My longtime friend and colleague Gareth Dyke has a new paper out this week on the feathers of early birds and the implications for flight. In short ths shaft (rachis) of soem of the flight feathers are a bit weak. He’s kindly (or foolishly) agreed to provide this little post on the origins of the paper:
When I was living in BC, Canada, I got talking to my friend Gary Kaiser, a retired marine ornithologist from the Canadian Wildlife Service. Gary had noticed – and published in his Inner Bird book (2007) – that compared to the overall wing length, the primary feathers of Confuciusornis are extremely long. Much longer in fact than any living bird. So, we started to work on this problem: how could a bird with such long primary feathers flap it’s wings.
This led Rob Nudds (University of Manchester) and I to the issue of the rachis diameter of bird primary feathers. One day in Dublin in 2009 we were sitting in the pub (as we often do: Rob loves to drink Beamish – a stout from Cork that is hard to buy in Dublin … this involves lots of walking around and checking taps in pubs as most, of course, sell Guinness) thinking about flapping in fossil birds. Based on this conversation, I went to Munich and Frankfurt later that year and made some measurements – the rest is in the paper.
Archaeopteryx has shorter feathers relative to the rest of it’s wing – still quite long (but within the range of living birds) – but has an extremely thin rachis. The simple modelling we have done, assuming similar keratin properties in fossil bird feathers and living feathers, shows quite clearly I think that the feathers of Archaeopteryx and Confuciusornis must have been structurally weaker than those of living birds. People will argue with the approach, and of course lots of palaeobiologists and biomechanists have lots invested in the flapping flight of Archaeopteryx in particular – it’s funny though that no-one has really examined the strength of the feathers of these bird before.
Finally, although we didn’t say this in the paper, I think that this analysis combined with our 2009 Evolution paper shows that the wing-assisted-incline-running (WAIR) hypothesis for the origin of bird flight cannot be correct … (but that’s the subject of another blog).
Nudds, R.L. and Dyke, G.J. 2009. Forelimb posture in dinosaurs and the evolution of the avian flapping flight-stroke. Evolution 63(4): 994-1002.
Dave Hone's Archosaur Musings 
Very interesting study.
It supports the conclusion of Senter (2006) on scapulocoracoid orientation in Archaeopteryx and Confuciusornis.
Senter, P. 2006. Scapular orientation in theropods and basal birds, and the origin of flapping flight. Acta Palaeontologica Polonica 51 (2): 305–313.
Awww, I really liked the WAIR hypothesis. Science marches on!
Interesting indeed. I hope the paper is open access so those of us without academic affiliations can check it out.
I wonder how the strength of the feathers of Archeopteryx and Confuciusornis compares with, say, Microraptor, or even clearly non-volant maniraptorans with well-preserved feathers. If they were stronger, than it suggests Avialans at least developed better aerial capabilities than their Deinonychosaur cousins. If not it gets even more interesting.
Sorry it’s in Science, so not open access. As I’ve said before though, researcher *want* people to read their papers. Just e-mail one of the authors and ask for a copy (for this or anything).
As for the other taxa, no, they just deal with Archaeo and Confusci.
In the case of Confuciusornis, this is pretty interesting in that it flies in the face of the published taphonomic arguments for powered flight. For example, the evidence from stomach contents and bill morphology that it ate fish (though doesn’t appear to be a wader or swimmer) and is often found together in large flocks that seem to have been killed simultaneously over a lake. I suppose they could have been perching on branches overhanging the lake (and using a phugoid gliding arc to catch fish, or something…?). I don’t have the paper yet, are any of those previous arguments addressed?
Not by the looks of things. This is Science though, it’s a 2 page paper so they really can’t go into much detail and is really just tacking one, previously unconsidered, aspect of the evidence for flight. They certainly do make a point of noting that there is a mosaic of evidence and this is just one bit of it, so it’s not ignored as such, but not specifically explored.
Nice work guys! Frankly, I was always skeptical about WAIR in mid-Mesozoic theropods – the energy expenditure is considerable, and while the Neornithes that usually do it do tend to have an “Archie-type” wing shape (short and rounded), they also tend to have a strong sternal keel and associated powerful breast musculature – in essence, WAIRing today is a specialty of burst-flyers. And one thing that is clear from the raw physical constraints is that Archie was not a burst flier over prolonged distances. It could have generated the necessary energy by going anaerobic, but with your data (but see below) there is nothing to suggest it could flap as much and as strongly as required. And in addition, its habitat does not seem to have included much “I”s to WAIR up. So, as a supplementary mechanism, I see no problem with WAIR, but as a main cause for the evolution of bird flight, I have always remained unconvinced. Mechanical and physiological constraints were not fully tested, and the comparison suffered from insufficiernt taxon sampling.
Add to that the theory (by and now quite robust) that all Archie specimens, with the possible exception of the Solnhofen one (“Wellnhoferia”), were not at all close to maturity. Galliformes (the #1 WAIRers perhaps) are notably precocious even as hatchlings – most extreme in Megapodidae -, but there is no good reason to assume the flight capabilities of Archie increased significantly as it grew (relationship of body mass to airfoil size – think “Bergmann’s Rule”, which is based on the same allometry phenomenon).
To me, the best explanation for Archie is still a barely volant theropod that scrambled up some moderately elevated lookout spot – a cycad trunk or a rock -, perhaps WAIRing a bit in the process, but was essentially a “non”powered glider that used powered flight only to get an advantageous vector at launch. Such a scenario has the advantage of avoiding the big problems with either “ground-up” (insufficient power output) and “tree-down” (insufficient climbing capability and possibly insufficient trees), while providing a major evolutionary advantage that is lacking in pure WAIR (which as an escape mechanism needs to be sustained).
Another thing that has not yet been tested (IIRC) regarding “ground-up” is the influence of ground effect on a level-ground attempt by Archie to get airborne. It was physically/physiologically capable of lifting itself off the ground (whether it actually did habitually is another thing), but was it able to get out of ground effect (i.e., rise more than ~half a meter off ground) off its own power?
In any case, Archie does not seem to have much bearing on the evolution of powered bird flight – phylogenetically, it is close enough. But biogeographically it was separated from about anything else by far too much water and would require far too large a ghost record for me to be comfortable with as close an association with modern birds as presumed by those wo expand Aves to include Archie.
As regards Confuci, now that is one intriguing critter. The wing shape screams “put airflow over me, and a considerable amount thereof” – the only living thing that comes somewhat close are Apodidae I think (which also have a vestigial tail), but these are among the most accomplished flyers in the history of life. It is puzzling that the wing shape seems to work better at higher airspeeds, yet the critter had no apparent primary capability to achieve these. Tree-down high-speed gliding and/or something coming close to dynamic soaring (think seabirds) is perhaps the least troublesome null hypothesis here.
Chiappe et al (BAMNH 1999) strongly favored a running start over anything that had to do with trees, but as I read them even the weird “squirrel” pose seems less falsifiable here than for any other early bird. Not to say it is likely, but it cannot be ruled out yet, particularly as the new data weighs quite heavily against anything that has to do with running-flapping.
But Confuci also had this extremely robust shoulder joint region. So it was capable of reasonably powerful flaps, or at least its internal anatomy could exert/withstand considerable flight-related forces.
Is the rachis of Confuci remiges asymmetric or has dynamic-soarer-/swift-like apomorphies, when compared with those of Archie? That would be necessary for it to reach really high airspeeds – we can apparently rule out a flapping motion, but perhaps not all AOAs.
Matt – as regards the “gliding fisher” hypothesis, I thought about that too. Its mandible was fairly narrow and strangely keeled. If it were broad and flat, it couldn’t have worked, but as it was, dip/skim-fishing from a glidepath would finally explain that weird bill shape (and square well with loss of teeth). Stioll, data that agrees with a hypothesis is only half good…
Karl – yeah, a Microraptor comparison would be very nice.
Rob & Garry: you should go to the Naturalis (or any other place that has them) and check out Ophrysia superciliosa http://ip30.eti.uva.nl/naturalis/detail.php?lang=uk&id=44. It may have been a very, very good WAIRer (the scanty bits known about it suggest so) and it supposedly its remiges’ rachises were quite weak. (Even so, it was a relictual species and yielded far too quickly to pressure for its own good, so one may assume an environment with heavier predation pressure would not have allowed such a combination).