Although 2020 was an all around shit show for the world, there were a few interesting bits of paleontology. Perhaps none more interesting than this potential overhaul to pterosaur reconstructions. Now, I haven’t really written anything about pterosaurs on my site yet (aside from some basic rundowns of news stories), so this marks an interesting way of christening the occasion. Pterosaurs are a fascinating group of critters that showcase yet another example of how animals can get airborne. They are also enigmatic in their origin and evolution.
One thing that we seemed to be confident about for a long time now is that pterosaurs had fuzz. It wasn’t fur in the mammal sense, and it sure as hell wasn’t feathers (more on that in a minute), but instead a third (or fifth, when counting arthropods and some plants) form of insulative covering. At least, that’s what I thought we knew up until about last year.
It may just turn out that this seemingly rock-solid interpretation of pterosaur outer coverings has been wrong all along.
First impressions of pterosaurs
The first pterosaur was discovered by Cosimo Alessandro Collini back in 1784. Collini gave a remarkably accurate description of the specimen, though he never named it. As the animal came from marine deposits, Collini assumed it was a strange swimming reptile that used its enormous arms as paddles (Collini 1784). Although a strange interpretation by today’s standards, it wasn’t out of the realm of possibility given the knowledge of ancient life at the time. Even Compsognathus was once thought to have had flippers (Bidar et al. 1972). Cuvier (1809) would later redescribe the animal as a flying reptile and coin the now famous name, pterodactyl. This specimen became Pterodactylus antiquus. As it was a reptile, it was justifiably assumed to have been scaly.
The scale-covered interpretation of pterosaurs would continue for several decades until the late 1830’s with the discovery Rhamphorhynchus muensteri and the description of one specimen by paleontologist, Georg August Goldfuss (1831). Goldfuss made mention of how the specimen he was studying appeared to have hairs like a mammal. This short paper likely validated the interpretations of entomologist, Edward Newman (1843), who proposed that pterosaurs had to have been endothermic and thus, hair covered like bats.
Goldfuss and Newman’s interpretations still didn’t really take off much (no pun intended) until we hit the 1970s and the discovery of the small pterosaur, Sordes pilosus (Sharov 1971). Here was a small pterosaur that was surrounded by “hairy” fibres. Most of those fibres surrounded the wing membranes (patagia), but a few seemed to be in areas that were outside of the membrane. Pterosaurs looked to be fuzzy critters.
Despite the implications of the taxon’s name (Sordes pilosus = Hairy devil), this discovery was not exactly a slam dunk for fuzzy pterosaurs. At the time, it was known from a single specimen and as I mentioned, much of the fuzzy parts were within the wing membrane, suggesting a structural use as opposed to integumentary appendages (see: Bakhurina and Unwin 1995). As such, the switch over from scaly pterosaurs to fuzzy ones still took quite some time, with scale-covered pterosaurs still regularly seen in paleo-art of the late nineties.
2002 was the real turning point for pterosaur body coverings with the description of the first specimen of Jeholopterus ningchengensis (Wang et al. 2002). This was an anurognathid (the so called “frog mouthed” pterosaurs) with some incredibly well preserved fuzz all around its body. Not only did J. ningchengensis showcase fibres around the body, but fibre type varied between the wings and the body, suggesting the type of variation expected of integument. Kellner et al. (2010) reviewed the data on J. ningchengensis and other fuzzy pterosaurs. Their work separated the dermal wing fibres (actinofibrils / aktinofibrils) from the integumentary fibres (pycnofibres). At last, we had solid evidence that pterosaurs were fuzzy critters.
It is worth noting that in this same year, Tischlinger and Frey (2002) published on a specimen of R. muensteri with preserved soft tissues. The authors focused heavily on the preserved wing membrane and its actinofibrils. There is no real mention of outer coverings on the skin (as far as I can tell from translations of the German) and most of the presumed fibres appear more like dendrites (natural branching patterns of crystal formation). Nonetheless, this paper is often cited as an example of pterosaurs that preserved integument (e.g., Kellner et al. 2010), so I’ll leave it here for posterity.
The main point is that we can thank J. ningchengensis for the move to hairy pterosaur reconstructions
Although current fossil data indicated that pterosaurs were fuzzier beasts than we had once considered, pycnofibres were largely considered to be convergent structures to the hair of mammals and the feathers of birds.
But damnit all if the discoveries of 2002 didn’t play with the idea of feather homology too. In the original J. ningchengensis description, Wang et al. (2002) wrote:
The “hairs” of Jeholopterus bear some resemblance to the hair-like integumental structures of the feathered dinosaur Sinosauropteryx and Beipiaosaurus. The integuments in the latter two dinosaurs are fibre-like, with no branching structure. Besides, they are distributed along the full body. Based on phylogenetic result, it is now generally agreed that the fibre-like integuments in these dinosaurs may represent protofeathers.... Although no one has proposed that pterosaurs might have protofeathers there is, however, no direct evidence to argue for or against the homology between the “hairs” of pterosaurs and the fibre-like integuments in Sinosauropteryx.
This was a wishy-washy way of basically saying the evidence could go either way (or: you can’t prove they aren’t homologous).
In that same year, Qiang Ji and Chongxi Yuan published a description of two pterosaur specimens from the Daohugou Biota that the authors described as being covered in “protofeathers, or feather-homologous structures.” (Ji and Juan 2002)
Lastly, Steven Czerkas and Qiang Ji described a new pterosaur specimen called Pterorhynchus wellnhoferi, which not only preserved a soft-tissue head crest, but protofeathers…apparently (Czerkas and Ji 2002). According to the authors:
The distal end of a wing membrane is preserved in Pterorhychus which shows clear aktinofibrils that are aligned in parallel rows. However, at right angles to the aktinofibrils are minuscule pinnate fibers which though imperfectly preserved, resemble the larger integumentary structures from the body. These tiny tufts on the wings are set close together in rows and the diamond or V-shaped pattern caused from their general outlines are distinctly visible throughout. These tufts extend across the entire width of the membrane. They are also preserved more as three dimensional structures, whereas the aktinofibrils are preserved two dimensionally as stains within the matrix. Several of the tufts show distinct filaments that emanate from a round base, like a calamus. Therefore, the evidence suggests that the external surface of the pterosaur wing was not naked, but covered by tiny pinnate fibers which would have looked much like a fine layer of velvet.
The integumentary structures seen in Pterorhynchus bear a striking similarity to that of a
natal down feather with only the notable absence of having the additional barbules branching from the barbs…. This absence is significant all the more because without the barbules, the barbs emanating from a calamus represents the hypothetical “Stage II” structure speculated as being an incipient step in the evolution of feathers…
Homologizing pterosaur pycnofibres with theropod protofeathers was a huge claim, and one that required some extraordinary evidence to back it up.
Kellner et al. 2010 deliver a cold shower
In 2010, Alexander Kellner and colleagues provided a thorough redescription of J. ningchengensis. This redescription included a detailed analysis of the preserved soft tissues. As with previous studies, this study focused largely on the wing membranes and their extent along the body. However, Kellner et al. did spend a fair amount of time looking at the hairy coverings and provided a detailed account of their location and shape. It was this paper that coined the term: pycnofibre.
This pterosaur hair, which is not homologous to the mammalian hair (a protein filament that originates deep in the dermis and grows through the epidermis), is here called pycnofibre (from the Greek word pyknos, meaning dense, bushy). The pycnofibres are further formed by smaller fibrils of unknown nature. They were possibly mostly composed of keratin-like scales, feathers and mammalian hair.
Given the eight years that had passed since the glut of pterosaur integument papers from 2002, Kellner et al. had a chance to go over the claims of feather homology in a fair amount of detail. What they found was that all the claims that had been made at this point, were based on a superficial similarity to the protofeathers of basal coelurosaurs (i.e., type I protofeathers). As I have brought up on multiple occasions, type I protofeathers are essentially unbranching cylindrical shafts, easily comparable to mammalian hair, insect setae, and even the trichomes of many plants; at least on a superficial level. Any claims of homology at this level require a detailed examination that at least looks at the ultrastructure of the material. At the time of Kellner et al. 2010, this had not happened. Kellner et al. pointed out that the integument assumed to be protofeathers from the specimens described by Ji and Yuan (2002) looked exactly the same as what Kellner et al. had just described for J. ningchengensis, and thus the assumption of feather homology is unwarranted. As for the P. wellnhoferi description from Czerkas and Ji (2002), Kellner et al. said this:
The rounded structures interpreted as calamus, the basal structure of a true feather, are intriguing. However, they are not hollow and have distinct convex outlines, contradicting their interpretation as true feather (or feather-like) calami and might not be directly related to the soft tissue.
As it stood, pycnofibres were just another instance of convergent fuzz development.
Yang et al. 2019 provide the extraordinary evidence
Aside from wishful thinking, pycnofibres standing as separate structures from protofeathers was largely unchallenged after Kellner et al. 2010. That all changed in 2019 with the publication from Zixiao Yang and colleagues. The authors published a paper specifically on the integument of pterosaurs, using data from two as yet unnamed anurognathids (likely species of Jehlopterus). According to the authors, these specimens preserved unambiguous homologies to bird feathers
Here, we report remarkably well-preserved pycnofibres in two anurognathid pterosaurs, and demonstrate—using evidence from morphology, chemistry and macroevolutionary analyses—that the preserved pycnofibres bear key features of feathers: monofilaments, two types of non-vaned grouped filaments, bilaterally branched filaments that were previously considered unique to maniraptoran dinosaurs, and preserved melanosomes with diverse geometries
Yang et al. (2019) identified four types of pycnofibres.
- Curved monofilaments 70–430 µm wide found on the head, neck torso, limbs and tail
- Bundles of curved filaments similar in length, forming a brush-like surface distally. They range in size from 80–180 µm wide. They are seen on the proximal forelimb, neck proximal tail and plantar surface of the metatarsals
- Straight to slightly curved, distally tapered central filaments with short lateral branches diverging from the midpoint. Width was 50–450 µm wide and these were found only on the head
- Tufts of curved filaments that diverge proximally. They are 70–130 µm wide and found only on the wings
To further solidify that these structures were integumentary appendages and not external structures unrelated to the animals, Yang et al. used scanning electron microscopy to search for and identify potential organelles that would be equivalent to melanosomes. The authors identified several structures that could be interpreted as phaeomelanosomes (red-black producing melanosomes). Fourier Transformed Infrared (FTIr) Spectroscopy was used to determine the chemical makeup of these microbodies in the filaments, further supporting a biological origin to these structures. FTIr spectroscopy also indicated the presence of keratin in these structures, further supporting their role as integumentary appendages.
Yang et al. go into further detail on why their interpretations cannot be confused with things such as superimposition of filaments on top of one another. All in all, they presented a very strong case for these structures being biological and part of the integument of the animals.
As for the homology to feathers, well this is what they wrote:
The filamentous integumentary structures in our anurognathid pterosaurs are thus remarkably similar to feathers and feather-like structures in non-avian dinosaurs. Intriguingly, cylindrical (type 1), radially symmetrical branched (types 2 and 4) and bilaterally symmetrical branched (type 3) filaments clearly coexisted in individual animals; these structures may represent transitional forms in the evolution of feathers, as revealed by developmental studies…
As you can see, it’s not exactly an air tight case. All the amazing work done earlier to show that these structures were biological, kinda falls apart when it comes down to homologizing with feathers. The authors essentially rely on superficial comparisons only (replete with extensive caveats).
Nonetheless, this was the first legitimate attempt to homologize (proto)feathers with pycnofibres, with the implications being that the origin of feathers is in fact extremely ancient, going back as far as the base of Ornithodira or possibly Archosauromorpha (depending on where pterosaurs finally wind up sitting on the evolutionary tree).
The paper’s implications received extensive news coverage at the time, with many articles heralding the new information and many a paleo-artist happily enfluffening anything that fell between pterosaurs and birds.
Critique of the pycnofeather proposal
Liliana D’Alba (2019) wrote a comment to Yang et al. in the same issue of Nature Ecology and Evolution. In the comment, she rightfully pointed out that the feather interpretation was the weakest part of the authors paper.
As is often the case in palaeontological studies, even when the preservation of the fossils
is remarkable, Yang and colleagues’ interpretation of the pterosaur skin appendages is still based entirely on gross filament morphology. As a consequence, the abstraction and documentation of all the details of the filaments cannot escape some degree of subjective interpretation, which has left previous claims of pterosaur pycnofibres open to debate…
D’Alba argued for the use of SEM or synchrotron scanning to help answer questions about some of the branching patterns asserted for the pycnofibre types proposed by Yang et al.
Aside from that, I would add that the large gap between pterosaurs and coelurosaurs (the first theropods to show filaments) is filled with scaly animals, further hurting the case for homology, as the filaments fail the Patterson’s second test of homology, congruence (Patterson 1982). To put it another way, it’s not monophyletic.
Unwin and Martill pull the pendulum back…way back.
This brings us to the crux of today’s post. Last year, David Unwin and David Martill published a formal critique of Yang et al.’s interpretation of pterosaur pycnofibres.
Unwin, D.M., Martill, D.M. 2020. No protofeathers on pterosaurs. Nat. Ecol. Evol. 4(12): 1590-1591.
In the paper, the authors point out what D’Alba had already warned of. Namely, that the bunching up of decomposing fibres can produce optical illusions that make simple structures appear more complex superficially. Unwin and Martill point out that pterosaur wing membranes also contained melanosomes, and the presence of said melanosomes does not stand as strong support for pycnofibres being homologous to feathers, or even filaments (melanosomes in general, are pretty widespread along the body).
Unwin and Martill make note that the varied morphology described by Yang et al. 2019 is unique to these relatively derived taxa and that nearly every other case of pterosaur pycnofibre preservation shows simple monofilaments. The point being that complex pycnofibres equivalent to protofeathers don’t really jibe with pycnofibre distribution in pterosaur (or protofeather distribution in theropods, for that matter).
From here, Unwin and Martill go into detail on how actinofibrils can be confused with pycnofibres when they degrade, and are more sinouous than they are given credit for. That the pycnofibre variations seem randomly distributed among monofilamentous pycnofibres is also strange, compared to the more regionalized distribution seen in hairs, feathers and scales. Unwin and Martill (2020) argued that this suggests either confusion with adjacent actinofibrils, or degradation of either fibres.
The critique of the biochemical analysis is the most telling for me. As I mentioned above, the original paper from Yang et al. used FTIr spectroscopy to determine the chemical makeup of the pycnofibres. Their results showed that there was a strong signal for keratin. But, as astute readers of this site know, there are lots of types of keratin out there. All amniotes have keratin-coated epidermes. It’s the family of that keratin that matters. In this case:
The amide I peak at 1,650 cm–1 is more consistent with α-keratin (characteristic of extant mammal hair…) than β-keratin (the primary keratin in extant avian feathers…). This signal may be original or diagenetic; the molecular configuration of keratin…and other proteins… can alter under mechanical stress and changes in hydration levels. — Yang et al. 2019
That’s the clincher. The keratin signal is alpha keratin rather than beta keratin. Reptile scales are comprised of beta keratin with an alpha keratin “hinge”. Bird feathers are almost entirely beta keratin. Mammalian hair and skin, on the other hand, is only alpha keratin. Our understanding of the evolution of keratin proteins indicates that beta keratin is a sauropsid synapomorphy, and alpha keratin is the plesiomorphic condition (Dhouailly 2009). This makes pycnofibres unique in that their keratin makeup neither matches protofeathers nor scales. In this way, pycnofibres are closer to mammalian hair than any other sauropsid integumentary appendage is.
Unwin and Martill noticed the same discrepancy:
The presence of α– rather than β-keratin is inconsistent with protofeathers…but is typical of more pliable regions of the epidermis in reptiles…and consistent with the glabrous, non-scaled integument of pterosaurs.
Note, the last part of the sentence there. This is the first official volley into a rather heretical new view of pterosaurs that we will go more into in a bit.
Yang et al. fire back
The critique by Unwin and Martill was not left unchallenged. As the authors had made a direct critique of the work by Yang and colleagues, it was only appropriate that the authors were given a chance to respond to the claims by Unwin and Martill. In that same volume of Nature Ecology and Evolution, Zixiao Yang and colleagues offered their counter to Unwin and Martill (Yang et al. 2020).
Namely, they rebut the following five points:
- Superposition or decomposition produces branched structures similar to those in the anurognathids — Yang et al. contest that the regional placement of these different fibre morphotypes makes it hard to use taphonomic degradation as the argument as one would expect a more global distribution of fibre decomposition.
- The anatomy and distribution of the integumentary structures are consistent with aktinofibrils, but not pycnofibres — Yang et al. point out that pycnofibre placement is consistent with other pterosaurs such as S. pilosus and J. ningchengensis, making it harder to argue for misasigned actinofibrils.
- Evidence for keratin and melanosomes are not indicative of pycnofibres but may reflect contamination from epidermal tissue — Yang et al. point out that melanosome placement is restricted to sites inside the fibres and not in any surrounding matrix, making it hard to argue for contamination. They also cite similar studies that have challenged melanosomes in dinosaur fossils, that were similarly supported by this restricted placement.
- Branching is not consistent with exclusively monofilamentous coverings in other anurognathids — Yang et al. counter with…well birds have varied feather morphologies so maybe pterosaurs did too.
- Homology with feathers cannot be demonstrated conclusively owing to the simple morphology of the pterosaur fibres — Yang et al. insist that the morphologies present in their specimens would still fall within the purview of early protofeathers.
Yang et al. largely did a good job tackling the arguments put forth by Unwin and Martill, but it really feels like they came out strongest with their first three counter arguments, using multiple points to validate their claims. The last two counter arguments are where things floundered pretty badly. Given that these are the two areas that D’Alba (2019) also dinged them for, it’s not too surprising. I will also add that the melanosome arguments, though well supported by Yang et al. in their rebuttal, should be viewed with caution. The presence of melanosomes and other organelles in fossils is still a contentious topic (see: Schweitzer et al. 2015, Vinther 2016 and McNamara et al. 2018 for examples). There remains a good chance that what has been proposed to be organelles, are really just taphonomic fragments of other unrelated microbes (or even unrelated melanosomes).
Unwin and Martill (2020) largely focused their short commentary piece on the alleged homology of pycnofibres to protofeathers, but there was more under the hood than the paper led on. Namely, Unwin and Martill have been working towards publishing a radical overhaul of pterosaur integument that would basically shave off all the pycnofibres and leave them naked. Though their response to Yang et al. only briefly mentions this, the paper received a fair bit of news coverage, and it was in this news coverage that this larger overall theme was made apparent.
As reported by the Daily Mail:
‘The idea of feathered pterosaurs goes back to the nineteenth century but the fossil evidence was then, and still is, very weak,’ said Dr Unwin, who is a pterosaur expert at the University of Leicester’s Centre for Palaeobiology Research.
However, Professor Martill and Dr Unwin believe that the proposed evidence for protofeathers in pterosaurs are actually just tough fibres that formed part of the flying reptiles’ wing membranes.
Detractors to Unwin and Martill’s claims have rightfully pointed out that this is a case of science by press release (essentially skirting passed peer review). I agree that this is somewhat shady. Paul Barrett did something similar when he had a press release come out regarding the extent of filaments in dinosaurs. The news article appeared a good year before the paper came out (Barrett et al. 2015). Similarly, the arguments made by Unwin and Martill in the press release are from their upcoming paper on pterosaur integument. So, there is a paper on the way. It’s just awkward and unfortunate that the press release has preceded the actual science.
Recently, Unwin and Martill gave a presentation at the 2020 SVP virtual conference, in which they officially argued for naked pterosaurs. Unfortunately, I was not willing to attend that year (the price for a glorified set of Zoom sessions did not sit well with me), and the presentation has yet to be made available. So, all we have to work with is the abstract (available here).
Identity, Homology, and Composition of Fiber-Like Strutures Associated with the Pterosaur Integument.
David M. Unwin and David Martill
Fiber-like structures are frequently preserved in association with fossilized remains of the pterosaur integument. Several fiber types have been recognized. Among the commonest are aktinofibrils, typically 40–100+ µm in breadth and present throughout the flight patagia, exhibiting the same patterns of alignment across Pterosauria. Occasionally partially mineralized in distal regions of the patagia, aktinofibrils were composite,helically-wound structures composed of much finer filaments a few microns in diameter. Comparable in size to aktinofibrils, but less common, are single-stranded, hairlike pycnofibers, seemingly branched in two specimens of the anurognathid Jeholopterus, that supposedly adorned parts of the cranium, neck, and body. Fiber-like structures have also been reported in cranial crests, foot webs, and tail flaps. The identity, homology, composition, and function of integumentary fibers is fiercely disputed. This study aimed to resolve these issues through analysis of 150+ specimens where the integument is preserved, representing >25% of known pterosaur species, 15 of the 20 principal lineages, and almost the entire temporal range of the clade. Details of the macro- and microstructure of fibers was obtained using light, UV and laser-UV photography, and binocular and scanning electron microscopy. Results of this study provide broad support for a new model in which pterosaur integumentary fibers of all types had a single common origin: dermal collagen. This idea is consistent with: (1) exceptionally preserved examples of cranial crests, wing membranes, and integument associated with the neck and body, which demonstrate that fibers were embedded within the integument, and formed part of the dermis; (2) calcification of fibers in the cranial crest and, occasionally, in distal parts of the flight patagia; (3) the composite construction of fibers, which were composed of much finer, helically-wound fibrils. Multiple specimens with soft tissues preserved in four different preservational modes, show that the integument had a glabrous, fine granular, or even polygonal external texture. Aktinofibrils and other collagenous dermal fibres (e.g., in cranial crests and skin associated with the neck and body) exposed by decay of the remarkably thin epidermis have frequently been misinterpreted as pycnofibers. External fibers fringing the jaws of anurognathids may be an exception, although branching, reported in one specimen, is likely an artifact of preservation.
That is one hell of a shot across the bow. Claiming fuzzless pterosaurs is no small hill to climb, but it does appear that Unwin and Martill have brought along a hefty amount of ammunition (including analyses of the ultrastructure for both actinofibrils and pycnofibres). Given the amount of work that has gone into this study so far, I suspect the final paper will be quite the tome. Unfortunately, as we are just now at the start of this new chapter in pterosaur research, we have a long way to go to see where everything settles.
The return of the scaly pterosaur…or not
The evidence brought forth by Unwin and Martill is very intriguing, but does it mean that our original interpretation of scaly pterosaurs was correct?
Well, no. Because, as Unwin pointed out in this earlier piece, we’ve never had evidence that pterosaurs were scaly.
Perhaps the most important point to emerge is that pterosaurs, unlike modern reptiles, did not have scales. Or, to qualify this slightly, in general, they did not have scales; as a beautifully preserved fossil of a Brazilian tapejarid shows, the underside of the heel of some (possibly all) pterosaurs was covered in small, diamond-shaped scales…
Apart from the undersides of the feet, and perhaps the “clawed” fingers, the skin seems to have had a relatively smooth, slightly leathery texture... — Unwin 2006
That the only evidence we have for scales in pterosaurs, comes from the bottom of their feet is telling. As we’ve learned in previous posts, the underside of bird feet are not covered in scales, but instead the underlying structures (reticulae) are likely aborted feather buds that have been heavily cornified. Did pterosaurs do something similar (e.g., heavily cornifying the epidermis on their feet)? An FTIr analysis of these scale impressions could tell us a lot here (e.g., are these also only alpha keratin?).
Note how this is one of the first appearances of the naked pterosaur argument from Unwin. Looking over Unwin’s previous works (e.g., Bakhurina and Unwin 1995), it’s apparent that the “no fuzzy pterosaurs” hypothesis has been in the works for a long time. That’s a good thing because it means the authors aren’t just shooting from the hip. Instead, they are doing their due diligence by collecting and interpreting loads of data. At least, that’s what it currently looks like. We’ll see when the final paper comes out.
Not quite a paper, but more than an abstract
Although the final paper by Martill and Unwin has yet to surface, Dave Unwin was kind enough to deliver his current presentation of the data at the 2020 annual meeting of the Palaeontological Association, which (handily) was recorded and made freely available online.
Martill and Unwin are swimming against a strong current right now (a lot of paleo folks like fuzzy pterosaurs). Nonetheless, I would hope that it is clear to anyone who watched this, just how much work and data collection have gone into their argument. I think that Martill and Unwin have brought forth a solid case in favour of naked pterosaurs, and I look forward to seeing the back and forth that will almost certainly come from this.
Just as soon as it gets published.
Bakhurina, N.N., Unwin, D.M. 1995. A Survey of Pterosaurs from the Jurassic and Cretaceous of the Former Soviet Union and Mongolia. Hist Biol. 10(3):197–245.
Bidar, A.; Demay L., Thomel G. 1972. Compsognathus corallestris, une nouvelle espèce de dinosaurien théropode du Portlandien de Canjuers (Sud-Est de la France). Annales du Muséum d’Histoire Naturelle de Nice. 1:9–40.
Collini, C. 1784. Sur quelques zoolithes du Cabinet d’Histoire Naturelle de S.A.S.E. Palatine et de Baviére, à Mannheim. Acta Academiae Theodoro-Palatinae, Mannheim, Pars Physica. 5:58-103.
Cuvier, G. 1809. Mémoire sur le squelette fossile d’unreptile volant des environs d’Aichstedt, que quelques naturalistes ont pris pour un oiseau, et dont nous formons un genre de Sauriens, sous le nom de PteroDactyle. Annales du Muséum national d’Histoire Naturelle, Paris. 13:424–437.
Czerkas, S. A. & Ji, Q. 2002 A rhamphorhynchoid with a headcrest and complex integumentary structures. In Czerkas, S. J. (ed). Feathered dinosaurs and the origin of flight. UT, USA: The Dinosaur Museum. pp. 15–41.
Dhouailly, D. 2009. A new scenario for the evolutionary origin of hair, feather, and avian scales. J. Anat. Vol. 214:587-606.
Goldfuss, A. 1831 Beitra¨ge zur Kenntnis verschiedener Reptilien der Vorwelt. Nova Acta Leop. Carol. 15:61–128
Ji, Q. & Yuan, C. 2002 Discovery of two kinds of protofeathered pterosaurs in the Mesozoic Daohugou Biota in the Ningcheng region and its stratigraphic and biologic significances. Geo. Rev. 48:221–224.
Kellner, A.W.A., Wang, X., Tischlinger, H., Campos, D.A., Hone, D.W., Meng, X. 2010. The soft tissue of Jeholopterus (Pterosauria, Anurognathidae, Batrachognathinae) and the structure of the pterosaur wing membrane. Proc. R. Soc. B. 277(1679):321-329.
McNamara, M.E., Kaye, J.S., Benton, M.J., Orr, P.J., Rossi, V., Ito, S., Wakamatsu, K. 2018. Non-Integumentary Melanosomes can Bias Reconstructions of the Colours of Fossil Vertebrates. Nature Comm. 9:2878.
Newman, E. 1843. Note on the pterodactyle tribe considered as marsupial bats. The Zoologist. 1:129.
Patterson, C. 1982. Morphological Characters and Homology. in Joysey, K.A., Friday, A.E. (eds): Problems of Phylogenetic Reconstruction. New York: Academic Press, pp. 21–74.
Schweitzer, M.H., Lindgren, J., Moyer, A.E. 2015. Melanosomes and Ancient Coloration Re-Examined: A Response to Vinther 2015. BioEssays 37(11):1174–1183.
Sharov A. G. 1971, Novyiye lyetayushchiye reptili iz myezozoya Kazakhstana i Kirgizii. [New flying reptiles from the Mesozoic of Kazakhstan and Kirghizia.] Trudy Paleont. Inst. Moscow. 130:104–113.
Tischlinger, H., Frey, E. 2002. Ein Rhamphorhynchus (Pterosauria, Reptilia) mit ungewöhnlicher Flughauterhaltung aus dem Solnhofener Plattenkalk. Archaeopteryx. 20:1–20.
Unwin D.M. 2006. The body inside out. in: The Pterosaurs from Deep Time. New York. Pearson Education Inc. pgs: 131–135.
Unwin, D.M., Martill, D.M. 2020. No protofeathers on pterosaurs. Nat. Ecol.
Evol. 4(12): 1590-1591.
Vinther, J. 2016. Fossil Melanosomes or Bacteria? A Wealth of Findings Favours Melanosomes. BioEssays 38(3):220–225.
Wang, X., Zhou, Z., Zhang, F., Xu, X. 2002. A nearly completely articulated rhamphorhynchoid pterosaur with exceptionally well-preserved wing membranes and ‘hairs’ from Inner Mongolia, northeast China. Chin. Sci. Bull. 47:226–230.
Yang, Z., Jiang, B., McNamara, M.E., Kearns, S.L., Pittman, M., Kaye, T.G., Orr, P.J., Xu, X., Benton, M.J. 2019. Pterosaur integumentary structures with complex feather-like branching. Nat. Ecol. Evol. 4:24–30.
Yang, Z., Jiang, B., McNamara, M.E., Kearns, S.L., Pittman, M., Kaye, T.G., Orr, P.J., Xu, X., Benton, M.J. 2020. Reply to: No Protofeathers on Pterosaurs. Nat Ecol Evol. 4(12):1592–1593.