Researchers detail how they married paleontology, biomechanics, computer simulations, live animal demonstrations, and even an Orobates robot
Nearly 300 million years ago, a curious creature called Orobates pabsti walked the land. Animals had just begun pulling themselves out of the water and exploring the big, dry world, and here was the plant-eating tetrapod Orobates, making its way on four legs. Paleontologists know it did so because one particularly well-preserved fossil has, well, four legs. And luckily enough, scientists also discovered fossilized footprints, or trackways, to match.
The assumption has been that Orobates—a cousin of the amniote lineage, which today includes mammals and reptiles—and other early tetrapods hadn’t yet evolved an “advanced” gait, instead dragging themselves along more like salamanders. But today, in an epically multidisciplinary paper in Nature, researchers detail how they married paleontology, biomechanics, computer simulations, live animal demonstrations, and even an Orobates robot to determine that the ancient critter probably walked in a far more advanced way than was previously believed possible. And that has big implications for the understanding of how locomotion evolved on land, not to mention how scientists study the ways extinct animals of all types got around.
Taken alone, a fossil skeleton or fossil trackways aren’t enough to divine how an animal moved. “The footprints only show you what their feet are doing,” says biomechanist John Hutchinson at the Royal Veterinary College, coauthor on the new paper, “because there’s so many degrees of freedom, or different ways a joint can move.” Humans, after all, share an anatomy but can manage lots of silly ways to walk with the same equipment.
Without the footprints, the researchers wouldn’t be able to tell with much confidence how the fossil skeleton moved. And without the skeleton, they wouldn’t be able to fully parse the footprints. But with both, they could calculate hundreds of possible gaits for Orobates, from the less advanced belly-dragging of a skink to the more advanced, higher posture of a crocodilian running on land.
They then used a computer simulation to toy with the parameters, such as how much the spine bends back and forth as the animal moves. “The simulation basically told us the forces on the animal, and gave us some estimates of how the mechanics of the animal may have worked overall,” says Hutchinson.
You can actually play with the parameters yourself with this fantastic interactive the team put together. Seriously, click on it and play along with me.
The dots in the three-dimensional graphs are possible gaits. Blue dots get high scores, and red dots get low scores. Double-click on one and below you’ll see that particular gait at work in simulation. You’ll notice that the red dots make for gaits that look a bit … ungainly. Dark blue dots, however, look like they’re a more reasonable way for a tetrapod to move. At bottom you’ll see videos of extant species like the iguana and caiman (a small crocodilian). It was observations of these species that helped the researchers determine what biomechanical factors are important, such as how much the spine bends.
A few other parameters: The sliders on the left let you monkey with things like power expenditure. Slide it to the right and you’ll notice the good blue dots disappear.
Here’s where things get tricky, though. Power efficiency is key to survival, of course, but it’s not the only constraint in biomechanics. “Not all animals optimize for energy, especially species that only use short bursts of locomotion,” says Humboldt University of Berlin evolutionary biologist John Nyakatura, lead author on the paper. “Obviously for species that travel long distances, energy efficiency is very important. But for other species it might be less important.”
Another factor is something called bone collision (which is a great name for a metal band). When you’re putting together a fossil skeleton, you don’t know how much cartilage surrounded the joints, because that stuff rotted away long ago. And different kinds of animals have different amounts of cartilage.
So that’s a big unknown with Orobates. In the interactive, you can dial the bone collision up and down with the slider at left. “You can allow bones to collide freely or just gently touch,” says Hutchinson. “Or you can dial it up to a level of 4 and allow no collisions, which is basically saying there must be a substantial space between the joints.” Notice how that changes the dots in the graph: The more collision you prevent, the fewer the potential gaits. “Whereas if you allow plenty of collision, there’s just more possibilities for the limb to move.”
Now, the robot. The team designed OroBOT to closely match the anatomy of Orobates. It’s of course simplified from the pure biology, but it’s still quite complicated as robots go. Each limb is made up of five actuated joints (“actuators” being the fancy robotics term for motors), while the spine has eight actuated joints that allow it to bend back and forth. In the interactive, you can play with the amount of spine bending with a slider at left, and see how dramatically that changes the gait. Also, take a look at the video of the caiman in there to see just how much its own spine bends as it moves.
The beauty of the simulation is you can run all kinds of different gaits relatively quickly. But not so with a robot. “Running too many experiments with a physical platform is quite time-expensive, and you can also damage the platform,” says coauthor and roboticist Kamilo Melo of the Swiss Federal Institute of Technology Lausanne. Running simulations helped whittle down the list.
“In the end we have several gaits we know are quite good, and those are the kinds of gaits we actually test with the real robot,” adds Melo.
What they found was that given the skeletal anatomy and matching trackways, it was likely that Orobates walked fairly upright, more like a caiman than a salamander. “Previously it was assumed that only the amniotes evolved this advanced terrestrial locomotion,” says Nyakatura. “That it is already present in Orobates demonstrates that we have to assume that locomotor diversity to be present a bit earlier.” An important confirmation from the trackways: There are no markings that would correspond to a dragging tail.
So thanks to a heady blend of disparate disciplines, the researchers could essentially resurrect a long-dead species to determine how it may have walked. “Because they have brought digital modeling and robotics and all those things together to bear on this one animal, we can be pretty confident that they’ve come up with a reasonable suggestion for how it moved,” says paleontologist Stuart Sumida of California State University San Bernardino. He’s got unique insight here, by the way: He helped describe Orobates in the first place 15 years ago.
It’s key to also consider where Sumida and his colleagues found the fossil, in Germany. Around 300 million years ago, there was no running water at the dig site. And it’s running water that paleontologists typically rely on to preserve specimens in mud. “This was an utterly terrestrial environment that just happened to flood occasionally,” says Sumida. “And so you get a very unusual snapshot of what life was like not in the water.”
The upright gait of Orobates, then, would make sense. “This is a thing that walked around with great facility on the land, and this is exactly what the geology suggested,” says Sumida. What that means, he adds, is that Orobates and perhaps other early land-going species adapted to their environment faster than expected.
As the Bee Gees once said: “You can tell by the way I use my walk, I’m a comfortably terrestrial early tetrapod, no time to talk.”
By Matt Simon
This article was originally published on Wired.com and has been republished under Creative Commons
A Crocodile-Like Robot Helps Solve A 300-Million-Year Mystery