Rethinking Turtle Origins - Odontochelys

ResearchBlogging.org

What an exciting time it must be to work on basal turtles. Hot on the heels of Chinlechelys and Eileanchelys comes a new basal turtle, Odontochelys semitestacea, from the Norian (~220 Ma) of China. The article and commentary came out today in the journal Nature and you can also read about it here and here.


Odontochelys predates other earliest known turtles by at least 5 million years and is from marine rocks suggesting a marine origin for turtles. The complete fossil (see these photos from the Nature News website), known from four specimens, possesses a ventral plastron but not a dorsal carapace suggesting that the plastron formed first (but see argument by Reisz and Head, 2008).

Finally, as also hinted at by the name, Odontochelys is the first known turtle to possess teeth.

Whereas, the marine origin is surprising given the terrestrial nature of other Triassic turtles overturning what was thought to be a stable hypothesis, the teeth are a nice find, but not so surprising given that turtles must have originated from a toothed reptilian ancestor.

As to the ventral carapace forming first, Reisz and Head (2008) argue against the interpretation provided by Li et al. (2008); however, based on a statement made by Li that "here, in our hands, there is an ideal missing link for turtle evolution. It has no osteoderms on its back, but only ossified neural [central] plates and expanded ribs." I wonder if what we are looking at is a preservational artifact. Maybe the plastron fully ossifies earlier on in ontogeny. Not being a turtle specialist or seeing the specimens I may be completely wrong, but that is the explanation that first popped into my head.

Sounds like there is a lot more exploration and work to be done on basal turtles.

REFERENCES

Chun Li, Xiao-Chun Wu, Olivier Rieppel, Li-Ting Wang, Li-Jun Zhao (2008). An ancestral turtle from the Late Triassic of southwestern China Nature, 456 (7221), 497-501 DOI: 10.1038/nature07533

Robert R. Reisz, Jason J. Head (2008). Palaeontology: Turtle origins out to sea Nature, 456 (7221), 450-451 DOI: 10.1038/456450a

9 comments:

  1. Bill...I don't know how I missed this! If you have access to the paper, could you send it to me? This is...incredible!

    ReplyDelete
  2. Indeed, the plastron apparently ossifies earlier than the carapace in living turtles. Also--minor quibble-- Proganochelys and Kayentachelys have palatal dentition right? So Odontochelys is technically the first known turtle with marginal dentition....

    ReplyDelete
  3. Neil,

    You are correct Odontochelys is the first with marginal dentition.

    Bill

    ReplyDelete
  4. I disagree with the idea that there's no carapace. First, as the paper explains in detail, the neural plates are there, apparently sutured to each other or anyway making the back immobile; it also explains that they are separate bones, not outgrowths of the vertebrae. Second, the paper claims that costal plates are absent and instead the ribs are broadened. Judging from the photos and the interpretative drawings, the costal plates are present and (sutured or more probably) fused to the ribs -- they just aren't sutured to each other, which is why the ribcage/carapace was able to disarticulate. The paper says that in extant turtles the costals form as outgrowths of the ribs (though by intramembraneous rather than endochondral ossification), rather than as separate bones that later fuse to the ribs, but a few sentences before they explain that the neurals form as outgrowths of the vertebrae in some extant turtles but not in others! I conclude that ontogeny can evolve. I think the carapace is there -- or rather its parts are.

    (Ironically, the dwarf pareiasaur Pumiliopareia really does have broadened ribs, as opposed to normal ribs fused to costal plates, as far as I can tell from a drawing in lateral view.)

    It is true that the peripheral, nuchal & pygal plates are absent. But that really could be a reduction as suggested by Reisz & Head.

    Keep in mind that the plesiomorphic lifestyle of turtles now optimizes as ambiguous: aquatic on the one side (Odontochelys), terrestrial on the other (Chinlechelys, Proganochelys, Proterochersis). One or the other must be apomorphic, but we can't simply assume that that must be the terrestrial one just because Odontochelys is older.

    Much more discussion over at Pharyngula.

    BTW, could someone please convince Adam Yates to allow name/URL comments at Dracovenator? I'd love to comment there, but I'm certainly not going to the trouble of getting a Google/Blogger ID or an OpenID just for that purpose.

    ReplyDelete
  5. I think the Reisz & Head hypothesis is the most likely, but it awaits more fossils. Just because you find some specimens in a marine depositional environment doesn't mean they lived there (Ticinosuchus and Compsognathus anyone?)

    One minor correction - you state in your first paragraph that the Norian is ~200 Ma. This is actually the Triassic-Jurassic boundary, and not in the Norian stage. The dating in the paper is interesting. They say the fossils are lower Carnian in age, but also have said they are ~220 Ma. This does not fit. The fossils are biostratigraphically dated, so I'd go with the qualitative "lower Carnian" age. Based on the data of Furin et al. (2006), this would make the fossils ~235-233 Ma.

    ReplyDelete
  6. Oops... I did mean to write 220 Ma.

    Thanks for catching that.

    David,
    I would have to agree. I don't believe that the dorsal carapace is simply non-existant.

    ReplyDelete
  7. I think that the "220 Ma" is in error rather than the "lower Carnian" part. The authors state the age is "late Carnian" based on biostratigraphically significant marine fossils (invertebrates and microfossils). Given that the stages of the Triassic are defined on marine microfossils, I think its likely that the strata are truly late Carnian. Its just that they used an obsolete timescale to convert that to a numerical date for the press release.

    ReplyDelete
  8. I think that you are correct. That is what happens when you base your blog post mainly on the press release material.

    They did state that is predated other finds by 15 million years and that would also suggest a date around 230 MA which is the Late Carnian. I had modified this to 5 million years difference in my post correcting what I thought was an error, but I did not correct the right error.

    ReplyDelete
  9. The ages of the southern China Triassic vertebrate faunas has been a source of contention and confusion for awhile (I know: confusion about the age of a Triassic vertebrate fauna--shocker). And matters aren't helped any by inconsistent stratigraphic nomenclature.

    "Lower Carnian" appears to be the emerging consensus, based on conodonts and ammonites, but apparently the bivalves suggest an "Upper Ladinian" age.

    As far as the environmental preferences of Odontochelys go, it's interesting that the Falang biota lacks any terrestrial vertebrates unlike some of the Triassic faunas in the area. On the other hand wood is relatively abundant, although generally well encrusted by crinoids.

    ReplyDelete

Markup Key:
- <b>bold</b> = bold
- <i>italic</i> = italic
- <a href="http://www.fieldofscience.com/">FoS</a> = FoS