Lesothosaurus is a rhynchosaur…Henodus is a turtle…and other misfits by default nesting.

I have often noted the “by default” nesting of pterosaurs and Vancleavea in the archosaurs, when the larger gamut study indicates they nest elsewhere, with lizards and thalattosaurs, respectively. I can speak with authority here because the large reptile tree represents the only large test of these misfit nestings based on smaller studies that excluded the actual related forms.

When you have a very large gamut family tree with full resolution,
you can play with it to your heart’s content. 
Earlier I removed all lepidosauromorphs – but turtles and pterosaurs – and noted that they nested together within the Enaliosauria, the marine archosauromorphs. Turtles and pterosaurs??? This odd bit of nesting should have been taken:
1) as a lesson in trying to shoehorn in taxa that clearly do not belong together, like pterosaurs and archosaurs. Yet, given the opportunity to nest with dinosaurs or archosaurs, by virtue of eliminating all fenestrasaurs, all tritosaurs and all lepidosaurs, pterosaurs STILL went with their closest relatives, according tothe large reptile tree: turtles.
2) and to add insult to injury, within the new Archosauromorpha, pterosaurs nested with pachypleurosaurs, a marine taxa far from the archosaurs that has NEVER been under consideration before in traditional studies. It took several rounds of elimination to finally nest pterosaurs with archosaurs. That’s how bad that mismatch really is.
Here I’ll take the misfits the other way,
by removing all archosauromorphs, but one, and seeing how they nest within the Lepidosauromorpha. You might find these amusing and instructive in light of the current nesting of “strange bedfellows” discussed earlier.
Lesothosaurus, the ornithischian, nests as a sister to Hyperodapedon, the rhynchosaur.
Herrerasaurus, the theropod, nests between Trilophosaurus and Mesosuchus.
Dimetrodon, the synapsid, nests at the base of Diadectes and Orobates, and far from the putatitve synapsids, now shown to be closer to Milleretta the lepidosauromorph: Casea and Cotylorhynchus.
Plesiosaurus, the plesiosaur, nests between Adriosaurus and Boa, the pre-snake and snake.
Henodus, the placodont, nests with Proganochelys, the turtle, even without any carapace or plastron characters in the matrix.
Obviously all these nestings are bogus,
but you wouldn’t know that unless you realized they nested with widely known excluded taxa. Some nestings, like Caseasauria with Synapsidsida and mesosaurs with pareiasaurs are also widely known and accepted, but they’re wrong. They cannot be confirmed with the large reptile tree.
These experiments show how wrong things can go when you try to mix pterosaurs with archosaurs, etc. etc. etc.
Sorry to keep harping on a sour note,
but a whole raft of professional paleontologists really needs to forget tradition and start testing for ALL possibilities before assuming an inclusion set is valid. Otherwise, all you get are the “strange bedfellows” we discussed earlier in a 9-part series starting here. This professional quagmire really needs to come to an end.
Thanks to TK for suggesting something like these experiments in phylogeny.

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.

3 thoughts on “Lesothosaurus is a rhynchosaur…Henodus is a turtle…and other misfits by default nesting.

  1. How about an alternative hypothesis- your particular analysis is too poorly constructed to nest these taxa correctly when taxon selection changes. Maybe if you had characters correctly ordered and included more of them, your lone archosauromorph would nest sister to Lepidosauromorpha instead of deeply within it.

    Ordering is very important here. Take your character 144- “Number of sacral vertebrae: (0) two; (1) 3 or 4; (2) one or none; (3) 5 or more.” The cost for putting a coelurosaur with five sacral vertebrae inside some clade like snakes with none or one sacrals is one step. Let’s rearrange it to “Number of sacral vertebrae: (0) one or none; (1) two; (2) 3 or 4; (3) 5 or more” and make it ordered. Now, placing the coelurosaur inside snakes is three steps, since you have to go through states 1 and 2 to get to state 0. And if you had a better analysis that also separated taxa with no sacrals from taxa with one sacral, and taxa with three sacrals from taxa with four sacrals, it would take five steps. Since you have numerous multistate characters, and none are ordered, these situations could add up fast.

    Now I agree taxon inclusion is important and may be leading to some errors in traditional phylogenies, but your tests here do not demonstrate that. How would you differentiate your hypothesis that taxon inclusion leads to inaccurate results in a good analysis, with mine that taxon inclusion leads to inaccurate results in your analysis because it is not good, only by using your analysis? You can’t without fixing your analysis and testing again.

    Finally, some of your pairings are plausible so don’t help make your point.

    Excluding the controversial pterosaurs and turtles, the trilophosaur-rhynchosaur clade are the only lepidosauromorphs in your tree with most paleontologists think are actually archosauromorphs. So that Lesothosaurus and Hererrasaurus would group with them seems logical.

    Similarly, synapsids have been suggested to be sister to diadectomorphs by some, so finding Dimetrodon by Diadectes and Orobates makes sense. Sure nesting by caseids would be a bit better, but still the result isn’t bad.

    And placodonts have nested close to turtles in several runs of Reippel and de Braga’s matrix, so that’s no big surprise either.

    I’d be interested in a list of which characters make turtles nest with pterosaurs. That could lead to insights on why these pairings are occuring.

  2. Of course dinos nest with rhynchosaurs. That was expected given Carrolll 1988 based on Benton, not sure what year. All pre-computer. But dinos nest better with archosaurs. And rhynchosaurs nest better with sphenodontians when the opportunity is presented. That’s the whole point. With regard to turtles and pterosaurs, you’re welcome to take the matrix, delete the lepidosaurs (sans ptero and turtle) and see for yourself. No doubt its not one character or two, but the weighting of a suite of traits.

  3. I think your repeated assertion that palaeontologists don’t test the relationships you suggest is a bit…well its just not true. Admittedly some havn’t been tested e.g. Tetraceratops (never tested outside synapsids), but a great many have. As just one example Hill (2005) uses a (mostly) genus-level taxon list which covers 80 taxa from almost all the major groups in Amniota with charater list almost 3 times the size of yours to show that Caseids do not go with Milleretids and Bolosaurids, Rhynchosaurs do not go with Rhynchocephalians, Ophiacodontids are not the sister to Therapsids, Synapsids do not go with Archosaurs, Captorhinids do not go with Lepidosaurs, Mesosaurids do not go with marine reptiles…I could go on.

    These relationships arn’t widely accepted because they’re ‘Traditional’. They’re accepted because they have been shown so convincingly to be true by modern analyses. Every relationship that is widely accepted is supported by a much longer list of characters than than any list which you produce trying to disproove it. And yet every time someone points this out to you, you just say ‘Someone else can add these characters to test as inclusive as mine’. Firstly, this is not an answer, its an evasion. Secondly, Hill has. And for the most part, he came up with the widely accepted relationships.

    Hill, R. V. 2005. Integration of Morphological Data Sets for Phylogenetic Analysis of Amniota: The Importance of Integumentary Characters and Increased Taxonomic Sampling. Sytsematic Biology (54): 530-547

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.