# News this Week

Science  28 Jan 2011:
Vol. 331, Issue 6016, pp. 382
1. # Around the World

1 - Basel, Switzerland
Melanoma Drug Extends Life
2 - Washington, D.C.
JPL Scientists Subject To Background Checks
3 - Tokyo, Japan
Solar Sail Mission Sails On
4 - Cairo, Egypt
Dead Queen Still Kicking Up Dust
Fight Rages On Against Invasive Rats
6 - Houston, Texas
7 - Washington, D.C.
A Call for Stem Cell Researchers to Share

## Basel, Switzerland

### Melanoma Drug Extends Life

The drug company Roche announced that an experimental drug increased the survival of patients with advanced skin cancer whose tumors had a specific mutation in a gene called B-RAF. RG7204, developed by the biotech company Plexxikon, targets a protein encoded by the mutated gene that drives cancer growth in about half of all patients with malignant melanoma. About 40,000 people worldwide die annually from this aggressive cancer, most less than a year after being diagnosed.

In the closely watched phase III trial, patients with the B-RAF mutation who received the drug lived significantly longer and their tumors grew more slowly compared with patients who received a standard chemotherapy drug, Roche officials announced in a press release. But the company did not disclose how much longer patients lived on average. Full results will be reported later this year at a meeting. Patients receiving the standard drug will now be offered the option of switching to RG7204.

## Washington, D.C.

### JPL Scientists Subject To Background Checks

The U.S. Supreme Court has upheld the use of background checks by the government on scientists and other workers at NASA's Jet Propulsion Laboratory (JPL). The decision caps a legal battle begun in 2007, when 28 scientists and engineers at JPL, which is owned by NASA but operated by the California Institute of Technology in Pasadena, filed suit against new screening procedures announced by NASA. They argued that questionnaires asking about drug use, counseling, and “trustworthiness” were too intrusive and harmed the open environment at JPL. As contract employees, they said, they should be subject to less scrutiny than government employees who work with classified material. An appeals court agreed, but the Supreme Court unanimously disagreed.

NASA will now have to decide whether to reinstate the policy, which has been on hold since 2007. http://scim.ag/NASA-checks

## Tokyo, Japan

### Solar Sail Mission Sails On

The IKAROS solar sail mission was formally extended, having passed all performance tests planned for its original 6-month life. Launched by the Japan Aerospace Exploration Agency (JAXA) last year on 21 May (Science, 7 May 2010, p. 677), IKAROS, with its 20-meter diagonal, 0.0075-millimeter-thick polyimide sail, became the first craft propelled through space by the pressure of photons in sunlight. “We achieved complete success,” says mission manager Osamu Mori of JAXA's Institute of Space and Astronautical Science in Sagamihara, near Tokyo. JAXA will fund the mission for another year so scientists can try to pull off some advanced navigational tricks, such as varying the sail's angle to the sun. Such experiments will help in planning more ambitious solar sail missions, says Mori. Separately, NASA announced last week that its solar sail mission, NanoSail-D, which was thought to be malfunctioning, successfully deployed its sail and is working as planned.

## Cairo, Egypt

### Dead Queen Still Kicking Up Dust

Egypt's Supreme Council of Antiquities demanded the return of a famous bust of Queen Nefertiti that has been in Berlin's Neues Museum since its discovery in 1912 by German archaeologists. Egypt has been trying to get her back since the 1920s—Adolf Hitler refused to send her back in the 1930s—but now council chief Zahi Hawass has made the request in writing to the Prussian Cultural Heritage Foundation, which oversees the museum. Egypt maintains that the discoverer misled Egyptian authorities after its discovery, but the initial German reaction was dismissive.

### Fight Rages On Against Invasive Rats

Conservationists have stepped up their war against alien rats in the Galápagos. Officials with Ecuador's Galápagos National Park announced that with conservationists from various nonprofit organizations, they had begun carpet-bombing the archipelago's smaller islands with rat poison systematically released from a helicopter. Rats first arrived as stowaways in Western sailing ships and are a problem because they eat native tortoise and bird eggs. Although conservationists have been killing rats for years using bait and traps, the new strategy aims for “100% eradication” from nine islands and islets, including Jervis and Beagle islands, the officials announced in a statement. To protect a native bird that might otherwise eat the poisoned rats, conservationists captured 20 Galápagos hawks and plan to keep them in captivity for 2 months. They said that “mitigation steps will be taken” to protect the sole endemic rodent, a mouse found on Santiago Island.

## Houston, Texas

When MD Anderson Cancer Center in Houston, Texas, admitted some cancer patients from the United Arab Emirates (UAE) for treatment, officials there had no idea it would lead to the largest gift in the hospital's history. Now, the Khalifa bin Zayed Al Nahyan Charity Foundation, formed by (and named after) the UAE's president, has announced it is giving MD Anderson $150 million. The money will be used for a new building and work in pancreatic cancer and personalized medicine. Some$25 million will go to study biomarkers, genetic patterns in tumors that could guide treatment for individual patients. “There's been a lot of stops and starts” in the biomarkers field, with many not panning out, admits Raymond DuBois, the cancer center's provost and executive vice president. “We're at a critical stage now,” and this infusion of money, he says, will surely help.

## Washington, D.C.

### A Call for Stem Cell Researchers to Share

Researchers working with stem cells should follow the example of their colleagues in genetic sequencing and clinical research, setting up global networks for sharing data, materials, and intellectual property, according to a report released by an international consortium on stem cells and ethics. The Hinxton Group recommended setting up a publicly available global stem cell registry that would include a cell line's characteristics and information on how it was derived. Stem cell banks and cell repositories should be expanded and should coordinate their work, the report says, and funding agencies and journals should make data and material sharing mandatory. A database of stem cell–related patents is also urgently needed, the group says, to help scientists deal with the thorny thicket of intellectual property that has grown along with the hot field. http://scim.ag/hinxton-group

2. # Noted

>The Kafkaesque bureaucracy of the Framework Programme, the European Union's multibillion-euro research funding system, is getting simpler. A specific E.U.-approved accounting system will no longer be required, small business partners can now be paid a flat rate, and multiple sets of rules and procedures will be eliminated. http://scim.ag/no-kafka

3. # Newsmakers

## Japan Prizes

TOKYO—A breakthrough drug and a computer operating system shared the laurels of one of science's top honors, the Japan Prize.

The Bioscience and Medical Science prize goes to Tadamitsu Kishimoto and Toshio Hirano, both of Osaka University, for identifying the protein interleukin 6, elucidating its immune functions, and determining its role in rheumatoid arthritis. The pair ultimately helped develop a drug that treats the debilitating disorder by blocking the protein's activity.

Kenneth Thompson and Dennis Ritchie take home the Information and Communications prize for developing the UNIX computer operating system in the 1960s and '70s while at Bell Laboratories in Murray Hill, New Jersey. UNIX set new standards for simplicity and ease of adapting software to different computing platforms. The source code was also distributed to users so that they could contribute improvements, marking the beginning of the open systems concept. Thompson is now at Google Inc.; Ritchie is retired.

## Canadian-Born Geologist Tapped for Key E.U. Research Post

Experimental volcanologist Donald Bruce Dingwell was born and raised in Canada, but he may soon be playing a big role in European research funding. Science has learned that he has been chosen as the next secretary general of the European Research Council (ERC), the E.U.'s funding agency for individual basic researchers. But negotiations about his contract are still ongoing, cautions ERC President Helga Nowotny.

Dingwell is currently director of the Department of Earth and Environmental Sciences at Ludwig Maximilians University Munich in Germany. As secretary general, he would be a liaison to ERC's executive agency, a managing body ultimately controlled by the European Commission, on behalf of ERC's scientific council, which sets strategy. ERC previously decided to scrap the position of secretary general but reversed that decision pending an overhaul of its organizational structure. http://scim.ag/dingwell

## They Said It

This week, our Facebook fans chimed in on a story of a dino with just one finger (see p. 384). Some of our favorite comments:

“Thanks for flipping us the prehistoric bird!”

“It's ok as long as the dino doesn't have to do math homework.”

“And that was how the binary system was created, right?”

4. # ERC Still Looking for Women

The European Research Council (ERC) announced that 266 “advanced scientists” won its latest round of funding, with British and German researchers grabbing the most grants. But just 9.4% of the winners are women, an imbalance seen in previous ERC funding rounds that is a source of continuing frustration to the organization.

5. # By the Numbers

40% — The percentage of U.S. high school seniors who scored below the basic level of achievement in science on the 2009 National Assessment of Educational Progress. Only 1% of 12th-graders scored at the advanced level on the test, a quadrennial measure of reading, math, and science that is known as the Nation's Report Card (http://scim.ag/NAEP-2009)

12 million — The average extent, in square kilometers, of Arctic sea-ice cover during December 2010, according to the World Meteorological Organization. That makes it the lowest on record for December, about 1.35 million square kilometers below the 1979–2000 average.

12% — The percentage of first-year U.S. college students who, when asked about seven common types of quantitative reasoning activities, said they had never done any of them, according to the 2010 National Survey of Student Engagement.

6. # Random Sample

Long before he became a bioengineer, Ingmar Riedel-Kruse was a typical 12-year-old video game aficionado. “I wasted a year of my life doing that—‘wasted’ in a good sense,” he recalls. He soon moved on from playing with computers to programming them and, eventually, to biotechnology research. About a year ago, while reading a Wikipedia article about the history of video games, he had an inspiration: Why not start gaming with actual, live microbes?

After “lots of trial and error,” Riedel-Kruse says, he and his colleagues at Stanford University developed a basic game console that nudges paramecia around a microfluidic chamber with chemical gradients or mild electric fields. A microscope camera pipes images of the wriggling protozoans into a laptop game window. Games include soccerlike “Ciliaball”; “Pac-mecium,” in which paramecia gobble virtual yeast dots while avoiding a cartoon zebrafish larva; and “Pond Pong,” in which two players bat the microbes back and forth by releasing chemicals from a needle tip. Descriptions of the whole microarcade, including games involving yeast and DNA, appear this month in the journal Lab on a Chip.

Riedel-Kruse says putting biotic games on the Internet could give researchers a way to crowdsource real-time biology experiments to online players. He's also working on bringing them into schools to inspire future scientists—after all, it worked for him.

7. Anthropology

# A New View Of the Birth of Homo sapiens

1. Ann Gibbons

New genomic data are settling an old argument about how our species evolved.

For 27 years, Chris Stringer and Milford Wolpoff have been at odds about where and how our species was born. Stringer, a paleoanthropologist at the Natural History Museum in London, held that modern humans came out of Africa, spread around the world, and replaced, rather than mated with, the archaic humans they met. But Wolpoff, of the University of Michigan, Ann Arbor, argued that a single, worldwide species of human, including archaic forms outside of Africa, met, mingled and had offspring, and so produced Homo sapiens. The battle has been long and bitter: When reviewing a manuscript in the 1980s, Wolpoff scribbled “Stringer's desperate argument” under a chart; in a 1996 book, Stringer wrote that “attention to inconvenient details has never been part of the Wolpoff style.” At one tense meeting, the pair presented opposing views in rival sessions on the same day—and Wolpoff didn't invite Stringer to the meeting's press conference. “It was difficult for a long time,” recalls Stringer.

Then, in the past year, geneticists announced the nearly complete nuclear genomes of two different archaic humans: Neandertals, and their enigmatic eastern cousins from southern Siberia. These data provide a much higher resolution view of our past, much as a new telescope allows astronomers to see farther back in time in the universe. When compared with the genomes of living people, the ancient genomes allow anthropologists to thoroughly test the competing models of human origins for the first time.

The DNA data suggest not one but at least two instances of interbreeding between archaic and modern humans, raising the question of whether H. sapiens at that point was a distinct species (see sidebar, p. 394). And so they appear to refute the complete replacement aspect of the Out of Africa model. “[Modern humans] are certainly coming out of Africa, but we're finding evidence of low levels of admixture wherever you look,” says evolutionary geneticist Michael Hammer of the University of Arizona in Tucson. Stringer admits: “The story has undoubtedly got a whole lot more complicated.”

But the genomic data don't prove the classic multiregionalism model correct either. They suggest only a small amount of interbreeding, presumably at the margins where invading moderns met archaic groups that were the worldwide descendants of H. erectus, the human ancestor that left Africa 1.8 million years ago. “I have lately taken to talking about the best model as replacement with hybridization, … [or] ‘leaky replacement,’” says paleogeneticist Svante Pääbo of the Max Planck Institute for Evolutionary Anthropology in Leipzig, lead author of the two nuclear genome studies.

The new picture most resembles so-called assimilation models, which got relatively little attention over the years. “This means so much,” says Fred Smith of Illinois State University in Normal, who proposed such a model. “I just thought ‘Hallelujah! No matter what anybody else says, I was as close to correct as anybody.’”

## Evolving models

Stringer and others first proposed Africa as the birthplace of modern humans back in the mid-1980s. The same year, researchers published a landmark study that traced the maternally inherited mitochondrial DNA (mtDNA) of all living people to a female ancestor that lived in Africa about 200,000 years ago, dubbed mitochondrial Eve. She caught the attention of the popular press, landing on the cover of Newsweek and Time.

Additional studies of living people—from Y chromosomes to snippets of nuclear DNA to the entire mtDNA genome—consistently found that Africans were the most diverse genetically. This suggests that modern humans arose in Africa, where they had more time to accumulate mutations than on other continents (Science, 17 November 2006, p. 1068). Meanwhile, ancient DNA technology also took off. Pääbo's group sequenced first a few bits of Neandertal mitochondrial DNA in 1997, then the entire mitochondrial genomes of several Neandertals—and found them to be distinct from those of living people. So ancient DNA, too, argued against the idea of mixing between Neandertals and moderns. Over the years the replacement model became the leading theory, with only a stubborn few, including Wolpoff, holding to multiregionalism.

Yet there were a few dissenting notes. A few studies of individual genes found evidence of migration from Asia into Africa, rather than vice versa. Population geneticists warned that complete replacement was unlikely, given the distribution of alleles in living humans. And a few paleoanthropologists proposed middle-of-the-road models. Smith, a former student of Wolpoff's, suggested that most of our ancestors arose in Africa but interbred with local populations as they spread out around the globe, with archaic people contributing to about 10% of living people's genomes. At the University of Hamburg in Germany, Gunter Brauer similarly proposed replacement with hybridization, but with a trivial amount of interbreeding. But neither model got much traction; they were either ignored or lumped in with multiregionalism. “Assimilation got kicked so much,” recalls Smith.

Over time, the two more extreme models moved toward the middle, with most multiregionalists recognizing that the chief ancestors of modern humans arose in Africa. “The broad line of evolution is pretty clear: Our ancestors came out of Africa,” says biological anthropologist John Relethford of the State University of New York College at Oneonta. “But what happens next is kind of complex.”

## Genes from the past

Then in May 2010 came the Neandertals' complete nuclear genome, sequenced from the bones of three female Neandertals who lived in Croatia more than 38,000 years ago. Pääbo's international team found that a small amount—1% to 4%—of the nuclear DNA of Europeans and Asians, but not of Africans, can be traced to Neandertals. The most likely model to explain this, Pääbo says, was that early modern humans arose in Africa but interbred with Neandertals in the Middle East or Arabia before spreading into Asia and Europe, about 50,000 to 80,000 years ago (Science, 7 May 2010, pp. 680, 710).

Seven months later, on 23 December, the team published in Nature the complete nuclear genome of a girl's pinky finger from Denisova Cave in the Altai Mountains of southern Siberia. To their surprise, the genome was neither a Neandertal's nor a modern human's, yet the girl was alive at the same time, dating to at least 30,000 years ago and probably older than 50,000 years. Her DNA was most like a Neandertal's, but her people were a distinct group that had long been separated from Neandertals.

By comparing parts of the Denisovan genome directly with the same segments of DNA in 53 populations of living people, the team found that the Denisovans shared 4% to 6% of their DNA with Melanesians from Papua New Guinea and the Bougainville Islands. Those segments were not found in Neandertals or other living humans.

The most likely scenario for how all this happened is that after Neandertal and Denisovan populations split about 200,000 years ago, modern humans interbred with Neandertals as they left Africa in the past 100,000 years. Thus Neandertals left their mark in the genomes of living Asians and Europeans, says co-author Montgomery Slatkin, a population geneticist at the University of California, Berkeley. Later, a subset of this group of moderns—who carried some Neandertal DNA—headed east toward Melanesia and interbred with the Denisovans in Asia on the way. As a result, Melanesians inherited DNA from both Neandertals and Denisovans, with as much as 8% of their DNA coming from archaic people, says co-author David Reich, a population geneticist at Harvard Medical School in Boston.

This means H. sapiens mixed it up with at least two different archaic peoples, in at least two distinct times and places. To some, that's starting to sound a lot like multiregionalism. “It's hard to explain how good I feel about this,” says Wolpoff, who says that seeing complete replacement falsified twice in 1 year was beyond his wildest expectations. “It was a good year.”

And yet the interbreeding with archaic humans seems limited—from 1% to 8% of some living people's genomes. Stringer and many others don't consider it full-scale multiregional continuity. “I think interbreeding was at a low level,” says Slatkin, who says that if there had been a great deal of admixture, the genetic data would have revealed it already. Low levels of interbreeding suggest that either archaic people mated with moderns only rarely—or their hybrid offspring had low fitness and so produced few viable offspring, says population geneticist Laurent Excoffier of the University of Bern in Switzerland.

In any case, Reich notes that at least 90% of our genomes are inherited from African ancestors who replaced the archaic people on other continents but hybridized with them around the margins. And that scenario most closely backs the assimilation models proposed by Smith and Brauer.

Of course, it's possible that future data will overturn today's “leaky replacement” model. Slatkin says he cannot rule out an alternative explanation for the data: The “archaic” DNA thought to have come from mating with Neandertals could instead stem from a very ancient ancestor that we shared with Neandertals. Most modern humans retained those archaic sequences, but Africans lost them. But Slatkin says this “doesn't seem very plausible,” because it requires modern human populations with the archaic DNA and those without it to have been partially isolated from each other in Africa for hundreds of thousands of years. And it seems even less probable that Melanesians and Denisovans are the only groups that retained a second set of archaic DNA motifs from a common ancestor shared by all modern humans, Neandertals and Denisovans. If those explanations do prove true, replacement would not be falsified.

In the wake of the big genome studies, other researchers such as Hammer are scrutinizing DNA from more living humans to further test the model. Researchers are also trying to pinpoint when admixture happened, which has significant consequences. At just what point did we evolve from archaic humans to become “modern” humans? “There are still archaic [genetic] features floating around until amazingly recently, until 40,000 years ago,” says Hammer. He wonders whether the process of becoming modern took longer and was more complex than once thought. “There's no line you can draw and say everything after this is modern. That's the elephant in the room.”

Meanwhile, paleoanthropologists are searching for fossils in Asia that might belong to the enigmatic Denisovan population—and might yield more ancient DNA. Paleoanthropologist Russell Ciochon of the University of Iowa in Iowa City and Wolpoff say there are several known, ambiguous fossils in Asia that might be candidates for early Denisovans. “I believe things were going on in Asia that we just don't know about,” says Ciochon. “Before this paper on the Denisovans, we didn't have any insight into this. Now, with this nuclear genome, I find myself talking about ‘the Denisovans.’ It's already had an impact.”

As for Stringer and Wolpoff, both now in their 60s, their battle has mellowed. Their views, while still distinct, have converged somewhat, and they shared a beer at a Neandertal meeting last year. “The reason we get on well now,” says Stringer, “is we both think we've been proved right.”

8. Anthropology

# The Species Problem

1. Ann Gibbons

Our ancestors are now thought to have mated with at least two kinds of archaic humans at two different times and places. Were they engaging in interspecies sex, or does the fact that they were able to produce offspring mean they were all members of the same species?

Our ancestors had sex with at least two kinds of archaic humans at two different times and places—and those liaisons produced surviving children, according to the latest ancient DNA research (see main text, p. 392). But were the participants in these prehistoric encounters members of separate species? Doesn't a species, by definition, breed only with others of that species?

These are the questions paleogeneticist Svante Pääbo dodged twice last year. His team published two papers proposing that both Neandertals and mysterious humans from Denisova Cave in Siberia interbred with ancient modern humans. But the researchers avoided the thorny question of species designation and simply referred to Neandertals, Denisovans, and modern humans as “populations.” “I think discussion of what is a species and what is a subspecies is a sterile academic endeavor,” says Pääbo, who works at the Max Planck Institute for Evolutionary Anthropology in Leipzig, Germany.

The question of how to define a species has divided researchers for centuries. Darwin's words in On the Origin of Species still hold: “No one definition has satisfied all naturalists.” However, many scientists use the biological species concept proposed by Ernst Mayr: “groups of actually or potentially interbreeding natural populations, which are reproductively isolated from other such groups.”

The draft versions of the Neandertal and Denisovan nuclear genomes show low levels of interbreeding between each of them and modern humans. Apply Mayr's definition strictly, and all three must be considered Homo sapiens. “They mated with each other. We'll call them the same species,” says molecular anthropologist John Hawks of the University of Wisconsin, Madison.

But that's a minority view among paleoanthropologists. Many consider Neandertals a species separate from modern humans because the anatomical and developmental differences are “an order of magnitude higher than anything we can observe between extant human populations,” says Jean-Jacques Hublin, a co-author of Pääbo's at Max Planck. In the real world, he says, Mayr's concept doesn't hold up: “There are about 330 closely related species of mammals that interbreed, and at least a third of them can produce fertile hybrids.”

There's also no agreed-upon yardstick for how much morphologic or genetic difference separates species. That's why Pääbo's team avoided the species question a second time with respect to the Denisovans. These hominins are known only from a scrap of bone, a single tooth, and their DNA. They are genetically closest to Neandertals. The genetic distance between Denisovans and Neandertals, in fact, is only 9% larger than that between a living Frenchman and a living San Bushman in Africa, both of whom belong to H. sapiens. But so far Neandertals seem to have low genetic diversity, based on the DNA of six Neandertals from Russia to Spain. To Pääbo's team, that makes the difference from the Denisovans significant.

Also, the Denisovan tooth doesn't look much like that of a Neandertal. So the team considers them a distinct population but declined to name a new species. “Why take a stand on it when it will only lead to discussions and no one will have the final word?” asks Pääbo.

9. Physiology

# Going the Distance

1. Elizabeth Pennisi

One challenge to studying migration energetics: Not every bird will fly for hours in a wind tunnel.

LONDON, CANADA—In a dimly lit room, Chris Guglielmo reaches into a cloth bag and gently pulls out a tiny bird. He checks that its feathers and wings are intact, then puts his hand inside a Plexiglas tunnel and releases it. All eyes in the room are mesmerized by the blur of beating wings as this yellow-rumped warbler flits about the chamber. An 8-meter-per-second headwind keeps the bird in place as it begins what Guglielmo hopes will be a multihour flight.

Moments later, the warbler perches on a net at the upper end of the tunnel. It can be coaxed back into the air, but it's clear that it prefers to stay put. “I'm a little bummed, but it's how it goes,” says Guglielmo's graduate student, Alex Gerson, as he recaptures the bird.

A $9.3 million grant has bought Guglielmo's group at the University of Western Ontario here a state-of-the-art bird research facility, complete with aviaries, surgery room, a bird-sized magnetic resonance imaging (MRI) machine, and a$1.5 million wind tunnel, the only one in the world in which temperature, humidity, and barometric pressure can be controlled. A year old, the tunnel presents researchers with an unprecedented opportunity to probe the mysteries of migration in exquisite detail.

Guglielmo and Gerson want to understand fuel and water use by migrants. In the field, they follow birds that stop over to stock up during the fall and spring migration. What the birds eat can affect their ability to reach their final destination, and conservationists are keen to know which diets work best—and possibly provide them. “It's not enough, if you care about wild birds, to go out and count them and study their habitat,” says Guglielmo. “You must know more about the mechanisms,” and that's where the wind tunnel comes in. “It's providing us with a tool to answer questions that we could only answer indirectly [before],” says Scott McWilliams, a physiological ecologist at the University of Rhode Island, Providence.

Wind tunnels provide a controlled environment in which researchers can take before-and-after physiological measurements to assess the toll flight takes on fat stores, protein mass, water, and even immunological function when birds fly long distances. In addition, high-speed cameras and techniques for visualizing air flow enable Guglielmo's colleagues to assess the biomechanics of flapping wings, giving a better handle on the link between aerodynamics and migration.

But money can't buy birds that are keen to fly, and Guglielmo and his team have had mixed success finding willing avian partners. An early project using starlings did well, despite taking place when the tunnel was not quite finished. It netted “Super,” a bird that always cooperates and will even fly into the wind tunnel on its own accord. But a study involving robins took months to identify five somewhat cooperative fliers; switching to Swainson's thrushes worked better. One immunological project involving a shorebird called a ruff is stranded because the birds show no inclination to take to the air. And Guglielmo has just started testing warblers to see if high-protein or high-carbohydrate diets make a difference in energy use during flight. “There's a lot of trial and error on what species of birds to use and how to find individuals that will fly,” says Guglielmo.

## Eat, drink, and fly

When Guglielmo gets discouraged, he thinks about Blue, a wind-tunnel record holder. In the mid-1990s, field ecologists Marcel Klaassen and Åke Lindström wanted to learn more about how stopovers influence subsequent migration. “In the wild, you can study a bird only a few days and then it's gone, literally gone,” explains Lindström, based at Lund University in Sweden. Until that time, researchers had only tested nonmigratory birds, such as homing pigeons and small parrots, in short wind-tunnel flights, “so we didn't know if what we found out was relevant” to migratory birds, he says. Lund had a newly built wind tunnel, one that later served as inspiration for Guglielmo's.

On the advice of a wind-tunnel expert, Lindström and Klaassen were trying to train thrush nightingales to fly from a stick for a food reward, but the birds were not cooperating. Getting one to stay in the air for even just a minute “was cause for celebration,” Lindström recalls. That all changed when someone suggested removing the stick.

Blue looked back for his perch, wavered a little, then settled down to flying. In one trial, Blue kept going for 16 hours, showing no signs of slowing down even when the researchers decided it was time to go home to bed. Over the next 2 months, Blue made seven 12-hour flights, yielding data that led to five research papers on fuel use. The work showed that the estimated power requirements for migratory flight were less than those for nonmigratory flights and suggested that birds burn protein as well as fat on their long journeys.

Searching for eager fliers like Blue, Guglielmo knew that the wind tunnel itself was not the problem. Already, McWilliams had had a successful run for his experiment on the effects of fatty acid composition and dietary oxidants on exercise performance, even as the tunnel was being finished. In the summer of 2009, McWilliams's team captured 120 starlings, feeding some of them food rich in olive oil (a monounsaturated fat) and others a diet high in canola oil (a polyunsaturated fat). His earlier experiments showed that migrating birds will choose fruits, such as viburnum or bayberry, that are high in polyunsaturated fats, even though metabolizing polyunsaturated fats generates more oxidative stress than burning monounsaturated fats. McWilliams wanted to determine whether birds eating the canola oil diet flew more efficiently, burning less fat per hour, than birds eating olive oil, making the extra stress worth it.

The researchers look at how birds fuel their flights by determining the change in the ratio of fat to lean mass, and the rate at which the birds expend energy. Researchers have traditionally assessed fat content by checking the color and condition of the belly. A more accurate measurement using x-rays requires the birds to be anesthetized. The Western Ontario lab has a far more convenient tool, however: a custom-built quantitative magnetic resonance scanner that can produce an accurate readout of the lean mass, fat mass, and body water content quickly and noninvasively. It takes just 2 minutes, with the bird temporarily immobilized in a Plexiglas tube. “It's no more difficult than weighing a bird,” says Guglielmo. “It's really a new way to do these experiments.”

To assess how much energy the birds expend, the researchers inject birds just prior to flight with water containing heavier isotopes of hydrogen and oxygen. They then take blood samples before and after flight and use the change in the relative amounts of these stable isotopes to estimate carbon dioxide production, a proxy for energy expenditure.

McWilliams was eager to begin testing the effects of the different diets on flight energetics, but construction had fallen behind schedule, and the experiment was delayed for weeks. By September 2009, he couldn't wait any longer. “It was a little touch and go,” he recalls. In between cycles of construction, he and his colleagues trained birds to fly in the wind tunnel and then, for the experiment, let each bird fly as long as it wanted, hoping for hourlong or longer flights. McWilliams is still analyzing the data, but he says The Nature Conservancy is eager to know the answer so it can better manage the preserves used by migrating birds.

## Hit or miss

Guglielmo and Gerson knew from experience that finding the right species for their experiments, and the right individuals, was hit or miss. Gerson is tackling another aspect of flight energetics: water balance. Studies of birds that fly across deserts indicate that to make up for water loss, the birds break down their organs, harvesting water from the metabolism of proteins in muscles or the gut, for example. Gerson has been testing this idea by examining the change in body water, fat mass, and lean mass in birds flying in the wind tunnel.

American robins, common migrants of optimal size, seemed ideal. “But they didn't work out so well,” says Guglielmo. Still, he and Gerson did learn something from robins: Hand-raised birds were no more likely to cooperate than wild-caught birds that were trained. So Gerson started trying whatever birds he could get his hands on, red-eyed vireos, gray catbirds, yellow-rumped warblers. Their conclusion: “Fat birds and calm birds fly,” quips Guglielmo.

Swainson's thrushes seemed to work well. One went more than 5 hours nonstop, the equivalent of about 180 kilometers, using about 0.25 grams of fat per hour. Gerson has been testing them in 10% and 80% humidity, checking weight loss and using the MRI scanner to measure changes in water content and lean and fat mass before and after the flights. Preliminary results show that “if the bird flies in dry conditions, it will burn more of its lean mass to liberate more water,” he says.

Fat and calm is no guarantee of success, however. Silke Nebel, a postdoc, wants to check how the state of the immune system affects flight. There is some evidence that birds with infections don't fly as well, and it is important to quantify the handicap, particularly for epidemiologists trying to predict how far avian-borne diseases will spread. The tradeoff that may exist between immunocompetence and flight performance “is very difficult to study in a natural setting,” she points out.

The wind tunnel is too small for flying ducks and geese, so Nebel wants to use ruffs as stand-ins. However, the ruffs, which come from a captive colony, showed no inclination to migrate, despite having put on weight and being used to captivity. Nebel has tried hatching, raising, and training young. Even so, at 2 months old, these youngsters fly no more than a few minutes at a time. Her last hope is that as they mature, the birds will develop larger flight muscles that will let them stay in the air longer.

## Flying without a net

Guglielmo settled on the yellow-rumped warbler because it is small, and the species is known to adapt well to captivity. The yellow-rumped warbler varies its diet seasonally, feasting on insects in the spring but turning more toward fruits in the fall. Guglielmo is looking at how diet affects body composition and fuel use during flight.

For these experiments, he mixes up pans of casein- or sugar-rich agar and, after the mixtures harden, runs the gel through a food mill to make small, appetizing “worms.” Guglielmo calls the high-protein fare his Atkins diet for birds; those on it weigh about 2 grams less than birds feasting on high carbohydrates.

When Guglielmo first let the warblers try out the tunnel, they didn't show much inclination to keep flying and instead kept landing on the upwind net. In earlier experiments, the researchers had tried everything—making that end of the tunnel dark, flashing strobe lights, even spraying water jets to discourage the birds from perching there. This night, they are pulling out all stops and taking the net down altogether. They tested a bird for 20 minutes and it didn't head upwind too far, so now they are ready to see if the warblers will fly long distances. The next night, one bird lasted 45 minutes before landing on the tunnel floor—an okay flight time but not ideal. But the evening after that was a different story. “Purple-black,” named for the color tag on its leg, “found his groove and flew for 6 hours,” Guglielmo e-mailed the next day. “We had a great night.”

10. Physiology

1. Elizabeth Pennisi

The wind tunnel at the University of Western Ontario in Canada is the centerpiece of a new building that includes indoor and outdoor aviaries, acoustic chambers, behavior-observation rooms, and environmental chambers for controlling light cycles, as well as temperature and humidity.

Wind tunnels have come a long way since former DuPont president and amateur ornithologist Crawford Greenewalt built one of the first 50 years ago to study hummingbirds. A fan at the end of his 46-centimeter “tunnel” kept hummingbirds stationary while he filmed them flying; those high-speed images impressed amateurs and academics alike.

In 1994, Swedish biologists greatly improved on the concept, borrowing from designs used in engineering. Air was recirculated through a 20-meter-long rectangular chamber. Just upstream of the 2-meter-long test section, the tunnel widens and then narrows, which greatly reduces turbulence and makes it easier to assess how power requirements for flight vary with speed. In this setting, Anders Hedenström of Lund University in Sweden and his colleagues adopted an engineering technology that uses fog and lasers to visualize airflow. They documented for the first time what happens to air as its moves across a real bird's wing. The tunnel can be tilted to allow birds to glide or “climb” to higher altitudes. Over the years Hedenström and his colleagues have studied flight in 15 species, as well as in bats.

A copy of the Lund tunnel was built at the Max Planck Institute for Ornithology in Seewiesen, Germany, in 1999. There, 150 birds from 25 species have been evaluated.

When Chris Guglielmo started at the University of Western Ontario in London, Canada, in 2005, he realized those tunnels were too far away for him to include wind-tunnel studies routinely in his research. Others on campus wanted better bird-care facilities for their research, so he teamed up with nine colleagues and proposed the Advanced Facility for Avian Research. Much to their university president's surprise, the Canada Foundation for Innovation approved this grant application in 2007.

The wind tunnel is the centerpiece of a new building that includes indoor and outdoor aviaries, some of which have small pools of water for shorebirds and waterfowl, acoustic chambers, behavior-observation rooms, and environmental chambers for controlling light cycles, as well as temperature and humidity. Made of steel, two stories high and 12 meters long, the tunnel looks quite imposing. Birds fly in a 1-meter-high by 1.5-meter-wide octagonal Plexiglas cylinder, almost completely covered with blankets to discourage the birds from getting too close to the walls. Pressure, temperature, and humidity in the tunnel can all be modified to simulate different conditions, including high altitude.

The whole flight section is encased in a steel room, with air locks. Once the room and wind tunnel are locked down, “in a few minutes, you can go up to over 23,000 feet,” says Guglielmo. “There's a lot of theory about what happens when birds fly at [high] altitudes, but nowhere in the world had we had a place to test that empirically.”

Guglielmo and his colleagues are still developing protocols for low-pressure research, but they have already simulated weather fronts. Scott MacDougall-Shackleton, also of Western Ontario, and Guglielmo put the birds into the tunnel at night and just before the lights go on, they cause a drop or rise in air pressure and temperature and observe how that affects the birds' behavior when they wake up. It seems that when the birds sense bad weather coming on, they are much quicker to start feeding, says Guglielmo.

The tunnel also doubles as a studio for understanding the mechanics of bird flight. Working with Roi Gurka of Ben Gurion University in Beer-Sheva, Israel, Gregory Kopp and graduate student Adam Kirchhefer of Western Ontario have filled the tunnel with oil droplets and set up a laser field in front of a high-speed video camera that captures the movement of the droplets—and the air—as a bird flies. A second camera is trained on the bird itself. In particular, they are studying the movement of air at the wing tips. For these studies, a particularly cooperative starling named Super wears an orange-tinted mask to protect its eyes from the laser. “A few good seconds of video will keep you going for a long time,” says Guglielmo. And a sample size of one or two is sufficient. “It's a different ball game when you want to look at migration,” where one needs multiple birds willing to fly for hours at a time.