Discover, November 1998Link
Nancy Simmons went down to the jungles of French Guiana in 1991 with the seemingly simple mission of determining how many species of bats live in one place. Each afternoon she and her colleagues headed out from their camp and set up their nets, sometimes arranging them artfully over the opening of hollowed tress where bats like to roost. Then they waited for the sun to fall.
"The forest is noisy and active at night," says Simmons. "Armadillos come walking by, and kinkajous are knocking around in the tress above your head. You sit there in the dark and you wait, and you never know what is going to come next." When the bats emerged from their roosts, Simmons would hear them thump into the nests. At times she could even hear the chatter of their teeth as they tried to gnaw their way free.
Simmons surveyed her site for three field seasons. She was overwhelmed by the results. "We ended up getting 78 species of bats from an area that was within a three-kilometer walk of our camp, just a pinprick on the map." One species had previously been reported only from a site 800 miles away in Brazil; another was altogether new. How so many kinds of bats can live in one place is an utter mystery. What stunned Simmons in particular was that her forest site, which has relatively poor soil, probably doesn't harbor much diversity. She's sure that there are many lusher places in the world where even more bats live side by side.
The great variety of bats seen today is just part of the larger evolutionary mystery surrounding these creatures. Simmons, who works at the American Museum of Natural History in New York, is a paleontologist by training. Before she entered the jungle to look for live bats, she had spent years studying a miserably scrappy record of bat fossils. Bats are not good candidates for paleontological study. When they die, they usually disappear. Often they are eaten by scavengers; if not, they decompose on the ground. The muscles and membranous skin that make up their wings rot away. Any wonderful equipment they might carry to navigate by the echoes of their own cries--a strange leaf-shaped nose, or giant frilled ears--vanishes. Soon all that remains are their frail, thin bones, which rarely end up in a place like a nice quiet lake bed where they can fossilize. As a result there are only a handful of complete bat fossils in the entire world.
Paleontologists like Simmons have studied those few relics assiduously because, living or dead, bats pose one of evolution's supreme puzzles. A few other mammals may be able to glide from tree to tree, but only bats have evolved true powered flight. In fact, only two other vertebrates--birds and pterosaurs--ever managed this feat. But with little fossil evidence to go by, it's hard to figure out how bats did it.
Judging from the bats around the world today, flight was a spectacularly good move, evolutionarily speaking. Biologists have identified 925 bat species--a quarter of all mammal species known. Among this diversity you can find, not surprisingly, a great diversity in ways of finding food. The bat most Americans probably know best is the little brown bat, which roosts in attics and barns and prowls for insects at twilight. Like many other species, little brown bats use echolocation to hunt. They shriek in pulses, and to avoid deafening themselves, they use muscles to freeze the bones of their middle ear during their calls. Between pulses, they relax those ear muscles and listen for the echo; the time that it takes the sound to return lets tie bat get a fix on a flying insect. They alter their flight path to intercept the insect, and snatch it up by extending a pouch running between their legs.
Other bats have an even more sophisticated location system. Leaf-nosed bats send their cries not through their mouths but through their nostrils, which are sculpted into ornate folds that help focus the sonic beams. And rather than sending out pulses and pausing to listen for the echoes, they emit almost continuous sounds. Instead of freezing their ear muscles, leaf-nosed bats are simply deaf to the frequency of their own calls; they can hear the calls' echoes because, via the Doppler effect, the frequency of the reflected sounds is shifted ever upward as the bats fly toward their prey.
But bats don't necessarily need echolocation to hunt. Some species use it only to stay aware of their surroundings. To find prey, they use more familiar modes of heating, listening for frogs croaking or for insects crashing through leaves; then they swoop down and pluck their meal from ground or tree, using their mouth rather than a leg pouch. Still other bats don't hunt at all. Some eat fruit, while others drink nectar from flowers, like hummingbirds.
Most bats belong to a suborder known as microchiropterans ("little bats"), to distinguish them from the megachiropterans, the giants of the bat world, with some boasting a five-foot wingspan. Generally, these larger bats prefer a diet of fruit. As a result, they are hugely important ecologically--by spilling seeds on the ground as they eat or by scattering them in their droppings, the bats help the fruit trees spread over a wide territory. Only a few megachiropterans can echolocate, and they do so very differently from the microchiropterans--in one species, for example, sound is produced by clicking the tongue. The rest navigate through the night solely with the aid of their oversize eyes.
The oldest bat fossils are 50 million years old, and paleontologists assume bats had been flying around for a few million years before that. Those fossils, found near the town of Messel in Germany and the Green River in Wyoming, are exquisitely complete. In both places the bats fell into lakes, where they were quickly covered by sediment and left undisturbed. These vanished species, sprawled on their mudstone slabs, dearly had what it takes to be a bat. Their arms are almost three times as long as their torsos, and their hands are elongated into fans that could have supported their wings. In younger rocks, however, the fossil record trails off into a mumble. "It's one of these strange cases," says Simmons, "where tight at the beginning we have great fossils and now we have this wonderful living diversity, and in between we don't know much aside from the fact that they were alive."
WHAT DID THE FIRST bat look like, and how did k survive? And how did bats diverge into the forms they take today? Piecing together the story of bat evolution requires researchers to determine in what order bats' marvelous equipment developed. In the 1980s one intriguing scenario was constructed by Brock Fenton, a zoologist at York University in Ontario. Fenton suggested that bats became echolocaters first and only later acquired the ability to fly. He pointed out that tree shrews, which are close relatives of bats, use ultrasonic calls to communicate with one another. It's possible that a shrewlike bat ancestor running through trees at night on four legs, catching insects to survive, might have found it could hear the faint echo of its calls as they bounced off a bug flying past. Its descendants might have started to reach for passing insects, and evolution would then have favored protobats with longer and longer arms, and with webbing between their fingers to help them catch their prey. Their calls would have become more focused and powerful, and they might have been able to start gliding from tree to tree. Eventually their descendants could fly under their own steam and catch insects in midflight.
That model came under fire, however, when John Speakman, a zoologist at the University of Aberdeen in Scotland, discovered how expensive an evolutionary development like echolocation can be. He and his colleagues were able to measure the metabolic cost of an echolocating pulse for a perching bat and found that it cranked up the animal's metabolism by a factor of nine. And that was for a stationary bat--how could a bat handle both that cost and the huge metabolic cost of flight simultaneously?
But when Speakman then measured the energy expenditures of bats while they are flying, he got a surprising result. "What we found out is that the cost of flying is no greater in an echolocating animal than in a nonecholocating one," he says. "There must be something fancy going on." Bats, it turns out, can use the energy required for flying to echolocate for free. "The wings come up passively--they are blown upward by the airflow--and as they come to the top of the wingbeat cycle, the animal has to contract its muscles in order to slow the wings down and reverse the direction in which they move." The contracting muscles not only move the bat's wings but also squeeze its rib cage, pushing the air out of its lungs and through its voice box, creating the echolocating pulses. These results have led Speakman to question whether the earliest bats could have evolved echolocation before flight, since it takes so much effort while perching. Instead he has suggested that flight and echolocation evolved hand in hand, improving by little steps together.
As with any transition in evolution, you can craft all sorts of plausible scenarios for how bats came to be. The only way to choose between them is to see if they match up with the way in which the evolutionary tree of bats branched into its many lineages. The genealogy of bats is a murky matter, though, and has inspired some pretty exotic ideas. In 1986, for example, John Pettigrew of the University of Queensland in Australia suggested that the megachiropterans weren't bats at all but were actually closer to primates. He based his idea on the observation that the big eyes of fruit bats are connected to their brains by bundles of nerves that strongly resemble those belonging to us big-eyed primates. According to Pettigrew, these primate cousins must have evolved their wings independently from microchiropterans.
Largely in response to Pettigrew's work, there was an avalanche of bat research in the late 1980s and the 1990s. Some people compared bat genes, others compared the patterns of muscles from species to species, others looked at their teeth. Simmons was one of those involved. With the help of paleontologist Jonathan Geisler, also of the American Museum of Natural History, she gathered together the findings of other bat experts and examined many species of bats herself. She studied the most complete fossils, comparing them with living bats, as well as with closely related nonbats, such as tree shrews and colugos, which are gliding mammals that live in Southeast Asia. Eventually she compared 30 groups of bats and bat relatives, using over 200 points of comparison--the most ambitious attempt ever undertaken.
Based on her results, Simmons adds her support to the idea that megachiropterans are indeed bats and not primates with wings, as Pettigrew claimed. The similarity of their wiring to that of the brains of primates, she says, is just a sign that both groups got their start using vision to live nocturnally. Thus all bats seen today, from large, big-eyed fruit eaters to small, echolocating insect hunters, descend from a common ancestor. Yet she also concludes that the megachiropterans did split away from other bats long ago, perhaps a few million years before the appearance of the fossil bats of Germany and Wyoming. Those fossil bats, she says, are actually on the path toward microchiropterans.
That means, Simmons points out, that the common ancestor of all today's bats and the known fossils could already fly. And that megachiropterans, while they have undergone a fair amount of evolution since those early days, are in some ways relic representatives of those first forms. Those ancient bats, she reasons, couldn't echolocate. The few megachiropterans that can echolocate today seem to have evolved the ability independently from their smaller cousins.
The earliest microchiropterans probably were a lot like the megachiropterans, depending on giant eyes to get them through the night. Gradually, however, they began to shift from vision to echolocation. According to Simmons, you can actually see the gradual rise of echolocation in the fossil record. Echolocating bats today use large muscles to control the shape of their throats while making their calls, and these muscles have to be supported on a bony strut, which in turn is anchored to the bat's skull by a pair of large bones. The most primitive known fossil bat, Icaronycteris, had these big anchoring bones as well. It also had a rather large cochlea--the shell of bone that houses the inner ear. Only bats that echolocate have both an oversize cochlea and oversize throat bones.
In other words, Icaronycteris could echolocate. But its cochlea wasn't all that big, suggesting to Simmons that it couldn't use its echolocation with much precision. Rather than focus on prey, early microchiropterans like Icaronycteris may simply have used echolocation to make out their surroundings. "They were probably using it for orientation and were listening for the sounds that prey make--the sound that an insect makes walking, or landing with a leafy, smashing sound."
A closer relative to today's microchiropterans is the 50-million-year-old Paleochiropteryx, from Germany. Paleochiropteryx, according to Simmons, shows that evolution had already endowed some bats of that time with a very sophisticated kind of hunting. Its cochlea was as big as those found in today's echolocaters, demonstrating that it could use sounds precisely enough to home in on prey. And next to Paleochiropteryx's heel, you can make out something that the more primitive fossil bats didn't have: the traces of a spur of cartilage. This spur, called the calcar, is an anchor for the netlike membrane that stretches from a bat's tail to its legs and that they use to catch insects.
That would mean that rather than representing a late embellishment, aerial insect hunting actually evolved in bats more than 50 million years ago. Scraps of fossils suggest that many of the families of microchiropterans were established within just a few million more years.
Taken together, all these clues point to an early development of flight--perhaps 60 million years ago--in a nocturnal common ancestor, followed by swift evolution of echolocation in the lineage that led to the microchiropterans, followed by an abrupt explosion of bat diversity. The ability to hunt flying insects at night may have given these early microchiropterans the opportunity to thrive in an ecological niche that no vertebrate had ever mastered before. They could specialize in different kinds of bugs and different kinds of strategies.
As for all the microchiropterans that don't hunt flying insects today, Simmons considers them to have actually reclaimed older ways of life. "At any time in evolution, there are a lot of ecological niches out there. There are some insects that are best located by echolocation, and there are other prey, like frogs, that are best located by listening to their call. The best way to catch katydids and grasshoppers is to listen to them crashing into the leaf litter. The fossil forms are long gone, and these other families, to take advantage of resources that nobody else was taking advantage of, are going back to listening to prey."
Simmons's work has earned praise from her peers, but some have doubts about the interpretation. "Animals don't leap off branches into the darkness in the hope that they're going to land somewhere," says Speakman. "They need some sensory development in order to do that." He also questions whether a big-eyed lineage of bats could give up their vision for the sake of echolocation. "If you're already committed to a visual system, you can perhaps cram in a small, rudimentary echolocation system, but you can't make the switch all the way over to a full-blown echolocation system." Speakman suggests instead that the earliest bats may have evolved flight during the day and then, for whatever reason, moved over into a nocturnal life. At that point, megachiropterans evolved powerful eyes, and microchiropterans evolved echolocation. "I've shifted the evolution of flight into the daylight so they don't have to leap in darkness," says Speakman.
The only way to test the ideas of Simmons and Speakman and others will be to get more evolutionary information. "Tomorrow somebody may come to me and say, `I have this animal that doesn't quite have wings, it's on its way.' That could happen," says Simmons. She would predict that such a protobat would have small ears and big eyes and oversize hands with membranes between its fingers--something like a colugo. But considering how hard it is to find bat fossils, it may be quite some time before that somebody comes to Simmons. In the meantime, she will be returning to the jungles of South America (this time Peru) to catch more living bats. There are enough mysteries flying around in the night to keep her busy.
Copyright 1998 Carl Zimmer