Here’s a picture of a great blue heron raising its left leg. That joint in the middle, which most people think of as the “knee”, is actually the ankle. The true knee is just about hidden by the feathers on it belly, because the heron is holding its thigh bone horizontally against its body (as shown in pink below). So even though herons look statuesque and upright, they’re actually crouching, as is every other living bird.
Birds evolved from two-legged meat-eating dinosaurs but the earliest of these stood upright, with thigh bones held almost vertically below their hips. At what point did they start to crouch, and why?
Scientists have speculated about this question for decades, but Vivian Allen from the Royal Veterinary College found a clear answer by building virtual dinosaurs. He used medical scanners to reconstruct the skeletons of 17 species, representing offshoots of the lineage that eventually gave rise to birds. These ranged from a crocodile, to dinosaurs like Tyrannosaurus, Allosaurus and Velociraptor, to early birds like Archaeopteryx, to a modern chicken.
By putting virtual flesh on these skeletons, Allen showed that their crouching stance evolved gradually, just as their centre of mass (or centre of gravity) moved forward towards their heads. That seems obvious, but the reason for this weight shift was unexpected. Everyone assumed it was because the dinosaurs’ tails became lighter and shorter. Instead, Allen showed that it was because their arms became bigger.
It’s a slightly ironic result since most people studying the origin of birds already tend to focus on the arms, which slowly evolved from grasping, fuzz-covered limbs to powerful flying wings with flat vane-like feathers. Steve Gatesy from Brown University was the first person to seriously focus on the hind-limbs back in 1990, and his ideas on the evolution of dinosaur leg muscles and running gaits have since become textbook material.
John Hutchinson helped with some of this work but he still saw the origin of the birds’ crouched posture as an open question. Sure, people had compared different dinosaur skeletons by eye, but he wanted some hard numbers for their mass and their centre of mass. Why? Because it’s the key to understanding how dinosaurs moved. “The centre of mass is sort of a shorthand for the whole animal,” says Allen. “You can think of locomotion as using parts of your body to exert forces on your environment to move your centre of mass somewhere.”
Estimating centre of mass is easier said than done. You need to reconstruct a fully-fleshed animal from its bones, so you need exceptionally preserved skeletons and some way of estimating how much flesh surrounded them. Hutchinson started doing this in 1999 and Allen, his PhD student, eventually took up the baton. Fourteen years and 17 species later, their database was ready. “This paper means a lot to me because of that perseverance,” says Hutchinson. “I’m happy Viv stuck with it and believed it could be done, when many colleagues, even in our own lab, would scoff at the audacity!” (You can read Hutchinson’s own post on the study.)
The technique sounds simple: plaster virtual flesh upon virtual bones. “But we knew that these fleshy reconstructions are at least partly subjective, and therefore very difficult, if not impossible, to get right,” says Allen. To get the best estimates, he scanned living animals to see how much their fleshy outline differed from their skeletal one, and used these to build a range of different models. He made each body part—head, arms, tail, and so on—as big and small as they could plausibly be, and combined them to give a range for each animals’ mass and centre of mass.
Some scientists argued that centre of mass gradually moved towards the head over the entire bird line, while others said that it suddenly leapt forward as the earliest proto-birds developed big chests and powerful flapping muscles. Allen’s data supports both ideas to a degree.
He showed that centre of mass did move forward at an accelerated pace after the rise of small maniraptoran dinosaurs like Velociraptor and Deinonychus, roughly coinciding with the origin of flight in this group. But this was just the tail end of a smooth, long-running transition. The group’s posture and centre of mass had already been changing well before any of them took to the skies.
“Once again, this shows that there is no discontinuity between a “dinosaur-style” and a “bird-style” animal,” says Thomas Holtz Jr, a palaeontologist from the University of Maryland. “There is no real morphological moment where you see “Aha! This stopped being a dinosaur and started being a bird right here!” It’s the second decade of the 21st Century, so this shouldn’t be surprise to anyone…”
When Allen analysed the influence of individual body parts, the tail turned out to be unimportant. Instead, the size of the arms (and, to a lesser extent, the head and neck) were strongly correlated with centre of mass.
The result was so surprising that Allen and Hutchinson spent around two years checking the stats and convincing themselves. But the data kept on telling the same story: As the dinosaurs developed beefier arms, first to grasp prey or climb and then eventually to fly, their centre of mass moved forward, and their legs became more crouched. Adding mass to the front of their bodies was more important than taking it away from the back.
Next, Allen wants to use his virtual models to move from studying the evolution of dinosaur stance to understanding the evolution of their movements. That means not just fleshing out a digital skeleton, but simulating muscles, articulating joints, and the physical forces acting upon the moving limb. And the team hopes that others will join in too. They have shared all the images and methods from their study so that other scientists can add to them, or to use the data for their own purposes.
Reference: Allen, Bates, Li & Hutchinson. 2013. Linking the evolution of body shape and locomotor biomechanics in bird-line archosaurs. Nature http://dx.doi.org/10.1038/nature12059