A critical evaluation of Tianyulong confiusci – part 2

Facebooktwittergoogle_plusredditpinterestlinkedintumblrmailby feather

Continuing from yesterday, the following is what I garnered from the Tianyulong confiusci specimen announced last week.

When looking at the fossil, a couple preliminary questions came to mind.

  1. Is the fossil real?

  2. Is the integument real feathers/protofeathers?

Is the fossil real?

Main slab for _Tianyulong_ with highlights showing the breakage in the slab
Main slab for Tianyulong with highlights showing the breakage in the slab. Click the image to enlarge.

The initial paper gives no mention of how the fossil was collected (i.e. if it was collected by local farmers – as most of these fossils are – or if it was found in the field), so it is hard to tell how many hands this fossil has passed through before it was described. The specimen is broken into at least 3 different slabs (as shown in the first pic. Highlights [mine] show where breaks occur). The first, and most obvious, is across the top of the body, separating the dorsal vertebrae from the rest of the fossil. The second break, is a little less obvious. It appears to neatly separate the anterior part of the body, from the posterior part (pretty much right before the hip). It’s hard to tell from the photos, but this section might have been glued together. Whether this was before it reached the scientists, or after is left unclear. So there is room for suspicion there. The characters used to determine heterodontosaurid affinities come exclusively from the skull. The preservation of the hip makes it very hard to tell what one is looking at. The ischium appears quite a bit thicker than in Heterodontosaurus tucki. This could be chalked up to generic difference, or even an ontogenetic one. The authors mention the presence of extensive ossified tendons on both dorsal and ventral sides of the caudal vertebrae. This is actually unusual for an ornithopod. Ossified tendons tend to be arranged in a lattice-like geometry throughout the dorsal portions of the caudal verts, but not the ventral side. Tianyulong not only has ossified tendons on both dorsal and ventral sides of the caudals, but they are arranged in a parallel fashion rather than the more typical lattice work. This sounds much more like what one would expect to see in a dromaeosaur, not a heterodontosaurid. Especially since the eponymous Heterodontosaurus lacked ossified tendons. This would make this tendon arrangement both unique for heterodontosaurs, and unique for ornithopods.

Incidentally, there is yet another crack that separates this section of the tail from that of the proximal (and apparently tendonless) portion of the tail. It doesn’t look like the crack goes all the way through the slab, but this can’t be verified from the photos. Nonetheless, this is yet another cause for skepticism.

Another bit of strangeness is the presence of an apparent stain along most of the skeleton. It appears as a lighter, white colour, and is found within the body cavity, and along the back and tail. This might have been caused by the dissolving of the soft tissue. Whatever it is, this stain cuts off all the apparent filaments from the rest of the skeleton (save one small section that will be described later). In fact, there is one part where the stain appears to cut ? rather sharply ? right through the tail filaments. This cut is at an angle to the tail, thus not following the body contour at all. In fact, it almost looks like a deep gouge like that caused by a shovel, or (in this case) a trowel. Perhaps this was a casualty of the preparation/excavation.

After looking the fossil pictures over, I have to say that Tianyulong more than any other “feathered dinosaur” before it, has the potential to be a chimera.

Is the integument protofeathers/feathers?

Well, the answer is an emphatic no to the latter. These are definitely not feathers.

So then are they protofeathers?

In the paper, Zheng et al mentioned that the filaments bear a similarity to both the “quills” on Psittacosaurus , and the protofeathers of Sinosauropteryx. Curious; I decided to compare the three.

Comparison of the _Psittacosaurus_ "quills" (top), _Sinosauropteryx_ protofeathers (left) and _Tianyulong_ "filaments" (right)
Comparison of the Psittacosaurus “quills” (top), Sinosauropteryx protofeathers (left) and Tianyulong “filaments” (right). Click the image to enlarge.

Right off the bat, I’d say one can dismiss any real relationship to the protofeathers of Sinosauropteryx. The filaments on Tianyulong are similar only in the sense that they don’t branch at all. Short of that, the size, and density of Tianyulong‘s filaments are quite different from those of S.prima (being wider, longer and more loosely packed).

When compared to the “quilled” Psittacosaurus, a much greater similarity can be seen as both filaments are rather long. The Psittacosaurus “quills” however, are quite a bit thicker, and seem to show up within the skin, while Tianyulong‘s filaments don’t touch the skeleton at all, save for the same spot where the strange (possible) groove is found.

Some folks have stated that the large filaments are focused on the caudal portion of the body, just like in the “quilled” Psittacosaurus specimen. I would caution against this. Most of Tianyulong‘s body is not preserved. Unlike the Psittacosaurus specimen, where one could tell that these “quills” appeared only on the tail, there is very little evidence for the same arrangement in Tianyulong. I would extend this caution to statements about Tianyulong being completely fuzzy too. There are some filaments found by the dorsal vertebrae and under the cervicals. However, these filaments are much removed from the body. The dorsal patch does not follow the arch of the vertebrae; instead lying more anterior to the bones. As for the ventral patch, unless one wants to posit a double chin on Tianyulong, they also don’t actually associate with the bones, nor do they follow the body contour.

The caudal filaments are strange in their own right. Like all the rest of these filaments they don’t follow the body contour (compare, for instance how the protofeathers of Sinosauropteryx follow the body rather tightly). In fact many of these filaments seem to be tangled amongst each other.

Note there is yet another apparent break in the slab, between the filaments.

If everything is arranged correctly, then these filaments seem to be tangling up with filaments that would have emerged much further up the back. Also unlike the singular “quills” on the Psittacosaurus, these thinner filaments all appear to protrude from the same narrow area. Instead of being more evenly spaced along the caudal vertebrae, they all bunch up by the proximal caudals. If these filaments did belong to the living animal, then it would appear that Tianyulong was brandishing a “smokestack haircut” long before Kid from Kid and Play ever did.

Final verdict:

Readers will no doubt have noted my extensive use of quotes around certain instances of protofeathers, as well as the mention of quills in the infamous Psittacosaurus specimen. I do so because of the questionable assignment of these filaments to those particular structures. In doing so, I am following in the steps of David Hone, who also suggested that one be cautious with one’s interpretation of some of these Yixian fossils (though my view is a little more extreme). Many of them have been described briefly, with little follow up work. The Psittacosaurus with the “quills” is a particularly nasty case. It received a quick right up in Nature, before it was discovered that the specimen was illegally collected. Now there is a veritable “shit storm” surrounding the fossil. This has resulted in it becoming a pariah that no journal dare touch. A result that has essentially put a halt to any further research for now. It’s unfortunate, as the identity of the Psittacosaurus filaments remains in limbo (not everyone is “happy” with the diagnoses of quills).

As for Tianyulong, there appears to be a fair amount of evidence to suggest the animal might have died on a plant, or was possibly being devoured by nematode like parasites prior to death. As for being protofeathers, they appear as unlikely in Tianyulong, as they do in Psittacosaurus. The relationship to the protofeathers of Sinosauropteryx prima, appear to be at the most basal geometric level (i.e. they are both straight and unbranching).

Still, what if everything is genuine? What kind of implications would that hold for dinosaurs?

For my answer to that, stay tuned.


Facebooktwittergoogle_plusredditpinterestlinkedintumblrmailby feather

3 Responses to A critical evaluation of Tianyulong confiusci – part 2

  1. I think you are being unduly harsh here. You seem to imply that a group of professional researchers (including one involved in the exposure of Archaeoraptor) never even considered that it might be a fake. You never write in a paper “we carefully examined the specimen, its preservation, taphonomy, anatomy and history to make sure it was not a fake”, you only mention problems, not a lakc of them.

    The preservation of many of these specimens is typified by breaks, cracks, missing pieces, crushing and variations in preservation quality – thinks like the absence of ossiified tendons in one part only tells us that they are missing. One does not have the space in a Nature paper to discuss things in detail and these are quite probably absent through destruction or lack of preservation and might (I have not seen the specimen, only photos) be eaily noted through marks on the remaining vertebrae. Even so, dromaeosaurs do not have ossified tendons on the tail, but elongate chevrons and zygopohoyses and these are easy to tell aparp, especailly when the rest of the tail is still there. The skull might be the only think to diagnose it as a heterodontosaurid specifically, but the pelvis, vertebrae and pes are all characteristic of ornithischians.

    In short, being sceptical is good, but I think it’s a mistake to try and second guess experience researchers who have been wroking directly on the fossil material. They do know their stuff and people are wary of fakes.

  2. The need for brevity in a Nature paper makes sense. It would have been nice though, if the researchers had put some authenticity testing information in somewhere like the supplementary material.

  3. I’ll keep this short as i have aa blog post prepared on this very subject, but in general fakes are easy to spot and the kinds of authentication various people have been suggesting online are either impractical or impossible. It is unrealistic to expect people to move valuable and fragile specimens to a distanct location to have them CAT scanned say, and the money has to be found to pay for this proceedure. You simply can’t treat every fossil the way you would like.