Yeah. As Witton & Naish point out, even with the very high degree of bone pneumatization we know about, we're left requiring soft tissue densities that are a fraction of those found in living animals. How do you make flight muscles a quarter the density of water?

Grant Hutchison

Exactly. And there doesn't seem any reason to think they'd be that light. Why should they have such low wing-loadings? There's no reason to believe they were long-distance soar-without-flapping types. Maybe they flew to get away from tyrannosaurs and travel over rivers, hills, forests, etc., but just walked on flat plains?

Well, the reason to think they're soar-without-flapping types goes back to the observations reported in the OP, as well as theoretical stuff that's been around for a while. The energy cost of flapping flight seems to scale as a greater-than-unity power of body mass; whereas the cost of soaring scales as a less-than-unity power.
There's going to be an upper limit of body mass at which the energy cost of flapping flight exceeds the maximal exercise capacity of the beast involved: a descending spiral of increasing flight muscle weight and decreasing aerodynamic performance.

Grant Hutchison

A bit more on pterosaur flight and control.

Most importantly, like most fliers, all Pterosaurs could deform their wings and move them independently. Other adaptations:

The earliest Pterosaurs, the small Rhamphorhynchoids, had long tails which acted much like the long tail of a cheetah I think. Many also had vanes at the end of their tails, which might have acted like rudders.

A short but good discussion on vanes and wing deformation by two biologists:
http://www.askabiologist.org.uk/punb...ic.php?id=1282
And the wiki page on Rhamphorhynchoids has good pictures on the right hand side:
http://en.wikipedia.org/wiki/Rhamphorhynchoidea


The later Pterosaurs, the Pterodactyloids, lost their tails, perhaps trading some stability for greater manoeuvrability.

A few, like the large Pteranodon developed large cranial crests.
http://museumvictoria.com.au/prehist...teranodon.html
This perhaps worked directly as a rudder:
http://www.jstor.org/pss/1303424
or indirectly as a balance for the head:
http://www.jstor.org/pss/2417241 (Sorry, last two are abstracts)
The wiki page for pteranodon: http://en.wikipedia.org/wiki/Pteranodon

My view is that having a large control surface in front of the centre of gravity is asking for trouble. Unless you are sailing, of course. Then, that is exactly where you would want a control surface to be:
http://en.wikipedia.org/wiki/Nyctosaurus
http://dinosaurs.about.com/od/aviand...yctosaurus.htm

Imagine! A sail on a bird!
Having said that, a quick perusal of the literature shows that there is no consensus on the use of Nyctosaurus's large cranial crest.


Finally, the large Azhdarchids. I'm curious about their flight stability (see above posts), while in Grant's excellent reference (reproduced here) http://www.plosone.org/article/info%...l.pone.0002271 the author says they might not have been stable on water.
Finally, the sheer lift generation of pterosaur wings, seems to be (almost) beyond dispute. Even the heaviest Azhdarchids had acceptable wing loadings (above discussion by Grant and Vultur). So soaring flight, given favourable wind conditions (nod to OP article), appears to be the minimum flight capability of the pterosaurs.

Annoying nitpicking post:
Yeah, reversed curvature would be wrong of me. It's just that camber, while desirable, is not necessary for generating lift, especially if we're taking about secondary control surfaces. All you need is a positive angle of attack (or negative in this case).

And this is nitpicking, because I know you know this: you mentioned flat plates earlier in this thread

Also, to choose a different example, aircraft can fly upside down...

Yes, I wasn't clear about it, but I was guessing you wanted your swan to be airworthy without motors.
Flat-plate lift produces a lot of drag, so you would be turning that little tail into a sort of airbrake, producing a levering moment along the length of the swan in order to keep its head up. Given a swan's difficulty getting into the air in the first place, that would seem to be rather cruel.

Grant Hutchison

Lol. Shame I'm not royalty...