The Hidden World Moves

Hidden worldNikon has handed out awards in its second annual Small World In Motion competition–a contest for the best video of the world we can’t see with our naked eye.

Here are a few of my favorites…

A tiny animal called a rotifer lives inside a tube attached to a plant. It beats hairs called cilia to pump food into the tube, where it can be ground up by the rotifer’s jaw-like structures.

Here, kidney cells grow in a dish into the distinctive branched anatomy that lets kidneys filter waste out of blood.

Cells move by tearing down and rebuilding their internal skeleton. This video of a cell shows the filaments that make up the skeleton as glowing sticks. (Here’s an article I wrote about cell skeletons in the New York Times.)

The entire gallery of winners and runners-up is here.

In The Beginning Was the Mudskipper?

In 1893, the Norwegian zoologist Fridtjof Nansen set off to find the North Pole. He would not use pack dogs to cross the Arctic Ice. Instead, he locked his fate into the ice itself. He sailed his ship The Fram directly into the congealing autumn Arctic, until it became locked in the frozen sea. Nansen was convinced that the ice itself would drift up to the pole, taking him and his crew along for the ride.

For two and a half years they drifted with the pack. It gradually became clear to Nansen that The Fram had stopped moving north and was now traveling east instead, back towards Europe. He leaped out of the ship and tried to sled up to the pole, only to discover that the ice he was now traveling on was moving south. Only four degrees away from true north, he decided to retreat. He bolted back for Franz Josef Land.

The Fram meanwhile continued to drift east. After several months, it broke free of the ice, and the crew sailed the ship south to the island of Spitzbergen. There on the bare flats they saw a giant balloon.

Its pilot was a young Swedish engineer named Salamon Andrée. Andrée had decided that ships like the Fram could never reach the Pole, and that flight offered the only hope. He had convinced the king of Sweden and Alfred Nobel to pay for a balloon which he had brought by ship to Spitzbergen. And there he mixed tons of sulfuric acid and zinc to create hydrogen gas, which filled his silk canopy for four days. But gales hit the island before he was ready to launch the balloon, and then the Fram arrived with stories of how Nansen was racing on sleds towards the pole. Andrée let the canopy fall back to the ground.

When he got back to Sweden, Andrée discovered that Nansen had actually failed and had returned to Norway. He began to plot a second attempt. He returned to Spitzbergen in 1897 and this time he succeeded in launching his balloon. For a few days Andrée floated north with his crew of two, bobbing up and down with the sudden changes in temperature and moisture of the Arctic atmosphere. But as he crossed over the edge of the polar ice, the voyage took a turn for the worse. The balloon became burdened with rain and snow, until the guidelines dragged across the ice, until the gondola bounced like a ball on the ground, until the balloon came to a rest.

For a week the crew huddled in cramped fog. Andrée decided to pack sledges with food and a collapsible boat, which they dragged over the drifting ice. Hauling them across sloshing leads, they hoped, like Nansen, that they could find refuge in Franz Josef Land. But the ice wandered in the wrong direction under their feet, and after two months of this polar treadmill they reached a little hump of Arctic rock called White Island. In 1930 whalers came to the island and discovered their decrepit boat, their journals, and Andrée’s corpse still sitting in the snow.

But in 1897, no one knew where Andrée had gone. His fellow Swedish scientists searched for him by ship in the following summers, first travelling around Spitzbergen, and then heading to Greenland. As the pack ice opened, they traveled for eight weeks along its eastern edge in their sail- and steam-powered ship. They mapped the tentacled coast, and in one fjord along an elephant-backed mountain they named Celsius Berg, the explorers found bones.

They weren’t the bones of Andrée and his crew. They were the bones of fish that had been resting in the Greenland rocks for over 350 millions years.

Other fossils of these fish had been found elsewhere in late Devonian rocks, but to those who studied that era, Greenland was a revelation. It was as if a new continent suddenly appeared on the map: other Devonian rocks were hid for the most part under a woody, bushy carpet in places like England and Pennsylvania, while the mountains of Greenland were mercilessly bare. Unfortunately the new fossils were also so remote that only some greater pretext–like the search for a famous explorer–could get the paleontologists to this far corner of the Arctic.

Another rationale came about thirty years later, when Denmark and Norway began competing in the late 1920s for control of Eastern Greenland, and the oil and minerals that it might hold. The Danes brought Swedish scientists with them, and they found bones of more fish, including lobe-fins, as well as a few things they didn’t know what to make of, simply marking them as “scales of a fish-like vertebrate of uncertain affinities.”

These expeditions were a bit less brutal than Andrée’s and Nansen’s trips. The scientists still traveled in wooden steamers with three square-rigged masts, and while they could now bring a hydroplane for their surveys they still wore polar bear suits when they flew. In 1931 an energetic 22-year old geologist named Gunnar Säve-Söderbergh was put in charge of the expeditions. For sixteen hours a day he could climb mountains, throwing rocks into his rucksack and sketching out stratigraphy along the way.

He had a book of numbered tags made for the expeditions, P. for fishes, and A. for amphibians–a supremely confident system, considering that no one had ever found a Devonian amphibian. The fossil record of land vertebrates with legs–known as tetrapods–only went back about 300 million years and stopped cold.

That first summer, Säve-Söderbergh made his way around the northern slope of Celsius Berg and found more fish. In the cones of fallen rocks below the mountain’s eastern plateau, he also found more than a dozen scraps of a flat skull that didn’t look like any species of fish he had seen before. Optimistically, he marked them with A. tags.

Back in Stockholm that fall, Säve-Söderbergh slowly worked the bones free of the hard sandstone, painting them with alchohol and balsam to reveal the sutures between the bones. Looking down on the flat roof of the skull, he could see that some of the bones were patterned like the skulls of a group of fish known as lobe-fins–represented today only by lungfishes and coelacanths. Many naturalists argued that tetrapods had evolved from an lobe-fin ancestor. But Säve-Söderbergh could also see that it had some traits–like a long snout–that had only been found in early tetrapod fossils.

Looking at that skull, Säve-Söderbergh realized that he had found the earliest tetrapod. He named it Ichthyostega–“fish plate”–after the top of the animal’s skull.

The discovery was a great hit in Denmark, not only with the politicians who wanted to tighten their grip on Greenland, but with the public as well. In celebration one newspaper cartoonist drew a trout with dog legs carrying a pipe-smoking caveman, as snakes encircled moutain peaks and elephants flapped their wings overhead.

Säve-Söderbergh spent the following few summers mapping more of the region by foot, boat, and Icelandic horse. Fossils practically fell out of the rocks for him–mostly fish but on rare occasion another piece of Ichthyostega. The strange scales that had been found in 1929 turned out to be Ichthyostega’s ribs, massive and overlapping like Venetian blinds of bone. His assistants, particularly a student from the University of Upssala named Erik Jarvik, found more Ichthyostega skulls. One skull, unearthed in 1934, was so handsome the paleontologists brought it back across the Atlantic resting on a blue velvet pillow.

After five years Säve-Söderbergh was appointed a professor at the University of Uppsala, but in that year he was diagnosed with tuberculosis. He lingered in bed, managing to write a few papers about some of the fish he had collected, and died in June 1948 at only 40. The summer of Säve-Söderbergh’s death, the expedition to Greenland finally found the legs and shoulders and tail of Ichthyostega. At last it had most of a body.

In the 64 years since Säve-Söderbergh’s death, scientists have discovered many more early tetrapods and their extinct lobe-fin relatives. They’ve found some of these beasts on return trips to Greenland. But they’ve also found other species in places like Pennsylvania, northern Canada, Latvia, and, most recently, Nevada. Together, these fossils now offer an illuminating look at one of the most crucial transitions in the history of life. Without it, we’d still be fish in the sea.

Despite all the new company Ichthyostega has enjoyed lavish attention ever since its discovery, thanks to the quality and quantity of the fossils it left behind. For decades, Erik Jarvik pored over the fossils, and then, after his death, Jennifer Clack of the University of Cambridge and other paleontologists took a look for themselves.

Ichthyostega’s legs, while short and squat, had the elbows, knees, ankles, wrists, and toes that qualified it as a tetrapod. (Strangely, it had seven digits on its feet.) Its spine was sturdy, its hips and shoulders massive, its skull rigid. Yet Ichthyostega’s rigid skull still shared some traits with the flexible skull of lobe fins. It had a distinctive suture in the skull at the same place where a lobe-fin skull has a hinge. Under the tetrapod palimpsest its ancestry could be seen.

Ichthyostega’s tail was a similar mix of tetrapod and fish. Tetrapods have simple tails consisting of a long series of tapering vertebrae encased in flesh (ours has dwindled to a mere sprout, the coccyx). A lobe-fin’s tail, the motor that the animal uses to move through water, is a much more elaborate affair. Each vertebra has two long rods, one on top and one below. Attaching to each of these rods are more slender bones, called radials, and attaching to the radials is a wide fan of fin rays: a completely differerent kind of bone called dermal bone that also makes up scales. This complex anatomy allows a fish to set up waves in its tail either forward or backwards, to let it dart through the water or suddenly brake.

The bottom of Ichthyostega’s tail had a simplified tetrapod form, but the top still retained all the geegaws of a fish. It was, in a sense, still half in the water.

Clack and her colleagues have used the anatomy of Ichthyostega to figure out what it did in life–and, by extension, to get some clues to how the tetrapod body plan evolved. Long before Ichthyostega came on the scene, lobe-fins were already evolving some of the crucial pieces of that body plan–legs and wrists, for example. These ancient relatives lived unquestionably like fish, using gills to breath and depending on water to support much of their weight. It’s clear, in other words, that even though the tetrapod body is very good for getting around on land, it didn’t start evolving on land.

How far had things gotten by the time Ichthyostega showed up 360 million years ago? Clack has found that Ichthyostega’s ear was tuned for hearing underwater. But when she and her colleagues looked at a series of Ichthyostega skeletons, going from young to old, they found a different story. As the animals matured, their shoulders changed shape, providing more space for anchoring arm muscles. It’s possible that they spent a lot of time in the water when they were young and then spent more time on land when they became adults.

In 2005 Clack and her colleagues did a thorough study of Ichthyostega’s trunk–its spine and rib cage. They concluded that the tetrapod was weirdly stiff, unable to bend from side to side. They suggested two possible ways for Ichthyostega to get around on land. It might walk, but without bending its body the way, for examplpe, a salamander does. Or it might mimic an inchworm. It would bend its spine upwards, reach forward with its front legs, and then straighten out, pushing forward with its hind legs.

Today in Nature, Clack offers more clues to this puzzling creature. She collaborated with John Hutchinson of the Royal Veterinary College, an expert on biomechanics, and his postdoctoral researcher Stephanie Pierce. They have brought Ichthyostega back to life through a detailed computer reconstruction.

They started by making high-resolution scans of its fossils, which they then assembled into a virtual skeleton. Hutchinson has, over the past decade, figured out a way to estimate how animals moved based on this kind of reconstruction. By placing virtual muscles on the virtual bones, he can estimate their range of motion. Hutchinson knows his models are reliable, because he can test them on living animals. His estimates for the movements of animals such as otters and alligators are close to how they really move.

Here are a couple videos showing their results. I’ll explain them below.

The whole body:

The hind leg in action:

Simply put, Ichthyostega could not have been very impressive on land. No matter how hard it tried, it could not walk with its back legs. The limbs could move forward and back, but they could not swivel into a position that would allow Ichthyostega to plant its feet on the ground. Its forelimbs were a little more useful. It could bend its elbows. But its shoulders had little range of motion.

Combined with their rigid trunk, these new findings lead Clack and her colleagues to conclude that the best living analogy for Ichthyostega is a mudskipper. Mudskippers are not lobe-fins. Instead, they are ray-finnsed fish, more closely related to goldfish or trout. In an independent transition from the ocean, they evolved the ability to move around on land by crutching along on their front pair of fins. As the delightful video below from David Attenborough shows, mudskippers are quite successful in their peculiar ecological niche, crawling on muddy beaches, sucking up food from the muck, and then swimming through their underwater burrows to care for their young. But they are hardly an inspiring vision of tetrapods emerging on land.

Clack’s new study stands in intriguing contrast to one that I blogged about in December. University of Chicago scientists reported then that lungfish–our closest living aquatic relatives–can walk underwater with their pelvic fins–which correspond to the hind legs of tetrapods. The Chicago team argued that hind-leg-driven walking could have started out long before the tetrapod body evolved. Clack and her colleagues, on the other hand, propose that hind legs came late to the terrestrial party.

But if there’s one thing that the past couple decades of fossil-hunting has made clear is that the origin of tetrapods was not some linear march of progress. Starting about 380 million years ago, some lobe fins independently evolved tetrapod-like traits in a grand, unplanned experiment. Different species ended up with different combinations of those traits, perhaps adapting them to different ways of getting around underwater or on land. Ichthyostega might be a good model for the ancestor of all living tetrapods. Or it may have been a very weird beachcomber with hind legs that were only good as underwater paddles. To find out, scientists need to build more virtual skeletons of early tetrapods. And they need to head out to find more fossils.

Let’s just hope that they don’t have to follow doomed explorers to find them.

For more information about the discovery of tetrapod evolution, see my book At the Water’s Edge, from which parts of this post were adapted.

[Images: skeleton from Clack’s site. Skull from Tree of Life. Images of the Fram and the failed balloon from Wikipedia.]

A Scientific Jonah: My profile of Joy Reidenberg in tomorrow's New York Times

For anyone in the US who likes to know what it’s like inside a giraffe (hands up, people), it was frustrating to discover the show Inside Nature’s Giants airing on British TV. The best we could manage were snippets on YouTube. Now the show is here in the States. The other day I spent some time with one of the main scientists of the show, Joy Reidenberg, an anatomist at Mount Sinai School of  Medicine. I’ve written a profile of her, both as a researcher who’s discovering fascinating new things about whales, and as that most improbable thing: a celebrity anatomist. Check it out.

Be sure to take a look at the extras on the page, such as the podcast, video, and graphic instructions for how to dissect a 50-ton whale.

[Photo courtesy of Joy Reidenberg]

Swans and stem cells: winners of this year's Imagine Science Film Festival

For the third year in a row I had the pleasure of serving as a judge for the Imagine Science Film Festival. Along with fellow judged neuroscientists David Eagelman and Darcy Kelley and documentary filmmaker Robb Moss, I watched a slew of short films that touched in one way or another on science. The awards were just announced, and so I thought I’d hunt around for some online sites where you can watch them, either as previews or in their entirety. Here’s what I found: (more…)

Swimming robots and flexing bones: My new story for the New York Times

Biomechanics is the science of flesh and bone–how birds fly, sharks swim, muscles twitch, and tendons spring. In January, I went to a fascinating session at the annual meeting of the Society for Integrative and Comparative Biology where biomechanics experts talked about how they’ve been trying to turn their insights about biomechanics into commercial products. One of the most surprising examples came from Charles Pell, a North Carolina inventor, who explained how surgical tools could be much improved by taking biomechanics into account. I later paid Pell a visit at the offices of his company, Physcient, to find out more about their first creation: a rib spreader that promises to spread ribs without breaking them. The result was an article which appears in today’s New York Times. Check it out.

[Image: Gray’s Anatomy]

When plants become ultrafast killers, it's time to slow the camera down

In my National Geographic article last year on carnivorous plants, I mentioned one particularly swift killer, the bladderwort. This aquatic plant grows little suction traps that can be triggered by passing animals. In a new paper in the Proceedings of the Royal Society, French researchers take the closest look yet at these ultrafast killers. They find that the door to the traps buckles like a popped bubble of chewing gum–but can then almost immediately swing back shut. Along with the new study on jumping fleas I wrote about last week, this is evidence of how far we’re just starting to explore the world of quick biology.

Science News has a nice write-up, and here is an excellent YouTube video provided by a co-author of the study, Philippe Marmottant, a physicist at Joseph Fourier University in Grenoble, France–complete with computer simulation, rubber-cap demos, and groovy soundtrack.

The flea's mighty jump

If you could jump like a flea, you’d be traveling 3,000 miles an hour within a thousandth of a second. You can’t do it, so how do they manage? That’s the 350-year-old question I consider in an article in The New York Times. Check it out.

Epic Gulp

Ed Yong has written a great post on a new paper on how blue whales snarf up half a million calories in every gulp. The paper is the latest in a series put out by Jeremy Goldbogen of Scripps Institution of Oceanography and his colleagues. If you want to dig back into the earlier research, check out my 2007 New York Times article and this Loom post on how a physicist who studies parachutes helped solve the mystery of big gulps.

Spiderman's Bats

This spring I blogged about some marvelous videos made by scientists at Brown University in their quest to understand how bats manage to be bats. Turning your hands into membrane-lined wings makes for some awkward trade-offs. Moving around on the ground, for example, gets to be a special challenge. Bats have not simply evolved a single solution to these trade-offs, however. Instead, they’ve explored lots of different compromises. While many bats can only creep awkwardly on the ground, for example,  vampire bats can actually gallop.

IMG_6162Today one of the Brown scientists, Dan Riskin (who has just set up a lab at City College in New York),  published a new study on another extraordinary solution to being a bat. Most bats are well-adapted to roosting by hanging upside-down from long claws. But that is not a universal rule. A few bats have evolved a Spiderman strategy.

These bats have pads on their wrists and ankles that they clamp onto surfaces. The sucker-footed bat of Madagascar (Myzopoda aurita), for example, roosts on  the inner wall of a rolled-up leaf, its head pointed up instead of the usual down.

The Spiderman strategy has evolved many times in animals. Many insects and spiders can cling to walls, as can frogs and gecko lizards. Mammals–not so much. It’s true that many mammals have pads on their hands and feet that add friction to their grip. But sucker-footed bats are among the few mammal species that can really stick to vertical surfaces.

Last year, Riskin and Paul Racey of the University of Aberdeen went to Madagascar to figure out how sucker-footed bats manage this feat. They filmed the bats climbing sheets of glass and brass. As the bats climbed, the scientists tried to drag and pull them around to measure the forces they generated. They came to a surprising conclusion:

The sucker-footed bats of Madagascar, despite their common name, do not actually suck at all.

Suction works in a distinctive way. If you pull a suction cup away from the surface it’s attached to, the cup will strongly resist your efforts. But you can drag the cup along the surface with much less force. When Riskin and Racey tested the bats, they found that the animals could easily be pulled off a surface. Dragging the bats, however, required much more force. In fact, they calculated that a single wrist pad is so strong that it could hold the weight of eight bats.

But there’s one important caveat to the strength of the bat pads. Riskin and Racey measured strong forces when they pulled down on the bats. But when they pushed the bats upwards on a surface, the pads peeled off right away.

These results indicate that a sucker-footed bat sticks to a leaf by gluing its wrists and ankles to it. Riskin and Racey observed that the pads glistened with some kind of fluid. If they tried to dry the pads off, they got wet again before long. It’s possible that this fluid serves as the glue.

While glue (or, more technically, wet adhesion) may be useful for sticking to surfaces, it poses a challenge of its own: once stuck, a bat needs a way to get unstuck. It appears that sucker-footed bats produce a fluid that’s just sticky enough to keep them clamped to the side of a leaf. But they can peel away easily from one end. This technique may explain why sucker-footed bats defy the heads-down rule among roosting bats. If they tried to hold onto a leaf with their heads pointed down, they’d slide off.

The sucker-footed bats were not the only bats to evolve the Spiderman strategy. In Central and South America, there are four species of disk-winged bats (Thyroptera) that can also cling to leaves with pads. Riskin studied disk-winged bats in graduate school and demonstrated that they can form a seal with the disks on the leaves. In other words, they really do suck. Combined with his latest results, the lesson is clear: there’s more than one way to mimic Spiderman when you’re a bat.

Reference: Daniel K. Riskin and Paul A. Racey, “How do sucker-footed bats hold on, and why do they roost head-up?” Biological Journal of the Linnean Society, in press.

How To Be A Snake [Life in Motion]


If you stroke a snake, its skin feels slick and slippery (ed: or smooth, at any rate). Yet according to a new study by scientists at New York University and Georgia Tech, snakes actually depend on friction to move.

Snakes crawl by contracting the muscles that run along their body and pushing against the ground. Recently David Hu and his colleagues took a close look at that snake-surface interface. They anesthetized snakes and lay them on a board. By tipping one end of the board, they could see how well a snake’s body could hold onto the surface thanks to friction alone, without any extra forces generated by the snake’s muscles.

Hu and his colleagues discovered that snake scales can actually create a lot of friction by catching on tiny bumps on the surface they’re lying on. (They only feel smooth if you stroke them tailward.) The scientists found that the scales can generate twice as much friction if a snake is sliding forwards than if it is sliding sideways.


To see if they were right, the researchers built a mathematical model of a snake on the basis of their observations. They then changed some of the variables, such as the smoothness of the surface on which their mathematical snake crawled to predict how a real snake would perform.

Here, for example, is what happens when a milk snake tries to slither across a smooth plastic surface. Without any bumps on which it can catch its scales, it crawls in place.

Hu and his colleagues then let their snakes crawl on a rough surface, but first put them in a cloth sleeve.The snakes could push against the surface, but because they couldn’t lock their scales onto it, they again slithered in place.

The model Hu and his colleagues created slithered a lot like real snakes do, as shown in this simulation (the red dot shows the center of mass).

snake-gelatin.jpgBut the scientists recognized how they could make the model match reality even more closely. In their original model, the snake lay completely flat against the ground. That’s not how snakes actually slither. They only make contact with the ground at a few spots along the length of their bodies. This picture shows a snake crawling across a plate of gelatin. The photo is lit by polarized light, which creates bright reflections where it hits places where the snake is pushing against the gelatin. Rather than creating a long, snake-shaped stretch of light, the snake creates just a few patches where it is pushing against the plate.

The researchers decided to see what happened if they let the snakes in their model lift up their bodies the way real snakes do. Hu and his colleagues found that their snakes slithered 35% faster and boosted their efficiency by 50%. In this movie, the body is colored red whe the snake has lifted its body, and blue where it is concentrating its weight on the ground. The model works better because the snakes can press their weight only on the spots where the force of friction is highest in the backwards direction. By continually redistributing their weight, the snakes can slither as quickly and efficiently as possible.

A snake may look a little silly trying to crawl while wearing a sleeve. But such humiliations can help us appreciate just how graceful snakes really are.

Reference: David L. Hu et al, “The mechanics of slithering motion.” PNAS.

All images and videos copyright Grace Pryor, Mike Shelly, and David Hu/Georgia Institute of Technology. Source.

How To Be A Bat [Life in Motion]

Bat in wind tunnel from Carl Zimmer on Vimeo.

When the evenings get particularly thick with mosquitoes where I live, I sometimes sit out in the yard with my daughters and look up at the fading sky. Before too long, a single bat will usually flit out of the nearby trees and start flying circles around the house, scooping up bugs along the way. We can barely make out the bat’s wings as it takes its laps, a flicker of membranes. And so it was a revelation to spend some time earlier this week with two Brown University biologists, Dan Riskin and Sharon Swartz, watching slow-motion movies of bats in flight. There’s a lot going on up there.

Bats evolved about 50 million years ago from squirrel-like ancestors. They probably made their first forays into the air as gliders. Like living gliders, they used flaps of skin to increase their surface area, letting them glide further. Their hands evolved long spindly fingers that were joined by membranes. Some early bat fossils suggest that they may have shifted from gliding to alternating between gliding and bursts of fluttering. Eventually bats evolved sustained powered flight.

Bats evolved a way to take advantage of the same laws of physics birds use to fly. And many scientists who have studied bat flight in the past have basically treated bats like leathery birds. Yet there’s no reason to assume that this should be so. After all, it would not be surprising to find that the way the feathers on a bird’s wing react to air pushing against them are different from the way the stretchy membranes on a bat react. Birds don’t have wing surfaces connecting their front and back legs, like bats do. And while birds only have a couple joints in their wing skeleton, such as at the elbow and wrists, bats have lots of knuckles they could, in theory, bend selectively to alter their wing surface. Bats also have lots of sensitive hair cells on their wings that appear to track the speed and direction of the air flow, and the information they get from the hairs may help them make fine adjustments to their wings many times a second.

Bat on flower from Carl Zimmer on Vimeo.

And when scientists like Swartz and Riskin study bats, they discover, in fact, that bats are not birds. Bats fly more slowly than birds, but they maneuver more effectively. Bats fly cheap compared to birds. A hovering bat use 60% less energy than a hovering hummingbird. These sorts of discoveries suggest that if you’d like to make an agile, efficient, and tiny flying robot (and who doesn’t?) it might be worth looking for some inspiration in bats.

The problem with looking to bats for inspiration is that scientists are only starting to figure out bat aerodynamics. What’s really challenging to figure out, however, is the difference between the aerodynamics of birds and bats. Riskin and Swartz use lots of tools to find the answer. They paint bright dots on bats and then film the animals as they fly in wind tunnels. The biologists can then use computers to create models of the bat wings and calculate the speed and direction of each dot at each instant of flight. They can spray mist into a tunnel and then film the swirls the bats leave in their wake. From this data on real bats, bat researchers can then test out simulations on computers to see if they produce the same forces and swirls of air as they see in their wind tunnels.

Bat vortex from Carl Zimmer on Vimeo.

A close look at these movies reveals that bat flight is just too complex for simple labels, like upstroke and downstroke. The shoulder of a bat starts rotating upwards before the wrist, which move up before the fingers. The fingers on each hand don’t move in sync with each other. A joint on the left wing is often out of sync with the corresponding joint on the right wing.

Physicists like to treat wings as rigid surfaces because the math involved causes fewer headaches. But that’s a gross simplification when it comes to bats. The bones in a bat’s hands are surprisingly flexible, and the skin of the bat wing is never fully stretched out during its stroke. In fact, the region of the wing close to its body actually balloons out to double its surface area during each flight stroke. Bats probably use this ever-changing wing surface to control their lift and drag, so that they can make tight maneuvers without stalling.

Bats have clearly evolved a sophisticated flight system, but they face some awkward challenges when they’re not flying. Birds only need two limbs for flying, leaving their remaining two relatively free to land and walk around on the ground. Bats, on the other hand, make their hind legs part of their wings, and so natural selection has to strike a compromise between several different functions. And while birds can stop flying by using their feet to land on the ground, most bats have to use their feet to hang upside down.

To figure out how bats manage this feat, Riskin turned the typical biomechanics lab upside down. Scientists can measure the forces of a running animal by putting a force-sensitive plate on the floor; Riskin put his on the ceiling. In his recordings of landing bats, he’s discovered two strategies. In one species that lives in caves, the bats make an elegant backwards flip combined with an upside-down cartwheel, so that they can land with just two feet.

2 point landing from Carl Zimmer on Vimeo.

In a species that hangs from trees branches, the bats use a very different technique. They swoop in without a cartwheel, and bring both feet and both hands upward to grab onto the tree. And they hit the tree hard. The cave bats land with a force that’s twice their body weight; the tree bats generate forces as high as eleven times their body weight.

Bat making 4 point landing from Carl Zimmer on Vimeo.

This discovery (published this week in the Journal of Experimental Biology) illustrates an important fact about bats–a bat is not a bat is not a bat. Bats live in many environments and are adapted to eating many different kinds of food, from moths to fruit to cow blood. They’ve adapted to these different ways of making a living, in part by evolving different ways of moving around. If you’re a bat flying towards a wall of rock, you don’t want to hit it too hard. But if you can grab a branch that can absorb the shock, you can skip the fancy acrobatics.

That same lesson emerges from how bats behave on the ground. With their delicate legs yoked together by their wings, you might expect that bats don’t do very well on the ground. And indeed, most species won’t win any track and field medals. When Riskin puts a typical bat on a treadmill, they stumble around. If the treadmill goes too fast they start to lose all control. It’s likely, then, that the ability to walk efficiently and to run was lost in the early evolution of bats. But millions of years later, that ability evolved once more in at least two species.

One place where bats have taken to the ground again is New Zealand. The remarkable isolation of New Zealand left it without big predators and without any mice or other ground-dwelling mammals. One species, the New Zealand short-tailed bat, has adapted to this niche. While it can still fly, it now moves around comfortably on the ground in search of bugs, nectar, fruit, and pollen.

Riskin found that New Zealand short-tailed bats walk comfortably on a treadmill, using the same pendulum-like movements that other walking mammals use to save on energy. But when other mammals have to move faster, they break into a run so that they can store extra energy in their tendons as they hit the ground. The New Zealand short-tailed bat can’t make the transition from walking to running.

But another species of bat can make that switch. A vampire bat will walk on the ground to sneak up on its victim. If its victim tries to get away, it can scramble in pursuit. Riskin found that if he put vampire bats on treadmills, they can walk like New Zealand short-tailed bats. But when he speeds up the treadmill, they suddenly switch to a bizarre form of running. Instead of pushing off with their hind legs, like a squirrel, they use their long, heavily muscled arms. It’s a mammal version of front-wheel drive versus rear-wheel drive.

Vampire walking from Carl Zimmer on Vimeo.

Vampire running! from Carl Zimmer on Vimeo.

The difference between the two species of ground-moving bats is not surprising when you consider where they live. Bats on New Zealand didn’t pay any cost for evoling into slow walkers, because life was pretty easy (at least before humans showed up with their rats and other assorted camp followers). But vampire bats evolved in a more competitive environment where they had to adapt to moving prey.

Once bats evolved flight, in other words, they did not stop evolving. Their movements have been changing in astonishing ways for millions of years, and will continue to change as long as bats fly, walk, or run across the Earth.

The Flesh of Physics

Muybridge horse infinite repeat

Our bodies are bunches of atoms, and like any rock or star or other bunch of atoms, we have to obey the laws of physics as we move. But each species obeys those same laws in its own way. My cat leaps onto my desk most mornings, his grace unblemished by the paper clips and computer cables he kicks onto the floor. A maple tree outside bends in the wind, a happy medium between flopping over and snapping in two. A hawk arrives at the tree and lands precisely on a branch. On their own, our eyes cannot tell us much about the different ways in which living things move. We can’t see the invisible vortices of air spiraling behind a hawk, the stresses experienced by different parts of the leaning maple, the thrust and torque generated by my cat as he rises into the air.

The first glimpse into this invisible world came in 1872. Leland Stanford, a railroad tycoon and the founder of Stanford University, spent a lot of time watching his race horses run. He was sure that when they trotted, there were moments when all four legs left the ground. Legend has it that he even bet $25,000 that they did.

Stanford paid a famed landscape photographer named Eadweard Muybridge to find out if he was right. Muybridge had horses trot down a path strung with threads connected to a row of cameras; when the horses snapped the threads, the cameras snapped the pictures. It took Muybridge years to perfect a shutter fast enough and film sensitive enough to capture the images (he also needed some time off to defend himself–successfully– against the charge that he had murdered his wife’s lover). But in 1877 he was finally able to give Stanford his answer. Horses do indeed bring all their legs off the ground during each cycle of a gallop. Later, Muybridge built contraptions that could display his pictures in quick succession. His moving pictures brought the horses back to life.

Eventually Muybridge made his way to the University of Pennsylvania, where he photographed many other four-legged animals. He found that whenever they ran, they lifted all their legs off the ground at once. Even two-legged humans did. That complete lack of contact with the ground, in fact, came to define the act of running. Muybridge’s photographs also revealed other rules. When four-legged animals walk rather than run, their feet usually hit the ground in the same pattern: hind left, front left, hind right, front right. Here’s a diagram of the cycle in a walking horse.


Muybridge opened the way to the scientific study of life in motion. These days, biologists can film animals with high definition video cameras and use computers to calculate the speed and direction in which different body parts move. They can put sensors on animals or have them run over force-sensitive plates to measure the thrust they generate with their muscles. Instead of Muybridge’s flickering photographs of horses, we can enjoy their glacial grace in movies like this, from researchers at the Royal Veterinary College:

These superior tools have allowed scientists to discover some of reasons that animals move the way they do. The cycle of footfalls in a walking dog or a walking elephant, for example, is the best way to keep a four-legged animal stable. Walking is not just stable, but also efficient, because it turns animals, in effect, into pendulums. A pendulum can swing for such a long time because it continually recovers some of its energy. On its downward stroke, it’s powered by the force of gravity; when it reaches the lowest point of its arc it has so much energy that it can counteract gravity and swing upward. When you walk, your body behaves like an upside-down pendulum: the foot you plant in front of you is the pendulum’s axis, your center of mass the hanging weight. In the beginning of your stride you work against gravity, vaulting your center of mass upward with your leg until you reach your highest point. Gravity then takes over, and your body swings downward until your other leg hits the ground. The next stride is even easier. You can use the energy given to you by gravity to vault yourself into your second and all successive steps, just as a pendulum reclaims its energy in each swing.When you run, however, you stop behaving so much like a pendulum and begin behaving more like a pogo stick. Now when you first plant your leg, your body sinks down on it instead of rising up. Your leg actually acts as a brake for your body, and so your center of mass is at its lowest point when your acceleration is lowest. Meanwhile your tendons are acting as springs. As they stretch and snap back, they store and release energy, just like the spring in a pogo stick, and propel you upward and forward.There are many other ways to move around, of course. If you’re a cockroach or a centipede, you can use more than four legs. It turns out that many of the same rules that govern walking and running among us vertebrates also apply to invertebrates. Meanwhile, other researchers are discovering the rules behind other kinds of motion, like flying, jumping, and swimming.

For all the advances in biomechanics, however, it turns out that a lot of people still live in a pre-Muybridge universe. A team of biologists, biophysicists, and a veterinarian in Hungary recently did a survey of the depictions of animals in museum displays and other places. In each case, the researchers determined whether the poses of the animals followed the basic rules for how four-legged creatures move.The grades they handed out were pretty dismal. Museum displays were wrong 41% of the time. Taxidermy catalogs were wrong 43% of the time. Animal toys were wrong half the time. And, incredibly, coming in dead last were animal anatomy books–63.6% wrong.

Here, for example, is an illustration of a horse not being a horse. B is a diagram showing its limbs. C and D show two real poses it could have taken.


And here’s a picture of an aardwolf in a museum display doing what no self-respecting aardwolf would do.


I was surprised that there are so many biomechanical mistakes out there, especially on such a simple matter of how to position an animal’s legs. To be fair, a lot of the biomechanical misktakes in museums are baggage they carry from the past. Today museums are following Hollywood’s lead and are working with biomechanics experts.

John Hutchinson of the Royal Veterinary College has done some pioneering work on how dinosaurs walked, and his research is the basis of an exhibit called Be The Dinosaur. Here’s a sample of the computer simulations the exhibit has to offer.

Tyrannosaurus Walk Cycle from Tom Spilman on Vimeo.

I first became fascinated with biomechanics in the mid-1990s, and I often dreamed of embedding movies on the pages of my articles. Words could only go so far, and photographs couldn’t go that much further. Most of my futuristic dreams have not come to pass, or have proven to be banal disappointments. But when it comes to writing about biomechanics, the future is here, and it is good. This will be the first of what I hope is a long line of blog posts about life in motion, illustrated with moving images that Muybridge could not imagine.

Reference: Horvath et al.: “Erroneous quadruped walking depictions in natural history museums.” Publishing in Current Biology, Vol. 19, No. 2, January 27, 2009

Muybridge Portrait: Smithsonian

Aardwolf and walking horse courtesy of Gabor Horvath and Adelinda Csapo