The TR-J Terrestrial Extinction Actually Early Jurassic?

Adam Yates most recent post over at Dracovenator and a new abstract by Zeigler and Geissman has got me thinking more about faunal transitions between the Late Triassic and Middle Jurassic. As I stated in an earlier post, Chinle Formation faunal composition remains relatively consistent from the oldest to youngest localities and it is not until you get into the uppermost units of the formation and higher that you start to see some changes. Lucas and Tanner (2007) provides a good documentation of faunal change in the western U.S.A. through this interval and demonstrates that the lowermost Dinosaur Canyon Member (Moenave Formation) and the basal portion of the Wingate Sandstone (both units previously argued to be Jurassic in age and in the Glen Canyon Group) are most likely latest Triassic in age. This is based several lines of evidence including magnetostratigraphy, lithostratigraphic correlation, and biostratigraphy (the presence of phytosaur body fossils and pseudosuchian trace fossils). The upper Moenave, upper Wingate, and the Kayenta Formation lack these fossils. In addition, Lucas and Tanner (2007) place the youngest known Chinle Formation fossil assemblage (the Ghost Ranch Coelophysis Quarry) in the Rock Point Member, which they consider to be laterally equivalent to the base of the Wingate and the lower Dinosaur Canyon Member. They also consider this assemblage to be latest Norian in age based on palynology and the presence of the aetosaur Aetosaurus.

Zeigler and Geissman (2008) argue that based on magnetostratigraphy that the Ghost Ranch Coelophysis Quarry is not in the Rock Point and that it may be even younger than previously supposed. As I have noted previously, Zeigler (2008) correlates the site (using magnetostratigraphy) with the lower Moenave and now Zeigler and Geissman (2008) suggest that the uppermost Chinle Formation is at least Rhaetian and may even be Hettangian in age! This would extend the range of phytosaurs and other non-crocodylomorph pseudosuchians into the Early Jurassic. Thus there would be no terrestrial Triassic/Jurassic extinction, at least not in western North America.

Furthermore, Adam Yates recent post suggests that there may have been an end Early Jurassic extinction that spelled the end of coelophysoids and basal sauropodomorphs, followed by the rise of tetanurans and eusauropods in the Middle Jurassic. If Zeigler and Geissman and Yates are correct there would have been two major faunal turnovers in the very short period of time (approx. 30 million years) encompassing the Early Jurassic. In the earliest Jurassic we would see the disappearance of non-crocodylomorph pseudosuchians and the rise of a dinosaur dominated fauna, including the first basal sauropodomorphs in North America (which are not found in the Late Triassic of that continent*). Approximately 24 million years later we get the Early-Middle Jurassic turnover discussed by Yates and an explosion in dinosaurian diversity. Very interesting and the reason why research on the vertebrate fossil record of the lower Glen Canyon Group in becoming very important and needs to be expanded.

*Note: the only purported evidence of Late Triassic sauropodomorphs in North America are the ichnotaxa Tetrasauropus and Pseudotetrasauropus (e.g., Lucas and Tanner, 2007); however, Rainforth (2003) has determined that these taxa probably represent tracks made by pseudosuchians.


Lucas, S.G., and L.H. Tanner. 2007. Tetrapod biostratigraphy and biochronology of the Triassic–Jurassic transition on the southern Colorado Plateau, USA. Palaeogeography, Palaeoclimatology, Palaeoecology 244:242–256.

Rainforth, E.C. 2003. Revision and re-evaluation of the Early Jurassic dinosaurian ichnogenus Otozoum. Palaeontology 46, 803–838.

Zeigler, K.E., and J.W. Geissman. 2008. Magnetostratigraphy of the Upper Triassic Chinle Group and Implications for the Age and Correlation of Upper Triassic Strata in North America. Geological Society of America Abstracts with programs (online).


  1. Perhaps this mass extinction was also partly responsible for the rise of eurypod thyreophorans and neornithischians. :-)

  2. Soooo...what-if the opposite? The Tr-J Mass Extinction was more of a nonevent? It could be that there were multiple places that faunal turnover at different rates and at different times. Didn't they JUST find a bunch of dinosauromorphs that survived much later than they thought at Ghost Ranch (Hayden Quarry) supposedly late triassic locale?

    That would run contrary to the biostrategraphic approach, but...I have to wonder. Or perhaps we were witnessing a gradual faunal turnover, but because of sampling biases, we thought we had a mass extinction on our hands?

    Your explanation is more parsimonious, but...just something to test.

  3. I think the Zeigler & Geissman data is very interesting, but it only demonstrates that the a variety of uppermost Chinle units are not correlative. Vertebrate biostratigraphy is simply not precise enough to correlate magnetostratigraphic records.

    You can always slide magnetostratigraphic records up and down relative to each other and come up with a best fit that looks good. But this is mostly subjective unless you have precise geochronologic constraints to tie them together. The thicknesses of each chron are largely meaningless unless you can correct for basin fill rates for every record.

  4. Thanks for the excellent comments. Will and 220mya are correct, this is all very tentative, something I quickly concluded after reading Adam Yate's post and contemplating the Zeigler and Geissman abstract. However, the point is that when all of these units are the future focus of more detailed research (especially geochronological) we may get some surprising results.

  5. Bill - you are definitely correct. We have alot to learn about the precise age and correlation of various Chinle Formation units. In fact, I would say that is the case for most early Mesozoic terrestrial assemblages around the world.


Markup Key:
- <b>bold</b> = bold
- <i>italic</i> = italic
- <a href="">FoS</a> = FoS