Scott Edwards photographed in the ornithology collection

Scott Edwards in the MCZ ornithology collection, a kiwi at his elbow. A cassowary stands on a nearby table. | Photograph by Stu Rosner

How Birds Lost Flight

Scott Edwards discovers evolution’s master switches.

How did emus, ostriches, and kiwis end up flightless? What chain of events resulted in these birds diverging from the species that soar through the air and moving to a wholly terrestrial, and very successful, existence? For a long time, scientists thought that the crucial changes occurred in a distant ancestor shared by these birds, a creature that lived about 80 to 100 million years ago, before the rupture of the Gondwanan supercontinent. With that landmass’s breakup, the theory went, populations of flightless birds were separated and spread around the globe. This helped explain why species divided by oceans shared this rather peculiar trait. Flightlessness ran in the family. So when scientists discovered some years ago that DNA evidence made it unlikely this ancestor had ever existed, it raised some serious issues. How could evolution have shaped all these winged yet earthbound birds, all so apparently similar, without a common descent? It was a question that fascinated Scott Edwards, Agassiz professor of evolutionary biology and curator of ornithology at the Museum of Comparative Zoology. And few people were better placed to tackle it.

In the fall of 1986, having graduated from Harvard College in June, the 24-year-old Edwards bought a car and drove it 3,000 miles around the Australian outback, collecting hundreds of birds. By then a graduate student at the University of California, Berkeley, he was studying, among other species, the grey-crowned babbler, a sprightly bird with a barking call. In these birds’ genes, their history was written—if he could figure out how to read it.

Berkeley at that time was an epicenter of the nascent field of molecular evolution. Until recently it had taken months to sequence a gene, but clever, imaginative people like Allan Wilson, a professor of biochemistry, had glimpsed what was possible. Humans and chimps were genetically 99 percent identical, wrote Wilson and his student Mary-Claire King in 1972, suggesting their last common ancestor was not too far in the past. He and his colleagues, including 2022 Nobel laureate Svante Pääbo, were some of the first to use a new type of swift sequencing, drawing on the polymerase chain reaction, or PCR, on DNA from ancient human remains, eventually aiding in the assembly of evolutionary trees for Homo sapiens. Edwards knew he wanted to use these new tools on birds. Birds have a particular magnetism—these creatures that breathe with nine air sacs, have relatively enormous brains, and are known to dance as humans do. “They are eerily like mammals—I think we see part of ourselves in them, especially their behaviors,” said Edwards. In the study of evolution, birds had the potential to provide a comparison to the evolution of humans, a chance to see whether universal forces might be acting on both.

With samples from babblers, Edwards revealed that with just a small scrap of genetic material, he could see where populations had diverged from each other, helping to illuminate how recently they had been one. In the decades since he has sequenced whole genomes, laid the groundwork for the modern practice of building family trees for species, and, in his latest tour de force, shown the process by which flight can be lost, over and over and over again, by surprisingly deft mechanisms. In the process, he has developed tools that could uncover other instances of convergent evolution—when traits in unrelated lineages evolve separately to resemble each other—in any group of organisms. “Scott has always worked right at the cutting edge of molecular evolution,” said Mary Caswell Stoddard, an ornithologist and professor of evolutionary biology at Princeton University. Edwards keeps his eyes on the horizon, on the cusp of what’s scientifically possible.


A “Titanic” Mentor

Edwards grew up in the leafy neighborhood of Riverdale in the North Bronx, and on the banks of the Hudson River, there was bird-spotting aplenty. A neighbor took him out birding the first time when he was in sixth grade. “That neighbor’s probably long since stopped birdwatching,” said Edwards wryly. “He wasn’t one of those hardcore birder types.”

But for Edwards, the hobby persisted, encouraged by his mother and aunt. It came to his rescue at Harvard, when a punishing experience with organic chemistry prompted him to take a gap year. “It was the only C on my transcript,” he recalls. “My ego was crushed.” He volunteered in the collections at the Smithsonian Institution, working at a bagel shop to pay the bills. Then he went to Hawaii, where he volunteered at a national wildlife refuge, measuring the eggs of seabirds, and to California, where he surveyed birds in Six Rivers National Forest. He got a taste of what it meant to be a scientist, so that by the time he returned to college, he knew this was what he wanted to do. After graduating in 1986, he went on to Berkeley, where he worked with Ned Johnson, curator of ornithology at its Museum of Vertebrate Zoology.

Edwards was on the hunt for a new way to peer into the genomes of birds.

Berkeley was also where he met Allan Wilson, who was already a titanic figure in biology. Wilson had described the possibility of molecular clocks, using the rate at which DNA tends to mutate to calculate how long it had been since species or individuals had shared a common ancestor. He’d been working on the idea of a “mitochondrial Eve,” the concept that all humans are descended from a single woman who lived more than a hundred thousand years ago. But he was also deeply interested in birds, said Mary-Claire King, now a professor of genome sciences at the University of Washington.

“They were very close,” said King of Edwards and Wilson. “They loved talking about birds and the evolution of birds, how one could explore this in a molecular way….I remember thinking, Oh, this is going to be perfect. Allan finally has a scientific son, someone who cares about exactly the things he cares about.” The year before Edwards graduated, Wilson passed away from complications of treatment for leukemia. But his way of thinking shaped how Edwards thought about science.

“He was at the height of his powers,” Edwards said. “Everyone’s pretty convinced that if he was still living, he would have shared the Nobel Prize. It made me realize how fleeting not just your life but your scientific career can be. You just want to make sure every project, every paper [is] impactful.” Wilson set a standard, too, for making bold claims using both data and a shrewd intuition. “He lived on the edge like that,” said Edwards. “It was very exciting.”

Edwards himself was on the hunt for a new way to peer into the genomes of birds. The immune system offered a powerful point of entry, he realized. Every vertebrate has a set of genes called the major histocompatibility complex, or MHC. When a pathogen attacks a cell, MHC proteins alert the immune system by gripping fragments of the invader and displaying them on the cell’s surface. In mammals, there are more different flavors, or alleles, of MHC genes than of any other gene known to science.

Did birds have a similarly broad variety of alleles? Edwards wondered. And could the MHC, taken as a microcosm of the genome, tell him about birds’ evolutionary dynamics?


Evolution in Real Time

The red-winged blackbird’s three-note call is a familiar sound throughout North America, its red and yellow shoulders a sprightly counterpoint to the drab winter grasses of the marsh. When Edwards was a new professor at the University of Washington in 1994, he and his students painstakingly sequenced portions of the red-winged blackbird’s MHC. They were generating what were then very long sequences. What they saw was intriguing. “No one had sequenced an avian genome,” he said. “And it was like, wow, this is complicated—there’s lots going on here.”

The songbird’s MHC was a kaleidoscope of activity, its sequences incredibly varied and its structures complex. There was a great deal of noncoding DNA sprinkled throughout the genes, the team found. What’s more, they saw signs of a process called concerted evolution, in which one gene duplicates itself and then uses its duplicate to replace, wholesale, another gene in the genome. It was a mysterious finding, because the MHC seems to require diversity, while concerted evolution reduces it. There were other ways, Edwards and his students eventually discovered, for these genes to maintain their diversity. They found, as well, that avian MHC genes were indeed as varied as those in mammals. Evolution seemed to be shaping vertebrate immune systems in similar ways.

Scott Edwards with students in the ornithology collection
Edwards with students in the ornithology collection in 2018 | PHOTOGRAPH COURTESY OF SCOTT EDWARDS

Around the time Edwards began to work on the MHC, a bacterial pathogen called Mycoplasma gallisepticum leapt from chickens into wild birds in North America, hitting the U.S. house finch population hard. “The pandemic took out about half the eastern population,” said Geoff Hill, an ornithologist and professor of biology at Auburn University. “The whole population structure changed because of the pathogen.” It was a moment when scientists could see in real time the kind of evolution that shapes genomes. Edwards, who had sequenced portions of the house finch MHC already, worked with Hill to study how both the birds and the pathogen were altered by the experience.

Remarkably, their work uncovered one of the fastest mutation rates ever observed in bacteria. Analyzing samples of the bacteria from the first 12 years of the disease’s ravages, Edwards, Hill, and their colleagues found that the pathogen possessed just 2 percent of the diversity present in the original chicken version. When pathogens cross into a new species, the work suggested, vast portions of their genome could turn over almost immediately.

As fascinating as the glimpses of evolution provided by the immune work were, by 2003, when Edwards moved to Harvard, he was prepared to work on a still larger scale. People were starting to talk about whole genomes. For his dissertation in 1992, Edwards had sequenced a genetic region about 400 base pairs long. In contrast, by 2003, the Human Genome Project had sequenced nearly all the 3 billion base pairs of the human genome. Could whole genomes be produced for birds as well? “For a long time, it seemed out of reach,” said Edwards. But in 2010, as gene sequencing technology advanced rapidly, he was part of the team that published the whole genome of the zebra finch, only the second avian genome ever sequenced, after the domesticated chicken.

With whole genomes increasingly commonplace, the kind of work Edwards did began to change. “One of the central tools in evolution is a phylogeny, the relationships between the species,” he said. “Once you’ve got that, you can study all kinds of different traits—plumage color, body weight, longevity—and lots of different kinds of questions. At this point, the genome was the playground.”


Understanding Flightlessness—and Beyond

Imagine a phylogeny as a family tree. All organisms alive today are at the tips of branches, but trace the branches back, and they intersect with others, each junction representing the last common ancestor of the group. During the last 20 years, Edwards has worked to understand these branchings in birds and other organisms and the forces that drove them.

For example, evolutionary biologists found something disturbing early in the process of building phylogenies from genetic information. If they built a family tree for a species using one gene—if they diagrammed the divergence of descendants from an ancestral version of the gene over time—and then did the same with a separate gene, the trees might not be the same. Which tree was the right one? Were two species siblings, as one gene might indicate? Or distant cousins, as indicated by another?

Edwards was able to help resolve this conundrum. In 2009, he published a landmark paper explaining that discrepancies of this sort are to be expected, because of the vagaries of genetic drift—sometimes alleles are lost in one group and not another, and these odd-seeming trees result. Neither tree contains the real story of the species. The best way to resolve what truly happened is to build a tree using many genes—ideally the whole genome. His insights have fundamentally shaped the study of evolution, said Sarah Otto, an evolutionary biologist and professor at the University of British Columbia.

Colleagues say Edwards continues to be the person who brings the very latest in population and molecular genetics to the study of birds, as well. Increasingly, he is collaborating with statisticians to advance the study of genetics. And a wave of whole bird genomes published during the past 10 years has enabled Edwards to start working to understand the genes that control flight.

When it became clear there could not have been a single flightless ancestor for all of today’s flightless birds, Edwards saw a way to get some answers. “I thought, why don’t we look at this question across the whole genome?” he said. “Not only could we confirm or refute this new phylogeny, but we could actually see what genes were driving the loss of flight.”

Using 11 newly sequenced whole avian genomes, he and his collaborators looked for areas that were conserved across birds broadly. Then they zeroed in on regions that were changing rapidly in flightless lineages, presumably linked to the loss of flight. These regions tended to amplify or silence the expression of particular genes, they found. And in lab experiments with chickens, the regions seemed to control the development of fore and hind limbs, altering them in ways that may be linked to flight. Something as complex as flying could, in fact, be switched on and off without radically reshaping the genome, the results suggest. Changes in these master switches could pivot the body plan of an entire species.

Edwards and his collaborators published their results in Science in 2019. Their work suggests that not only has flight been lost many times, it’s been lost in flightless birds around the world through similar types of changes. “It was such a beautiful paper,” said ecologist Gunnar Kramer, who joined Edwards’s lab as a post-doctoral fellow not long after.

That work led Edwards and his Harvard colleague Jun Liu to develop software enabling the detection of hitherto unknown master switches in genomes of all kinds. Kelsie Lopez, a graduate student in Edwards’s lab, points to these innovations as a major advance. “The whole point is trying to find these regulatory regions, which are really complex,” she said. “The tools, the application, and the work are important for the bigger picture.” Working together with Emma Farley, a professor at University of California, San Diego, Edwards and Liu have been delving into what makes these enhancer regions work. “What’s the grammar of these enhancers? What’s changing in them to cause changes in gene expression?” Edwards said. The results might well apply to species beyond birds.

Edwards’s group is also exploring the construction of pangenomes: using sequences from many individuals within a species or population to see what might be glossed over in a whole-genome sequence from a single specimen. Lopez and Bohao Fang, a post-doctoral researcher in Edwards’s lab, are both working on different pangenome projects. Most pangenome work so far has been performed by scientists studying bacteria and crop plants. But Fang remembers advice Edwards gave his lab group about choosing research ideas: “Genomics moves fast. If it could be done five years ago, think again.”


In 2020, as the world stood still, Edwards was in motion. More than 30 years after he drove alone through Australia, he got on his bicycle and set out across the United States with a Black Lives Matter sign on his bike. George Floyd had just been murdered. Edwards had been thinking about such a trip for a while, and it felt like the right moment, as an African American, to do something.

Scott Edwards on his bike while cycling across the United States
Edwards in Wyoming during a solo bike ride across North America in 2020. | Photographs courtesy of Scott EdwardS

“It was a time when I think a lot of people felt a little helpless in terms of trying to understand what was going on with the social fabric of the country,” said Kramer. “And Scott did this really bold thing and took an action. It was really powerful.”

All throughout his career in science, Edwards had been aware of an absence. “At Berkeley, there were not a lot of African American students in zoology,” he recalls. “I was aware of it, but it didn’t weigh heavily on me then because, working in Allan Wilson’s lab, he had a pretty diverse lab and progressive approach.”

But over the years, that sense that few other people of color were following this path grew. “As I progressed in my career, I’ve become more aware that we need to do something,” he said. More than 20 years ago he began working to fund students of color traveling to the scientific meetings he attends. Alongside his work on evolution, he has been a persistent force for inclusion in science.

And so, decades after his solo wanderings on the other side of the globe, Edwards spent 76 days going coast to coast in the summer of 2020 with his BLM sign. Sarah Otto set up a GoFundMe, and Edwards raised more than $60,000 to support minorities in science. “It was pretty much just like jumping off a cliff or a diving board,” Edwards said. “You’re not sure how it’s going to turn out. But you think, I have to do this now.”

It was an experience that felt essential to him, a boundary, like so many he has encountered in science and in life, that needed to be crossed. He met people he never would have met otherwise, and all along the way, he spotted birds.

“I’m definitely,” he reflects, “an explorer at heart.” 

Corrected February 25, 2024: the common ancestor of the flightless paleognaths lived about 80 to 100 million years ago before the breakup of the Gondwanan supercontinent, not 200 million years ago, before the breakup of the Pangaean supercontinent.

Science journalist Veronique Greenwood wrote about the microbiome in the November-December 2023 issue.

Read more articles by Veronique Greenwood

You might also like

Harvard Overhauls Disciplinary Procedures

To cope with violations of University statement on rights and responsibilities

Harvard’s Development Chief Departs

Brian Lee to step down at end of 2024

Immigrant Workers— America’s Engine?

Harvard economist Jason Furman on immigration and the U.S. economy.

Most popular

Mechanical Intelligence and Counterfeit Humanity

Reflections on six decades of relations with computers

Harvard Overhauls Disciplinary Procedures

To cope with violations of University statement on rights and responsibilities

The Placebo Phenomenon

An ingenious researcher finds the real ingredients of “fake” medicine.

More to explore

Relabeling Medical Definitions for Obesity in the United States

For obesity patients, improved treatments and a nuanced understanding of the disease may lead to better health.

How Was Brooklyn Bridge Park Planned?

Michael Van Valkenburgh and the making of Brooklyn Bridge Park

The Mystery Behind an Incan Tunic

Unraveling an Inca masterpiece’s secrets