Feathers, filaments and theropod dinosaurs

Article metrics

One of the hottest debates in palaeontology is whether birds evolved from dinosaurs. A study of two exceptionally well-preserved specimens of a theropod dinosaur from China — complete with skin, internal organs and eggs — provides new clues to the origin of feathers.

Ever since John Ostrom resuscitated the idea in the 1970s, palaeontologists have been piling up the evidence in favour of theropod dinosaurs as the ancestors of birds. Recently, the gap between theropods and birds has been narrowed even further. Important avian characters, such as the furcula (wishbone), have been discovered in theropods thought to be close to birds1. And typical theropod characteristics, such as an enlarged claw on digit two of the foot, have been found in early birds2. The final, clinching fact would be the discovery of evidence of feathers — a defining feature of birds — in theropods.

Cue Sinosauropteryx prima, the so-called ‘feathered’ dinosaur from China (Fig. 1). Reports of this sensational discovery first appeared3 in late 1996. Although few scientists have yet seen the fossil material, some are already incorporating Sinosauropteryx into models for the origin of feathers and bird flight4. Still others argue5 that the ‘feathers’ are merely an artefact of preservation. On page 147 of this issue6, Chen, Dong and Zhen present the first detailed description of Sinosauropteryx, and show that the integument (skin) bore discrete filamentous structures. But are they feathers?

Figure 1: Reconstruction of Sinosauropteryx by Michael W. Skrepnick.

Three exceptionally well-preserved specimens of this dinosaur have been recovered from the Early Cretaceous Yixian Formation of China. Two of these are described by Chen et al.6, who show that Sinosauropteryx bore discrete structures that could have been ‘proto-feathers’.

Chen et al. describe two almost complete, near-adult individuals, with evidence of soft tissues including the integument, the eyes and possibly some internal organs7. They are, without doubt, the best-preserved dinosaur remains yet found. These, along with a third specimen, were recovered by Chinese farmers in Liaoning Province, China (P. J. Currie, personal communication). They were found in beds of the Yixian Formation, part of a thick sequence of lake sediments intercalated with volcanic deposits. During the past six years, these sediments have yielded many superbly preserved fossils. The result is an almost complete Early Cretaceous (145-97.5 million years ago) continental biota, composed of plants, insects, fish, lizards, turtles, pterosaurs, dinosaurs, mammals and numerous birds8 — the latter often preserved with their plumage intact9.

In life, Sinosauropteryx was about the size of a large chicken and distinguished by its deep, narrow body, remarkably long tail and rather short, stout forelimbs (Fig. 1). Like its close relative Compsognathus, from the Late Jurassic of Europe (163-145 million years ago), Sinosauropteryx has a very specialized hand. Its massive first digit bears a large claw that might have served as a killing tool. Last meals provide further evidence of an active, predatory lifestyle — a lizard in the gut region of one individual6 and a tiny mammal in the third specimen (P. J. Currie, personal communication). The body of the larger individual described by Chen et al. also contains another surprise — two small, oval structures which, judging by their shape, size and position, are almost certainly eggs. This is the first reasonably convincing record of this type of association for any dinosaur and, if correctly identified, provides incontrovertible evidence that this individual was female. Moreover, the relatively small size of the eggs and the possibility of paired oviducts6 suggest that, unlike modern birds (which have a single oviduct and, in general, smaller clutches of relatively large eggs), theropods had a more reptile-like reproductive system with two oviducts that produced larger numbers of relatively small eggs.

The most striking features of the fossils are the well-preserved remains of the integument, which forms a dark halo above the skull, neck, back, hips and both sides of the tail. Small patches of integument also occur on the skull, and are associated with the forelimbs, rib-cage and legs. The halo is composed of many coarse, sinuous filaments, which are possibly hollow and up to 40 mm long6. The filaments seem to be discrete structures and, although often matted and tangled, are not encased in skin or remnants of the decaying dermis, as some have suggested5. Branching of the filaments — a construction that is also typical of feathers — has been reported6,10, but the topography of this branching is not yet clear.

This brings us to the critical question: are these structures some kind of ‘proto-feather’? Chen et al.6 and others4,10 clearly favour this idea. They draw comparisons with the plumules of modern birds, which are relatively simple structures without barbules or hooklets. But any argument for homology between the feathers of birds and the integumentary structures of Sinosauropteryx needs to be supported by more than general similarities in structure and position. High-resolution microscopy and biogeochemical tests might provide some answers, but they will not solve all of the problems. Moreover, if Sinosauropteryx bears proto-feathers, we might expect similar (or perhaps even more feather-like) structures to have been present on at least some of the theropods that are more closely related to birds than is Sinosauropteryx (Fig. 2). Exceptionally well-preserved remains of the integument are now known for two of these dinosaurs — the ornithomimosaur Pelecanimimus11, and a small, unnamed maniraptoran theropod from Brazil12. In both cases, however, there is no evidence of the filamentous structures found in Sinosauropteryx.

Figure 2: The relationships of Sinosauropteryx to other dinosaurs.

Sinosauropteryx is not as closely related to birds as many other theropods, yet it is the first non-avian theropod that seems to show evidence of feather-like structures. (Modified from ref. 10.)

So, it seems that we still do not have absolute proof that some dinosaurs were feathered. Or do we? The Liaoning deposits have yielded three examples of another putative dino-bird intermediate, Protarchaeopteryx. According to a preliminary report by Ji and Ji13, these individuals have well-preserved evidence of true feathers. While we wait for further details of these Chinese fossils, we might consider the irony of the present situation — nothing for hundreds of years and then, suddenly, a whole flock of evidence.


  1. 1

    Norell, M. A., Mackovicky, P. & Clark, J. M. Nature 389, 447 (1997).

  2. 2

    Forster, C. A., Chiappe, L. M., Krause, D. W. & Sampson, S. D. J. Vert. Paleontol. 16, (Suppl.) 34A (1996).

  3. 3

    Hecht, J. New Scientist No. 2052, 7 (1996).

  4. 4

    Padian, K. J. Vert. Paleontol. 17, (Suppl.) 68A (1997).

  5. 5

    Geist, N. R., Jones, T. D. & Ruben, J. A. J. Vert. Paleontol. 17, (Suppl.) 48A (1997).

  6. 6

    Chen, P., Dong, Z. & Zhen, S. Nature 391, 147–152 (1998).

  7. 7

    Ruben, J. A., Jones, T. D., Geist, N. R. & Hillenius, W. J. Science 278, 1267–1270 (1997).

  8. 8

    Hecht, J. New Scientist No. 2094, 32-35 (1997).

  9. 9

    Hou, L., Martin, L. D., Zhou, Z. & Feduccia, A. Science 274, 1164–1167 (1996).

  10. 10

    Currie, P. J. in Encyclopedia of Dinosaurs (eds Currie, P. J. & Padian, K.) 241 (Academic, San Diego, 1997).

  11. 11

    Briggs, D. E. G., Wilby, P. R., Pérez-Moreno, B. P., Sanz, J. L. & Fregenal-Martinez, M. J. Geol. Soc. (Lond.) 154, 587-588 (1997).

  12. 12

    Kellner, A. W. A. Nature 379, 32 (1996).

  13. 13

    Ji, Q. & Ji, S. A. Chin. Geol. 238, 38–41 (1997).

Download references

Author information

Rights and permissions

Reprints and Permissions

About this article

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.