September 2008

Gossamer Wings and DNA

Tracking 2 million years of butterfly evolution in the Arctic Circle

By Marjorie Howard

In the Yukon in the late 1890s, gold diggers arrived with hopes of striking it rich in the Klondike. Last summer, Francie Chew found her own gold there in the grassy tundra, but the precious commodity she and her colleagues sought had fluttering wings.

Click on the play button to begin the audio slide show.

To view the accompanying slideshow go to

Armed with large nets, they caught white and yellow butterflies, part of their research to better understand the long-term changes in local species, which will help chart the evolutionary history of butterflies.

An energetic woman with a ready smile, Chew, a professor of biology, has been fascinated by butterflies and by biology for most of her life. In one corner of her office in Barnum Hall is her butterfly net. Plants line the window sills, and amid photos of students and family are pictures of butterflies. A butterfly mobile hangs from the ceiling.

She has long been aware that not everyone considers butterflies as fascinating as she does. “Butterflies evoke all kinds of comments from people, from ‘Why would you bother’ to ‘You study insects—eww.’

For Chew, butterflies offer a window onto how much of the natural world is interconnected. She spends most of her research time studying the effects of invasive plants on native butterfly populations, including the effect on butterfly and caterpillar behavior and growth when exotic plants take over a habitat.

Last summer, with funding from the National Geographic Society and Tufts’ New Directions Faculty Research Award, she journeyed from the Yukon Territory north to the Arctic Circle for a five-week camping trip, roughing it with a group of researchers and undergraduates from Stanford University.

Sleeping in tents and cooking outdoors, Chew and her colleagues were joined by her husband, Richard Bryan, an engineer “and a great butterfly catcher,” she says, who helped maintain the group’s electronic equipment, including a global positioning system that helped them navigate their way through the tundra.

Chew and her colleagues chased and caught butterflies, mostly in the genera Colias and Pieris, on the spongy tundra. “Catching butterflies is not so difficult, normally,” she says, “but when you’re on the tundra, which is made up largely of peat moss, it’s like walking on a sponge or an air mattress. You cannot run fast.”

The extended net handles created a long-lever arm, making it somewhat easier to catch the butterflies when running quickly was difficult. The butterflies were labeled and stored in a freezer fueled by liquid nitrogen.

Back in the lab, they get to work. “We take a little leg or a piece of the thorax or mid-section, which is very rich in DNA, and we amplify the DNA and sequence it,” she says.

Chew is interested in how glaciations affected butterfly populations in North America. Over the past two million years, four major glaciations had profound effects on the continent. As the glaciers moved south, the butterflies that survived did the same. Later, when the glaciers retreated at the end of the last ice age about 15,000 years ago, “some of those butterflies that moved south recolonized northern regions,” Chew says.

But in parts of northwest Canada and eastern Alaska, glaciers never formed—it was too dry. In this area, known as Beringia, scientists have found traces of pollen left by grasses and other plants. “You can take pollen from bog or lake sediment, and people can identify what plant species were there,” she says. “The pollen tells us there was a good deal of vegetation and that it was happy enough to reproduce. Perhaps butterflies and other insects were also there.”

It is believed that during the ice ages, the Beringian butterflies and the ones pushed south by glaciers branched off from each other. Using DNA sequences, scientists can measure how long different groups of butterflies living in different places may have been isolated from each other.

“We’d like to know, for example, which species of butterflies in North America are most closely related to which other species,” Chew says. “DNA sequences are a way to reconstruct what we think happened by way of asking: how different are the DNA sequences in this species from the DNA in other species?”

The clues from the butterflies Chew and her colleagues collected will help reveal how and where the butterflies evolved, but the work will take several years to complete. Chew is planning to use a sabbatical next year to go to Stanford to work with colleagues on the DNA sequencing. Given the thousands of years of history she’s studying, a few more is not too much to wait to solve this puzzle of butterfly evolution.

Marjorie Howard may be reached at

Article Tools

emailE-mail printPrint