A Brief Moment in Kulinda

Not that long ago, a new fossil locality was discovered in Kulinda, Chita (Chininskaya Oblast), Russia, and it has the potential to confirm a recent hypothesis: that the filamentous integument of many theropods, found apparently in some ceratopsian dinosaurs, may be the base state from which all dinosaurs arose. Moreover, it would strengthen the argument that “dinofuzz” may be the same as pterosaur integument (“pterofuzz,” called pycnofibres). Thus, animals like lagerpetontids, silesaurids, even tiny, scaly Scleromochlus taylori, might be fuzzy. A new dinosaur from Kulinda represents a bold entry in the debate over dinosaur integument, a popular topic on this blog and one that evokes strong emotional reactions from many.

The KO, as Kulindadromeus zabaikalicus, as illustrated by Andrey Atuchin (2012). Used in promotion of the description by Godefroit et al. (2014).

The locality of Kulinda in Chita, Russia, has produced hundreds of bits of leg, arm, skull, back, tail since it was discovered in 2010. Many of the bits imply a single animal, a small ornithischians about 1.2-1.5 meters long. They speak of a ornithischian whose anatomy presents it as an early offshoot, arising not long after the ornithischians split off from the rest of the dinosaurs. It’s very old, in relative terms; the Ukureyskaya Formation has been dated to between the Middle and Late Jurassic (Sinitsa, 2011), which places it amongst some of the oldest ornithischians (one of the earliest offshoots, heterodontosaurids, are known from older beds, and “fabrosaurs” as well, but their integumental anatomy is less well-known). The heterodontosaurid Tianyulong confuciusi is also known from fuzzy specimens, however these remains were from younger specimens (in the late Early Cretaceous) and there was always a doubt about how filaments might persist. Might they merely be convergent?

There are always doubts. Some argued that these integumental structures were intradermal, not extradermal, being comprised of structural collagen. These “some” are few in number, but regardless of their fewness they are nevertheless loud. Through this discovery, they (including Theagarten Lingham-Soliar) made efforts to decry these structures as features within the skin and akin to ichthyosaurs and other animals which lacks filamentous integument. But they have been mostly occupied by Sinosauropteryx, the great boogeydino. This would change in 2014.

An international team led by Royal Belgian Institute’s Pascal Godefroit and Institute of Natural Resources, Ecology, and Cryology’s Sofia Sinitsa, set out to describe these remains, many of them fragmentary and partial and none complete enough to offer a whole animal. In 2013, the paleo world was tantalized by the subject of this research, which would be revealed at the 2013 meeting of the Society of Vertebrate Paleontology in Los Angeles, California, USA. At the time, images floated online along with a tantalizing name, “Kulindadromeus.” The images were hard to make out, but they implies that there were some unusual features of the tail and legs. Unfortunately, due to unforseen circumstances, Godefroit et al. were unable to present their apparently “fuzzy” ornithischian, and so what was to be the highlight of the 73rd SVP meeting was absent. We were made to do.

Late into 2013, however, things changed. First, Vladimir Alifanov of the Paleontological Institute at the Russian Academy of Sciences, wrote a short impromptu paper describing integumental structures of a novel form not too dissimilar to that Godefroit’s team mentions, followed by a longer paper in 2014 in which Alifanov joined with Sergei Savel’ev of the Russian Academy of Medical Sciences to discuss specimens collected from Kulinda as the basis of two distinct small ornithischian dinosaurs, granted the named Kulindapteryx ukureica and Daurosaurus olovus.[n1] Both of these ornithischians were treated to what might be termed a cursory description and lead in. The real prize, though, were the integumentary structures, and in their paper Alifanov and Savel’ev — and in a later paper Savel’ev and Alifanov — discussed the preservational and integumentary implications of these structures in a more refined manner.

At this time, however, Godefroit et al. were in the process of working a paper through the journal Science, and because of the locked down, tight-lippedness of Science‘s publishing process, little of this was revealed for some time. Their public discourse was brief, but sharp: Alifanov was describing material that was collected and brought from Chita and taken to Moscow, whereas all of the remaining material which Godefroit and his team were working on remained in Chita; this became an affront, an ongoing legacy of the centralization of scientific work in the old Russia (business as usual). Despite the localization of museums, which coauthor Yuri Bolotsky (a frequent coauthor of Godefroit’s) worked in at the Institute of Geology and Nature Management, Blagoveschensk — near Chita, but far, far from Moscow — the dispersal of the material purported a rift in workers interested in the site and its contents.

Eventually, Godefroit et al. produced a paper in 2014 that described many of these remains as indicative of a single morph, an ornithischian they deemed Kulindadromeus zabaikalensis. And it is this name that has persisted in the online discussion to date (but see [n1]; I won’t get into this further here). It is important to note that the structures being described by both sets of authors, and the skeletal material they use are indicative of likely only one animal, and that the material housed in Moscow likely pertains to the same taxon as that in Chita. If this bed contains a mass accumulation of a single, early ornithischian, it is useful to at least discuss the integument as it belongs to the “Kulinda ornithischian” (or “KO”) — which is what I, the eternal fence sitter, will call it moving forward here.

The KO, from Godefroit et al. (2014). Represented by nearly complete coverage of the skull and skeleton, including integument associated with most portions of the body. Elements described by Alifanov & Savel’ev (2014) relate to portions of the skull, forelimbs, hindlimbs, and pelvis, and their variation is not substantially different from what is shown here.

At first glance, the KO doesn’t look very spectacular, with a general body plan typical of small, early ornithischians: A large head and shortish snout, relatively simple “phylodont” teeth, large orbits indicated a largish eye; very short forelimbs and short hands, which indicate something more derived and less prehensile than in heterodontosaurs; long legs and long feet, with long claws; a pelvis with a short obturator process of the ischium, resembling heterodontosaurs somewhat, but with a very ornithopodan-like ilium; the tail lacked large ossified tendons, with abrupt narrowing in the first third or so of the tail base – the rest of the tail was a thin rod-like feature. But what is important about the KO is not its skeleton so much as the preservation of dermal structures around the limbs.

The Ukureyskaya Formation at Kulinda (representing the lower part of the Formation) represents a nearshore lacustrine or estuarine-deltaic environment, perhaps a lowland freshwater marsh, which saw frequent volcanic activity, resulting in tuffs and tuffaceous modification of the sand- and siltstones that form the beds at Kulinda. In many ways, this environment resembles the Yixian Formation, though much older. And so it is no surprise that this environment might be conducive to preserving the outlines, and even detailed integumental structures, of its preserved fauna. And in this case, boy howdy (archaic, I know).

There are three major types of integument preserved, all superficial. The first is the presence of rounded, tubercle-like structures on the arms and legs. These structures resemble the apparent dermal scales or scutellae on the limbs of a specimen of a Psittacosaurus species (lujiatunensis) from Liaoning, China (Mayr et al., 2002).

Integument of the KO, from Godefroit et al. (2014). Extra dermal structures vary in the form of closely packed rounded tubercular “scutae” (A), large plank-shaped structures with lobate interlocking features (B), fine filaments with roughly parallel arrangement, especially around the leg where they are longest around the periphery of the preserved blocks (C-F) or on the body wall (G-H). Additional types include the parallel-arranged filaments arising from a scuta-like base (not shown).

The second type is one of the more interesting structures to be found in any dinosaur, and occur only on the tail, in the form of elongated paired plates arranged along nearly the entire tail. Multiple tail specimens of the KO, many of them nearly complete, demonstrate this, and they are exclusive to the tail. These structures resemble if anything the dorsal scute armor of armadillos (though ossified in the mammal), the dorsal scutes of the dinosauriform Scleromochlus taylori, and that of many crocodilians, including the presence of overlapping, backwards-oriented prongs possibly marking the dorsolateral surface. Thus, these plates may have been L-shaped in section.

The third type is the most controversial, and involves the presence of elongated filamentous structures adjacent to the tubercles, extending distally from them, and are present only on the limbs. They are arranged in sets of four, distally splay outward, and begin above the middle of the tubercles. Filaments across the remainder of the body are looser, and not connected to underlying structures, like tubercles, but rather seem to arise on the dermis directly. These appear similar to those integumental structures as in Tianyulong confuciusi.

But what these are and what they might be related to is the subject of debate.

Some have speculated — for quite some time, now — that the filamentous integument in some basal ornithischians may be similar to the integument in some theropods, and in turn that of pterosaurs. This makes sense, in a way: As we move basally in both theropodan and ornithischian lineages, complex and filamentous integument continues to appear. Parsimoniously, the simpler scenario is that fuzzy integument arose once on the avian stem, occurring in the pterosaurian and dinosaurian lineages, than occurring multiple times and producing divergent morphologies only later in the history of the branches. A concise morphological analysis of stem-avian integument of the fuzzy kind has been approached, but tends to focus too much on the current morphology of the structures (including implying something specially about flattened structures that appear ribbon-like, but show no evidence of barbs, thus implying a solid vane). Lately, work has shown that details of preserved structures can be lost in birds (Foth, 2012, which I wrote about here), leaving one to question how well we should trust the preserved morphology and low levels of detail as being authentic.

The KO and other fluffy ornithischians also raises questions regarding the extent and distribution of such integument in other dinosaurs, and in the pattern in which this integument was distributed. But before we can figure out how it all began, we need to figure out what the pattern and distribution of living dinosaur feathers is. Such a study has never been done, to my knowledge, and may go a long way to helping figure out how early birds developed and modified their integument to become the feathers of today

It seems important to link ornithischian integument to theropods, and from that to pterosaurs, to give dinosaurian integument context and a deep history. Romantically, we’ve begun extending these structures onto every ornithodiran, regardless of whether integument is preserved. Art has run ahead of the science, perhaps with great prognostic powers but also with great heedlessness to the security that science brings.Some have even claimed that we should base our work on the secure knowables, the integument we have, but never on extrapolation of exterior factors, such as environment. Inference of integument from high latitudes have snuck in, where I was wont to joke that we might inevitably find a furry ankylosaur. While I might add that we cannot exclude this possibility currently, we have some constraints and should not be hasty to discount them, and one of those happen to be close relatives. The KO offers several close relatives amongst ornithischians (in phyletic space, at least) with extra-integumental structures that suggest that what the KO preserves may be analogous — but as to whether they are homologous has not been established, yet some will barrel onwards.

Older sediments than the Jurassic ones that have produced the KO (and even more recently, Dracoraptor hanigani (Martill et al., 2016) are needed, and they tend not to preserve enough substantial detail to clear the fog that surrounds this important possible link. Enter debate. What to make of the KO structures, which are generally well preserved but not in enough detail to determine even if the filaments are hollow, if filaments at all? That there were two types of body “fuzzage” (sparser, coarser on the limbs and associated with tubercles; finer, denser over the head, neck, body) seems mostly uncontroversial, but the current research on the structures, built from the two sets of papers produced, differed on details.

Both papers with both Alifanov & Savel’ev on the structures implicated a rather peculiar morphology directly contradicted by both photos of the material at hand and by Godefroit et al.’s more thorough treatment of the data. The first group implied, for example, that the “filaments” were flattened, ribbon-like, and arose from the margins of the rounded tubercles, and the authors implied these structures fit neatly into the old, but long discredited, idea that feathers arose from a fully-developed scale that differentiates into a frayed structure. The work of various others have demonstrated that a more likely origin for the feather structure is from a scale, but that both scale and feather are distinct trajectories: neither comes from the state that resembles the other. Avian scutellae are rather stunted forms of feathers, which occurs after the feather differentiates from the state where the placode it comes from would otherwise produce a scale. The feathers-from-scales hypothesis seems alive in some circles despite having been effectively tossed out a decade ago.

But coming into this, Lingham-Soliar has also produced a paper in which he was determined to show that the structures identified by both groups of authors were intradermal, rather than extradermal. Moreover, he referred to them as evidence of collagen fiber bundles. This is the go-to explanation at this point for any elongated filaments preserved anywhere on the body of a non-avian dinosaur (or, in parlance: a “dinosaur,” excluding birds from the complex entire), regardless whether it forms a halo or are scattered over the body [n2]. The hasty generalization doesn’t follow from careful analysis but broad comparisons of form. As Godefroit et al. (2014b) responded, TLS’s arguments were broadly superficial, lacked concern for the wealth of fossils collected at Kulinda in which distinct integumental regionalization was consistent (where structures appear on the body), but also that the identification as collagen was, loosely, in error: collagen forming dense bundles of fine resolution, these other structures were very widely spaced, single filaments and rather thick at that.

A bizarrely integumented animal from the near-shore at what is now Kulinda may finally have been laid to rest so long ago, but only to revise controversy (and dispute from amongst a very small cadre of critics) on the homology of analogous structures in non-avian dinosaurs, birds. Yet as if the comedy that surrounds the KO could not get any richer, Alifanov & Savel’ev followed forth with yet more from Kulinda: therein, they described a partial limb with integumental remains, and claimed the similarity of the limb to an ornithomimosaur forelimb, naming the fossil Lepidocheirosaurus natatilis; the name amusingly suggesting a scale-limbed lizard that swam, due to the environment, presuming that environment of deposition reflected environment of life habits. Yet as Andrea Cau pointed out not too long thereafter, the authors seem to have confused a foot for a hand, and when placed toward the hinder of the two sets of limbs, it fell rather close to limb material described for Kulindadromeus zabaikalicus. Needless

From Cau: A): I resti dell’arto di Lepidocheirosaurus. B): ricostruito come una mano. C): piede di Heterodontosaurus con indicate le parti corrispondenti in Lepidocheirosaurus. [A) Remains of the limb of Lepidocheirosaurus. B) reconstruction of the manus. C) pes fragments of Heterodontosaurus and the corresponding material of Lepidocheirosaurus.]

Thus the identification of large scales on the manus in an ornithomimosaur, seeming at odds with recent discovery of quite the opposite, feathery kind, becomes rather conventionally sauropsidan, finely to coarsely scutellate as seen in many birds, lizards, and even crocs. Other factors suggesting a pedal identity include the weak curvature of the unguals. [I cannot but note that another case that likely misidentifies a pes as a manus is that of Martill et al.’s Dracoraptor hanigani, as pointed out by Mickey Mortimer.]

The peculiarities of the KO remind us that integument is both rare, and at once informative and mysterious. We know relatively little about how it preserves and certainly less about its variety. But even more cautious than assuming integumental form and homology, the material we base our arguments on must be properly and fully placed in their context of rarity and uncertainty. It will only be through repeated discoveries that we can resolve, individually dinosaur by dinosaur, their integumental status, while holding them in context to the taxa that bracket them. Pterosaurs had bodies largely covered in very fine filaments, and these seem to extend onto the wings in some taxa, but where preservation either fails to demonstrate the surface of the membrane, not in others; some ornithischians appear covered in a fine fuzzage, others tubercular skin with large quill-like structures; some theropods have a fine filament body covering, and others with varying types of odd, perhaps actually pennaceous but “blade-like” or fused-vane structures on tail, back, etc.; and of course birds. Yet other ornithischians show extensive dermal preservation in the form of either large swaths of fine tubercular scutellae, or larger ones, or very large ones immersed in a sea of tiny nodes, or ossified dermal structures in such and perhaps arising from the previous. No two hadrosaur taxa seem to have the same dermal pattern, if the saurolophine census (Bell, 2012) holds. Dermal morphology seems to define anylosaurs to a fair degree, especially of the skull; the same holds true for some crurotarsans, especially aeotsaurs and phytosaurs (parasuchians). So we shouldn’t expect this of ornithischians, and when there is high variability in integumental form, even more so.

So there should be much caution in making blanket statements about consistency in integument. It’s fine to vary your art; art exists to ask, and to answer. But it’s also fine to check yourself. If you’re gonna paint a woolly dinosaur, there should either be a humorous aside in there somewhere, or you’ve got some damn nice evidence.

[n1] I will not at this point entangle myself in the nomenclature issues involved beyond saying that a given person’s opinion will strongly depend on how firm you feel the ICZN should be taken and that the process of post-publication peer review can result in disuse of taxonomy. I don’t like raising up a system of code-based nomenclature, use it, support it, only to then throw it down when someone names something in a way that is disagreeable, as happened for dinosaurs with the designation of “Syntarsus” rhodesiensis as Megapnosaurus rhodesiensis (the inane “big dead lizard,” coined by an entomologist, both points used to discredit its use). Problematically, the current situation suggests heavily that were we to follow the letter of the ICZN, the names put forward by Alifanov and Savel’ev in 2014 would have priority over Godefroit et al.’s name, regardless of how you feel personally. The accusation of “claim jumping” that may come from the practice of the former to pre-empt the latter has unfortunate socially political ramifications. However this sorts out, the majority of workers who must deal with these consider Godefroit et al. the “right claim,” and use their name accordingly; Wikipedia gives the illusion this is settled, and takes a side very firmly, even lumping Lepidocheirosaurus into Kulindadromeus, along with Kulindapteryx and Daurosaurus, essentially committing “taxonomic activism.”
[n2] TLS has failed to determine the difference in an actual bird, and continues to fail to determine what fossilized collagen should look like in any otherwise terrestrial or aerial animal preserved in the Liaoning lagerstätten, such as mammals, birds, lizards, or what-have-you.

Alifanov, V. R. 2014. The discovery of Late Jurassic dinosaurs in Russia. Doklady Earth Sciences 455 (2): 365-367.
Alifanov, V. R. &, Savel’ev, S.V. 2014. [Two new ornithischian dinosaurs (Hypsilophodontia, Ornithopoda) from the Late Jurassic of Russia.] Палеонтологический Журнал 4: 72-82. [Russian edition; English edition: Paleontological Journal 48 (4): 414-425.]
Alifanov, V. R. &, Savel’ev, S.V. 2015. The most ancient ornithomimosaur (Theropoda, Dinosauria), with cover imprints from the Upper Jurassic of Russia. Paleontological Journal 49 (6): 636-650.
Bell, P. R. 2012. Standardized terminology and potential taxonomic utility for hadrosaurid skin impressions: A case study for Saurolophus from Canada and Mongolia. PLoS ONE 12 (2): e31295.
Foth, C. 2012. On the identification of feather structures in stem-line representatives of birds: Evidence from fossils and actuopalaeontology. Paläontologische Zeitschrift 86:91-102.
Godefroit, P., Sinitsa, S. M., Dhouailly, D., Bolotsky, Y. L., Sizov, A. V., McNamara, M. E., Benton, M. J. & Spagna, P. 2014. A Jurassic ornithischian dinosaur from Siberia with both feathers and scales. Science 345: 451-455.
Godefroit, P., Sinitsa, S. M., Dhouailly, D., Bolotsky, Y. L., Sizov, A. V., McNamara, M. E., Benton, M. J. & Spagna, P. 2014. Response to Comment on “A Jurassic ornithischian dinosaur from Siberia with both feathers and scales.” Science 346: 434.
Lingham-Soliar, T. 2014. Comment on “A Jurassic ornithischian dinosaur from Siberia with both feathers and scales.” Science 346: 434.
Martill, D. M., Vidovic, S. U., Howells, C. & Nudds, J. R. 2016. The oldest Jurassic dinosaur: A basal neotheropod from the Hettangian of Great Britain. PLoS ONE 11 (1): e145713.
Sinitsa, S. M. 2011. [New data on dinosaurs from Zabaikal’ya]. pp. 173-176 in Sinitsa, S. M. (ed.) Environmental Cooperative Studies in the Cross-Border Ecological Region: Russia, China, and Mongolia (Institute of Natural Resources, Ecology and Cryology, Siberian Branch of the Russian Academy of Sciences, Chita, Russia.) [In Russian.]
Savel’ev, S. V. & Alifanov, V. R. 2014. [A new type of skin derivatives in ornithischian dinosaurs from the Late Jurassic of Transbaikalia (Russia).] Докладй Академий НАУК 456: 182-184. [in Russian]