# News this Week

Science  17 Nov 2000:
Vol. 290, Issue 5495, pp. 1270
1. SOLAR SYSTEM EXPLORATION

# Push to Revive Pluto Mission May Mean Competition for JPL

1. Andrew Lawler

For a planet that some astronomers argue is too small even to be called one, Pluto is asserting a surprisingly strong pull on Earth. An unusual coalition of scientists, activists, and politicians is pressuring NASA to rethink a September decision to put a 2004 mission on hold because of budget constraints. The growing clamor is shaking up the planetary science community, which is also preparing for a mission at mid-decade to Europa, a moon of Jupiter. The biggest impact may be felt at the Jet Propulsion Laboratory (JPL) in Pasadena, California, which could face serious competition for the first time in decades on contracts to build planetary missions.

NASA managers have paired the Pluto and Europa missions, although they are separate undertakings. JPL estimated a few years ago that it would cost about $650 million to launch both missions, but a more conservative approach in the wake of two Mars failures has sent the cost soaring to as high as$1.4 billion. That figure prompted NASA space science chief Ed Weiler to put the freeze on Pluto (Science, 22 September, p. 2018).

Now the agency may ask outside organizations to come up with cheaper and faster alternatives for sending a probe to Pluto. This week a NASA scientific advisory panel backed the idea of an open competition if JPL can't come up with a better plan by the end of the month. The panel also urged the space agency to delay work on the Europa mission, despite its glamour, if that would free up money for Pluto.

Researchers and White House staffers are excited about the proposal to search for signs of life in oceans that may lie underneath Europa's icy crust. But Pluto's advocates say that delays could mean the forfeiture of important data. A 2004 Pluto launch would take advantage of Jupiter's transitory position to rocket the spacecraft and its plutonium generator to its destination in 8 to 12 years; any delay would push the arrival time at Pluto beyond 2020. By that time, researchers fear that the planet's atmosphere, which cools as Pluto recedes from the sun, may have frozen and collapsed, burying clues about the origin of the solar system. “If Pluto waits, science will be lost,” says Alan Stern, an astrophysicist at the Southwest Research Institute in Boulder, Colorado. “If Europa waits an extra 18 months, no science will be lost.”

NASA managers say they would consider an open competition for the Pluto shot but that delaying Europa is more problematic. “We like the idea of competition,” says Jay Bergstralh, acting science chief of NASA's solar system exploration division, noting that a request for proposals could go out as early as January. But delaying the Europa mission—tentatively slated for a 2006 launch—would require White House approval. And given the current uncertainty over who will be in the White House, Bergstralh adds, “the timing couldn't be worse.”

The open competition and the Europa delay were proposed in a letter due out this week from the agency's solar system exploration advisory panel, chaired by planetary scientist Michael Drake of the University of Arizona in Tucson. “If polled, the science community likely would agree that the Europa mission is more important,” says Drake. But “most people also like the idea of going to the edge of the solar system.”

Among those who particularly like the idea are scientists and engineers at three organizations eager to prove that they have the right stuff to build potentially lucrative and prestigious planetary missions. Managers at Johns Hopkins's Applied Physics Lab (APL) in Laurel, Maryland; Lockheed Martin of Bethesda, Maryland; and Ball Aerospace of Boulder, Colorado, all say they have plans to bid on a Pluto mission. Each organization has a part in NASA's Discovery program, which provides experience with small planetary missions, and each would like a slice of JPL's much larger planetary pie. Also, each plans to use the spacecraft structure developed for Discovery missions as the basis for their Pluto entries. “Don't underestimate the innovative ideas which could arise out of competition,” says Stamatios Krimigis, APL's space science chief.

But given the constraints of orbital mechanics, launch vehicles, propulsion, and electrical systems, JPL managers say they are skeptical that any dramatic new cost cuts will emerge from the competition. “We've looked at everything conceivable—even every wacky idea,” says Richard Terrile, JPL's chief scientist for the outer solar system. “I don't think you will see [budget] numbers very different from ours.” But he adds that bringing a contractor on board could increase the program's political clout. “Industry can bring in more congressional districts and get more political support than we can, and that's a good thing.”

Weiler's decision to put Pluto on hold in September triggered a campaign by scientists and members of Pasadena's Planetary Society to boost congressional interest in the once-obscure mission. It appears to be having the desired effect: On 28 October, Representative James Walsh (R-NY), who chairs the House panel that approves NASA's budget, noted that “Pluto remains the only planet in the solar system” unvisited by spacecraft. He also said that the mission would provide data on the small bodies beyond Pluto that populate the Kuiper belt. Walsh has asked NASA for an explanation for the announced delay to “clear up any confusion on the part of our members.”

That pressure has put NASA managers in a delicate position. Agency officials say they would like to keep Pluto on track despite rising costs and an uncertain budget. But they don't want to get into a fight with the White House over the Europa mission. Proponents hope to tip the balance in favor of the outermost planet without upsetting those more focused on Europa. “NASA is going to see it will be greatly to their advantage to restart [Pluto] expeditiously,” says Stern. “They missed a bet—this mission is a keeper.”

2. NEUROBIOLOGY

# Heretical View of Visual Development

1. Ingrid Wickelgren

In the 1960s, two Harvard neurobiologists turned thinking about brain development on its head, showing that experience itself could alter the structure of the brain as it continues to mature after birth. David Hubel and Torsten Wiesel found in monkeys and cats that seeing out of both eyes is necessary for the normal arrangement of so-called ocular dominance columns—neat columns of brain cells that respond to visual activity from one eye or the other. This work led to the deeply rooted belief that the columns form as a result of visual activity—a belief now being called into question.

In work described on page 1321, Duke University neuroscientists Justin Crowley and Lawrence Katz report that ocular dominance columns in ferrets appear long before the columns can be modified by visual experience. Even more heretical, they present data suggesting that neural activity from the eyes has little influence over the formation of the columns. They propose instead that innate molecules that guide growing axons to their locations in the developing brain may be primarily responsible for building these columns.

“It's a very beautiful study,” says Harvard neuroscientist Carla Shatz. “There's no question that they showed that ocular dominance columns form earlier than we thought.” But Shatz and others contest the conclusion that neural activity is not required for constructing the columns, arguing that there are other explanations for the researchers' findings. “I'm agnostic about that [conclusion],” Shatz adds.

If the Duke team's conclusions are correct, however, they may once again drastically alter the way scientists view how the brain matures. What's more, if the purported molecular signals that shape sophisticated visual structures can be found, they may provide a new handle on the cause of myriad visual or other brain-related birth defects that now go unexplained.

But those are big “ifs.” For years, research confirmed and extended Hubel and Wiesel's findings. In 1978, three of their Harvard protégés, Simon Levay, Michael Stryker, and Shatz, showed that in cats, the ocular dominance columns don't appear in the visual cortex, a region near the back of the brain, until 3 weeks after birth. That coincides exactly with the “critical period” in the animals, the time when Hubel and Wiesel had found that shutting one eye could disrupt the structure of the columns. So the scientists concluded that visual stimulation from both eyes is essential for the columns' initial formation.

But later work cast doubt on this analysis. Biologists showed several years ago, for example, that the columns are present at birth in monkeys and so must form before the critical period when they could be modified by visual experience. Still, researchers refused to retreat entirely from the idea that neural activity is essential, arguing that the spontaneous firing of neurons from the retina before the eyes open shapes the columns.

Three years ago, Crowley and Katz set out to test this hypothesis in ferrets. These animals were ideal because their nervous systems at birth are 3 weeks less developed than those of cats, enabling the detection of earlier developmental events. In work published last year, the researchers removed both eyes from newborn ferrets in an attempt to cut off any neural activity from the eyes that might influence the development of the columns.

Next the researchers wanted to find out whether that operation had in fact blocked column formation. When the animals reached adulthood, the researchers injected a tracer into eye-specific cells in the lateral geniculate nucleus (LGN) of the brain, through which visual signals travel en route to the cortex. Once the tracer traveled to the cortex, it revealed neatly ordered ocular dominance columns. So, says Crowley, “we concluded that retinal activity could not be as important as we thought it was” for forming the columns.

But still, questions remained. Crowley and Katz wondered whether they had removed the eyes early enough—that is, before the columns had formed. So in the current study, they used their tracer technique to time the formation of the columns. They found that the columns were not present at birth but appeared as early as 16 days later, which is equivalent to a week before birth in cats. “In pushing the day back,” says Stryker, now at the University of California, San Francisco, “it becomes even clearer that the formation of the ocular dominance columns precedes the onset of plasticity” in the critical period.

But even though lack of input from both eyes hadn't affected the columns, skeptics argued that it might take an imbalance in input—for instance, having signals from one eye but not the other—to disrupt ocular dominance formation, as it does later during the critical period. To counter such claims, Katz and Crowley removed just one eye from each of six ferrets at an age when the LGN axons had linked up with the cortex but before the columns had formed. The doubters were wrong again: These ferrets still formed normal-looking ocular dominance columns.

Even that finding has not closed the case against neural activity, however. Shatz points out that the Duke scientists did not check to see whether the columns they saw after they removed one eye alternated in responding to either the right or left eye as they should. Thus, it could be—as Shatz has found in the past in cats—that removing one eye altered the visual system such that seemingly normal ocular dominance columns formed, but they all responded to the remaining eye. And Stryker points out that the study did not rule out the very real possibility that activity in the brain—say, from the LGN—is necessary for column formation even if that from the eye isn't.

Katz concedes that these arguments are valid, although he doesn't believe they explain his findings. He's now aiming to identify the molecular signals that he thinks guide the formation of ocular dominance columns. Says Katz: “My gut feeling is that we need to look in a different direction.”

3. NEUROSCIENCE

# An Antibiotic to Treat Alzheimer's?

1. Laura Helmuth

NEW ORLEANS—An antibiotic once used to treat traveler's diarrhea might battle Alzheimer's disease as well, researchers announced here last week at the Society for Neuroscience's annual meeting. The drug dissolves Alzheimer's-like plaques in mouse brains, apparently by trapping the copper and zinc that stud these deposits. A clinical trial to test whether the drug helps people with Alzheimer's is already under way.

Deposits called amyloid plaques riddle the brains of people with the disease—who now number about 4 million in the United States and may reach 14 million by 2050. The plaques' primary ingredient is a protein called β amyloid (Aβ) that occurs normally in the body and doesn't appear dangerous in its soluble form. “We're not trying to stop [Aβ] from being produced—we're trying to stop it from being accumulated,” says Ashley Bush of Massachusetts General Hospital in Boston.

To do so, Bush has been studying the effects of copper and zinc on Aβ. About a decade ago, he found that these metals allow enzymes to cut Aβ from a larger protein; he later discovered that they influence the properties of Aβ itself. Adding zinc to the protein in a test tube, for instance, can cause clumps of Aβ to form, and compounds that bind to copper and zinc can dissolve amyloid plaques in postmortem brain tissue from people who died with Alzheimer's disease.

In search of a candidate drug, Bush's team screened dozens of antibiotics and anti-inflammatory drugs known to bind metals. An antibiotic called clioquinol proved the most potent, efficiently dissolving plaques in postmortem tissues and reducing amyloid's ability to clump together. To test whether clioquinol could clear up amyloid plaques in the brain of a living animal, the team gave the drug to young mice engineered to develop Alzheimer's-like deposits. The drug appeared to inhibit plaques from forming: The animals developed fewer plaques overall, and some 30% developed no detectable plaques. In a second study, the drug appeared to clear up plaques in mice old enough to have developed substantial deposits. Those given the drug for 9 weeks had 50% less amyloid deposited in their brains than untreated animals had.

The mouse studies are “very impressive,” says Alzheimer's researcher David Morgan of the University of South Florida in Tampa, showing “a dramatic reduction [of plaques] in a very short time.” Other approaches, including vaccinating mice against Aβ, also clear plaques from mouse brains (Science, 9 July 1999, p. 175).

The Food and Drug Administration approved clioquinol decades ago as an antibiotic. The drug was used for about 500 million patient days, Bush says, but was pulled from the market after a few people developed an acute vitamin B-12 deficiency while taking the drug. Although the B-12 deficiency can be addressed with supplements, Morgan cautions that the drug has not been tested for the long-term use necessary to help people with Alzheimer's.

A phase II clinical trial of the drug in people with mild or moderate Alzheimer's, coordinated by Colin Masters of the University of Melbourne in Australia with technical assistance from Bush, should start to answer questions about both the safety of long-term use and, perhaps, the drug's efficacy. The team expects to have the results from the trial in about a year.

4. PARTICLE PHYSICS

# The Final Tally Leaves LEP a Probable Loser

1. Oliver Morton*
1. Oliver Morton is a writer in Greenwich, England.

LONDON—The future of research at CERN, the European particle physics laboratory near Geneva, is proving only slightly easier to resolve than the identity of the next occupant of the White House.

Due to be dismantled in September, the lab's Large Electron-Positron (LEP) collider was granted a 1-month reprieve to continue experiments suggesting it had found evidence for the Higgs boson, a theoretical particle that physicists have coveted for decades. Last week LEP scientists presented the new data to CERN's management, saying the case for the Higgs was strong enough to merit another, longer reprieve. But management didn't budge. On 8 November, CERN's director-general announced that LEP would be scrapped to make room in the 27-kilometer-long tunnel to begin installing the Large Hadron Collider (LHC).

With LHC waiting in the wings, LEP scientists this summer had pushed the machine's beam energies to the max. On 2 November, after a month of extra operation, the team crammed into CERN's auditorium to review the latest evidence. The scene was not unlike a recount. One event once seen as evidence for the Higgs was withdrawn, while another event was added. Evidence gathered last summer was recalculated and revealed to be less significant than thought. But the new evidence was clearly stronger. Summing up the case for the Higgs, Peter Igo-Kemenes drew loud laughter and prolonged applause when he dryly remarked that “all this is very exciting.” The four experimental teams jointly called for another year's operation to settle the matter.

But the elation evaporated at a private meeting straight after of the committee of scientists from CERN and elsewhere who oversee LEP. “It was subdued,” one committee member told Science. “Nobody banged on the table. And that reflects the data, which is 50-50.” Although models gave the combined results only about a 1% probability of being due to chance, the LEP committee knew that such statistics do not tell the whole story (Science, 29 September, p. 2260). The signal was getting stronger with time, the panel agreed, but there was no guarantee that the trend would continue.

The committee did agree that spending $62 million to keep LEP running for another year was worth it. But not at the cost of delaying the LHC by a year—a decision with financial and political ramifications, because the LHC program includes partner countries with no stake in LEP. Thus the committee split on LEP's future. This came as a blow to rank-and-file LEP scientists, who argued that the LEP committee should have focused on the machine's scientific potential and disregarded the impact on LHC. CERN's senior management—the directorate—then took up the issue. A consensus favored moving ahead with LHC, but the decision was not final. Four days later, the lab's research board, of which the directorate forms a subset, was convened. According to LEP experimentalist David Miller of University College, London, one of two outside scientists on the board, this panel, too, was split. CERN's management team opposed an extension, whereas many of those representing the lab's various experiment committees and specialist divisions were in favor. “It was a painstaking meeting,” Miller says, in which every aspect was scrutinized. “In no way was the directorate trying to rush through its own ideas.” But in the end the research board remained split, leaving the decision in the hands of director-general Luciano Maiani. George Kalmus, chair of the Science Policy Committee that advises CERN's ultimate masters, the council of delegates of the member states, told the director-general that an e-mail poll of his top-level committee revealed a small majority in favor of shutting LEP down. And that is what Maiani decided to do. Originally, the director-general left the final ratification of that decision to a council meeting scheduled for December. Then, perhaps to head off further discussion, a special committee of the council was called for 17 November, presumably to sign the death warrant. In a very unusual move, the executive committee of CERN's staff association spoke out against the decision, expressing “astonishment and incomprehension” and saying that the case against LEP had not been made clearly enough. So although LEP looks dead, it's not over till it's over—as they say in Palm Beach County, Florida. 5. GEOPHYSICS # Core Takes a Page From the Sea Floor 1. Richard A. Kerr There's probably no stranger place in all Earth's layered interior than the intersection of the molten iron of the core with the bottom of the rocky mantle, 2900 kilometers down. This boundary produces the most dramatic effects on seismic waves traveling below the near-surface. Exactly what the core and mantle might be doing to each other to affect seismic waves so powerfully, however, has remained a mystery. Now one group of geophysicists is suggesting that the boundary most resembles an inverted sea floor, with liquid-iron-laced sediments collecting on the roof of the core. The case is made on page 1338 of this issue by geophysicist Bruce Buffett of the University of British Columbia in Vancouver, seismologist Edward Garnero of Arizona State University in Tempe, and mineral physicist Raymond Jeanloz of the University of California, Berkeley. They argue that a slow, inverted rain of precipitates rising to the core-mantle boundary and settling into a kilometers-thick layer might explain a variety of observations, from a subtle nodding of Earth's axis to seismic speed zones at the boundary. “I actually kind of like the idea,” says mineral physicist Quentin Williams of the University of California, Santa Cruz, adding that core sedimentation might explain at least some of the “very weird structure in the bottom 40 kilometers of the mantle.” But he's not completely sold: “It's a speculative story that's going to be very difficult to verify.” Difficulties seem inevitable, because painting a portrait of the core-mantle boundary depends on such indirect geophysical evidence. Williams and others believe mantle rock next to the hot core melts partially. Such melting, they say, could explain “ultralow-velocity zones”—thin patches of mantle at the boundary that slow seismic waves passing through them a whopping 10% to 30%, depending on the type of wave (Science, 31 January 1997, p. 614). But other, even more indirect probes hint at a different sort of boundary. A decade ago, Buffett analyzed the subtle nodding or nutation of Earth's rotation axis. From the magnitude of the annual nutation, they concluded that there is probably a thin layer of something at the boundary that magnetically connects the fluid core to the solid mantle. The power driving such a connection would come from the tidal forces of the moon and sun that shake Earth. That shaking sloshes the planet's fluid interior about, carrying the core's magnetic field through a presumed conducting layer at the boundary. That in turn would tug on the solid mantle and make Earth nutate ever so slightly. Buffett likens monitoring this sort of nutation to feeling a shaken jug of milk for signs of its contents. Geodecists Piravonu Mathews of the University of Madras in India, Thomas Herring of the Massachusetts Institute of Technology, and Buffett have analyzed a longer record of Earth nutation, soon to be published in the Journal of Geophysical Research, and “the nutation evidence seems to be getting stronger,” says Buffett. That implies there really is a layer conductive enough to link core and mantle magnetically, he says. How did a layer more electrically conductive than molten mantle come into being? Buffett, Garnero, and Jeanloz invoke inverted ocean-style sedimentation. The core has been cooling for billions of years, with pure iron crystallizing onto the solid inner core. Both the cooling and the crystallization should tend to drive impurities dissolved in the outer core out of solution, the team reasons, the way cooling and evaporation of a hot brine will precipitate salts. Being less dense than iron, the sandlike particles coming out of solution would rise to the core-mantle boundary the way particles in the ocean sink to the bottom. As the boundary sediment layer thickens, the pressure of the underlying sediment would slowly squeeze out most of the molten iron caught between particles. But taken together, the iron throughout the sediment pile would provide enough conductivity to explain the nutation results, the group calculates, as well as enough fluid to produce the observed slowing of seismic waves. Other geophysicists are intrigued but cautious. “It's a very plausible idea,” says David Stevenson of the California Institute of Technology in Pasadena. “It's also obviously a speculative thing.” One problem, he says, is that mineral physicists know so little about the behavior of materials at core temperatures and pressures. And telling what's going on from the surface will be difficult, although perhaps not impossible. Seismic techniques may help distinguish mantle from core sediment the way they can separate ocean sediments and crust, says Peter Olson of the Johns Hopkins University. The work could end up adding a layer to the “onion” model of Earth structure that now includes surficial sediments, crust, mantle, core, and inner core. 6. UNIVERSITY MANAGEMENT # Debate Flares Over Tracking Foreigners 1. Andrew Lawler Thousands of foreign students flock to U.S. campuses each year, providing critical intellectual muscle as well as revenue for science and engineering departments. But Congress, the Administration, and members of a national commission are worried that some of those students may be using their education as a cover for terrorism, leading to a proposal to watch them more closely. Academics have responded with alarm. They say the terrorist threat is overstated and that the surveillance proposal could turn into an administrative nightmare that could harm U.S. institutions. At the root of the controversy is a pilot program designed by several southeastern colleges and universities and the U.S. Immigration and Naturalization Service (INS). The effort, called the Coordinated Interagency Partnership Regulating International Students (CIPRIS), is an electronic system intended to replace the current patchwork of record-keeping on foreign students. In June, a panel of 10 national security experts recommended using the program to track when foreign students alter their workload or change their major field of study, among other things. It's that proposal, from the National Commission on Terrorism chartered by Congress, that has elicited an outcry from university officials. “We absolutely support electronic reporting” of foreign students, says Catheryn Cotten, who directs the international office at Duke University in Durham, North Carolina. But the increased monitoring, she adds, could force institutions like Duke into “a megareporting system … that is invasive at a level we can't manage.” Some 900,000 foreign students and scholars are studying in the United States, spending about$9 billion annually on tuition and other expenses. The CIPRIS program stems from a 1996 immigration law ordering the INS to set up an electronic tracking system to collect detailed information, such as current address and academic status. Under CIPRIS, universities would be required to let the INS know if a student dropped below a full course load, and why. INS began the pilot program with 21 colleges and universities, intending to expand it nationwide by the start of 2003.

The commission report warned that “a small minority may exploit student status” to conduct terrorist activities. To reduce that threat, it proposed that INS also gather data on when students change their majors—from, for example, English to nuclear physics. “We have to be careful not to lump all foreign students together,” says Yonah Alexander, a terrorist specialist at the Potomac Institute for Policy Studies in Arlington, Virginia, who spoke last month at a meeting on science, human rights, and national security hosted by the American Association for the Advancement of Science (AAAS, which publishes Science). “But the risk is real.” Commission member Richard Betts of Columbia University says that “it's not exactly paranoid” to suggest that modest efforts to gather more data would help authorities. “We're not talking about black FBI vans trailing students to their dorms,” he adds.

But most university officials say that increased scrutiny is unnecessary and could be extremely burdensome. The terrorism commission “vastly overstates the case,” says Victor Johnson, a policy expert at the Association of International Educators in Washington, D.C. “There is no evidence that foreign students pose a threat.”

The commission's proposal to track a student's major has also come under withering fire. “At the undergraduate level, that information is relatively useless” given the frequent changes many students make to their majors, says Cotten. And changes at the graduate school level may be less helpful than first appearances might indicate. “We've got ourselves into a Cold War warp” by focusing on nuclear weapons development, says Mary Good, dean of engineering at the University of Arkansas and AAAS president. A terrorist, she says, “can do just as much damage in computer science.”

CIPRIS would allow the INS to invalidate the visas of students who do not show up on campus or drop a full course load for unauthorized reasons. The INS and universities that participate in CIPRIS are negotiating the details and when the data would be collected.

Johnson and others are worried that a more vigorous effort may “turn colleges and universities into the eyes and ears of the federal government.” That change could make foreign students wary of choosing U.S. universities, he adds, at a time when “this is one of the few areas where we have a trade surplus.” Betts doubts that any new system would affect enrollments, but adds that “if it deters a few, that may be the unfortunate price we pay.” However, Good is less confident of a benign effect. “They've become a real part of the research establishment,” she says about the foreign students. “We can't lose that in this shuffle.”

7. TOXICOLOGY

# Arsenic Researchers Face Isotope Shortage

1. Jocelyn Kaiser

Molecular biologist Barry Rosen has been making good headway studying the mechanism by which arsenic causes cancer. But he's hit a roadblock. Although he's got funding from the National Institutes of Health, he can't buy the reagent that he needs—arsenic-73. The Department of Energy's (DOE's) Los Alamos National Laboratory in New Mexico is the only place in the world that sells the radioactive isotope, and they've run out. And it could be the middle of 2001 before Rosen, at Wayne State University in Detroit, and some two dozen other labs around the world are back in business.

To Rosen and others, the shortage comes at a particularly bad time. In May, the U.S. Environmental Protection Agency (EPA) proposed a costly clampdown on arsenic levels, a natural contaminant, in drinking water. EPA is also funding a burst of research on how arsenic causes cancer, because pinning down this elusive mechanism could reveal whether the limit needs to be so stringent. To understand the mechanism, researchers are using arsenic-73 to find genes that metabolize arsenic and to explore how these metabolites enter cells and damage DNA.

But Los Alamos, which makes the isotope by smashing protons from an accelerator into a rubidium bromide target, hasn't produced any arsenic-73 since the source of these protons—a tritium production program—ended in early 1999. A new isotope production facility was slated to open early next year, but the massive fires that swept through the region in the spring have pushed back the scheduled completion date to mid-2002. Los Alamos ran out of arsenic-73 inventory around July. “None of us knew about it until it was too late” to make other plans, says Marc Mass, an EPA toxicologist. There are alternative tracers, he says, but they're costly and too insensitive for some experiments.

Los Alamos officials say they can make arsenic-73 at another accelerator, then purify it at Los Alamos. But it took a flood of phone calls and letters from scientists to convince DOE officials to make it a priority, and it may be another 6 months before there's any arsenic-73 available, says Gene Peterson, manager of the lab's Isotope Production Program. The lab has to finagle beamtime on an accelerator—most likely one in Canada—prepare the target, he says, then cut through the red tape for transporting radioactive materials. Peterson says it's often not easy to predict shortages of radioisotopes used by only a handful of researchers, but that the new facility should make it easier to catch up because it won't depend directly on other experiments for protons.

That's little comfort to arsenic researchers, who are at their wits' end. Toxicologist Vas Aposhian of the University of Arizona, Tucson, who bought up the last few millicuries of arsenic-73 this summer, says, “We're going to scream bloody murder and call our congressman and senator” when the lab's supply runs out in a few weeks. Miroslav Styblo, a biochemist at the University of North Carolina, Chapel Hill, is trying to persuade colleagues at an accelerator in his native Prague to make a batch of arsenic-73. But “so far,” he says, “we don't have realistic promises.”

8. ELECTION 2000

# Uncertainty Wins by a Landslide

1. David Malakoff*
1. With reporting by Gretchen Vogel.

Scientists seeking order from the chaos of the U.S. election results may have a long wait. With the closest presidential election in a generation still undecided as Science went to press and the winners in several closely watched Senate and House races also unknown, it may be months before researchers and science lobbyists know who will fill influential posts in Congress and the new Administration for shaping science policies and budgets.

Amid the confusion, however, were a few results that election watchers could count on. They included the narrow defeat of Representative James Dickey (R-AR), a prominent opponent of government support for stem cell research, and the comfortable margin that returned Representative Vern Ehlers (R-MI), one of two physicists in the current Congress. Voters also approved several state ballot initiatives that will channel some funds from tobacco lawsuit settlements to research-related activities.

Those clear-cut decisions, however, were overshadowed by the chaos in the presidential race that has focused on Florida. That standoff was mirrored in New Jersey, where Democratic Representative Rush Holt—Congress's other physicist—was locked this week in legal wrangles with Republican Dick Zimmer over disputed ballots. On election night, Holt, a former official at the Princeton Plasma Physics Laboratory in New Jersey, appeared to have won by 56 votes, but later tallies prompted both candidates to claim a lead. “Rush likes to say that, as a scientist, he likes to get the facts,” campaign spokesperson Peter Yeager said as the controversy swirled this week. “So that's what we are doing, counting votes and trying to get the facts.”

If Holt loses, he would join stem cell opponent Dickey as one of the few incumbent House members to be defeated. Each year since 1995, Dickey has successfully pushed a controversial spending provision that prohibits federal funding for research that creates, harms, or destroys a human embryo. When the National Institutes of Health (NIH) earlier this year announced guidelines for use of human embryonic stem cells, which are derived from human embryos, Dickey threatened to sue if NIH actually awarded any grants (Science, 1 September, p. 1442). Dickey's defeat won't end that threat, however, notes Richard Doerflinger of the National Conference of Catholic Bishops in Washington, D.C. Returning lawmakers, including House majority whip Tom DeLay (R-TX), are likely to pick up where Dickey left off.

The election also heightened speculation about who will lead the committees involved in crafting science policy and budgets. In the House, James Sensenbrenner (R-WI) is attempting to jump from the chair of the Science Committee to the head of the higher profile Judiciary Committee. If Sensenbrenner is successful, Representative Sherwood Boehlert (R-NY), currently the science panel's second-ranking Republican, would likely inherit the top seat. Boehlert, an avid environmentalist whose district includes the U.S. Air Force's Research Laboratory in Rome, New York, has said he would bring “a lot of ideas” to the post, without offering specifics.

Another vacant seat is the chair of the House appropriations subcommittee that oversees NIH's spending. The post was held by retiring Representative John Porter (R-IL), a major backer of efforts to double NIH's budget. Observers say there is no clear replacement.

In the Senate, Porter's counterpart, Senator Arlen Specter (R-PA), hinted before the election that he may also abandon his spending panel post. If he does, his replacement is uncertain. A victory by Democrat Marcia Cantwell over two-term Republican Slade Gorton in Washington state—a race still too close to call as this issue went to press—could leave each party with 50 seats. If that happens, Democrats may demand to fill a number of committee leadership slots, although such power-sharing is unprecedented. With the statewide race likely to remain unsettled until late November, “it's getting agonizing just trying to figure out all the possible scenarios,” says one science lobbyist.

Legislators planned to return to Washington this week to try to finish off several spending bills for the current fiscal year—including funding for NIH—that were left hanging after protracted negotiations with the White House broke down a few weeks before the election. But the budget talks have now been pushed back until early December, and the uncertainty over the presidential results makes it impossible to predict how those negotiations will turn out and when Congress will adjourn.

Meanwhile, voters in five states—Arizona, Arkansas, Montana, Oklahoma, and Utah—approved ballot items that will allow officials to begin spending billions of dollars won in state lawsuits against tobacco companies. Although all call for spending some of the money on health care and antismoking efforts, only Arkansas's specifically mentions research, with about 60 million slated for several state-funded universities. 9. ASTROPHYSICS # Astronomers Spot Their First Carbon Bomb 1. Robert Irion HONOLULU—Astronomers love watching things blow up, but they've never seen a blast quite like the one described here last week. Carbon on the surface of an ultradense star detonated in a 3-hour thermonuclear explosion, according to a report at a meeting of the American Astronomical Society's High Energy Astrophysics Division. If confirmed, the burst would be the first known cosmic explosion fueled solely by carbon rather than hydrogen or helium. That prospect, says theorist Lars Bildsten of the University of California, Santa Barbara, is “very exciting from a nuclear physics standpoint” for its potential to verify or revise models of carbon combustion. The blast came from a waltzing pair of stars called a “low-mass x-ray binary.” In such a system, a dwarf star orbits closely around a neutron star, a stellar corpse that packs the mass of one or two suns into a dense ball just 20 kilometers wide. Gas from the dwarf flows into a hot spiraling disk around the neutron star. Some gas hits the star's surface, forming a compressed slurry of hydrogen, helium, and a few heavier elements. When pressures and temperatures get high enough within the thickening layer, the elements can flash-fuse in a thermonuclear explosion. Then, the layer rebuilds and the flash repeats after some interval, usually hours or days. This process continues indefinitely, although the timing changes drastically depending on the orbital dynamics of the two stars. Satellites see most explosions from such systems as mild x-ray flares that last 10 or 20 seconds. Last year, however, astronomers detected four flares that broke the mold. First, researchers at the Space Research Organization Netherlands (SRON) in Utrecht used the Dutch-Italian BeppoSAX satellite to find evidence for a long burst from each of three known low-mass x-ray binaries. The events lasted 500 times longer and unleashed 500 to 1000 times more energy than the generic short pops. “It's a new class of events that will challenge the theorists,” says SRON astronomer John Heise. Heise's colleague Erik Kuulkers, however, stops short of claiming that the explosions are carbon-based. The hydrogen-rich dwarf stars in those binary systems don't dump the necessary raw ingredients onto their neutron star companions, Kuulkers believes. That's not the case with a fourth long burst, found by Tod Strohmayer of NASA's Goddard Space Flight Center in Greenbelt, Maryland. On 9 September 1999, NASA's Rossi X-ray Timing Explorer (RXTE) satellite picked up a powerful flare from 4U 1820–30, the tightest known low-mass x-ray binary. The two stars whirl around each other once every 11 minutes within a volume of space just slightly larger than the planet Jupiter. The white dwarf feeds nearly pure helium to the neutron star, as its hydrogen gas was stripped long ago. That helium slowly builds the bomb, Strohmayer says. The helium layer grows 20 or 30 meters thick before it explodes, a process that can happen a few times a day at 4U 1820–30. Each blast leaves some carbon, one of the main ashes of helium fusion. Those ashes, in turn, mantle the neutron star with several hundred meters of carbon after about a year, according to Strohmayer's scenario. When the base of the layer reaches some critical temperature threshold, it ignites a carbon bomb that rages for hours. “This thing is 1000 times more powerful than the helium bursts,” Strohmayer says. “It may blow apart the entire accretion disk.” If that's the case, astronomers might gain their best insight yet into how disks of hot gas behave when they spiral into compact stellar remnants. New material from the frantically orbiting white dwarf would quickly replenish the blown-apart disk, Strohmayer says. RXTE's data may contain x-rays flowing from the disk during that reassembly, exposing the physics of the process as never before. Even more tantalizing to theorists is their first close look at the details of a real carbon detonation, rather than one based on computer codes. Already, there are hints that prior theoretical calculations don't quite align with the stars. Recent work by Bildsten and former student Edward Brown, now a Fermi Fellow at the University of Chicago, suggests that temperatures within the rind of a neutron star should fall well short of the billion degrees or so needed to ignite a relatively thin layer of carbon. A much thicker layer—and the resulting higher pressures at its base—might suffice to trigger the bomb. However, Brown notes, the rates of matter transfer in 4U 1820–30 imply that such a layer would require a century to accumulate. That would make Strohmayer's observation lucky indeed. Even so, Bildsten and Brown concur that the likeliest explanation is a carbon blast. As for reconciling the theoretical and observational differences, Bildsten says, “it's a fun problem for us to ponder.” 10. CLINICAL TRIALS # Planned Ritalin Trial for Tots Heads Into Uncharted Waters 1. Eliot Marshall A new study of whether a popular but controversial drug is safe and effective for preschool children also raises ethical concerns about using young subjects in clinical trials Doctors like to say that medicine is an art as well as a science. But the science seems a bit thin when it comes to giving drugs to young children. Physicians write millions of prescriptions for children each year without solid evidence that the therapy they're offering is safe or effective. To fill in that knowledge gap, U.S. scientists are gearing up for a major clinical trial intended to measure the effects of a popular stimulant on a previously untested population—children aged 3 to 6. But in doing so, they are also running up against one of the hottest ethical issues in research. The study, funded by the National Institute of Mental Health (NIMH) in Bethesda, Maryland, and set to begin in December, will recruit about 300 preschoolers diagnosed with attention deficit hyperactivity disorder (ADHD). Almost half will receive Ritalin, or methylphenidate (MPH), a drug used to help older children get along with playmates and fit into school routines. The scientists involved in the study admit that they are concerned about the drug's effect on the children's still-developing personalities and brains, as well as their inability to give informed consent. But they believe that such trials are the only way to answer concerns about rising use of the drug among this population. “We've put the study through an ethical wringer” to win approval, says NIMH director Steven Hyman. However, the demand for better scientific data runs counter to growing concern about human subjects research. “There is a big clash between the pressure to include children in clinical trials [and] discomfort about enrolling young children” in studies, says Norman Fost, a pediatric researcher and bioethicist at the University of Wisconsin, Madison. The ethical problems are reduced if the child volunteers have a chance of benefiting from experimental therapy, says Fost, who also notes that NIMH can be trusted to provide rigorous oversight. ## Drugs without data The idea for undertaking a trial like this arose 3 years ago, when psychiatrist Laurence Greenhill of the New York State Psychiatric Institute met with other academic researchers and NIMH to discuss the lack of safety and dose information for treating children under 6 with Ritalin. He proposed a large multicenter study that would enlist preschoolers in a test that would establish the best dose for very young children—a group rarely tested for any type of drug use. Peer reviewers at the National Institutes of Health (NIH) gave the proposal, called the Preschool ADHD Treatment Study (PATS), a favorable score in 1998. It then wended its way slowly through a series of ethics and policy review panels. Its importance seemed to grow, as people became increasingly aware of the need for better scientific data for some popular prescription drugs. Companies have run very few drug trials that include young children, but this hasn't stopped doctors from writing prescriptions. By one estimate, 94% of drugs given to children are prescribed “off label” in this way. In 1997, the Food and Drug Administration (FDA) drew up a “short list” of the 10 most widely prescribed products used for children without FDA approval. They ranged from the antiasthma drug Albuterol, approved only for children aged 12 and older, to the antidepressant Zoloft, approved for those 16 and older. Ritalin is on the list, prescribed 226,000 times in 1994 for off-label uses. Early this year, epidemiologist Julie Zito of the University of Maryland, Baltimore, reported in the Journal of the American Medical Association (JAMA) that the off-label use of psychoactive drugs by youngsters has increased steadily. Using Medicaid and insurance data, she traced a tripling from 1991 to 1995 in the use of psychoactive medicines among children 2 to 4 years old. MPH topped the list, and Zito estimates that there may now be 150,000 to 200,000 U.S. children in this age group taking it. Commenting on Zito's findings in the same issue of JAMA, Harvard University psychiatrist Joseph Coyle asked whether psychoactive drugs might affect the development of visual processing, language, motor skills, and memory of young children. The “disturbing prescription practices” documented by Zito, he said, deserve “more thorough investigation.” Hillary Clinton and White House health policy staffers picked up the message in March, urging officials to warn the public of the risks of off-label use of psychoactive drugs. With such encouragement, NIMH moved ahead and on 30 September awarded6 million to Greenhill and colleagues at five other psychiatric centers.

PATS is part of a broader effort to get better information on how prescription drugs affect children. Since the mid-1990s, advocates for children's health have been lobbying to get clinical researchers to include young volunteers in their studies, just as others campaigned to increase the representation of women and minorities in research. Now they've succeeded, and both FDA and NIH are taking steps to enroll thousands of children in clinical trials.

A 1997 law giving FDA new authority for such trials allows it to reward companies with exclusive marketing rights to a drug for 6 months if they agree to study the safety and efficacy of treating children. An April 1999 FDA rule requires companies seeking approval of a new drug to run tests that include children, if children are suitable candidates for therapy. In the psychiatric area, an FDA official said recently, the agency has already requested pediatric studies of drugs for posttraumatic stress, mania, social anxiety, and “premenstrual dysphoric disorder.” Soon, it may ask companies to study pediatric drugs for “conduct disorder” (a type of aggressiveness), panic disorder, and schizophrenia.

In parallel, the National Institute of Child Health and Human Development has created a 13-site network to provide technical support for more than 50 industry-sponsored trials of nonpsychiatric drugs. A recent announcement boasts of access to a pediatric population with 160,000 inpatient admissions and 2.3 million outpatient visits annually. NIMH has set up a seven-site clinical network to support similar work called the Research Units on Pediatric Psychopharmacology. All of these changes, a recent NIH announcement notes, have produced “an unprecedented surge in the number of pediatric drug trials.”

But PATS is the first major trial for preschoolers. “We are breaking new ground with this study; there was no prototype,” said Benedetto Vitiello, chief of NIMH's branch of child and adolescent treatment and preventive intervention, who spoke last month at a public forum on the PATS trial in New York City.*

## Trials and tribulations

Designers of the PATS trial have faced some novel challenges. One of the first was to develop a definition of ADHD for 3-year-olds. A child may be included only if an experienced clinician using an agreed-on set of behavioral criteria makes the ADHD diagnosis, a parent and a teacher both rate the child as troubled, and the child has exhibited symptoms for 9 months. Another hurdle is overcoming skepticism about the diagnosis, which has been controversial for decades. Today, ADHD is said to be a common disorder, affecting as much as 5% of the school-age population.

But critics question whether it makes sense to rely so heavily on chemicals, because ADHD isn't defined by a biochemical or even physical abnormality. These doubters range from the die-hard variety, like the Bethesda, Maryland, psychiatrist Peter Breggin, to moderate skeptics like pediatrician William Carey of the Philadelphia Children's Hospital. Carey has written that the “assumption that ADHD symptoms arise from cerebral malfunction has not been supported, even after extensive investigations.” The “very fuzzy” diagnosis of ADHD for school-age children includes a broad range of normal behaviors, he says.

Breggin, meanwhile, has signed up as an expert witness for parents of ADHD children who this year filed lawsuits against the manufacturer of Ritalin and psychiatric organizations in several states, alleging that they conspired to promote the drug. Breggin and California neurologist Fred Baughman Jr. blasted the use of Ritalin in congressional testimony on 29 September. Baughman called the ADHD diagnosis “a total fraud.” Enrolling young children in a trial of MPH, he adds, is “outrageous” and “immoral.” Robert Findling, a psychiatrist at Case Western Reserve University in Cleveland, Ohio, who conducts research in this field, dismisses such views as “nihilist” and says that they don't carry much weight among the experts.

Greenhill acknowledges that ADHD is “not a well-defined psychiatric disorder in this age group.” Findling agrees but says that ambiguity should not be an excuse for inaction. “Just because we don't know what causes it doesn't mean it's not a problem,” he argues. “This is not just benign fidgetiness. … Parents will tell you how awful it is that these kids can't be taken out in public because they're so impaired. We know they suffer for years and years.”

Hyman agrees that researchers have not identified any distinguishing biological hallmarks of ADHD. But he says that the disorder is well defined in behavioral terms, that ADHD children who fail to receive treatment often suffer life-changing harm, that older ADHD kids appear to benefit from drug therapy, and that no “really gross side effects” have been documented. All this adds up to a strong argument for the trial, he believes. “Without good clinical data, every child who receives this medication represents an uncontrolled experiment,” says Hyman. “That is entirely unacceptable.”

Assessment of efficacy is another major issue. How will researchers know whether a 3-year-old is functioning “on-task” during therapy—one of the goals of giving MPH? Greenhill explains: “We're going to set up a laboratory classroom, and we'll observe common tasks done in nursery school,” such as stacking blocks and stringing beads on a thread. Children will be asked to sit in a circle and take part in group events. The test will be whether the child is “compliant” and participates or “attends for a few seconds before drifting away and doing everything else in the room.”

The PATS researchers have taken other steps to allay qualms about the effects of MPH therapy itself and the difficult issue of getting informed consent. Each family will begin the trial with a 10-session “training” period in which researchers will attempt to treat ADHD with nonchemical therapy. Only if this fails will a child be assigned randomly to drug therapy or a placebo group. To address concerns about the effect of MPH on young children, only very low doses of MPH will be used in the initial stage—so low that a planning memo calls the level “homeopathic.”

In fact, Greenhill anticipates that this initial dose may have no effect. Once safety has been demonstrated, children will receive gradually higher doses until they reach a “best dose,” which they will receive for 40 weeks. No child will be enrolled who lacks the language skills to “indicate that he or she is in distress,” and parents will be given information and asked to give fresh consent on behalf of their child at each of the five stages of the trial.

After what Vitiello calls “zillions” of safety and ethics reviews and funding approval, the trial received a final green light from the data safety monitoring board. Greenhill says the clinics will begin recruiting the first subjects in a few weeks.

• *Research Forum on Optimal Strategies for Developing and Implementing Psychopharmacologic Studies in Preschool Children, Annual Meeting of the American Academy of Child and Adolescent Psychiatry, 25 October, New York City.

11. CLINICAL TRIALS

# Enforcers Halt NIH Study Called Less Risky Than Outdoor Play

1. Eliot Marshall

How much risk should healthy children face in a research experiment? The answer can land even the best clinicians in hot water, especially if they read the rules differently from the government officials who must enforce them. Such a disagreement led to the sudden suspension last week of research on childhood obesity at the National Institute of Child Health and Human Development (NICHD), part of the National Institutes of Health (NIH) in Bethesda, Maryland. The crackdown was imposed by a new enforcement group called the Office for Human Research Protections (OHRP), which disagreed with researchers on how to interpret the words “minimal risk.”

The study in question, by NICHD pediatrician Jack Yanovski, is examining genetic factors that affect body weight. Since its approval in 1996, the trial has enrolled more than 190 children aged 6 to 10. Some are obese and others are normal children of obese parents, considered at risk for obesity. The plan was to follow them for 15 years and collect data and blood samples, and perform x-rays and abdominal magnetic resonance imaging scans.

The controversy arises from an invasive procedure that is part of the study. It requires children to stay overnight in the hospital; two intravenous catheters are inserted and clinicians infuse insulin and sugar while taking blood samples. This “clamp” study was designed to manipulate blood sugar levels over a specific range (80 mg to 200 mg per deciliter) and measure each child's response. Children were paid up to $270 per visit. The government overseers who approved the study in 1996 assumed that the children were exposed to “minimal risk,” according to a 3 November letter from OHRP enforcement officer Michael Carome to NIH intramural research chief Michael Gottesman. Federal guidelines don't allow young children to be in a trial that has no therapeutic benefit—like this one—unless the risk is minimal. The standards were intended to ensure that a nontherapeutic experiment doesn't pose a greater risk than the child would encounter in a normal visit to the doctor, says Norman Fost, an ethicist at the University of Wisconsin, Madison, who helped draft the federal standards 2 decades ago. According to minutes from NICHD's ethics review, members felt that the children probably were exposed to bigger risks by “playing actively on sidewalks and streets” than they faced in the intravenous blood sugar study. But 4 years later, OHRP rejected this logic, concluding that the clamp study is not like a routine visit to the doctor. Yanovski has been instructed to consider how to make amends for being out of compliance with federal rules. This may mean informing everyone in the trial that the research involves more than minimal risk. OHRP wants a decision and a report by 8 December. NICHD will provide both, says Gottesman. But he insists that the research is benign and that it “will be extremely important for understanding the physiology of obesity and learning how to intervene to prevent it.” Yanovski, he adds, is “an enormously thoughtful, caring pediatrician” who makes Marcus Welby, the avuncular TV doctor, look mean. 12. NINTH INTERNATIONAL CORAL REEF SYMPOSIUM # Reef Migrations, Bleaching Effects Stir the Air in Bali 1. Dennis Normile NUSA DUA, INDONESIA—More than 1500 marine biologists, ecologists, conservationists, and community activists met at this Bali resort from 23 to 27 October for the Ninth International Coral Reef Symposium. Topics included the dynamics of reef fish communities and the lingering impacts of coral bleaching. ## Staying Close to Home Marine biologists have long known that adult fish that inhabit coral reefs don't stray far from home. But their larvae were presumed to be scattered by currents to distant shores, where they join existing communities, settle, and mate. This “open system” was thought to result in a thorough mixing of the gene pool, characterized by widespread homogenous populations. In the last few years, however, some scientists have begun to question that assumption. “The evidence is accumulating that reef fish communities are closed systems,” in which the vast majority of larvae return to the reef where they were born, says Robert Warner, a marine biologist at the University of California, Santa Barbara (UCSB). If that new vision holds up to scrutiny—and there are skeptics—it would not only change ideas about fish population dynamics but would also have implications for protecting marine biodiversity and managing fisheries. Domingo Ochavillo and colleagues at the University of Southern California in Los Angeles have employed genetic tools to look at interbreeding among three different populations of the rabbitfish Siganus fuscescens that live within 350 kilometers of contiguous reefs along Luzon Island in the Philippines. When the researchers examined the mitochondrial DNA, says Ochavillo, they found that each population had a distinct pattern, suggesting little genetic mixing among them. The group also found that late-stage larvae settling on a particular patch of reef shared the same genetic pattern as the local adults. Ochavillo and his colleagues obtained further evidence after releasing 500 larvae at a point 2 kilometers from shore. “They started swimming toward where they had been born,” says Ochavillo, leading him to conclude that “larval dispersal may not be as widespread as usually assumed.” Other studies support this so-called self-recruitment, in which larvae return to the community of their origins. Robert Cowen, a marine biologist at the University of Miami, fed data on actual Caribbean currents and information on survival times for larvae at sea into an ocean circulation model. Although seasonal currents affect the outcome, currents typically don't bring larvae close to suitable habitats within their limited larval phases. “You end up with relatively few viable individuals to populate downstream locations,” says Cowen. On three different reefs surrounding St. Croix in the U.S. Virgin Islands, UCSB marine biologist Stephen Swearer has been studying bluehead wrasse, reef fish with distinctive vibrantly blue heads. To track their movements, Swearer examined larval otoliths, bony structures in the ear that grow over the course of a fish's life and incorporate trace elements from the water it inhabits. “They are like flight recorders, allowing you to determine where a fish has been,” Swearer says. The results, he says, showed that the majority “had spent little time in pelagic waters” and never ventured far from the reef where they spawned. Work by Paul Barber, a Harvard University postdoc, suggests that these self-recruitment patterns have lasted for millennia. Barber gathered numerous specimens of three species of mantis shrimp from throughout Indonesia's nearly 2 million square kilometers of territory. Barber found that for each species, individuals from a given region shared distinctive genetic markers not found in other regions. Even though the three species were analyzed separately, the regions defined by the genetic groups tended to be the same. And each of these regions could be associated with deep ocean basins that would have been separate bodies of water during the last ice age, when sea levels were lower. “Despite 10,000 to 15,000 years of modern oceanic conditions, time and water movements have not erased regional genetic differences,” Barber concludes. If reef fish indeed live in closed communities, different strategies may be needed to manage and protect them, says Stephen Palumbi, a molecular biologist at Harvard University. At present, protected areas allow fish to grow larger and produce more eggs. But if those larvae remain close to home, local communities have a compelling economic reason to host a protected area. “We can make the case that hosting a marine protected area will benefit local fisheries,” Palumbi says. Limited larval movement also means that each region needs its own protected area to protect biodiversity. Peter Sale, a marine ecologist at the University of Windsor, Canada, says that what is missing from the discussion is a clear definition of the scale at which fish populations are open or closed. He suggests that the scale might vary by reef conditions, water movement in particular locations, and fish species. Previous studies conducted on smaller scales of roughly 100 square meters with fish communities large enough to form reproductive units found open populations with all newly settling larvae coming from outside. Current studies, he says, are looking at groups of populations on much larger scales. The line that separates open and closed populations “is a very important question,” he says, whose answer will affect where protected areas are located and how big they should be. ## Bleaching Takes a Toll on Reproduction When coral reefs regained their colors shortly after the massive bleaching triggered by the 1997–98 El Niño-La Niña event, reef scientists cheered. They hoped that the return of the characteristic brown, yellow, and rusty hues was a sign that the corals had recovered. But two new studies have reinforced previous suggestions that such recovery may be only skin deep. Bleaching, it seems, can damage the coral's reproductive capacity, meaning full recovery could take much longer than expected. One line of evidence comes from Makoto Omori, a marine ecologist at Tokyo University of Fisheries. When hit with an environmental stress, such as the warm waters brought by El Niño, corals expel zooxanthellae. The departure of these symbiotic algae turns the coral white—thus the term bleaching. In an ongoing study of coral genetics, Omori had gathered samples of five coral species from reefs off Okinawa for several years before the punishing 1998 bleaching. In samples collected before the event, more than 90% of the coral's spawned eggs were successfully fertilized in a lab experiment. But that rate dropped to 42% for samples gathered from the reefs in June 1999, a year after widespread bleaching. The same year, Omori and his colleagues also found that concentrations of sperm in reef seawater declined more quickly than usual after spawning. Omori believes that the coral are producing fewer and less fit sperm, which perish before finding an egg. The data “match field observations of dramatically reduced production of new coral polyps in Okinawa in 1999,” he says. At least one species also had a reduced fertilization rate in summer 2000, he notes. Areas within Australia's Great Barrier Reef took a similar reproductive hit, reported Selina Ward, a marine biologist at the University of Queensland in Brisbane. In March 1998, a month after the bleaching, Ward and colleagues began periodic sampling of 200 colonies around Heron Island Reef. Roughly half of these had bleached. By July, 23% of the bleached colonies had died and the rest had regained color, suggesting recovery. But in November, just before spawning season, Ward found that the colonies that had regained color produced half as many eggs and testes as colonies that had not bleached. “A lot of the corals [that had been bleached] had no eggs at all, and some had no testes,” she says. This low egg production continued the next year for large areas of the coral. Earlier research suggests that bleaching hurts coral reproduction in two ways, says Ward. Rising water temperatures alone appear to damage reproductive organs, whether or not the corals bleach. And when temperatures rise enough to trigger bleaching, the corals are effectively starving themselves, because they rely on zooxanthellae to supply their energy. “We know that in most organisms the first thing cut [under starvation conditions] is reproduction,” she says. These new findings have reminded the community of work done following a 1987 bleaching event in the Caribbean, says Mark Warner, a marine biologist at the University of Georgia, Athens. “These papers solidify the evidence,” he says, “and they're important because they came from two different geographical locations and involve different species.” Warner adds that the new evidence “makes the severity of the 1998 bleaching event more alarming. If the corals that are still alive can't reproduce over the next year or two, it raises questions about the chances of reseeding the population.” 13. VIKTOR HAMBURGER # A Lifelong Fascination With the Chick Embryo 1. Kathryn Brown* 1. Kathryn Brown is a writer in Alexandria, Virginia. Since the early days of developmental biology, Viktor Hamburger has inspired scientists. And at age 100, he isn't done yet Breakfast is always at 8 a.m. A short walk, 11:30. At 4 p.m., a book—Thomas Mann, perhaps, or Goethe. Such is the carefully choreographed day of Viktor Hamburger. Inside his St. Louis home, Hamburger divides up the hours. Blue eyes dart to his watch again and again, instinctively. There is time for biographies, news, visitors. Yet there are few hours to waste. Hamburger is 100. For a lifetime, Hamburger (pronounced “Hawmburger”), an embryologist, has made the moment count. Working with the simplest of tools—small glass needles, a basic microscope, a camera—he has asked, and helped to answer, some of biology's biggest questions. What do nerve cells gain from their targets? Do neurons die naturally? What causes an embryo's earliest movements? Hamburger did not have genomics or the polymerase chain reaction to help him tackle those mysteries. What he had was curiosity, discipline, and a sharp eye. At least three times—at ages 34, 50, and 66—Hamburger turned developmental biology on its head. He also weathered some storms, including rejection by his native country and, at least once, his peers. In this, Hamburger's 100th year, accolades fall like rain. At the Society for Neuroscience meeting in New Orleans last week, colleagues honored his accomplishments. In October, Washington University in St. Louis hosted the “Viktor Hamburger Centenary Symposium.” And just a few months earlier, the Society for Developmental Biology gave him its first lifetime achievement award. Gerald Fischbach, director of the National Institute of Neurological Disorders and Stroke, calls Hamburger “a true intellect, always wry and warm.” Dale Purves, a neuroscientist at Duke University, adds: “Viktor is one of the most influential scientists in neurobiology and development.” Hamburger was a biologist from the beginning. He grew up comfortably in the quiet German town of Landeshut, now part of Poland. His parents, who owned a textile factory, collected art and entertained often. Meanwhile, Hamburger and his two brothers spent plenty of time outdoors, scrambling over rocky hills and wading in ponds. By age 10, Hamburger was scooping up the spring eggs of frogs and salamanders so he could watch the eggs develop in an aquarium. Still, it was partly luck that he landed in developmental biology. In 1920, Hamburger enrolled in zoology at the University of Freiburg. At the time, renowned embryologist Hans Spemann headed the department. “And so, I had no choice,” jokes Hamburger. “I became an embryologist.” In 1932, Hamburger, then an instructor at Freiburg, met his lifelong lab partner: the chick embryo. He accepted a Rockefeller Fellowship to join Frank Lillie's lab at the University of Chicago, leaving the familiar German hills for the “powerful, impressive, but somewhat scary new scenery” of the United States. At Chicago, Hamburger learned to saw a window in a chick shell and, with glass microneedles, cut and transplant budding limbs on the embryonic chick inside. He had a basic question: In the embryo, how do developing limbs influence the nerve centers that stimulate them? Answers soon began to emerge. When Hamburger removed a chick embryo's limb bud, the adjacent part of the spinal cord seemed to shrink. But wherever he transplanted an extra limb bud, the nearby spinal tissue appeared to swell. Some growth factor, Hamburger reasoned, must flow from developing peripheral organs, like limbs, back to the nerve cells that innervate them. That insight would eventually lead to two of his major contributions. But his Chicago adventure was rudely interrupted. In April 1933, a letter arrived from the chancellor at Freiburg informing Hamburger that he was out of a job. Adolf Hitler had come to power—and banned Jews from teaching at universities. Hamburger hurried to Germany to collect his wife and young daughter. His professional fate was secure—he soon accepted one of several offers and moved to Washington University. But the exile took a personal toll. Hamburger lost contact with many family members, including his brothers, for years. He found solace at the lab bench. “I spent morning and evening—and sometimes hours after dinner—on my research,” Hamburger recalls. He also forged an extended family of faculty members. As head of the zoology (later called biology) department for 25 years, he organized weekend outings in the Missouri hills. Researchers would often bring their children along, gathering salamander eggs for embryology classes or identifying flowers in the meadows. “Viktor had a genuine regard for the people he worked with, and we were all a big family,” remembers Doris Sloan, one of Hamburger's two daughters and a geologist at the University of California, Berkeley. ## Embryonic art The academic family grew in 1947, when Hamburger invited Italian neurobiologist Rita Levi-Montalcini of the University of Turin to spend a year in his lab. Working together, they repeated his earlier limb bud experiment—and soon reached a conclusion. Developing limbs, they suggested, pump some trophic factor back to the spinal cord that prevents the nerve cells inside from dying. Along the way, they noticed something striking: lots of nerve cells dying normally. At the time, researchers assumed that decaying neurons reflected biology gone awry. But Hamburger and Levi-Montalcini discovered that in most areas of the nervous system, many more neurons are born than will ever be needed. Regulated cell death then shapes the nervous system, matching innervation centers to targets, like limbs. That discovery paved the way for today's studies of neurodegenerative disease. Buoyed by their discovery, the researchers kept on with their original task: finding the trophic factor, or growth protein, that developing tissues must send back to nerve cells. They might have labored for decades. But at just that time, Hamburger says, “a rare gift from Heaven” arrived: a reprint from Elmer Bueker, one of Hamburger's former Washington University students. Bueker had discovered that implanted tumor tissue, like developing tissue, sends some maintenance protein back to neurons—quickly doubling the size of some neural regions in experimental chick embryos. This easy-to-measure effect was just what the team needed. Within months, Hamburger and Levi-Montalcini had planted mouse tumors in their chick embryos and confirmed that the tumors secreted a molecule boosting nerve cell growth. Soon after, they recruited biochemist Stanley Cohen. In the 1950s, Cohen and Levi-Montalcini finally isolated and identified the mystery molecule: nerve growth factor (NGF). Levi-Montalcini devoted the rest of her career to characterizing NGF, spending more than 2 decades at Washington University. But Hamburger had little interest in the biochemical minutiae of NGF. Instead, he turned his gaze back to the chick embryo, charting its stages of development like a parent sketching day-by-day portraits of a newborn. His description of normal chick development became one of the most widely cited papers in biology. Looking back, researchers say Hamburger was uniquely equipped for the job, with a sharp visual sense that came from art as much as science. As a child, he was surrounded by the art—and artistic friends—collected by his parents. And as an adult, Hamburger befriended his own circle of artists and dancers, filling his home with original paintings. “If you confine yourself too much to science, you lose a lot,” Hamburger says now. “A lot of scientists become very abstract and forget about the visual world in which they live. The eyes are important.” Indeed, it was Hamburger's keen eyes that led to another landmark discovery in the 1960s. He had long noticed the herky-jerky movements of embryonic chicks in their earliest days. A chick's right wing, say, or left foot would randomly quiver in uncoordinated fashion. This early, awkward movement, Hamburger reasoned, must be spontaneous, without any urging from sensory stimuli. But that idea was theoretical heresy: Behavioral psychologists were convinced that, from the beginning, movement is simply a response to sensory stimulation. To set the record straight, Hamburger spent the decade of his 60s tracking the chick embryo's movements, from its first bend of the head at 3.5 days. Working with both chicks and rats, he and his colleagues also removed key sensory neurons that normally stimulate the embryos—and found that the embryos still twitched and quivered as before. “In characteristic fashion,” says Washington University developmental biologist David Kirk, “Viktor turned this field on its head in just a few years.” The research ultimately showed that in vertebrates, at least, adaptive motor behavior is preprogrammed and develops in the growing embryo without any sensory feedback. ## Like clockwork Kirk arrived at Washington University in 1969, when Hamburger had reached retirement age. “He told me he was going to have one more 3-year grant cycle and then be done,” Kirk says. “But every time that grant came close to expiring, he'd say, ‘I can't quit now.’ It just never occurred to him to slow down, as long as there was another question.” And there was always another question. In the 1980s, Hamburger finally left the lab bench—but not the university. He lunched with colleagues at the medical school every Tuesday for almost a decade; when he could no longer drive, he came by cab. “Viktor was like clockwork,” recalls Fischbach, who was at Washington University at the time. “He would arrive at the same time every week to eat the same kind of sandwich, wrapped precisely the same way.” Hamburger came to talk about science: how it should be done, and why it should be remembered. “He had a wonderful sense of humor,” adds Fischbach, “and you never knew when he'd burst out in a guffaw.” That sense of humor was put to the test in 1986, when the Nobel Prize committee stunned biologists by awarding the Nobel Prize in physiology or medicine to the discovery of growth factors, including NGF—but not to Hamburger. Instead, the prize went to Levi-Montalcini and Cohen alone (Science, 31 October 1986, p. 543). “This research could not have been done without Viktor,” comments Jean Lauder, a developmental biologist at the University of North Carolina, Chapel Hill. His omission from the prize, adds Purves of Duke, was “extremely unfortunate.” But Hamburger took the slight in stride. When Lauder phoned him following the announcement, he told her: “That's all right—I don't have time to go to Stockholm right now. I need to finish proofreading the bibliography for my Spemann book.” By that time, Hamburger had launched the last major effort of his career: a series of books and articles reviewing embryology's history, including an English translation of papers by his German mentor, Spemann. Hamburger wrote it all by hand, reliving a century's worth of science in even, steady pen strokes. “I never did own a typewriter,” he reflects. “I guess I trusted myself more than a machine.” Recounting his achievements in print, he understood all that he had accomplished. “I know what I've contributed,” Hamburger says. And so do his colleagues. Visiting Hamburger last month, Fischbach expected to find him frail or fatigued. Had life finally caught up with the man who helped define it? Yes, Hamburger had lost some hair and some hearing. He moved slowly. And yet there he sat, grinning in his chair, telling stories, gesturing past stacks of books yet to read. This, perhaps, is Hamburger's most personal contribution. “Viktor,” says Fischbach, “has taught us how to overcome life's challenges.” 14. EUROPEAN SPACE AGENCY # New Science Chief Must Juggle Missions and Politics 1. Andrew Watson* 1. Andrew Watson writes from Norwich, U.K. Space scientists hope David Southwood can balance ESA's research with its widening interests Early in his career, David Southwood says, he had to choose whether to apply his training in plasma physics to fusion research or to space. “I didn't have any doubt which way I wanted to go,” he says. “I particularly find the solar system fascinating, because I'm interested in why the place we live is the way it is—why the Earth is like it is, why the planets are like they are.” That career choice, Southwood says, has landed him “one of the most interesting jobs in Europe, and equally so in space science”: Next May, he will take over as the European Space Agency's (ESA's) science director. Southwood, 55—who has spent most of his career at Imperial College in London, where he headed the physics department from 1994 to 1997—will take on a clutch of ambitious projects that ESA hopes to pull together in the coming decade. Two bright stars are European participation in the Next Generation Space Telescope—the Hubble's replacement, which is scheduled for launch in 2008—followed the next year by a half-billion-dollar mission to Mercury called Bepi-Colombo. Others to fill out the constellation include an orbiting gravity wave detector, a satellite for monitoring storms in Earth's magnetosphere, and a mission to determine the positions of stars with high precision (Science, 22 September, p. 2019). But perhaps the most daunting task Southwood faces is to walk a political tightrope: He must balance the aspirations of scientists from ESA's 15 member states with calls to tie the agency more closely to the business and security sectors (see sidebar). “The scientific challenges are great, but I also like the political challenges of getting a European program together,” says Southwood. Experts say he's up to the task. Southwood “has a good background of relevant experience,” says Britain's Astronomer Royal, Martin Rees. Adds Stamatios Krimigis of the Johns Hopkins University Applied Physics Laboratory, “You know he is a leader when you meet him.” Southwood is no stranger to ESA politics. He has contributed to several ESA projects, including the Cluster mission to explore the interplay of the solar wind and Earth's magnetic field, and the SOHO mission to study solar storms. And he led the team that built a magnetometer for NASA's Cassini mission, now on its way to Saturn. “I became a theorist and then realized it was more fun working with experiments as well,” he says, “and ultimately I became an experimentalist by leading the building of instruments.” Perhaps crucial to Southwood's appointment was the time he spent earning his management stripes in ESA's Earth Sciences Division from 1997 until last April. Southwood is credited with breathing life into a mundane monitoring effort by creating the Living Planet program, leading to upcoming missions that will study everything from ocean circulation to Earth's gravitational field. The overhaul is widely viewed as a success story, says Krimigis, but Southwood faces a tougher job in getting ESA's upcoming big-ticket missions to fly on budget without diluting the science. Toward this end, Southwood says he gained valuable experience at the instrument bench. “You have to get a firm hold on the engineering of a project if you are going to get the project done to time and to cost,” he says. Says Krimigis, “I wish him good luck.” At ESA, Southwood will succeed the highly respected Roger Bonnet—“an incredibly hard act to follow,” he says. With European countries struggling to reconcile their national aspirations with their pan-European goals, Southwood views both ESA and CERN, the European particle physics laboratory near Geneva, as models for how a wider European integration might work. “If Europe can't get its act together in something like science, big science where it has to, what's the hope of it doing something in more politically complex areas?” he asks. “European space activities have to be successful if they are going to make people feel positively about Europe.” 15. EUROPEAN SPACE AGENCY # Getting More Out of Space 1. Andrew Watson* 1. Andrew Watson writes from Norwich, U.K. Space is too important to Europe to be left to scientists alone, according to a report on the future of the European Space Agency (ESA) released last week. The report (available at http://www.esa.int/) concludes that better coordination between ESA and the business and defense sectors is essential to Europe's development. “We see the need to integrate space efforts, or space activities, with European political and other activities much more clearly than has been the case in the past,” says former Swedish Prime Minister Carl Bildt, who led the study. The report refers to the$2.5 billion Galileo project—a satellite navigation system intended to rival the U.S.'s Global Positioning System (GPS) and Russia's GLONASS system—as a case in point. The first of Galileo's 30 satellites is slated for launch in 2004. For Europe, Galileo offers self-sufficiency in global navigation and a break from foreign military hegemony. Telecommunications and other commercial opportunities seeded by this joint ESA-European Union project are expected to be huge. Galileo “will give strategic autonomy to Europe in a sector which is vital for the evolution of the European economy,” says ESA Director-General Antonio Rodotà. But the report points out that at present there is no mechanism to decide how to limit access to Galileo in times of conflict. Other unsettled questions that ESA should get involved in include how to guarantee a return on private-sector investments and how to organize a single public entity for operating it.

The report from the “Three Wise Men”—the other two are Jean Peyrelevade, president of the Paris-based Credit Lyonnais bank, and Lothar Späth, CEO of JENOPTIK AG, an optoelectronics firm in Jena, Germany—offers few surprises, says ESA science director-designate David Southwood. “The aim, I am sure, is to try to wake European policy-makers and politicians up to the fact that space capability and space-derived information provide part of a modern developed society's infrastructure,” he says. His only reservation is that in broadening ESA's horizons, science could lose out. “Space science cannot be ignored or downgraded,” he says, “as one broadens the perception of space's use.”

16. DEMOGRAPHY

# A Billion and Counting: China's Tricky Census

1. Daniel Walfish*
1. Daniel Walfish is a freelance writer in Beijing.

Officials try to keep politics out of the world's biggest enumeration as they gather data on mobility, fertility, and other sensitive demographic indicators

BEIJING—It was a few minutes past midnight on 1 November when the police knocked on his door. “Census,” they announced as their reason for rousting the young Beijing pedicab driver out of bed. The police were helping census takers locate migrant workers who might otherwise flee or avoid participating in the largest enumeration in history. But when the driver failed to produce the appropriate papers, he says the police also fined him \$6 and threatened to send him back to his home in Henan province. “Their attitude was violent,” says Bai, who gave only his last name. “They were not conscientious or considerate.”

The brusque behavior of the police is an unintended offshoot of the fact that census-taking is serious business in China. This month some 6 million clipboard-carrying workers knocked on 350 million doors in the fifth census since the founding of the People's Republic in 1949. For weeks Beijing's streets have been covered with slogans like “truthful reporting is every citizen's duty” and “the census is good for the country and good for the people.” The central government has gone to great lengths to emphasize the benefits of providing accurate data to set national policies on housing, education, the environment, and other social issues. “The census is very important for us,” says Chen Shengli, a spokesperson for the State Family Planning Commission, one of a score of ministries anxiously awaiting the results, expected out in preliminary form in February.

For millennia, the central government has sought accurate demographic information to manage its vast population. The first four PRC censuses were remarkably good by international standards, says Judith Banister, the former head of the U.S. Census Bureau's China branch and now a professor at the Hong Kong University of Science and Technology. The counts, she says, were bolstered by a strict household registration system, a relatively immobile population, and a vast supply of low-cost labor. But more than accuracy is at stake. The wrong numbers could have political ramifications if they raise questions about compliance with such policies as the one-child-per-family rule. They could also—despite stern warnings to the contrary—lead to reprisals at the local level.

This census will produce a much fuller picture of Chinese society than the last one, done in 1990. It marks the first time that Chinese statisticians have felt confident enough to use a long form, says Y. C. Yu, former head of demographic and social statistics at the United Nations and a trusted adviser to the Chinese government. A randomly selected 10% of the population is being asked additional questions about their movements, housing, sanitary and cooking facilities, employment, education, and fertility rates. At the same time, census officials rejected requests from researchers to include questions about income and household possessions after pilot runs showed that the Chinese, like citizens in many other countries, would refuse to cooperate.

Answers to the long form are intended to illuminate the numbers and migratory patterns of China's populace, including the 60 or so million people who have left their farms and villages in search of a better life since China began to relax economic and social controls in the 1980s. This “floating population” represents a potential powder keg for a government trying hard to control urbanization and maintain order. But using the census to monitor their movements is tricky. Asking how many family members are away, for example, may lead to an accurate total for migrants, but it won't say anything about where they are and what they are doing. At the same time, respondents are unlikely to implicate themselves in any wrongdoing. “People employing peasants or [illegally] housing migrants really have an incentive not to tell the truth,” says Dorothy Solinger, a political scientist at the University of California, Irvine.

To reassure the public, the government has repeatedly proclaimed that its census takers are ordinary citizens, that the information will remain confidential, and that the census won't be used as an excuse to send migrants back to their hometowns. But as Li Xiru, a census manager at the statistics bureau, acknowledges, census takers are told to “find out the real situation” by talking to other grassroots sources. That's what happened in Bai's case, where the same police who often treat migrants harshly were enlisted to make sure those migrants were counted.

Despite the government's promises, many citizens equate the census-taking process with the police. “I don't like the census because I'm an illegal resident of Beijing,” says a recent college graduate surnamed Wan who worries about being discovered. Indeed, rising expectations of privacy have become a problem for the government in this year's enumeration. “People's ideas have changed,” says Qian Yukun of the National Bureau of Statistics, “so their degree of cooperation has also changed.”

Another sensitive issue the census will explore is the country's fertility rate and ratio of boys to girls at birth. Although researchers and family planners are desperate for such data, accurate information is hard to come by. Families worried about being fined for births that violate China's strict family-planning policies may try to hide children. Local officials who fear the consequences of not enforcing quotas may contrive to depress the reported counts for their villages. As a result, says Banister, past censuses have undercounted children younger than 6 by about 5%.

For the past decade, the statistics bureau, which also runs the census, has done an annual survey that samples 0.1% of the population. The results are routinely adjusted at the national level—by as much as 15% for birth figures—to correct for underreporting. Last year's figure put the country's population at 1.26 billion. Census numbers are never altered, however, as they are based on an actual enumeration rather than a sample. That situation has led some demographers to speculate about what might happen if the 2000 census produces a number below 1.26 billion. An official in the statistics bureau recognizes this potentially embarrassing development. “Of course,” he admitted reluctantly, “the count shouldn't be low.”

The government has tried to address the problem of undercounting by telling local officials that they will not be punished if the census results in a number that exceeds their quota. On the other hand, a similar pronouncement in 1990 did not prevent actions against some unfortunate cadres, say some researchers. The problem, they add, is the central government's tenuous hold over local officials. The center “can issue these wonderful-sounding pronouncements,” explains Susan Greenhalgh at the University of California, Irvine, “but the localities often do just what they want to do.”

There are also disagreements within the central government. Qian of the statistics bureau says the government has told families that fines for registering extra children will be reduced this year to foster an accurate tally. Not so, says Chen of the State Family Planning Commission, who insists that families have never been given amnesty of any kind for excess births.

One legacy of the single-child policy is a distorted gender ratio at birth. The patterns are aided by the widespread use of ultrasound machines in the countryside to determine the sex of the fetus which, in some cases, lead to aborting females. A 1995 survey showed an alarmingly high ratio of 118 boys to 100 girls aged 0 to 4, compared to the normal birth ratio of 106.5:100. Some foreign researchers believe that infanticide is a factor behind the imbalance, although Chinese demographers tend to dismiss the idea and to focus instead on selective neglect. In any case, the census is expected to quantify what some officials and researchers see as a potentially huge problem—the social consequences of a staggering surplus of unmarried males.

At one point demographers hoped that the lopsided sex ratio reflected an underreporting of girls by couples who were reserving an official space in the family for an unborn boy. But there is a growing consensus that, for the past 15 years at least, most of the “missing girls” never existed.

At the other end of the life cycle, this year's census will update mortality data that are hard to come by because China does not have reliable death registries. This undercount of the dead inflates the expected life-span. Huang Rongqing, a demographer at the Capital University of International Economics and Trade in Beijing, offers several explanations for the undercount, including peasants who break the law by burying instead of cremating their dead and factory bosses employing illegal migrant workers. In addition, Huang notes that discussions of the dead are taboo among some of China's minorities and that peasant parents are afraid of being blamed for the death of an infant.

The statistics bureau hopes to publish complete results of the census in 2002, along with samples of raw household-level data. That information is eagerly awaited by researchers everywhere. They are heartened by promises from the bureau to do better than in 1990, when similar sampling data proved to be virtually useless. But the census is also a test of the ability of the body politic to paint an accurate picture of itself. A Beijing taxi driver named Hu expects his countrymen to pass with flying colors. “Of course people will report truthfully,” he says. “Not reporting truthfully is wrong, isn't it?”