[Continuing from my first posting on Duane Nash’s idea that some scales on some dinosaurs may have contained air (or at least gas), thereby providing thermal insulation…]
I think Duane has in mind hollownesses within nodular scales that were sometimes filled with air (or other gas) and sometimes with blood, or sometimes increasing or decreasing amounts of each:
“Through vasoconstriction[,] blood flow can be diminished to this layer creating an insulating, vacuum sealed layer of air visceral to the outer skin and which insulated dinosaurs from temperature extremes. Additionally, this layer could be vasodilated and engorged with blood to facilitate heat shedding or heat uptake into or from the environment respectively.”
But I think it’s important to remember that the “air” concerned probably wasn’t going anywhere in between hot or cold episodes, and also that it probably wasn’t possible to change its pressure at will, so you couldn’t just pump blood into the air-filled volume. Of course, if the mass of air remained the same but the temperature changed dramatically, then the pressure would inevitably change within an air-tight cell. I think blood supply changed when the animal was hot; it might have done what we do and just widened up the blood capillaries near the surface of the skin when it wanted to cool down, or perhaps it might have pumped more blood close to the surface of cells (through channels in the bone) only on the shady side.
But no blood in the central hollownesses. I’ve looked at the heat images of those sunbathing crocodylians (caimans) in Duane’s post, and because we don’t quite know enough about the circumstances, I don’t think it’s possible to say much, other than it’s not too surprising to see such images of “crocs” sunbathing, possibly still quite warm from the night before (the zoo was near the equator), with a slight breeze cooling the tips of the scales. That doesn’t mean the channels within stegosaur plates did not contain blood – perhaps they were internal arteries that fed exterior veins, but if so, those channels never contained air.
There is a big problem about how to get air into such hollow spaces within nodular scales. I suspect it is very difficult to produce air spaces without access to outside air. If it were easier, mammals would have air in the centre of their long bones that needed to be accelerated and decelerated the most (e.g. tibia/fibula). Birds can sometimes do it with some long bones (doubt if it ever gets below the knee) because they have outgrowths of their airsac system which can channel the air. But in order for birds to be born with air in their lungs they have to poke their nostrils into the airsac within the egg a day or two before hatching. The original liquid is drawn back into the membranes, and some evaporates, but outside air replaces it. I’m pretty sure hollow bones below the neck are never air-filled until after birth, and I don’t think even birds can just magic up bubbles entirely within the body… usually.
But if it’s true that Longisquama‘s feathers were in the form of long squashed balloons with seams down the middle, then air must have got in somehow! AND… not only would ornithiscian dinosaurs have presumably used the same method or at least achieved the same effects to air-fill what were in all likelihood homologous structures, but also, one would expect the method to be inherited from dino-ancestors with similarly primitive feather types. As I have explained at length, “dinosaurs” evolved repeatedly from small tree-living gliders, quite often from an ancestror with a slightly different feather type. Thus, troodonts and dromaeosaurs evolved from Archaeopteryx/Anchiornis-grade forms (powered flight by then, perhaps even Anchiornis to some extent) with modern feathers; Sinosauropteryx etc. evolved from animals with more primitive but already fibrous feathers; maybe there was an earlier descent of “theropods” earlier in the Jurassic, already with fibrous feathers of a slightly different sort… and before even that, there were dinosaur clades that came from types with non-fibrous feathers. That’s why we don’t see fibres in the same way and to the same extent on ornithiscians, but we may well be seeing some of these air-filled structures.
How did air get into these structures? Incidentally, air-filled structures may well have been part of the heritage of later fibrous-pellaged types, but presumably lost somewhere along the way when more modern feathers evolved. I have always supposed Portugese-men-o’-war can actually produce bubbles entirely internally – I think I’ve heard that claimed – and if so then it must be possible. I don’t see how Longisquama could have produced air-filled feathers by communicating with the internal pulmonary system, or to have inflated them with outside air. (Of course, it might well not be air as such but some other gas such as CO2 or O2.)