News this Week

Science  08 Oct 2004:
Vol. 306, Issue 5694, pp. 206

You are currently viewing the .

View Full Text

Log in to view the full text

Log in through your institution

Log in through your institution


    Parkfield Keeps Secrets After a Long-Awaited Quake

    1. Richard A. Kerr

    Last week's moderate-to-strong earthquake in central California has justified seismologists' belief that Parkfield (population 37) was the place to wait for a sizable quake they could study. “It's right in the very middle of our network,” says geophysicist Malcolm Johnston of the U.S. Geological Survey (USGS) in Menlo Park, California, about the densest fault-monitoring system in the world. It cost more than $10 million over 20 years. “We got great stuff,” says Johnston.

    But they didn't get it entirely right. When seismologists began the Parkfield Earthquake Prediction Experiment in the 1980s, they expected to capture the next magnitude 6 in unprecedented detail within a few years. Instead, they had to wait 2 decades, a delay that casts additional doubt on models of predictable seismic behavior. And far from providing practical experience in the nascent science of short-term earthquake prediction, Parkfield 2004 seems to have given no warning that would lend hope to the field of short-term quake forecasting. All in all, Parkfield has driven home the point that even one of the world's best behaved fault segments can be pretty cantankerous.

    Twenty years ago, the 25-kilometer section of the San Andreas fault that runs under the town of Parkfield seemed like a model seismic citizen. Earthquakes of about magnitude 6, noted two USGS seismologists, had ruptured the same Parkfield segment of the San Andreas in 1857, 1881, 1901, 1922, 1934, and 1966. The average of 22 years between recurrences seemed reliable enough (after rationalizing 1934's “early” arrival), so the next quake in the series should arrive in 1988, give or take 5 years. The National Earthquake Prediction Evaluation Council, a federal committee advising the USGS director, had concurred with that long-term forecast.

    But the accuracy of that “give-or-take” forecast had long ago come into question. Now, 16 years after the forecast's most probable date, official quake forecasts say the likelihood of the next Parkfield quake occurring in 2004 was just 5% to 10%. The delay only reinforces the idea that “earthquake recurrence is less regular than had been hoped,” says seismologist William Ellsworth of the USGS in Menlo Park. “There are real practical limits to the type of forecast we made at Parkfield.”

    Back at last.

    The Parkfield earthquake (largest red circle marking its starting point among aftershocks) took far longer than average to recur on the San Andreas fault (red line) and gave no obvious warning of its return.


    The limits of quake forecasting became clearer still when seismologists looked at the magnitude-6.0 event on 28 September, which caused little damage to the sparsely populated region 75 kilometers inland from the coast. Seismologist Ross Stein of USGS Menlo Park recalls a number of 1980s ideas about quakes that would have favored predictability. They included the idea that quakes could recur with some regularity; that the more time a fault had to build up strain, the larger the eventual quake would be; and that the same fault segment would rupture in the same “characteristic” quake—the same magnitude and same section of fault—each time.

    Of these and other optimistic quake ideas, “the only one still alive at Parkfield is the characteristic earthquake,” says Stein. The quake's timing certainly wasn't regular. And to judge by the amount of fault strain accumulated in the intervening 38 years, Parkfield 2004 should have released 20 times the energy that it did and have been a magnitude 6.7.

    Even the characteristic aspect does not hold up in detail, Stein notes. The same 25 kilometers of fault broke as in 1966 and 1934, producing a similar-magnitude quake. But in 2004 the rupture started at the southeast end of the segment and ran northwestward, the opposite direction from those that struck in '34 and '66. Parkfield earthquakes—once considered among the most regular of quakes—“are certainly not peas in a pod,” observes Menlo Park's Johnston.

    Unfortunately for the prediction experiment at Parkfield, the individuality of quakes there extended to geophysical activity before the main shock, activity that seismologists once hoped could be used to predict the main event. The 1966 Parkfield main shock was preceded by a number of possible and even certain precursors. They included a flurry of microearthquakes 2 to 3 months before, cracks in the ground along the fault at least 11 days prior, and a magnitude-5.1 foreshock 17 minutes ahead of the main shock. A magnitude-5 foreshock preceded the 1934 Parkfield quake by 17 minutes as well.

    Nothing obvious heralded the 2004 Parkfield quake. “At the moment, nothing has jumped off the screen,” says Ellsworth. A vastly improved seismometer network at Parkfield detected no foreshocks down to magnitude 0, says Robert Nadeau of the University of California, Berkeley. (Magnitudes can be even smaller and negative.) Johnston reports nothing obvious from the dense networks of creepmeters, magnetometers, and strainmeters scattered along the fault. The only possible precursor being discussed is a slow, subtle straining around the fault that began on 27 September. Johnston thinks it may be the long-sought signature of a main shock's very beginnings, so-called nucleation. Colleagues are reserving judgment.

    Despite all the disappointments, seismologists haven't lost faith in their quest to understand how earthquakes behave. “The [Geological] Survey bet the farm, lost, was humbled, but stuck it out,” says Stein. “In the end, it was the right choice.” Earthquake prediction aside, the recording of strong ground shaking in unprecedented detail creates a great opportunity to learn how to build safer, more quake-resistant buildings, says engineering seismologist Anthony Shakal of the California Geological Survey in Sacramento. “Our science advances on the basis of great data,” adds Stein, and that is what they got.

  2. 2004 NOBEL PRIZES

    Axel, Buck Share Award for Deciphering How the Nose Knows

    1. Greg Miller

    The sweet smell of success greeted Richard Axel and Linda Buck this week as the two U.S. neuroscientists were awarded the 2004 Nobel Prize in physiology or medicine for their pioneering work on the sense of smell.

    The pair first worked together as professor and postdoc in Axel's lab at Columbia University in New York City and have since worked independently to answer fundamental questions about how the brain notices odors wafting through the air. Both are now investigators of the Howard Hughes Medical Institute. Their work has enticed researchers from other fields to study olfaction. “They're magnificent scientists who made a key discovery that opened a big area of research,” says Solomon Snyder, a neuroscientist at Johns Hopkins University in Baltimore.

    That discovery, reported in a landmark 1991 paper in Cell, was the first description of olfactory receptors, the proteins responsible for turning a smell into something the brain can understand. The receptors are embedded on the surfaces of neurons at the back of the nasal cavity. When the receptors bind to odorant molecules sucked into the nose, they trigger a biochemical cascade that ultimately generates a nerve impulse that transmits information to the brain. The paper described a family of about 1000 genes that encode olfactory receptors in rats. The receptor proteins belong to a large class of proteins already familiar to researchers for the variety of roles they play in cell signaling.

    Smells like Stockholm.

    Richard Axel (left) and Linda Buck share the 2004 Nobel Prize in physiology or medicine for their research on olfaction.


    Some previous work had suggested that olfactory receptors belonged to this class—G protein-coupled receptors—but the sheer number of olfactory receptors was far greater than anyone had expected, says Columbia's Stuart Firestein, who was not involved in the research. The human visual system, he points out, is able to distinguish myriad colors using only three types of receptors—ones tuned to blue, green, and red. (Subsequent research has revealed that humans have fewer working olfactory receptor genes than rodents—only about 350.) “The work was clearly a breakthrough,” says Peter Mombaerts of Rockefeller University in New York City, who joined Axel's lab as a postdoc after reading the 1991 paper and went on to start his own olfactory research laboratory.

    Identifying the receptors paved the way to understanding how information about smell is organized in the brain. Independently, Axel and Buck, who is now at the Fred Hutchinson Cancer Research Center in Seattle, Washington, determined that each olfactory receptor neuron expresses one—and only one—olfactory receptor protein. This provided an essential clue to understanding how the brain distinguishes smells. Each odor activates a unique combination of olfactory neurons, allowing the brain to distinguish, say, a good apple from a rotten one.

    Axel, 58, and Buck, 57, are both known among colleagues as extremely thorough scientists. “Richard will never publish anything unless it's a really important step forward,” says Snyder. The same goes for Buck, who becomes only the sixth woman to win the physiology or medicine Nobel in its 103-year history.

    Although the duo's work has answered important questions about the sense of smell, it has also posed additional puzzles. Researchers have just begun to make inroads, for example, toward understanding how an olfactory neuron chooses which receptor gene to express (Science, 19 December 2003, p. 2088).

    The layered mysteries of the olfactory system are part of the draw for Buck. “It's a wonderful, never-ending puzzle,” she says. “I can't think of anything else I'd rather be working on.”


    Russia, Reluctantly, Backs Kyoto

    1. Andrey Allakhverdov,
    2. Vladimir Pokrovsky*
    1. Allakhverdov and Pokrovsky are writers in Moscow.
    2. With reporting by David Malakoff.

    MOSCOW—After a heated debate last week, the Russian cabinet approved the Kyoto Protocol and sent it to the State Duma, the lower house of the Russian parliament. Observers expect the Duma to ratify it, and if it does, the treaty clears a critical threshold on its way to being accepted as international law. But this will likely do little to quell the fierce debate among Russian researchers and some officials over the merits of the treaty and its ability to reduce greenhouse gas emissions. Also unclear is how firmly Russia and other signatories would enforce the agreement.

    Kyoto supporters have been lobbying Russia for years to support the treaty because the country is responsible for 17% of 1990 greenhouse gas emissions, the levels on which the protocol is based. The treaty only comes into force when enough countries have signed up to account for 55% of 1990 emissions. With the United States, the world's biggest emitter, opting out, the protocol would collapse without Russia's participation.

    Geographer and biologist Mikhail Zalikhanov, a member of the Duma committee on environment, says that he thinks the Duma will ratify the treaty, but with some provisos. “At the moment I cannot say exactly what these reservations will be, but in the current situation Russia will not benefit from the ratification and may lose much in the future,” he says. The treaty requires Russia to stay below its 1990 emissions level until 2012.


    But European nations have been pressuring Russia to sign up. Prior to last week's cabinet meeting, Russian President Vladimir Putin met with European Commission President Romano Prodi, while Russian Prime Minister Mikhail Fradkov met with the acting prime minister of the Netherlands, Gerrit Zalm, currently president of the European Union. Fradkov told reporters that the protocol would have trouble in the Duma and might have to be amended. Putin's economic adviser Andrey Illarionov was even more pessimistic, saying forced reductions in industrial output would cost Russia $1 trillion by 2012. “This is a very bad day for the economy and the environment—and civilization,” he told a meeting in Washington, D.C., last week.

    Opposition in the scientific establishment surfaced earlier this year when Putin asked the Russian Academy of Sciences (RAS) to examine the treaty. A panel of 25 prominent researchers and experts, including RAS President Yuri Osipov and Illarionov, concluded in May that the protocol does not have any scientific basis and would be ineffective in stabilizing greenhouse gas emissions.

    In the short term, however, Russia may be able to cash in on the treaty. Russia's greenhouse gas emissions, which fell dramatically after the collapse of the Soviet Union in 1991, have yet to come back up to 1990 levels. According to Yuri Israel, director of the RAS Global Climate Institute, “we can even count on profiting from selling the greenhouse gas quotas to other countries.” Experts disagree, however, on how long it will take Russian emissions to rise again to 1990 levels. Illarionov predicts it will happen as early as 2007, perhaps forcing Russia to buy emissions credits from other nations. “Those expecting Russia to be a net seller of CO2 emission credits will be greatly disappointed,” says Illarionov. Israel thinks the country will make at most $100 million.


    Louse DNA Suggests Close Contact Between Early Humans

    1. Elizabeth Pennisi

    Lice may be the bane of teachers trying to stop the parasites from leaping from head to head, but their persistent association with people is proving a boon to researchers probing modern human origins. Because lice are species-specific parasites, their history is thought to parallel our own. Now a genetic analysis of head lice suggests that two distinct species of early humans had close physical contact after a long period of isolation. “The work [gives] us an indirect but informative new window on modern human origins,” says paleontologist Chris Stringer of the Natural History Museum in London.

    Stringer and others have argued that our species, Homo sapiens, migrated out of Africa and quickly replaced other human species, such as H. erectus in Asia, without interbreeding. A competing theory, multiregional evolution, contends that modern humans appeared when Homo sapiens from different geographical regions mated with each other as well as with archaic Homo populations, blurring regional and species boundaries. A middle-ground proposal suggests that as modern humans from Africa spread across the globe, they interbred with archaic humans, but that only African genes persisted. After analyzing lice data, Dale Clayton, an evolutionary biologist at the University of Utah, Salt Lake City, says that the history of these pests best fits the third hypothesis.

    For the work, Clayton and postdoc David Reed, now an evolutionary biologist at the University of Florida, Gainesville, compared mitochondrial DNA from lice, primarily Pediculus humanus, to existing data on human evolution. They analyzed six louse species, including two from humans, three from other primates, and one from a rodent. They used the sequences of two mitochondrial genes plus morphological traits to draw the louse family tree, which they then compared to the Homo tree.

    Evolutionary partner.

    Researchers itching to track human origins are turning to lice for answers.


    Because lice never leave their human hosts, the lice data are “a completely independent line of evidence” that helps confirm human prehistory, says Clayton. For example, according to the parasite's DNA, lice specific to humans and lice specific to chimpanzees appeared 5.6 million years ago, confirming previous work suggesting that the ancestors of chimps and humans diverged at about this time, Reed, Clayton, and their colleagues report in the 5 October online Public Library of Science. The lice also suffered a dramatic population decline and then recovery about 100,000 years ago, a bottleneck that parallels the story inferred from human genes. “The degree to which [the louse] tracks human history [is] amazing,” says Reed.

    The data also revealed that two genetically distinct lineages of P. humanus appeared about 1.18 million years ago. One subspecies is now distributed worldwide and infects either the head or the body, whereas the other only inhabits the New World and only lives on scalps. Clayton argues that the two lice subspecies must have diverged at about the same time that two human lines—perhaps Asian H. erectus and the African ancestors of H. sapiens—became established. The fact that the lice grew so far apart genetically suggests that they had little or no contact with each other —which implies that their human hosts were also separated. Consequently, “long-term gene flow such as is envisaged in the multiregional model is ruled out from these data,” says Stringer.

    But the data do suggest that there must have been some contact among different kinds of early humans. Today, there is only one species of human—but two subgroups of lice. So the lice thought to have been living on H. erectus must have jumped to H. sapiens at some point before H. erectus went extinct, perhaps as late as 30,000 years ago. The researchers think the shift occurred through skin-to-skin contact, as might occur during fighting or sex.

    Some researchers are convinced by this scenario. “The pattern they found is as clear as a bell,” says anthropologist Henry Harpending of the University of Utah, who was not involved with the work. But Milford Wolpoff of the University of Michigan, Ann Arbor, author of the multiregional hypothesis, calls the new study a “fringe explanation.” He notes that the divergence of the louse subspecies does not necessarily imply a million-year separation, because populations can diverge without isolation. He adds that the story “doesn't work at all with our studies,” which he says indicate frequent contact between different archaic humans.

    Clayton and Reed hope to pin down the question of contact among human species by studying the genetic history of lice transmitted almost exclusively through sexual intercourse. “If we get pubic lice, which are a different genus, and get the same results, then we would know that there is something very interesting going on,” says Clayton.


    Janelia Farm to Recruit First Class

    1. Jocelyn Kaiser

    Neuronal circuitry and imaging technologies will be the focus of the new Janelia Farm Research Campus of the Howard Hughes Medical Institute (HHMI). This week HHMI begins recruiting staff in these fields for its $500 million, 280-scientist institute in Ashburn, Virginia, scheduled to open in late 2006.

    Janelia Farm director Gerald Rubin says he wants to recreate the close-knit feeling of legendary labs such as the Laboratory of Molecular Biology in Cambridge, U.K., where well-funded investigators free of grant-seeking pressures work in small groups (Science, 9 May 2003, p. 879). There will be at least one difference: Janelia will emphasize technology. Last week, a hardhat-clad Rubin showed off the vast concrete bays and corridors of Janelia's main building at a bucolic site along the Potomac River, about 64 kilometers from Washington, D.C. It could accommodate the largest nuclear magnetic resonance machine or microscope, but at this point, he says, “we have no idea what we're going to put in it.”

    To decide on Janelia's research focus, HHMI held five workshops earlier this year and asked scientific leaders to think about problems tough enough to require 100 people working for 10 years. The advisers ruled out areas such as membrane proteins, figuring that they could be studied at existing labs. But the “challenging” and “highly interdisciplinary” problem of how a fruit fly assesses motion and distance to land softly on a wall made the cut, Rubin says. So did building new optical and other microscopes for imaging subcellular structures and living systems.

    One workshop participant, molecular biologist Eva Nogales of HHMI, the University of California, Berkeley, and Lawrence Berkeley National Laboratory, hopes Janelia's teams will devise new detectors and computational methods for imaging nonhomogenous macromolecules. “It could be a quantum leap in what is being done right now,” Nogales says.

    Applications for the first batch of Janelia's 24 group leaders—biologists, chemists, engineers, computer scientists, and physicists are all invited—are due 15 December. But be warned: Appointments, although renewable beyond the initial 6 years, will be untenured. “We want people who say, ‘Give me some resources and get out of my way,’” says Rubin. “That will appeal to some people [but] scare the daylights out of others.”


    T. rex Clan Evolved Head First

    1. Erik Stokstad

    Like reclusive celebrities, tyrannosaurs have risen to evolutionary stardom while keeping their origins shrouded in mystery. Now, the most primitive tyrannosauroid yet discovered has revealed the basic blueprint from which Tyrannosaurus rex and its kin evolved. The fossils, so well preserved that one even shows a “protofeather” fuzz covering the body, are described this week in Nature. Among other details, they show that tyrannosaurs began evolving the deadly design of their heads before their bodies morphed into powerhouses. “I think people are going to be tremendously excited about this,” says Matthew Carrano of the Smithsonian Institution. “It's certainly going to clarify a huge amount about the evolution of tyrannosaurs.”

    Paleontologists have found about a dozen species of tyrannosaurs. Most lived late in the Cretaceous Period, which ended 65 million years ago. Isolated bones have been found from older and more primitive tyrannosaurs, but not all have been accepted as ancestors. The new specimens—one fairly complete skeleton, plus parts of two others—come from western Liaoning Province in China. “It's the best primitive tyrannosauroid that we have,” says Thomas Holtz of the University of Maryland, College Park.

    Forging a head.

    Dilong paradoxus sported downy “protofeathers” and an advanced T. rex-like skull.

    CREDITS: X. XU ET AL., NATURE 431, 680 (2004)

    After farmers unearthed them, the specimens were studied by Xing Xu and colleagues from the Institute of Vertebrate Paleontology and Paleoanthropology in Beijing, along with Mark Norell of the American Museum of Natural History in New York City. Teeth and other features pegged the roughly 135-million-year-old creature as a tyrannosauroid. The skull has many familiar attributes, including bones shaped like those that apparently helped later tyrannosaurs launch swift, bone-jarring ambushes. The team dubbed the new creature Dilong paradoxus for “surprising emperor dragon.”

    Those surprises include features that distinguish D. paradoxus from its descendents. Its small body, 1.6 meters long, gives researchers a chance to study which aspects of T. rex's anatomy are truly tyrannosaurian rather than due to gargantuan size. And compared with T. rex, D. paradoxus had relatively long arms. Maybe developing the head for attacking—a safer approach than hands-on grappling with prey—enabled D. paradoxus's descendants to grow larger and handle bigger prey, speculates Oliver Rauhut of the Bavarian State Collection of Paleontology and Geology in Munich, Germany.

    Another previously unknown feature of tyrannosauroids is the soft pelt of 2-centimeter-long fibers, called protofeathers. These have been found in more primitive ancestors outside the tyrannosaur group, but large tyrannosaurs appear to have sported reptilelike scales instead. Norell proposes that smaller tyrannosaurs needed fuzz to stay warm but that their larger descendants, like modern elephants, shed their insulation to keep from overheating.


    Drugmakers Test Restrictions on Generics in U.S. Programs

    1. Jon Cohen

    Last spring, the Bush Administration triggered howls of outrage from AIDS researchers and activists around the world when it insisted that U.S. government programs could only use drugs approved by the U.S. Food and Drug Administration (FDA) to treat HIV-infected people in poor countries. Many saw the policy as a thinly veiled effort to favor big pharmaceutical companies over the manufacturers of cheaper generic drugs. The issue came to a full boil in July at the international AIDS conference in Bangkok, where several leading AIDS researchers lambasted the policy and AIDS activists disrupted a talk by Randall Tobias, the Administration's global AIDS coordinator (Science, 23 July, p. 470).

    In the next few weeks, some—but not all—of the heat could be taken out of this dispute. In an effort to defuse the issue, the Administration announced a plan in May for FDA to put applications from manufacturers of generics on a fast track, with a decision in 2 to 6 weeks. That commitment is about to be put to the test: Science has learned that South Africa's Aspen Pharmacare submitted an application to FDA in early September for six generic anti-HIV drugs it manufactures, and two Indian companies, Cipla and Ranbaxy, plan to file applications soon. If approved, these companies' drugs could eventually be used in the President's Emergency Plan for HIV/AIDS Relief (PEPFAR), a $15 billion program that aims to treat 2 million people in 15 developing countries over the next 5 years.

    Critics, however, are unlikely to be assuaged. For one, it may already be too late for any generics to qualify for the next round of treatment under PEPFAR: Companies that want to supply FDA-registered drugs to the program must submit their proposals by 15 October, and “it's going to come down to the wire whether we're registered by then,” says Stavros Nicolaou, a senior executive at Aspen. Moreover, generic drugs that have not been submitted for FDA approval would still be ruled out.

    Demand for access.

    Demonstrators in Bangkok protest policies that block the use of certain generic drugs in HIV/AIDS programs.


    Many AIDS researchers also question the rationale underlying the Administration's position: that generics might not contain “bioequivalent” doses of the branded drugs, allowing the virus to develop resistance more easily. “There's no biologic basis in the fear that slight differences in bioequivalence will make the slightest difference in effect,” says Bernard Hirschel, head of the HIV/AIDS unit at the University of Geneva in Switzerland. Herschel points out that the World Health Organization has already approved many generic drugs, and “it's hard to see that there's any substantial difference between the WHO and the FDA processes.” Insisting on brand-name drugs, he says, limits the availability of the most effective treatment strategies, sows confusion, and stymies cutting-edge treatment research in developing countries because U.S.-funded research projects must also use FDA-approved drugs.

    Daniel Berman, who coordinates a drug-access campaign for Médecins Sans Frontières, argues that the FDA requirement could inadvertently increase drug- resistance problems. Berman notes that fixed-dose combinations—ideally, one pill that combines three AIDS drugs—are easier for people to take and thus lead to better adherence to regimens, a critical strategy to avoid resistance. No big pharmaceutical companies make this fixed-dose combination, nor does Aspen. “The politics of the Bush Administration have prevented the easier treatment of patients,” charges Berman.

    Mark Dybul, chief medical officer for Tobias's office, says, however, that several African clinicians, citing adverse experiences in the past with poorly made generic drugs, have thanked him for insisting on FDA approval. “In the long run I think everyone will recognize this is the right decision,” says Dybul. “Our policy is safe and effective drugs at the lowest possible cost for everyone in the world.” He adds that the cost of branded drugs is only about three times that of generics and is dwarfed by the costs of building infrastructure and training clinicians how to use the treatments. “People are making this huge noise about a relatively small amount of money,” he says.


    A Slanted View of the Early Universe

    1. Charles Seife

    In the Atacama Desert of northern Chile, a microwave telescope has taken the best snapshot of an exquisitely faint echo of the big bang. In a paper published online by Science this week (, astronomers present detailed pictures of the polarization of the cosmic microwave background (CMB)—the faint and ubiquitous image of the fiery universe when it was less than 400,000 years old.

    “A photon is left with the imprint of the last few times it scatters” off the cloud of glowing gas in the infant universe, says telescope team member Carlos Contaldi, an astrophysicist at the Canadian Institute for Theoretical Astrophysics in Toronto. “[Polarization] is a very clean picture” of the structure of the cosmos when it was extremely young.

    The telescope, known as the Cosmic Background Imager (CBI), has been observing the CMB for years in hopes of testing theories about how the universe was born. Two years ago, CBI presented what was then the best picture of the CMB (Science, 31 May 2002, p. 1588). Even after the CMB-observing Wilkinson Microwave Anisotropy Probe (WMAP) blew most of its competition out of the water (Science, 19 December 2003, p. 2038), CBI still held an edge in observing very small features in the CMB.

    Now, the CBI team has released the results of nearly 2 years of observing polarization in the CMB: the extremely hard-to-spot directionality of incoming light. First detected by another instrument in 2002 (Science, 27 September 2002, p. 2184), the polarization paints a sharp picture of the early universe, as it remains relatively unchanged during a photon's multibillion-year journey to Earth. “It shows that [the primordial gas] was behaving exactly as we expected it to behave,” says team member Anthony Readhead, an astronomer at the California Institute of Technology in Pasadena.

    Even stronger confirmation is expected when WMAP releases its own polarization results, probably within weeks, and other ground-based experiments follow suit. But CBI will still provide data about features in the CMB that are too small for the other experiments to resolve.


    Astronomers Eager for a Swift New Vision of the Universe

    1. Robert Irion

    A long-awaited satellite should find scores of gamma ray bursts, sparking a rapid response from telescopes that span the globe

    “Swifts fly expertly on their first try,” a writer for National Geographic once observed about the graceful, darting birds. Astronomers trust that those words will hold true for a satellite called Swift, which hopes to start flitting around space next month in search of gamma ray bursts—the biggest explosions in the universe.

    “This is the first astronomical satellite that can rapidly change direction with its own onboard brains,” says principal investigator Neil Gehrels of NASA's Goddard Space Flight Center (GSFC) in Greenbelt, Maryland. That agility will allow Swift to swivel its “eyes” onto a new burst within minutes. What it “sees” should yield insights about the earliest incandescent moments of each explosion, thought to arise from especially violent supernovas that form black holes at their cores.

    The satellite also will send notice of every burst to a fleet of telescopes on the ground, from robotic instruments that respond in seconds to the planet's most powerful telescopes. The unprecedented reach of this network promises to lift the veils on what drives the tightly beamed blast waves.

    “Swift will take us from burst-by-burst science to very deep studies using hundreds of bursts,” says astrophysicist Joshua Bloom of the Harvard-Smithsonian Center for Astrophysics (CfA) in Cambridge, Massachusetts. Bloom hopes that Swift's most distant bursts will let scientists peer back to the first few hundred million years of cosmic history: “The early universe may have been a ripe petri dish for making what we think were the first gamma ray bursts. They are our best hopes for probing this very hot era in cosmology.”

    BAT's eyes

    The $250 million mission, a collaboration between NASA and institutions in Italy and the United Kingdom, was scheduled to take flight in December 2003 before several delays—most seriously, a 5-month overhaul of electronic components to make them more resistant to radiation. Recent damage to Florida's Kennedy Space Center from hurricanes Frances and Jeanne then pushed the launch from September to early November.

    The mission features a compact assembly of three telescopes. Swift will catch gamma ray bursts with its Burst Alert Telescope (BAT), which Gehrels calls “the most sensitive gamma ray imager ever.” Like someone staring upward to watch for meteors, BAT's gaze will encompass a large swath of the heavens (about 1/6) at any one time. An array of 32,768 cadmium-zinc-telluride detectors, covering a half-meter square, will register electronic blips from incoming gamma rays.

    Bird's eye.

    Swift will pivot in orbit to view evanescent gamma ray bursts.


    Because gamma rays are so energetic, they would pierce through the optics of a traditional telescope. Instead, BAT will interpret a geometrical pattern created by a “coded aperture mask”: a screen above the detectors with randomly placed square lead tiles. “A burst from a particular point on the sky will cast a unique shadow [through the tile pattern] onto the detectors,” says astrophysicist Craig Markwardt of GSFC. The satellite's software will calculate the location well enough for Swift to reorient itself toward the burst within about a minute.

    After the adjustment, the satellite's two other telescopes will zero in on the burst's rapidly changing cascade of energy. One telescope will gather x-rays to scrutinize the burst's internal chaos and its superheated interaction with material around it. The other telescope, sensitive to ultraviolet and optical light, will help gauge the burst's overall energy and its approximate distance from Earth—typically, several billion light-years.

    During each step, Swift will beam the burst's location and characteristics to the ground for e-mail flashes to astronomers worldwide. With each alert, instruments will race to that parcel of the sky “like an Oklahoma land rush,” says astronomer John Nousek of Pennsylvania State University, University Park, site of the mission control center. “All scientists will get all of the information as fast as is robotically possible.”

    The rush could happen often: Mission scientists estimate that Swift will spot 100 to 150 gamma ray bursts a year. But its two research telescopes aren't likely to observe the first critical seconds of many explosions. The sun, moon, or Earth could be too close for a safe view, and the satellite's pivot speed will be too slow to catch the initial flare for all but a few events.

    Ground patrol

    For the fastest response, the Swift team will rely on automated telescopes now deployed across the globe. Teams on six continents have built small new telescopes or have adapted larger existing telescopes to respond to Swift's electronic prompts— often within mere seconds. The web of ground teams, 39 and counting, will compose the most sweeping coordinated response to a satellite's observations. “It's become a cottage industry,” says astronomer Kevin Hurley of the University of California, Berkeley, who coordinates the follow-up effort. “Everything is now in place to reap all of the benefits of studying bright new sources that last only a half-day or so.”

    One such ambitious project is the Robotic Optical Transient Search Experiment (ROTSE), which has identical autonomous telescopes in Australia, Namibia, Turkey, and Texas. At least one of the 0.45-meter telescopes should be able to zip to a Swift position in less than 10 seconds. That's an advantage because efforts to track a burst's optical or infrared emissions from the ground must take place at night. “With apologies to our British colleagues, the sun never rises on the ROTSE array,” quips astronomer Donald Smith of the University of Michigan, Ann Arbor.

    Similar efforts in California, Chile, Europe, Hawaii, Japan, and elsewhere will provide global coverage of any given burst as Earth rotates. Even well-equipped amateur astronomers could provide useful results, says Hurley. But everyone expects the squadron of automatons to have growing pains. “These robotic telescopes are incredibly hard to operate and to keep running,” Smith says. “It's like Whack-a-Mole: As soon as you fix one thing, something else pops up.”

    Provided that some of the robots work as advertised, astronomers expect to see the first fires of gamma ray bursts more clearly than ever. That's critical for unraveling what happens at the heart of a titanic supernova, says astronomer Derek Fox of the California Institute of Technology in Pasadena. “At very early times, you observe the blast wave a short distance from the central engine,” he says. “The later you look, the less memory it has of the initial explosion.”

    Coded pattern.

    A lead-tile mask will cast a unique gamma ray shadow on Swift's detector for each burst.


    Although the robots will have the best shot to catch a burst's first sparks, the world's largest telescopes will join the act, too. Plans call for one of the 8.2-meter telescopes of the European Southern Observatory's Very Large Telescope array in Chile to swing to a new burst within 15 minutes or so, when feasible. One of the 10-meter Keck Telescopes in Hawaii will respond to some bursts as well. These mammoth mirrors gather so much more light than other instruments do that they will nail down the distances to the explosions—especially the faintest ones near the edge of the observable universe.

    The deepest probes?

    Indeed, the prospect of detecting such faint bursts is the stuff of dreams for astrophysicists. Currently, quasars—the active cores of galaxies, powered by supermassive black holes—are the brightest steady sources that astronomers can see in the young universe. These reach back to within about 1 billion years of the big bang. But in its first 20 minutes of raging energy, a gamma ray burst is 1000 times brighter than any quasar, says Bloom of CfA. “We may see when the first stars were dying,” Bloom says. Such a burst, far earlier than the quasar era, would illuminate all other matter between it and Earth to give astronomers the deepest possible cosmic probe.

    But it's not clear that the universe's earliest stars actually unleashed gamma ray bursts. Such stars were different beasts, with virtually no heavy elements and perhaps far more mass than later generations. If Swift sees no bursts within the first few hundred million years after the big bang, it will have revealed something fundamental about how those stars lived and died, Bloom notes.

    Another profound riddle that Swift may address is the origin of the shortest gamma ray bursts. A whole class of bursts flashes for fractions of a second, then vanishes (Science, 30 November 2001, p. 1817). Astrophysicists speculate that these events might arise from something never before observed, such as two neutron stars crashing together. “We're all really curious about what these are,” says Hurley. “No one has found a [glowing remnant] yet.” If Swift can do that, it may open a new window on the violent universe.

    With such rewards ahead, the Swift scientists are itching to fly. “It will be like waiting for the next firework to go off on the Fourth of July,” says Nousek of Penn State. “It's going to be a treat.”


    Gamma Ray Bursts: New Cosmic Rulers?

    1. Robert Irion

    One class of stellar explosions, called type Ia supernovas, erupts with surprising uniformity. They probably arise from white dwarfs that explode when they exceed a well-known threshold of mass. By correcting for subtle variations, astrophysicists turned the supernovas into “standard candles”: cosmic light bulbs of similar brightness (Science, 24 November 1995, p. 1295). That property has made type Ia supernovas the premier probes of the accelerating expansion of space, one of astronomy's landmark finds in recent years.

    At first glance, it seems unlikely that gamma ray bursts could serve the same purpose. Gigantic spinning stars—the favored progenitors of gamma ray bursts—have wildly varying masses, spin rates, heavy elements, and other properties. When the stars die, those factors apparently spawn bursts whose energies vary as much as 100,000 times from one burst to the next.

    But astrophysicists have found a physical pattern hidden within that drastic range. Each burst churns out light that peaks at a unique frequency. A spectral plot reveals that crescendo as a bump in the number of photons at that energy. Each burst also has a total output of energy: its “wattage.” For the best-studied bursts, researchers can derive that output by accounting for whether the explosion channeled its emissions toward us along a needlelike cone or a wider spray (Science, 30 November 2001, p. 1816).

    Those two quantities—peak frequencies of energy and total energy—are tightly correlated for gamma ray bursts, according to astronomer Giancarlo Ghirlanda of the Brera Observatory in Italy and his colleagues. “There is a very small scatter. It convinces us that something significant is going on,” Ghirlanda says, although theorists have no idea why the relation exists.

    Still, the correlation is so striking that just 15 gamma ray bursts already reveal the mass content of the universe and its expansion nearly as well as type Ia supernovas and other techniques, Ghirlanda says. His team confidently calls gamma ray bursts “new rulers to measure the universe” in the 20 September Astrophysical Journal Letters. A team from Nanjing University in China, led by Zigao Dai, reached a similar conclusion.

    Other astrophysicists are wary. A couple of noteworthy bursts don't fit the correlation, and the overall statistics are still shaky, say CfA astrophysicist Joshua Bloom and graduate student Andrew Friedman. “The biggest problem is the small number of bursts so far,” Friedman says. Swift's cornucopia of bursts should settle the debate, both sides agree.


    Robotic Telescopes Give Kids a Cosmic Classroom

    1. Daniel Clery

    Thanks to the Internet, schoolchildren can view the heavens via professional-caliber remote-controlled observatories. But are they ready for astronomical prime time?

    CAMBRIDGE, U.K.—At the mountaintop Haleakala Observatory on the Hawaiian island of Maui, a gleaming new telescope waits to peer deep into the cosmos. With a mirror only 2 meters wide, it is not in the front rank of such instruments, but it is a serious piece of research equipment. It is also entirely robotic: It can be controlled from a computer anywhere in the world, and no one need be on site from one week to the next. But astronomers eager to get their hands on it will have to wait their turn; this telescope was designed and built to be used by schoolchildren in the United Kingdom.

    The telescope and a twin still undergoing final tests and adjustments at Siding Springs in Australia are part of the Faulkes Telescope Project, an unprecedented effort to get children excited about astronomy in the hope that they will stick with science and mathematics as their education progresses. “It's not just about getting kids into astronomy. It's very rich in all sorts of disciplines,” says Dill Faulkes, the cosmologist turned software entrepreneur who put up $18 million to create the project.

    Faulkes is not alone. Another new British scope is setting aside observing time for schools' use. In the United States, a handful of long-standing projects are putting smaller scopes into the hands of schoolchildren and laying grand plans for networks of telescopes spanning the globe. “We can compete with MTV and get them hooked into something useful,” says astronomer Carl Pennypacker, founder of the U.S.-based Hands-On Universe.

    Astronomers may get a piece of the action, too. The Faulkes project hopes to team groups of students with professional astronomers to do some real science. The challenge “is to find a way of bringing kids and teachers up to a professional level,” says David Bowdley, educational programs manager for the Faulkes project. “And I'm sure professionals would like to get their hands on [the scopes] too.”

    Faulkes's motivation was simple. He was grateful for the free state education he'd received up to doctorate level, which led first to postdoctoral work and then to a successful career in the software industry. About 5 years ago he decided to give some of the wealth he'd created back to education. “I was concerned about children moving away from science in schools,” he says. After discussion with the U.K.'s astronomy funding body and staff at the Royal Greenwich Observatory, he settled on building a telescope in Hawaii so that children could see the night sky live during school hours. “Being able to observe in class time is a tremendous advantage,” says Bowdley.

    No toy.

    The Faulkes Telescope on Maui takes real astronomy into schools.


    The scopes are provided by the Liverpool-based company Telescope Technologies Ltd., which has pioneered building 2-meter instruments using a production-line approach to reduce costs (Science, 22 March 2002, p. 2203). The company's prototype instrument, the Liverpool Telescope, is sited in the Canary Islands. Its owner, Liverpool John Moores University, is devoting 5% of observing time to school groups through a project called the National Schools Observatory (NSO). The Liverpool Telescope became fully operational over the summer, and now NSO hopes to begin enlisting schools in earnest this term. “I haven't spoken to a teacher yet who doesn't want to do it,” says NSO's Andy Newsam.

    The Faulkes project also plans to ramp up its operations. It has been working with about a dozen schools to iron bugs out of the software and develop curriculum materials. Any school can sign up for the project. For about $340 they get three half-hour sessions, during which they can do what they like with the telescope, plus some offline observing time.

    Just before the summer vacation, Tim O'Brien, an astronomer at the Jodrell Bank Observatory near Manchester, tried the system out in a weeklong astronomy project with a class of 10-year-olds. “You have to prime them what to expect,” he says. “They're used to seeing things on a computer. You need to show them that this is a live telescope.” The class rifled through star charts, catalogs, and books and then picked a handful of objects to observe, including a supernova that exploded only a few weeks before. With the help of Webcams, the children get to see exactly what's going on. “When you move the scope, a light comes on in the dome, and you get a view of it changing position. That got a ‘Woooo,’” he says. “It's a field trip in your classroom,” says Faulkes's operations manager, Edward Gomez.

    Projects in the United States have been getting that sort of reaction for years. Philip Sadler of the Harvard-Smithsonian Center for Astrophysics (CfA) in Cambridge, Massachusetts, was one of the founders of the MicroObservatory project, which has taken a different approach from the Faulkes project by creating a network of four 15-centimeter scopes that snap pictures all night from a list of requests from students. “We found ease of use was important,” Sadler says. “They really want pictures of objects with which they have a connection: the sun, the moon, planets, and constellations.” Begun some 15 years ago, the MicroObservatory now takes about 50,000 pictures per year. A full half of the pictures taken, Sadler says, are of the moon: “It's a way in for them to the mysteries of astronomy.” The young observers also take a lot of pictures of the dirt around the telescope and of nearby cacti, Sadler says, but they soon learn that the object they're after has set: “It's important to fail. You learn more from failure than from success.”

    The Hands-On Universe (HOU), based at the Lawrence Hall of Science at the University of California (UC), Berkeley, grew out of a research project hunting for supernovae and has been operating since the 1990s, mostly using a 60-centimeter scope at the Yerkes Observatory in Williams Bay, Wisconsin. But Pennypacker, a scientist and educator at UC Berkeley and Lawrence Berkeley National Laboratory, says the project is “on the verge of a major expansion.” HOU has begun setting up 36-centimeter scopes in far-off locales so that students can use them live during class. The project now has one each in Hawaii and Australia and hopes to have another two in Australia within months. “In 10 years there will be hundreds,” Pennypacker predicts.

    High hopes.

    Dill Faulkes says sky-watching can spur a broad interest in science.


    Telescopes in Education (TIE) took a similar tack. It started in 1993 by automating a retired 60-centimeter telescope at the Mount Wilson Observatory in Southern California, which was donated by NASA, and making it accessible through the Internet. Now the project relies largely on 36-centimeter scopes at Mount Wilson as well as in Australia and Chile. TIE director Gilbert Clark is a firm believer in giving direct control of the scopes to students, comparing it to driving a Ferrari rather than taking a taxi ride. “You learn a lot behind that wheel,” he says.

    Is that learning mainly inspirational, or can schoolchildren do real science? “Most teachers are not interested” in research, Clark says. But if students have ambitious plans, TIE puts them in touch with astronomers. “They produce rather miraculous things sometimes,” he says. For the Faulkes project, doing science is part of the plan. Bowdley says that over the summer, secondary- school students, under supervision, made measurements of asteroids accurate enough to be submitted to the International Astronomical Union's Minor Planet Center, which keeps track of such objects. School groups can do valuable work making follow-up observations of fast-changing objects such as supernovae and gamma ray burst afterglows. “The more data you can get, the better, and they can make as good measurements as [those of] a professional astronomer,” says O'Brien.

    The astronomers and educators involved in these projects have little doubt that they are helping forge the scientists of the future. Sadler gets a kick out of meeting young astronomers at CfA who got their first taste of the stars through the MicroObservatory. Even so, most of the projects struggle to keep going on shoestring funding provided by the likes of NASA, the National Science Foundation, and the departments of Energy and Defense. TIE, for one, has had to cut back the number of students it can handle over the past several years. Says Pennypacker: “It's been hard, but it's been fun and it's been worth it.”


    Institute Sparks an Italian Renaissance in Mouse Biology

    1. Gretchen Vogel

    A young lab in the hills north of Rome is making its mark in mouse genetics—and in the science landscape of Italy

    MONTEROTONDO—Coffee, beer, and genetically altered mice are the staples of many modern biology labs, but in an up-and-coming institute outside of Rome, they have been raised to an art form. At the European Molecular Biology Laboratory (EMBL) campus in Monterotondo, Italy, a top-of-the-line espresso machine fuels work throughout the day, top German brews available at Friday beer hours provide a weekly chance to unwind, and the mice dwell in a sleek new 900-square-meter addition to the lab.

    The campus is home to six research teams and more than 20,000 mice, which bear genetic defects in dozens of genes. The groups study a grab bag of biological themes, including genetic influences on anxiety, the role of inflammatory genes in atherosclerosis, and the effect of cytoskeletal genes on brain development. “We're free to do anything we want as long as it has to do with the mouse,” says director Nadia Rosenthal, a developmental geneticist who left Harvard to lead the fledgling lab in 2001.

    Bright outlook.

    Nadia Rosenthal has led the EMBL outpost in Monterotondo since 2001.


    After initial growing pains, the young laboratory in the Tiber Valley 20 kilometers north of Rome is starting to make its mark on mouse biology. “They're definitely hitting the international community,” says developmental biologist Marianne Bronner-Fraser of the California Institute of Technology in Pasadena, who visited the campus earlier this year. The lab still has to prove its value in the long term, says geneticist H. Lee Sweeney of the University of Pennsylvania in Philadelphia, who collaborates with Rosenthal. “It needs to be productive over a period of time, and they haven't been fully functioning long enough. But I think people recognize now that it's going to work and there's tremendous potential,” he says.

    A few years ago, the picture was not so rosy. The institute was started in 1996 more out of political than scientific necessity: Italian authorities had complained they weren't getting their money's worth from participation in EMBL and had threatened to withdraw from the 17-country organization. As an incentive to keep Italy on board, EMBL proposed opening a campus outside Rome that would be devoted to mouse biology. At the beginning, there was funding for only three groups. The initial director, Klaus Rajewsky, kept his main lab in Cologne and was on site only part time. Few people even knew the campus existed. “We went through some rough times,” says EMBL director Fotis Kafatos. “It was difficult to recruit when funds were so limited.”

    Rosenthal says she received plenty of discouraging advice as she contemplated moving to Monterotondo. “I had heard that efforts to establish the campus were not going according to plan, and everyone had an excuse for why it wasn't working,” she says. High on the list was what both Italians and outsiders see as an unfriendly climate for science in the country, with labs burdened by layers of bureaucracy and limited funding. “Even my Italian scientist friends said, ‘Don't go there. It's suffering from necrosis,’” Rosenthal says. But a sense of adventure and a minor midlife crisis—she had just turned 50—prompted her to take the plunge, she says.

    The risk has paid off handsomely. Rosenthal has overseen the expansion of the institute's size and international profile. In combination with the European Union-funded European Mutant Mouse Archive, which moved in next door in 1999, Monterotondo is increasingly seen as a center for mouse biology in Europe. “It is playing a significant role in the international scene” working to make mouse models of human disease, says Bob Johnson, head of the British Medical Research Council's new Mary Lyon Centre in Harwell, U.K.

    And Rosenthal is enjoying herself. She and her nine-member group probe the effects of the hormone insulin-like growth factor-1 (IGF-1) and related molecules on muscle growth and regeneration. She continues to work on the muscle-bound “Schwarzenegger mice” that grabbed headlines several years ago. The mice carry an extra copy of a gene that codes for IGF-1, which not only bulks up their leg muscles but also seems to aid in wound healing and delay some signs of aging. She is now focusing on the role of the gene in heart muscle repair and regeneration. “This is exactly the kind of place where I want to come to work every morning,” she says. “I love it.”

    Mouse house.

    Genetically modified mice (right) make their home at the Monterotondo campus


    Stem cell biologist Claus Nerlov, who arrived just after Rosenthal, notes that the atmosphere has changed radically since the early days. “It used to be considered Siberia by people in Heidelberg [EMBL's headquarters],” he says. “Now they're starting to get jealous, which is good.”

    Kafatos is also pleased. “The campus is incredibly stimulating and abuzz with excitement, and it has gained the respect of the scientific community,” he says. “As far as I'm concerned, it has been a great success.” Kafatos, who has known Rosenthal since she was an undergraduate student in his lab and who helped recruit her to the post, attributes much of the success to her “ambitious but cooperative” leadership style. Bronner-Fraser agrees. “There's an interactive spirit there you don't see many places. They're all doing different things, but they're all somehow working together.”

    The newest recruit, Cornelius Gross, says it was largely Rosenthal's enthusiasm and collegial attitude that persuaded him to turn down university positions in the United States and join the Monterotondo campus. Gross studies the interaction of environmental and genetic factors, especially those related to serotonin signaling, in brain development and anxiety. He has found an unlikely collaborator on the campus in Walter Witke, who studies the genes that code for cell structure proteins. Witke's team found that one of those genes, profilin 2, is expressed only in neurons, and that mice with that gene knocked out had a strange phenotype: They seemed normal until they gave birth, when mouse mothers lacking the gene turned out to be completely uninterested in their offspring. The researchers suspect that the mutation affects the release and uptake of neurotransmitters, and the Gross team is now helping to characterize the mice using a battery of behavioral tests.

    Such unexpected collaboration is exactly what Rosenthal hopes to see. The wireless Internet network that enables lab members to answer e-mail or download research articles while enjoying the sunshine in the lab's courtyard encourages mixing among the groups. “You see two students sitting and talking in the courtyard, and soon two mice are getting mated that I never would have dreamed of,” Rosenthal says.

    Each group gets 500 free cages for their mice, and all receive most of their funding from hard money from EMBL, freeing them from pressures of grant writing. As at the main lab in Heidelberg, group leaders have a limited tenure at EMBL. They receive an initial 5-year contract that can be renewed only once for up to four more years. “All that fits into a ‘paradise for a decade’ idea,” Rosenthal says. “The promotional pressure is gone. There is deliberately no ladder to climb,” which also helps encourage a collaborative spirit, she says.

    It hasn't all been paradise. Although as an EMBL outstation the lab is free from much of the Italian government's notorious bureaucracy, Rosenthal has faced plenty of red tape while importing equipment and building the new mouse house. However, says Andrea Ballabio of the Telethon Institute of Genetics and Medicine in Naples, the Monterotondo lab sets a good example for science in Italy, where many researchers complain bitterly that fossilized organizational structures keep young scientists from getting ahead and stifle innovative research. “It's very important to Italy to have an EMBL campus,” he says. “They have the potential to influence Italian science” by recruiting top young scientists to the region and by “establishing a model” of an institute free from most of the constraints of government bureaucracy.

    The growth may continue. The campus received enthusiastic reviews from an international review team in September, and Kafatos says that it is possible the facility could expand in coming years. Bronner-Fraser says that would help strengthen the institute's remaining weak points. “They need to recruit a few more top postdocs,” she says, and they still need to work to become better known outside Europe.

    Although Italian politicians wanted a concrete return for their contributions, EMBL had its own reasons for establishing a foothold in Italy, Kafatos says. “We were keen to see a Europeanization of the research activities in Italy. The fact that we were able to inject [the EMBL] culture and let it take over so fast is really extraordinary,” he says. The espresso may have helped.


    Risky Business

    1. Jeffrey Mervis

    Can the U.S. government do a better job of betting on long shots in science? NSF and NIH hope the answer is yes

    Duke University neuroscientist Erich Jarvis won the National Science Foundation's (NSF's) prestigious Waterman Award for outstanding young researchers 2 years ago. But despite his early success, the assistant professor sounds like a battle-hardened veteran of the struggle for federal funding—in his case, for work on vocal learning. He certainly knows what it's like to have his ideas shot down.

    For example, Jarvis has cracked the code used by reviewers to undercut a grant proposal, especially the one that begins, “This is a very ambitious proposal. …” He's learned that those words, seemingly in praise of a novel scientific idea, are actually the kiss of death. And he sees irony in being penalized for trying something that nobody else has attempted—in other words, for proposing the sort of cutting-edge science that federal agencies profess to welcome. “You learn the hard way not to send high-risk proposals to NSF or [the National Institutes of Health], because they will get dinged by reviewers. Instead, you're encouraged to tone down your proposal and request money for something you're certain to be able to do.”

    That iron rule may be changing, however, at least for a few scientists. On 22 to 23 September, Jarvis was one of 15 outside scientists who spent 2 days telling a few members of NSF's oversight body and agency staff exactly what's wrong with the current system. They also suggested how NSF might become more receptive to the handful of ideas that have the potential to set the scientific establishment on its ear.

    Although the fruits of that meeting may not show up for years, if at all, on 29 September nine scientists received a more immediate payoff from NIH: $500,000 a year (in direct costs) for 5 years, with no strings attached. The money is part of a new program, the Director's Pioneer Awards, meant to allow researchers to pursue innovative ideas ( NSF and NIH are also working together on a separate initiative, mandated by Congress, to foster interdisciplinary research with long horizons. That work is inherently high risk, say government officials, who have scheduled a meeting next month to ask outside scientists how best to achieve that goal.

    “You learn the hard way not to send high-risk proposals to NSF or NIH.” —Erich Jarvis


    Together, these efforts represent a small but potentially significant move to alter conventions in grant reviewing, which many say favor timid, incremental steps over profound leaps of intuition. At the same time, the initiatives are quite modest, highlighting just how hard it is for federal agencies to encourage risk taking while remaining responsible stewards of taxpayer dollars. “We've been hearing for years that our process doesn't recognize work on the fringes,” says NIH Director Elias Zerhouni, whose $25-million-a-year Pioneer Awards program is a tiny piece of his road map initiative to reform the $28 billion biomedical behemoth. “So rather than continuing to debate it, I said, ‘Let's test that hypothesis and see how many scientists have good ideas that are not part of our portfolio.’”

    Real-time feedback

    Zerhouni is concerned that scientists don't even bother to submit their best ideas to government agencies anymore. This is part of the larger question of whether those agencies—and the outside reviewers they rely upon to help make funding decisions—would even recognize what the NSF workshop participants labeled “potentially transformative research.” Indeed, anecdotal evidence from the Pioneer Awards suggests that NIH may be missing the boat. “None of the 21 finalists thought that NIH peer review was ready for their idea,” says Stephen Straus, director of the National Center for Complementary and Alternative Medicine, who helped plan and implement the program. Indeed, several of the winners told Science that, despite receiving NIH funding for other projects in their labs, they had been forced to scrape together meager funding for their big idea.

    “I never even put in a proposal because the chances of getting an R01 [NIH's bread-and-butter award for investigator-initiated research] would have been zero,” says Steven McKnight, who is studying how the metabolic cycle in yeast influences circadian (or more frequent) cycles within the cell. “It's a new and unpopular idea, and it has no sex appeal—metabolism is boring—but I think it's pretty important,” says McKnight, chair of the biochemistry department at the University of Texas Southwestern Medical Center in Dallas.

    Another winner, Rob Phillips of the California Institute of Technology in Pasadena, suspects that he may have benefited from instructions to all reviewers to “suspend their usual paradigm” because of the risky nature of the proposals they were judging. A theoretical physicist now working on biological questions, Phillips is hoping to complete a “mathematicized” version of classic texts, including The Molecular Biology of the Cell by Bruce Alberts et al., that will illustrate how the laws of physics can be used to explain cellular behavior. “It's a scary project,” he says, “and I feel like a salmon swimming upstream, with the bears ready to rip me out of the water. But I'm committed to doing it, and this award gives me the resources.”

    The NSF workshop, held in Santa Fe, New Mexico, was convened by a task group of the National Science Board (NSB), which is mulling a formal study of the issue. The participants—some of them junior faculty members, some distinguished professors and national community leaders—offered heaps of personal testimony that parallels what NIH has learned on its own about the difficulty of funding novel ideas. But to their credit, the researchers heeded the advice of NSB member and workshop chair Nina Federoff, a biologist at Pennsylvania State University, University Park, and avoided turning the meeting into a mass whine about funding disappointments. Instead, the scientists suggested several ways to send the community a message that NSF wants to fund more transformative research.

    Many speakers endorsed a scheme to have the agency take a second look at unsuccessful proposals receiving both high and low scores, suggesting that some reviewers may have missed the significance of the idea being pitched. Another popular idea was to put investigators on call for reviewers to query during the course of a panel meeting, or give applicants a chance to respond to questions from an initial mail review before their proposal was taken up by a second panel. In each case, the proposed changes are driven by the assumption that high-risk research, because of its novelty, requires a more careful assessment by the agency.

    One ongoing experiment by NSF's biology directorate offers a partial answer to Zerhouni's concern about being ignored. Since 2000, program managers in the division of molecular and cellular biology have asked reviewers to flag any proposal that they believe is high risk. Although the percentage is very low (see graphic), there was a sudden leap this year in the number of such projects reviewers identified; the spike could mean that more scientists now think the agency will be receptive to risky ideas. In addition, the data show that a high-risk proposal stands a better chance of being funded than a run-of-the-mill submission. Maryanna Henkart, division director, hopes to understand the factors affecting those success rates, including any characteristics of the investigators themselves.

    At the edge.

    NSF biology reviewers saw a spike this year in the number of proposals that they considered high risk.


    Nitpicking conformists

    The shortcomings of the review process were a big concern to both NSF workshop participants and NIH officials designing the Pioneer Awards program. The peer- review system that allocates most public monies for basic research tends to reward scientists for finding flaws in the work of others rather than encouraging them to take risks. Two obvious reasons for that behavior, say researchers, are that scarce resources create a zero-sum game and that experts can prove their preeminence by tearing down other proposals in their field. “We profess to be seekers of truth,” says biomedical engineer and workshop participant Gerry Pollack of the University of Washington, Seattle. “But our scientific culture reinforces the idea that opposing the mainstream is bad.”

    Aware of peer review's leveling effect, NIH officials took steps to neutralize it when they designed a process for choosing the Pioneer Awards. First they created a separate pot of money—a total of $125 million, if Zerhouni keeps the competition going for 5 years, as promised. That eased concerns that the awards were siphoning off money from existing programs.

    The second change was to request short summaries—two pages in round one, and up to five pages for the second round— describing the new idea and its significance. “We felt it was important to focus on people, not projects,” Zerhouni explains. Unlike a 25-page R01 proposal, the Pioneer Awards' submissions were not critiqued for their methodology or technique because none of the proposals contained that level of detail.

    The biggest change from business as usual was in the selection of reviewers and what they were asked to do. Rather than assemble panels steeped in the proposer's subdiscipline, NIH chose distinguished scientists from across many areas. Then they told the reviewers to rely on their wisdom and weigh the project's contribution to the big picture.

    This advice dovetailed with NIH's invitation to applicants to think big. That's in stark contrast to the work described in a typical R01 grant application, where a scientist is likely asking for support for an incremental piece of a project that is largely done. “NIH has funded great people over the years,” says Straus. “But we tend to fund the next step, and then the step after that. It's a slow and risk-averse strategy.”

    “[The metabolic cycle] has no sex appeal, but I think it's pretty important.” —Steven McKnight


    The Pioneer Awards will let a few scientists take giant steps into the unknown. Some 1300 people applied, and 240 moved on to the second round after an up-or-down vote by at least two reviewers. After a second winnowing, the finalists were then flown to NIH for face-to-face interviews with a panel of luminaries.

    The nine winners, all men, are in their 40s and 50s. Most are tenured professors at elite institutions, and all but two have current NIH grants, in some cases as many as three. Straus notes that two winners owe their good fortune to decisions by individual institute directors to supplement the director's pot of money.

    The vanishingly small success rate—0.7%—has led some scientists to accuse NIH of tokenism. And whether even that handful are true pioneers won't be apparent for several years, Straus says. Asked what an acceptable rate of failure might be for the program, Zerhouni replied that “one big win” might well justify the total expenditures on 50 or so scientists. However, he promised that “they will be monitored more closely than any other project” because of the compelling interest in knowing whether the federal government can become more hospitable to innovative research ideas.

    Jarvis is also paying close attention to the Pioneer Awards, which he learned about only after his Duke colleague, biochemist Homme Hellinga, received one. He's also got a big idea—teaching a chicken, say, to imitate sounds as part of an effort to develop new tools for vocal learning, a subset of human language, that can be used to repair speech problems. Jarvis figures that it's too radical for NSF's current funding mechanisms, but he's hoping that the NIH program will spur NSF to come up with something similar. In the meantime, he's thinking about competing for the next round of Pioneer Awards, to be announced this winter. “I think it's a great idea,” he says, “and I'd love to get one.”


    Westerners Put Their Chips on 2007 Indian Moon Mission

    1. Pallava Bagla*
    1. With reporting by Andrew Lawler in Boston.

    Developing countries have started their own moon race, and scientists from cash-strapped developed countries are hoping to hitch a ride

    NEW DELHI—Western researchers often beat a path to developing countries to study endangered species, ancient civilizations, or traditional medicine, among other subjects. Now it's time to add planetary science to that list. Five scientists from around the world are jostling to get their experiments aboard an Indian spacecraft, Chandrayaan-1, that is slated to fly to the moon in September 2007.

    “Chandrayaan offers a very cost-effective means to gather critical and unique data on the moon while forging new cooperative relationships in lunar exploration,” says one of the finalists, Paul Spudis of the Johns Hopkins University Applied Physics Laboratory in Laurel, Maryland. Another finalist, Manuel Grande of the Rutherford Appleton Laboratory in Chilton, U.K., says he welcomes “the increasing opportunities for flying experiments on emerging space-nation launch vehicles and satellites.”

    The Indian probe is part of a second race to the moon, and this time the competition is not limited to two superpowers. The Indian Space Research Organization (ISRO) is reserving a 10-kg slot for a foreign research team aboard Chandrayaan-1 (Hindi for Voyage to the Moon), which will orbit 100 km above the lunar surface for a minimum of 2 years. The four Indian instruments will map the lunar topography and conduct x-ray and gamma ray spectroscopic studies. Some 30 scientists from 11 countries responded to an ISRO solicitation earlier this year to join the mission, and last month the list was whittled to five.

    Mission to India.

    Scientists from these five countries hope their experiments will be aboard Chandrayaan-1 when it's launched by India's Polar Satellite Launch Vehicle (inset).


    The 525-kg Chandrayaan is a bit larger than the 367-kg European Space Agency Smart-1 mission launched last year, but much less ambitious than the 1600-kg, 13-instrument orbiter Japan intends to send to the moon in 2006. However, it is not clear if Japan can meet that launch date. China is also planning a mission for as early as 2007, although details about the experiments and scope of the project are not known. NASA intends to launch a lunar orbiter in 2008 as part of a new initiative to return humans to the moon. But funding for the project is in question, and last week a congressional panel expressed concern that the orbiter plan might shortchange science.

    Given these uncertainties, space researchers say they welcome the chance to vie for a spot on the Indian probe. And the benefits cut both ways. The competition is designed to ensure “maximum scientific knowledge about the moon,” says ISRO chair Gopalan Madhavan Nair. Former ISRO chief Krishnaswamy Kasturirangan says it should also “enhance India's status as a potential partner in future space exploration.”

    Madhavan says that there may be room for more than one foreign payload on the mission, depending on size and power requirements. A decision is expected later this fall. Still in the running are:

    • Spudis, who proposes a radio technology instrument to measure scattering properties of the surface; this experiment can confirm the presence of water ice in the lunar polar regions up to a depth of a few meters. These deposits were first detected by the U.S. military's Clementine mission in 1994 and again by NASA's Lunar Prospector in 1998, although their total volume, thickness, and composition remain unknown.

    • Tsvetan Dachev of the Solar-Terrestrial Influences Laboratory at the Bulgarian Academy of Sciences in Sofia, wants to measure solar wind particle flux and map radiation around the moon. His instrument is cheap, small, and uses little power, he says.

    • Stas Barabash of the Swedish Institute of Space Physics in Kiruna has a joint proposal with India's Anil Bhardwaj of the Vikram Sarabhai Space Centre in Thiruvananthapuram to image the moon's surface composition and magnetic anomalies.

    • Urs Mall of the Max Planck Institute for Solar System Research in Katlenburg-Lindau, Germany, wants to build a near- infrared spectrometer to study the geological and mineralogical aspects of the lunar surface. It is aimed at the mysterious asymmetry that gives the moon a thicker crust on the far side and a thinner crust on the side facing Earth.

    • Grande, who proposes a high-quality x-ray spectroscopic map of the moon to shed light on “the key questions of the origin and evolution of the moon.”

    The winner must bring his own funds to the table to build and deliver the hardware to ISRO by early 2007, says Subash Chandra Chakravarty, program director of ISRO's space sciences office. The entire mission is expected to cost just under $100 million.

    Foreign scientists don't seem concerned about partnering with an organization that has never flown beyond Earth's orbit. “ISRO has the full capability to carry out the Chandrayaan-1 mission successfully,” says Barabash. And Barabash is spreading his risks. He also is working with China on a joint European-Chinese experiment called the Double Star Polar Satellite, which currently is studying the effects of the sun on Earth's environment. “I do have the experience of working with an ‘untried’ space program. And this experience is very positive indeed,” he says.


    Circling In on a Vulture Killer

    1. Fiona Proffitt,
    2. Pallava Bagla

    Scientists blame Asian vulture declines on a veterinary drug

    In the 1980s, Oriental white-backed vultures (Gyps bengalensis) were probably the world's commonest large birds of prey, circling India's skies in the millions. By devouring dead livestock, they and other vultures perform a vital task in many Asian countries: removing rotting carcasses that could spread disease to humans.

    Today, this cleanup squad is imperiled: Numbers of white-backed and long-billed vultures (Gyps indicus) have declined by more than 99% and 97% respectively in India since 1992, with similarly drastic declines recorded in Pakistan and Nepal and among the rarer slender-billed vultures (Gyps tenuirostris). It's “one of the fastest population declines recorded for any bird species,” says Rhys Green, a conservation biologist with the Royal Society for the Protection of Birds (RSPB) in Bedfordshire, U.K.

    After years of seeking an explanation for the vulture deaths, a surprising theory emerged in May 2003 at a conference in Hungary: Researchers identified a veterinary drug used on hoofed livestock as lethal to the scavenging birds. The hypothesis remains controversial, but a new study out this month offers further support for it. And last month, one of India's states announced that it would phase out the drug. But no one knows if it is too late to save the birds.

    Scientists initially suspected that vultures were succumbing to a viral disease, explains veterinary pathologist Andrew Cunningham of the Zoological Society of London (ZSL). In 2003, however, a consortium of scientists from the United States and Pakistan linked diclofenac—an anti-inflammatory drug used to treat livestock on the Indian subcontinent since the 1990s—to vulture deaths in Pakistan. Postmortems of 259 white-backed vulture carcasses from the Punjab province found that 85% had visceral gout—a condition caused by buildup of uric acid crystals on the internal organs, usually as a result of kidney failure. Tests on a subsample showed that those with gout had residues of diclofenac in their kidneys, and 13 of 20 captive vultures fed diclofenac-treated livestock also developed gout and died.

    “This was the first veterinary drug implicated in a large-scale effect on wildlife populations,” says Green. The results, published this February in Nature, were met with initial skepticism, particularly in India. It “was not intuitively apparent that there could be enough contaminated carcasses to cause a massive population decline,” says study leader J. Lindsay Oaks, a veterinary microbiologist at Washington State University, Pullman.

    Imperiled scavengers.

    Flocks of Gyps vultures are now becoming a rare sight in India.


    But further work by a consortium of scientists from the U.K., India, and Nepal—published online 21 July in Biology Letters—found tell-tale gout and diclofenac residues in a high proportion of dead and dying white-backed and long-billed vultures collected in India and Nepal. This demonstrated that the diclofenac problem reached beyond Pakistan, says Green.

    He and colleagues at RSPB, ZSL, and the Bombay Natural History Society (BNHS) in Mumbai have now used computer models of vulture demography to confirm that the rapid decline in populations of white-backed and long-billed vultures in India, Pakistan, and Nepal could be largely, if not entirely, attributed to diclofenac poisoning. According to their calculations, reported in the October issue of the Journal of Applied Ecology, less than 1% of carcasses would have to carry a lethal dose of diclofenac to account for the declines. “Every time a vulture feeds on a carcass, it's like Russian roulette,” says Green. “The trigger is pulled about 120 times per year, so even if a small proportion of the chambers are loaded, a lot of vultures are going to get killed.”

    But some raise questions about diclofenac usage. “There are large areas of India where vulture declines have been reported, but where there is minimal veterinary care for livestock,” veterinarians Joshua Dein of the National Wildlife Health Center in Madison, Wisconsin, and P. K. Malik of the Wildlife Institute of India in Dehradun told Science in an e-mail.

    Green discounts that argument as unsubstantiated. Estimates of total diclofenac sales by Vijay Teng, vice president of Indian pharmaceutical company Neovet, suggest that the drug is widely used in India, with 20 million large-animal doses sold per year, the equivalent of 5 million large animals treated.

    Some Indian scientists, like P. R. Arun and P. A. Azeez of the Sálim Ali Centre for Ornithology and Natural History in Coimbatore, India, maintain that it is “premature” to conclude that diclofenac is the sole cause of vulture declines. Other factors may be contributing, they say, and more needs to be known about how diclofenac affects the birds.

    “You don't need to know the mechanism to prove it's killing vultures,” counters Cunningham. Vultures may be particularly prone to diclofenac poisoning because they eat the liver and kidney of livestock, where the drug is likely to be more concentrated, he suggests.

    At summit meetings in Kathmandu, Parwanoo, and Delhi earlier this year, veterinarians, scientists, government officials, and representatives of conservation groups and pharmaceutical companies agreed that diclofenac should be phased out. The state government of Gujarat, India, was the first to act, announcing last month that it will cease purchasing veterinary diclofenac. There is hope that safe alternatives could be on the market “within months rather than years,” says Deborah Pain, head of RSPB's international department.

    Even as researchers try to nail down whether diclofenac is the only vulture killer, there's a major effort to establish captive breeding programs to rebuild the bird populations. A breeding center in Haryana state, India, already houses 39 vultures, and there are plans to build centers in West Bengal, Pakistan, and Nepal by early 2005. There's no time to waste: Finding enough birds to stock the Haryana center is already proving tough, warns ornithologist Vibhu Prakash of BNHS.

    Indeed, with some vulture populations halving each year, “the possibility to do anything to conserve them is rapidly disappearing,” says Green.