Today’s guest post comes from Michael Doube. I found out about Michael’s work though my friend John Hutchinson at the Royal Vet College in London who regular readers might well recognise as a major dinosaur researcher, though John often dabbles in extant organisms as a basis for his palaeontological research. In this case Michael and his team were looking at trabeculae, those little supporting threads of bone. Their pattern of distribution can potentially tell us quite a lot about both extant and extinct species and Michael explains:
There’s a stiff, supportive foam inside many bones, trabecular bone. Typically it’s found in condyles, the femoral head – the sort of place where load has to be spread out over a large area like near articulations.
In short, we found that trabecular bone is different in large and small animals (birds, mammals and one crocodile). Larger animals have thicker trabeculae that are further apart than in smaller animals but surprisingly they don’t have a higher volumetric density – there isn’t a change in the amount of bone per unit volume as a function of animal size. There are some pronounced differences between mammals and birds, which are as you would predict; birds (especially flighted ones) have much less bone per unit volume than mammals. The one reptile, a Nile crocodile (Crocodylus niloticus) that actually had trabeculae where we were looking, was not particularly different from mammals of similar body mass. Because reptilian bone development is a bit different to mammalian bone development (variable or absent epiphyses, slow mineralization of cartilage), reptiles tended to have a lot of calcified cartilage or woven bone in the ‘epiphyses’ where mammals and many birds have trabecular bone. The differences between crocodylian/mammalian and bird trabecular scaling raise the intriguing question of when this
difference evolved – in earlier theropods, early birds or modern birds?
We used finite element modelling and applied synthetic, equal loading to come up with a mechanical explanation for the differences in trabecular morphology we observed between animals of different size. Because the volume fraction basically stayed the same, the apparent modulus – the bulk stiffness of the models – was similar. But increased thickness of trabeculae appeared to be related to reduced strain within trabeculae.
We speculate that this might relate to bone’s ability to add material where mechanical loads are high and remove it where mechanical loads are low, and that allometry of trabecular geometry might be an
interspecific manifestation of this phenomenon.
Another important point is that there are physiological limits to trabecular thickness – they can’t be so thick that the embedded osteocytes are too far from an exchange surface (this distance is about 0.25 mm, meaning maximum thickness is about 0.5 mm). In the largest specimen we studied, a 3.4 tonne Elephas maximus, practically all the trabeculae were at this maximum limit of 0.5 mm thickness. In even larger animals, I suspect that increased volume fraction might appear, or changes in lifestyle that reduce gravitational load might be necessary, e.g. using buoyancy force of floating in water (like whales).
What also surprised me was that even the smallest animals (3 g Suncus etruscus) had trabeculae, despite their whole femur being smaller than an elephant trabecula.
As far as I’m aware, there’s not any good 3D data on dinosaur trabeculae, so I don’t know if these trends would hold in dinosaurs. There are enough specimens known that similar measurements could be done
on dinosaurs. Good 3D preservation, and image contrast between fossilized bone and matrix (or even lack of matrix) would be required. There’s a basic imaging limitation that you have the fossil’s matrix embedded between fossilized trabeculae, and ordinary polychromatic X-ray microtomography has real trouble resolving the difference in X-ray attenuation between the two phases. Also, there’s typically a resolution limit in cone-beam systems which operate by projecting a shadow onto a CCD detector. The higher the ratio between source-detector and source-specimen distances the greater the magnification, but the smaller the imaged volume becomes. This means that you can only scan small samples to get adequate resolution (in our system this was about 30 mm diameter maximum). If you have a large specimen, you have to cut a sample out of it, which is not possible for really valuable material, like a large, intact fossil femur.
People do cut up dino bones, but it would be a pain. Finally, most scaling studies use body mass as the animal size variable; this is unknown for dinosaurs, in common with most of the animals in our study. Instead, we used femoral head radius as the size variable, which could be used for fossils too.
Doube M, Klosowski MM, Wiktorowicz-Conroy AM, Hutchinson JR, Shefelbine SJ. 2011. Trabecular bone scales allometrically in mammals and birds. Proc. R. Soc. B doi:10.1098/rspb.2011.0069