A recent paper (Liu et al. 2015) on a new specimen (BGPDB-R0001) of the basalmost placodont, Diandongosaurus, (IVPP V 17761), brings up the twin problems of taxon inclusion/exclusion without the benefit of a large gamut cladogram, like the large reptile tree (580 taxa) to more confidently determine inclusion sets in smaller more focused studies (anything under 100 taxa).
Let’s start by making the large reptile tree go bad.
Liu et al. used a traditional inclusion set (Fig. 1 on left) that included suprageneric taxa and taxa that were unrelated to one another in the large reptile tree (Fig. 1 on right). To illustrate inherent problems, I reduced the taxon list of the large reptile tree to closely match that of Liu et al. See them both here (Fig.1).
Although many taxa on the left and right of figure one are similar, many nest differently.
Let’s start with the problems
in the cladogram on the right, in the reptileevolution.com incomplete cladogram
- Prolacerta nests basal to squamates (Iguana) and Triassic gliders (Kuehneosaurus).
- Trilophosaurus nests between squamates and rhynchocephalians.
- Turtles nest with archosauriforms and both close to rhynchosaurs, none of which are related to each other in the large reptile tree. This is the wet dream of many turtle workers intent on matching DNA studies that place turtles with archosaurs, a clear case of DNA not matching morphology.
is basically in the correct topology, remarkable given that 540 or so taxa are missing.
The problems in the cladogram on the left,
from Liu et al include:
- Turtles nest between Triassic gliders and placodonts (and not the shelled ones proximally). This is Rieppel’s insistence on a force fit. Is the insertion of turtles the reason for other tree topology disturbances here and on the right? Not sure…
- Hanosaurus, a derived pachypleurosaur close to nothosaurs nests with Wumengosaurus, a pachypleurosaur/stem ichthyosaur.
- Liu et al. nested Diandongosaurus with headless Majiashanosaurus (which is correct) but then nests both at the base of the nothosaurs (which is not validated by the large reptile tree). The large reptile tree nested Diandongosaurus at the base of the placodonts, between Anarorosaurus and Palatodonta + Majiashanosaurus. Shifting Diandongosaurus to the base of the nothosaurs adds 32 steps to the large reptile tree.
Perhaps what the Liu et al team need is a subset of the large reptile tree. That would help them drop those turtles from placodont studies. They don’t belong. When cladograms go bad, sometimes there are included taxa that should not be there. Colleagues, make sure to check your recovered sister taxa to make sure they look like they could be sister taxa. After all, evolution is about slow changes over time.
Liu X-Q, Lin W-B, Rieppel O, Sun Z-Y, Li Z-G, Lu H and Jiang D-Y 2015. A new specimen of Diandongosaurus acutidentatus (Sauropterygia) from the Middle Triassic of Yunnan, China. Vertebrata PalAsiatica. Online Publication.