On Gossamer Wings

To Duke graduate student Joshua S. Rose, who studies their ecology, these elegant predators are as poetic in form and function as their species names. They fully deserve to be called by their mythical-sounding moniker: dragonflies. Rose enjoys describing the talents of his chosen study subjects. "They're incredibly agile, which is particularly amazing because fossil dragonflies from the Jurassic period of dinosaurs don't look much different from dragonflies today," he says. "And yet, you'd think that the true flies that are their prey, and which evolved much more recently, might be more maneuverable. But the dragonflies are just a heck of a lot faster and more powerful."

Modern dragonflies, fortunately, are much smaller than their ancestors, says Rose. Fossil dragonflies dating to the Permian period some 250 million years ago boasted wingspans of two-and-a-half feet. If those ancient adults were formidable, the dragonfly larvae, called "nymphs," were downright horrific, he says. Stalking the floors of ancient forests, the voracious nymphs measured more than a foot of downright nastiness. Beneath their heads nestled a hinged, armlike structure called a "labium" that ended in fanglike pincers.

Fans of the Alien movies, whose hellish monsters sport similar extensible dentition, will appreciate the deadly function of such a dining appliance. Like the movie aliens, a nymph encountering prey slashed out with its labium, snatching the hapless animal back to shred it in powerful mandibles. "If these ancient forms were still around today flying in the air, and their nymphs stalking the land, many biologists believe humans would still be living in caves," says Rose.

Fortunately, today's dragonflies are definitely benign, at least to larger animals. Belying their nicknames of "devil's darning needles" or "horse stingers," they do not bite. And, although the nymphs have retained their rather alarming eating habits, almost all species are now aquatic, spending their infancy in ponds and streams terrorizing worms, tadpoles, and small fish. The modern nymphs have retained their ancestors' decidedly uncouth breathing method of drawing water into their anuses to oxygenate gills inside their rectums. What's more, to escape danger, they can jet-propel themselves by forcefully expelling that water--bringing to mind a rather crude phrase occasionally used by humans as an insult. Fortunately, when adults emerge from their larval exoskeleton, they develop a more aesthetic breathing mechanism, drawing air through trachea along their body segments.

As fascinating as dragonflies are in their own right, Rose has found them invaluable subjects for his ecological studies. Unfortunately, observing the creatures has been no easy task, and the hearty scientist has needed all his energy to trek through fields, swamps, and forests in Florida, Texas, and Hawaii, tracking the elusive creatures. Unlike such "convenient" insects as butterflies, which flit obligingly about flowers, dragonflies--among nature's fastest insects--zoom about randomly in search of tasty in-flight meals. And these aerobatic insects have never been raised successfully in captivity, ruling out laboratory studies.

Despite the creatures' wandering habits and occupation of often-unpleasant terrain, Rose has remained undeterred. "One thing about being a field biologist is that you tend to get a certain affection for stomping around and getting wet and muddy and getting bitten by mosquitoes and poison ivy, and whatever else happens to be out there," he says, gamely.

The ecological mystery that compels him to chase dragonflies is the question of what happens to alien species when they invade new territory. It's an especially important ecological question given the widespread mixing of species in today's global biological mixmaster of international transport. "The most well-known cases are exotic species that are introduced and wind up wreaking havoc on the native ecology," he says. He cites examples ranging from the aggressive Japanese plant kudzu, known popularly as "the plant that ate the South," to the brown tree snake, which invaded Guam and virtually wiped out native forest birds.

Rose's dragonfly research takes a different tack from most studies of alien species. "In my work, most people assume that I'm looking at how the new species affect the dragonflies that are already there," he says. "But I'm actually looking in reverse. I'm looking at how the new species are affected by all the species that are already there." He has taken this unusual approach because of the reality that invading species can have a more complex effect on ecosystems than just endangering them. "It's usually assumed that any exotic species that gets established is going to be a pest, because when you hear about exotic species, you hear about the ones that have run roughshod over native ecosystems and done horrendous damage. But there are a lot of exotic species that don't really do much to the native species, and to some extent can even coexist with them."

The development of this coexistence by the new species, he says, can offer useful ecological lessons. "From an evolutionary point of view, a species colonizes a new area and then adapts to that new area and the species it's coexisting with. So, I'm looking at a range of dragonfly species in a given area, including the invading species, and trying to detect the first hints of adaptation."

Thus, Rose has waded through the Everglades of Florida, tramped the Rio Grande Valley of Texas, and pushed through the tropical forests of the Hawaiian Islands to gather data for his studies. And it has been extraordinarily hard, he says. Not only must he track up to twenty species of the peripatetic insects at once, but he must also record as many details as possible about their habitats and the "ecological space" they occupy. For example, he must detect whether a given invading species might change its preferred perch in grassland as little as a few inches in height. "I'd probably be very close to finishing my Ph.D. at this point, except that my original plans for statistical analysis were based on getting good data on, say, how high the animal is and what percentage of time it spends over water versus dry land," says Rose. "But once I got out into the field, those idealistic notions of mine went out the window." He discovered that standing in a dragonfly-rich environment meant tracking a confusing swirl of dozens of insects and trying to scribble down up to ten pieces of data on each one.

Now, after years of soggy feet and writer's cramp, he has gathered enough data to begin a statistical analysis to attempt to detect changes in the habits of invading dragonfly species. "If there is a change, the next step is to determine whether this is an actual adaptation--something that's a permanent evolutionary change in the dragonfly--or whether it's what biologists call 'phenotypic plasticity,' where the dragonfly has a certain amount of flexibility in its ecology that it can change from day to day without any permanent new genetic or evolutionary consequences."

In his research, Rose represents what is both old and new about science and scientists. His approach is "old" in the sense that it reflects the kind of observational study done before the age of the DNA sequencer and the computer. "Observational research was the accepted mode of research a century ago, and there's still a lot to be learned from it. Unfortunately, it's looked down upon in many university settings, where the trend is toward controlled laboratory experiments."

Rose's research is also "new" in that it recognizes that nature cannot necessarily be reduced to a laboratory experiment. "Basically, life is multivariate. One argument used by people like me who are going back to observational research rather than hard-core controlled experiments is that many scientists have tended to make conclusions black or white, with no overlap between them. So, when they look at the ecology of animals like the dragonfly, they might try to see whether it's determined by one factor, eliminating all the others. But in real life, that's not the case. Ecology and behavior are affected by the interaction of a whole bunch of influences."

Fortunately, says Rose, Duke's biology department has offered him the freedom to foster such independent thought. "They have a very strong belief here in letting the graduate student pick his or her own research projects. That's not true in a lot of schools. They'll either hand a student a project for them to do, or they'll just pay lip service to independence for graduate students, but not give them quite enough rope to hang themselves."

In fact "hanging themselves" is what all creative scientists do--that is, take chances on truly creative projects that may or may not work. "That was the case when I got here," says Rose. "I came up with numerous ideas for research, and almost all of them got shot down. But finally I came up with the idea for studying dragonflies." In particular, he says his adviser, biology research professor Peter Klopfer, offered just the right balance of guidance and independence to foster the young scientist.

In his teaching, Rose uses as an object lesson the rather bizarre habits of dragonfly nymphs. He recounts his delight in introducing the nymphs to students in associate professor Alexander Motten's biodiversity class. "We'll bring a live dragonfly nymph into the lab the week we study insects," he recalls, "and as the students watch, we drop a worm in front of it and, bam! In the blink of an eye the worm goes from swimming around to being chewed up by the nymph. The students are really amazed that such bizarre alien beasts are crawling around in their backyard garden ponds and local drainage ditches."

Rose hopes to see some of the rarer species of dragonflies he has yet to encounter in his travels, like the cobra clubtail, a native of North Carolina streams and rivers. "There's a whole family of dragonflies that have swollen knobs at the end of their tails," he says. "And the cobra clubtail gets its name because, when it raises its tail up, the tail looks like the spreading hood of a cobra."

For Rose, such spectacular species also illustrate how humans have failed to respect the natural wealth of species on the planet. "Clubtails need clear running water with a high oxygen content," he points out. "But there's hardly a river or stream in North Carolina that hasn't been dammed or channelized and lined with concrete. So, many of these clubtails are becoming harder and harder to find."

More broadly, he says, all of the thousands of dragonfly species have yet to be identified. "There was a new dragonfly species discovered in this country only a few years ago. So, if we don't know what's in the United States, imagine what we might find in Brazil or Colombia or Peru."

"There's definitely a whole new frontier of knowledge out there, but part of the problem is that as universities angle more toward molecular biology and laboratory experiments and away from observational natural history, there aren't many scientists left who are going out into the field and looking for these species," he says. Such knowledge might even lead to practical benefits, says Rose, citing a case in Southeast Asia in which dragonfly nymphs were added to drinking-water barrels to eat disease-carrying mosquito larvae. With some adaptation, dragonflies might even find use in mosquito control in this country, he says. "Local dragonflies wouldn't work because they breed in wetlands, while mosquitoes breed in small containers. But one could envision importing tropical species that breed in the small amounts of water, such as tree holes."

Rose and his fellow scientist-naturalists see dragonflies as significant beyond such a mundane use as mosquito eradicators. The ancient, elegant insects--those known and the multitude yet undiscovered--constitute an object lesson about human ignorance of the stunning diversity of our planet's creatures. And they offer a humbling lesson about the power of evolution to engineer creatures so perfectly adapted that they have survived for millennia, far longer than humans have walked the Earth.