Maybe Asilisaurus had shorter arms

Updated September 3, 2020, 
but should have been updated much earlier. Asilisaurus and Lotosaurus are both poposaurids, taxa that nest outside Archosauria in the LRT (Fig. x).

Figure 5. Subset of the LRT focusing on the Poposauria and surrounding clades.

Figure 5. Subset of the LRT focusing on the Poposauria and surrounding clades.


Figure 1. Asilisaurus. Above: as reconstructed by Nesbitt et al. (2010). Below: With shorter arms, more in accord with the thickness of the bones. The related Lotosaurus was a quadruped and had much more robust forelimbs. Other sister taxa had relatively short arms.

How Long Were the Arms of Asilisaurus?
(Nesbitt et al. 2010) is a basal dinosaur with incomplete forelimbs. It was originally reconstructed with long forelimbs in a quadrupedal configuration (Fig. 1). The question is: were the arms reconstructed too long? The humerus includes both ends but has a broken middle. The ulna/radius does not include the proximal ends.

Asilisaurus is the oldest known reptile in the dinosaur lineage, according to Nesbitt et al. (2010) and Wikipedia, but the large reptile tree finds Lotosaurus* is also a dinosaur that is just as old. Both lived during the Anisian period of the Middle Triassic (245-237 mya). Asilisaurus is a sister to Silesaurus in alll studies.

Considering only the Nesbitt drawings (my only data), it appears that the fore limbs may have been drawn too long in order to force a quadrupedal configuration. Why not let the bone diameters help determine their lengths? Moreover, sister taxa, like Pisanosaurus and Poposaurus, had short forelimbs. Silesaurus may have been bipedal. Lotosaurus was a sister taxon with robust forelimbs and a quadrupedal configuration. If Asilisaurus were indeed quadrupedal more robust forelimbs might be expected following these patterns. Even Pisanosaurus has more robust forelimbs.

*Nesbitt (2007) suggested Lotosaurus was a poposaurid, more closely related to Shuvosaurus, not to Xilosuchus, but in the same large clade of rauisuchians. The large reptile tree nested poposaurids within the Dinosauria, but Xilosuchus and Arizonasaurus, the other two finbacks, with the rauisuchids.

Nesbitt SJ, Sidor CA, Irmis RB, Angielczyk KD, Smith RMH and Tsuji LMA 2010. Ecologically distinct dinosaurian sister group shows early diversification of Ornithodira. Nature 464 (7285): 95–98. doi:10.1038/nature08718PMID 20203608.

23 thoughts on “Maybe Asilisaurus had shorter arms

  1. Asilisaurus is the oldest known dinosaur in the dinosaur lineage” – Nope. The “dinosaur lineage” is the stem-lineage for Aves (when “Aves” is the crown for birds). The stem-lineage extends all the way back to the point it branches from the next stem-lineage leading to a different crown group, which in this case is the Crocodylia lineage (in the Nesbitt analysis, this begins with Aetosaurus et al. on the tree). If we only treated dinosaurs to exclude birds, this is still true in the phylogeny presented. Agilisaurus kongwe is not the oldest member of the bird-stem, as Marasuchus lilloensis and Lagerpeton chanarensis/Dromomeron roemeri/gregorii (all included in the phylogeny noted) and Pterosauromorpha are more basal on the stem leading to the split with crocodilians.

    However, I should note that incomplete means merely incomplete — one can try to question this, and on its face you’d be correct. But the authors are basing their reconstruction on Silesaurus opolensis, a taxon in which the limb bones really are that ridiculously long, thin, etc. There is no substantive reason to question the length of the bones due to incompleteness and use “slenderness” as a viable metric: Silesaurus opolensis presumes this “slenderness” is not only possible, the close affinities of the two taxa infer it to be true on a level 2 inference (close relationship and similar morphology). Nesbitt et al. (2010) write:

    Although incomplete, the estimated lengths of the humerus and radius indicate
    Asilisaurus had elongate forelimbs like Silesaurus.” – pg. 97

    • Blame Wikipedia for the oldest dinosaur line. And by the way, Jaime, you may be confusing chronologically oldest with phylogenetically more basal. I did say dino. I don’t see you listing any dinos older than Asilisaurus there. And with regard to your other argument on arm length, that’s why we I said, “Maybe.” We’ll look for more data to find out which is correct.

      • Jaime, while we’re at it… you seem pretty sure that pterosaurs are in the lineage of dinos. Maybe you can shed some light on the gradual accumulation of pterosaurian traits in closest kin? I’ve been waiting for so long for someone to step up to the plate. Brother, you’re up!

  2. Even if you don’t end up with complete forelimbs there are other ways to test if Asilisaurus was capable of bipedalism. Calculating the center of gravity and/or calculating the load-bearing strength of the hind limbs could give you some roundabout ways to test your idea.

  3. Dave, you ask that I “step up” to support pterosaur evolution, especially as a basal stem-avian taxon. I do not need to. My comments were derived following the phylogenetic analysis of Nesbitt et al., presented in the paper I (and you) cited in this post, modified as it is from earlier analyses and more complete as a result (addition of taxa, etc.) in some respects. It is unnecessary for me to support that without simply pointing at it. You’ve been asked to publish your own data, and have been strongly encouraged to do so. My response was on the relative distribution of taxa as the “stem” of lineages, and nothing more, using Nesbitt merely as a graphical representation.

    I have a challenge for you:

    Without using your “DGS” technique, can you support your character codings, present them to symposia, or publish them? If yo8u have to use “DGS”, could you base this on verifiable methodology that seeks to eliminate pariedolia and confirmation bias? If you cannot, I will continue to put MORE stock in the analyses that recover the typical “psuedosuchian” and “dinosauromorph” phylogenies that aren’t yours. If you can, and seek to publish them, I will support you in your efforts to present this data as equivalent. As of now, your results are NOT equivalent (i.e., not scientific).

    • How many taxa of the 300 do you suppose I have used DGS on? Very few. If 30 that’s 10%. And what portion of those taxa did I score differently than prior studies? A finger? A skull? That’s a minority within the minority.

      The great, grand majority of my data comes from the literature. We can remove every suspect taxon and you’ll still have the same tree. I do that all the time. You can try it too with my matrix.

      Back to the Nesbitt paper, this widespread refusal to look at Tanystropheus, Huehuecuetzpalli and Proganochelys as possible sisters to pterosaurs is with precedent. It’s human nature. No one wants to be wrong even though science is all about figuring out what’s right and what’s wrong. There’s no testing in your camp. There’s nothing but testing here. I cut. Pare. Switch. What have you. Jaime, don’t fall for the propaganda. Test the opposition yourself. If you think my results are NOT scientific it should be relatively easy to come up with a better tree that includes the new candidates.

      You guys are all recalcitrant.

      “I’m not responsible for your reality” — cute phrase I first heard today that I’m going to be using a lot from now on.

      • “We can remove every suspect taxon and you’ll still have the same tree.”
        “… refusal to look at Tanystropheus, Huehuecuetzpalli and Proganochelys as possible sisters to pterosaurs…”

        Do you recover the same tree as Nesbitt (and earlier trees) by removing these taxa?

        Agree with Jamie about publishing or presenting data. Submit an abstract to SVP for example. Obviously folks are interested in what you have to say…

  4. No, Rob, my tree, like a fractal amoeba – every part when broken apart retains its integrity and resolution. I recover my tree, or parts thereof. Why would you think I would recover Nesbitt’s tree? The Nesbitt tree has been demonstrated to be a mix of this and that. That’s why there’s problems with resolution, some taxa are strange bedfellows and there’s no gradual accumulation of derived traits. Doesn’t anything sink in?

    I have published, but since 2003 virtually all attempts at publishing have been blackballed, with only a few exceptions. My published works are listed at

    • You consider the exclusion of those taxa as notable and possibly topology changing. Are they so important they could cause a rearrangement of the tree that Nesbitt presented? Would this tree perhaps more closely resemble your tree?
      If so then the removal of those taxa preventing Nesbitt’s tree from looking like yours from your tree should result in your tree looking more like Nesbitt’s. Right? You have pointed out some things about Nesbitt’s that you disagree with (not all of which I agree with) – fair enough. But still – wouldn’t this be an easy way to test against your tree without having to code new characters for all your taxa currently in your matrix?

      Is there any evidence of the blackballing? Or have your papers/abstracts just not been accepted?

    • I find the idea of a “fractal” tree that retains its topology regardless of how fragmented you make it improbable, or more so highly suspect as to the methodology used when splitting the tree. One begins to suspect the characters and codings are deliberated specifically to enforce topologies at this point, as such an effect is, in my experience when working with Sereno’s 2000 trees, the only possible way to achieve this effect. This methodology is flawed.

      The solution is not removal of taxa, regardless of suspicion. One includes taxa or even specimens regardless of their fragmentariness, as literature on “wildcard” and incomplete taxa has shown. This is not where the flaw lies, but rather on formation and inclusion of characters. This is where the fundamental difference in yours and Nesbitt’s analyses lie — and in fact with everyone else including your critics.

      • The solution indeed is not removal of taxa but addition of taxa. Even so, once a tree is established with x number of taxa it’s easy to test characters and taxa by removing sets of them to see if and how the topology changes. The flaw does not lie in the formation of characters. You can’t make this stuff up with 60,000± character scores. To prove this point, as I’ve done before, I can remove characters down to 40% of them and the topology doesn’t change. It’s really up to you and others now to test this. Otherwise, as I’m being accused right now, I’ll be accused of some sort of bias. You and others have to test this independently.

  5. Rob, apparently you don’t understand the concept behind and in which I present a new tree in which all sister taxa look similar to one another, a gradual accumulation of traits is demonstrated for all taxa and it is chronologically ordered (with a few anachronisms like Sphenodon and Huehuecuetzpalli). If Nesbitt included several choristoderes, several Youngina/Youngoides and several lizards, then I’m hypothesizing that his tree would shift more toward mine. Right now he has Euparkeria nesting between Chanaresuchus and Parausuchus, among other strange bedfellows listed last week. How is this tenable?

    Once again, removing what taxa? Please be specific. I did not follow your logic train.

    • There’s a fundamental problem with trying to affirm affinities of taxa by “shape matching.” You aren’t using RFTRA, R or PCA to compare the shapes of things, or even basic qualification of your over-representation of proportion-based characters; you’ve not supported these with the raw math they need. You’re simply saying “these look more similar” and use this to try to claim that certain fitted taxa are NOT allied, or others are (seriously, how long have we debated your attempting to throw aquatic taxa together? WAAAY too long). Your argument FAILS to discern the difference between basal grades developing into different descendant groups (such as basal sauropterygians), but instead try to lump them all into one so-called cohesive group. No specification but similar form. It’d be like me trying to pretend that all toothless dinosaurs were a clade to the exclusion of their ancestors with teeth:

      Oviraptoridae, Caenagnathidae, Ornithomimidae, Pygostylia; excluding Caudipteryx zoui, Incisivosaurus gauthieri, Harpymimidae/Pelecanimimus polyodon, and ALL non-avian Paraves, meaning Archaeopteryx lithographica. Some of these have gastroliths, some don’t. Some have hypocleidia on their furcula, some don’t. Some have extensive pneumatic vertebrae and skulls, some don’t. And they aren’t all consistent. It is, in fact, the intrinsic issue at play with all the various phylogenies playing around with where exactly Xiaotingia zhengi goes, messing with troodontids, dromaeosaurids, unenlagiids, and “archaeopterygids.”

      The solution to this problem is MORE CHARACTERS, BETTER DATA, BETTER CHARACTERS … and above all, the ability to qualify the data used in the first place. As of now, I cannot reproduce your methods for “interdigital lines,” baby pterosaurs, integument preservation on a host of levels, and even wing-finger claws or missing bones. Longisquama insignis is the poster-child for what needs to be fixed about this methodology, and it needs to be handled NOW, before this work can be taken seriously.

      • Also, you’re applying a Larry Martinism here by holding up one character in taxa I’m not even studying. Get it together brother. See the big picture, all 228 traits. Check out the nexus file before making judgements. Come up with an alternate scenario and let’s test yours against mine. It’s that easy. Or support Nesbitt’s nestings if you can. Tell me why Parasuchus nests with Eudimorphodon.

  6. Jaime, you’re stuck in a logic rut. You’re balking at a very simple concept: More taxa. Just let it happen. Also, let me help you find the interdigital lines. Send me your problem. I’ll send you the solution and help you find your own solutions. Why is Longisquama a poster child? One of the so-called plumes is a complete hind limb. The subdivided feather shafts at the backbone are some of the toes. The furcula is a sternal complex. The tail is behind another plume. The bumps on the head is a rotated parietal. Come up with a different interpretation and we’ll talk about it. Baby pterosaurs? You’re talking about embryos in eggs? Or known juveniles? Let’s get this settled. Send me your problems.

  7. Dave, I didn’t pick one trait, I picked a suite of features you ascribe to form, specifically calling you out on how you manage to use a host of proportion-based characters to affirm aquatic taxa as sister taxa. “Aquatic bauplan” is NOT a single trait (I should hope not, anyways).

    Further, “Come up with a different interpretation and we’ll talk about it.” Seriously, Dave. At this point YOUR analysis is the one that differs from the others presented out there; you are the outlier to this analysis. Such things bear greater scrutiny. You seek to offer a more comprehensive analysis, but rather than make your analysis more rigorous, your feature-finding and confirmation more exact, your methodology more transparent, you are simply reaffirming the old process of using pixellated images in Photoshop. The problems with your analysis are legion, and they’ve been raised before (Darren Naish did a decent job of pointing them out). By biggest problem with the analysis is that even when I use better imaging or personal examination, I cannot find what you see. I asked you to try to use this technique for better known taxa, such as Archaeopteryx, WAAAAY back on the Dinosaur Mailing List, but nothing came of it.

    You want specifics? I directly challenged your interpretation of Jeholopterus ningchengensis “vampire teeth” by telling you what you spied as a ridiculously long tooth was, in fact, a pair of closely associated teeth. Authors who’ve examined the material in person have never affirmed this discovery, and despite presentation and discussion against your “discovery” since SVP, you’re now modeling the “vampire” as though it were true.

    You find pterosaur “maxillae” as distinct from their premaxillae along with little extra external bony nares. Anhangueridae? Wukongopteridae? Pterodactylidae? Ctenochasmatidae? Each of these taxa you’ve reconstructed skulls for with extra bony nares, and each of these has a little fragment of the nasal bone dangling from the posterior nasoantorbital fenestra. Not one of these affirms your argument. This is compounded from the problem that you draw cracks as though they were bone margins. There is a difference between doubting the “modern paradigm” and “tilting at windmills” when it comes to reconstructing the skulls of pterosaurs when they are not in the best of conditions, but it gets worse when you apply this technique to skulls that are in ridiculously good condition, such as Anhanguara santanae, in which you place giant whopping holes in the rostrum where there are none.

    As I said before, you “find” little claws on fourth manual digits — they don’t exist, and centuries of other people pouring over these very slabs with microscopes cannot find what your poor-resolution photographs seem to pick up. You ignore pixellation artifacts, the result of resolution problems. Moreover, you “find” little fossil babies scattered on slabs (not in “eggs”) — these also do not exist.

    The problems stem from the “DGS” process — or rather, your idea that photographs can trump personal examination. Correct this process, then we can work on phylogeny.

    • Jaime,

      Let’s take this point by point.

      With pictures and responses.

      First: Show me with a tracing the two teeth you say are present in Jeholopterus. Send me any problems you have with DGS and we’ll look at them together. I don’t recall you doing that yet. Sorry. Getting old.

      Second: Proportion based characters are measureable and easy to see. They work.

      Third: Darren did a decent job of discussing other interpretations of my self-rejected work, and all preceding, which was his theme. Not much against the web-site as is. If you can find a decent criticism of the current website, let me know.

      Fourth: Send me your interpretations of cracks vs. sutures on pterosaur rostra.

      Fifth: Send me your distal wing tips that don’t have an ungual.

      Let’s have a real discussion with evidence.


      • Dave, you and I HAD that discussion years back about Jeholopterus ningchengensis, before you published your argument in Prehistoric Times, and before your SVP presentation. I disagreed with it then, and pointed out the distinction, when you brought this subject up on the Dinosaur Mailing List.

        Proportions are relative to ecomorphologic conditions. They are ridiculously plastic, and vary among even very closely related taxa. For example, just look at a range of analyses that have been performed on Cuban anoles, where populations of the same species will differ in proportions on the basis of how low or how high above sea level they are, how high up a tree they prefer to go, etc. This includes tail to SVL, limb size, relative width of the trunk, etc. The issue is not that proportions are BAD for analyses — I did not ask to to just blindly remove them — but that you have to find phylogenetically informative ones, those that do not tend to ecomorphism, such as limb proportions in aquatic reptiles, “long necks” with smaller, toothier heads, and so forth.

        You entirely misunderstand the concept of what I am going at. Your finding of frills and such wasn’t just “self-rejected,” it was heavily supported by your bias for it and your “technique” for years, and justly criticized by most of the peers to whom you looked for confirmation. NO ONE recovered the same structures … if any. Yet you use the same argument FOR the “technique” for all of these other extraneous elements: you say you find “sutures,” but they turn out to be cracks; you say you find missing bones, frills, invisible babies, but not retract those (but not very loudly); you say you find extra unguals on wing digits, but the margins are just blurs because they are a pixel or less wide, and thus wildly open to interpretation. It’s not the website, it’s the subject matter, and it’s the same for this as with “”, and it’s that which Darren was getting at — one and the same.

        I don’t need to send pictures. You need to stop relying on photos to overinterpret or counterinterpret with such unorthodox conclusions. You are free to use these photos to inquire, question and attempt gain answers by looking for more data — that’s why they’re published — but you use them as actual data in themselves, not the product of distortion, parallax, color effects, resolution, much less the problems (often unseeable) of things like finish, preservation degrees, damage and other taphonomic effects. Mickey Mortimer had to devote a whole blog post to correct you on the maxilla/premaxilla margin for Herrerasaurus ischigualastensis, and that was one small photo! The point is that you cannot rely on photos when you are trying to do primary research, it should always lead you to examination first and foremost. Such is Science when the subject is broken, damaged, and cracked fossils. We’ll ignore the bizarre notion of an external bony naris not in confluence with the nasal bones, especially when the posterior margin of the nares retract to rear of the antorbital fenestra and is reduced to a splint and unassociated with the maxilla in all sorts of Pterodactylus/Ctenochasma-grade pterosaurs, and that forcing some “hole” in the snout to suddenly accomodate the nostril when there is NO animal on this planet that even suggests this is likely from which to form a confirming argument. Even if there were “holes” in the maxilla/premaxilla of pterosaurs, there is NO reason to suspect first that they are even bony nares. You start with what is known, and you work from there; seek to disprove what is known, and try not to fail.

        NONE of your pterosaurs has a mdIV ungual. Why would I need to bring up a fossil that LACKS it in the specific either when it is YOU who attempts to disprove the “orthodoxy” with a flawed methodology? YOU are supposed to be the one producing the counter argument, and that argument is supposed to be SOUND. It ISN’T. Rather than perfect and make the methodology clear, including how you discriminate pareidolia, confirmation bias, and other error from it, you simply continue to “use” it. The problem lies even further: if someone does manage to prove lack of an ungual in ONE specimen, will you concede the methodology is flawed? or will you double down and simply argue either the person is mistaken, or JUST THAT ONE specimen lacks the ungual, perhaps merely preservational bias? Even intimating this question is like you saying “there’s two suns,” and I go “there’s only one sun!” and you go “prove to me there’s not two suns!” and no matter how many photos of just one sun I bring you, you say it doesn’t disconfirm your argument, there’s another sun not seen, those two suns are never in the sky together, they faked the photos, etc.

        Heads, you win; tails, I lose. That’s what it feels like.

  8. Jaime, you and Darren make it impossible to defend if you do not attack specifics. Get back to being a professional. Stop ranting and stop bringing up analogies about two suns. Find a taxon or two that does not belong where I recovered it in the reptile tree and tell me where it should go. Send some evidence, yes photographs, that defend your position. We can’t rely on what someone says they saw, we have to see the same evidence, in this case, photographs in lieu of having the specimen in front of both of us together. If you have evidence that I fake photos, make it public. And please make sure it’s from high rez originals from the last two years. When I found the premaxilla/maxilla suture on Herrerasaurus I supported it with similar morphologies on close kin. That validates it until new data says otherwise.

    And besides, the vast majority of my data comes straight from the literature. Your remarks about Cuban anoles are important and pertinent, but if something similar is present in my trees you need to point out exactly where. Otherwise your remarks strike at the hypothesis of maximum parsimony, which everyone supports.

    • If you really have redefined the premax/max contact – publish it. I know you say you are blacklisted, etc. Why not try anyway? At the least you’ll see what other reviewers have to say, maybe some unfamiliar with you.

    • One more very important point, Jaime… you mentioned the disutility of using bone proportions (I use phalanges, limbs, head/neck/ tail/ torso), If you’re hypothesis is correct, then how do you explain the recovery of two lineages of snakes in the large reptile tree, both apart from other legless lizards? Whatever I’m doing is working. If it’s so off target, it should be easy to find the target. If you don’t like it, don’t be afraid to provide evidence and specifics. Then we’ll have something concrete to talk about.

Leave a Reply

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

You are commenting using your 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.