1995. Late Triassic (Carnian-Norian) Tetrapods from the Southwestern United States. New Mexico Museum of Natural History and Science Bulletin 4, Pp. 153-163.
Sill WD 1974. The anatomy of Saurosuchus galilei and the relationships of the rauisuchid thecodonts. Bulletin of the Museum of Comparative Zoology 146: 317-362.
Sues H-D, Nesbitt SJ, Berman DS and Henrici AC 2011. A late-surviving basal theropod dinosaur from the latest Triassic of North America. Proceedings of the Royal Society Bpublished online
Walker AD 1961. Triassic reptiles from the Elgin area: StagonolepisDasygnathus and their allies. Philosophical Transactions of the Royal Society B 244:103-204.
Walker AD 1964. Triassic reptiles from the Elgin area: Ornithosuchus and the origin of carnosaurs. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 248(744): 53-134.
Yates AM 2003. A new species of the primitive dinosaur Thecodontosaurus (Saurischia: Sauropodomorpha) and its implications for the systematics of early dinosaurs. Journal of Systematic Palaeontology 1(1):1-42. wiki/Daemonosaurus wiki/Shuvosaurus

20 thoughts on “Poposaur palates

  1. Those palates are almost completely hypothetical or just plain wrong. Basically none of the elements in Daemonosaurus are correct, except possibly the parasphenoid which we don’t know the shape of in ventral view. Your ectopterygoid is probably part of the pterygoid in lateral view, your palatine is part of the lacrimal and matrix, etc.. The palatines of Effigia are all only partially exposed and illustrated in oblique views, so you don’t know their shape, and indeed what you draw contradicts Nesbitt’s description (e.g. that they indicate a reduced choana). The maxilla, pterygoid and palatine of Shuvosaurus are far more fragmentary than you show, the vomer’s unknown, and the ectopterygoid should articulate laterally. Your “Thecodontosaurus” is Pantydraco, where the palatine has mostly broken margins, the ventral shape of lateral bones is unknown (ditto for Daemonosaurus), as is the articulation between palatal elements. Note Dzik did illustrate the missing palatal bones of Silesaurus, but only in outline.

    To answer your Daemonosaurus question, in the actual ectopterygoid, not enough of the jugal process was preserved/exposed to know if it was anteroposteriorly longer than the pterygoid process. But you think the ectopterygoid as part of the coronoid process of the mandible because you believe your tracing low res photos more than Sues et al.’s observations of the actual specimen.

    Oh, and that Shuvosaurus illustration is from Lehane’s thesis. Probably best to not post it, since it’s unpublished. Your habit of using figures without citing the source (e.g. Silesaurus here being from Dzik, 2003) is bad enough.

    Btw, new post up about your “theropods”

  2. Just like in math class, Mickey, you need to show your work to be helpful. Again, the ectopterygoid of Daemonosaurus is atop the coronoid process. Both are present. Dzik did illustrate the missing palate bones, but, like mine, they were totally imaginary. Were his correct? If so, by what reason?

    • And again, that supposed ectopterygoid is just part of a larger bone which is probably part of the pterygoid. Why you think your tracings of low res photographs are better than Sues et al.’s interpretations based on studying the specimen itself is beyond me. Just pointing out all of the errors you make in this one skull would be a day’s time, let alone tracing the probable correct interpretations, let alone doing any of this for all of the est. 2000 errors you made in the dinosaur codings of your analysis.

      For Silesaurus, you wrote above “The missing ectopterygoid and palatine were not illustrated for Silesaurus.”

      My last post responding to your analysis is now up. It covers just what characters support your phytodinosaur clades and why nearly all are based on miscodings and/or missing taxa, while the few that do work usually work as well or better in the traditional phylogeny. I also go over Daemonosaurus’ characters that nest it in Theropoda.

  3. It would be helpful if these same poposaur palates were reconstructed by another worker. Perhaps you can point me to that reference. Otherwise, these may be the first and only such reconstructions. If you don’t do the reconstruction using in situ elements, how can you replicate or falsify my test in a scientific manner? For Silesaurus note that some of the elements were originally drawn in as hypothetical. With regard to Daemonosaurus, you are approaching it from a list of theropod traits, whereas I am approaching it from a phylogenetic analysis. I will be more willing to accept your phylogenetic analysis when Theropoda is directly derived from basal Archosauria and the various plant-eating clades are derived from theropods along the lines of Daemonosaurus, which includes a suite of traits found in phytodinosaurs. Thank you for your thoughts.

    • The problem (and this is true of many of your reconstructions) is that sometimes without seeing the material yourself, the published images just aren’t enough to allow a reliable reconstruction. It’s rare to be presented with all of the views necessary to reconstruct elements in dorsal or ventral perspective, for example.

      But even just looking at Effigia’s pterygoid, which is the one articulated palatal element actually shown in ventral view, you get it utterly wrong. Just look at my comparison here- http://postimg.org/image/80hr1y699/ . Scaled to the same element length, your palatine process is a third the width it should be (the disarticulated vomer overlaps part of it in the specimen), the ectopterygoid articulation (colored green in my pic, as the ectopterygoid would cover it, though the element itself is disarticulated elsewhere) is about half the length it should be, you confusingly think the transverse flange is narrow and projects far lateral to the ectopterygoid articulation, your quadrate process is longer and half the width it should be, you missed the ventral process of the quadrate that bisects the pterygoid’s quadrate process. You get the missing/hidden parts completely wrong, coloring the base of the quadrate transverse processes that way, but actually it’s the middle of the palatine process, most of the basipterygoid process and a bit of the quadrate process that can’t be seen. Note although the palatine looks like it’s in the same position as in your reconstruction, Nesbitt notes it is displaced and rotated, so that’s not its true position or shape.

      This is yet another example of how flawed your tracings/reconstructions are, and why tracing all the thousands of times you’re wrong would be a life’s work. How many more examples would you need to doubt the general utility of it when done by yourself?

      How can you even honestly say I didn’t approach Daemonosaurus from a phylogenetic analysis?! I coded it for all of the characters in your phylogenetic analysis and ran that. Then added the characters used by Sues et al. to place it in their phylogenetic analysis, coded it, and ran that. If that isn’t approaching it from a phylogenetic analysis, what is?!

      Note too I already tested Daemonosaurus as a phytodinosaur in the various matrices. It never helps Phytodinosauria become more parsimonious, and usually makes herrerasaurids, Eoraptor and/or Eodromaeus phytodinosaurs too. Your interpretation of it as being phytodinosaur-like is just wrong.

      You’ll be more willing to accept my phylogenetic analysis when it matches yours? Do you realize how unscientific that is? You’ll be more willing to accept my analysis when the results match yours, ignoring that I already used all of your characters and corrected them. It’s only the results you care about, not the validity of the data used to get those results. You’re as bad as the BANDits, who know their result and don’t care about the massive amount of data refuting it or how inaccurate their interpretations are. To them, birds evolved flight from trees and had manual digits II-III-IV, no matter what other data there is. They’d be more willing to accept BAD if dinosaurs were arboreal and had II-III-IV, so maniraptorans get in now.

      Remember when you said “You have your assignment: Nest Daemonosaurus with theropods while including Heterodontosaurus and Massospondylus.” Now that I’ve done just that, in excruciating detail, you say you aren’t willing to accept it because I got the answer you challenged me to, and not the answer you like. Surely you’re better than that, David.

  4. There’s something wrong with your work if you don’t get a gradual blend of traits, readily visible in a series of reconstructions that look like one could morph into the other, with allowances, of course, for autapomorphies. Your tree has to make sense in a holistic sense and right now it doesn’t. I know this is frustrating for you. You’re also making fictitious accusations in your emotional fury that are inappropriate.

    • First, I want to note you didn’t address how/why you got Effigia’s palate so wrong, and what that means for trusting the rest of your reconstructions. Also note you never explained how I didn’t “approach Daemonosaurus from a phylogenetic analysis.” Like so many critiques, these merely fade into the ether.

      I’d like to ask why you, and you alone, are the only paleontologist to think it matters if cladograms result in a phylogeny where change is “readily visible in a series of reconstructions that look like one could morph into the other”? No one else seems to think that. Do you have some special insight into phylogeny and evolution all of the phds in the field somehow lack? Are there references out there saying given our incomplete fossil record (note a third of the taxa I used in my critique weren’t known before 2000), we should expect such gradual change in readily visible form? I’d say no. This is just your idea of how evolution should work, and you reject conclusions that disagree with it. Why do you think your idea is reliable compared to everyone else in the field? Think of other fields where one amateur thinks they have the answer while all the professionals are wrong. How often is that amateur correct?

  5. Mickey, you asked, “I’d like to ask why you, and you alone, are the only paleontologist to think it matters if cladograms result in a phylogeny where change is “readily visible in a series of reconstructions that look like one could morph into the other”? No one else seems to think that.”

    Simply because that’s the way evolution works. Small steps. Minimum changes are most parsimonious. (I realize I’m preaching to the choir here:) This is the bedrock upon which we trust and test phylogenetic analysis. Putting a series of skulls or feet or overall reconstructions in a series should show where red flags exist or where your tree appears to echo actual evolutionary events. I don’t know what you’ve done to mess up your tree, separating theropods from basal archosaurs, but that’s something that needs a good second look at. I know this is frustrating for you. I wish you every success.

    • But PAUP’s already finding the minimum changes in my tree by definition. I am getting the “most parsimonious trees” after all. What you’re arguing is that it’s only/mostly the gross similarity that could be seen in a reconstruction that should count toward parsimony. Or else that there are additional characters you’re seeing in reconstructions but which aren’t in your matrix, in which case feel free to list them. In either case, the ball’s in your court to justify the importance of gross similarity over detailed similarity, or list the other characters you’re seeing but not using.

      And can you please stop falsely claiming it’s my results which are frustrating to me? What’s frustrating is having you propose an explicit test for your phylogeny, me spending days performing that test, then you rejecting it because my conclusions didn’t match yours. Which was the very thing being tested.

      Just what test could convince you that your dinosaur phylogeny is wrong?

  6. MIckey, I’m looking for results that produce a more gradual accumulation of character traits in derived taxa than in my results. Fewer steps. More parsimony. More logic. Your results don’t do that for whatever reason.

    • But my results ARE more parsimonious than yours. I demonstrated that extensively in my recent blog entries, describing exactly what you got wrong, like coding Effigia as having stout maxillary teeth or Silesaurus as lacking gastralia or [insert 2000 more examples here]. So apparently a more parsimonious cladogram in fact doesn’t convince you that you’re wrong. I’m beginning to think there’s nothing anyone else can do that could convince you.

  7. Thank you for the gastralia data on Silesaurus. I was working on the holotype and did not consider the new specimens. I checked on my Effigia data and I don’t show any teeth, nor do I code for “stout” maxillary teeth. You must have input an error. The gastralia change did not affect the topology of the tree.

    • Character 112- “mx_tooth_size / ‘>2x_deeper_than_wide’ not,” . You have Lotosaurus, Shuvosaurus and Effigia coded as state 1- maxillary teeth not more than twice as deep as wide.

      As for Silesaurus’ gastralia, that’s a start. Note this is one of the characters falsely placing Lotosaurus as a silesaurid in your matrix. Now you can fix the nine other taxa you miscoded for gastralial presence/absence and you’ll be half a percent closer to having an accurate matrix.

  8. This is where your coding perverts mine. In character 112 you have to have mx teeth twice deeper than wide or not. That means all conditions other than twice deeper is “not.” That includes toothlessness. Perhaps if you recode again using this strict yes/no dichotomy you will find a tree that echoes actual evolutionary events.

    • This is where you don’t understand phylogenetic analyses. By coding both stout teeth and toothlessness as one state, you’re saying toothlessness is more similar to stout teeth than it is to slender teeth. Which it’s not. Remember, PAUP doesn’t actually care whether state 0 or 1 is primitive, so groups can be diagnosed due to sharing state 0 too. So in your tree as of January, you had ‘Paraornithischia’ diagnosed in part by “maxillary teeth stout or maxillary teeth absent”. Those are two different conditions that aren’t related to each other. It’s as if your state were “eyes not green”, and it was grouping red-eyed taxa with eyeless taxa simply because they weren’t green-eyed.

      By “perverting” your matrix, I was actually fixing it. Just think of how little your character state makes sense as an evolutionary event. Is there a “tooth stout or absent” gene that could be inherited? Ask any phylogeneticist whether it’s proper to combine “toothless” with a tooth feature in one state, and you’ll see they agree with me. I can call Marjanovic over if you need confirmation.

  9. I’m sorry, Mickey. There is no “maxillary teeth stout” anywhere in my matrix. This is your rewording of my coding. Bottom line: when you can line up a series of taxa that gradually produce theropods, ornithischians, sauropodomorphs and poposaurids let me know. If instead your tree produces no solution to the origin of various dinosaur subclades, then it’s time to get back to the chalk board. I’m eager to see a logical solution.

    • Surely you understood I was shortening “maxillary teeth less than twice as tall as wide” to “teeth stout” to aid discussion. If your one state is “mx teeth twice deeper than wide”, what else do you think the opposite of that would be?

      Do you think I’m just making this rule up about having each state be a definite morphology and not “anything else than the other state”? Jenner (2002) wrote an entire article about it. As he said “The general problem of unspecified character states in phylogenetic parsimony is the incorrect suggestion of similarity in morphologically dissimilar taxa, and the unsupported assumption that the disparate morphologies united within a trash can character state represent a clear alternative to the other coded character state.” It’s even free online here- http://dpc.uba.uva.nl/ctz/vol71/nr01/art06 Please read it, learn from it, fix your analysis.

      Do you think it’s not important to follow criteria the experts in the field set forth? How can you expect to get an accurate cladogram when you don’t even try to follow the basic guidelines of making valid character states?

  10. Somehow I got a more accurate cladogram than you did. It provides logical and tenable solutions both in whole in and in part. How? I used the best data available, which is sometimes crappy. As always, if you have better specific data, like the gastralia issue, please send it along. My guess as to the key to our disagreements probably centers on the different ways we code poposaurid characters. When that’s resolved our trees will look more like each others’. Best.

    • How can you just know your cladogram is more accurate? You don’t follow the proper procedures and you don’t care. You have your tracings demonstrated to be incorrect time and time again and you don’t care (still didn’t even aknowledge your Effigia pterygoid wrongness I see). You proposed a test of your data, then dismiss the test because it resulted in a different tree than yours. Face it David, you’re a pseudoscientist. There’s no way you could ever be convinced you’re wrong and you refuse to learn. So I’m done trying to teach you. Have fun drifting into obscurity like John Jackson, Stephen Pickering and Peter Mihalda. My only regret is I wasted so much time thinking maybe if I educate you, you could actually do something useful with your energies. But if you’d rather wallow in your ignorance and arrogance, so ahead.

  11. No, there is a way and it’s happened time and again. This morning, for instance, it happened when you mentioned gastralia and Silesaurus. I checked it out, found the new data and I changed my data matrix. I also asked for correct or updated data for other taxa. You can start by listing the reptiles that need or lack gastralia that you mentioned. Unfortunately your reply (see above) included more name calling and negative associations. I encourage you to keep working on your tree because parts of it are not logical or gradual. Let’s take the emotion out and get back to Science. Share data and the results should start to merge.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.