# News this Week

Science  05 Mar 1999:
Vol. 283, Issue 5407, pp. 1422
1. FRANCE

# Physics Prize Falls Foul of Middle East Politics

1. Michael Balter

PARIS—A bitter dispute has broken out among some of France's leading physicists over a decision by the French Physical Society (SFP) to block the award of a prize named after a Lebanese scientist to an eminent Israeli researcher. The emotional battle has detonated an explosion of charges and counter-charges among the close-knit community of France's physics elite.

The physicist who originally created the Rammal Medal—in honor of a gifted Lebanese physicist, Rammal Rammal, who made his career in France but died while still young—accuses the SFP of caving in to pressure from Lebanese officials and academics to withhold the prize from theoretical physicist Daniel Amit of Hebrew University in Jerusalem. But SFP officials, and some members of the prize jury, insist that the jury's vote to award the prize to Amit was invalid, arguing that a Lebanese juror was discouraged from attending its meeting. To complicate matters further, officials at the French embassy in Beirut and the foreign affairs ministry in Paris became embroiled in the affair, possibly exerting an influence on the outcome.

The dispute is all the more poignant because Amit is an outspoken opponent of Israel's continuing occupation of southern Lebanon. He spent 2 weeks in a military prison in 1982 for refusing to serve in the Israeli army during its invasion of that country. Indeed, all parties to the controversy agree that, aside from his nationality, Amit was an ideal candidate for the annual prize which, according to its founding statement, is intended to reward a physicist from a Mediterranean country who has “by his life and his [research] activities given a new and modern form to the currents of scientific exchange in this region of the world.”

The jury, made up of physicists from France and other countries, voted to award the 1998 medal to Amit on 12 October last year, subject to the approval of the SFP, which administers the prize. But in the wake of a concerted campaign of opposition to the award which began in Lebanon almost immediately afterwards, SFP officers decided early last month to overturn the decision and leave the medal unattributed for 1998. The SFP has not publicized this decision and has distributed a brief communiqué only to those immediately involved and “anyone who asked about it,” according to physicist Roger Balian, who was president of the SFP until early February.

Nevertheless, on the basis of interviews with all the major players in the dispute, as well as dozens of letters and stored e-mail exchanges provided by both sides, Science has pieced together the story of a feud that has been emotionally scarring to all concerned.

The controversy centers around the actions of Gérard Toulouse, a noted theoretical physicist at the Ecole Normale Supérieure (ENS) in Paris. Toulouse created the medal to honor Rammal, to whom Toulouse was a close friend and mentor. Rammal, a talented scientist who was well known and respected in the French physics community, died in 1991 at age 39 of complications from an earlier heart transplant operation. Over the years, the winners of the prize—which includes no monetary award—included physicists based in Italy, Turkey, Tunisia, Spain, and Egypt.

Toulouse—who was also a member of the prize jury—told Science that he had long believed the medal must eventually go to an Israeli because of the many excellent physicists there. But in previous years this suggestion had met resistance from other jurors concerned about possible adverse reactions in Lebanon. This viewpoint had been expressed most strongly by jury member Hafez Kobeissi, secretary-general of Lebanon's National Council for Scientific Research (CNRSL). This resistance, Toulouse now says, “was a betrayal of the spirit of the medal.”

In early 1998, however, just before the first of the jury's two deliberation sessions for that year, Kobeissi died. At the meeting, which took place in May, Toulouse proposed Lebanese physicist Ragi Abou-Chacra, then also with the CNRSL, as Kobeissi's replacement. But there are sharp differences of opinion about what happened next. According to some jury members, Toulouse discouraged Abou-Chacra from attending the crucial meeting in October that chose Amit, knowing that an Israeli would not be chosen if a Lebanese juror was present. “Gérard Toulouse had a certain desire to keep Ragi Abou-Chacra away from any discussions,” says ENS physicist Claude Taïeb, secretary of the jury. Abou-Chacra, now dean of science at Saint-Joseph University in Beirut, says that Toulouse “advised me not to attend the meeting,” with the result that he was “not given a chance to express my opinion … that we cannot award the Rammal Medal to a scientist from Israel, regardless of the identity of that person, unless peace is established between Israel and Lebanon.”

Toulouse insists that Abou-Chacra's appointment to the panel was supposed to have begun only in 1999. He points to an 11 August 1998 e-mail he received from ENS physicist Michèle Leduc, who later became president of the jury, in which Leduc endorses Toulouse's suggestion that Abou-Chacra join the panel the following year. Although Leduc says she now believes Abou-Chacra was officially appointed to the jury just before the 12 October meeting, she adds that she does not believe Toulouse deliberately kept him off the jury but only took advantage of his absence. And Toulouse admits that Abou-Chacra's absence provided a “window of opportunity” to propose Amit for the medal.

With only 10 of the jury's 24 members present at the 12 October meeting, Amit received five votes, with one vote for a Greek physicist and four abstentions. In a later letter to Balian, Leduc wrote that Amit “obviously would have received the majority, and maybe even the unanimity [of the vote] if no member of the jury had feared the Lebanese reaction.” This view is supported by jury member Miguel Virasoro, director of the International Center for Theoretical Physics in Trieste, Italy, who comments in his personal capacity that “everyone, absolutely everyone was agreed that Amit was ideal.”

The jury then forwarded its choice to the SFP for ratification. At the same time, Abou-Chacra was informed of the result by another jury member, ENS physicist Franck Laloë. Just over a week later, letters protesting the decision began pouring into Balian's office from Lebanon, signed by political leaders, the Islamic Association, the League of Professors, and leaders of the CNRSL. In addition, a flurry of articles appeared in the Lebanese press—including at least one by Abou-Chacra himself—demanding that the vote be overturned.

At this point, the French embassy in Beirut appears to have taken notice of events. Leduc says she received “two or three” telephone calls from the embassy's attaché for scientific and cultural cooperation, Henri Genaud. “He was annoyed at the reaction in Lebanon, and he wanted it to stop,” Leduc told Science. “He was in favor of suspending the medal to calm things down.” In an e-mail to Balian, dated 10 November 1998, Leduc describes her conversations with Genaud more explicitly: “The embassy is very clear: They think that the harmful effect of this award to Amit will be significant for the political, scientific, and university relationships between the two countries.”

Genaud says he became involved in the affair in a “personal capacity” and not as an official representative of the embassy. But a more intimate involvement by French foreign affairs officials is implied by a 25 January 1999 e-mail from Genaud to Balian. Shortly before, SFP officials had finally decided that Amit would not get the prize, and Balian had asked Genaud to look over a draft communiqué announcing this decision. In the message, Genaud states that he had contacted “the services of the French foreign affairs ministry, the Lebanese CNRS, the Lebanese ministry of higher education … and the Lebanese Prime Minister Sélim Hoss to obtain, on the one hand, a green light on the text of the communiqué and, on the other hand, assurances as to its publication and the reactions that it would possibly draw.”

Genaud suggested to Balian a number of changes to the communiqué, including cutting out a reference to Amit's refusal to serve in the Israeli army of occupation in Lebanon. In the end, however, the SFP opted for a shorter version of the communiqué, which said that the prize would not be given in 1998 due to the “serious and multiple difficulties” which had arisen, and did not mention Amit.

Balian insists that political pressure had no effect on the SFP's decision which, he says, was based solely on the irregularities in jury membership. “The French government cannot interfere in this affair, it is private.” But Leduc says the embassy's attitude was indeed part of the reason she and some other jury members urged the SFP not to approve the choice of Amit. “There were pressures from the embassy, from the CNRSL, and from other organizations,” she says. “We would have been able to resist political pressures if the jury had functioned normally.” On the other hand, Leduc adds, if the jury had “functioned normally” she is certain “the vote would have been different.”

But ENS physicist Antoine Georges, a jury member who strongly supported Amit, says “it seems very shocking to me that the SFP could decide to not give a medal on the basis of the nationality of the candidate.” And Toulouse says, “The Lebanese never had the chance to know who Daniel Amit was. They said we must wait until the political situation is better, but that is completely contrary to the scientific spirit, which is to be in advance.” Virasoro agrees: “One cannot renounce principles because there could be reactions or, even worse, because there is pressure.”

Balian counters that awarding the medal to Amit would have “created hostility between Lebanese and Israelis, which contradicts the aim of the medal. … We cannot give this medal to an Israeli without a lot of psychological preparation.” And this sad affair has now called into question the future of the prize: The SFP is studying whether to continue its sponsorship of the Rammal Medal.

Such a decision would not sit well with Rammal's family, some of whom live in southern Lebanon. Ali Rammal, the Lebanese physicist's younger brother—an information technologist who lives in Paris—recently wrote to the SFP's current president, Jean-Paul Hurault, expressing the family's “profound surprise” at the recent turn of events and asking Hurault to “clarify your position concerning the future of the medal.” Ali Rammal told Science that although the family has “no opinion either for or against” the choice of Amit, “we want the spirit of the medal to be respected.” Moreover, he adds, “The guardian of the spirit of the medal today is Gérard Toulouse.”

As for Amit, he is philosophical about not receiving the prize. “All actors involved are lifelong, dear friends of mine,” he told Science. “I have no shadow of a doubt that they are all acting with constructive, ethical concerns in mind. Unfortunately, as human beings, we must learn to live with morally unresolvable situations.”

2. CELL BIOLOGY

# New Clues Found to Diabetes and Obesity

1. Dan Ferber*
1. Dan Ferber is a writer in Urbana, Illinois.

For the 15.7 million Americans with type 2 diabetes, good health means daily vigilance. To head off the eye, kidney, and heart damage the disease can cause, sufferers must follow strict diet and exercise regimes to prevent their blood sugar levels from soaring. Because those measures don't work for everyone, however, some people also need drugs to keep their blood sugars in check. And with nearly 200,000 people dying of diabetes complications each year, better drugs are still sorely needed. On page 1544, a research team based in Canada reports that it has identified a major new target for such a drug—and possibly for anti-obesity drugs as well.

The team, led by molecular biologist Brian Kennedy of the Merck Frosst Center for Therapeutic Research in Pointe Claire-Dorval, Quebec, and biochemist Michel Tremblay of McGill University in Montreal, came to that conclusion by creating a line of mice lacking an enzyme called protein tyrosine phosphatase-1B (PTP-1B). Those animals, the researchers found, are more much sensitive to insulin's blood sugar-lowering effects than control animals. Because type 2 diabetes is thought to result from an inability to respond to insulin, rather than to an inability to make the hormone as is the case for the type 1 form of the disease, the findings raise the possibility of treating type 2 diabetes with drugs that block PTP-1B activity.

The mutant mice also turned out to undergo a more surprising change: Unlike normal mice, they could eat a high-fat diet without gaining much weight. The researchers do not yet understand this connection, but the result suggests that PTP-1B-blocking drugs might be useful for treating obesity, too. Phillip Gorden, director of the National Institute of Diabetes and Digestive and Kidney Diseases, calls the findings “very interesting and very important.”

In the current work, Kennedy, Tremblay, and their colleagues were following up on test tube studies by their group and others showing that PTP-1B removes certain phosphates from the receptor that transmits insulin signals to the cell interior. The addition of those phosphates, which occurs when insulin binds the receptor, touches off a cascade of enzyme reactions inside muscle and liver cells. This tells the cells to take up glucose and sock it away as the storage carbohydrate glycogen, thus lowering blood sugar concentrations. Removal of the phosphates by PTP-1B should therefore turn off the signal cascade, and that's what researchers found in the test tube studies. To see whether the enzyme does the same in the body, the McGill team inactivated the PTP-1B gene in live mice. That “was the way to show whether the enzyme was important or not,” Tremblay says.

And important it was. Mice lacking the gene maintained normal blood glucose levels after a meal, even though they had half as much insulin in their blood as normal mice. In addition, a shot of insulin caused some of the PTP-1B-deficient mice to move so much glucose into their cells that they passed out from low blood sugar—something that never happened to the wild-type mice that received the same dose, Kennedy says. Together, those results showed that the knockout mice were more sensitive to the hormone than their wild-type cousins.

The group also showed that these effects are due to increased insulin receptor activity in the knockout animals. The receptor is a tyrosine kinase, an enzyme that when activated, in this case by insulin, adds phosphates to residues of the amino acid tyrosine in its target proteins. The researchers found that in the absence of PTP-1B, the receptor attached 2.5 times as many phosphate groups to the next protein in the insulin signaling cascade than it normally does.

So far, all the results had been in healthy mice, rather than diabetic ones. Obesity predisposes to type 2 diabetes in ways researchers do not fully understand. So to see if knocking out PTP-1B helps diabetic mice become more insulin-sensitive, the researchers tried to induce the condition by fattening both normal and mutant animals on rodent chow with 10 times the normal amount of fat. Only the normal mice became obese and showed signs of diabetes. “We expected both [strains] to become fat,” Kennedy says, “but right off the bat it became obvious that the knockout mice didn't gain as much weight.”

Equally important, the mice without PTP-1B appear healthy. Because tyrosine phosphatases may help check cell growth, “you might have had a beneficial effect on insulin signaling, but you also might have had tumors,” says endocrinologist Jeffrey Flier of Harvard Medical School. But the PTP-1B knockout mice have now passed the advanced age of 2 years and show no signs of cancer.

Still unclear is how enhanced signaling through the insulin pathway protects against obesity, although the researchers speculate that it might boost energy consumption by liver and muscle cells. Also unknown is whether PTP-1B overactivity plays a role in excess weight gain in normal animals—or in people. But even if it doesn't, that might not matter for developing an anti-obesity drug, says diabetologist Barry Goldstein of Thomas Jefferson University in Philadelphia: “The fact that the results are so clean, that there are apparently no other phenotypic changes, makes [PTP-1B] a very exciting drug target.”

3. ANTHROPOLOGY

# Genetic Clues Revise View of Japanese Roots

1. Dennis Normile

KYOTO, JAPAN—Conventional wisdom traces the peopling of the Japanese archipelago to two waves of migrants. The first, the ancestors of the Jomonese, a people who lived cut off from the Asian mainland for 10,000 years and presumably account for many of the distinct cultural and ethnic traits of modern Japanese, are thought to have originated in southeast Asia and island-hopped to southern Japan about 30,000 years ago. The second wave, the Yayoi, originated in northern Asia and traveled down the Korean Peninsula to Kyushu some 2300 years ago, bringing with them rice paddy agriculture and metal tools. What happened next is not clear, including whether the Yayoi mixed with, displaced, or were overwhelmed by the Jomonese.

But a new study of the origins of the Jomonese and Yayoi turns that explanation on its head. Speaking at a conference here last week,* Michael Hammer, an anthropological geneticist at the University of Arizona, Tucson, suggested that the early Jomonese likely originated in central Asia and crossed over a northern land bridge, while the Yayoi may have had roots in southeastern Asia before they headed north and arrived in Kyushu from the west. At the same time, Hammer's work, which is based on an analysis of the Y chromosome in some 2500 men in 60 populations around the world, reinforces the prevailing view that modern Japanese are a hybrid of these two earlier cultures. “These are extremely interesting results, and I expect they will stimulate further work using the Y chromosome,” says Keiichi Omoto, an anthropological geneticist at the International Research Center for Japanese Studies (IRCJS) in Kyoto.

Hammer, a leader in the use of the Y chromosome in anthropological studies, tackled the question of Japanese roots by using two sets of genetic variations, called haplotypes, as markers for the different populations. He used DNA from the Y chromosome because it is the largest nonrecombining block in the human chromosome, which means it is passed down through the male line unchanged except for mutations. One of the Y variants, called YAP, is highly represented in the Ainu of northern Japan and in Okinawans. These two groups are believed to have been the least affected by Yayoi immigration and, thus, seen as possibly characteristic of the Jomonese. Hammer figured a different haplotype, designated as 1J and common among both Koreans and Japanese, might be a Yayoi indicator.

His findings support the theory that the occurrence of the two haplotypes should vary with distance from northern Kyushu, the Yayoi point of entry. YAP occurs in the population more frequently with increasing distance from northern Kyushu, while the ratio of the population with the 1J haplotype is highest in northern Kyushu and decreases with increasing distance. “These results are consistent with the hybridization theory,” Hammer concludes.

The studies trace only the male lineage, but other studies presented at the meeting that analyze the female line also support the hybridization theory. Hammer's work goes a step further, however, by exploring the origins and migration patterns of the Jomonese and Yayoi. Relying on blood samples already collected around the world for other purposes, he found that YAP, or a closely related variant, only showed up in populations from Japan, Tibet, and sub-Saharan Africa. Hammer believes that the sub-Saharan variant evolved after the variants found in both Tibet and Japan. Variants of the 1J haplotype were common in Japan and Korea, but also appeared in Manchuria and southeast Asia.

To explain these patterns, Hammer theorizes that YAP originated in central Asia 50,000 years ago. People carrying YAP dispersed across the east and west, perhaps under pressure from new waves of immigrants. Eventually YAP peoples were pushed to the fringe areas of Tibet and Japan, and all traces of YAP in central Asia were subsequently obliterated by mixing. The ancestors of the Jomonese crossed a land bridge to Japan about 30,000 years ago and were cut off from the mainland when water levels rose about 12,000 years ago. Another group migrated all the way to Africa, he speculates.

The 1J haplotype emerged in southeast Asia and was carried north and east, Hammer believes, eventually spreading to the Korean peninsula and Japan. Tracing the Yayoi to southeast Asia “needs much more work,” he admits, adding that his theory nevertheless is consistent with the origin 8000 years ago of paddy rice agriculture in south Asia.

Support for Hammer's scenario splits along specialty lines. Omoto notes that a northern origin for the Jomonese “fits very nicely” with his own earlier work using classical genetic markers such as enzymes. But Kazuro Hanihara, a physical anthropologist also at IRCJS, says, “I'm not able to agree.” The morphological evidence, such as skull shapes and teeth characteristics in fossils found throughout Asia, tie the Jomonese to southeast Asia, he believes.

Even harder for everyone to swallow is the premise of a migration to Africa. “There is no fossil evidence for any migration [of any type] into Africa,” says Chris Stringer, a physical anthropologist at the Natural History Museum, London. Omoto says a more convincing theory would posit that YAP originated in Africa and was somehow retained in a few isolated populations. Hammer's answer is that it may be time to reexamine the conventional view that migrations were always out of Africa and never back in.

Scientists hope to find ways to resolve this and other conflicting evidence being gathered by geneticists, anthropologists, and archaeologists. William Wang, a linguist at the City University of Hong Kong, says, “Each discipline provides just one window to the past. We need several viewpoints to get an accurate picture.” Stringer says that finding a way to blend the genetic and anthropological evidence into a consistent picture of Japanese origins, which is presumed to rest on two distinct waves of migrants, “could set an example for [work on] other regions” with more complex migratory patterns.

• * Recent Progress in the Studies of the Origins of the Japanese, 22 to 23 February, Kyoto.

4. SCIENTIFIC MISCONDUCT

# Chinese Journals Pledge Crackdown

1. Justin Wang*
1. Justin Wang writes for China Features in Beijing.

BEIJING—Chinese journals and scientific societies have embraced a new code of conduct designed to reduce the incidence of plagiarism, fabrication, and other acts of misconduct. The policies, adopted last month at a national meeting here, are meant to alert editors and authors to a problem that Chinese authorities see as a threat to their rising investment in research.

The campaign, organized by the China Association for Science and Technology (CAST), is the most visible to date on this sensitive topic (Science, 18 October 1996, p. 337). Two 1997 cases, involving duplicate publication of research from its magazines, prompted the association's Committee of Morality and Rights of Science and Technology Researchers to convene a meeting of representatives from several hundred scientific societies and journals. On 1 February the group endorsed a seven-part “Moral Convention.”

The one-page statement asks journal editors to refrain from publishing poor-quality, “from-a-buddy” articles, to reject articles of questionable authorship, and to weed out multiple submissions. It suggests that authors found to have committed plagiarism, fabrication, or falsification of data be warned in writing, followed by a boycott of future articles, notification of their home institution, and public disclosure of their misdeeds. CAST is also thinking about asking all journals signing the convention to reject any articles for up to 10 years from authors found guilty of misconduct, and to make their names public.

In addition to recommending ways to stamp out misconduct, the convention also affirms the role of authors and seeks to promote better communications between journals and the scientific community. It asks journals to notify authors of the status of their submissions within a reasonable period of time and to respect their “rights and interests.”

Chinese journals and science officials have long been concerned about scientific misconduct, especially plagiarism, but the two 1997 incidents brought the issue to a head. In one case, an associate professor at the Higher Education Research Center of Nanjing Teachers' University copied an entire article on pay disparities in the labor market from Science and Technology Guide, a monthly CAST publication that is widely circulated, and published it in another, less prominent journal. The plagiarism was discovered after the two journals merged their editorial offices and CAST became publisher of both journals. The second incident involved a faculty member at the Institute of Higher Education of Tongji University in Shanghai, who copied an article in the Guide about chaos theory. The plagiarism was spotted by a reader.

Both plagiarists were identified in a May 1997 article in the Guide, which has decided not to accept any more submissions from the authors. “There must be no compromise over dishonesty and no cover-up. Taking pity will harm the cause of science,” says Cai Decheng, former standing vice president of the Guide.

Chinese scientists and journal editors see the convention as a useful tool and a necessary step in combating misconduct. “These cases of misconduct have ruined scientific values and damaged academic standards,” says Zhang Yutai, first secretary of the CAST Secretariat. But some scientists worry that it will not be sufficient to root out the problem. “The burden of proof is mainly on the journals themselves,” notes one director of CAST who requested anonymity. “But it is difficult for editors to raise copyright or other legal issues with the wrongdoer.” Journals that decide to conduct investigations often get little help from the home institutions, notes one editor: “Some institutions and universities cover up the wrongdoing to protect their own reputation.”

The process needs to go a step further, agrees Tsou Chen-lu, a professor of biophysics and former head of the National Laboratory of Biomacromolecules in Beijing. “What we need is a convention on morality and behavior of Chinese science researchers that builds upon this convention,” says Tsou. Without a broad national policy, he and other scientists fear that self-interest may stifle efforts to root out misconduct.

5. PARTICLE PHYSICS

# Surprising Asymmetry Seen in Kaon Decays

1. James Glanz

CHICAGO—Once again, nature is teasing particle physicists. The Standard Model of subatomic particles, a body of theory that has survived several close shaves over the past few years, suggests that a lopsidedness in the laws of physics called direct CP violation should be small. Now data from millions of particle decays in the Tevatron particle accelerator at the Fermi National Accelerator Laboratory (Fermilab) near here point to a much larger asymmetry. Either the Standard Model is showing a crack—something theorists have long hoped for—or they have been applying the model incorrectly.

Either way, “it's a shocking result,” says Bruce Winstein, a University of Chicago physicist on the experiment, called KTeV. For more than 30 years, CP violation has been known in a so-called indirect form, first observed by the University of Chicago's James Cronin and Princeton University's Val Fitch at Brookhaven National Accelerator Laboratory. Fitch and Cronin studied the decays of particles called kaons into the lighter pions. They showed that a long-lived kaon, called K-long, decayed about once out of 500 times into two pions instead of three, which would be forbidden if the laws of physics were unchanged when all particles are changed into their antimatter counterparts and space is reflected about all three of its axes. It was a new and unexpected asymmetry, and its discovery won Fitch and Cronin the 1980 Nobel Prize in physics. [Hints of similar asymmetry were recently seen in the decay of particles called B mesons at Fermilab (Science, 18 December 1998, p. 2169).]

Theorists later came up with an explanation for CP violation within the Standard Model by assuming that a K-long consists of a slightly lopsided mixture of an ordinary neutral kaon and its antiparticle, which added a kind of CP “impurity” in an otherwise symmetrical state. The same explanation predicted that CP violation would still be evident, but smaller, in decays of the neutral kaons themselves. So experimenters went looking for such “direct” CP violation.

But that would require keeping track of many more decays at once, including some that went to neutral pions, which are much more difficult to pick out than the mostly charged pions of the original experiment. It took decades of improvements in detector technology—and the design of a clever experiment—to pick out slight differences in the way kaons and antikaons decayed into several combinations of pions, says Sunil Somalwar, a KTeV collaborator at Rutgers University in Piscataway, New Jersey.

Because the team did a blind analysis of the data—eliminating any human bias by adding an unknown offset that was removed at the very end—they themselves did not know the answer until a few days before the University of Chicago's Peter Shawhan gave a seminar at Fermilab on 24 February. “It was one of the most remarkable moments I've ever experienced in a physics seminar,” says Chris Quigg, a Fermilab theorist who is not a member of KTeV. “When he ripped the Post-It off” a viewgraph, revealing the value, “there was a quarter-note rest in the whole audience during which nobody breathed; and then a big gasp, collectively.”

The number was far larger than suggested by earlier, sketchier results from the Tevatron, although some results from CERN, the European particle physics lab in Geneva, Switzerland, had also suggested a large value. In one respect, the large asymmetry offers solace to devotees of the Standard Model: It definitively rules out an alternative explanation of the Fitch-Cronin results based on a postulated “superweak” force that would have stirred up the asymmetric mixture. “What this new result does is basically throw that out as an explanation for CP violation,” says Fitch. On the other hand, the asymmetry, about 300 times smaller than in the Fitch-Cronin experiment, is also several times larger than the Standard Model had suggested.

Still, theorists should not give in just yet to the temptation to look beyond the Standard Model, says Quigg. The calculations that predict a value for direct CP violation involve poorly known quantities like the mass of the strange quark. If, as some theorists have recently proposed, the strange quark is much lighter than has been assumed, all could be well again, he says. More data are on the way from CERN and the Tevatron, so particle physicists should soon know whether the new results are more tease than promise.

6. BIG SCIENCE

# Panel Backs Next-Generation Synchrotron

1. Robert F. Service

GAITHERSBURG, MARYLAND—A key federal panel last week recommended continued research toward a “fourth-generation” synchrotron, a machine capable of creating x-ray pulses billions of times more intense than current designs. The instrument could revolutionize many fields of science, from figuring out protein structures to understanding the physics of materials, by providing more detailed snapshots and movies of the atomic structures of molecules and materials. It's unlikely to be built in the next decade, however.

Last summer, the Department of Energy (DOE) asked the panel—made up primarily of university and industry-based scientists—for advice on how to proceed with novel synchrotrons. The panel's go-ahead is expected to prompt DOE to spend up to $8 million a year this year and next on research and design ideas for the machine. If given another go-ahead in 2001, DOE would likely ask Congress for about$100 million to build a test-bed facility at the Stanford Linear Accelerator Center. A full-scale facility, designed to accommodate around 1000 users a year, is expected to cost $1 billion to$2 billion.

While several different designs have been proposed for a fourth-generation synchrotron, the panel threw its initial support behind a so-called “hard x-ray free-electron laser,” which would use gyrating beams of electrons traveling through a linear accelerator to create its x-ray pulses. The synchrotron would produce much higher intensity pulses than existing third-generation machines can, and the x-ray light waves in each pulse would be “coherent,” with the crests and troughs of the waves traveling in lockstep. That property, the panel concluded, “has the potential to open new areas of science that are likely to be well beyond what can be anticipated by current scientific knowledge and predictions.” One likely beneficiary is the nascent field of x-ray holography, where the blast of coherent x-rays triggers the emission of additional x-rays from atoms in a sample. By tracking how these waves interfere with one another, researchers can determine the location of nearby atoms in a three-dimensional sample in one step rather than the complex multistep approach used today.

The panel's endorsement “is a major event for the synchrotron community,” says David Moncton, director of the Advanced Photon Source, a third-generation synchrotron at Argonne National Laboratory in Illinois. But to justify the enormous expense, “a more effective case for the science must still be made,” says panel chair Stephen Leone of JILA—formerly known as the Joint Institute for Laboratory Astrophysics—in Boulder, Colorado. He's confident that can be done. Whether the promise of great science is enough to convince Congress is another matter: DOE's budget is already feeling the strain of another major project, the $1.3 billion Spallation Neutron Source, which is expected to be completed in 2005. 7. QUANTUM MECHANICS # Entangled Trio to Put Nonlocality to the Test 1. Andrew Watson One of the strangest claims of quantum mechanics is that two particles can be “entangled”—inextricably linked at birth. In theory, a measurement on one entangled particle is linked to a degree that defies common sense to a measurement on the other, even though the pair may have traveled to opposite sides of the cosmos. Now physicists at the University of Innsbruck in Austria have created the same eerie link among a trio of photons, so detecting two of the photons preordains the result of the third measurement. The feat, which the Innsbruck group reported in the 15 February issue of Physical Review Letters, allows researchers to close some loopholes in tests of the strange predictions of quantum mechanics. By studying pairs of entangled photons, physicists have already tested the quantum prediction that a measurement on one of the pair will instantly affect the outcome of a measurement on the other, even if they have traveled great distances apart since being created. But these tests have to be run over and over to be sure these “nonlocal” effects aren't due to chance, and purists find such statistical evidence dissatisfying. The entangled trio opens the way to a single measurement that will give one result if nonlocality is true and another if it is not. “This could be like a single shot test of quantum mechanics,” says Vlatko Vedral of Britain's University of Oxford. Because of these stakes, says Daniel Greenberger of the City University of New York, City College, a “race” was on to create such three-photon states. The winning Innsbruck team “did a phenomenal job,” he says. “I think it's very significant,” agrees Vedral. Besides allowing a yes-no test of quantum nonlocality, three-photon entanglement should also offer more efficient quantum communications, says senior team member Harald Weinfurter. Quantum communications, which promises to be more efficient and secure than normal optical signals, uses entangled states to pass information between participants using carefully prepared sets of photons shared among them in advance. “We are moving toward quantum communications,” says Vedral. “It's got implications for quantum computing and all kinds of fundamental experiments,” adds Greenberger. The Innsbruck experiment begins with the same kind of crystal that spawns entangled photons in pairs. When a photon is fired into it, the crystal can split the photon into two daughters that each have half the frequency of the parent. Their common parentage means that the photons' properties are linked. For example, if the first one is horizontally polarized, the other has to be vertical. However, according to quantum mechanics, such properties remain indeterminate until they are actually measured, so if a measurement on one photon finds that it has vertical polarization, its sibling instantly “knows” that its own polarization is horizontal. To entangle three photons, Weinfurter and his colleagues direct a high-frequency laser beam onto the photon-splitting crystal and wait for two photons to cleave simultaneously, giving two entangled pairs. Each time this happens, three of the four photons pass through a system of polarization-sensitive beam splitters and other optical elements, which tangle together the photons in such a way that it is impossible to tell them apart. “We interfere the particles in such a way that in the end you cannot decide any more which of the particles belongs to which pair,” says Weinfurter. Each entangled trio then heads toward three single-photon counters, each with a polarization filter in front of it. These are primed to look out for the trio, amongst other photons, by the detection at a fourth detector of the fourth, unentangled photon of the two pairs. The orientation of the polarization filters is set so that, if the photons are entangled, the counts in two of the detectors are correlated with those in the third—so simultaneous detection in all four detectors flags three-photon entanglement. Three independent photons would show no such correlation. Recent theoretical work by the University of Calgary's Richard Cleve and others suggests that entangled trios could make quantum communication systems more efficient, reducing by a third the amount of communication required to share information. Equally tantalizing for quantum purists is the possibility of a simple yes-no test of nonlocality. Three-photon entanglement means that the experiment in effect registers a photon in one detector if nonlocality is operating, but in a different one if it isn't. “It's no longer a statement about probabilities, but it's really a statement about one event,” says Weinfurter. The team has already made a first stab at the measurements and is analyzing the results, he says. The early news: “Quantum mechanics is correct.” 8. INDIA BUDGET # Big Increase Seen as Answer to Sanctions 1. Pallava Bagla NEW DELHI—Indian researchers are feeling buoyed by a new budget unveiled last weekend that hands science its largest increase of the decade. A 20% hike that would benefit both civilian and defense sectors is seen as a shot in the arm for domestic efforts to overcome foreign sanctions imposed in the wake of last spring's nuclear tests. These large increments “reflect India's determination to fight … the sanctions and denial of technology,” says Raghunath A. Mashelkar, director-general of the Council of Scientific and Industrial Research (CSIR). The increases stand in sharp contrast to last year's budget, which favored the atomic, space, and defense R&D sectors but didn't provide enough for other departments to even keep pace with inflation (Science, 5 June 1998, p. 1520). Science and Technology Minister M. M. Joshi told Science that this year's planned outlay of$2.56 billion is proof that the prime minister's slogan of “hail science,” coined after the blasts, “was not a hollow promise.” Still, not everyone is pleased. M. G. K. Menon, a physicist and former science and technology minister, says that the overall budget “lacks any bold new initiatives,” such as downsizing the general bureaucracy, and that it fails to invest sufficiently in civilian R&D. “The government has its priorities all wrong” through its emphasis on strategic research related to national security, he says.

To be sure, defense research still receives the lion's share of the government's science and technology investment, rising by 20% to $696 million. That figure constitutes 6% of the country's overall defense budget, its highest share ever, and reflects the government's attempt to beef up indigenous military technologies. Two related sectors also fare well. Space research is scheduled to rise by 16%, to$439 million. The increase will fuel a program to develop a booster for the country's new geosynchronous communications satellites and to create a second launch site. Atomic energy programs will receive a 32% boost, to $384 million, and funding for nuclear power plants, including a prototype fast breeder reactor, will also rise. The Bhabha Atomic Research Center in Mumbai, India's leading laboratory for nuclear weapons research as well as for civilian-related projects, gets a 33% hike, to$134 million. In addition, the Department of Atomic Energy announced a new National Center for Applied Mathematics and Radiophysics at the Tata Institute for Fundamental Research in Mumbai, although no details were available at press time.

On the civilian side, the department of science and technology, which funds academic research, is receiving an 18% hike, to $164 million, and the department of biotechnology's budget will rise by 9%, to$30 million. Civilian electronics garnered a 33% increase, to $52 million, reflecting India's push to compete in world markets. The budget for agricultural research and education is slated to rise by 26%, to$303 million, with the largest increases going for research on wheat, rice, and pulses.

Some of the loudest applause is coming from CSIR, which runs a chain of 40 laboratories catering to the needs of industry. Its 30% boost, to $199 million, is the largest in its 50-year history. “For the first time, our R&D budget is looking healthy,” says Mashelkar, who singled out for praise a tripling of its$1.25 million program to find commercial applications for biologically active agents in plants.

The new budget also contains new initiatives in grassroots innovations, vaccines, and biodiversity. A $5 million foundation is being set up to encourage small inventors to pursue commercialization of their ideas, along with a national registry to recognize the achievements of those without the means to apply for patents. The vaccine initiative is most likely to benefit efforts to develop vaccines against rotavirus and cholera, which are already undergoing clinical tests. A new National Bioresources Board is expected to consolidate under one roof all research related to bioresources and to expand survey and taxonomic activities. “This would put India in a very strong position to exploit the age of biology in the next century,” says Manju Sharma, secretary of the department of biotechnology. 9. LYME DISEASE # Patients Scarce in Test of Long-Term Therapy 1. Eliot Marshall Scientists are having trouble enrolling enough patients in a$4.2 million study to test whether long-term antibiotic treatment is effective against chronic symptoms of Lyme disease. Ironically, while researchers were discussing this problem in a suburb of Washington, D.C. last week, officials in Connecticut—ground zero for the disease—held a hearing to increase pressure on insurance companies to cover extended Lyme disease therapy.

About 13,000 U.S. citizens get tick bites each year that transmit a corkscrew-shaped parasite (Borrelia burgdorferi) that causes inflammation and joint pain—Lyme disease. But medical experts differ on how to treat very long bouts of arthritis, fatigue, and memory loss that some patients experience—a syndrome known as Chronic Lyme Disease (CLD), which is a controversial diagnosis.

The current study, run by a team led by Mark Klempner of the Tufts University School of Medicine in Boston, offers CLD patients a 90-day course of antibiotics. It's the largest study of its kind, aiming to bank 45,000 tissue samples for future research (Science, 13 October 1995, p. 228). But, as its leaders revealed in an interim review last week, patient recruitment is lagging.

After a year of advertising, says Klempner, only 57 subjects had been enrolled. The goal is to get 260 by the time the study ends in 2 years. More than 1200 people have expressed interest, and 700 have come in for screening. But only one in 10 who appear in the clinic fits the study's strict criteria.

A patient is enrolled in the double-blind placebo study, funded by the National Institute of Allergy and Infectious Diseases (NIAID), only after testing positive in a blood antibody test for Borrelia or having evidence of a tick bite rash. Patients who have received extensive intravenous antibiotic treatment are excluded, as are, for ethical reasons, those who have never received any treatment. The criteria are designed to focus selectively on CLD. Once enrolled, patients are randomized into a placebo or treatment group. The treatment group gets 30 days of intravenous ceftriaxone followed by 60 days of oral doxycycline.

On 25 February, Klempner asked a gathering of NIAID project managers, members of an advisory panel, and patient advocates for ideas on how to speed up enrollment. But he also offered an encouraging note: Many CLD patients have a distinctive enzyme in their spinal fluid that's been seen before in patients with neuroborreliosis—when the organism invades the central nervous system. The enzyme, he suggested, might be useful as a disease marker.

After the meeting, Klempner and NIAID Lyme program director Philip Baker said they would not relax study criteria to get more patients. That would wreck the protocol, Baker said. Instead, at extra cost, NIAID will add a recruitment site in Connecticut to existing sites in Boston and New York, and possibly others in New Jersey and Maryland. Klempner also sought help from advocacy groups such as the controversial Lyme Disease Foundation (LDF) of Hartford, Connecticut.

Karen Vanderhoof-Forschner, chair of LDF, which has questioned conventional CLD therapy, agreed to open LDF's membership to recruitment. In addition, Klempner says, some physicians who specialize in CLD may cooperate. Vanderhoof-Forschner told Science that she thought some eligible candidates had not volunteered for the study because they are suspicious of the researchers. But it's also possible, Klempner concedes, that there aren't as many CLD patients as expected.

That possibility might surprise the attorney-general of Connecticut, Richard Blumenthal. One day before Klempner solicited ideas to enroll more patients, Blumenthal held a hearing in Hartford to help people seeking reimbursement for long-term antibiotic therapy for Lyme disease. “Most insurance companies will not cover treatment of longer than 4 weeks of intravenous antibiotics,” Blumenthal says, “despite the judgment of treating physicians” that the drugs should be provided. Blumenthal is drafting a bill to “mandate” insurance coverage for what the doctor orders.

Meanwhile, Klempner says that his team will try to “remove barriers” to enrollment. Conducting free screening clinics “within the walls of advocacy organizations,” he says, should provide access to a new cohort of patients.

10. STEM CELLS

# Harnessing the Power of Stem Cells

1. Gretchen Vogel

The isolation of human embryonic stem cells could, in theory, allow scientists to cultivate any of the body's tissues in the lab, but realizing this potential won't be easy

When developmental biologist Gary Anderson walks into his lab at the University of California, Davis, in the morning, he's sometimes greeted by an amazing sight: a plate of cardiac muscle cells pulsing in rhythm. To Anderson, though, the sight is frustrating. He'd rather see no change at all: the same pig embryonic stem (ES) cells he left the night before, staying immature indefinitely. For reasons Anderson does not yet understand, however, they spontaneously differentiate into any of a variety of cell types—sometimes a disorganized blob of nervous tissue or cartilage, sometimes the heart cells. “We're constantly battling their desire to differentiate,” Anderson says.

That powerful drive to become something else is a signature characteristic of stem cells, the versatile cells that are one of the hottest and most controversial topics in developmental biology. The current excitement was touched off 5 months ago, when two groups reported that they had isolated the first lines of human ES cells—cells that can not only differentiate into all types of tissue, but can, under carefully controlled conditions, be maintained continuously as undifferentiated cells in lab cultures (Science, 6 November 1998, pp. 1014 and 1145). The work has sparked some controversy because one group extracted the cells from donated human embryos from a fertility clinic, while the other used tissue from aborted fetuses. But it could ultimately open the way to growing replacements for many types of tissue or even organs damaged by disease (see Perspective on page 1468).

That's what has made human ES cells such a hot commodity. But they are not the only kinds of stem cells that researchers have been cultivating. Over the past few years, scientists have also isolated stem cells from various tissues in animals and humans, including bone marrow and even brain, that are not “totipotent”—meaning that they can't form all tissue types, as ES cells do—but can produce a narrower range of cells, such as the various blood, muscle, or nerve cells. What's more, recent findings suggest that these partially differentiated stem cells may be more versatile than thought.

Whichever kind of stem cell they work with, researchers hoping to channel them into becoming certain kinds of tissue—say, dopamine-producing neurons for implantation into patients with Parkinson's disease, heart muscle cells to repair damaged hearts, or insulin-producing cells for diabetes patients—will have to solve a great many problems first. Put simply, they will need to understand and control the process of cell specialization that, for example, turns stem cells into beating heart cells in Anderson's lab. “Right now, we can get spontaneous differentiation,” says Martin Pera, a developmental biologist at Monash University, in Melbourne, Australia, but “we don't understand how to control that process” to produce cells of a specific lineage.

View this table:

To accomplish their goal of growing replacement tissues in the lab, researchers will have to identify the signals that tell a stem cell to become one tissue or another—one of the central mysteries in developmental biology. “We generally know the vocabulary” of growth factors and other molecules that help determine cell fates, says stem cell researcher Roger Pedersen of the University of California, San Francisco. “At least we know what to listen for.” But development also requires direct communication between cells and their neighbors, and Pedersen adds, “Eavesdropping on the particular conversation going on between the small group of founder cells in the embryo and their environment is not easy.” Once the conversation is known, researchers will have to take the next step: determining whether they can reproduce those signals for cells growing in lab cultures.

## Blood and brains and bone

The challenges are large, but already researchers have some promising clues. Much of the success so far has come in directing the differentiation of the tissue-specific cells. For example, a decade after scientists first isolated the stem cells that form blood cells from mouse bone marrow, researchers have learned how to identify and select such hematopoietic stem cells from human bone marrow, support them as they develop into blood precursor cells, and then use an array of growth factors and other regulatory proteins to guide their development into red cells and other mature blood cells.

Scientists also have promising clues about how to guide the development of neuronal stem cells. Developmental neurobiologist David Anderson of the California Institute of Technology (Caltech) in Pasadena and his colleagues, working on cells derived from fetal rats, have shown that several proteins, including neuregulin and bone morphogenic protein 2 (BMP2), can nudge neural stem cells toward becoming neurons, the neuronal support cells known as glia, or even smooth muscle. The researchers don't yet know exactly how the molecules are working, however.

And it takes more than just the right molecules to get stem cells to differentiate. For example, researchers have found that even with the best neuronal growth factors known, they could at most get only about half of the neural stem cells in their cultures to become neurons. Developmental neurobiologists think that may be because a developing brain cell relies on a specific combination of signals from its neighbors, reflecting its three-dimensional position in relation to other developing brain cells, to determine the specific type of cell it becomes.

“Essentially, to induce that differentiation [into neurons], you need the other cell types present,” says developmental geneticist Robin Lovell-Badge of the National Institute for Medical Research in London, who is among those working on the problem. “It's never going to be a simple story.” Caltech's Anderson agrees: “We are far from being able to take a cell from the brain and differentiate it, at will, into a desired cell type.”

Even mechanical forces can affect stem cell development. Cell biologist Daniel Marshak and his colleagues at Osiris, a biotechnology company in Baltimore, have been working on mesenchymal stem cells, which produce bone, cartilage, fat, and muscle. They and others have found that compressing the cells encourages bone development, while a tensile force—essentially stretching the matrix surrounding the cells—encourages development of tendons. Cartilage, on the other hand, will only form if the cells are clumped together or grown on a three-dimensional support, says Marshak.

Although scientists are not as adept at controlling the more primitive ES cells, they have made some progress there, too. Because human ES cells have been available only for a few months, most of this work has been done on mouse ES cells and cell lines from teratocarcinomas. These testicular or ovarian tumors grow from undifferentiated cells that have properties similar to those of ES cells. The tumors are masses of varied tissues, sometimes sprouting hair and even teeth, and although the cells are good for study, they aren't as promising for human therapies as ES cells are because of their tumor-forming potential.

So far, researchers have found that they can induce ES and teratocarcinoma cells to produce both neuronal and mesenchymal cells. In fact, some hints from developmental biologists suggest that neuronal cells may be a “default” fate for undifferentiated cells—meaning that's what they form when left to their own devices. Perhaps because of that, it is relatively easy to turn ES cells into neurons. Adding retinoic acid—which helps drive nervous-system development in the embryo—at the proper time “will make 90% of the cells go down the neural lineage path,” says David Gottlieb of Washington University in St. Louis. Those cells, in turn, differentiate spontaneously into neurons and glia.

Producing mesenchymal cells from ES or teratocarcinoma cells requires a different signal: BMP4. With the right factors, Gottlieb says, “you can tilt their fate in the desired direction.” No one is yet sure, however, which factors to use to encourage the formation of the third major type of tissue, endoderm, which gives rise to the gut, lungs, and inner organs. But developmental biologist Douglas Melton at Harvard University has been able to select a few mouse ES cells that differentiate spontaneously into endoderm cells.

He can also persuade those endodermal cells to become pancreatic cell precursors—just a few steps away from the insulin-producing cells needed by diabetes patients—by exposing them to cells from the “pancreatic bud” dissected from another mouse. Those cells “send out signals saying, ‘Come join us and become pancreas,’” Melton says. Of course, to make human pancreatic cells, scientists would have to reproduce those signals without the help of already-developing tissue, since work with fetal tissue raises serious ethical, moral, and perhaps legal issues, Melton notes.

## Unexpected versatility

But ES cells may not be the only hope for cultivating a wide range of tissues. There are signs that the more differentiated kinds of stem cells might have greater potential for tissue replacement than researchers had assumed. In a major surprise that seems to have overturned some long-standing beliefs about developmental biology, scientists reported earlier this year that neural stem cells from the mouse brain, when transplanted into mice whose bone marrow has been largely destroyed, develop into blood cells (Science, 22 January, pp. 471 and 534).

Although amphibians and other “lower organisms” can reprogram their cells to regrow body parts—new tails, for example, and even new limbs—developmental biologists have long assumed that mammalian cells are not so versatile. The standard view held that once a more specialized cell forms, its fate can't be altered so that it becomes something else. But “there is very little data to support [that assumption],” says developmental neurobiologist Ron McKay of the National Institute of Neurological Disorders and Stroke. The new observations suggest “that cells can make unexpected jumps between fates,” he says. “It's going to be interesting to see where those jumps can occur.”

Neuronal stem cells may not be the only ones able to switch fates. Last year, for example, developmental biologist Darwin Prockop of the Allegheny University of the Health Sciences in Philadelphia and his colleagues reported in the Proceedings of the National Academy of Sciences that when bone marrow stromal cells that would normally form muscle and other mesenchymal tissues are injected into the brains of mice, the cells become glia. And Juan Sanchez-Ramos of the University of South Florida in Tampa and his colleagues have preliminary evidence, presented at a meeting last year, that exposure to neural growth factors in culture can also send marrow cells down the road to becoming neurons.

Many stem cell researchers are greeting the neuronal stem cell-to-blood cell results with caution, however. “It's a preliminary analysis,” says developmental hematologist Leonard Zon of Children's Hospital in Boston. The blood cells may have grown not from neuronal stem cells, he says, but from a more primitive stem cell also in the sample.

But if the results are confirmed, the stem cell versatility could open the way to cell therapies that don't rely on embryonic stem cells, with their ethical and legal entanglements. A sample of a patient's own bone-marrow cells, drawn through a needle inserted into the hip bone, could be manipulated in culture and perhaps used to replace neurons in the brain or damaged heart muscle.

Such applications, however, face the same hurdles as other potential stem cell treatments. More immediately exciting, say many scientists in this area, is the chance to study what it is about a cell that allows it to remain malleable and able to change its fate in response to environmental cues.

Indeed, the efforts to harness stem cells' potential for biomedical applications is a boon for cell and developmental biologists. With cultured stem cells, cellular changes “that have only occurred in the complex context of the early embryo are now happening before your eyes in a dish,” says Washington University's Gottlieb. And those changes seem to be very similar to the ones that happen in the developing embryo, producing normal-looking neurons that make synapses and heart cells that set up rhythms. “What we're looking forward to is a much greater level of understanding and control,” he says. “It's a field in which chapter one has been written. I'm looking forward to chapter two.”

11. PROFILE

# A Man in a Hurry

1. Joseph Alper

While pushing research on stem cells and other areas that might help an aging population, Michael West has also sparked controversy

For Michael West, the man whose support of stem cell research led to three dramatic announcements last fall, aging is a driving force. But it's not so much his own aging that worries him. Instead, it's the fact that the population as a whole is growing older, placing an increasing demand on the nation's health care system. “Addressing the medical problems associated with aging has got to become our number-one priority for research in this country,” says West, who is president and chief executive officer of Advanced Cell Technology (ACT) in Worcester, Massachusetts.

And because aging is relentless, West himself has been relentless in funding and promoting cell technologies that could rejuvenate diseased tissues. In so doing, he has earned praise for his vision and energy as well as criticism for his bulldog approach. Although he's now heading up ACT, West is the founder of Geron Corp. of Menlo Park, California, a biotechnology firm that funded the two groups that last November reported isolating human embryonic stem (ES) cells (Science, 6 November 1998, pp. 1014 and 1145). These cells might have wide application in the clinic because they can be maintained in lab culture and, theoretically at least, can differentiate into all types of tissues (see p. 1432).

“Mike is a big idea guy,” says Jerry Shay, a cell biologist who worked with West at the University of Texas Southwestern Medical Center in Dallas, “but he is also a guy willing to do what needs to be done to make sure that important ideas actually get translated into action.” To West, that sometimes means courting controversy.

The ES cell work had already stirred debate because one group extracted the cells from aborted fetal tissue and the other from embryos discarded by a fertility clinic. Barely a week later, West added more fuel to the fire. He announced that ACT co-founder James Robl of the University of Massachusetts, and his then graduate student José Cibelli, had produced what appeared to be ES cells by first fusing human nuclei with cow oocytes whose own nuclei had been removed and then allowing these hybrid cells to divide. (See Science, 20 November 1998, p. 1390.) West announced this, he says, because he was leery of the legality of the work, and wanted to gauge the reaction of both the public and the U.S. government before he committed company funds to continue this research.

And react they did. President Clinton asked the National Bioethics Advisory Committee (NBAC) to review the medical and ethical aspects of the work. And much of the scientific community slammed him for publicly discussing data that had not yet been published in a peer-reviewed journal—stirring public debate about results that were not even documented. “I argued at length with Mike about going public with this,” says John Gearhart at Johns Hopkins University School of Medicine, one of the two researchers whose work on stem cells West supported while at Geron. “First, you don't publish scientific work in The New York Times, and second, I think this approach of doing interspecies research is inappropriate.”

But West shrugs off the criticism. “NBAC's immediate review found that there were no totally new ethical issues raised by this research as long as it is not intended to produce an embryo with the potential to develop into a child,” which West says the hybrid cells could not do in any case. “As a result, we're now free to pursue this avenue of research aggressively, and we do intend to publish our results in the scientific literature.”

West wasn't always driven to cure the ills of aging. In fact, he got a slow start on a scientific career. After graduating from Rensselaer Polytechnic Institute in 1975, West went to work in the family business selling trucks, which he eventually parlayed into a truck-leasing business. He soon became interested in gerontology, especially the question of why the ordinary somatic cells of the body have a finite lifetime. When he sold the truck business, he was free to turn to science. “I made a fair amount of money on that transaction, which has given me the freedom to do what I've wanted to do in science without having to worry about making a living,” says West.

A false start in graduate school at the University of Arkansas for Medical Sciences—he showed that one of his Ph.D. adviser's papers was incorrect—further delayed his research career. But he obtained his Ph.D. at Baylor College of Medicine in 1988, then decided to go to medical school there. He never finished, however, because his interests took a new turn.

At Baylor, he met Jerry Shay and long-time colleague Woodring Wright, who were among those who had linked shortening of the telomeres, specialized DNA sequences capping the ends of the chromosome, to cellular aging. The idea is that when the telomeres hit a certain length, cells go into senescence and die. Cancer cells have a way of restoring their telomeres, however, and thus can continue dividing when normal cells would stop.

West became fascinated by the telomere research and its potential to short-circuit the aging process and possibly cancer as well. “Mike sought us out when he arrived here for medical school, and we were smart enough to let him work in our lab at night and on the weekends, when he wasn't in class,” says Shay with a laugh. West was so taken with the potential of the work that he quit school in 1991 and started Geron to explore ways of exploiting it in the clinic. “He didn't want to waste time finishing the last year of medical school. He wanted to start work on this immediately and see what telomere biology could do for human medicine,” Shay says.

West also got the firm to branch into the field of stem cell research, which led to last fall's developments. By then, however, he had left because of differences in corporate philosophy. Although neither Geron nor West would comment on his leaving, others involved with Geron who know West speculate that he pushed the firm too hard into areas of research that other executives felt were peripheral to its main mission. “Mike was the founder, and I'm sure he was frustrated with the fact that the people running the company weren't pushing certain research projects forward,” says one academic researcher who receives funding from Geron. Says another, “I think it was a huge mistake for Geron to get rid of Mike, even though there were certainly clashes going on over research direction.”

Since that departure, West has founded another business, Origen Therapeutics in South San Francisco, which is attempting to use avian stem cells to develop improved chicken varieties. That led him to his current job with ACT. During a meeting with officials from Avian Farms, the nation's largest poultry producer and ACT's financial backer, West learned that Avian Farms was looking for someone to run ACT. In short order, he had won the job, reportedly investing in the company, too.

As president, West is pushing ACT in two directions. The company's original goal was to use cloning and gene transfer technologies to engineer cows that produce pharmaceutical proteins in their milk. Two cloned cows born last year testify to the firm's success in this area; under West, ACT will continue to focus most of its efforts on this relatively noncontroversial research.

But West is also determined to move forward with the cow-human embryo work. Cibelli, who is now a senior scientist at the company, is currently repeating his earlier work on a large number of cells. West believes that if this approach succeeds, it will solve both ethical and practical problems. “First, we wouldn't be using [human] fetal tissue or frozen embryos,” he says, “and second, should we eventually be able to turn stem cells into human organs for transplantation, we would be able to use a patient's own somatic cells to make the fusion and avoid any issues of rejection.”

But a few critics say that in his enthusiasm, West slights the ethical dimensions of the work. “Michael West is a man who sincerely believes that what he is doing is right because it will ultimately benefit human health,” says Richard Doerflinger, a theologian who works on pro-life issues for the U.S. Catholic Conference in Washington, D.C. and (like West) has testified before Congress on this issue. “But he really doesn't acknowledge the moral issues involved in what he's doing.”

West disagrees. “First, I disagree that this is a moral issue, because these cells cannot become human beings. If they could, I would not be promoting this research. So then, where's the morality in blocking studies that can benefit millions and millions of people?“

12. EVOLUTIONARY BIOLOGY

# Can Mitochondrial Clocks Keep Time?

1. Evelyn Strauss

New data fuel fundamental challenges to the accuracy of molecular clocks, although researchers say they are tackling the problems

Put a scientist on the analyst's couch, say the word “mitochondria,” and she's likely to blurt out “powerhouse of the cell” in response. But if she happens to be an evolutionary biologist, she might instead connect this organelle with a different type of power: the ability to illuminate evolutionary events deep in the past. For over two decades, biologists have been using mitochondrial DNA (mtDNA) to time the divergences of organisms from each other and to map human migrations. Such “molecular clock” studies have suggested that modern types of mammals and birds shared the Earth with the dinosaurs, that animals evolved hundreds of millions of years before their first fossils, and that “mitochondrial Eve,” our common female ancestor, lived about 200,000 years ago in Africa.

The DNA sequences pouring in from sequencing projects have fueled the effort and extended the clock approach to many genes in the cell nucleus. But the wash of data has uncovered some troubling facts. It's now clear that in many cases, the main assumption underlying molecular clocks doesn't hold up: Clocks tick at different rates in different lineages and at different times. And new work on the biology of mitochondria suggests that their evolution may be more complicated than researchers had suspected (see special issue beginning on page 1475).

“There's an emerging consensus that there are significant rate heterogeneities across different lineages,” says John Avise, an evolutionary geneticist at the University of Georgia in Athens. “How big they are and how to deal with them is very much a matter of concern.” Even those who once embraced the clocks are now somewhat skeptical. “Sure, there are mitochondrial clocks. A lot of them,” says Wesley Brown of the University of Michigan, Ann Arbor, who no longer uses mtDNA sequences to time ancient divergences.

But even as scientists cast a newly critical eye on the clock results, researchers are proposing and applying sophisticated statistical methods to deal with the clock's idiosyncrasies. “People don't assume that a gene is evolving at a constant rate,” says Blair Hedges, an evolutionary biologist at Pennsylvania State University in University Park. “We can test for different rates in different lineages and if a gene doesn't pass the test, you can throw it out.” He and others say clocks can yield useful information even if they don't work perfectly. “A year ago people would have said we're sunk” if some of the model's assumptions didn't hold up, says David Penny, a computational biologist at Massey University in Palmerston, New Zealand. “Now it's just a nuisance. We have to add variability into our estimates.”

Although researchers have varying degrees of confidence in the current statistical tests, biologists agree that clocks are worth fixing. “If you want to get a handle on the timing of events where there's no fossil record, this is your only option,” says Gregory Wray, an evolutionary biologist at the State University of New York, Stonybrook. “Is this method so flawed that you want to abandon it completely? The debate boils down to whether you want to throw the baby away with the bathwater.”

Back in the late 1970s, mtDNA seemed the perfect choice for peering into the past. In multicellular animals it is almost always inherited from the mother, so researchers can track a maternal lineage, useful when trying to identify a single common ancestor. There are thousands of mitochondria in every cell, so mtDNA is abundant and relatively easy to obtain, even from partially decayed samples. Nuclear genes lack these advantages but tend to evolve more slowly than mtDNA, making it possible to extend the analysis further into the past.

For the clock to work with either sort of DNA, nucleotide changes must tick away steadily so scientists can convert the number of nucleotide differences seen between two organisms into the number of years since they diverged. Different genes evolve at different rates, depending on the selective forces upon them, but the model requires only that each gene's clock maintains its own rate.

Early work hinted that this might not always be true, and now a plethora of data shows that many genes don't conform to this model. For example, the nuclear gene that encodes an enzyme called Cu,Zn superoxide dismutase (SOD) has a variable rate of evolution depending on what groups of organisms are measured, according to a 1997 study by evolutionary geneticist Francisco Ayala of the University of California, Irvine. He found that the gene evolved very quickly among different species of Drosophila–5 times faster than it did among multicellular organisms in general. A different pattern of varying rates emerged for another gene, which encodes the metabolic enzyme glycerol-3-phosphate dehydrogenase (GPDH).

Even on a single branch of a phylogenetic tree, rates can fluctuate over time. For example, one stretch of DNA within the Drosophila male fertility gene Odysseus (Ods), has changed more in the past 500,000 years than in the preceding 700 million years, according to work by evolutionary geneticist Chung-I Wu at the University of Chicago and his colleagues (Science, 20 November 1998, p. 1501). If researchers assumed a standard rate, they would conclude that the last 500,000 years spanned a longer period of time than the previous 700 million.

Other problems arise in simply counting nucleotide substitutions, says Charles Marshall, a molecular paleobiologist at the University of California, Los Angeles. Certain substitutions are more likely to happen than others, depending on the organism, the gene, and a nucleotide's position in the gene. And a single change in nucleotide identity doesn't necessarily mean that only one substitution has taken place—there could have been multiple hits at the same site.

## Paleontologists strike back

Such complexities may underlie some of the surprising results from clock methods, says Marshall. Many analyses of mitochondrial and nuclear DNA have arrived at dates that match those from fossils, but there are a few serious discrepancies.

For example, fossils record a burst of innovation that marked the emergence of many modern groups of animals about 530 million years ago, in the so-called Cambrian explosion. But molecular studies suggest that animals were diverging at least several hundred million years earlier (Science, 25 October 1996, p. 568).

One explanation is simply that creatures, especially soft, simple ones, can easily elude fossilization, so that multicelled animals could have lived before the Cambrian explosion without leaving a record. But paleontologists counter that the common ancestor of animals had a vascular system, a central nervous system, and body wall muscles, and that such a complex creature would eventually leave some trace. “Where are they going to hide this sophisticated animal for half a billion years?” wonders Doug Erwin, a paleobiologist at the Smithsonian's National Museum of Natural History in Washington, D.C.

Still, he and others caution that discrepancies between the molecular and fossil evidence don't necessarily mean that one type of data is wrong. Molecular data estimate the point at which genetic intermixing stops—usually some time before the appearance of diagnostic physical differences, which is what fossils record. “We're actually looking at two different things,” says Erwin. “When the arthropods and the chordates split, the first species don't look like flies or J. Edgar Hoover. They look essentially identical. The morphological differences that we later recognize as being one or the other evolve later.” But in this case, half a billion years of evolution is a bit much for even many molecular biologists to swallow, especially because the molecular estimates themselves have bounced around, from 1200 million years ago to 990 and even down to 670, suggesting that perhaps the fossils are right after all.

Fossils and molecules are also at odds over when most modern orders of birds and mammals appeared. Paleontologists haven't generally found them before about 65 million years ago, after the great Cretaceous-Tertiary extinction wiped out the dinosaurs. But molecular studies using both mitochondrial and nuclear DNA conclude that many living species diverged much earlier, up to 130 million years ago (Science, 1 May 1998, p. 675).

The molecular biologists argue that fossils documenting this early radiation have perhaps not been preserved or yet discovered. “Fossils are starting to get found on the opposite sides of barriers where people used to think nothing existed,” says Alan Cooper, a molecular evolutionist at Oxford University. “Now [from fossils] we've got parrots and seabirds in the late Cretaceous.”

But a recent study by paleontologist Mike Foote of the University of Chicago and his colleagues (Science, 26 February, p. 1310) suggests that the existing fossil record doesn't have huge gaps. The researchers assessed the quality of the record during the late Cretaceous, when molecular clock results suggest that modern placental mammals should exist. They concluded that the large number of other mammal fossils from that time make it implausible that modern placental mammals are there but undiscovered. “If the record really stinks, almost every species you find will be from single fossils,” says Foote. “But the empirical record is something like 10 to 100 times greater than what would be required to allow for a 65-million-year gap in the fossil record.” Hedges disagrees with that conclusion, noting that Foote's work assumes constant speciation across the Cretaceous-Tertiary boundary. “They ignored the biological effects of the asteroid impact that wiped out the dinosaurs,” he says.

## Riddles of recombination

New information about the complexities of mitochondrial biology is also raising new questions about the mtDNA clock. Conventional wisdom has it that mitochondrial DNA comes only from the mother's egg. But electron microscopy and DNA detection studies have revealed that the sperm's mitochondria can enter the egg, says evolutionary geneticist Adam Eyre-Walker of the University of Sussex in Brighton. And several papers to be published this week make the surprising suggestion that, contrary to what scientists have thought, sperm-contributed mtDNA can recombine with that from the mother. If that's true, a single recombination event could instantly insert or erase multiple changes in a piece of DNA, throwing off the clock. Such a phenomenon may also explain how some people can have two different versions of mtDNA in their cells (Science, 2 January 1998, p. 28).

In work to be published in the Proceedings of the Royal Society of London B, Erika Hagelberg, a molecular geneticist at the University of Otago in Dunedin, New Zealand, and her colleagues report that when they sequenced a fast-evolving region of mtDNA from 452 inhabitants of some western Pacific islands, they found three distinct groups. In the inhabitants of one island, however, they detected something peculiar: a high frequency of a very rare nucleotide substitution, seen in individuals from all three groups. It seems unlikely that this mutation had occurred independently in the three lineages on this island but nowhere else, says Hagelberg. Instead, she suggests that the mutation arose once on the island, in a male, then recombined with egg mtDNA during fertilization. The male descendants of this lineage then spread the mutation into the other two lineages, again by recombination during fertilization. Such recombination is “probably ticking away in the background in other parts of the world,” she says.

In an independent study, to be published in the same journal, Eyre-Walker and his colleagues draw similar conclusions. They looked for instances of multiple changes at the same site in mtDNA. “People have always interpreted that in terms of hypervariable sites,” says Eyre-Walker. But in their 29 samples, the team found that these multiple hits occurred far more often than would be expected by random mutation. Instead, “it could all be due to recombination,” he says. If so, “estimates of rates of nucleotide change are going to be incorrect.”

Recombination could also be bad news for use of mtDNA in other questions of human ancestry. For example, in 1997 researchers led by Svante Pääbo of the Max Planck Institute for Evolutionary Anthropology in Leipzig retrieved ancient mtDNA from a Neandertal and concluded that the sequence was so distinct that Neandertals didn't contribute mtDNA to modern humans (Science, 11 July 1997, p. 176). But recombination tends to mix up lineages and create a more homogeneous population of DNA sequences over time. So if mitochondrial recombination occurred in our ancestors, one might expect greater diversity in mtDNA sequences in the past—and greater disparity between ancient sequences and the more homogeneous sequences of today, says Hagelberg. That would mean, she says, that “the Neandertals might be more closely related to present-day humans than the mtDNA data suggest.”

“We should think about recombination as a possibility,” agrees Pääbo, “but it's not proven by this work. It may still be possible to explain the mutations with sites that mutate frequently.” He notes that Neandertals were confined to Europe and western Asia and so, if they gave rise to modern humans, they should be closest kin to people of these regions. But “Neandertal DNA is equally different from that of people everywhere in the world,” making it unlikely that they were ancestral to Europeans, he says.

Recombination could also cause problems for mitochondrial Eve. Studies of mtDNA from living people on various continents show a surprising homogeneity, suggesting that we are all descended from a woman who lived a mere 200,000 years ago in Africa. But such homogeneity might be due to recombination rather than a common recent ancestor, says Hagelberg. Pääbo concurs. “Mitochondrial Eve is the one woman who carried the ancestral mitochondrial DNA,” he says. “There was no such woman if there was recombination.” Until scientists determine how frequently recombination in mtDNA occurs, however, “it's hard to know how it would affect these types of analyses,” says Penny.

Despite these difficulties, researchers are already working to correct the clock's timekeeping. For starters, clock proponents argue that they can identify genes of variable evolutionary rate and avoid them. “Name a gene and there are going to be lineages that are fast or slow,” says Hedges. “We do a statistical test to identify those problems.” And some genes tend to evolve more steadily, such as those that encode serum albumin and enolase, he says.

Not everyone is convinced that the current statistics are powerful enough. But researchers such as Cooper are modeling the error bars in clock analyses, allowing them to answer well-defined questions with confidence. And researchers such as Jeffrey Thorne at North Carolina State University, Raleigh, are working to beef up the statistics. Thorne and others are developing new tools that allow for different probabilities for various nucleotide substitutions and for changes in the rate of substitutions at certain times during evolution. “One idea is that you expect closely related species to have more similar rates of molecular evolution than more distantly related species,” says Thorne. “So we try that and then use statistical methods that are available to evaluate how well the model fits the data.”

Even as the statistics improve, clock studies are benefiting from all of the sequencing under way, says Hedges. “We're gathering lots of genes from lots of organisms,” he says. Because rates vary along branches in different directions for different genes, “the thing to do is to use a lot of genes.” Far from giving up on clock analyses, researchers are still set on extracting whatever layers of meaning they can.

13. GENETICS

# Y Chromosomes Point to Native American Adam

2. Ricardo Braginski*
1. Hurtado and Braginski are science writers in Buenos Aires, Argentina.

A man who lived some 20,000 years ago carried a set of genetic markers that are now found in up to 85% of native American males

Many of the native peoples around the Americas—from southern Argentina to North America—can trace their heritage back to a single founding father, according to a group of Argentine researchers. Three years ago these researchers, along with another group, discovered the traces of a single common ancestor in the Y chromosomes of native American men in South and Central America. Now they have traced this ancestral father's influence throughout the Americas. The studies by Nestor Bianchi and his colleagues at the Instituto Multidisciplinario de Biologia Celular (IMBICE) of La Plata suggest that the Native American Adam lived roughly 20,000 years ago and is the common paternal ancestor of at least 85% of all native Americans in South America and almost half those in North America.

“We now know that if you are a Native American individual and you have [a particular Y chromosome] mutation, you can trace your ancestry back to a single individual who came to the Americas sometime during the settlement of this continent,” says Peter Underhill of Stanford University, who led the other group that discovered this common Y chromosome variant. The age of the ancestral New World father falls within the broad range of other genetic estimates for when the Americas were first thought to be settled, although there's no undisputed archaeological evidence for settlement before about 12,500 years ago. And the mutation's abundance suggests that many of the first Americans arrived in a single migration from an ancestral Siberian home—a home that other researchers are identifying.

Since 1995, Bianchi and his colleagues have been tracing Native American genealogy by studying variations in the Y chromosomes of volunteer donors from more than 20 native populations around the Americas. Two men who carry the same markers on their Y chromosomes must be descendants of the same male ancestor, says Bianchi, because of peculiarities in the way the Y is inherited. Most chromosomes exist in pairs, which allows them to recombine—swap pieces—during the formation of the sperm cells. As a result, a son who has inherited a particular chromosomal marker from his father won't always pass it to his offspring. But the Y escapes this scrambling, because it lacks a counterpart. “It is transmitted intact, like last names, from the father to the male children, over the generations,” Bianchi says.

Three years ago, the La Plata researchers and Underhill's group independently announced the discovery of a common Y haplotype—a common set of markers—among indigenous groups in Central and South America. They proposed then that most extant male aborigines in this region are the offspring of a single patrilineage. Now Bianchi and his colleagues have broadened their search, analyzing Y chromosomes from a much larger sample—a total of 200 men—from North America as well as South and Central America.

The researchers looked at two different marker sites, each of which exists in multiple variants, or alleles, along with a mutation that Underhill had detected in the Y chromosomes of Amerindian men. In last December's American Journal of Human Genetics they reported what they had found: A specific combination of two marker alleles and the mutation turned up in between 40% and 95% of the men they studied, depending on the population. All of the men who carried this common haplotype, they concluded, are the descendants of a New World Adam who colonized the Americas thousands of years ago.

The next step was to determine how long ago he walked the earth. Although most native American males carry the three markers, their Y chromosomes differ at many other sites, presumably because the chromosomes have gathered mutations since the Adam first passed on his copy. By working out how long it would have taken to accumulate those differences, the researchers could determine how long ago the original father, carrying the ancestral Y, had lived. By using the mutation rate seen in other parts of the Y chromosome, they figured that the New World Adam had lived, very roughly, 22,500 years ago. (The results have wide error bars, from a minimum of 13,700 years to a maximum of 58,700 years).

Bianchi and his colleagues note that Native Americans in North America are less likely to carry the telltale combination of markers than are South American natives, perhaps because North American natives have a greater admixture of African or European genes. The presence of other Y types could also indicate that the first Americans arrived from Asia in several different waves, as some anthropologists have proposed. Indeed, in work that is in press, Andrew Bergen, a researcher at the National Cancer Institute, reports that he has identified a second, less common Y chromosome, suggesting that the New World Adam had a rival.

The La Plata group now hopes to identify specific mutations that took place in the ancestral haplotype after it entered the Americas and correlate them with geographic regions, perhaps finding clues to the migratory pathways of those first settlers. Meanwhile, other researchers are looking farther afield, seeking the original Asian home of the New World Adam.

Two papers in the American Journal of Human Genetics–one published in February and the other this month—compare the Native American Y chromosome with genetic material from natives in Asia. The researchers who led the two teams—Fabricio Santos of the Federal University of Minas Gerais, Brazil, and Tatyiana Karafet of the University of Arizona, Tucson—say that they have identified a set of chromosome markers that looks like the ancestor of the New World Y in two small Siberian ethnic groups: the Kets from the Yenissey River Basin and the Altaians from the Altai Mountains. And in Y chromosomes from populations in other parts of Asia and Europe, they have found clues that the ancestral Y's own precursor originated in central Asia, then in ancient migrations spread eastward into Siberia and as far west as England.

Other genetic evidence has also hinted at a central Asian population that spread both east and west (Science, 24 April 1998, p. 520). Says Andrew Bergen, one of the researchers: “In essence, the ancestral founding Y chromosome found its way to America, and also supplied Europe.”

14. CLIMATE ANTHROPOLOGY

# Taking Global Warming to The People

1. Kathryn S. Brown*
1. Kathryn S. Brown is a science writer in Columbia, MO.

Thanks to ever-finer tools for calculating the costs of climate change, scientists are making a case to deal with global warming now

Cruising down northern Vietnam's Red River on a research boat 2 years ago, Mick Kelly started to get a queasy feeling. It wasn't seasickness that troubled him—it was the sight of a feeble earthen dike meant to protect people south of Ha Long City if a storm-whipped South China Sea were to push the river over its banks. Kelly, an atmospheric scientist at the University of East Anglia, United Kingdom, knew that the risk of a calamity is likely to grow: Computer models suggest the South China Sea could creep tens of centimeters higher over the next century and spawn more frequent and violent storms. Hadn't local officials considered building a better barricade, Kelly asked his Vietnamese colleagues? Their answer: Politicians had indeed debated the project but splurged for new roads instead.

When the 2000 scientists on the United Nations Intergovernmental Panel on Climate Change issued their landmark report 4 years ago spelling out the risks of global warming, they cited possible shifts in vegetation and storm patterns. Now, researchers such as Kelly are focusing on the human dimension of those changes. They have launched dozens of projects to assess how various segments of society—farmers, forestry managers, and politicians, for example—are bracing for future climate events. And even as negotiations drag on over implementing the climate change treaty signed so far by 76 nations, scientists are stepping up efforts to help communities devise ways to cope with, and even benefit from, global warming. In Vietnam, says Kelly, “it's the human dimension that will determine just how much an area remains at risk from sea-level rise, not the physics.”

Although forecasts of regional effects of global warming are still far from precise, last year's vicious El Niño events have helped focus attention on the kinds of local devastation that might be in store. Also highlighting the need for contingency planning is a report released last month by the nonprofit Pew Center on Global Climate Change, which forecasts a bleak future for the U.S. heartland. If global warming shifts agricultural patterns, the report states, some farming communities could become “ghost towns as people seek economic opportunity elsewhere.”

As a sign of the emergence of what some call “climate anthropology,” the National Oceanic and Atmospheric Administration (NOAA) and several other U.S. agencies are placing an increasing emphasis on projects with a human dimension, which are expected to total $85 million next year. “We're finding that the effects of climate look totally different when you add people to the equation,” says Michael Hall, director of NOAA's Office of Global Programs. “A physicist who forecasts that La Niña will cause heavy rains in Indonesia might stop right there.” But if you factor in land-use practices—for example, planting an extra rice crop in the winter season because of the increased rain, or catastrophic floods in a region denuded by clear-cutting—“you begin to get a fuller view of climate's good and bad sides,” Hall says. A developing field. Scientists are hoping to ground such socioeconomic forecasts in lessons drawn from present-day calamities. Working last year in Ceará, a semi-arid state in northeast Brazil, anthropologist Timothy Finan of the University of Arizona in Tucson and his colleagues interviewed some 500 farm families struggling through a year-long El Niño-related drought that scorched the region's corn and bean fields. Finan's team found that most farmers had failed to comprehend—or had not even seen—the jargon-laden drought forecasts issued by the government, so they did not stockpile food. The researchers also observed that the government had distributed food and water only when household shortages threatened mass starvation. “If global warming scenarios are right, they've got to learn how to live in a drought environment,” says Finan, whose group has advised the Ceará government to plan ahead by creating nonfarming jobs, drilling wells for isolated farmers, and desalinating groundwater for drinking. With funding from the World Bank, Ceará's government is taking steps toward implementing some of these recommendations. “In a sense, our job is to get people to plan, to think ahead,” Finan says. “And that's happening right now.” While one Brazilian state is moving to minimize the toll of future droughts, Vietnam's social and economic trends are only exacerbating its vulnerability to flooding. Shore-hugging Vietnamese towns have for centuries relied on dikes and mangrove wetlands for protection from typhoons. Since 1994, Kelly and East Anglia economist Neil Adger have queried 250 farmers and government officials about the upkeep of this storm shield. The duo has documented an alarming trend: People increasingly are clearing coastal mangroves for shrimp farms and croplands. Officials realize the danger of losing mangroves, Kelly says, but they feel pressure to boost the economy. In response, Kelly, Adger, and Nguyen Hoang Tri of the Vietnam National University in Hanoi have analyzed the costs and benefits of conserving mangroves. They found that in providing honey, shrimp, fish, and crabs, as well as a defense against storm surges, mangroves are well worth keeping. “Mangroves are a win-win solution for ecology and economics,” says Kelly, who has shared his findings with local officials. For many developing countries, scaring up the resources to do such analyses is tough. “Climate change is a very different reality in the developing world, much lower on the list of concerns,” says Max Campos, an advisor to Costa Rica's National Climate Change Program. Researchers such as Campos—who has studied the possible effects of climate change on Costa Rican crops and water supply—rely on a trickle of money from the U.S. Country Studies Program, which has spent$40 million over 5 years to help countries assess and prepare for climate change.

In Gambia, for example, program funds have allowed water resource manager Bubu Jallow to assemble aerial photos, topographic maps, and land-use data to estimate global warming's impact on Banjul, the capital. Although seas swelling more than a meter, on average, could submerge the coastal city, Jallow found that smaller rises threaten Banjul's water quality and beach property. His research is meant to undergird Gambia's coastal management plan, which is likely to include sea walls to block the rising tide and lines of palm and mangrove trees to stabilize eroding beaches. “There's little expertise in Africa to develop or apply global warming research,” says Jallow, “and it's difficult to get the funds from national governments. So the Country Studies program is critical.”

Politicians aren't the only people whom researchers must win over in making a case for what life may be like in a warming world. Since 1996, Carlo Jaeger, a sociologist who splits his time at the Darmstadt University of Technology in Germany and the Swiss Federal Institute for Environmental Science and Technology in Duebendorf, has overseen some 300 focus groups in seven European regions. The rap sessions included select groups—venture capitalists and tourist business owners, for instance—or ordinary folks plucked randomly from phone books. Jaeger has found that people of all stripes are surprised at the scientific uncertainty over the magnitude of global warming and the havoc it might wreak. It's essential, he concludes, that scientists explain better how to cope with a changing world.

Calculating the costs. In the United States, some companies seem to be getting the message. Twenty-two firms, including DuPont, Boeing, and Shell, work with the Pew Center to find ways to lessen and adapt to warming. The center's new report, “Agriculture and Global Climate Change,” suggests that although global warming overall will do little harm to the U.S. food supply, certain regions and crops may be in for a big change. Warmer temperatures in North America might nurture crops in northern states but parch southern crops. Wheat yields could drop by 20% and corn by 30%. Soybean production could plummet 40%–or rise 15%, depending on soil and air conditions. Some landscapes could get a new reputation: California's Napa Valley, for example, may no longer be wine country.

Some companies are already positioning themselves to reap global warming's harvest. In 1997, plant biologists at Pioneer Hi-Bred International Inc. in Des Moines, Iowa, began collaborating with a team led by Cynthia Rosenzweig of Columbia University and NASA's Goddard Institute for Space Studies to model how temperature and rainfall changes could alter crop yields and water supplies worldwide. Pioneer Hi-Bred and its partner companies plan to use data from the ongoing project to reevaluate existing markets and tap new ones over the next 30 years. “This research finally has become real, with companies using it to decide where they will strategically place their investments,” Rosenzweig says. “This isn't just an academic model—it's a tool for economic development.”

Scientists hope that communities, too, can make better use of improved climate models. They're now trying to craft Integrated Assessment Models—analyses that estimate global warming costs—that are more realistic, says Hadi Dowlatabadi, director of the Center for the Integrated Study of the Human Dimensions of Global Change, at Carnegie Mellon University in Pittsburgh. In the past, for example, scientists have estimated the economic toll of rising seas by calculating the value of shorefront property lost to encroaching water. But that's too simple, says Dowlatabadi, who predicts that shore homeowners won't just pick up and move because the tide creeps a bit higher each year. (His forecasts, rosier than most, assume that low-lying coastal cities could, for instance, build sea walls to protect property.) Dowlatabadi's models focus instead on the costs that homeowners, insurance companies, and local governments would bear following hurricanes and other damaging storms—events that could increase in frequency with warming. This approach, he argues, better depicts global warming costs and gives coastal communities real-life data they can use to plan shoreline development and disaster strategy.

Researchers also have to do a better job of modeling how people might adapt to warming, says Robert Mendelsohn, an economist at Yale University. In a new book, The Impact of Climate Change on the United States Economy (Cambridge University Press, 1999), Mendelsohn suggests that warming may boost the U.S. economy as farmers and orchard owners, for example, switch to vegetables and fruits that flourish in warmer weather. He predicts that by 2060, U.S. agriculture—thanks to warmer temperatures and more rainfall—could come out as much as \$41 billion a year ahead. That is a far cry from the billions of dollars that most economists predict agriculture will lose to warming every year. Mendelsohn does say that some sectors—such as energy—may suffer, and that his models do not address important quality-of-life issues, such as sweltering heat in the south and disease-carrying mosquitoes invading the north.

Such questions linger, but climate anthropologists appear to be making strides at relating global warming models to everyday lives. Michael MacCracken, head of the U.S. Global Change Research Program's National Assessment, which coordinates studies on the potential regional effects of climate change, recalls a recent meeting with homeowners and officials in New York. At first, city dwellers asked why they should care if the midwestern economy were to slump during global warming. But they soon realized that many of New York's goods—cars, food, and even elevator equipment—are imported from the Midwest. “This is a human ecosystem we're talking about,” in which environmental and economic forces are intertwined, MacCracken says. “People are beginning to understand that.”

15. SCIENCE INTERVIEW

# Claude Allègre: Back to the Wall, But Still Fighting

1. Michael Balter

France's research minister came to his job with grand plans to bring more fluidity to the nation's privileged research system. Although he now faces fierce resistance, he is determined to tough it out

PARIS—When geochemist Claude Allègre was appointed France's minister of national education, research, and technology in June 1997, French scientists thought they at long last had a colleague in their corner. Now many researchers are not so sure. Allègre has challenged many of the sacred cows of the French research system, including the “researcher-for-life” status of scientists employed by large public research organizations such as the Centre National de la Recherche Scientifique (CNRS). The minister's insistence that these civil service researchers develop closer ties with universities and industry has met fierce resistance, including a historic mass meeting of CNRS researchers last December (Science, 18 December 1998, p. 2162). Moreover, Allègre has been fighting on two fronts, as thousands of teachers have taken to the streets in recent weeks to protest his attempts to reform the nation's secondary schools. Some teachers' union leaders have even called for Allègre's resignation.

In an interview in his Paris office, Allègre outlines his rationale for the changes he wants to see in French research, and underscores his determination to push ahead with the reform program. An edited transcript follows.

Q: You are obviously a controversial person in France at the moment. Are you surprised at this strong reaction to your efforts to reform the nation's education and research establishments?

A: No, I am not surprised at all. There is resistance from some of the teachers' unions but, at the same time, the parents and students are completely with me. It is very difficult to reform education in this country, but I have the confidence of the government, so I am not very troubled about the resistance.

Q: Can you imagine any circumstances in which you would resign?

A: No. We are a democratic country, the government is not appointed by the unions. The pressure you are referring to is greatly amplified by the press, but it is not very strong on the street.

Q: At your request, Prime Minister Lionel Jospin recently appointed two Socialist parliamentary deputies to conduct an inquiry into research and the universities. What is the purpose of this inquiry?

A: The reasons are simple. I want to bring the research organizations into closer contact with the universities, and I want researchers to be mobile and not remain full-time researchers for life. That has met great resistance. So I have asked the deputies to study it. They have 6 months to do it. I do not want to negotiate at length with the CNRS about it.

Q: There has been some speculation in the press that the deputies might conduct the national debate on the future of French research that many scientists have called for, but which up until now you have opposed.

A: They can do that if they want. This is clearly stated in their assignment letter. But I am not for a national debate on everything. It should be centered on bringing together the universities and research organizations, and increasing mobility for researchers. I am not opposed to a debate on that, it doesn't bother me. The reason I do not want a national debate is simple, we have already had two [in the 1980s and 1990s], and at the end nothing happened. We debate a lot and then nothing changes. The reforms will come, I have no worries about that. But making reforms is difficult.

Q: You have used the phrase “revolutionaries of the status quo” to characterize some researchers who have actively opposed your reforms. What do you think is behind this resistance? Are the researchers sincere about wanting change but not in agreement on the details, or do you think there is a real resistance to change?

A: Oh yes, there is a real resistance. France, and also Italy, are the only countries to have this kind of system, with full-time researchers for life [civil servant scientists]. That does not exist in the United Kingdom, nor in Germany, nor in the United States or Canada. Our researchers think it is just fine. They don't want to have to teach. They don't want to have to go into industry. There is a real fundamental resistance there. Other [ministers] have tried before me, and they have never really succeeded.

My goal is not to squelch the researchers. I find that our system of having full-time permanent researchers for a period of their lives is a good one. In the United States this does not exist, but I am not sure that the American system is better, because the researchers are always under a great deal of pressure. For part of one's life, it is a good thing to do research without being under pressure. But there are very few people who can do full-time research all their lives and still be good at it.

Q: So you want some of the older researchers to step aside in favor of the younger ones?

A: I want experienced researchers to think about continuing their careers as both researchers and professors. There are two issues with respect to young researchers. The first issue is to hire them. The second, after they are hired, is to give them the freedom to be scientifically autonomous. In the United States, when you are an assistant professor, you write your own grant proposals. In France, you are in a laboratory, there is a director, and the director is in charge of everything. In the United States that situation exists in certain sectors, for example high-energy physics or certain sectors of biology, but it is rare. I have asked each [French] research organization to create a program for young researchers, and they all have done it. I am also creating a national program for young scientists, a competition based on grant proposals, and the researchers who win this competition will have the right to one or two postdocs and access to equipment. This will favor the blossoming of young research teams. I want to rejuvenate French research.

Q: If money were given directly to young researchers, that would be a radical reform in France. Do you think these measures could finally break the “mandarin system” that you and others have said rules French research?

A: Yes. No one dares to openly oppose these measures, but in fact there is a great deal of resistance.

Q: Resistance from directors of laboratories, of institutes …?

A: Yes, from the directors, but not only the directors. Also from scientists who do not do much research anymore themselves, but who need young researchers to do it for them.

Q: Many French researchers say “yes, we are for reforms.” But there is a perpetual complaint about the lack of adequate funds to keep laboratories going, and a great fear that if researchers must find more and more of their funds from contracts with industry, then basic research is in danger.

A: No, fundamental research is the first priority for us. And we have asked the research organizations to increase the [proportion of their budgets used] for running the laboratories. All labs should receive a budget increase of 10% by the year 2000. But the CNRS has allowed itself to fall into a spiral where 80% of its budget is taken up by salaries. The CEA [Atomic Energy Commission] was in the same situation 10 years ago, but the CEA has made a choice, it has reduced its personnel.

Q: Scientific personnel?

A: Yes, to give more funding to the laboratories. The CNRS has not had the courage to do that. [The ministry later clarified that it is not its policy to encourage the CNRS to reduce researcher numbers.]

Q: You have said that you want to bring the research organizations and the universities together. But 80% to 90% of CNRS laboratories are already located on university campuses and many graduate students do their theses in CNRS labs. What specifically do you want, for example, in the case of labs that are already at the universities?

A: For those labs already at the universities, nothing much will change. But I want more CNRS researchers to become professors, and I want a certain number of professors to spend their sabbatical years in the CNRS. I want the collaboration between the two sectors to be more intimate. They are still too separate for my taste.

Q: At the CNRS mass meeting last December, chemist Henri-Edouard Audier said that if the teaching hours of university instructors were cut in half, all the problems of mobility between the CNRS and the universities would disappear. How do you respond?

A: We are trying to lighten their load, it is a real problem. Especially for the maîtres de conferences [assistant professors] who do research. We are studying that, but it is not easy.

Q: You have recently talked about giving the universities a percentage of the public research budget, an overhead, to incite them to recruit the best research teams to their campuses, something that is not done now. But since virtually all the universities belong to the state, can they compete with each other for the best scientists, as American universities do, for example?

A: This happens even now in France. A university can offer a post of professor to someone who is a maître de conferences elsewhere. So an attractive promotion is a possibility. What is different is that American culture is a culture of mobility and taking risks, and French culture is not. We are not the descendants of those who crossed the Atlantic, we are the descendants of the ones who stayed behind. American culture is to go West in covered wagons, to push back the frontier. Our culture is to construct a house as quickly as possible … and never a wooden house like in America, but in stone. It's a static symbol. The American builds his house out of wood because he knows he is not going to live in it his whole life.

Q: You have earned a reputation for speaking very frankly, and some have even called you a provocateur. Are you doing it deliberately to make things move?

A: Yes. My answer is yes.