# News this Week

Science  23 Feb 2001:
Vol. 291, Issue 5508, pp. 1460
1. PALEOANTHROPOLOGY

# Scientists Spar Over Claims of Earliest Human Ancestor

1. Michael Balter

PARIS—Brigitte Senut took the weathered fossils from her safe and laid them out, one by one, on her desk. To an untutored eye, they might not look like much: broken femurs, bits of lower jaw, several teeth—13 fossils in all. But these relics have caused one of the biggest sensations in the field of human evolution in years. In two papers scheduled for publication in the 28 February issue of the Comptes Rendus de l'Académie des Sciences, a team led by Senut, a paleontologist at France's National Museum of Natural History, and geologist Martin Pickford of the Collège de France claims that these 6-million-year-old bones unearthed in Kenya represent our earliest known ancestor.

If that's true, “Millennium Ancestor”—so called because the bones were found last year—would predate other leading candidates by some 2 million years. But Senut and Pickford have a more drastic shake-up in mind for the human family tree. They believe that all australopithecines—hominids which include the famous skeleton Lucy, whose species is thought to be one of our direct ancestors—should be relegated to a side branch in favor of their specimen. Millennium Ancestor appears to have been a bipedal primate—perhaps the first of its kind—at home equally on the ground and in the trees. Because the fossils date to a period when the human and ape lineages are thought to have split, any primate remains from that time could shine a strong light on our murky origins. The findings also enflame an ongoing debate about what constitutes a hominid.

Such dramatic claims require big-time proof, of course. So press conferences in Nairobi in December and Paris earlier this month left most colleagues unmoved until they could see the description in a peer-reviewed journal. On 15 February, the French Academy lifted the embargo on the papers, giving the community its first good look at the riveting bones, which Pickford and Senut have labeled Orrorin tugenensis (Orrorin means “original man” in the local dialect). In the meantime, some researchers contend that the team collected the fossils illegally, charges that Pickford and Senut have strongly denied (Science, 15 December 2000, p. 2065; 9 February, p. 986).

The verdict on the significance of the fossils, in a poll of more than a dozen experts, is split. “This discovery is very exciting and very important,” says Brian Richmond of the University of Illinois, Urbana-Champaign. He says that Pickford and Senut have “come the closest yet to finding evidence of the base of the human family tree.” Milford Wolpoff of the University of Michigan, Ann Arbor, agrees, calling Orrorin a “great discovery, one with key information about hominid origins and early evolution.”

Others, however, doubt that the bones even belonged to a hominid—a loose classification that currently includes the australopithecines and the genus Homo—or even that the species they belonged to walked on two feet. “The case for a hominid is weak,” argues Lucy co-discoverer Donald Johanson, director of the Institute of Human Origins at Arizona State University in Tempe. Indeed, says David Begun of the University of Toronto, the fragments cannot reveal whether Orrorin was “on the line to humans, on the line to chimps, a common ancestor to both, or just an extinct side branch.”

The one thing experts do agree on is the date of the fossils, which were found last fall in lake and river sediments in the Tugen Hills of northwestern Kenya. “The dating might ultimately require some fine-tuning, [but 6 million years] is at least in the ballpark,” says John Kingston of Emory University in Atlanta, who has conducted detailed studies of the area's geology. Indeed, the site's potential importance has been recognized since at least 1974, when Pickford discovered a single molar there that he and some others believe belonged to a hominid.

“I have been expecting this discovery for 25 years,” says Yves Coppens of the Collège de France, a co-author of one of the Comptes Rendus papers. “One tooth was enough to know that hominids were there.”

Indeed, the team argues that the teeth found among the remains are key evidence for Orrorin's status as a human ancestor. The molars are thickly enameled, small, and squared—features retained in modern humans. Most australopithecine molars, although also thickly enameled, are much larger. That evidence impresses Wolpoff, who points out that the oldest potential hominid before Orrorin was a 4.4-million-year-old australopithecine whose molars—although small like Orrorin's—have unhumanlike thin enamel.

Other researchers are unswayed by the dental evidence. Begun says that because some early Homo species had large molars, Orrorin's small molars alone are insufficient to sideline australopithecines: “If big molars exclude Australopithecus from a close relationship to Homo, they also exclude most of early Homo.” Others assert that enamel thickness varies so much from one species to another that it may not be a valid measure for evolutionary relationships.

But there are more than teeth for the community to chew on. Pickford and his colleagues also believe that Orrorin's femurs have several features ancestral to later Homo. One of the three retains its head, which fits into the pelvis. The team points out that the femoral head—although smaller than that of modern humans—is nevertheless much larger than Lucy's. According to Pickford, this implies that Orrorin's femurs were built to support its upper body in a bipedal stance long before australopithecines arose. He concludes that this makes Lucy—who lived 3 million years ago and had smaller femoral heads—an unlikely human ancestor.

To some experts, Orrorin's femurs push back the earliest evidence for bipedalism by almost 2 million years. “Upright walking goes way back in prehistory,” says Ron Clarke of the University of Witwatersrand in Johannesburg. Johanson notes that other features of Orrorin's femurs, including grooves where muscles and ligaments needed for walking on two feet might have attached, could be evidence for bipedalism.

But the femur argument has gotten a lukewarm reception from others. They point out that many male specimens of Lucy's species—Australopithecus afarensis—have much larger femoral heads than Lucy's. (Few researchers agree with Senut's contention that these larger specimens belong to another species.) And Alan Walker of Pennsylvania State University, University Park, argues that the lower femur—not found among Orrorin's remains—would be more likely to make the bipedalism case by revealing the structure of the knee.

These conflicting views reflect the fact that experts lack a clear definition of a hominid, says Jeffrey Schwartz of the University of Pittsburgh. But that only means researchers seeking to penetrate our shadowy origins will be debating Pickford and Senut's find for years. Says Leslie Aiello of University College London: “If half of what they are claiming is true, it's fantastic.”

2. JAPAN

# Fusion Scientists Urge Closer Look at ITER

1. Dennis Normile

TOKYO—Japan's scientific community has always appeared to be four-square behind the $5 billion International Thermonuclear Experimental Reactor (ITER). But last week, the first cracks in that unified front appeared as the country's leading fusion researchers gathered to discuss the megaproject's potential impact on the country's fusion research efforts, with some urging a fresh look at other options. “We really should have gotten more involved sooner,” says Osamu Motojima, one of the fusion scientists raising concerns about ITER. Scientists are worried that the recent merger of Japan's two major science agencies will put other projects more directly in competition with ITER for funding. The debate comes as the major ITER partners—Japan, Europe, and Russia—prepare to select by next year a site and budget for the giant machine, which has a troubled history. When researchers first proposed the project in the early 1980s, there was substantial support in the United States and elsewhere for the idea of harnessing nuclear fusion, the process that fuels the sun and stars, to produce energy on Earth. Some government budgetmakers were shaken, however, after scientists estimated that it would cost$10 billion to build a tokamak—a doughnut-shaped device in which a magnetic field contains a superheated ionized gas, or plasma—capable of containing the violent reaction. Despite efforts to cut costs by scaling back the device, the U.S. Congress abandoned ITER in 1998, leaving the three remaining partners to complete the project on their own.

They have since finished basic design work on the slimmer version, flippantly called ITER Lite. Each is now preparing to propose a candidate site for the reactor, with a final decision on the location and funding to come by the end of 2002. Completion is not expected before 2013.

Many Japanese scientists—including a solid majority of the 300 who stayed to the end of the meeting—would like to see the device built in their nation. “I think virtually everyone is convinced of [ITER's] scientific and engineering feasibility,” says Kenro Miyamoto, a plasma physicist and professor emeritus of the University of Tokyo. There is also widespread agreement that the next big step for fusion research will be a facility to study an actual burning plasma.

But there is scattered opposition to ITER. And some scientists wonder if it is the best bet for plasma studies. “ITER is one candidate, … [but] we need to investigate other alternatives,” says Motojima, director of research at Japan's National Institute for Fusion Science (NIFS). The institute operates the Large Helical Device, a $650 million facility that confines plasma in a magnetic field created by spiraling coils instead of the plain rings of a tokamak (Science, 20 March 1998, p. 1846). Although such helical devices “are a decade or two” behind ITER's tokamak technology, Miyamoto says the government should still fund research into alternatives. And more than a quarter of those at the meeting felt that ITER's scientific details need to be vetted more carefully before the partners move ahead. Their concerns are fueled, in part, by the January merger of the Science and Technology Agency (STA) and Monbusho, the Ministry of Education, Science, Sports, and Culture. STA traditionally funded big science projects, including ITER, while Monbusho handled virtually all other fusion work at universities and other institutions. The former STA's budget currently provides$139 million for ITER R&D and other projects, while Monbusho spends $108 million on NIFS and other academic fusion work. But with all fusion funding now coming out of a single bureaucratic pot, some researchers worry that paying up to 60% of ITER's overall cost could squeeze out other research. “It would really be a shame if Japan does not maintain its current position at the forefront of a broad range of approaches to fusion,” says Atsuo Iiyoshi, former director-general of NIFS and now president of the private Chubu University in Nagoya. He notes that NIFS's Large Helical Device has been closing the technological gap with tokamaks, and that Osaka University's Institute of Laser Engineering is making steady progress in using lasers to crush fuel pellets to the point of igniting fusion. These and other university-based research facilities, he says, could define the characteristics of a power-producing reactor. Adding to concerns, Iiyoshi notes, was the government's decision to eliminate a Monbusho advisory council that had staunchly supported university-based fusion research. “We're in a transition period, and it's hard for researchers to see where decisions are being made,” he says. Where the discussion will lead is not clear. Shuichi Takamura, a Nagoya University electrical engineer and a key organizer of last week's meeting, says research leaders hope to issue some sort of report. “We're still discussing what the next step should be,” he says. Iiyoshi suggests that any decision on ITER be delayed for half a year or so to allow the Japanese government to work out a comprehensive strategy for fusion research. But with the government still firmly backing ITER, further delays are unlikely. Miyamoto predicts that the situation “will be resolved within a couple of months.” That means Japan's fusion science community must act quickly if it wants its voice to be heard. 3. 2002 SPENDING # First Bush Budget May Put Science on Diet 1. David Malakoff, 2. Jeffrey Mervis The Bush Administration's first budget request to Congress may leave many scientists feeling a little flat. White House officials will release a preliminary spending proposal next week for the 2002 budget year that is expected to boost biomedical and military science but hold down new spending at the National Science Foundation (NSF), NASA, and the Department of Energy (DOE). Rumors about the plan, which White House officials were still assembling as Science went to press, have alarmed some science groups and members of Congress, who were expecting spending hikes for nonbiomedical science as well. “It looks like the budget's starting point is not going to mean boom times for science,” says David Goldston, staff director for House Science Committee chair Sherwood Boehlert (R-NY). “The way [the proposal] is unfolding raises concern,” adds Senate aide Cheh Kim, who works for the appropriations subcommittee that oversees the budgets of NSF and NASA. That panel is led by Senators Kit Bond (R-MO) and Barbara Mikulski (D-MD), who last year launched a campaign to help others catch up to recent increases in biomedical research spending by doubling NSF's budget, now$4.4 billion, by 2006. The 2002 budget covers the fiscal year that begins on 1 October.

The NSF doubling effort, however, is expected to get little support in the plan that will be released on 28 February. Knowledgeable sources say that the White House whittled down NSF's initial double-digit request to 1%, which the agency then countered with an appeal for a boost of 6% to 7%. The final request will probably fall below the predicted inflationary rate of 3% to 4%, sources predict.

At the same time, NSF director Rita Colwell seems to have salvaged at least a chunk of her plan for a fivefold increase over 5 years in mathematics research. Sources say that the mathematics division may garner up to one-third of the agency's total projected increase for research. “The budget is a disaster for NSF as a whole, but she stood up for mathematics,” says one NSF official.

NSF is also expected to benefit from a slice of the president's education initiative. Although most of the media's attention has focused on proposals for testing and accountability for elementary and secondary schools, NSF officials and members of Congress have also lobbied hard for a component that would involve higher education, in particular teacher training, as well as programs to strengthen the country's technological workforce.

Other nonbiomedical science agencies also face stagnant spending. NASA's $14 billion budget will reportedly barely keep pace with inflation. DOE's$3.2 billion Office of Science could get squeezed by an even smaller overall agency increase, as officials channel funds to other Bush Administration priorities, such as weapons technology and improving security at national laboratories. Department of Interior officials are also said to be mulling significant cuts in science programs—such as those run by the U.S. Geological Survey—to finance a Bush campaign promise to spruce up national parks.

Other research funders—from the Environmental Protection Agency to the National Oceanic and Atmospheric Administration—are also expected to fare poorly. The belt-tightening is a result of a presidential promise to deliver a $1.6 trillion tax cut, plus increased spending on education and the military, while keeping a lid on spending. Administration officials say they want to boost discretionary spending (which does not include required spending on social welfare programs) by 4% to about$663 billion. But nearly two-thirds of the planned $26 billion increase would go to education and defense, leaving little money for other programs. One winner in the initial budget battle, however, appears to be the National Institutes of Health (NIH). Bush is expected to follow through on a campaign promise to keep biomedical research spending on a path to double by 2004; doing that would technically require a 15% increase this year ($3.4 billion) for NIH. Defense science also may jump, with Bush promising to increase the Pentagon's R&D budget by 6%. How much of that increase would go to basic research, however, remains unclear.

If the details seem unpredictable right now, so does Congress's reaction. Key lawmakers—including Senator Pete Domenici (R-NM), leader of the Senate Budget Committee that will formulate Congress's initial spending blueprint—have already complained about its penury. Boehlert and Representative Vern Ehlers (R-MI), another science committee leader, also raised concerns about the slim pickings for science during a 14 February meeting with White House Budget Director Mitch Daniels. “He was fully aware of the issue,” says one congressional aide.

Still, some science leaders worry that the grim budget outlook is another sign that the new Administration is not taking science seriously. Exhibit A, they say, is the absence of a science adviser. “Concrete gets poured fast on budgets,” so it is critical that researchers have a voice inside the White House, noted Jack Gibbons, who served as Bill Clinton's first science adviser, at a meeting last week of the National Academy of Sciences' public policy committee. D. Allan Bromley, who held the job in the first Bush Administration, agreed. “All of us hoped and believed there would be [an adviser] by now,” he said. But for the moment “there is no evidence that we will see anything for a few more months.” A source familiar with the search agreed that a selection “was not imminent” but said that “making quality appointments takes time.”

4. EUROPEAN SCIENCE

# Max Planck Takes an E-Publishing Plunge

1. Vivien Marx*
1. Vivien Marx is a science writer who lives in Boston and Cologne, Germany.

In the evolution from the Gutenberg era to the so-called Ginsparg era of electronic publishing, European researchers have mostly watched from the sidelines. But big changes are afoot. Germany's Max Planck Society, which runs the country's flagship network of 80 research institutes, is about to launch a publicly accessible electronic publishing center that will enable its scientists to post published papers—and findings before they are peer-reviewed. The center will also develop new tools for information dissemination and electronic management, and negotiate on behalf of all the institutes to cut deals for cheaper access to electronic journals.

For years, the United States had a virtual lock on e-publishing. In 1991, Los Alamos National Laboratory's Paul Ginsparg shook up scientific communication with the world's first preprint server, where physicists post articles before they are peer reviewed. More recently, Harold Varmus, former director of the U.S. National Institutes of Health, created a stir among biologists by backing the launch of “PubMed Central,” originally conceived as both a free Internet archive of published papers and a preprint journal, although the latter idea is now on indefinite hold (Science, 14 July 2000, p. 223). Now European researchers are beginning to carve out territory in cyberspace (see table). The Max Planck's new Center for Information Management (CIM), which is just now getting under way, is one of the most comprehensive of these efforts.

View this table:

Previously, individual Max Planck institutes have dipped their toes in the e-publishing waters. In 1997, for example, the Institute for Gravitational Research in Potsdam began publishing a peer-reviewed Web journal of review articles in its field. Contributors pledge to keep their articles updated, hence the journal's name: Living Reviews (http://www.livingreviews.org/). “But the society [as an entity] has not been present” in the e-publishing arena, says Jürgen Renn, co-director of the Institute for the History of Science in Berlin.

CIM's roots can be traced to a pilot project in which the Max Planck Society in 1999 negotiated access to 1200 electronic journals and tallied usage and approval among its 3000 scientists. With a positive response to this initial foray into the e-world, Renn and his colleagues asked Rick Luce—director of Los Alamos's digital library initiative, Library Without Walls (lib-www.lanl.gov/lww/welcome.html)—to help organize a more ambitious effort: the CIM. A six-person team, primarily computer scientists, will staff the $1.5-million-a-year operation, which will be based at the Institute of Plasma Physics in Garching. CIM plans to create an archive of publications by Max Planck authors. But perhaps the most exciting of CIM's projects will be to post preprints of articles from all Max Planck scientists on a server freely accessible to the public before they are submitted to traditional journals for peer review. The society does not know when the preprint operation will be up and running. Some Max Planck researchers are skeptical that the society will be nimble enough to keep pace with the rapidly evolving world of Web publishing. Others caution that the CIM must take care—particularly with its preprint server—to label the sources of the material it distributes electronically. “It will be important to differentiate between peer-reviewed papers and what amounts to private communication that a scientist may post,” says Peter Fromherz of the Institute for Biochemistry in Martinsried. As the society gets its act together, the U.S.-based Open Archives Initiative (http://www.openarchives.org/) will hold a meeting in Berlin next week on creating the architecture necessary to link emerging European preprint archives, including the one at Max Planck, with those in the United States. CERN, the European laboratory for particle physics near Geneva, will get in on the action too next month with a meeting, co-sponsored by Open Archives, on archive melding. A decade late, perhaps, but the e-publishing revolution is finally crossing the Atlantic. 5. NEUROBIOLOGY # A Discriminating Taste for Bitter 1. Kathryn Brown* 1. Kathryn Brown is a writer in Alexandria, Virginia. Life has many bitter moments—sometimes of the culinary kind. Now, a new study suggests that our taste cells are much better at distinguishing between bitter flavors than some researchers have thought. On page 1557, University of Miami biologists Alejandro Caicedo and Stephen Roper report that—contrary to one popular theory—taste buds recognize the many unique bitter flavors that land on your tongue. Your mouth, they say, knows the bitter of beer from a bitter pill any day. “In terms of evolution, this work makes good sense,” says Sue Kinnamon, a neurobiologist at Colorado State University in Fort Collins. “It suggests that bitter taste perception involves multiple cells and mechanisms.” This could be important, she adds, in a world with many different toxic compounds, which tend to taste bitter. Indeed, a well-developed system for recognizing bitters could enhance survival. Although there are five basic tastes—sweet, sour, salty, bitter, and umami (MSG)—researchers so far have identified the receptors for only umami and bitter. Taste has been tricky to study, because scientists don't know how to grow taste cells in the lab. Indeed, the bitter receptors were discovered just last year by two groups, one led by Nicholas Ryba at the National Institute of Dental and Craniofacial Research in Bethesda, Maryland, and Charles Zuker at the University of California, San Diego, and the other by Linda Buck of Harvard Medical School in Boston. That work showed that the bitter receptor family consists of 50 to 100 related proteins, each of which seems to respond to a different bitter flavor. Because Ryba and Zuker's group found that individual taste bud cells express the genes for most of the receptors, they concluded that the cells couldn't discriminate between the many different bitter compounds they encounter. In this scenario, cells would send the same “bitter” signal up to the brain no matter what. But Caicedo and Roper weren't so sure that all bitters taste the same. To pursue their hunch, they decided to catch the taste cells in action. When a receptor is activated by a bitter compound or other stimulus, it triggers a spike in calcium concentrations inside the cell, which in turn causes the cell to release its neurotransmitter. So first the researchers injected a fluorescent marker of calcium activity into taste cells taken from a rat's tongue. They reasoned that if the cells could distinguish between bitter flavors, some bitters would cause the telltale calcium boost—and an accompanying rise in fluorescence—while others would not. Then, one at a time, Caicedo and Roper added five common bitter compounds—cycloheximide, denatonium benzoate, quinine hydrochloride, sucrose octaacetate, and phenylthiocarbamide—to the solution bathing the marked taste cells. Sure enough, 65% of the cells fluoresced strongly in response to just one of the bitter compounds. About 25% of the cells responded to two compounds, whereas just 7% reacted to three or more of the bitters. Cell responses to the different bitters also varied in amplitude, length, and sensitivity. “It appears that different taste cells are tuned to different bitter compounds,” says Roper. “These cells are not generalists, as some suggest.” At this point, however, the researchers can't explain the specificity of the taste cells' responses, given that each one makes so many different bitter receptors. Even so, says David Smith, a neurobiologist at the University of Maryland, Baltimore, the study moves the field of bitter taste perception past molecular guesswork into real-time physiology. “I think the use of calcium imaging to visualize many cells at once is a big step in the right direction,” Smith says. The study, he adds, clearly refutes the idea that a given taste-receptor cell responds to many different bitter flavors. Ryba cautions that questions linger, however. Although the study is provocative, he says, its conclusions “go further than the data allow.” For one, calcium imaging is an indirect—and somewhat imprecise—measure of bitter receptor activity. What's more, he adds, the pattern of taste-cell responses might change if more bitter compounds were tested. “When you look at five compounds, you may not see much overlap in taste-cell activity,” Ryba remarks. “When you look at 25, that overlap might be considerable.” Caicedo and Roper agree that they focused on the “big bitters,” or most common bitter compounds, but they predict their results will hold up in further studies. They note that the amount of a bitter compound needed to provoke a calcium response in their test cells correlates with the amount that affected rat behavior in previous tests—an indication that the result reflects what's happening in living animals. “We're very interested in expanding this work,” Caicedo adds. “We have a lot of questions still to answer.” 6. WOMEN IN SCIENCE # Court to Hear Charges by Harvard Researcher 1. Andrew Lawler BOSTON—Harvard University goes to court next week to defend itself in a sex-discrimination suit brought by a researcher at its school of public health. Barring an unexpected last-minute settlement, it would be the first such case against Harvard brought to trial by a scientist and only the second such case to be heard by a jury. It will also shine a spotlight on Harvard Provost Harvey Fineberg, who attended last month's meeting on equitable treatment of women at elite U.S. research universities (Science, 2 February, p. 806). The suit, by biomathematician Tamara Awerbuch-Friedlander, alleges that the school refused to promote her because of her sex and then harassed her for complaining about that decision. Fineberg, who was dean of the school of public health until 1997 and is seen as a strong candidate to succeed retiring Neil Rudenstine as president, declined to comment on the case. But he disputed Awerbuch-Friedlander's account in a 1998 deposition, explaining that “there were controversies over the qualifications of the candidate” and that the field of biomathematics—her specialty—”did not appear to have sufficient priority for a faculty appointment.” Awerbuch-Friedlander arrived at Harvard in 1983 as a postdoc from the nearby Massachusetts Institute of Technology. In 1989, an internal panel recommended 4 to 1 that she be given a tenure-track assistant professor job in the biostatistics department. But Fineberg overruled the internal committee's recommendation for an appointment—by his own account a very rare occurrence. And a biomathematics position never materialized. Awerbuch-Friedlander still works as a lecturer at Harvard, currently supported by a small grant from an outside foundation. Several women faculty members at the school declined to comment on the case but praised Fineberg as a positive force for change as dean. “He's been wonderfully supportive of women,” says molecular biologist Leona Samson. “When Harvey came, there were practically no tenured women, and by the time he left there were three female department chairs,” says another. Biologist Bruce Demple, chair of an internal committee on the status of women, says that “the school has been willing to commit resources to recruit female faculty.” In his deposition, Fineberg said that 34% of 125 faculty searches conducted during his 13-year tenure ended with the hiring of a woman. “I believe the situation has been rectified,” he added. Awerbuch-Friedlander paints a different picture. She says that after the tenure decision, Harvard cut off her phone, warehoused her office materials, and refused her requests for administrative support because she was a woman. In court filings, Harvard officials acknowledge some of the events described but say that their actions did not constitute harassment and that there was no pattern of sex discrimination. “We didn't have a job for her, and she didn't get the message,” says one Harvard faculty member. Although they declined to comment on this case, several women at the school say that there are gender-related problems. Julia Walsh, a former health professor now at the University of California, Berkeley, says she left several years ago in frustration over gender issues; another tenured faculty member is about to do the same. And Demple says that he was disappointed in the lack of response to a 1996 study on promotion rates by his panel, which has been hobbled by the reluctance of senior women to participate. In 1993, Awerbuch-Friedlander switched to the population and international health department, and the next year she filed a complaint with the Massachusetts Committee Against Discrimination, which was rejected by the committee. In 1997, she filed suit in the Middlesex County Superior Court. The trial is set to begin on 26 February. Neither side expects an out-of-court settlement, although court filings by Harvard describe a$100,000 offer that Awerbuch-Friedlander refused. Court documents show that she is asking for a guaranteed 5-year position as senior lecturer, $550,000 in lost wages, and$200,000 in lost benefits.

7. ASTROPHYSICS

# Cluster Reveals Earth's Rippling Magnetic Field

1. Barbara Casassus,
2. Alexander Hellemans*
1. Casassus is a writer in Paris;
2. Hellemans is a science writer in Naples.

PARIS—Four satellites flying in unison have revealed a hidden wild side to Earth's magnetosphere, the magnetic field enveloping the planet that acts like a gigantic deflector shield against blasts of solar radiation. The unprecedented view, unveiled here last week at European Space Agency (ESA) headquarters, could help scientists devise better defenses against crippling magnetic storms.

ESA launched the quartet of identical spacecraft last summer, 4 years after the original set of satellites was lost in an explosion seconds after lift-off (Science, 28 June 1996, p. 1866). The satellites of the resurrected mission—nicknamed Salsa, Samba, Rumba, and Tango—each carry 11 instruments designed to produce the first three-dimensional maps of the magnetic fields and plasmas surrounding Earth.

Project scientists are thrilled with the data so far. “We can see things we couldn't possibly see before,” says André Balogh of Imperial College in London, the principal investigator of the fluxgate magnetometer experiment. The magnetometer has two sensors on each craft that measure the intensity and orientation of Earth's magnetic field lines. Outside experts also are impressed. “I am surprised the team has been able to extract such exciting observations so soon after launch,” says Alan Gabriel of the Institut d'Astrophysique Spatiale in Orsay and president of the French sun-Earth research program.

The Cluster spacecraft began gathering data soon after crossing the magnetopause—the outer edge of the magnetosphere, where the influence of the sun's magnetic field takes over—on 8 November. Chancing upon one of the most violent solar storms in 25 years, the spacecraft watched a barrage of particles from the sun, carried on a supercharged solar magnetic field, compress the magnetosphere to about half its usual size. It was the first time that this phenomenon has been measured in detail, says Cluster project scientist Philippe Escoubet of the European Space Research and Technology Centre in Noordwijk, the Netherlands.

More data came pouring in last month when the satellites crossed a polar cusp, a funnel-shaped gap in the magnetosphere through which charged solar particles reach the atmosphere and set off the northern and southern lights. Refuting the classic view of polar cusps as relatively stable, the satellites found the northern cusp gyrating wildly like a top, moving at speeds of up to 30 kilometers per second.

The Cluster spacecraft have found that the magnetopause, thought to be smooth, is actually corrugated and undulates like an ocean wave buffeted by wind, says Nicole Cornilleau-Wehrlin of the Centre d'Etude des Environnements Terrestres et Planétaires in Vélizy. “For years, we had been trying to find out what happens to this shield,” says Cornilleau-Wehrlin, whose instruments on the Spatio-Temporal Analysis of Field Fluctuations experiment detected waves in the magnetosphere that extended for 1000 kilometers and rippled along the magnetopause away from the sun—the first proof that these waves exist, says Escoubet: “That was not possible with a single spacecraft.”

Cluster's findings could soon have some practical benefits as well. The sun is entering the peak of its 11-year cycle of activity, which is expected to bring powerful solar flares that trigger magnetic storms in Earth's atmosphere. Such storms can disrupt radio and satellite communications. “Cluster is well positioned at the most complicated phase of the solar cycle to try and work out what the solar storms do to the magnetosphere,” says Balogh. A better understanding of these processes, he says, could lead to the development of early warning systems that would enable satellite operators to shut off their equipment before electrical circuits are damaged.

8. PLANETARY SCIENCE

# Strange Doings on a NEAR-Struck Asteroid

1. Richard A. Kerr

LAUREL, MARYLAND—Researchers here are puzzling over the last pictures returned by the NEAR Shoemaker spacecraft as it descended to its final resting place on the surface of asteroid Eros. At a press conference here last week at Johns Hopkins University's Applied Research Laboratory, team members showed pictures that reveal that something—no one knows quite what—is shaping the surface of Eros into bizarre “ponds” with “beaches” marked by “footprints.” Something else is populating the surface with boulders. “Our jaws are just hanging out,” says NEAR imaging team member Clark Chapman of Southwest Research Institute in Boulder, Colorado.

Never designed to land, NEAR Shoemaker made “perhaps the softest [planetary] landing ever” on 12 February in what was expected to be a mission-ending descent to the surface of the 33-kilometer-long asteroid. To the surprise of everyone, the spacecraft continued to beam a radio beacon back to Earth after touchdown. Telemetry was still being received 2 days later, prompting NASA to extend the mission for up to 10 days. With the barrel-shaped, half-ton spacecraft apparently propped on two solar panels, the gamma ray spectrometer was fired back up in hopes of refining the surface-composition measurements made from Eros orbit, according to spectrometer team leader Jacob Trombka of NASA's Goddard Space Flight Center in Greenbelt, Maryland.

NEAR Shoemaker's picture-taking days are over, because its telephoto lens is nearly in the dirt. But the last images it sent back should keep planetary geologists busy for years. “I never would have imagined you'd see some of these things on an asteroid,” says Chapman.

The mysteries start with an abundance of huge boulders—perhaps a million of them larger than 8 meters—visible on the surface. One explanation being considered by planetary dynamicist Erik Asphaug of the University of California, Santa Cruz, and his colleagues is seismic shaking: Large impacts might so shake Eros that the surface debris would settle like mixed nuts in a can, with the big, heavy bits rising to the top and the smaller ones falling to the bottom. This “Brazil-nut effect” might have caused boulders completely buried in the surface debris to rise into view, they say.

Another mechanism probably accounts for how the very finest material not only separated out but found its way to low spots, notes imaging team leader Joseph Veverka of Cornell University. Somehow, the finer looking material has filled in low spots to form flat deposits resembling ponds. A transition zone from smooth pond to rougher surroundings resembles a beach, notes Chapman. And in the last image returned by NEAR Shoemaker from an altitude of 125 meters, some spots the size of a footprint—though irregular in shape—seem to have collapsed a few centimeters, as if the fine material was somehow compressed.

One way fine material might move around would be for sunlight to charge it up electrostatically, notes Veverka. That could levitate dust and allow it to move downhill, as happens on the moon to an inconsequential extent and as has been suggested for Jupiter's moon Callisto. But nobody's betting on the accuracy of any of these theories. “We're facing processes we're not familiar with,” says Veverka. “I truly don't know what's going on.”

9. PALEONTOLOGY

# Whiff of Gas Points to Impact Mass Extinction

1. Richard A. Kerr

Two hundred fifty-one million years ago, as the Permian period gave way to the Triassic, Earth experienced its greatest mass extinction ever. Ninety percent of all marine species, including the last of the trilobites, disappeared, while on land pervasive extinctions opened the way for the rise of the dinosaurs. But despite the magnitude of this “mother of all mass extinctions,” its cause has remained mysterious.

A new analysis of rock that marks the Permian-Triassic (P-T) extinction now suggests that it was caused by the hypervelocity impact of an asteroid or comet similar to the one thought to have killed off the dinosaurs 65 million years ago. The evidence that some catastrophe triggered the P-T extinction has been building for the last 5 years. Although it was once thought to have lasted for 8 million years, it now appears to have occurred in a geological heartbeat—perhaps even instantaneously. So sudden does the extinction now appear, in fact, that many paleontologists presume it had a single, abrupt cause—a mega-volcanic eruption, a catastrophic release of toxic chemicals from the ocean's depths, or an impact. But no one had been able to implicate such a catastrophe by placing it at the geologic moment of extinction.

That's where the new work comes in. On page 1530, geochemists Luann Becker of the University of Washington, Seattle, Robert Poreda of the University of Rochester in New York, and their colleagues report that they have detected the noble gases helium and argon apparently trapped in the molecular cages of carbon “buckyballs,” or fullerenes, extracted from rock laid down at the P-T extinction. Analysis of these gases shows, the researchers say, that their isotopic compositions are much more like those found in meteorites than on Earth. Thus, they conclude that a giant impactor delivered the chemicals to Earth just when the extinctions occurred.

Some researchers find the argon and helium analyses persuasive. “It's the noble gases that make the case” for an impact, says physicist Robert Pepin of the University of Minnesota, Twin Cities, who works on noble gases in meteorites. Still, claims of finding buckyballs—closed lattices made of nothing but 60 or more carbon atoms—in natural samples such as impact debris and meteorites have been controversial. Indeed, the suggestion that they provide a marker for a P-T impact recalls the early days of the controversy over the impact at the Cretaceous-Tertiary (K-T) boundary, 65 million years ago.

The first clue to the K-T impact, discovered in 1979, was an abundance of the element iridium at the geologic instant of the mass extinction. Because iridium is plentiful in meteorites, the iridium-rich deposit suggested impact debris, but some researchers argued that the layer could instead have been produced by the iridium-rich exhalations of volcanoes.

Fullerenes are also proving to be a suggestive but unconvincing impact marker. Previous work by others showed that they are present in rock at the K-T boundary, a finding confirmed by Becker and her colleagues, who also detected them in two meteorites. Together these findings suggested that fullerenes are impact markers like iridium. That prompted Becker and her colleagues to look for the compounds in rock at the P-T boundary at the classic site at Meishan, South China, and at Sasayama in southwest Japan. The researchers did in fact detect fullerenes in boundary rock, but not in similar rock a few centimeters to meters above or below the boundary. But fullerenes can have more mundane sources than meteorites. They are produced by forest fires and even by the mass spectrometers used to separate and identify them.

In the case of the K-T mass extinction, the clincher was the discovery of shocked quartz, distinctively veined crystals made only in the extreme pressures of large, hypervelocity impacts. Shocked quartz has not been confidently identified at the P-T, but noble gases may yet serve to make the case. Because of their structure, fullerenes can trap gas atoms like birds in a cage. When Becker and her colleagues then analyzed the gases trapped in fullerenes from P-T-boundary rocks, they found that the abundance of helium-3 jumped 50-fold above what it was above or below the boundary. The ratio of helium-3 to helium-4 entrapped there was typical of that found in meteorites—not in earthly atmosphere and rock. And the ratio of argon-40 to argon-36 in boundary fullerenes is well below that of air and approaches that of meteorites. The recovery of such fullerene-encapsulated gases, says Becker, is “the best case for an extraterrestrial event coincident with the P-T extinction.” And she adds, “it was likely the trigger.”

Researchers who study fullerenes aren't so sure. “The [fullerene] work of Luann Becker and colleagues has been a bit controversial,” notes microscopist Peter Harris of the University of Reading, U.K. “Some people have found it hard to accept that fullerenes can survive for billions of years.” Although Becker claims to have detected fullerenes in two other impact deposits and in two meteorites, he notes, only the K-T fullerenes have been found by an independent group, despite a number of searches. Still, Harris has recently reported that transmission electron microscopy reveals what look like fullerene molecules in a meteorite, so he is “fairly convinced” that the fullerenes and noble gases mark an impact at the P-T. But “I'm perhaps in a minority,” he says.

Among noble gas workers, the reception has been warmer, however. Geochemist Kenneth Farley of the California Institute of Technology in Pasadena calls the anomalously low ratio of argon-40 to argon-36 “astounding. I can't imagine how you could have any other interpretation” than an impacting meteorite that carried in the noble gases. “There appears to be an extraterrestrial component in the [P-T] boundary layer,” agrees Pepin. “I think they've demonstrated that rather convincingly.” Still, even noble gas workers want to see more. “This result needs to be replicated by somebody else,” says Farley, “as any such measurement does.”

10. ASTROPHYSICS

# New Headaches for U.S.-Russia Experiment

2. Andrey Allakhverdov*
1. Pokrovsky and Allakhverdov are writers in Moscow.

MOSCOW—A tug-of-war over 60 tons of precious gallium is threatening to undermine a major neutrino experiment. This week, officials at the Baksan Neutrino Observatory in Prielbrusye are asking a court to stop a government order to sell off some of the liquid metal, which serves as an underground detector. It's the latest round in a long-running battle over the fate of the material, which has a market value of $500 to$600 per kilogram.

A child of the Cold War, the $60 million Soviet-American Gallium Experiment (SAGE) is one of the largest collaborations between Russia and the United States. Its 60-ton gallium detector sits in a mine shaft in the Caucasus, deep below Mount Andyrchi. Run since the mid-1980s by Moscow's Institute for Nuclear Research, the detector studies neutrinos streaming from the sun. Low-energy neutrinos can transform gallium nuclei into germanium-71 atoms, which are extracted and counted. SAGE is best known for confirming an unpredicted shortfall of solar neutrinos. The tussle over the silvery white metal began in 1997, when the Ministry of Fuel and Power Production asked the Cabinet for permission to sell the gallium, at a third of its market value, to Russia's State Research, Development, and Design Institute of Rare-Metal Industry (GIREDMET) plant in Moscow. It presumably would resell the gallium to foreign buyers—it's used in gallium-arsenide semiconductors—and reap the profits (Science, 11 April 1997, p. 193). SAGE officials caught wind of the impending gallium grab and organized a protest letter from 12 Nobel laureates to then-Prime Minister Viktor Chernomyrdin, followed by an appeal from U.S. Vice President Al Gore. The strategy worked: Chernomyrdin halted the transaction. Later that year, however, a deputy prime minister decreed that at least 7 tons of gallium should be handed over to the fuel ministry. Project scientists resisted, arguing that the detector's sensitivity would be so diminished that the experiment would no longer be worth running. (It's slated to continue through next year.) Shortly after, thieves bungled an attempt to break into the observatory and steal the gallium (Science, 14 November 1997, p. 1220). Last summer, President Vladimir Putin told Prime Minister Mikhail Kasyanov to review that order, which had not been carried out. In the meantime, the fuel ministry and the GIREDMET plant lodged a complaint against Baksan, arguing that observatory officials were interfering with efforts to procure and sell 7 tons of gallium. In December, the Arbitration Court in Moscow ruled for the plaintiffs; a hearing on the observatory's appeal was scheduled to begin on 22 February. But the GIREDMET plant isn't waiting for the court's decision, which could take weeks. Earlier this month, GIREDMET experts, who accuse Baksan officials of “squandering,” or hoarding, the gallium, showed up to measure the metal while escorted by local police. The process involves removing the liquid from its tank and weighing it. Partway into the exercise, however, the GIREDMET team gave up and read the calibration marks on the tank. Vladimir Gavrin, Baksan's director, believes that he has smoothed things over with local authorities: “Very soon, the militia understood that there was no squandering of the gallium, and we started to treat each other with respect.” But he can't say the same for the GIREDMET staff, who he claims were intent on finding some infraction that could be used to justify the gallium's removal. If the observatory loses its appeal, Gavrin says that his last hope is a government decision to rescind the order. 11. ENDANGERED SPECIES # West's Energy Woes Threaten Salmon Runs 1. Robert F. Service The combination of a dry winter and a power shortage could be bad news for endangered salmon in the Pacific Northwest. Last week, California's energy crisis forced the Bonneville Power Administration (BPA), the region's energy supplier, to exceed federal guidelines for the release of water through its turbines. But with reservoir levels already low, the utility might not have enough water available this spring and summer to help juvenile salmon on their run to the sea. “What we see time and time again is that when the going gets tough, fish take it on the chin,” says Rob Masonis, who heads northwest conservation efforts at American Rivers in Seattle. “That's untenable and irresponsible. We need a real commitment to salmon recovery in the region, not just a few museum fish in the river,” he says. BPA spokesperson Dulcy Mahar concedes that the spring water releases may fall short, but says that the agency has no choice. BPA is required to supply power to its customers. In this case, releasing extra water was the cheapest way to do it. “We are seeking to appropriately balance the needs of fish and electricity consumers during a serious drought,” says acting BPA administrator Steve Wright. In normal years, BPA buys power from California suppliers during the cold winter months, when demand peaks in the Northwest, and sells it back to California in the summer, when demand peaks there. This year, however, California hasn't had a megawatt to spare. What's more, because of low rainfall and smaller-than-normal mountain snowpacks, BPA's system of 29 federal dams has been able to generate only about 80% as much power as usual. The agency has been forced to buy the excess at market rates, at up to 10 times the usual price, putting a big dent in reserves earmarked for repaying its federal mortgage. Says Mahar: “The stability of BPA is at risk.” Ironically, these developments come on the heels of a decision in December that was intended to balance energy and environmental needs. Choosing not to back a plan to breach four dams on the Snake River to aid salmon recovery, President Clinton instead ordered several agencies to coordinate efforts to increase water releases from reservoirs in the spring and summer. The releases were meant to be part of an overall plan to speed and cool rivers to aid fish migration, restore damaged habitat, limit fishing, and prevent the overproduction of hatchery-reared fish, which can replace wild stocks. It's now unlikely that those spring guidelines will be met. After declaring an energy emergency twice this winter, BPA has increased water flows at some dams by as much as 60%. Although the need for excess releases should end as winter ebbs, Mahar says that they may reduce springtime river flows by 1.5%. BPA fisheries biologist Bill Maslen doesn't think that the small drop in flow will have much effect on juvenile salmon migration. But Chris Ross, a fisheries biologist with a National Marine Fisheries Service (NMFS) office in Portland, Oregon, says they make an already bad water year even worse for the salmon. “We're in the thick of trying to figure out what it means,” says Lynn Krasnow, another fisheries biologist at NMFS. Even a season of good rains, however, is unlikely to make the problem evaporate. It will be years before power plants fueled by natural gas, now under construction in California, Oregon, and Washington, come on line. That leaves hydropower with the burden of filling the energy demand for a growing region—and of keeping its salmon population afloat. 12. HIGH-ENERGY PHYSICS # B-Meson Factories Make a "Number From Hell" 1. Robert Irion SAN FRANCISCO—Humanity—and everything else in the universe—exists because matter and antimatter forged in equal amounts during the big bang may have decayed into slightly different sets of particles, giving matter a competitive edge. This tiny imbalance of one part per billion arose from a process called charge-parity (CP) violation, and there's a vigorous debate among particle physicists about its origin. New data reported here last week* at the annual meeting of the American Association for the Advancement of Science are at odds with the imbalance predicted by the reigning model of particle physics—but not by enough to settle the argument. “It's the number from hell,” says Stewart Smith, a physicist at the Stanford Linear Accelerator Center (SLAC) in California, home to one of the two experiments. Physicists discovered a simple form of CP violation in 1964 within the decays of K mesons, which are short-lived mixtures of matter and antimatter. For the last 2 years, teams at SLAC and the High Energy Accelerator Research Organization (KEK) in Tsukuba, Japan, have probed for a deeper signal of CP violation in B mesons, the heavy brothers of K mesons. Special machines dubbed “B factories” create tens of millions of B mesons by smashing electrons into their antimatter counterparts, positrons. However, only about one out of 10,000 collisions are “golden events”—pairs of B's and anti-B's that spawn an easily measurable spray of certain mesons and offer the clearest signature of CP violation. As of January, physicists had seen 630 such events at SLAC and 260 at KEK. That's enough for a preliminary analysis, reported SLAC physicist Patricia Burchat. The Standard Model, which describes nature's basic particles and their interactions, predicts that the dimensionless value of CP violation should be 0.72 on a scale from −1 to 1, in which 0 represents symmetry between matter and antimatter. SLAC's value to date is 0.34, but the error range is large: ± 0.20. That means there's a 5% chance (twice the error bar, or two standard deviations) that the real value could match the prediction of the Standard Model, but it also could be 0. “It's not the most exciting of possible values,” admits Burchat. KEK's preliminary number, 0.58, is closer to the Standard Model value, but with a bigger error bar of ± 0.33. Both teams presented their results in more detail this week at a conference in Ise-Shima, Japan. However, a “meta-analysis” of all B-meson decays in the world to date offers some intriguing results. After combining data from SLAC, KEK, and other facilities and weighting them according to their errors, Burchat derives a value of 0.48 ± 0.16. “That just squeaks in at three sigma [standard deviations] above zero,” she says. But neither of the two B-factory teams can make that claim by itself, says physicist Chris Quigg of the Fermi National Accelerator Laboratory (Fermilab) in Batavia, Illinois. “The first people who do that will have a party,” he observes. Despite the uncertainties, other physicists applaud their colleagues' rapid progress. “It's starting to get interesting,” says Fermilab theorist Joseph Lykken. “We're almost at the point of challenging the Standard Model and its explanation of CP violation.” Theorist Michael Dine of the University of California, Santa Cruz, had hoped for something else: “It's depressing. I desperately wanted it to be 0.” That result, far out of whack with the Standard Model, would have worked well with a sweeping but untested theory of particles and forces called supersymmetry, Dine says. Burchat notes that imminent upgrades will lead to vastly improved statistics. For instance, SLAC's B factory has churned out 23 million pairs of B's and anti-B's so far, but physicists expect it to produce 80 million in 2002—far beyond the machine's initial goals. There is some urgency to do so: The new Tevatron accelerator at Fermilab may start spitting out billions of B pairs per year when it is turned on next month. However, the swarms of B's will be embedded within a complex tangle of other particles from the collisions of massive protons and antiprotons. That jumble will make the analysis far more complex than at SLAC and KEK—and in all likelihood add further fuel to the debate. • * 2001 AAAS Annual Meeting & Science Innovation Exposition, San Francisco, 15–20 February. 13. PAPAL SCIENCE # Science and Religion Advance Together at Pontifical Academy 1. Charles Seife Some 80 eminent scientists convene biennially to bend the ear of the pope—and help shape global attitudes toward new technologies VATICAN CITY—The Casina Pio Quattro is a distracting location for a scientific meeting, to say the least. Outside the templelike structure, high on a hill behind St. Peter's Basilica, immaculate gardens teem with squawking parrots. Inside, gleaming marble and brilliant frescoes glorify God and the heirs of St. Peter. A statue of the goddess Cybele, temporarily hidden in a thicket of scaffolding, graces the Casina, whose walls are adorned with quotations from Seneca and Cicero. Pontifical Academy of Sciences Created: 1936, by Pope Pius XI Number of members: 80 Purpose: “To promote the progress of the mathematical, physical and natural sciences and the study of epistemological problems relating thereto.” Recent publications: • The origin and early evolution of life • Geosphere-biosphere interactions and climate • Food needs of the developing world in the early 21st century The pagan motifs are fitting, because the Casina—erected in the 1550s as a residence for Pope Pius IV—now serves as a meeting place for the religious and secular worlds. It is the headquarters for the Pontifical Academy of Sciences: 80 esteemed scientists appointed for life to make their cumulative collective wisdom available to the pope. Members run the gamut of disciplines, backgrounds, nationalities, and religious beliefs. Twenty-five of them are Nobel laureates. Some of the most famous scientists in the world are members, such as physicists Stephen Hawking and Carlo Rubbia, astronomers Martin Rees and Vera Rubin, biologist David Baltimore, and numerous others. Every other year, the group gathers in the Vatican Gardens for a plenary session at which members hold forth on the state of science and the world, pass resolutions for improving the latter, and renew acquaintances. They have, by all accounts, a heavenly time. As member Joseph Murray, a 1990 Nobel laureate who performed the first kidney transplant, puts it, “Every day is like Christmas.” If so, the gift giving is mutual. The pope gets access to the scientific expertise of people at the top of their fields in astronomy, cosmology, genetics, and other areas that interest the church. In return, the scientists get the ear of one of the most important people in the world—and, through him, a chance to influence whether people accept or reject new knowledge and technology. In recent years, the academy has weighed in on both urgent practical issues—environmental concerns, neurological research, breast feeding, fertility —and deeper ones, such as the origins of life, the implications of genetics, and the formation of galaxies in the early universe. And it has been credited with influencing the church's positions on issues ranging from gene splicing to evolution. “All these things raise big, important philosophical questions,” says academy member Peter Raven, director of the Missouri Botanical Garden in St. Louis and new president of the American Association for the Advancement of Science (publisher of Science). “If you don't have good scientific advice, you're going to make all sorts of screwy statements. No organized body of people can operate efficiently without some input of science.” At its latest plenary session, held last November, members prepared to make an announcement that could shift the debate over one of today's most controversial science-policy issues, the use of genetically modified foods. “Broadly speaking, the Pontifical Academy is very similar to a national academy,” says member Charles Townes, inventor of the laser and winner of the 1964 Nobel Prize in physics. In some ways it's better, Raven adds. Its small size allows it to meet and make decisions as a whole, unlike most national academies, yet it also fields an unusually broad range of scientific expertise. With members drawn from 27 countries, the Pontifical Academy has served as the inspiration for other international academies. “The idea for the Third World Academy of Sciences was actually born in this academy,” says chemist C. N. Rao, a member of the Pontifical Academy and president of the Third World Academy. “This campus was its birthplace.” The Pontifical Academy itself traces its ancestry to the Accademia Lincei. Founded by the 18-year-old son of a duke in 1603, the Lincei was named in hopes that science would enable people to perceive nature with eyesight as keen as a lynx's. Over the centuries, the Lincei assumed numerous incarnations, both secular and pontifical. In 1936, during the reign of Mussolini, Pope Pius XI split the modern Pontifical Academy off from the Lincei, which survives as the Italian national scientific academy. Since its founding, the Pontifical Academy has numbered among its members such scientific luminaries as Alexander Fleming, Niels Bohr, Chandrasekhara Raman, and Werner Heisenberg (elected in 1955). (Prominent nonmembers have included Albert Einstein, James Watson, Francis Crick, and Wolfgang Pauli.) Candidates are nominated and elected by the members, although technically they are appointed by sovereign act of the pontiff—who looks for more than mere scientific eminence. The institution's charter specifies that members must possess “acknowledged moral personality.” Members say they don't know how the church verifies that clause. Townes says he found out about it when his nomination was under consideration in the early 1980s. A woman from the church phoned and told him that members were required to have good moral character. “How do you decide if I do?” Townes asked. “We have our ways.” Once on board, members share a simple set of duties: to meet and talk. Members say they choose their own speakers and topics and debate issues freely. “The Catholic Church is supporting this academy,” says Crodowaldo Pavan, a geneticist at the University of São Paulo and a member of the academy. “They pay for this meeting and don't say what we should say—they give us total freedom.” Yet religion does make its presence felt. “You have to be respectful of it to work in this body,” Raven says. At times, he acknowledges, “there is a little bit of tension there.” The academy's November session drew an eclectic mix of people: physicists, biologists, philosophers, theologians, and clerics. Dominican robes and Jesuit collars mingled with suits that had seen more than their share of chalk dust. From 8:30 in the morning until past six in the evening, members sat at dark wooden desks while their colleagues and guest speakers lectured from a podium at the open end of the chamber on such subjects as the biology of the brain, technology for predicting and mitigating natural disasters, the evolution of the universe, and how Christianity influenced Isaac Newton's discovery of his laws of motion. Speakers delivered their talks in English, French, and occasionally Italian; no interpreters were provided. The audience listened alertly to the scientific presentations, but many shifted uncomfortably in their seats during talks such as “Natural Theology in the Light of Modern Cosmology and Biology,” in which guest speaker Richard Swinburne, a philosopher at Oxford University, outlined a probabilistic argument for the existence of God. According to Pavan, this session, which coincided with the Jubilee—a spiritual celebration held every 25 years—featured an unusually high dose of theology. “We don't usually talk about religion,” Pavan said apologetically. What the Pontifical Academy does talk about, at length, is policy. Twenty years ago, for example, it left its mark on the debate about recombinant DNA technology. “There was a great deal of alarm at the time whether it was appropriate to use,” says Alexander Rich, a biophysicist at the Massachusetts Institute of Technology. But academy members, including Baltimore, drafted a statement supporting the use of recombinant DNA. The pope followed with his own statement giving the nod to genetic research. “Some said, ‘Let's not unleash this technology at all,'” says Sheldon Krimsky, a science policy scholar at Tufts University in Medford, Massachusetts, and a nonmember of the academy. “The Vatican's position on [recombinant DNA] helped to blunt ideological opposition to the use of recombinant DNA technology.” The academy has also long advised the pope about the science behind nuclear weaponry and the probable environmental effects of a nuclear war. Some observers credit papal diplomacy in the late 1970s with helping change the superpowers' nuclear doctrines. Herbert York, a physicist and the former head of Lawrence Livermore National Laboratory in California, who was active in formulating nuclear-weapons treaties at the time, says the pope was having difficulty developing a statement that would have any effect. The academy briefed the pope so that he could speak at a United Nations Educational, Scientific, and Cultural Organization meeting in Paris. “The speech, as far as I'm concerned, was a bust—it was highly formal,” adds York. But that doesn't mean that the Vatican had no effect. The Catholic bishops in various countries came out with several statements that were influential in reformulating nuclear policy. “The whole Catholic Church took a stance and wrote a number of documents that were seriously considered,” says York. The Vatican's antinuclear efforts continue to this day. “By picking our issues carefully, our views can be very influential,” Raven says. Although members may differ with church policy on specific issues, they quickly develop a sense of which agendas to push and which to soft-pedal. Contraception, for example, is out of bounds, but overpopulation is very much a live issue. “The pope has come out stating that population is a problem,” Townes says. “He didn't before.” (The church stresses education and general improvement of living standards as indirect means of controlling population.) Occasionally, however, the academy ventures into territory that the church finds sensitive. In 1987, for instance, members drew up a protocol for radiocarbon dating of the Shroud of Turin. Church investigators ignored key parts of the recommendations, and the shroud's provenance remains controversial today. The academy had better luck in the early 1990s, when it pressed the church to reexamine the case of the 16th-century scientist Galileo Galilei, whom the Inquisition tried for spreading the then-heretical idea that Earth moves through space. Three hundred and fifty-nine years after Galileo's conviction, the church admitted that its prosecutors had failed to “interpret with great circumspection the biblical passages that declare the Earth immobile,” although it faulted Galileo for making unproven assertions. The pope decried the “tragic mutual incomprehension” that had caused Galileo's imprisonment. “The church looked silly doing it so late, but they finally did it,” Rao says. “It is a nice thing.” In 1996, the pontiff took a further step toward shedding the church's ancient antiscientific taint when he declared evolution “more than just a theory.” Members say the Pontifical Academy deserves part of the credit for counseling the pope to make the statement and pushing him to reconcile the Catholic faith with Darwinian biology. At the November meeting, the academy continued its activist tradition by agreeing to draft a statement supporting the responsible use of genetically modified foods. The church has long been cautiously sympathetic to the use of gene technologies because of their promise for feeding the hungry in Third World nations, but it is wary of the ethical and theological consequences. “[Biotechnologies] cannot be evaluated solely on the basis of immediate economic interests,” Pope John Paul II told a gathering of farmers 2 days before he addressed the Pontifical Academy. “They must be submitted beforehand to rigorous scientific and ethical examination, to prevent them from becoming disastrous for human health and the future of the Earth.” Academy members hope that a carefully worded statement, expected to be issued this month, will prompt the pope to voice his own support for genetically modified foods. “This might have a lot of consequences for those living and working in developing countries—and it might make it possible for a country like France to continue working on this stuff,” says French physicist Paul-Marie Germain. “I regret that we have not said anything before.” At the end of the meeting, the church had the last word. Dressed in a scarlet tunic, cape, sash, and beanielike zuchetto, Cardinal Paul Poupard, president of the Pontifical Council for Culture, pronounced that “the right understanding of the links between science and faith is absolutely essential if scientists are to avoid foundering in dire straits.” He then invited the assembled scientists to join him in the study of Christ as the “supreme science of life.” Some of his listeners, however, privately demurred. Even devout members say they are wary of theological influence on science. “There is no scientific teaching in Genesis,” says George V. Coyne, a Jesuit priest, member of the Pontifical Academy, and director of the Vatican Observatory (see sidebar on p. 1473). Coyne decries attempts to invoke God as a catch-all explanation for mysteries in cosmology and evolution, or to cite science to prove God's existence. “The understanding of origins has nothing to do with the existence of God or not, but it has a lot to do with my understanding of God,” he says. Pavan draws a firmer line. “My point of view is that religion and science are two parallel worlds that should not try to cross. Put together, the two would not work,” he says. Yet the Pontifical Academy shows that they can at least move in the same direction. 14. PAPAL SCIENCE # Vatican Observatory Takes Long View of Exploring the Heavens 1. Charles Seife VATICAN OBSERVATORY, CASTEL GANDOLFO, ITALY—For more than 400 years, the astronomers of the Vatican Observatory have kept the church's gaze fixed on the heavens. Begun with the modest aim of tracking the course of the sun, the observatory has become the Vatican's eye on a few frontier areas of modern astronomy. The headquarters of the observatory is at Castel Gandolfo, the pope's summer residence in the Alban Hills several kilometers southeast of Rome. Father George V. Coyne, director of the Vatican Observatory, spends half the year here in his office atop the pope's bedroom (he jokingly warns visitors not to toss beer cans over the ledge) and half on Mount Graham in Arizona, since 1993 the home of the 1.8-meter Vatican Advanced Technology Telescope (VATT). The telescope was the first to sport a mirror made in a rotating furnace, which slopped the glass into the parabolic shape required, saving weight and material. It also drew fire from protesters who feared that its construction would threaten the habitat of an endangered red squirrel (Science, 14 July 2000, p. 228). The Arizona observatory is just the latest incarnation of a scientific enterprise that started within the Vatican walls. The first was the Tower of the Winds, a small, boxy tower that juts from a building overshadowed by the dome of St. Peter's. The tower—probably the most ornate scientific laboratory the world has ever known—was built in the 16th century by order of Pope Gregory XIII to help Jesuit astronomers assess the need for calendar reform. Its interior is one large, lavishly decorated scientific instrument. The walls are covered with frescoes showing the storm on the Sea of Galilee described in the gospel of Luke, as well as astronomically accurate paintings of the signs of the zodiac. On each wall, one of the four winds huffs away; by day, sunlight streams through a hole in the mouth of the south wind. By charting how the sun moved across the paintings and against a meridian line in the floor of the room, Jesuit astronomers calculated how far the old Julian calendar had strayed from astronomical reality, a discrepancy the pope corrected by wiping 10 days out of existence in October 1582. “The Jesuits did such a good job that they gave us spaghetti and wine and told us to keep doing it,” Coyne says. Church- sponsored astronomers have been scanning the firmament ever since. In 1868, Father Angelo Secchi became the first person to classify stars by their spectra, lumping them into four major classes. His method paved the way for current spectral classification schemes. By the end of the century, Pope Leo XIII had established an official Vatican Observatory on the Vatican grounds. In the 1930s, the astronomers moved to Castel Gandolfo in hopes of finding darker skies; 50 years later, they moved to Arizona. Today's Coyne-operated observatory fields a team of eight Jesuit astronomers and eight staff members; it shares its telescope with the University of Arizona, which has other telescopes on Mount Graham and at other sites. The topics studied there range from galactic tidal forces to Coyne's own specialty, cataclysmic variables—stars that undergo dramatic changes in brightness. Results appear in scientific journals. “They are a very good crew of professional astronomers,” says Mark Sykes of the Steward Observatory in Tucson, Arizona. “Though few in numbers, they contribute quite a bit, and they do a lot of public outreach.” In one long-standing line of research, Father Guy Consolmagno has spent decades trying to determine whether asteroids are solid chunks of rock or piles of gravitationally bound rubble, long a matter of debate (Science, 4 July 1997, p. 30). Using simulations, meteorite samples, and images from the VATT and other telescopes, Consolmagno—working with astronomer Daniel Britt of the University of Tennessee, Knoxville—has determined that some asteroids may be at least 35% empty space. They have speculated that asteroid 16 Psyche is more void than rock. The Vatican Observatory also grapples with issues that most astronomers shy away from. Coyne occasionally explores the theological implications of astronomical research in papers such as “Evidence for the Existence of Extra-solar Planets: Challenges for Religious Thought,” which appeared in 1999 in an anthology published by the Templeton Foundation (Science, 21 May 1999, p. 1257). 15. COMPUTER MODELS # Paleontologists Learn to Shake Up Virtual Bones 1. Erik Stokstad A sophisticated engineering tool is providing insights into how fossil skulls and other bones worked in real life In their dreams, some paleontologists wire strain gauges to the skulls of extinct killers and ram the skulls into chunks of meat to simulate a deadly attack. Understanding the stresses on the skulls might help clarify the ecology and evolution of predators. But fossil skulls are too rare to be sacrificed and are now made of rock, not living bone. Thanks to computing power and some engineering savvy, however, paleontologists have a safer way to turn their most precious fossils into crash-test dummies. The technique is called finite-element analysis. Engineers have used it for decades to investigate forces on an object, be it a bridge or a booster rocket. They recreate the structure with a mesh of small, linked polygons, then designate its elasticity and other material properties at each tiny site. Force is applied, and as each element nudges its neighbors, the computer solves simultaneous equations to calculate the compression and tension. By making a finite-element model and loading it in various ways, engineers can study how a structure might fail and use those insights to optimize its design. In the late 1970s, biomedical researchers began to use the technique to study, among other things, the biomechanics of bone and the strength of tooth implants. More recently, some biologists and paleontologists have begun to tackle questions about animal mechanics, such as the strength of horse hooves—and test ideas about anatomy, evolution, and ecology. In this week's issue of Nature, a theory about the hunting strategy of a dinosaur is backed up by a finite-element model of a fossil with unprecedented detail. “In the past, people just waved their arms” when talking about the biomechanics of ancient creatures, says Don Henderson, a functional morphologist and postdoc at the Johns Hopkins University School of Medicine who works on computer modeling in paleontology. “Now you can actually make it rigorous.” The first major effort to examine fossils using finite-element analysis began at Cambridge University in the United Kingdom. Paleontologist David Norman and his graduate student, Ian Jenkins, heard about the technique in 1997 from Jeff Thomason, a biologist at the University of Guelph in Ontario, Canada, who was presenting findings about the evolution of the mammalian hard palate. Jenkins went on to use the technique to study a group of saber-toothed reptiles, called gorgonopsids, that were among the first carnivores on land. The abundant gorgonopsids had apparently outcompeted other saber-toothed predators of the time, such as the therocephalians. Jenkins hoped to find their competitive advantage in some subtle feature of their skulls. To test his hypothesis, Jenkins made a simplified computer model of the gorgonopsid skull. Using data from dogs, he subjected the virtual skull to forces involved in attacking prey. The strain was concentrated along numerous sutures in the palate that acted like shock absorbers. These and other features would have absorbed the shock from the blow inflicted by their powerful jaws. Other saber-toothed reptiles sported similar peculiar sutures, it turned out, but theirs were less pronounced. Evidently, gorgonopsids were built to bite harder. “When that came out, I was hugely pleased,” Jenkins says, because it suggested a reason for their success: a wider diet that allowed them to better adapt to changes in the faunas during the late Permian, some 250 million years ago. Now a postdoc at the University of Bristol, U.K., Jenkins is using finite-element models to examine what filled the niche when gorgonopsids went extinct at the end of the Permian. Jenkins's simplified models only scratched the surface of the finite-element method. Emily Rayfield, a graduate student at Cambridge, is working on a virtual skull with nearly 250,000 elements. Her subject is a remarkably complete Allosaurus found in Wyoming in 1991 and dubbed “Big Al.” Like the skulls of many other theropod dinosaurs, an Allosaurus skull is a lattice of small, thin bones that looks both light and strong. Rayfield wants to understand how its 80-centimeter-long skull functioned as a tool and a weapon. To estimate how much force Big Al's jaw muscles would have exerted, Rayfield examined muscle scars on the bones and compared them with scars and muscles on living dinosaur relatives, such as birds and crocodiles. Then she reconstructed the muscles in clay on a cast of the skull. The strength of a muscle depends on its cross-sectional area and on the metabolic rate of the animal that flexes it. Rayfield measured the thickness of the clay muscles and, to be on the safe side, calculated muscle strength for a range of metabolisms, guessing that the true value lay somewhere in the middle. She found that Allosaurus's jaws were only a fourth as powerful as those of a modern alligator. “I was really surprised that it had such a weak bite,” Rayfield says. She infers that Allosaurus didn't crunch with the kind of bone-splitting bites that the thicker headed Tyrannosaurus rex seems to have used to dismember carcasses. Rayfield then used her finite-element model to test the strength of Big Al's skull. By applying force through six teeth in the virtual head, she found that the upper jaw and the rest of the skull could have withstood a load of up to 6 metric tons—26 times the maximum force from clenched teeth, Rayfield and colleagues report in the 22 February issue of Nature. Why such drastic overengineering? Rayfield suspects that the skull had to absorb such large forces when Allosaurus collided with its prey. She imagines Allosaurus running into a fleeing victim with jaws agape, slamming its upper teeth in like a hatchet and then using its strong neck muscles to rake out flesh with its teeth. “At first glance, it seems like it would be a weird approach to biting,” says Tom Holtz of the University of Maryland, College Park. But it fits with the observation, published last year, that Allosaurus could open its jaws extremely wide. “It's appealing to see that the mechanical analysis is consistent,” he says. Such consensus is reassuring, because paleontological modeling is far from an exact science. The material properties of fossil bones and the applied forces are all estimates, and figuring out muscle strength is just part of modeling a complicated motion. As a result, paleontologists say they are on guard against the “garbage in, garbage out” effect. “I wanted solid answers,” recalls Michael Fastnacht, a doctoral student at Johannes Gutenberg University in Mainz, Germany. “But when I went to the engineers, they said, ‘Oh, no, even we don't get those.'” What's more, there's rarely a way to test a theory short of building a scale model of a dinosaur skull out of modern bone. Still, for all the caveats, Fastnacht and colleagues are pressing ahead with plans to investigate the bony crests that grace the tops of some pterosaur skulls. The crests have been proposed as rudders for flying or buttresses to strengthen the snout during a bite. Other paleontologists talk about modeling foot bones of dinosaurs, and paleoanthropologists are interested in using the technique to study chewing and hip function in primates. It's not Jurassic Park, they acknowledge, but detailed answers from this high-tech approach may help bring these old bones to life. 16. SUPERCONDUCTIVITY # Material Sets Record for Metal Compounds 1. Robert F. Service Magnesium diboride, one of the simplest compounds around, superconducts at nearly twice the temperature of its closest metallic rival The discovery of ceramic superconductors in 1986 lit a fire under physicists worldwide. Experimentalists stayed up for nights on end concocting new ceramic mixtures and testing the results. Theorists jumped at the latest data and racked their brains trying to explain how the newfound ceramics could conduct electricity without any losses at temperatures far above those of the conventional metallic variety. Physics meetings took on the aura of jam-packed rock concerts. Now a new discovery, although clearly more modest, has the superconductivity community abuzz again. At a meeting last month in Japan,* researchers led by physicist Jun Akimitsu of Aoyama Gakuin University in Tokyo announced that they had discovered a boron-containing metal compound that superconducts at 39 K, nearly twice the temperature of the previous metallic record holder. Although some ceramics can superconduct at temperatures up to 96 degrees higher, most metallic compounds are better at carrying current across gaps between grains of material and thus make better wires. “This is the highest observed [superconducting temperature] of any intermetallic compound,” says Paul Canfield, a physicist at Iowa State University in Ames, and the Department of Energy's Ames Laboratory. “That's a big hairy deal.” Researchers have spent decades looking at boron-containing compounds for hints of superconductivity, because theory suggests that boron's light weight should give any compound it is in a relatively high superconducting temperature. Yet somehow they overlooked one of the simplest compounds around, magnesium diboride (MgB2), a tan powder that can be purchased from standard laboratory chemical suppliers. “I'm really amazed that they didn't find it before,” says Jorge Hirsch, a superconductivity theorist at the University of California, San Diego. “It's like putting cinnamon in your magnetometer and finding it superconducts,” marvels Canfield. In an e-mail exchange, Akimitsu declined to provide details of his team's discovery or explain why they decided to look at MgB2, because the team currently has a paper on the subject under review. In any case, the result has already been replicated by other teams in Japan, the United States, and the United Kingdom. Now physicists are racing to make sense out of MgB2's abnormally high superconducting temperature. Materials superconduct when electrons inside overcome their usual repulsion and pair up, with electrons effectively taking on the size of the paired structure. That property allows them to surf through a material's crystalline lattice without banging into atoms that would slow their progress. According to the “BCS” theory of superconductivity, first outlined in 1957 by John Bardeen, Leon Cooper, and Robert Schrieffer, this pairing occurs in metallic superconductors as a result of a kind of electronic water skiing: The movement of one electron creates vibrations in the surrounding atomic lattice that then sweep another electron along in its wake. But this link between electrons normally breaks when the temperature rises much above 20 K. The heat produces extra vibrations that act like rogue waves sending the trailing electron skiers careering in all directions. In high-temperature ceramic superconductors, electron pairing is widely thought to be due to the magnetic behavior of atoms in the material, although this remains in dispute. So figuring out what is keeping electron pairs together at nearly 40 K in the nonceramic MgB2 has become the latest contest in the most competitive area of materials physics. “Tally ho,” says Canfield. “The chase is on.” Adds Hirsch: “I can't sleep. It's extremely exciting.” Hirsch's insomnia may be brief, however. A series of early reports suggests that MgB2 is most likely a BCS superconductor, albeit a very good one. In one paper posted to the Los Alamos physics preprint server on 3 February and accepted for publication at Physical Review Letters, Canfield and his Iowa State colleagues present experimental evidence of its similarities to other intermetallics, most of which can be explained by the BCS theory. The Iowa State team found that the top superconducting temperature of MgB2 increases by a degree when they make the material with the lighter isotope of boron, boron-10, rather than boron-11. This “isotope effect” is a classic signature of a BCS superconductor. “This does not prove it's a BCS superconductor, but it supports it,” says Canfield. Other results are bolstering that support. Another Iowa State team—this one headed by physicist Doug Finnemore—reports in another Los Alamos preprint that other standard tests done on MgB2, which track the way materials conduct heat and behave in a magnetic field, show that the material's behavior closely resembles that of Nb3Sn, a popular intermetallic superconductor. And both Finnemore's team and one headed by David Larbalestier at the University of Wisconsin, Madison, report that MgB2 can transport large electrical currents between separate grains in the powdery material, again a behavior similar to Nb3Sn. Still, not everyone is ready to call off the hounds. For BCS theory, says Hirsch, MgB2 is “a big outlier,” and it's not clear what makes its electron pairs stick together at such a high temperature. No matter what the mechanism, MgB2 could generate an even greater buzz in the real world. Despite the hype that accompanied the earlier high-temperature superconductors, low-temperature metallic superconductors continue to dominate the applications arena, because these materials can be fashioned into wires that carry large currents. Among the metallic superconductors, niobium-based superconductors reign supreme, because wires made from it are durable and can carry huge electrical currents. Yet niobium is expensive, whereas magnesium and boron are cheap. Last week, Canfield and another team posted another preprint reporting that they've already made MgB2 wire filaments that superconduct up to 39 K. As a result, magnesium diboride could find itself the superconductor of choice for a wide range of applications, such as the wires that make up the high-field magnets in magnetic resonance imaging (MRI) machines. That could make this newcomer far more useful than its high-temperature cousins. • * Symposium on Transition Metal Oxides, 10 January, Sendai, Japan. 17. CHINESE ACADEMY OF SCIENCES # In China, Publish or Perish Is Becoming the New Reality 1. Ding Yimin* 1. Ding Yimin writes for China Features in Beijing. A new program is funneling money and resources to a chosen few who are found to be highly productive—but at the expense mainly of older researchers BEIJING—Being one of the chosen is paying off for Wu Xiangping. The 39-year-old astrophysicist at the Beijing Astronomical Observatory (BAO) received a fivefold boost in pay last year for his theoretical work on the existence and function of dark matter, which is funded by three government agencies. He also secured a long-term, renewable contract as a team leader and the authority to pick the rest of his research group. Things haven't worked out so well for a colleague, Li Xiaocong, who went through the same review that boosted Wu's salary and status. Two years ago, the 52-year-old researcher was forced to retire from the observatory's solar activities forecasting group and take a job at another institution that doesn't make use of her scientific skills. However, she's working on a paper that she hopes will restore her to the job she loves. Wu and Li are two of the 305 scientists at BAO who have been through the first phase of a decade-long self-improvement regimen at BAO's parent body, the Chinese Academy of Sciences (CAS). Launched in 1998, the Knowledge Innovation Program (KIP) is the academy's attempt to tame a sprawling empire of 123 institutes that for almost 50 years provided not just jobs but a lifetime social support system for its 40,000-member workforce. That cradle-to-grave approach became an anachronism as China moved toward a market economy, leading CAS President Lu Yongxiang to order institutes to shrink their workforces, shed their nonscientific functions, eliminate unproductive research programs, and focus on their best scientists, including luring back those from overseas (Science, 30 January 1998, p. 649). The shake-up is doing exactly what it set out to do: Create a cadre of elite researchers like Wu by channeling new funding to a chosen few. The first phase of the reforms has led to a 50% cut in staff at 76 targeted institutes. The second phase aims at a slimmed-down CAS of 20,000 researchers spread across 80 institutes. And it doesn't end there: KIP has also enshrined the principle of ongoing performance reviews, with the lowest performers—especially those older than 50—getting the boot. At the same time, some scientists complain that the quantitative measures being applied are too rigid to span all types of scientific activity. A key element in the reform is the government's promise to spend more on those who are the most productive. CAS has allocated nearly$600 million for phase I, a sharp increase over previous spending rates, and has budgeted $1.2 billion over 10 years, a figure that could grow. The support has meant an extra$15,000 per researcher, a princely sum. With one-quarter going for salaries, scientists in the program can earn three to four times more than co-workers not chosen for KIP. (Under the reform, some CAS researchers have kept their jobs and receive small amounts of money from their institutes' regular research budgets but nothing from KIP.)

In addition to selecting for quality, CAS hopes to deepen the talent pool by bringing in more young scientists. By 2005, the academy plans to create positions for 20,000 graduate students and 5000 postdocs and visiting scholars. That would be a big jump from current levels of 12,000 graduate students and 1500 visiting scholars. It also plans to invite 500 outstanding young scientists from abroad to work as long-term CAS employees. If deemed a success, the approach is likely to be copied by other scientific institutions, universities, and high-tech businesses.

## The chosen

Although CAS determines the overall guidelines that each institute must follow, individual institute directors have great leeway in implementing the reforms. Most use some form of objective review, combined with age guidelines.

For instance, officials at a key national lab of the Institute of Zoology in Beijing began annual evaluations of staff based on the number of projects worked on, research grants obtained, international conferences and collaborations, students being advised, and papers published in Chinese and foreign journals tracked by the Science Citation Index (SCI). The last, a compilation of the Institute for Scientific Information in Philadelphia, is seen as an external measure of quality. “This system can help us judge a researcher's work more objectively and fairly,” says Li Dianmo, the lab's director. Li says it's also the best way to determine future eligibility. “Only half of us can be chosen each year,” he notes about a process that now involves annual reviews.

Some 60% of the 10,000 “chosen” scientists are 45 years of age or younger, the average age that CAS has set for a research group. However, there are exceptions. At the Institute of Geography and Resource Science in Beijing, for example, senior scientists still play a prominent role, because geography is a “science [that] needs experience and accumulation,” says Liu Jiyuan, head of the institute. A select group, mainly CAS academicians and tutors of doctoral students, are no longer on the official job rolls but serve as consultants to young and middle-aged scientists on research projects.

That's not the case at BAO, after Director Ai Guoxing decided that no one over the age of 50 could be chosen for KIP. Ai went even further, sorting researchers not chosen for KIP (about 150) into three categories. “The first group [of 30] is still involved in scientific research but does not benefit from the program funding,” he says. “The second group [of 80] went to companies opened by the BAO to do product development and sales promotion. The third group [of 40], all above 50 years old, were asked to retire, and now they receive a pension.”

Zhu Cuilian, a middle-aged woman who works in the “solar activity forecasting” research group at BAO, fell into the first category. She and one other researcher do the work of five people previously in her group. Her workload has increased, she says: “My colleague and I take turns doing the forecasting on weekends and holidays.”

Whereas Zhu accepts the emphasis on youth—”I know that young people should be the main force of the Knowledge Innovation Program, and I am no longer young”—the transition has been much harder for Li Xiaocong. “I was worried about my living conditions when I was asked to retire,” says Li, whose husband was laid off at about the same time from a state-owned enterprise. “I loved my job, and I need money to support my son, who just entered the university.” Li says that there are many people like her who wanted to keep their jobs and did not feel old enough to retire.

Li, who receives a pension from BAO amounting to 80% of her former salary, has since found a temporary job elsewhere that does not make use of her astronomy training. But she harbors hopes of returning to BAO and knows that a new publication will greatly improve her chances of passing the next review. Toward that goal, she spends her spare time preparing one of her papers for publication in an SCI journal.

In contrast, KIP has been a real career builder for Wu. His pay jumped from $250 to$1250 a month after he won a CAS first prize for scientific findings and for publishing a paper in The Astrophysical Journal Letters. Each year he also receives $12,000 in research money from a state high-technology program and$25,000 from the National Committee for Natural Sciences fund, in addition to $55,000 from KIP. Wu also can choose the members of his research group, which now includes three postgraduate students and a longtime assistant, and can set their pay based on the number of papers published and any awards received. He is one of only two people in the group to receive a long-term BAO employment contract, which runs for 3 years and is renewable after a review. Even those with less stellar achievements are doing well. The salaries of employees selected for the program have gone from$150 to $475 a month, whereas the rest have received only small increases. The publishing incentives—ranging from$240 to \$1200, depending on the journal's reputation—have also worked, says Ai. The number of SCI-quality papers by BAO researchers has risen from 25 in 1998 to more than 70 last year. Even his students are part of the paper chase: They must publish four SCI papers before they are allowed to graduate, Ai says, and nobody has let him down.

## More than money

Not all scientists agree with the use of financial incentives to spur productivity. “I believe that many people choose science as their career because they love to do it” and not because of monetary rewards, says Yuan Yaxiang, vice president of the Academy of Mathematics and System Sciences and director of the Institute of Computation Mathematics and Scientific Engineering Computing. A better way to evaluate a scientist, he says, is to solicit judgments about the work every few years from a group of outside experts.

Some young scientists agree with Yuan. The 36-year-old Zhang Dexing, who just returned from the United Kingdom to head a biological lab at the zoology institute, says that quantity isn't everything. Sometimes, he notes, he might go 2 years collecting specimens and carrying out the necessary analyses before he can prepare a paper.

But other scientists see “more papers, more bonus” as an appropriate policy for a transition period. “We did not have a rational evaluation system for basic research in the past, and there were some people who did not work hard,” says Wang Huaning, a principal investigator at BAO. At least the number of papers published and citation rates track what a researcher has been doing, he says.

The bonuses also allow the authorities to reward talented researchers who might be tempted to go abroad to snare a higher salary. “All we have done is to build an environment for scientific research that is on a par with the world's advanced level,” says Bai Chunli, vice president of CAS. “Modern research and development is actually a war for more talented people,” he adds. “We are determined to win the war, with the help of the Knowledge Innovation Program.”