Of Chicken Wings and Dinosaur Hands

The idea that birds evolved from dinosaurs dates back to the 1800’s (Cope 1867; Huxley 1870), but it was largely abandoned for almost a century. The bird-dinosaur connection didn’t start gaining traction again until the 1960’s and 70’s, largely through John Ostrom’s discovery and detailed studies of particularly avian dinosaurs such as Deinonychus (a cousin of the more famous Velociraptor, see Ostrom 1976). The hypothesis was heavily vindicated in the mid-2000’s with the discovery of a variety of dinosaurs (largely from China) sporting feathers and/or protofeathers. It’s hard to see a fossil like the incredibly well-preserved specimen below, and not be struck by the clear connection between dinosaurs and birds:

Today, it is generally considered common knowledge that birds evolved from dinosaurs (in fact, a recent commenter on one of my YouTube videos chastised me for implicitly suggesting otherwise). However, there has been one major sticking point in this hypothesis.  Bird wings and dinosaur hands both show a reduction in the number of fingers, going from an ancestral five-fingered hand to a three-fingered hand.  But while paleontological evidence suggests that the bird-like dinosaurs (like Deinonychus) lost digits five and four, embryological evidence suggests that modern birds lost digits one and five:

This discrepancy has led some scientists to challenge the whole idea that birds evolved from advanced dinosaurs (e.g. Feduccia et al. 2005). But in a recent issue of the journal Science, Tamura et al. revisited the problem, and found new evidence that the bird hand actually retains digits one, two, and three, just like dinosaurs.

In four-legged vertebrates, digits (a.k.a. fingers and toes) begin to develop when the limb is little more than a bud sticking out of the embryo (you can see some images of developing limb buds in my post How to Build a Marsupial). Typically, the first visible digit in the developing limb bud is digit four.  Scientists have traditionally argued that the first visible digit in chickens is digit four, as it is in reptiles and mammals, meaning the three remaining fingers in bird wings would be digits two, three, and four.

Tamura et al. challenged this hypothesis using detailed cell-labeling and tissue graft experiments to see how gene expression controls digit specification.  In most limb buds, digits four and five develop in a region of cells collectively called the zone of polarizing activity (ZPA).  These cells release a gene called sonic hedgehog, which forms a gradient that moves up the limb bud, specifying the formation of digits three and two.  Below is a figure that sums up the results of the research performed by Tamura et al.  It compares development of a mouse limb, a chicken leg (hindlimb) and a chicken wing (forelimb):

Don’t be overwhelmed by the amount of data in this image; start by noticing the difference between the theoretical position (P) of digits shared in all early limb buds, and the actual digit (D) that develops.  In the chicken leg (hindlimb), position four (P4) stays within the ZPA (colored light blue) so it takes on the identity of the fourth digit (D4), just like in the mouse.  But in the chicken wing (forlimb), position four (P4) moves out of the ZPA, so it takes on the identity of the third digit (D3).  Similarly, because the ZPA has shifted, the gradient of sonic hedgehog (the grey curve that says “SHH”) now travels to position four and position three, turning them into digits two and three. 

This work shows that the presence of the first visible digit is not a reliable way of determining what the identity of each digit actually is. In a sense, this might seem like semantics, but by showing that chicken fingers have the same identity as dinosaur fingers refutes that last major challenge against the bird-dinosaur debate. Gives you something to think about next time you enjoy some buffalo wings…

Works Cited

Cope E. D. (1867). Account of extinct reptiles which approach birds. Proceedings of the Academy of Natural Sciences of Philadelphia: 234-235.

Feduccia, A., Lingham-Soliar, T., & Hinchliffe, J. (2005). Do feathered dinosaurs exist? Testing the hypothesis on neontological and paleontological evidence Journal of Morphology, 266 (2), 125-166 DOI: 10.1002/jmor.10382


Huxley, T. (1870). Further Evidence of the Affinity between the Dinosaurian Reptiles and Birds Quarterly Journal of the Geological Society, 26 (1-2), 12-31 DOI: 10.1144/GSL.JGS.1870.026.01-02.08


Ostrom, J. (1976). Archaeopteryx and the origin of birds Biological Journal of the Linnean Society, 8 (2), 91-182 DOI: 10.1111/j.1095-8312.1976.tb00244.x


Tamura K, Nomura N, Seki R, Yonei-Tamura S, & Yokoyama H (2011). Embryological evidence identifies wing digits in birds as digits 1, 2, and 3. Science (New York, N.Y.), 331 (6018), 753-7 PMID: 21311019

How to Build an Arthropod (An Arthropod Leg, at Least)

Liu J, Steiner M, Dunlop JA, Keupp H, Shu D, Ou Q, Han J, Zhang Z, & Zhang X. (2011) An armoured Cambrian lobopodian from China with arthropod-like appendages. Nature, 470(7335), 526-30. 

If you’ve ever wondered what a “typical” animal looks like, here it is:

Beetles are the largest group of animals in the world, constituting 400,000 species, or about 40% of all known animals on earth.  Beetles and other insects are part of the phylum Arthropoda, which encompasses animals that have an exoskeleton and jointed appendages.  Besides the insects, crabs, spiders, centipedes, scorpions, and the extinct trilobites are also arthropods.  Together, arthropods account for 80% of all known animals.

Genetic evidence suggests that the closest living relatives of arthropods are the Onycophorans, or velvet worms.  These animals show segmentation like the arthropods, but they lack rigid bodies or limbs.  Interestingly, the specialized, jointed appendages of arthropods seems to have been critical to their evolutionary success; although over a million species of arthropods are known, only about 70 living species of onycohporans have been described (Pechenik 2010).

Recently, a team of paleontologists from China and Germany published some exceptionally preserved Cambrian fossils that show a possible link between the onycohporan and arthropod bodyplans.  The animal, called Diania cactiformis, has a soft body like an onycohporan, but hardened, jointed limbs like arthropods.  Below is an image of one of the fossils from the publication, as well as a reconstruction of what the animal may have looked like:

Diania is not the first fossil animal to bridge the gap between onycohporans and arthropods.  During the Cambrian, a number of animals generically lumped together as lobopods evolved, which had soft, worm-like bodies with varying degrees of armor on their shoulders and legs (including Hallucigenia, which wins my nomination for greatest animal name in history).  But Diania shows more characteristics in common with true arthropods than any previously described lobopod. 

The authors of this paper are hesitant to say that this conclusively shows that hardened arthropod limbs evolved before hardened arthropod bodies, or speculate on why complex, hardened limbs might have evolved first.  Whatever the case, Diania is an incredible example that “missing links” in the fossil record do not typically look like a perfectly intermediate blend of two animals.  Instead they often have an unexpected mosaic of features, and show unique adaptations suited to their distinct environments.

Works Cited 

Liu J, Steiner M, Dunlop JA, Keupp H, Shu D, Ou Q, Han J, Zhang Z, & Zhang X (2011). An armoured Cambrian lobopodian from China with arthropod-like appendages. Nature, 470 (7335), 526-30 PMID: 21350485


Pechenik, J. A. (2010). Biology of the Invertebrates (Sixth Edition). New York: McGraw-Hill Higher Education.