# News this Week

Science  21 Feb 2003:
Vol. 299, Issue 5610, pp. 1160
1. 2003 BUDGET

# Science Agencies Get Most of What They Want, Finally

1. David Malakoff
1. With reporting by Jocelyn Kaiser and Jeffrey Mervis.

Science Agencies Get Most of What They Want, Finally

Let the spending begin. After a nearly 5-month delay, the U.S. Congress last week finished work on the 2003 federal budget. The $397 billion package holds some good news for scientists, including major increases for the National Institutes of Health (NIH), the National Science Foundation (NSF), and NASA (see next page). But there was little time to celebrate. The new spending levels come on the heels of a 2004 budget request by President George W. Bush that would dramatically slow growth for those and several other science agencies (Science, 7 February, p. 806). So once again, the next move is up to Congress. Lawmakers were supposed to finalize spending plans by 1 October, but election-year politics stalled work on all but two of the 13 measures that make up the government's$2 trillion-plus budget. After extensive wrangling, the House and Senate wrapped up all the loose ends on 13 February, including a 0.65% across-the-board cut to satisfy an overall White House ceiling.

Biomedical research advocates were relieved that NIH emerged relatively unscathed. The agency gets a $3.8 billion increase to$27 billion, a 15% boost that essentially completes a 5-year doubling. About a third of the new funds—$1.2 billion—will go to the National Institute of Allergy and Infectious Diseases, mainly for bioterrorism research. That 50% budget boost is smaller than the 57% increase requested by the president. Lawmakers also came up short on the National Cancer Institute, leaving more for some two dozen smaller institutes. They also added provisions that funnel some of NIH's cash to other programs within the Department of Health and Human Services. Still, “we're very pleased,” says Steven Teitelbaum, president of the Federation of American Societies for Experimental Biology. “There is no better investment than biomedical research.” View this table: NIH advocates can't rest on their laurels, however. The White House's 2004 request would produce just a 1.8% increase in the agency's budget, far below the 10% that Teitelbaum's group says is necessary to sustain programs. At NSF, an 11% boost to$5.3 billion provides a healthy start on the agency's goal to double in size by 2007. Its research account does even better, rising nearly 13%. Congress also was receptive to NSF's growing backlog of proposed major research facilities. Legislators handed out $30 million to begin EarthScope, a seismic monitoring network, and$118 million more for seven other projects. But for the second year in a row, they rejected NSF Director Rita Colwell's proposal for a cluster of ecological observatories, called NEON.

The foundation's education programs receive a 3% boost. Lawmakers sliced $73 million off the Administration's flagship initiative, partnerships between universities and local school districts to raise student scores in math and science, which was seeking$200 million. The savings allowed legislators to add $17 million to a$5 million program to reward universities for producing more science and engineering majors and to boost annual stipends for three graduate fellowship and training programs from $21,500 to$27,500.

Like NIH, however, NSF will have to work hard to get those new funds, because the White House's 2004 request is just 3.2% above its new budget. And some core research programs, such as biology, would actually face cuts from 2003 levels. “We're pleased that Congress values NSF, … [but] we still have a long way to go,” says Samuel Rankin of the Coalition for National Science Funding, noting that the request is some $900 million below what NSF needs to stay on its doubling track. Another target of doubling proponents, the Department of Energy's Office of Science, had a lackluster year. Overall spending will rise just 1%, to$3.3 billion. Lawmakers did give DOE's nuclear physics budget a boost to provide additional run time at two major user facilities and funds to plan for a new rare-isotope accelerator. They also set aside $3 million to start work on a supercomputer to rival Japan's Earth Simulator. In other directives, Congress bumped Texas A&M University from the inside track on a new, multimillion-dollar research center to combat terrorism (Science, 9 August 2002, p. 912), broadening language to allow all comers. And it gave the Forest Service the go-ahead to accelerate a controversial effort to thin forests in order to prevent wildfires. 2. SPACE RESEARCH # Amid Troubles, Station Science Gains a Center 1. Andrew Lawler The chronic problems facing the international space station have suddenly become acute: The shuttle Columbia is gone, NASA has grounded the rest of the fleet, and the investigation is under way. But despite these troubles, there was good news for space station science last week. Congress gave NASA a green light to create a private institute for research aboard the orbiting laboratory. Scientists and NASA managers are hailing the move as an important step in strengthening the research credibility of the massive project. The final 2003 budget numbers (see previous story) were hashed out as concerned lawmakers extracted a promise from NASA Administrator Sean O'Keefe to enhance the independence of the panel investigating the Columbia tragedy. O'Keefe also used the 12 February joint House and Senate committee hearing to defend his plan to keep the aging shuttle fleet flying well into the next decade while pursuing an alternative small space plane. And he promised to both keep the station occupied and resume station construction as soon as possible. The 2003 spending bill, which gives NASA$15.4 billion—a $513 million boost over the previous year and$414 million more than the Administration asked for—allows the agency to take the first step toward a nonprofit institute for space station research. Modeled on the NASA-funded Space Telescope Science Institute in Baltimore, Maryland, the organization would give the research community greater control over setting science priorities for the station and allocating research time. Several previous NASA and National Research Council (NRC) studies have backed the idea, but it drew criticism from Alabama and Texas lawmakers fearful that it would weaken the Marshall Space Flight Center in Huntsville, which conducts much of the station payload operations, and the Johnson Space Center in Houston.

To break the impasse, NASA's chief of biological and physical sciences, Mary Kicza, put together a review team including Marshall and Johnson officials; they examined various options and recommended a nonprofit, nongovernmental organization. “I know I have the support of Marshall Space Flight Center and its director,” Kicza says. “And I wanted to have the support of the research community.”

NASA sent the report last month to Congress, which gave its stamp of approval in last week's bill. Kicza hopes to select a contractor for the institute by the end of 2004. The institute will be funded initially for approximately 4 years, at an estimated annual cost in the tens of millions of dollars. Researchers hope the new institute will make decisions “on the basis of science” rather than economics, says Lynette Jones, a mechanical engineer at the Massachusetts Institute of Technology in Cambridge, who served on an NRC panel that examines station use. Kicza says she shares the concerns of legislators that research management of the space station “is much more complex” than of Hubble. But stronger ties with outside scientists will help overcome that hurdle, she argues.

The legislation also contains a plan for solar system exploration that both the Administration and Congress have endorsed. Most of the $110 million boost in space science—$95 million—goes to a Pluto-Kuiper belt mission that the White House thinks is too expensive but agreed to back in the 2004 request. The bill also replaces a mission to Jupiter's moon Europa with one that would examine Europa and two other large icy moons, using advanced propulsion and power technology.

Congress chopped $40 million out of a$296 million request NASA made in November for a revamped space-launcher effort. Government and industry officials, however, say that the White House is likely to ask for a massive funding boost this summer for an orbital space plane, perhaps as high as $2 billion—which doubtless would spark a heated debate in Congress. The bill also includes a healthy smattering of earmarks for projects not requested by the White House. Notable are$1.35 million each for a life sciences center at the University of Missouri, Columbia, and for expansion of the Maryland Science Center in Baltimore. The Senate panel that oversees NASA funding is led by Senator Christopher Bond (R-MO) and ranking member Barbara Mikulski (D-MD).

3. NEUROSCIENCE

# Snooty Exchanges Are Key to Mouse Society

1. Greg Miller

How would you go about sizing up an attractive stranger—by maximizing eye contact, looking for a smile, or maybe sneaking a peek at the shoes? As visual creatures, we humans rely heavily on our eyes in social situations. But as anyone who's ever walked a dog in a crowded park knows, other animals have different means of checking each other out.

For animals ranging from elephants to mice, olfactory cues called pheromones convey a wealth of information about the identity and mood of other individuals. A whiff of the right pheromones can spark a fit of rage or a bout of mating.

Now, on page 1196, neuroscientists report a major step toward understanding how pheromone cues are processed in the brain. By recording the electric chatter of neurons in the accessory olfactory bulb (AOB) of mice as they sniff another mouse, the researchers have discovered neurons that vary their activity in response to the sex and genetic strain of the sniffee. This suggests that the AOB extracts important social information from pheromonal cues.

“It's a fantastic paper because it deals with a living, behaving animal,” says Catherine Dulac of Harvard University. “That's very important for understanding how pheromone detection leads to changes in behavior.”

In mice, pheromones are detected by receptors in a small, cigar-shaped cavity at the back of the nose, the vomeronasal organ (VNO). As a mouse sniffs around, the VNO expands and contracts, pumping fluid over the receptors, which then relay messages to the AOB. Because the VNO is only active when the animal is, recordings from anesthetized mice have yielded limited insights into how pheromones tickle the brain.

To listen in on AOB neurons in sniffing mice, neuroscientists Minmin Luo and Lawrence Katz of Duke University in Durham, North Carolina, used an ultralightweight, remote-controlled recording setup developed by co-author Michale Fee of Bell Laboratories in Murray Hill, New Jersey, to record from the brains of singing finches (Science, 31 January, p. 646).

Thus wired, a male mouse was placed in a circular arena to which another, lightly anesthetized mouse was then added. The test mouse gave the sleepy stranger a vigorous sniff test, probing its head and derriere with his snout. Within seconds, many AOB neurons boosted or dampened their activity. These changes depend on what's being sniffed. For instance, one neuron in the brain of a mouse of the CBA strain grew six times more active when the sniffed animal was a male BALBc mouse. The same neuron was inhibited by a CBA female and somewhat inhibited by a female of a third strain. Dozens of other neurons were excited or inhibited by different sex and strain combinations. The researchers tested only male mice, but they suspect that females' neurons respond the same way.

The study answers a lingering question about pheromone detection that has never been resolved, says Dulac: how pheromones are transmitted. AOB neurons only responded when mice were actively sniffing another mouse, suggesting that direct contact is required and seemingly ruling out the possibility that the pheromones are carried through the air.

The information about sex and strain extracted by the AOB is passed to other parts of the brain that determine the social relevance of chemical cues and decide whether circumstances call for an attack or an attempt to mate, says Michael Meredith, a neuroscientist at Florida State University in Tallahassee. “I think it's great,” he says of the current paper. “We're breaking the code of this very mystical and quasi-magical field of pheromone communication.”

The AOB may extract even more detailed information from pheromone cues than the current study reveals. Katz and Luo suggest that some AOB neurons may even be tuned for the “pheromonal image” of specific individuals, analogous to how certain neurons in the primate brain respond selectively to specific faces. Other researchers say it's not far-fetched. At this point, there's just no telling what all the nose knows.

4. PALEONTOLOGY

# Primitive Jawed Fishes Had Teeth of Their Own Design

Jagged, straight, artificially whitened—in whatever form they take, teeth are a marvelous invention, enabling us to rip into drumsticks or chew a caramel with abandon. The fact that these complex structures are always organized—into sturdy rows of molars, for instance—has caused researchers to assume that teeth evolved just once, in the common ancestor of jawed vertebrates. But on page 1235, Moya Meredith Smith of King's College London and Zerina Johanson of the Australian Museum in Sydney present evidence that teeth evolved at another time, independently, in a group of extinct jawed fishes called the placoderms.

The finding doesn't reveal how teeth came about, but if they evolved more than once, scientists may need to shake up a significant portion of the vertebrate family tree. “The relationships of the main groups of jawed vertebrates are entering a state of flux,” explains Michael Coates of the University of Chicago.

Some 408 million years ago, armored fish known as placoderms ruled the Devonian seas. The first jawed vertebrates, they caught their prey with impressive bumpy gums or bony cutting blades and tusklike structures made of so-called semidentine. Teeth were believed to have evolved—as a better way to capture prey—in a relative of placoderms, one that gave rise to all of the other major groups of jawed vertebrates, such as sharks, bony fishes, and the extinct acanthodians.

Now, Meredith Smith and Johanson have evidence that a group of advanced forms of placoderms, the Arthrodira, had true teeth. Instead of a random assortment of small spikes called denticles, they sported conical structures arranged in rows. In studying the tooth wear on well-preserved specimens from western Australia, the scientists realized that two members of this group added new teeth to the end of a row, like a pattern seen today in lungfish. Meredith Smith was able to slice through a few teeth and show that they are made of regular dentine, not semidentine.

Still, placoderms differed from living toothed vertebrates in several ways. New placoderm teeth took root in the back of the mouth, not along the margin of the jaw, says John Maisey of the American Museum of Natural History in New York City. Meredith Smith thinks that new specimens may resolve the difference. In any case, the presence of teeth in advanced placoderms—and their absence in more ancestral forms—indicate that they originated independently of other jawed vertebrates, Meredith Smith and Johanson say.

Philippe Janvier of the National Museum of Natural History in Paris, as well as Coates, cautions that the independent origin of placoderm teeth depends in part on placoderm phylogeny, which he says has not yet been rigorously established. If the Arthrodira are not as advanced as is currently thought, their teeth may have arisen from the same ancestor as all of the other jawed vertebrates, not independently.

Meredith Smith and Johanson are planning to reanalyze the placoderm family tree. Meanwhile, they speculate that teeth might have originated three or more times among jawed vertebrates. For instance, because some acanthodians lacked teeth and early sharks may have too, the researchers believe that these and other groups could have acquired teeth on their own. That wouldn't surprise evolutionary biologist Jukka Jernvall, who studies tooth development at the University of Helsinki, Finland. “Multiple origin of all the things that have something to do with teeth seems to be an emerging theme in evolutionary biology,” he says.

Each case may not be entirely de novo, however. Meredith Smith and Johanson speculate that although many of the basic genetic tools needed to make teeth might have evolved just once, the coordination that creates ordered patterns may be unique to each group. “We hope our work will be a spur to considering the development of teeth across a wider range of jawed vertebrates,” Johanson says.

5. MICROBIAL FORENSICS

# Report Spells Out How to Fight Biocrimes

1. Martin Enserink,
2. Dan Ferber

DENVER, COLORADO—Catching bioterrorists will require changes from the crime scene to the courtroom, according to a new report that concludes that the United States is not sufficiently prepared to combat biocrimes. It may also require scientists to band together in a new discipline.

The report,* presented at the annual meeting of the American Association for the Advancement of Science here on 16 February (see reports on gravitational waves and subterranean coal fires on p. 1177), outlines ways to detect outbreaks rapidly, handle evidence correctly, and use it to track down and prosecute the perpetrators. It's intended as a wake-up call in the aftermath of the October 2001 anthrax letters that killed five people and paralyzed the U.S. Senate. The government response to past bioattacks has been “reactive,” says committee chair Paul Keim, a microbial geneticist at Northern Arizona University in Flagstaff who helped the FBI analyze samples after the anthrax attacks (Science, 30 November 2001, p. 1810). “We'd like a more proactive approach,” he says, one that unites federal agencies such as the FBI and the Centers for Disease Control and Prevention with local police, health departments, physicians, and scientists.

The rarity of biocrimes has prevented the buildup of sufficient expertise in microbial forensics, says panel member and FBI researcher Bruce Budowle. Nor are doctors and public health workers equipped to deal with evidence, he says, noting that valuable clues for criminal investigators, such as a bug's precise genomic makeup, may have little medical or epidemiological value.

The American Society for Microbiology convened the panel in June 2002 to bring together forensics experts and microbiologists. The group advises that first responders must learn how to secure evidence, and that reliable test kits to confirm an outbreak are needed. Because genomic information can help pinpoint the source of an organism, the genomes of at least three different strains of each pathogen and up to 20 for the nastiest ones should be sequenced. And it calls for more basic research into how to exploit genetic variation in forensic work.

Before a suspect is brought to trial, the panel says, labs must improve quality- assurance and -control procedures to make sure they have reliable evidence that stands up in court. And new forensic techniques should be scrutinized by independent scientists. The panel's recommendations would apply to a wide range of biocrimes, Keim says, such as using HIV as a murder weapon and tampering with crops and livestock produced for human food.

The report “is an excellent step” because it makes the “specific tactical recommendations [needed] to push this field forward,” says forensics expert Randall Murch of the Institute for Defense Analysis in Alexandria, Virginia, a former deputy director of the FBI forensics lab. But more is needed, he adds, including a national summit to establish microbial forensics as a new scientific discipline. The specialty would combine expertise in conventional forensics with knowledge from microbial genomics, phylogenetics, and informatics.

Some of the panel's recommendations are already being implemented, Budowle says. The FBI and the Department of Defense have established a collection of strains, for example, and beefed up biodefense research at the U.S. Army Medical Research Institute of Infectious Diseases in Fort Detrick, Maryland.

Microbial Forensics: A Scientific Assessment.

6. EVOLUTION

# How Global Change Shaped the Squirrel Family

1. Elizabeth Pennisi

Wild yet cosmopolitan, squirrels are a diverse bunch. Ranging in size from just 15 grams to 7.5 kilograms, they and their relatives are adapted to many settings. Some are ground based and some live in trees; a few even soar through the air. Evolutionary biologists now have good evidence that shifting continents and global climate changes have helped create the diversity of the 273 species in this family.

John Mercer and V. Louise Roth of Duke University in Durham, North Carolina, after studying how these species are related, constructed an evolutionary tree or phylogeny and dated major events. They found that key branches of the tree sprouted in parallel with geologic and climate changes, they report online this week in Science (http://www.sciencemag.org/cgi/content/full/299/5610/www.sciencemag.org/cgi/content/abstract/1079705).

In addition to offering a long-awaited phylogeny on squirrels, the results are exciting because they “provide a plausible and provocative scenario for the diversification of squirrels,” says Richard Thorington, a mammalogist at the Smithsonian Institution National Museum of Natural History in Washington, D.C. For example, the data show that squirrels became common in South America only after the continent became connected to North America.

Mercer and Roth began with a modest goal. For their research on morphological features in different-sized species, they needed to know how the species were related. That question turned up an unexpected result: The pygmy squirrel they were studying had branched off earlier than expected from the main tree. The researchers expanded their analysis, spending the next decade gathering tissue from museum and live specimens representing 50 of the 51 genera. “This is an excellent species sampling, encompassing the whole biodiversity of living squirrels,” says Emmanuel Douzery of the University of Montpellier in France. In an analysis Douzery praises, Mercer and Roth compared DNA sequences from three genes to determine which genera were close kin to others.

The work confirmed the researchers' initial view that the pygmy squirrel had branched off early. It also revealed surprises —for instance, that flying squirrels all evolved more recently from a single ancestor than some had thought.

Mercer and Roth then calculated when various genera evolved, based on the number of changes in the DNA bases. Next, they looked for correlations among certain global changes—such as shifting continents—and these landmarks in squirrel evolution. In addition to the insights into squirrel life in Africa and South America, the study indicates that in Southeast Asia, a rising sea level broke widespread ancestral genera of tree squirrels into isolated populations, which quickly evolved into separate entities.

The work is part of “a clear trend [to investigate] how speciation events may be explained by geological and environmental changes,” says Douzery. And that, adds Harvard University paleobiologist Lawrence Flynn, is leading to “a brave new world in which the data allow us to [envision] extraordinary scenarios.”

7. SCIENTIFIC WORKFORCE

1. Jeffrey Mervis

Twenty years ago, Phillip DeLeon attended a Massachusetts Institute of Technology (MIT) summer program for minority high school students that he says changed his life. “I don't know where I'd be now without [it],” says DeLeon, an associate professor of electrical engineering at New Mexico State University in Las Cruces. But this month, MIT announced that, under pressure from the government, it will now accept white students into that program and into another, similar effort to boost minority participation in the sciences. The Cambridge school's retreat comes at the same time that it is urging the U.S. Supreme Court to uphold race as a factor in admissions, in pending cases involving the University of Michigan.

Last May, the Department of Education's Office of Civil Rights told MIT that it was investigating both programs after receiving complaints that they discriminated against nonminorities. “They're right; the programs are racially exclusive,” says Robert Redwine, MIT's dean of undergraduate education, who disclosed the changes last week. Racial grouping has served MIT's goal of broadening minority involvement, he says, “by getting these students excited about science and engineering.” Each program serves about 60 African-American, Latino and Hispanic, and Native American students a year.

MIT officials count the programs, which are among the last of their kind at a U.S. university, as a huge success. Nearly one-third of the graduates from DeLeon's program—called MITES, or Minority Introduction to Engineering, Entrepreneurship, and Science—have gone on to attend MIT, for example, with graduation rates slightly higher than the norm. MITES introduced DeLeon, who came from a working-class family living in a Texas city with a “terrible” school system, to a world he never knew existed: “black and women engineers and other Hispanics who were [all] extremely talented.” Although he didn't attend MIT—“even after the scholarship, the tuition was more than my father's annual salary”—he says the experience convinced him to become an engineer.

Likewise, the Interphase program (which serves about a third of MIT's incoming class of minority students) helps those who might be intimidated by MIT's rigorous academic program, despite excellent test scores and grade point averages. “After you've been surrounded by a bunch of very bright minority students, you begin to have confidence in your ability to do this type of academic work,” says Paula Hammond, an associate professor of chemical engineering at MIT and an African American who was also trained at MIT. “Without that sense of community, it's very hard for them to feel comfortable at MIT.”

Given such positive feedback, university officials say they are reluctant to make changes but decided not to risk losing a legal challenge. Redwine said that he hoped both programs could be expanded (MITES depends heavily on outside contributions), but that he hadn't set a target for majority enrollment “because that would be a quota, and nobody wants that.” MIT hopes to add disadvantaged white students to the mix by using such criteria as the level of parents' education and the range of high school course offerings.

Outside experts worry that something may be lost if these minority-only programs are transformed. “MIT hung in there for a pretty long time,” says Daryl Chubin, vice president of the National Action Council for Minorities in Engineering (NACME). “I'm concerned about the number of minority students who won't be able to participate in such programs if they disappear.”

To forestall that, NACME has joined IBM, DuPont, Stanford University, and the U.S. National Academies in signing onto MIT's Supreme Court brief in the Grutter and Gratz v. Bollinger cases. Filed on 17 February, the brief argues that race-conscious criteria are a valid way to achieve socially beneficial diversity on campus. Jamie Lewis Keith, MIT senior counsel, says that “the important thing is that the court not limit our ability to take race into account” in striving to increase minority participation.

8. NEXT LINEAR COLLIDER

# Collision Course With Reality

1. Adrian Cho is a freelance writer in Grosse Pointe Park, Michigan.
2. With reporting by Gretchen Vogel in Berlin and Dennis Normile in Tokyo.

Particle physicists around the world have given themselves a year to settle on a design for a $6 billion particle smasher. But can they disentangle the technology from the politics? How does a physicist cook a chicken? Step One: Assume a spherical chicken. … That lab-bench chestnut overstates the case. Physicists may reflexively focus on abstract principles and overlook niggling details, but if they're hungry enough, they'll settle for poultry that's not perfectly round. Yet the general point is true: Physicists are an idealistic lot. Indeed, the world's particle physicists have set themselves a task that may be as quixotic as snaring a globular hen. Within a year, they will decide on the basic technology for a$6 billion particle smasher they all agree they need. And they say they will make that choice without regard to the politics of where to build the prize machine.

The 30-kilometer-long “linear collider” will smash electrons into their antimatter cousins, called positrons, with such violence that exotic particles several hundred times as massive as a proton may pop into existence. The collider could explain the very essence of mass or map out whole new dimensions. And for the sake of science, most physicists say they should dispassionately choose the basic design for the machine before haggling over whether to build it in North America, Europe, or Asia and how to split the costs. “Whatever we adopt as an international community, we all [should] have input into the final design,” says Jonathan Dorfan, director of the Stanford Linear Accelerator Center (SLAC) in Menlo Park, California, and chair of the International Committee for Future Accelerators (ICFA).

Practically, however, physicists may have better luck finding a Higgs boson in a haystack than separating the technology and politics of the linear collider, as researchers from different regions are pushing different designs. American and Japanese physicists have teamed up to work on a relatively conventional technology. After an unexpected setback, they hope to show by the end of the year that their idea works as planned. Meanwhile, German physicists and their international collaborators have designed a slightly lower-energy machine based on a more novel technology that they have already tested. They have submitted a formal proposal to the German government and are essentially ready to go, says Albrecht Wagner, chair of the board of directors of the German Electron Synchrotron (DESY) in Hamburg. “There are clearly two proposals,” Wagner says. “In all fairness, one is further ahead than the other.”

Some physicists argue that the German machine has a decisive edge. Others counter that the American and Japanese effort will soon catch up and will ultimately prove the better choice. Either way, most physicists agree that the global community will have to decide on the basic technology soon, but they don't always agree on why. Most say they must choose by early 2004 to complete the collider early in the next decade, in time to run in tandem with a complementary machine now being built in Switzerland. Some say that the German group is pushing the pace. And after more than a decade of contemplation, more and more physicists feel they need to get cracking if the grand idea is ever to become a reality, says Maury Tigner of Cornell University in Ithaca, New York, who heads the International Linear Collider Steering Committee. “There is a growing awareness that we'd better have made up our minds by the end of this year,” Tigner says, “or we might as well throw the whole thing down the drain.”

Yet no one knows precisely who will make the choice or how.

## Technologies hot and cold

Particle physicists in Asia, Europe, and the United States have agreed that the linear collider should be at the top of the community's wish list. The unprecedented consensus reflects physicists' conviction that exotic new particles, and perhaps even new dimensions of space and time, lurk over the horizon. The discoveries themselves will probably fall to the Tevatron, an accelerator that smashes protons into antiprotons at Fermi National Accelerator Laboratory (Fermilab) in Batavia, Illinois, or to the more powerful Large Hadron Collider (LHC) under construction at CERN, the European particle physics laboratory near Geneva. LHC will smash protons into protons starting in 2007 and should run for roughly 20 years. Only the electron-positron collider, however, can nail down the essential properties of the new particles and dimensions, physicists say (see p. 1171).

The proposed electron-positron collider would be unlike any ever built. For starters, it cannot be circular, as was CERN's Large Electron Positron Collider (LEP), which shut down in 2000. When charged particles go in circles, they lose energy by radiating light. The effect makes it increasingly difficult and expensive to push particles to higher energies. To reach energies several times higher than LEP, the new machine will have to consist of two opposing straight accelerators, each of which will have to set records for accelerating particles from zero to near light speed. And the pricey machine will have to be a global collaboration, as no one country or region of the world can afford it.

Accelerator physicists are developing two basic technologies for the innards of the machine. Within any accelerator, particles gain speed and energy by surfing electromagnetic waves. Much as sound waves ring in organ pipes, these electromagnetic waves ripple down elongated vacuum chambers called cavities (see figure, p. 1168). Physicists in the United States and Japan are developing room-temperature copper cavities similar to the ones that currently power a 3-kilometer linear accelerator at SLAC. German physicists and their collaborators are working on superconducting niobium cavities, which must be cooled to a few degrees above absolute zero. Neither design suffers an obvious fatal flaw, says SLAC's Gregory Loew, who recently led a technical review of the technologies for ICFA. But, Loew says, both designs still must prove that they're up to the task.

## Germany: A fast start

In the village of Rellingen, about 10 kilometers northwest of Hamburg, stands one of the most famous churches in northern Germany. An exemplar of Late Baroque design, the Rellinger Kirche has escaped damage from war and fire for two-and-a-half centuries. A cupola sits atop the octagonal building, and from the center of its ceiling a painting of “God's eye” gazes down. Someday, that eye may stare directly at a machine that will help unravel the very fabric of creation.

Known as TESLA (for Tera-electron-volt Energy Superconducting Linear Accelerator), the machine would run in a tunnel 8 meters directly below the church and stretch from DESY in Hamburg 30 kilometers toward the North Sea. It would also mark a triumph for the superconducting technology, which a decade ago couldn't pack the punch needed for the linear collider.

Within a cavity, the rate of acceleration depends on the strength of the oscillating electric field, or gradient. In the early 1990s, superconducting cavities could produce gradients of only a few million volts per meter, a few times more than the gradient needed to drive lightning through air. For the past 2 years, however, TESLA researchers have run cavities in their test facility that produce gradients of 23 million volts per meter. That's enough to produce collisions with energies of 500 billion electron volts, more than double LEP's limit and the minimum starting energy for the collider. “The TESLA guys have done a remarkable job in developing that technology,” says Robert Kephart of Fermilab. “They've made improvements that a lot of people didn't foresee.”

Just as important, the TESLA team has made substantial political progress toward transforming its design into a reality. In March 2001, TESLA researchers released a technical design report, a preliminary blueprint for the machine and, as for any large project, the crucial first step toward approval. In July 2002, the German Science Council gave the project a positive evaluation, and TESLA researchers have finished an environmental review of the site and calculated a detailed cost estimate.

TESLA researchers still have a plateful of work in front of them. Physicists generally agree that a linear collider must eventually reach energies of at least 800 billion electron volts. To do that, TESLA's cavities will have to produce gradients of 35 million volts per meter, a goal that should be obtainable within a year, says DESY's Dieter Trines. Politically, TESLA researchers must convince their colleagues around the world that TESLA is the right collider to build. DESY researchers had hoped that the German government would make a bold first move, approve the project, and ante up roughly half the cost, on the condition that other nations sign on. But on 5 February, German Research Minister Edelgard Bulmahn said that until there is stronger international support, the cash-squeezed German government could not commit itself to building the machine.

Some physicists argue that TESLA's technological and political advantages are so decisive that researchers from the United States and Japan should simply join in the German project. “From my point of view, we absolutely should jump into the machine and get it done quickly,” says Gordon Kane of the University of Michigan, Ann Arbor.

But even European physicists warn that continental politics may soon put the brakes on TESLA's smooth progress. Many physicists anticipate political friction with nearby CERN as TESLA competes for resources. Moreover, before German physicists and officials begin wrangling with their counterparts in the United States and Japan, they must persuade their European colleagues to sign on, says Ian Halliday of the Particle Physics and Astronomy Research Council, a British funding agency in Swindon. That will take time, he says.

All agree that the TESLA project cannot idle for long before it loses its precious political momentum. “Once you've put your proposal on the table and had it reviewed and gone to the government,” Halliday says, “if nothing happens for 2 years, you're dead.”

## United States: Looking for a champion

Although TESLA may be off to a fast start, many physicists think that the American and Japanese design will prove more desirable. Superconducting cavities stop superconducting if the gradients and concomitant magnetic fields grow too large, they point out, so TESLA might not be able to reach energies much above the hoped-for 800 billion electron volts. Conventional copper cavities ought to produce gradients at least twice as high.

Researchers at SLAC are collaborating with colleagues from the High Energy Accelerator Research Organization (KEK) in Tsukuba, Japan, to develop a common design. Their machine is known as the Next Linear Collider (NLC) in the United States and as the Japanese Linear Collider (JLC) across the Pacific Ocean. It employs cavities similar to the ones used for nearly 40 years in SLAC's linear accelerator. But the new cavities have inner dimensions that are only a quarter as big as those in the current machine, and they resonate with electromagnetic waves from a much higher frequency range, known as the X-band—all of which makes them more powerful.

The new cavities threw researchers a curve ball in 2000. When physicists tried to crank up the gradient, the electromagnetic fields would occasionally vanish as a kind of spark flashed through the cavity. Known as breakdown, this process gradually pitted the insides of the devices and ruined them.

SLAC and KEK researchers think they have solved the problem by changing the shapes of various parts to reduce surface currents and fields. They expect to have complete cavities that will reach gradients of 65 million volts per meter by the end of the year. “It's a matter of putting together pieces that have already been demonstrated,” says SLAC's Marc Ross. If the cavities make the grade, the X-band machine will be able to produce collisions with energies of more than a trillion electron volts, says SLAC's David Burke, and that could be a crucial advantage.

But although the X-band technology may be poised to catch up to its superconducting rival, NLC is struggling to gather political steam. In 1999, the U.S. Department of Energy (DOE) capped spending on NLC at $19.2 million annually and denied researchers money and permission to write an official conceptual design report. And the 2004 federal budget request contains no extra money for NLC. DOE spending on particle physics, flat for a decade at roughly$700 million per year, would have to increase by 30% if the United States is to build a collider and by 10% if it is to play a major role in a machine built elsewhere, says Fred Gilman, a physicist at Carnegie Mellon University in Pittsburgh, Pennsylvania, and chair of DOE's High Energy Physics Advisory Panel.

Nonetheless, American physicists are cautiously optimistic that their fortunes may quickly change. NLC has a shot, they say, if someone high enough in the current political administration takes up its cause and pushes it up the chain of command. All agree that the key first link in that chain of persuasion is Ray Orbach, director of DOE's Office of Science.

For his part, Orbach says that NLC has to compete with other major projects that DOE might fund in the next 20 years. He has called for a review of the proposals, which include an accelerator for radioactive nuclei and an international fusion experiment known as ITER (Science, 7 February, p. 801), and he expects a report in March, in time to set 2005 budget priorities. Still, Orbach says, by hammering out consensus, particle physicists are maximizing their chances of getting the collider. “I think the physics community is doing the right thing,” he says. “They're being methodical but not slow.”

## Japan: “It's our turn”

Although Japanese physicists are collaborating with the Americans, they insist that the collider should be built in Japan. “Science is not just between Europe and the U.S.,” says Hirotaka Sugawara, director-general of KEK. “Asians can claim that it's our turn to have such a facility.”

The Japanese researchers have also taken a more flexible approach to technology than their American counterparts have. Although most are working on the X-band design, a small team under the leadership of Tsumoru Shintake at the SPring-8 accelerator in Harima Science Garden City is independently developing an alternative. Similar to the X-band technology, it uses copper cavities that have slightly bigger inner dimensions and ring with lower frequency C-band electromagnetic waves. A C-band collider would produce collisions of only 300 billion or 400 billion electron volts—a third as energetic as an X-band machine's—but would be much easier to build. As a backup plan, the researchers envision starting with C-band cavities and swapping in X-band cavities when that more ambitious technology matures.

JLC does not have a site or an official budget, although researchers are spending roughly $5 million a year on research and development. To jump-start the project, Japanese physicists are writing a conceptual design report. “It hasn't quite been asked for by the Japanese government,” says KEK's Nobu Toge. “We're sort of volunteering it.” Japanese physicists are less sure than their Western counterparts about the need to choose a technology within a year. “We need a little more time for R&D,” says Sachio Komamiya of the University of Tokyo. And Sugawara questions the entire protocol of choosing a technology before starting in on the politics. Instead, he says, physicists should establish a central design group that would develop all the technologies in a unified way—even as researchers slug it out politically. ## How to choose? Most particle physicists argue, however, that governments will be more receptive to the project if researchers first agree on what they want. They will start deciding how to decide at a meeting in Appi, Japan, at the end of February. The technology choice will likely fall to a committee appointed by ICFA or the International Linear Collider Steering Committee. But choosing the committee members may be tricky, because most qualified experts are involved in one project or the other, says Brian Foster of the University of Bristol, U.K., who chairs both the European Committee for Future Accelerators and the European Linear Collider Steering Committee. “The committee that has to make the decision has so many vested interests that it's not clear how one makes a decision unless it's obvious,” Foster says. Unless one of the technologies unexpectedly fails, the choice could hinge on subtle tradeoffs between the designs. For example, an X-band collider should reach higher energies, but a superconducting collider should produce data at a greater rate. Comparing costs may be tricky, as American, German, and Japanese labs all use different accounting methods. In the end, the technology may simply follow the money. “If the German government comes up with the bulk of the money, it's natural to use the cold technology. If the Japanese government comes up with the bulk of the money, it's natural to use the warm technology,” Sugawara says. Of course, that's precisely the less-than-ideal eventuality that most physicists hope to avoid. They've given themselves a year to find a way around it. While they're looking, maybe they'll spot that spherical chicken, too. 9. NEXT LINEAR COLLIDER # Why Physicists Long for the Straight and Narrow 1. Charles Seife Europe's Large Hadron Collider should be powerful enough to find a new generation of particles, but not to get to know them well To outward appearances, there's little excitement in particle physics these days. All the particles predicted by physicists' Standard Model have been found and are being studied, so it might seem that scientists have little left to do but mop up the details. Researchers beg to differ: Far from having finished their work, they feel they are on the brink of something much bigger. The buzz centers on the Large Hadron Collider (LHC), a superpowerful proton smasher, slated to come online in 2007. Particle physicists are confident that this machine will help them break out of the straitjacket of the Standard Model into a new realm of particles. “It's going to inaugurate a new golden age in particle physics,” says Frank Wilczek of the Massachusetts Institute of Technology (MIT) in Cambridge. “There's a whole new world of phenomena” that LHC is likely to discover, he says, “with very rich and very fundamental implications.” Physicists such as Wilczek believe that, by the end of the decade, they will be studying a whole new cast of characters. The Higgs boson will explain how particles acquire their mass, and each of the known particles in the Standard Model will acquire a heavyweight “superpartner” with names such as the “squark” and the “selectron.” Not only will this revolution bring physicists closer to a “theory of everything,” but it may also identify the exotic dark matter that vastly outweighs the stars and galaxies that make up the visible universe. But in some ways, LHC is a crude instrument, a chainsaw that will tear up the high-energy frontier. The machine should find the new particles, but when physicists pick through the debris of shredded protons that the collider creates, they will be unable to figure out the exact characteristics of these new particles. So, in addition to their chainsaw, particle physicists also want a scalpel: a vast linear collider, 35 kilometers long and designed for smashing together electrons and antielectrons, rather than LHC's protons with protons. The linear collider will zoom in on the candidates and reveal the nature of the new beasts in the particle zoo. “There should be several, four or five” accessible to the new colliders, Wilczek says. “These are not theories that can hide themselves indefinitely.” ## The Standard Model and beyond LHC and the proposed linear collider are simple in principle but fiendishly complex to achieve. They use intense electromagnetic fields to give beams of charged particles a sustained push. The particles acquire more and more energy, moving faster and faster, and then smash into a beam of their own antiparticles moving in the opposite direction. Accelerators are the microscopes of high-energy physics. Because of the laws of quantum mechanics, the higher the energy of the collisions, the shorter-lived and more massive the particles they create. And just as biologists learned more about the cell as they got better optics on their microscopes, physicists learn more and more about the subatomic world as they get more capable and more energetic accelerators. As microscopes go, LHC is a monster. Currently under construction in a ring-shaped, 27-kilometer tunnel at CERN, the European particle physics laboratory near Geneva, LHC will cost$1.6 billion. It will smash together beams with a combined energy of 14,000 giga electron volts (GeV), roughly a billion times the energy of the electron beam that illuminates a TV screen. Each collision will create a swarm of particles, some of them debris from the original protons but the majority created from the raw energy of the collisions. When physicists analyze the tracks of those particles, they can figure out which ones were created and destroyed and which forces govern their interactions.

It was in just this way that physicists working with LHC's predecessor at CERN, the Large Electron-Positron Collider (LEP), and with the Tevatron at Fermi National Accelerator Laboratory near Chicago filled in the details of the Standard Model. This mathematical framework binds together all of the fundamental particles that make up matter—quarks, neutrinos, electrons, taus, and muons—with those that carry the forces between matter particles—photons, gluons, and the W and Z bosons. The Standard Model has been a stunning success. It allows physicists to calculate the properties of subatomic particles with incredible precision. But there are some problems with the model, problems that have been nagging physicists for decades. Many researchers think that LHC and the proposed linear collider will resolve two of the biggest of these problems and show the way to a more complete and logical theory.

The first of these problems is mass. Strictly speaking, the plain-vanilla version of the Standard Model doesn't allow particles to have mass at all. Whenever physicists plug masses into the equations of the model, other values swell up to infinity and become meaningless. In a way, mass just does not fit naturally into the model.

British physicist Peter Higgs came up with a solution in the late 1960s. He proposed a new particle, now known as the Higgs boson, to add “stickiness” to particles in the Standard Model. Higgs bosons, which are thought to exist everywhere in the cosmos, interact with other particles like screaming fans grabbing onto a passing celebrity. Just as the fans add an effective “resistance” to the celebrity's attempts to move, the attractive Higgs bosons interact with particles, giving them resistance to forces that accelerate them.

In 2000, LEP saw tantalizing hints of a Higgs boson at about 115 GeV, just the mass theoreticians were expecting it to have (Science, 22 September 2000, p. 2014). But because LEP was due to be dismantled so that LHC could be built in the same tunnel, researchers were not able to prove whether the Higgs “sightings” were real or just a statistical fluke. Even if the CERN results are wrong, most physicists feel that the Higgs is within reach and LHC will easily have the energy to find it.

The second major hole in the Standard Model lies in the “unification” of three of nature's forces: the strong and weak nuclear forces and the electromagnetic force. At very high energies and very small scales, these three forces get more and more similar—almost becoming the same thing. “The three run to a common strength, but the convergence is not perfect,” says David Burke of the Stanford Linear Accelerator Center in California. Theorists have long struggled to close this gap, and the theory that most now favor is known as supersymmetry, where every known particle has a doppelgänger. For example, the quark is twinned with the squark, the electron with the selectron, and the photon with the photino (see table, p. 1171). These partners are heavy and tend to be short-lived, so we don't see them, but they double the number of fundamental particles in nature and gently alter the predictions of the Standard Model. “If you add supersymmetry, you can make [unification] almost perfect,” says Burke.

There's another compelling reason to believe that the universe is swarming with as-yet-undiscovered supersymmetric particles. Cosmological observations, such as measurements of the cosmic background radiation, the distribution of galaxies, and the speed of distant supernovae, all provide an estimate of how much matter there is in the universe. All point to the fact that there is a lot of matter that is totally unaccounted for. About 85% of the mass in the universe cannot, for various reasons, be made up of any of the particles known to the Standard Model. The lightest supersymmetric particle is a prime candidate for this “exotic” dark matter. “It's quite possible that this dark matter will be made up of lots of copies of the lightest supersymmetric particle,” says MIT's Wilczek. “This makes the particle and all of supersymmetry's projections extremely interesting.”

The Higgs boson and the lightest supersymmetric particle are expected to be quite massive, compared with the well-known particles of the Standard Model, but physicists are confident that they are not so massive as to be out of reach of LHC and the linear collider. If they were, they would no longer plug the holes in the Standard Model they were meant to plug, nor would the theories behind them agree with existing observations.

## Worlds in collision

So, come 2007 or soon afterward, particle physicists are confident that they can pick through the debris of LHC's collisions and find both Higgs bosons and supersymmetric particles. But that is far from the end of the story. “Protons are complicated objects,” says Wilczek. They are full of quarks, gluons, and virtual particles that pop into and out of existence for very brief periods, thanks to Heisenberg's uncertainty principle. So the results of their collisions are very messy, and the analysis is quite difficult. U.S. physicist Richard Feynman likened the process to smashing two Swiss watches together to figure out how they work. “Analyzing what you get is extremely complicated,” says Wilczek.

That is where the linear collider comes in. Unlike LHC and its protons, the proposed linear collider will collide electrons with antielectrons, also known as positrons, both fundamental particles. Whereas colliding protons is like smashing two bean bags together, slamming an electron into a positron is more like firing a bean at another bean. Furthermore, when a fundamental particle collides with its antiparticle, it annihilates in a burst of pure energy. The entire spray of particles produced has the same progenitor: the energy of the collision itself. That makes an electron-positron collision a much “cleaner” environment than a proton collision, allowing physicists to glean much more precise information. But because electrons and positrons are much lighter than protons, the energies that an electron-positron collider reaches are not nearly as great as those reached by a proton-proton collider. “You trade off energy for clarity,” says Wilczek.

As a result, all particle physicists agree that the fine control of a linear collider will be a vital complement to the brute force of LHC. “The LHC instantly gives you a very high-energy reach, and with an electron-positron collider you get a very complete picture, greater than the sum of the two individual parts,” says Burke. “The fact is, you can't get all the information from one machine.” All that remains for physicists to do is persuade their governments that that clearer vision is worth several billion dollars.

10. NEXT LINEAR COLLIDER

# One Collider, Many Countries: How to Share the Wealth?

1. Daniel Clery

Scattering control rooms around the globe might give everybody a piece of the particle-smashing action—if physicists can handle it

It's 2014, and you're in the control room of the World Linear Collider (WLC)—the spanking-new particle smasher built by a collaboration of labs around the world to study the Higgs boson and various supersymmetric particles discovered by the Large Hadron Collider at CERN near Geneva. Late-afternoon light slants through the windows: It's shift-change time, and the half-dozen physicists and engineers in the room are viewing computer screens and making last-minute checks on the accelerator.

Outside, there is no sign of the 35-kilometer-long monster they are controlling, just office buildings and the wooded hills west of San Francisco Bay. No relief crew has arrived—nor will it. Instead, a video screen covering an entire wall shows an almost identical control room, shafts of bright morning light streaming through vertical blinds, in which half a dozen Japanese physicists and engineers are busily consulting screens while sipping their first coffees. Elsewhere on the screen, other images are arrayed: a third control room, empty, with lights dimmed; the spokesperson of the team whose experiment is currently running on WLC, dozing in her office; a table of numbers describing the state of the beam and the health of the collider. As the California team prepares to hand off, the number 10 appears in large figures in the center of the video wall. “Counting down,” the leader calls out. Everyone tenses slightly as the numbers tick away down to zero. Several screens go blank. “She's all yours, Setsuko,” says the U.S. leader as his team prepares to head home.

The technology to control such a huge and complex instrument from virtually anywhere in the world already exists. Even in 2014, however, nobody is likely to use it merely to avoid night shifts. The real rationale behind the scenario is more subtle: There will be only one WLC, or whatever it is eventually called, so accelerator physicists can't all live and work nearby. They will have to find some other way to ensure that everyone gets a piece of the action. Their solution is the Global Accelerator Network (GAN).

Albrecht Wagner, director of DESY, Germany's particle physics lab near Hamburg, came up with the idea of GAN in the late 1990s to run an entire accelerator remotely. The world's five big accelerator labs—DESY and CERN in Europe, Fermi National Accelerator Laboratory and Stanford Linear Accelerator Center in the United States, and KEK in Japan—were developing several rival designs for a linear collider. “They can't all have frontline machines, but we want to keep their accumulated know-how alive,” Wagner says.

Wagner found the seeds of a solution in his own lab. Particle physics detectors are huge machines, several stories high, weighing hundreds of tons and packed with cutting-edge electronics. Each is built by a group of researchers, usually many hundreds strong, from dozens of universities across the globe. If a worldwide team can build something as complex as a particle detector, Wagner argued, why not the accelerator itself? In March 2000, intrigued by the idea, the International Committee for Future Accelerators (ICFA) set up a panel to see if it would work. The answer, delivered at the end of 2001, was a qualified yes: Technically, it's doable, but getting such a huge team, scattered across the globe, to work together efficiently is another matter.

Today's accelerators are already largely controlled by dedicated computers on site. It makes no difference to the control computers whether they get commands from a building next door or from another continent. In principle, with enough bandwidth and enough diagnostic sensors to warn when something goes wrong, experts could run—and repair—an accelerator from anywhere in the world. The ICFA panel concluded that, compared with the cost of the hardware itself, such remote diagnostics won't break the bank. Nor will multiple control rooms, says ICFA panel member Ferdinand Willeke of DESY: “You just need 10 PCs [personal computers] and no specialized equipment.”

The ICFA panel estimated that a skeleton crew of about 200 would be enough to tend the collider. In existing accelerators, “breakdowns” are often remedied by technicians on the spot, sometimes with a phone call to an expert accelerator physicist. With efficient remote diagnosis sensors, the panel reckoned, there will be little need for experts on site. Only about 20 times a year will an expert need to fly in to fix something.

But what about the tricky politics of getting these labs—erstwhile rivals—to collaborate? Psychologist Gary Olson of the University of Michigan, Ann Arbor, who advised the ICFA panel, says that first you should ask if the partners are ready to work together. High-energy physics has a long history of collaboration, so Olson says he is optimistic about GAN. But physicists are “kind of behind” in adopting the technologies available to aid collaborative work, he says, such as online blackboards for sharing data, virtual “chat rooms,” and videoconferencing. In addition, “trust is a big issue. It will take a lot of work to build up trust and confidence,” Olson says.

When the ICFA report was presented at conferences last year, Willeke says it was enthusiastically received. He and colleagues at DESY and in the United States are now planning experiments to test the idea. Meanwhile, a new set of panels, one in each region, will look at all organizational aspects of setting up the linear collider. Deciding how to share the collider may be one of the easier problems on the agenda. “No one knows how to get such an international project going,” says George Kalmus of Britain's Rutherford Appleton Laboratory, who heads the European panel. Physicists may be good at pushing back the frontiers of knowledge, but getting world governments to back such a machine is something new for them. “If we can do that,” says Kalmus, “[it] would be an achievement.”

11. BIOTERRORISM

# Security Rules Leave Labs Wanting More Guidance

1. David Malakoff

Scientists criticize flaws in the U.S. government's plan to oversee research on material that could be used as bioweapons

The U.S. government's effort to regulate bioterror research appears to be off to a rocky start. Key elements of the sweeping new security rules are vague, confusing, and possibly counterproductive, say scientists and university administrators in comments* filed by last week's deadline. Research leaders are also frustrated that the government has yet to explain how it will conduct mandatory background checks, due to begin next month.

The rules “require a large number of activities in a short period,” notes the American Society for Microbiology (ASM), one of 100 groups, institutions, and scientists that have offered suggestions to the Centers for Disease Control and Prevention (CDC) and the U.S. Department of Agriculture (USDA). ASM warns that the process of complying may result in “delaying and possibly discouraging research.” Federal officials have promised to finalize the procedures in time for laboratories to meet a November deadline.

The new rules stem from the 2001 anthrax letter attacks, which killed five people and disrupted mail service for months. Last summer, Congress passed a bioterror bill that set a brisk timetable for improving security at laboratories working with nearly 100 “select agents”—viruses, bacteria, and toxins that could imperil people, farm animals, and crops. In December, CDC and USDA released nearly 50 pages of preliminary rules, covering everything from new locks on lab doors to government approval for certain genetic-engineering experiments (Science, 20 December 2002, p. 2304). The comment period ended last week; the rules, which also specify hefty fines and jail time for violators, started to take effect on 7 February.

Although researchers and biosafety experts give the two agencies credit for quickly cobbling together their plans, they say that plenty of fine-tuning is needed. For lab administrators, one of the most troublesome issues is who must undergo background checks. Agency officials estimate that up to 20,000 researchers at nearly 1000 facilities will need to be screened under rules that require prior government approval for anyone to have “access” to a select agent lab. But researchers working on a wide range of experiments often share space and expensive equipment, “with only a few actually handling infectious agents,” noted lab safety expert Emmett Barkley of the Howard Hughes Medical Institute (HHMI) in Chevy Chase, Maryland.

THE HOT LIST A Sampling Of Suggestions For Remaking The Government's Select Agents List

Prions (kuru, Creutzfeldt-Jakob)

Mistletoe lectin (similar to ricin)

All select agent genes and fragments

Don't regulate:

Cercopithecine herpesvirus 1 (CHV-1)

Plum pox potyvirus

Yersinia pestis Pgm and Lcr strains

Brucella spp. strains

Coccidioides

Shiga toxin from wild E. coli (pictured below)

Instead, HHMI and many universities recommend that the government allow laboratories to develop systems to limit access to select agents—by keeping them in locked freezers, for instance—without barring all unscreened researchers from the surrounding space. “Isolating scientists who handle infectious agents will be detrimental,” Barkley predicts. Isolating materials could greatly reduce the number of people needing background checks, noted David Zeman, head of veterinary science at South Dakota State University in Brookings. “In our lab,” he writes, “this would typically mean background checking about six people,” instead of 80.

ASM and some researchers are also skeptical of a CDC plan to require scientists to get approval from the Secretary of Health and Human Services for genetic-engineering experiments that might make a select agent more toxic or more resistant to known drugs. Government-funded scientists are already subject to that restriction under National Institutes of Health guidelines, which require the agency's Recombinant DNA Advisory Committee (RAC) to approve such experiments. But converting the RAC guidelines to regulations could produce inflexible rules that lag behind current science, ASM argues.

Critics say the rules are also vague about who will decide what constitutes a dangerous experiment. U.S. National Academy of Sciences chief Bruce Alberts suggests that agencies delay any determinations until after a report this spring from an academy panel. ASM recommends that the job be done by a new, high-level review panel stocked with microbiologists and biosafety experts.

That panel also could help sort out the increasing confusion over which agents—especially those that come in both lethal and nonlethal varieties—are covered by the new rules, says ASM. And it might handle appeals to add or subtract agents from the current list. Some scientists, for example, say the current rules appear to regulate some harmless, widespread strains used in vaccines.

Other researchers want the government to rethink how it will regulate diagnostic and clinical laboratories, which test thousands of tissue samples a day for pathogens. Under the interim rules, workers in these labs face fewer screening requirements. That's “sheer lunacy,” says W. Robert Newberry, a safety officer at Clemson University in South Carolina, because technicians are often in the best position to divert a select agent for weapons use. And requiring these labs to destroy samples containing select agents within 7 days could hamper future scientific and criminal investigations, says virologist Scott Weaver of the University of Texas Medical Branch in Galveston. “Our future ability to identify the source of a terrorist introduction [depends] on having collections of reference agents,” he wrote.

Several other proposals seem counterproductive, scientists noted. One would require all packages entering and leaving laboratories to be inspected, most likely by guards untrained in handling biological samples. Another appears to force labs holding select agents to post signs advertising that fact.

Agency officials have not said when they will issue a final set of rules. And the Department of Justice remains silent on what kinds of information laboratories must provide on researchers whom Justice wants screened, a process that is supposed to begin on 12 March. The government “has the obligation to move with the same degree of effort that the scientific community has shown in responding to these proposals,” says Barkley.

12. WEST NILE VIRUS

# Researchers Scramble to Track Virus's Impact on Wildlife

1. David Malakoff

As mosquito season approaches, researchers strategize about how to stalk West Nile virus and safeguard vulnerable species

EDGEWATER, MARYLAND—Two decades ago, California condors, ancient scavengers with mighty 3-meter wingspans, numbered just two dozen. Now there are nearly 200, thanks largely to breeding programs at several California zoos. But the condor now faces a threat that could undo those gains: West Nile virus.

Since arriving on North America's eastern coast in 1999, the mosquito-borne virus has swept west, killing hundreds of people—and hundreds of thousands of wild and captive birds (Science, 20 September 2002, p. 1989). Now, with the virus expected to hit California in full force this year, condor conservationists have taken an extraordinary step: They recently immunized several dozen of their precious captive birds with a new, experimental vaccine.

Saving endangered species was just one of the issues on the minds of the 100 researchers who gathered here earlier this month to develop a blueprint for studying the West Nile virus's impact on wildlife. Over the course of 2 days, they got an update on the virus's impacts so far, and they identified dozens of questions that must be answered before scientists can predict where West Nile virus might strike next and how best to stop it. “It's a fascinating and unpredictable virus,” says ornithologist Peter Marra of the Smithsonian Environmental Research Center, which hosted the summit. “But so far, there are many questions and not a lot of answers.”

There also isn't a lot of money: Some participants released a letter this week that admonishes the U.S. government for lackluster backing of West Nile science that could help public health officials better combat the threat.

At the meeting, researchers agreed that they do know one thing: West Nile has spread farther and faster than most experts expected. Scientists have found the virus or antibodies to it in 157 species of birds, 37 kinds of mosquitoes, and 18 other vertebrates, ranging from horses to alligators, noted Duane Gubler of the Centers for Disease Control and Prevention (CDC) in Fort Collins, Colorado. In just 4 years, it has spread to 6 of the 13 Canadian provinces and territories and 44 of the lower 48 U.S. states; the virus also appears to have penetrated northern Mexico.

But little is known about which kinds of wild animals are most susceptible to the virus, researchers said. In laboratory experiments, crows and their kin appear to be the hardest-hit birds, reported CDC's Nicholas Komar, who has infected members of more than two dozen species. Some crows died within just 4 or 5 days. Among more resistant species, survivors often carried the virus for weeks, suggesting that they could be important reservoirs that hold West Nile and move it across the landscape. Many researchers have speculated that migrating birds are responsible for the West Nile virus's speedy spread—perhaps moving it in a zigzag, north-south pattern during their annual flights.

So far, however, researchers have yet to find a migrating bird carrying live virus, notes Marra—but not for lack of trying. Marra is part of a team that has collected more than 8000 blood samples from 174 species in the mid-Atlantic states, Florida, the Caribbean, and Mexico. Other researchers are also monitoring migrants in their tropical wintering grounds, on the assumption that birds will move the virus south (Science, 7 February, p. 821).

Many samples are still waiting to be tested, however, highlighting another issue discussed at the meeting: a shortage of testing facilities. The few state and federal government laboratories that specialize in testing wildlife have been overwhelmed with corpses of crows, hawks, and other birds, Gubler noted. Overall, CDC says that health officials have reported nearly 125,000 dead birds, of which about 32,000 were tested for West Nile, with half showing signs of the virus. That source of data is drying up, however, because health departments often stop testing birds once they know West Nile has arrived in their area.

Existing numbers give biologists little sense of whether the virus is having a significant impact on bird populations. Researchers reviewing long-term survey data—such as annual winter and spring counts conducted by volunteers—have spotted temporary declines in some populations, but the dips are patchy and inconclusive, said John Sauer of the U.S. Geological Survey's Patuxent Wildlife Research Center in Laurel, Maryland.

One ornithologist, however, was able to track the virus's path through a population of 1100 crows in New York state that he has studied since 1988. In 2001, the virus had little apparent impact, noted Kevin McGowan of Cornell University in Ithaca. But last year, nearly one-third of some flocks died. Because many of his birds had been banded and followed for years, McGowan was able to study whether the virus hit certain demographic groups (young birds, for instance) harder than others. He also looked to see if the fatalities ran in families, perhaps because birds that lived and fed together were passing on the virus. But no clear pattern emerged, he says. He's readying for the possibility of another onslaught this year. “We crow researchers have had West Nile thrust upon us; we don't have much choice,” he says.

Other researchers are also preparing for the West Nile season, which they expect to begin when warmer weather arrives in the southeastern U.S. and coaxes mosquitos out of their winter lairs. To help guide their work, a group led by Marra and workshop co-organizer Robert McLean of the U.S. Department of Agriculture's National Wildlife Research Center in Fort Collins, Colorado, is drafting a research agenda. Among other actions, it proposes to draw on a multifaceted surveillance system—using everything from public health offices to wildlife hospitals and zoos—to track the virus and perhaps nail down how it spreads and where it hides during the winter. State and federal budget woes could hamper such work, but scientists hope West Nile's nasty reputation will convince the Bush Administration and Congress to add funds. In the meantime, zoos and endangered-species biologists are nervously watching West Nile's march.

13. AMERICAN ASSOCIATION FOR THE ADVANCEMENT OF SCIENCE MEETING

# Gravity Waves Elude First Scrutiny

1. Charles Seife

DENVER, COLORADO—More than 4000 scientists and 1000 members of the press gathered here 13 to 18 February to discuss a broad array of research and policy. More coverage in next week's issue and on ScienceNOW (www.sciencenow.org)

The physicists of the Laser Interferometer Gravitational-Wave Observatory (LIGO) are searching desperately for signs that the globe is being squashed by an interstellar traveler. They're looking for the telltale stretch-and-squish signature of gravitational waves distorting the fabric of spacetime—and Earth along with it. On Monday at the AAAS meeting, scientists presented the first results from LIGO. Although there is no sign yet of gravitational waves, LIGO scientists say the measurements so far at least suggest that the observatory is overcoming some troublesome teething pains.

The $365 million twin observatories—based in Washington state and Louisiana—beam lasers down 4-kilometer-long tubes to detect a tiny warping of space associated with the passing of a gravitational wave, a phenomenon predicted by Einstein's general theory of relativity. The lasers are at right angles to each other; a gravitational wave would likely shrink one and stretch the other, causing a faint flicker when the beams are brought together. Maintaining two facilities allows researchers to screen out local noise. Since LIGO began operating in October 2000, scientists have been struggling to reduce the amount of noise—such as the low murmur of seismic vibrations and the earth-rattling whine of loggers' saws in Louisiana —that shake the machine and prevent it from seeing the subtle signals of gravitational waves (Science, 16 August 2002, p. 1115). According to LIGO team member Rainer Weiss, a physicist at the Massachusetts Institute of Technology in Cambridge, these problems should be coming to an end soon. “Now, at the sweet spot, we're a factor of 10 away from our design sensitivity. Though we're not there yet, we're within shooting distance,” he says. Adds LIGO team member Al Lazzarini, a physicist at the California Institute of Technology in Pasadena So far, though, no gravitational waves have shown up in any of the project's first four results. The analyses concern different types of gravitational waves. The LIGO interferometers searched for a sudden burst of waves caused by the collapse of a dying star, for example, as well as for the increasingly high-frequency chirp of two massive stars spiraling into each other. In both cases, estimates Lazzarini, LIGO at its current sensitivity could see such an event in “the entire galaxy and our satellite galaxies, the Magellanic clouds.” Two other searches, for periodic sources analogous to electromagnetic pulsars and for a rumbling background of gravitational waves left over from the very early age of the universe, also came up short. But Lazzarini says that the measurements are beginning to surpass the best estimates previously set on the amount of gravitational radiation rattling around the cosmos. “It's still not astrophysically interesting,” he says, but he adds that the next run, scheduled to begin shortly, will start to look for gravitational waves with sensitivities well beyond those of other instruments. 14. AMERICAN ASSOCIATION FOR THE ADVANCEMENT OF SCIENCE MEETING # Subterranean Coal Fires Spark Disaster 1. Kathryn Brown DENVER, COLORADO—More than 4000 scientists and 1000 members of the press gathered here 13 to 18 February to discuss a broad array of research and policy. More coverage in next week's issue and on ScienceNOW (http://www.sciencenow.org/) By the time last summer's “coal seam” fire died in Glenwood Springs, Colorado, the damage was stunning. Researchers think that smoldering underground coal ignited tree roots—and in just weeks, the ensuing forest fire devoured 4900 hectares and caused$20 million of damage.

The same fierce blazes often hit coal-rich China, Indonesia, and India. But whereas the Colorado fire lasted weeks, these developing countries endure coal fires that burn for years—often polluting the air, erasing landscapes, and threatening human health. “Coal fires are relentless,” remarks Alfred E. Whitehouse, a project manager with the Office of Surface Mining in Washington, D.C. At the AAAS meeting, researchers detailed the global coal fire problem, as well as efforts to study and fight the blazes.

Geologists have dated the rubble left from coal fires—such as reddish “clinker,” baked shale covering burned coal seams—back 4 million years. Coal seams lace the earth. They can spontaneously ignite when oxygen and hot sunlight combine; then flames slowly devour the coal layer below ground. Mining adds to the combustible conditions—and researchers say devastating fires are on the rise.

Burning coal beds spew a brew of volatile chemicals, including arsenic, fluorine, mercury, and selenium. Few studies link these emissions with health problems, although rural Chinese residents have developed arsenic poisoning after cooking with coal. The greenhouse gas carbon dioxide escapes from burning coal, but emission estimates vary.

But the clearest danger is simply a hungry blaze. In 1997 and 1998, as drought gripped Indonesia, hundreds of coal fires damaged roughly 5 million hectares in East Kalimantan. That includes 5000 hectares of the pristine Sungai Wain Nature Reserve. “There aren't that many orangutans and Sumatran tigers left,” Whitehouse says. “This is where they live.”

Last year, Whitehouse finished directing a 5-year, \$1.5 million project, funded by the U.S. Agency for International Development, to train Indonesians to fight coal fires. Whitehouse described working with Kalimantan residents to inventory and extinguish coal fires. Using bulldozers and backhoes, the team dug out fires, doused the burning material, and built trenches to separate smoldering areas from unburned coal. Their success inspired Indonesia's Ministry of Energy and Mineral Resources to draft a formal coal fire plan and budget in 2000.

Before fighting a coal fire, you have to find it—and when the blaze is deeply buried, that can be tricky. New remote-sensing tools offer a “quantum leap” in the effort, says Paul van Dijk of the International Institute for Geo-Information Science and Earth Observation in the Netherlands. Working in the Wuda coalfields of Inner Mongolia last fall, van Dijk and his colleagues collected thermal data from the U.S. and Japanese ASTER satellite to pinpoint hot spots likely to be underground fires.