Dinosaur family tree: Langer et al. responds to Baron et al. 2017 in Nature

Earlier
Baron et al. revised the dinosaur family tree by uniting Ornithischia with Theropoda to the exclusion of Herrerasaurus + Sauropodomorpha. Then Baron and Barrett 2017 moved Chilesaurus (Fig. 1) from Theropoda to Ornithischia, confirming the earlier hypothesis advanced here in 2015, but in the context of uniting Ornithischia with Sauropodomorpha (= Phytodinosauria) to the exclusion of Herrerasaurus + Theropoda.

Figure 1. Chilesaurus and kin, including Damonosaurus and basal phytodinosauria.

Figure 1. Chilesaurus and kin, including Damonosaurus and basal phytodinosauria.

Today
Langer et al. 2017 argue, “we evaluate and reanalyse the morphological dataset underpinning the proposal by Baron et al.5 and provide quantitative biogeographic analyses, which challenge the key results of their study by recovering a classical monophyletic Saurischia and a Gondwanan origin for dinosaurs. Our international consortium of early dinosaur evolution specialists has come together to critically assess the Baron et al.5 dataset.”

The Langer team recovered a traditional Saurischia/Ornithischia tree, but noted it would take only 2-3 additional steps to enforce a Sauorpodomorpha/ Ornithoscelida split, as recovered by the Baron team – this after changing 2,500 scorings (10% of the dataset). The Langer team also confirmed the origin of dinos in southern Pangaea and left with three conclusions (my comments follow):

  1. There is currently great uncertainty about early dinosaur relationships and the basic structure of the dinosaur family tree. Not in the large reptile tree (LRT, 1119 taxa)/
  2. Dataset construction is key. No, taxon inclusion is the key. Neither the Baron team nor the Langer team included the correct outgroup taxa nor a long list of basal dinosaur taxa (see below) that direct the tree topology toward the phytodinosaur clade.
  3. It is important to use appropriate computational analytical tools before making macro-evolutionary claims. No, taxon exclusion will lead to wrong results. Trait selection matters, but not as much. Scoring correctly matters, but not as much. Employing decades old software does not matter because the math and statistics are the same. Remember, only a poor workman blames his tools so don’t  blame the “computational analytical tools” for poor macro-evolutionary claims.

Bottom line:
The Langer team used the same incomplete taxon list as the Baron et al. team did. So they were looking for their ‘keys’ beneath the bright lamp, while the keys were lost in the dark alley they ignored.

This happens so often.

And
Baron et al. 2017 reply. “This  extensive re-scoring results in recovery of the ‘traditional’ topology, although with less resolution and very weak support; their result is statistically indistinguishable from the possibility that our topology provides a better explanation of the data. This weak support, despite these extensive changes, suggests that the ‘traditional’ tree struggles to account for many character distributions.”

And they disagree with many of the re-scorings. Their re-scoring of just Pisanosaurus reproduced the clade Ornithoscelida in their revised tree.

Both presented trees were poorly resolved.
The LRT is fully resolved. Baron et al. defended the possibility of a Northern origin for dinosaurs. That big ‘maybe’ does not follow the data in the LRT.

On a similar, but side note
Biology Letters was kind enough to publish my reply to the Baron and Barret 2017 paper on Chilesaurus, but much of it has bearing for today’s discussion. Here is that letter in its entirety:

Baron and Bennett [1] nest Chilesaurus [2] as the sister group to Ornithischia, rather than a tetaneuran theropod as previously proposed [2]. Unfortunately, the Baron and Bennett [1] taxon list, like the Novas et al. [2] taxon list before it, did not include many of the taxa essential to resolve this issue.

A larger study of over 1060 taxa [3] includes more taxa essential to resolve this issue. The matrix was created using MacClade [4]. Analyses were run in PAUP 4.0b10 [5] using a heuristic search and a Bootstrap/Jackknife search for 100 random addition replicates. Scores are indicated on the webpage.

On that cladogram Chilesaurus (Late Jurassic) nests as a basal ornithischian in a clade that also includes Daemonosaurus (Late Triassic) and Jeholosaurus (Early Cretaceous). The latter two taxa were not included in Baron and Bennett [1]. This clade of three taxa nested as a sister to the Sauropodomorpha with Leyesaurus at its base. The analysis recovered the clade Phytodinosauria as the sister taxa to the Theropoda. Herrerasaurus was the outgroup to these two clades as basalmost member of the Dinosauria. Basal phytodinosaurs not nesting within Sauropodomorpha + Ornithischia include Barberenasuchus, Eodromaeus, Eoraptor and Pampadromaeus.

On that cladogram Silesaurus nests within a clade Poposauridae outside the Archosauria. The clade Archosauria includes only the Crocodylomorpha + the Dinosauria. Lagerpeton nests with Tropidosuchus and other proterochampsids. The pterosaur, Dimorphodon, nests with lepidosaurs like Huehuecuetzpalli, Macrocnemus and Cosesaurus (the last of which had an antorbital fenestra by convergence [6, 7]). None of these are archosauriformes nor prolacertiformes [contra 7]. The following pertinent taxa were not included in Baron and Bennett [1]: Daemonosaurus, Jeholosaurus, Haya, Barberenasuchus, Buriolestes, Segisaurus, Procompsognathus, PVL 4597, Junggarsuchus, Pseudhesperosuchus, Carnufex, Trialestes, Gracilisuchus, Scleromochlus, Decuriasuchus, Turfanosuchus, Poposaurus, Lotosaurus, Shuvosaurus, Effigia and Tropidosuchus.

Chilesaurus was first nested as a basal ornithischian in April 2015 [8] in an earlier version of the above analysis, then with fewer taxa. With the addition of more pertinent taxa the position of Chilesaurus is indeed well resolved contra the previous e-letter [9].

1. Baron MG and Barrett PM 2017. A dinosaur missing-link? Chilesaurus and the early evolution of ornithischian dinosaurs. Biology Letters 13, 20170220.
2. Novas FE et al. 2015. An enigmatic plant-eating theropod from the Late Jurassic period of Chile. Nature 522(7556), 331.
3. http://www.ReptileEvolution.com/reptile-tree.htm .nex file link on that webpage
4. Maddison DR., & Maddison WP 2001 MacClade 4.08: Analysis of Phylogeny and Character Evolution. Version 4.03. Sinauer Associates.
5. Swofford D 2002 PAUP 4.0 b10: Phylogenetic analysis using parsimony. Sinauer Associates.
6. Ellenberger P. and de Villalta JF 1974. Sur la presence d’un ancêtre probable des oiseaux dans le Muschelkalk supérieure de Catalogne (Espagne). Note preliminaire. Acta Geologica Hispanica 9, 162-168.
7. Peters D 2000. A Redescription of Four Prolacertiform Genera and Implications for Pterosaur Phylogenesis. Rivista Italiana di Paleontologia e Stratigrafia 106 (3): 293–336.
8.https://pterosaurheresies.wordpress.com/2015/04/28/chilesaurus-new-dinos…
9. King B 2017. Chilesaurus is not a basal ornithischian. http://rsbl.royalsocietypublishing.org/content/13/8/20170220.e-letters

References
Baron MG and Barrett PM 2017. A dinosaur missing-link? Chilesaurus and the early evolution of ornithischian dinosaurs. Biology Letters 13, 20170220.
Baron MG, Norman DB and Barrett PM 2017.
A new hypothesis of dinosaur relationships and early dinosaur evolution. Nature 543: 501–506;  doi:10.1038/nature21700
Baron MG, Norman DB and Barrett PM 2017. Baron et al. reply. Nature 551: doi:10.1038/nature24012
Langer et al. (8 co-authors) 2017. Untangling the dinosaur family tree. Nature 551: doi:10.1038/nature24011

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 )

Google photo

You are commenting using your Google 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.