On the Structure of Fossil Feathers

Here’s a short piece, in part inspired by discussions with Matt Martynuick at DinoGoss on the reasoning for why we attach terms like “stage 1” to fossil “feathers.”

I’m putting that in quotes largely because it also involves the issue that what we are seeing are not precisely feathers — they are parts of feathers. And how much of that feather is preserved is a question, one for which we may not have a clear answer. It is one that has brought various researchers to the conclusion that there is a developmental pathway for evolution of feathers that mimics the ontogeny of them. For Drs. Richard Prum and Alan Brush, it indicates a development that proceeds from a slender, singular filament, which resembles the development of a variety of structures such as scales, scutes, and feathers; towards a branching array of filaments, which develop as the single structure becomes tubular, and then splits toward the base; toward a structure in which each of these secondary filaments ends up fusing into a single, thicker central structure — a rachis.

A, feather development from a placode. B, basal filament development into a differentiate tube with the nascent tubular and barbed structure inherent before opening into a mature feather. C, evolution of feather form, as hypothesized in Prum and Prum & Brush. Modified from Prum, 1999.

Problematically, the fineness of some of these structures is such that they do not always get preserved, and it may be that they may get preserved in bizarre ways. Recent discoveries have produced data indicating melanosomes in fossil feathers (Zhang et al., 2010), and since then they’ve been investigated in other fossil taxa. However, the issue has never been secure based largely on the process of melanosome preservation: is it direct preservation, or the product of organisms that incidentally preserve key features that are nonetheless themselves highly indicative of melanosomes? This has left enough uncertainty that some groups (such as Lingham-Soliar, 2011 — mentioned here — or Feduccia parroting TLS in his recent book, Riddle of the Feathered Dragons — “dragons”: anything but to link dinosaurs and birds) have used this to support their previously argued proposition that birds cannot, under any circumstances, be dinosaurs (even so far, as Feduccia so easily uses in his book, the idea that some dinosaur groups are, in fact, “birds,” but that they cannot be dinosaurs).

Even before Feduccia got his book published, and in fact before TLS published his critiques of “so-called” integumental preservation, Dr. Julia Clarke of the University of Texas and an international team had published on a Late Eocene (~36mya) fossil penguin named Inkayacu peracasensis (Clarke et al., 2010). This new taxon joined a host of other early fossil penguins from the Eocene of coastal Peru, but this one was different. This one preserved integument, and it garnered quite the attention due to the detail of preservation, down to melanosomes of body and wing feathers. These details, likely to not gain the dismissive waive of TLS (Inkayacu peracasensis is, after all, securely a bird), are virtually identical to eumelanosome structure identified for Confuciusornis sanctus, Sinornithosaurus millennii (Zhang et al., 2010) and Anchiornis huxleyi (Li et al., 2010), and the authors performed a range of analyses that helped them reconstruct the coloration pattern of their penguin.

Fossil feathers of Inkayacu peracasensis Clarke et al., 2010 and natural feathers of emperor penguin, Aptenodytes forsteri.

This is the level to which feather preservation can reach, but not all of it is that good. In many cases, the preservation is so low in detail that merely a carbonized smear is left, or depressions in the sedimentary medium (as in Archaeopteryx lithographica, despite details of barbules indicating a clear closed-vane pennaceous feather). Much doubt has been raised on the nature of preservation of the integument of Sinosauropteryx prima, prompting proponents to argue for filamentous integument (Chen et al., 1998; Currie & Chen, 2001), suggesting but unable to demonstrate that the material represented more intricate integument than simple filaments.

A closeup of integument of a specimen generally referred to Sinosauropteryx prima, courtesy of National Geographic. The integument indicates open-vaned pennaceous feathers — type II of Prum, 1999 — and isolated feathers of this specimen have shown clearer demonstration of this morphology.

As I noted here, the integument of Sciurumimus albersdoerferi has revived the interest in Sinosauropteryx prima‘s integument, preserving as it does detailed features appearing as filaments. Recent work by Christian Foth (University of Rostock) has forced attention away from apparent preservation and instead to understanding that what we see preserved is not the whole story (this does not mean that Foth accepts TLS’s view that instead of feathers or “dinofuzz,” the integument is wholly dermal, comprised of collagen). Foth (2012). As identified by Foth, crushing of a three-dimensional body forces the feathers to splay into a “halo” effect around the body, and also incidentally causes the feathers to obscure the finer details.

The results of Foth’s compression analyses: clearly pennaceous feathers appear as “fluffy” filamentous structure with no definition beyond the rachis.

As concluded by Foth (2012, pg.100),

“examples of simple and aberrant feather morphologies can be taphonomic in origin, and thus not represent intermediate steps of feather evolution (see Benton et al. 2008). The overlapping nature of feather plumage, the two-dimensional state in which the feathers are usually preserved and other taphonomic conditions make identification of feather type difficult, and misinterpretations cannot be ruled out. To avoid or at least minimize such mistakes, it is necessary to expand current knowledge of the diverse morphology of recent bird feathers, and further investigate the changes which integumental structures undergo during decomposition and the fossilization process in similar experimental approaches. Fossil feathers and plumage should additionally be investigated under ultraviolet (UV) light, because of the higher contrast that can be achieved in photographic documentation [but see Hone et al. (2010) for application and limitations of this method]. The discovery of feathered non-avian dinosaur fossils, the clarification of their phylogeny and the identification of different feather types in the fossil record are essential to our understanding of feather evolution in particular, and bird evolution in general.”

I agree whole-heartedly with this. I am, however, cautioned by what is apparently preserved, and thus by the stricture of evidence, in what type of integument I ascribe to fossils. It is fine for me to speculate, where I can describe full “lips” or “cheeks” to this ornithischian or that theropod, or redact those features instead. When you move beyond the evidence, you have to be fairly reasonable in how far you go, especially by use of the rules of inference. Primary data is always preferred, but in the absence of such one could at least instead use the bracketing of preservation. Prum & Brush (2002) offered such a paradigm, which I’ve modified below:

Here, I’ve modified this by highlighting Tyrannosauroidea (in orange) to indicate preservation in Dilong paradoxus and Yutyrannus huali (Xu et al., 2012), “Dave” (in yellow), and the issues raised here about Sinosauropteryx prima (in blue). If, as Matt at DinoGoss suggests, we raise the preservation level of Sinosauropteryx prima to type III, this should insist that all later pre-avian dinosaurs also at least preserve this level (i.e., pennaceous feathers). Foth implies this may be the case, but as I quoted above, we simply haven’t determined this yet.

I have made the argument that NGMC 91 (“Dave,” a specimen often considered a juvenile version of Sinornithosaurus millennii) preserves detailed preservation, and that moreover this preservation helps reject the hypothesis of Paul that they together allow him to reconstruct fully pennaceous arms and legs for “microraptorians”. While I feel now my position was incorrect in regards to how much we should rely on the preservation in reconstruction, I am cautious enough to worry that we should still not extend our ideas of preservation as seems “reasonable.” It is fine, however, to speculate as an artist, as long as we respect the nature of speculation. Further, it is my current concern (as was Foth’s) that preservation shouldn’t favor unusual morphologies of feather (such as the EBFFs (elongated, broad, filamentous feathers) of Xu et al., 2009), and that this morphology may represent degrees of resolution rather than demonstrable morphology. Meaning, it is also possible that therizinosauroids preserve stage III feathers.

Instead, we should (at the least), support the minimal reasonable evidence for a morphology that doesn’t force us to speculate. Relatedness may inform potential preservation and reconstruction, but actual preservation should be preferred at every stage. While I thus regard it likely that Sinosauropteryx prima had stage III feathers, as supported by a related specimen, I do not think we should so readily reconstruct this as fact.

Benton, M. J., Zhou Z.-h., Orr, P. J. Zhang F.-c. & Kearns, S. L. 2008. The remarkable fossils from the Early Cretaceous Jehol Biota of China and how they have changed our knowledge of Mesozoic life. Proceedings of the Geologists’ Association 119:209–228.
Clarke, J. A., Ksepka, D. T., Salas-Gismondi, R., Altamirano, A. J., Shawkey, M. D., A’Alba, L., Vinther, J., DeVries, T. J. & Baby, P. 2012. Fossil evidence for evolution of the shape and color of penguin feathers. Science 330:654-657.
Chen P.-j., Dong Z.-m. & Zhen S.-n. 1998. An exceptionally well-preserved theropod dinosaur from the Yixian Formation of China. Nature 391:147-152.
Currie, P. J. and Chen P.-j. 2001. Anatomy of Sinosauropteryx prima from Liaoning, northeastern China. Canadian Journal of Earth Sciences — Revue canadienne des sciences de la Terre 38(4):1705-1727.
Feduccia, A. 2011. Riddle of the Feathered Dragons: Hidden Birds of China. Yale University Press (New Haven).
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.
Hone, D. W. E., Tischlinger, H., Xu X. & Zhang, F.-c. 2010. The extent of the preserved feathers on the four-winged dinosaur Microraptor gui under ultraviolet light. PLoS ONE 5(2):e9223.
Li Q.-g., Gao K.-q., Vinther, J., Shawkey, M. D., Clarke, J. A., A’Alba, L., Meng Q.-j., Briggs, D. E. G. & Prum, R. O. 2010. Plumage color patterns of an extinct dinosaur. Science 327:1369-1372.
Lingham-Soliar, T. 2011. The evolution of the feather: Sinornithosaurus, a colourful tail. Journal of Ornithology 152(3):567-577. [pre-press copy of the paper, dated 8 Nov 2010 here]
Prum, R. O. 1999. Development and evolutionary origin of feathers. Journal of Experimental Zoology 285:291-306.
Prum, R. O. & Brush, A. H. 2002. The evolutionary origin and diversification of feathers. The Quarterly Review of Biology 77(3):261-295.
Xu X., Wang K.-b., Zhang K., Ma Q.-y., Xing L.-d., Sullivan, C., Hu D.-y., Cheng S.-q. & Wang, S. 2012. A gigantic feathered dinosaur from the Lower Cretaceous of China. Nature 484:92–95. (PDF)
Xu X., Zheng, Z.-t. & You H.-l. 2009. A new feather type in a nonavian theropod and the early evolution of feathers. Proceedings of the National Academy of Sciences, Philadelphia 106(3):832-834.
Zhang F.-c., Kearns, S. L., Orr, P. J., Benton, M. J., Zhou Z.-h., Johnson, D., Xu X. & Wang X.-l. 2010. Fossilized melanosomes and the colour of Cretaceous dinosaurs and birds. Nature 463:1075–1078.