News this Week

Science  08 Dec 2000:
Vol. 290, Issue 5498, pp. 597

You are currently viewing the .

View Full Text

Log in to view the full text

Log in through your institution

Log in through your institution


    Asia Stays on Top, U.S. in Middle in New Global Rankings

    1. Constance Holden*
    1. With reporting by Michael Baker in Seoul.

    Eighth graders from the United States are still running in the middle of the global pack when it comes to science and math achievement, according to the latest results from the Third International Mathematics and Science Study (TIMSS). And Asian nations continue to lead the way, with Singapore and Taiwan emerging as the star performers among the 38 participating countries.

    The original TIMSS issued three reports, starting in 1996, on fourth, eighth, and 12th graders. The new findings, called TIMSS-R (for repeat), provide longitudinal data that allow countries to measure their progress over time (

    The news is not good for U.S. science and math educators, who have spent much of the decade pursuing reforms aimed at raising student achievement. Today's eighth graders look pretty much like the ones tested in 1995 (Science, 22 November 1996, p. 1296). “We were particularly interested” in seeing how student cohorts tested as fourth graders in 1995 did as eighth graders in 1999, says Larry Suter of the National Science Foundation.

    Compared with the 16 other countries that were in TIMSS both years, the U.S. cohort is the only country to show a “significant drop” in both science and math achievement as its students mature. Whereas in 1995 U.S. fourth graders tied with Austria for third place in science and were above average in math, they had slipped to below average in both subjects by the time they reached eighth grade. “It just shows that other countries do a lot better job of educating their students between elementary school and middle school than we do,” says Suter. The 19th-place ranking in science comes despite the fact that U.S. eighth-grade teachers are more likely to teach concepts in biological and physical sciences than their international peers.

    The hope that U.S. students would show benefits from the reforms is “a hope gone awry,” laments statistician William Schmidt of Michigan State University in East Lansing, who directs the U.S. portion of TIMSS. “Basically, we have not changed the middle school curriculum in any systematic fashion.” Although educators agree that U.S. math and science education need to be more “rigorous,” they fight over the definition. Schmidt sees “rigor” in terms of course content: Reform efforts haven't worked “because we haven't looked at the core issue of substance, rather than pedagogy,” he says. But Arthur Eisenkraft, president of the National Science Teachers Association, believes it's more about process, including “high expectations for students, experiments … and exams that not only test for knowledge of facts but how we know them.”

    One apparent irony on the international front is that some Asian countries are trying to move away from lectures and rote learning despite its apparent success in boosting test scores. Education officials in South Korea, for instance, told Science that “the students are all well trained to do well on tests,” so their impressive TIMSS scores “don't really mean much.” Educators there are focused instead on shortening the all-important college entrance exam and giving students more time for creative thinking.

    Singapore, at the top of the heap, has several factors in its favor, says Tham Tuck Meng, principal of one of the participating schools. That includes a rigorous curriculum, well-trained teachers, supportive parents, and abundant resources. “Principals have wide autonomy to plan curriculum according to student needs,” adds Tham about the 3.5-million-person city-state, the financial hub of Southeast Asia.

    One thing that emerges from this latest assessment is the relative insignificance of technology. The use of calculators in math class—a topic hotly debated in the United States—doesn't seem to shape the rankings. Most students in Hong Kong, which ranked fourth, use them, for example, but calculators are rare in Taiwan, Japan, and Korea. Around the world, the study found that computers are also not major pedagogical tools. One-quarter of students have Internet access, but only about 10% use it for class work in math or science. U.S. educators may get a better idea of what works and what doesn't in April, when the TIMSS-R results from more than two dozen states and districts are released.

    View this table:

    Panel Urges Further Study of Biotech Corn

    1. Jocelyn Kaiser

    This fall, when genetically modified corn that had not been approved for human consumption was found in taco shells in the United States, all hell broke loose. The corn, produced by Aventis CropScience for animal feed, was soon found in a range of foods, prompting product and grain recalls countrywide. Last week, the U.S. Environmental Protection Agency (EPA) convened a scientific panel to review evidence on whether this StarLink corn could harm sensitive people by causing allergic reactions. The panel found that the overall probability is “low,” but the EPA appears intent on investigating further before allowing the corn in foods, as Aventis has requested.

    At the center of the controversy is a bacterial protein known as Cry9C, the gene for which was added to StarLink corn to make it resistant to insects. Cry9C is one of several so-called Bt proteins, but it is more heat stable and harder for humans to digest than its kin—qualities that are typical of such allergens as peanuts. Although comparisons of Cry9c's structure with known food allergens turned up no signs of allergenicity, EPA's scientific advisers don't consider those tests conclusive. In 1998, when Aventis asked EPA to approve StarLink for human consumption, the agency limited its sale to animal feed and industrial uses.

    Then came the September scare when a coalition of activist groups detected the DNA coding for Cry9C in taco shells. Aventis soon pulled its seeds off the market. But facing the multimillion-dollar costs of tracking StarLink corn already in grain elevators and silos, the company asked EPA to allow it in food for the 4 years it will likely remain in the food supply.

    To bolster its case, Aventis gave EPA new studies showing that the blood of people with food allergies doesn't cross-react with Cry9C. They also presented evidence that the amounts of the protein showing up in contaminated food would be perhaps 1/100 the level needed to sensitize people to allergens such as peanuts. But the review panel didn't find these arguments persuasive. It's unknown exactly what makes a protein trigger allergic reactions, they argued, and allergists don't know whether there is a safe level for the protein. “Unfortunately, there are no valid animal models that will tell you that something's an allergen” in people, says Hugh Sampson, a New York University allergist on the panel.

    Evidence on Cry9C's allergenicity may be coming soon, however. Following the flurry of press reports over StarLink, at least 34 people reported to the U.S. Food and Drug Administration (FDA) that they had allergic reactions after eating corn products. The panel concluded that up to 14 of these 34 reports merit further investigation. The cases “have not been corroborated medically or scientifically,” notes allergist and panel member Marc Rothenberg of the University of Cincinnati. To begin to do that, FDA and the U.S. Centers for Disease Control and Prevention hope to test the blood of these 14 or so people for the presence of antibodies to Cry9C. “If even two have them, then that sort of gives you the answer” that Cry9C has the potential to cause allergies, says Sampson. But developing the protocol and test could take several months, says Karl Klontz of FDA.

    Whether EPA will grant Aventis the 4-year exemption is anyone's guess, but some observers don't expect a quick decision. Despite the panel's finding that the overall risks from Cry9C appear low, it urges agencies to follow up on illness reports and get better data on protein residues in foods. “A thorough assessment” will continue, says an EPA statement. “It's clear that EPA is going to sit on this petition for a while,” says Rebecca Goldburg of Environmental Defense. Whatever EPA decides, the agency will likely be cited in lawsuits now proliferating by people who allege allergic reactions from StarLink; some corn growers have also filed a class action suit, claiming they weren't warned about mixing StarLink corn with other corn. Last month, Cry9C did turn up in another variety, for reasons Aventis can't yet explain.


    Polio Outbreak Raises Questions About Vaccine

    1. Liese Greensfelder*
    1. Liese Greensfelder is a writer in Nevada County, California.

    The oral vaccine designed to protect children from polio has been fingered as the possible culprit in a recent outbreak of the devastating disease in the Dominican Republic and Haiti. The small cluster of cases marks the first polio outbreak in the Western Hemisphere in more than 9 years. It is also the first reliable report that a vaccine-derived polio strain may have reverted to a virulent form and spread contagiously. Although a massive vaccination campaign already in the works is expected to contain the outbreak, the unusual incident raises troubling questions about a vaccine that has been in widespread use for nearly 40 years.

    Three children in a rural area of the Dominican Republic about 80 kilometers from the Haitian border came down with paralytic polio in July and August. A single case has also been confirmed in Haiti, where the disease struck a child who lives 3 hours' hike from the nearest road. None of the children had been vaccinated; that means they acquired the vaccine-derived virus from someone who had. Although both regions are remote, there is enough traffic between them that person-to-person transmission could have occurred, says Ciro de Quadros, who directs the vaccines and immunization program for the Pan American Health Organization (PAHO).

    The oral polio vaccine is highly effective and easily administered. But because it employs live but weakened strains of the virus, its use results in vaccine-associated paralytic polio in about 1 of every 750,000 people who receive it, usually those with compromised immune systems. But there has been no evidence that this vaccine-induced form of the disease can be spread from person to person. The new outbreak demonstrates that the vaccine can indeed mutate to a virulent form and spread from person to person.

    When the laboratory that conducted the routine surveillance tests of one of the Dominican Republic cases found a virus that was unsettlingly similar to wild poliovirus type 1, health officials quickly sent samples to the U.S. Centers for Disease Control and Prevention in Atlanta. There, scientists sequenced the virus—and later samples from the three other victims—and found that they all significantly diverged from the vaccine strain and now clearly resembled the virulent wild-type. It's not yet known how this reversion occurred. PAHO has convened a group of scientists this week to study the data and recommend subsequent action.

    Although 73% of children under age 1 were vaccinated in the Dominican Republic in 1998, only about 20% of children had been vaccinated in the region where the outbreak occurred, says de Quadros. The World Health Organization (WHO) is striving for at least 90% coverage in its efforts to eradicate polio from the world by 2005. Whether this incident will delay the timeline remains to be seen, says Donald Henderson, an epidemiologist at Johns Hopkins University in Baltimore, Maryland, who led WHO's successful global eradication of smallpox. “We really want to zero in on this and check it out in great detail,” Henderson says. “It's now 9 years since we've detected any circulating wild virus in the Americas. This comes as a great surprise to everybody.”


    EMBL Rescued From the Financial Brink

    1. Lone Frank*
    1. Lone Frank is a writer in Copenhagen, Denmark.

    COPENHAGEN—Scientists at the European Molecular Biology Laboratory (EMBL) were breathing a sigh of relief last week after the topflight center in Heidelberg, Germany, announced a 25% budget increase for the next 5 years. The boost eases months of uncertainty over how the lab would comply with an order to pay employees back salary and provides a measure of stability to its renowned but embattled European Bioinformatics Institute (EBI).

    EMBL's governing council has approved a spending increase of $10 million a year, raising the lab's budget to nearly $50 million in 2001. “This is a major vote of confidence for EMBL,” says director-general Fotis Kafatos. Much of the money goes to bailing out EBI. The bioinformatics institute in Hinxton, U.K., has struggled to pull in enough funding under unfavorable European Union guidelines that tend to be more generous to investigator-initiated projects and to neglect research infrastructure. Some $6 million a year—60% of EMBL's budget increase—will go to EBI, covering 40% of its costs. Kafatos says he's confident that “the rest will come from outside,” citing possible collaborations with the U.K. Medical Research Council and other European funding agencies. The budget boost also will allow EMBL to establish a full-fledged center for mouse biology at its outstation in Monterotondo, Italy.

    Long a center of basic research, EMBL now is trying to find ways to market its findings for income. Enthusiastically endorsing this new direction, the council of 16 member states approved the establishment of an externally managed venture-capital fund and the construction of a 6600-square-meter International Technology Transfer Center in Heidelberg that will serve as an incubator for start-ups from EMBL and member states. “EMBL has not made the best possible use of technology transfer in the past,” says council chair Peter Gruss, a biologist at the Max Planck Institute for Biophysical Chemistry in Göttingen. Kafatos insists that EMBL's “academic culture will not be negatively impacted” by teaming up with industrial partners.

    The council's benevolence eases jitters over a court ruling last year that forced EMBL to pay back salary to dozens of employees, prompting fears that the lab might have to make deep cuts in research or even shut down (Science, 5 November 1999, p. 1058). In the wake of the generous budget increase, “people are much more optimistic,” says scientific coordinator Iain Mattaj. Indeed, adds Kafatos, “the outlook is not as unstabilizing as last year.”


    Program for Elites Draws Praise, Fire

    1. Richard Stone

    DRESDEN—The European Commission (EC) has launched a new program to bring western European scientists to elite research institutions in central and eastern Europe. But the German administrator who was instrumental in persuading the EC to consider such a program now contends that the money fails to address the real needs of such institutions, which are hungry for first-rate labs and equipment.

    Four years ago Wolf Lepenies, rector of the Institute for Advanced Study in Berlin, approached the EC on behalf of the Collegium Budapest, a multidisciplinary institute in Hungary. His modest plea for funds to host Western scholars for extended visits and put on international conferences eventually blossomed into a program unveiled here last week to support 34 top research centers in 11 nations that are candidates to join the European Union, from the Czech Republic to Lithuania. Each center, selected from 185 applicants, will get between $90,000 and $900,000 over 3 years to fund academic exchanges.

    By bolstering a handful of scientific bastions that have earned international reputations, the new “Centers of Excellence” program offers an oasis of stability for institutes that can't rely on substantial support in countries with fragile economies. The program, says Jerzy Langer of the Institute of Physics in Warsaw, “is really a far reaching and very wise move of Brussels,” headquarters for the EC. Encouraging mobility to these institutes “is a very important trend,” believes David Schindel, director of the U.S. National Science Foundation's Europe Office.

    But Lepenies and others say that its focus is too narrow. “We were thinking of creating a very specific atmosphere, supporting centers with a mix of disciplines,” says Lepenies about his idea, which was discussed at a 1997 meeting in Budapest. But to EC officials, he says, “a center of excellence just meant an excellent center.” Several meetings with former EC research chief Edith Cresson failed to alter this view, Lepenies says: “She never understood what it meant and didn't want to understand.” EC officials have a different view. “Brussels is not the place to create an atmosphere,” says Barbara Rhode, who points out that the EC strove for a broader vision than funding only the Collegium Budapest. “We have a responsibility of fairness and justness to everybody,” she says. Last spring the commission selected the winners—chosen according to criteria that included how well they are managed and the potential economic impact of their research on their home countries—but spent months hammering out contracts before money finally started flowing to some centers in November.

    Describing the program here at a NATO forum organized by the journal Nature, Rhode noted that the peer-reviewed selection process bared some disturbing trends in central and eastern Europe. Whereas a large share of proposals from bioscience centers won grants or just missed the cut, the physical sciences fared poorly. “We have the feeling that [the region's] physics, materials science, and engineering are getting weaker,” she says. Posting the worst showing were social sciences and humanities centers: Of 32 applicants, only two, including the Collegium Budapest, won funding. At the same time, countries that had been more aggressive in overhauling their scientific establishments after the Soviet Union dissolved tended to reap the biggest dividends: Four of five Estonian centers that submitted proposals got high marks, for example, and two won funding.

    Now that Brussels has put these 34 centers on a pedestal, observers say, it should also help out on big-ticket items such as new buildings and instrumentation. “Many centers need money for upgrading machines, not networking,” says Simeon Anguelov, a consultant to UNESCO's Office for Science and Technology for Europe. Although only E.U. member states are eligible for structural funds, the EC's Rudolf Meijer says that Brussels is considering other mechanisms for candidate countries. “This is certainly not something we have neglected,” he says.

    Lepenies isn't counting on Brussels to come through. He's helped organize a handful of institutes into a network called Agora, after the ancient Greek word for a meeting place. Agora will soon identify worthy projects, then lobby agencies and private donors for funding.

    Others, however, want Brussels to incorporate the centers program into its next 5-year research program, Framework 6, which will begin in early 2003. “It would be rather stupid not to continue it,” says Anguelov. “Forget all critics, forget all troubles,” adds Langer, “we really have to pray for its extension.”


    New 'Replicon' Yields Viral Proteins

    1. Eliot Marshall

    Twelve years ago, researchers identified hepatitis C virus (HCV) as the elusive pathogen that was causing liver disease in some people who had received blood transfusions. That had an immediate benefit: Donated blood is now screened for the virus, reducing the number of new HCV infections. But an estimated 1% of the U.S. population—more than 2 million people—has already been infected with the virus, which can persist in the body for many years. Most aren't aware of it, because the virus often produces no symptoms, or mild ones. But over time, it can damage the liver and increase the risk of cancer. Unfortunately, efforts to determine just how an infection develops—and how to combat it with antiviral drugs—have been frustrated by HCV's stubborn refusal to grow in the laboratory. Now, a team at Washington University in St. Louis led by Charles Rice has partly overcome that problem: On page 1972 of this issue, the researchers report the creation of an improved viruslike “replicon” that produces HCV proteins efficiently in the lab.

    The innovation has piqued the interest of both industrial and academic scientists, who are racing to develop treatments. It should make it possible “for investigators to study the effects of antiviral drugs and host control mechanisms that regulate HCV replication,” says Frank Chisari of the Scripps Research Institute in La Jolla, California.

    Rice's new system also may elbow its way into a highly contentious industrial arena, where biotech firms and pharmaceutical companies are battling for priority. Already, Chiron Corp. of Emeryville, California, has filed a string of lawsuits to protect its patents on HCV (Science, 2 July 1999, p. 28). Rice has begun making his replicon system available through a small company he founded, Apath LLC of St. Louis. Other groups are said to be developing similar replicons, and these, too, could become available. Apath is offering nonexclusive licenses to use its so-called “Blazing Blight 7” technology. The name refers to the system's blazing efficiency in producing HCV proteins and to the co-inventor at Washington University, Keril Blight.

    Rice, now at Rockefeller University in New York City, says his replicon builds on earlier work by Ralf Bartenschlager and colleagues at the Institute for Virology at Johanes-Gutenberg University in Mainz, Germany. In work reported last year (Science, 2 July 1999, p. 110), Bartenschlager's team took the HCV genome apart and reassembled it into a replicon, editing out parts and adding new pieces, including an antibiotic resistance gene that can be used to select the cell clones that produce the viral proteins.

    Replicons have drawbacks, however. Although they include genes that regulate some host-pathogen interactions, they do not include key genes that enable the virus to infect human cells and replicate normally (see diagram). And according to Rice, Bartenschlager's initial system is inefficient, producing HCV proteins in only about one in a million host cells.

    To improve the efficiency, Rice and Blight rebuilt the system using Bartenschlager's data from GenBank, looking for genetic mutations that might enable the replicon to be more productive. They found 10 interesting mutations, one of which, called S1179I, was outstanding. Replicons with this mutation produced abundant viral proteins in one out of 10 host cells. “That really makes a big difference,” Rice says. “It means you can do experiments over a long term” without having to rely on cumbersome cell selection techniques, and “it is going to allow us to do genetic studies on a much shorter time scale.”

    Chisari says that the Rice team's work provides “a major improvement in the efficiency of the replicon system that Bartenschlager developed.” He and Stanley Lemon, dean of medicine at the University of Texas Medical Branch at Galveston, note, however, that it will be important to develop a system that can produce all the important HCV proteins. Both Bartenschlager and Rice are working on such projects.

    Academic researchers also hope that the Apath replicon will be easy to obtain. Until now “it's been hard to get a system that was widely enough available so that people could play with it,” Lemon says. It will be great news, he continues, if this innovation means that the technology will now be widely available. Bartenschlager could not be reached for comment.

    Asked whether Apath would seek restrictions on academics' use of the new technology, Rice said he does not want to do anything that would “impede academic research.” Apath may ask for a 30-day prepublication review of scientific papers written by those who use the technology. And it may request that such investigators who produce patentable discoveries negotiate first with Apath on the intellectual property rights. But aside from that, Rice says, “I think that sharing material for academic research should be done with as few strings as possible.”


    Tiny, Feathered Dino Is Most Birdlike Yet

    1. Erik Stokstad

    Last November, in what was to prove an embarrassing blunder, National Geographic magazine trumpeted the discovery of a “missing link” between birds and dinosaurs. Archaeoraptor turned out to be a primitive bird with a dinosaurian tail glued on (Science, 14 April, p. 238). Even so, many paleontologists saw a silver lining in the debacle: The chimera consisted of two partial specimens interesting in their own right. Now Chinese paleontologists describe the dinosaurian half, and they say it reinforces the same message as the falsified Archaeoraptor did: Birds evolved from dinosaurs.

    The new fossil, dubbed Microraptor, is by far the smallest adult dinosaur yet discovered—about the size of a crow. Like some other dinosaurs from the fossil beds of northeastern China's Liaoning Province, the creature sported feathers on its body. It also has skeletal features that tighten the link between dinosaurs and birds, including clawed feet that may have been adapted for perching on branches. Such a lightweight creature was almost certainly a nimble climber, says Larry Witmer of Ohio University's College of Osteopathic Medicine in Athens. “I can't see how you could keep this thing out of trees.”

    Microraptor came to light when Xu Xing, a paleontologist at the Institute of Vertebrate Paleontology and Paleoanthropology (IVPP) in Beijing, heard about a dinosaur that Liaoning farmers had unearthed months before in the same fossil beds where Archaeoraptor had been found. Xu bought the specimen and some others for about $5000 and took them back to IVPP. He had studied Archaeoraptor, and when he examined the new dinosaur, the tail looked familiar. “I suspected that [the tail of Archaeoraptor and Microraptor] might be the same thing,” Xu says, “but I could not believe things like this would happen.” Xu contacted National Geographic, which published a retraction. Meanwhile, Xu and his IVPP colleagues Zhonghe Zhou and Xiaolin Wang continued to study Microraptor.

    Xu, Zhou, and Wang report in this week's issue of Nature that the creature belongs to the dromaeosaurs, dinosaurs that many paleontologists consider the closest dinosaurian relatives of birds. (Other paleontologists suspect that Microraptor may belong within another group of birdlike dinosaurs, the troodontids.) Like Sinornithosaurus, another dromaeosaur from Liaoning, Microraptor is surrounded by carbonized impressions that resemble the contour feathers covering the bodies of modern birds. But Microraptor does Sinornithosaurus one better: Close to its thigh, the impressions appear to stick out from a central shaft, called a rachis, which is a defining feature of true feathers. “These structures look exactly like feathers preserved on the bodies of birds in the same rock,” says feather expert Rick Prum of the University of Kansas, Lawrence. Bird-dino skeptic Larry Martin, also of the University of Kansas, says he sees only “broad filaments of something.”

    But Microraptor has other features that ally it with birds. For one, it is the first known adult dinosaur smaller than Archaeopteryx. (Fusion of certain bones indicates that the specimen was more or less full grown.) Its 47-millimeter-long trunk would fit in the palm of your hand. A dinosaur that small—unlike other known birdlike dinosaurs, such as the 2-meter-long Velociraptor—would have already accomplished a key step toward evolving into a creature light enough to take wing.

    The feet, too, resemble those of Archaeopteryx. Microraptor's claws are highly curved, and the first digit is positioned farther down the foot than in other dromaeosaurs. Tom Holtz of the University of Maryland, College Park, suggests that Microraptor's ancestors “may have been in trees long enough that it had developed specific traits associated with that life habit.” Witmer says that adaptation fits nicely with the idea that flight evolved not on the ground but in tree-dwelling animals—a minority view that he shares with Zhou and a few others.

    No one is sure about Microraptor's climbing habits. “Just because you have a curved claw doesn't mean you have to be up a tree,” says Jim Clark of George Washington University in Washington, D.C. “We don't know enough about what these animals were doing, and we certainly don't know enough to test ideas about whether the ancestors of birds were arboreal or cursorial.” With the controversy of Archaeoraptor just behind them, paleontologists may take a while to decide whether to place Microraptor out on a limb.

  8. JAPAN

    Human Cloning Ban Allows Some Research

    1. Dennis Normile*
    1. With reporting by Ding Yimin in Beijing and Michael Balter in Paris.

    TOKYO—Japanese legislators last week approved a ban on human cloning that leaves room for the use of certain techniques in basic research. The action comes at the same time officials in two other countries—China and France—aired similar proposals that would prohibit so-called reproductive cloning while recognizing the possible importance of the technology in combating disease and improving human health.

    The Japanese law, passed by the Diet on 30 November, bars the implantation into a human or animal womb of any embryos produced by transferring a human cell into an enucleated egg. It also proscribes implanting hybrid embryos, produced by matching a human cell with an animal egg or vice versa, and chimera embryos, in which early human and animal embryonic cells would be fused. Violators would face penalties of up to 10 years in prison or a fine of up to $90,000.

    At the same time, the Japanese law does not forbid creating such embryos. “I think it is a very reasonable bill, in that it will allow basic research to continue,” says Takashi Yokota, a researcher at the University of Tokyo's Institute of Medical Science, who is planning to use human stem cells to study basic stem-cell mechanisms.

    An official of the Science and Technology Agency says the bill was deliberately worded to allow in vitro research using cloning techniques and human cells. “There is a possibility of this technology having important medical research applications,” he notes. However, any such research will require the approval and oversight of a national review board. A government advisory panel is expected to work out details within the next year.

    On 2 December, a panel of scientists advising China's human genome project held an open discussion of issues relating to genetic research, during which several panelists expressed concerns about the dangers of cloning. In 1997 China's health minister announced that the government was opposed to human cloning and that scientists and doctors were not allowed to participate in any research toward that goal, although social scientist Qiu Renzong, chair of the genome project's ethics committee, noted that research on technology to generate human organs for use in medical treatment is not prohibited. The government is also drafting a law that would regulate genetic research in plants, animals, and humans. The proposal is expected to be presented next year to the State Council, says Wang Hanpo of the Ministry of Science and Technology.

    In France, Prime Minister Lionel Jospin announced on 28 November that the government was preparing bioethics legislation that would modify a 1994 law prohibiting all research on human embryos. The changes would preclude reproductive cloning but allow research on embryonic cells under strict conditions, including a requirement that the cells come from “spare” embryos from fertility clinics that would otherwise be destroyed and that the parents give full consent. The bill is expected to be introduced in the spring.


    Studies Trace Patchwork of Conflict Policies

    1. Bruce Agnew*
    1. Bruce Agnew is a writer in Bethesda, Maryland.

    For more than a year since the September 1999 death of a teenager in a gene-therapy clinical trial that had industry connections, scientists, ethicists, and government and university officials have been fretting about conflict of interest. But no one could say just how the nation's universities are policing such conflicts. Now, the answer is finally emerging: It all depends on a researcher's home university.

    Requirements for disclosure of outside financial interests vary widely from university to university, according to three recently published studies. Penalties for violations vary, too. Although most policies provide penalties ranging all the way up to termination, university officials have broad discretion in imposing them. All three research teams—headed respectively by S. Van McCrary of Baylor College of Medicine in Houston, Bernard Lo of the University of California, San Francisco, and Mildred Cho of Stanford University—found existing policies wanting.*

    McCrary's team collected information from 250 institutions that received at least $5 million in grants from the National Institutes of Health (NIH) or the National Science Foundation in 1998—about an 85% response rate. Cho's group targeted the top 100 institutions receiving NIH grants that year and obtained information from 89. Lo's team analyzed the policies of the top 10 medical schools in terms of NIH grants.

    View this table:

    Astonishingly, 15 institutions that received NIH grants in 1998 told McCrary's group that they have no conflict-of-interest policy—a flat violation of the 1995 Public Health Service (PHS) conflict-of-interest regulations. NIH deputy director for extramural research Wendy Baldwin calls this “a little mystifying”—and she is looking into it. Among those that have a policy, 91% in McCrary's survey adopt the PHS threshold that requires disclosure to local administrators of financial interests of $10,000 in annual income or equity, or 5% ownership of a firm whose prospects might be affected by a scientist's research. But only 34% require disclosure of research support from corporations, 73% ask about intellectual property such as patents, and 89% demand disclosures of potential conflicts relating to a researcher's spouse and minor children.

    In Cho's survey, 49 institutions require all faculty members to disclose financial interests, but 40 ask for such data only from principal investigators or those conducting research. Only 78 require family members to disclose. Several institutions ban any financial interest in companies that might be affected by a researcher's work, but one allows a researcher to own up to 50% of such a company.

    Lo's top 10 medical schools require all faculty members to file disclosures. Five demand information on all financial interests, not just interests above the federal threshold. One bans any financial ties to a company that manufactures the drug or device being studied in a clinical trial. Another prohibits trading in stock or stock options while a trial is under way (a practice that, at some point, would also violate Securities and Exchange Commission insider trading rules). Six require disclosure to the Institutional Review Boards; two of these demand that clinical-trial patients be told, too.

    Lo and his colleagues believe that university clinical researchers should be prohibited outright from holding any stock, stock options, or decision-making position in a company that might be affected by the trial. That standard is endorsed in a separate editorial in the same issue of The New England Journal of Medicine (NEJM) by Greg Koski, director of the Office for Human Research Protections in the U.S. Department of Health and Human Services, and Jeffrey Drazen, editor-in-chief of NEJM. Right now, it would be a hard standard to enforce: According to Lo and his colleagues, only one of the 10 schools even comes close.

    • *The McCrary and Lo studies appeared in the 30 November issue of NEJM, and Cho's study appeared in the 1 November issue of the Journal of the American Medical Association.


    Clinton Creates Huge Hawaiian Coral Haven

    1. David Malakoff

    A 2000-kilometer-long necklace of remote, coral-fringed Pacific islands has become the United States' largest ever protected area. President Bill Clinton last week signed an executive order creating the Northwestern Hawaiian Islands Coral Reef Ecosystem Reserve. Marine scientists say that the mega-reserve, which holds nearly 70% of the nation's reefs, will help protect some of the globe's most pristine ocean habitats.

    Coraled jewels.

    The new Northwestern Hawaiian Islands Coral Reef Ecosystem Reserve holds 70% of U.S. reefs.


    Last May, acting on a suggestion from a White House Coral Reef Task Force, Clinton ordered Administration officials to devise a plan for protecting the chain of small, mostly uninhabited islands that harbor endangered monk seals, sea turtles, and other fragile populations. After months of meetings with local fishing, conservation, and tourism interests, a committee crafted a plan designed to reconcile opposing views while strengthening protection.

    The new 35-million-hectare reserve, for instance, will allow commercial, recreational, and native Hawaiian fishers to maintain current catch levels, but will ban most other new exploitation, from drilling to coral harvesting. In addition, about 5% of the reserve will be set aside in 15 “preservation areas,” where most activities will be severely restricted. A council of scientists, fishing officials, and state leaders will shape its future, including a research agenda.

    The new reserve encompasses reefs that are “relatively undisturbed,” says invertebrate zoologist Scott Goodwin of the Bishop Museum in Honolulu, a recent visitor during a research expedition ( “This is what the main Hawaiian islands must have looked like before there was so much impact from human habitation,” he says. “A lot of species are probably going to benefit.”


    Crystals Branch Out Into Exotic Shapes

    1. Robert F. Service

    When it comes to designing individual molecules, chemists are remarkably adept at crafting complex shapes, such as the twists and turns of drug molecules that dock into precise cavities of proteins. But when researchers try to extend that complexity into larger solid materials such as crystals, frustration carries the day. Crystals tend to grow on all sides at once. As a result, they typically adopt simple shapes, such as sheets, spheres, and cubes—the delightful branches of a snowflake notwithstanding. Researchers recently managed to steer crystal growth to make thin rods by controlling the speed at which various faces of crystals grow. Now, a team from the University of California, Berkeley, has elevated this control to an art form, enabling them to make nanosized versions of a variety of shapes including rods, arrows, teardrops, and even four-armed tetrapods shaped like a child's jack.

    “It's really beautiful science,” says Jim Heath, a chemist at the University of California, Los Angeles, and a specialist in making nanoscale electronic circuitry. “Learning how to control size and shape is a tremendous challenge at this length scale.” While these shaped crystallites remain a curiosity for now, their unique shapes could be a boon for researchers trying to shrink computer circuitry and magnetic data storage to nanoscale dimensions.

    With such applications in mind, the Berkeley researchers—graduate students Liberato Manna and Erik Scher, working with lab director Paul Alivisatos—started by looking for a better way to grow inorganic rods. In recent years, numerous groups have shown that they can grow miniature forests of tiny inorganic trees on a surface from a vapor of precursor compounds. As the inorganics begin to link up, they form a crystallite in which the top surface grows faster than the slightly different crystalline arrangement of atoms on the sides, causing the crystallite to elongate into a rod.

    One problem, Alivisatos says, is that because the inorganic trees start growing at different times, researchers typically wind up with a wide range of sizes, from stubby young saplings to towering old growths. That variation can alter the trees' electronic and optical properties. And because the trees are anchored in place, they cannot be moved around. Researchers looking to chop down a few metallic trees and use them to wire up nanodevices are out of luck.

    To grow rods without having them planted on a surface, Alivisatos's team turned to two-part molecules called surfactants. The two portions of surfactants prefer hanging out in chemically different surroundings. Common soaps, for example, are surfactants that contain a “polar” chemical group that favors residing in water and a “nonpolar” group that prefers oils.

    For their crystal-growing experiments, the Berkeley researchers used two surfactants with similar polar-nonpolar pairs. The first chemical, abbreviated TOPO, contains a slightly polar phosphine oxide group connected to a trio of oil-friendly alkyl chains. The second, called HPA for short, has a more polar phosphine acid group connected to a single alkyl chain.

    Earlier this year, Alivisatos's team showed that by varying the amounts of TOPO and HPA and playing with the temperature of the solution and other growth conditions, they could create nanoscale rods from the semiconductor cadmium selenide. Although the exact mechanism is unknown, Alivisatos believes that the rods grow because as CdSe nuclei form in solution, they harbor different crystalline arrangements of atoms on different sides. One face contains alternating cadmium and selenium atoms and is relatively nonpolar, while other faces contain either all selenium or all cadmium atoms, which are more polar. Alivisatos suspects that the TOPO binds to the relatively nonpolar crystalline faces on the sides of a budding nucleus, while the HPA favors either the top or the bottom face, leaving the other face relatively exposed to add new atoms from the solution. As a result, the crystal grows in one direction.

    Now in the 1 December online version of the Journal of the American Chemical Society, Alivisatos's group reports extending their approach considerably. By adding more HPA to the mix, the researchers speed the growth rate of the fast-growing face. The result is an arrowhead shape that forms as new layers of atoms add themselves to the face before the previous layer is filled in. To make CdSe teardrops, the researchers start by growing short rods and then add a bolus of the cadmium and selenium precursor compounds to the mix. The extra inorganics cause the rods to widen at the tops, approaching spheres. Then, as all the inorganics are used up in the reaction, the crystals peter out at the top.

    Perhaps the most intriguing crystals are the nanoscale jacks, which resemble branched organic molecules called dendrimers that are being explored for uses ranging from drug delivery to solar cells. Like standard organic polymers, dendrimers are made up of repeating units called monomers. But unlike normal polymers, in which monomers link together in long, spaghetti-like chains, the monomers in dendrimers each have at least three arms that can link to separate neighbors. As new monomers are added, the branches multiply outward, causing dendrimers to grow into little plastic spheres.

    To date, inorganic chemists haven't had a similar molecular toy to play with. But by starting with low-temperature growth conditions, Alivisatos's team was able to create CdSe crystallites that harbored four separate fast-growth faces. When they then switched to the rapid growth conditions, rods sprang from all four faces. By then dropping the temperature again, the Berkeley team created new branch points at the end of each rod and then grew new rods on them. Alivisatos says that the inorganic versions of dendrimers are not nearly as complex or as large as the organic variety. “But we observed that it is possible to do this,” he says. And it may be possible to add other interesting ingredients as well, growing inorganic dendrimers from several different components.

    Even without all the extra branches, simple four-armed jacks could prove useful. Alivisatos notes that when dropped on surfaces, the jacks always rest on a tripod of three legs, leaving one rod pointing straight up. That could come in handy for wiring up neighboring electrical devices in future nanoelectronics and could help solar cells efficiently channel current from one electrode to another.


    Probing HIV's Elusive Activities Within the Host Cell

    1. Marina Chicurel*
    1. Marina Chicurel is a freelance writer in Santa Cruz, California.

    AIDS researchers are optimistic that preliminary findings on how HIV enters cell nuclei will lead to new therapeutics in the fight against the epidemic

    Like a lock-picking burglar, HIV slips into cells by tinkering with their outer membranes. Once inside, the virus makes its way to the nucleus, where it commandeers the cell's machinery and begins churning out copies of itself. Several promising efforts to outfox the virus with either vaccines or antiviral drugs are focusing on this first interaction—when HIV binds to and fuses with the cell—the idea being to stop the infection before it starts. But a handful of researchers have been eyeing HIV's trek from its entry site to the nucleus of its host cell. Their findings, although still preliminary, promise to reveal new—and badly needed—drug targets to fight the ever-mounting AIDS epidemic.

    It's one of the most exciting frontiers in AIDS research, says Robert Gallo, director of the Institute of Human Virology at the University of Maryland in Baltimore. And like any frontier, this emerging field is a wild place, replete with conflicting views and scientific tussles. Even so, says Gallo, “this research can lead to things we can't even yet anticipate. It'll open more and more ideas on understanding HIV infection, its replication cycle, and its capacity to cause disease. It's obvious that it also will give us new targets for blocking HIV infection, just as the fundamental work on how HIV enters cells gave us new insights.” The work may even provide new tools for gene therapy (see sidebar).

    For all the billions of dollars that have been poured into HIV research over the past 2 decades, this is largely uncharted territory. “All the interactions that the virus does inside the cell with the cellular components are not known in most cases,” says Ari Helenius of the Swiss Institute of Technology in Zürich. As researchers begin to explore this landscape, a common theme is emerging: Soon after viruses enter cells, they enlist components of their hosts to assist them in their journey. Stephen Goff of Columbia University in New York City, for example, has identified cell mutants that block the progression of HIV infection at various stages, suggesting that the virus depends on several intact cellular factors to cause infection. A growing number of studies indicate that this may be the case for almost every stage of HIV's long trek, which includes the production of a DNA copy of its RNA genome, transit through the cytoplasm to the nucleus, transport across the nuclear envelope, integration into a chromosome, and the final challenge, ensuring that its genes are efficiently expressed.


    Recent studies point to the host's scaffold of protein filaments—the cytoskeleton—as one of the key cellular components HIV enlists in its cause. Soon after HIV enters a cell, it forms a reverse transcription complex, a particle rigged to produce a DNA copy of the virus's RNA genome. In studies published in the Journal of Experimental Medicine in 1998, Mario Stevenson and his team at the University of Massachusetts Medical Center in Worcester suggest that the complex interacts with actin filaments of the cytoskeleton. What's more, the complex seems to require an intact filament network to perform its job efficiently. Stevenson thinks the complex might use the filaments to gain access to other factors in the cell that help it make the DNA copy, although other interpretations are possible.

    Beyond making a copy of its genome that is compatible with its host's, HIV must transport that copy to the host cell nucleus, where it directs the production of new viral particles. That is no easy task. HIV's site of entry can be as much as 20 micrometers away from the nucleus—a formidable distance for HIV reverse transcription complexes that are only between 50 and 120 nanometers in diameter. Yet HIV arrives at the nucleus within minutes after infection. “Now that probably doesn't happen by chance,” says Stevenson. “The virus must have a roadmap.”

    Thomas Hope of the University of Illinois at Chicago Medical School thinks he might have found the highway, at least. In unpublished work, Hope labeled HIV particles by fusing a viral protein called Vpr to green fluorescent protein, a marker used to track proteins in living cells. “What we saw was striking. The particles were moving in fairly linear paths within the cell,” he says. Then Hope lit up the cells' microtubules—filaments that, like actin, form part of the cytoskeleton—by injecting microtubule building blocks labeled with a different fluorescent tag. Most HIV particles appeared to be associated with the microtubules; in some cases, Hope could see particles moving in bursts along the lengths of microtubules, like stop-and-go traffic on a busy street.

    “I think the movies are very cool,” says John Coffin, a leading HIV expert at Tufts University School of Medicine in Boston, of the images Hope has captured. “But one has to be very cautious when one's results are cool.” Coffin notes, for example, that in a typical infection, less than 1% of viral particles successfully convert their host cells into HIV factories. That means that most of the particles that Hope labels are dysfunctional. But Hope has managed to partially solve this problem by using fluorescent deoxynucleotides to tag only those particles capable of synthesizing DNA; he is also timing his experiments to avoid monitoring nonfunctional particles that are quickly destroyed by the host cell.

    Breaking and entering

    After traversing the cytoplasm, HIV encounters its next formidable task: getting through the nuclear envelope. In nondividing cells, the only way in is through the nuclear pores, which normally allow entry to cargo no more than 28 nanometers wide. Although estimates vary, the HIV particles, called preintegration complexes, that arrive at the nuclear envelope may be twice this size.

    The long trek.

    Before it can wreak its havoc, HIV must enter the cell and make a DNA copy of its RNA genome. This copy is transported through the cytoplasm and moves across the nuclear envelope as part of a particle called a preintegration complex. Within the nucleus, the virus integrates into the host chromosome and begins producing more viruses.


    Most retroviruses solve the problem by bypassing the pores and infecting only dividing cells, taking advantage of the stage when the nuclear envelope breaks down during cell division. But not HIV. The virus's targets include both dividing cells, such as activated T cells, and nondividing cells, such as macrophages. “It's clear that the virus has gone to great lengths to do this,” Stevenson says. “So macrophages must be important for some aspect of the biology of these viruses.”

    Outside a host cell, HIV is exposed and vulnerable. So Stevenson thinks that by infecting macrophages, one of the first cells it encounters, HIV protects itself. In addition, Stevenson reported last year in Nature Medicine that HIV-infected macrophages secrete signals that attract T cells, HIV's preferred host, and prime them for infection. “Our suspicion is that macrophages actually disseminate the virus on to other cells,” he says.

    So how do HIV particles get through the nuclear envelope of macrophages and other nondividing cells? It's the $64,000 question, says Goff. Four viral components have emerged as candidates for helping move the particles into the nucleus, but their relative contributions remain elusive and controversial. “It's a ridiculously intense controversy,” says Didier Trono of the University of Geneva in Switzerland. “I say ‘ridiculously’ because I think we are much too arrogant in our assertions and we underestimate the complexity of biological phenomena.”

    One of the first hints of HIV's capacity to penetrate nuclei surfaced in 1991 when Michael Malim, now at the University of Pennsylvania in Philadelphia, reported in the Journal of Experimental Medicine that HIV was capable of integrating its DNA into the chromosomes of nondividing cells. A year later, Stevenson helped explain this finding: He reported in the Proceedings of the National Academy of Sciences (PNAS) that preintegration complexes could traverse their host's nuclear envelope. Stevenson's team then set out to search for ways in which the virus might be tapping into the cell's transport system. Like packages moving through a postal system, cellular proteins often bear address labels that help the cell route them to their proper destinations. So the researchers hypothesized that HIV could be sneaking into the nucleus by sporting a nuclear address label, a short sequence called a nuclear localization signal that is recognized by the cellular proteins that ferry cargo into the nucleus. Indeed, in 1993, the researchers found that a viral protein called the matrix protein contains such a sequence.

    A flurry of studies followed this initial observation, some supporting the matrix protein's role in transport, others offering evidence against it. Today the protein's role remains uncertain, but most agree it cannot be the sole mediator of HIV nuclear import. A study published in the EMBO Journal by Heinrich Gottlinger's lab at the Dana-Farber Cancer Institute and Harvard Medical School in Boston in 1998, for example, claimed that mutant viruses lacking most of the matrix protein are still capable of infecting nondividing cells.

    Buds and blebs

    A year after the matrix protein first emerged as a candidate for transporting HIV particles into the nucleus, another viral protein, Vpr, stepped into the spotlight. A team led by Stevenson and Michael Emerman of the Fred Hutchinson Cancer Research Center in Seattle, Washington, reported in PNAS that mutations in the matrix protein's nuclear localization signal didn't interfere much with the transport of HIV DNA into the nuclei of nondividing cells unless Vpr was mutated as well. Subsequent studies showed that Vpr sports sequences that may enable it to bind directly to the nuclear pores or to associate with cellular proteins that dock nuclear-bound cargo at the pores.

    Recent studies suggest that Vpr might also help disrupt the nuclear membrane itself. Working in Warner Greene's lab at the Gladstone Institute of Virology and Immunology of the University of California, San Francisco, Carlos de Noronha genetically engineered cells to produce Vpr and monitored their behavior using labeled proteins. In one set of experiments, described at the 2000 International Meeting of the Institute of Virology in Baltimore last September, they fluorescently labeled nuclear lamin C, a protein of the nuclear envelope that forms an array of filaments that enclose the nucleus like a burlap bag. “It was unbelievably striking,” says Greene. “When Vpr is present, the nuclear membrane begins to bleb, forming transient protrusions from the nucleus that look like solar flares. And then, intermittently, these blebs rupture, and soluble components of the nucleus flood out into the cytoplasm, and cytoplasmic soluble components flood into the nucleus.” Greene speculates that, among other things, this disruption may allow HIV preintegration complexes to sneak into the nucleus.

    In 1997, Trono's group came up with yet another candidate for carrying these HIV particles across the nuclear envelope: integrase, the enzyme that inserts HIV DNA into its host's chromosomes. As Trono described in PNAS, integrase also sports a nuclear localization signal, which when fused to marker proteins can transport them into the nucleus. Unpublished work from Malim's lab suggests that integrase is indeed essential for transporting HIV preintegration complexes into its host's nucleus. But Fabrizio Mammano at INSERM in Paris and his colleagues claim their work, published in August in the Journal of Virology, indicates otherwise. Mammano thinks that integrase's nuclear localization sequences are required for hauling integrase itself, as a single protein, across the nuclear envelope but are dispensable for importing bona fide preintegration complexes, which contain several components in addition to integrase, including DNA, Vpr, and the matrix protein.

    Adding yet another twist to the story, Pierre Charneau of the Pasteur Institute in Paris reported in the April issue of Cell that a short segment of viral DNA plays an important role in the import of HIV particles. Although DNA usually exists as a double-stranded molecule, a small region of HIV DNA called the “central DNA flap” is triple stranded. Charneau found that mutant viruses lacking this unusual structure pile up at the outer edge of the nucleus, seemingly unable to get in. One possibility is that the flap helps shape the bulky particles so they can crawl through the pores.

    Several researchers have developed working hypotheses to explain the complexity. “No matter what cell the virus gets into, it will face the need for traveling from the plasma membrane to the nucleus and getting through the nuclear envelope,” says Trono. “And those circumstances will most likely vary from one cell to another. I think this may well explain the diversity of mediators of nuclear import.”

    Others note that the problem of hauling a large particle through a small and selective pore is so onerous that HIV may need multiple mechanisms to do it efficiently. In support of this idea, cell biology studies have shown that the more nuclear localization signals a particle carries, the faster and more likely it is to be transferred across the nucleus. In addition, different signals may cooperate with each other or act sequentially. And because getting into the nucleus is of such vital importance to HIV, the multiple transport mechanisms may serve as backups for each other. “It's not uncommon for HIV to incorporate redundancy—to have multiple ways of getting a job done,” says Greene. Still others, like Emerman, have already placed their bets on a single candidate—in his case, on integrase.

    “I don't think everybody is right here, but I don't yet have a betting line on who might be and who's not,” says Coffin. “I'm standing aside waiting for the dust to clear.”

    Inside the black box

    While controversy reigns over nuclear import, mystery shrouds HIV's movements within the nucleus itself. “If the trip from the plasma membrane to the nuclear envelope is a gray box, the rest is a black box,” says Trono. “Biologists don't know much about intranuclear trafficking, in general.”

    Despite this ignorance, researchers are beginning to realize that, just as HIV co-opts cytoplasmic factors, it also takes control of nuclear factors. Trono's unpublished observations, for example, suggest that shortly after HIV enters a cell, it triggers factors involved in regulating gene expression to rush out of the nucleus. Trono stresses that the meaning of the observation remains unclear. But he hypothesizes that HIV might be recruiting these factors to ensure that its genes are expressed efficiently once they are integrated in the host's chromosomes. Most locations on a chromosome are not suitable for gene expression. Yet HIV's genes seem to be expressed very effectively, almost regardless of where they integrate. “That is quite a success that you need to explain,” says Trono. “These factors might help to create a sort of transcriptional honeymoon for the provirus right after it integrates.”

    Turning these emerging insights into therapeutics will clearly take some time. A few researchers are already trying to block viral nuclear localization signals, for instance. But because these sequences are similar to those of many cellular proteins that reside in the nucleus, some fear this strategy could lead to serious side effects. Also, the complexity of the challenge causes many drug companies to shy away from developing drugs that block multisite interactions between large proteins, says Coffin, such as those that occur as HIV travels through the cell.

    But Gallo is optimistic, comparing the study of HIV's trek through the cell to the early days of research on viral entry. “Look at what happened in a short time in that area,” he says, alluding to the new drug candidates spawned by viral entry research. “Who knows what can happen here. You can't predict what'll come, but that something will come is very likely.”


    Capitalizing on HIV's Talents for Gene Therapy

    1. Marina Chicurel*
    1. Marina Chicurel is a freelance writer in Santa Cruz, California.

    The emerging understanding of HIV's journey from the cell membrane to the nucleus may help researchers design better vectors for gene therapy for a variety of diseases. In the early 1990s, genetic engineers were using mouse retroviruses, which are incapable of infecting nondividing cells, as gene-delivery vectors. Despite their efforts to shape the retroviruses' genomes and endow them with the ability to infect nondividing cells, prime targets for gene therapy, the vectors failed repeatedly.

    Didier Trono of the University of Geneva, who was working on HIV and was also interested in developing vectors for gene therapy, reasoned that if he could understand how HIV managed to infect nondividing cells, he might be able to engineer the mouse retroviruses to do likewise. But as the complexity of HIV's nuclear transport emerged, Trono began to doubt that strategy. “At that time we were working on nuclear import and were saying, ‘Oh, gosh, there's matrix, there's Vpr, there seems to be another guy. … This complex may be very sophisticated; how can we think of just putting a nuclear localization signal somewhere and expect it to work?’” he recalls. So instead, working with Luigi Naldini and Inder Verma of the Salk Institute for Biological Studies in La Jolla, California, Trono decided to use HIV itself as a vector. They proposed modifying HIV to eliminate its ability to cause disease, while capitalizing on its talent for delivering genes into the nucleus.

    The radical proposal worked, leading to the development of a new class of vectors for gene therapy of nondividing cells (Science, 12 April 1996, p. 263). Further modifications to this original vector have now led to safer and more effective varieties that are beginning to prove their clinical potential. Last month, for example, a vector developed in Trono's lab, known as a lentiviral vector, was used to successfully treat monkeys suffering from a condition similar to Parkinson's disease (Science, 27 October, p. 767). And even better vectors may soon become available as researchers understand the mechanisms behind HIV's talent for delivering genes to the nucleus. Pierre Charneau of the Pasteur Institute in Paris describes in an April Cell article, for example, that adding a DNA flap to Trono's original HIV vector dramatically improved its efficiency. As Ari Helenius of the Swiss Institute of Technology in Zürich explains, “Viruses have had millions of years to evolve gene therapy.”


    A Dripping Wet Early Mars Emerging From New Pictures

    1. Richard A. Kerr

    The latest images from the Red Planet are suggesting that water ponded across its equatorial region eons ago, just when life might have been emerging

    Mars, water, life. That explosive mixture comes together again on page 1927 of this issue, where a sampling of hundreds of spacecraft images shows crisply detailed sediment layers on Mars. Although the authors offer more than one interpretation, the one they prefer has the sediments laid down beneath broad lakes and shallow seas at a relatively clement time in the planet's history. The images don't have the visceral impact of the springlike seeps reported earlier this year, but the geologic implications of the pictures plus supportive signs from earlier missions mean that these possible lake sediments will be prime candidates for NASA missions seeking signs of past life on Mars.

    An edgy face of Mars.

    Some erosive force, perhaps the wind, has sculpted layered sediments into stair-stepped hills inside an impact crater (left) and a chasm. Images are about 1.2 kilometers square.


    The new hints of extensive standing water on Mars as much as 4 billion years ago—about when life got started on Earth—comes after 3 decades of studies of layered martian terrains. “A lot of the pieces are not new” in the Science paper, says Mars geologist Michael Carr of the U.S. Geological Survey, but the new study “brings all these bits and pieces together with much better support for layered deposits and makes a good story out of it”—not that the story will lack controversy. “They make a good case that [the terrain] is layered,” says planetary geologist James Head of Brown University, “and the most likely interpretation is they are sediments. That's pretty impressive. The part that will be debated will be the origin of the sediments. Sediments don't necessarily mean water.”

    The controversy over how much water Mars ever had and when, if ever, liquid water flowed on the surface began with the first successful Mars orbiter, the U.S. Mariner 9 spacecraft. Arriving in 1971, it confirmed the dry, desolate, moonlike appearance of large areas of Mars imaged by earlier flyby spacecraft. But Mariner 9 also discovered so-called valley networks, reminiscent of river-carved valleys on Earth. Many researchers argued that groundwater oozing from the headwalls of valleys, rather than rain and running rivers, could sculpt such features, obviating the need for a “warm and wet” early Mars. But more signs of water—water that once stood in pools, lakes, and even oceans—appeared in many of the 50,000 images returned by the pair of Viking orbiters that arrived in 1976. Hints of a shoreline around the northern lowlands suggested to Mars geologist Timothy Parker of the Jet Propulsion Laboratory in Pasadena, California, that there had been an early ocean (Science, 4 December 1998, p. 1807). And reports of craters containing lakelike sediments, some of them layered, suggested that some impact craters had held water at some point in Mars's history. But the reality, extent, and age of oceans and lakes remained controversial.

    Then Mars Global Surveyor (MGS) arrived. Carrying the Mars Orbiter Camera (MOC), it began returning a torrent of images in 1999 so sharp that features as small as a few meters could be discerned, compared with the 25- to 50-meter resolution in Viking images. At Malin Space Science Systems Inc. in San Diego, Michael Malin, MOC principal investigator, and Kenneth Edgett scan the incoming images daily for features of geologic interest. Most dramatic were the springlike seeps that Malin and Edgett reported in Science last summer (Science, 30 June, p. 2330). The rivulets of sediment on impact crater walls showed that some fluid, presumably involving water as a brine or melted ice, briefly flowed from the ground within the past few million years. But Malin and Edgett were also seeing a steadily growing number of images, now in the hundreds, of layered terrains.

    Malin and Edgett compiled a survey of layered terrains that turned up a more consistent picture than seen before. As they report in their Science paper, layered rock is exposed at most longitudes in the martian equatorial region but is concentrated in five low-lying areas there, including the canyons of Valles Marineris and northern Hellas basin, a 2200-kilometer impact crater. Layers most commonly fill impact craters. In places, craters seem to have been completely covered by layered material—possibly akin to the easily eroded chalk Cliffs of Dover—that has since been partly eroded away, perhaps by the wind. The layered rock tends to come in three sorts—many thin, stacked layers each a few meters to tens of meters thick; one massive layer hundreds of meters to a few kilometers thick; or a thin, dark layer that often caps mesas where much erosion has occurred. Where they occur together, mesa-forming layers are on top, massive layers below that, and thin layers on the bottom. Neither the sources of all this sediment nor its means of transport are evident in the images. However it got there, it seems to have arrived early in Mars's history, according to Malin and Edgett, a time called the Noachian more than about 3.5 billion years ago.

    Malin and Edgett favor water as the motive force behind the layered sediments. “On Earth, water is most effective at producing” strong layering, says Malin. Add in the fact that the images show layered sediments have piled in, and even filled, some craters—where wind-blown dust would likely blow in and out again rather than accumulate to great depths—and the MOC images “strongly hint water was involved,” says Malin.

    In their preferred scenario, a thicker, warmer atmosphere than today's would have dumped rain on the highlands, and flowing water would have carried clay, silt, and sand into crater lakes and perhaps some shallow seas between craters. They place this scenario during the Noachian, the time of heavy, crater-forming bombardment, when the terrestrial planets swept up the last of the debris lingering from their formation (Science, 1 December, p. 1677). The heat of large impacts may have sterilized the surface as late as 3.9 billion years ago, but impacting objects also could have brought in life-giving water and left abundant thermal springs where life may have gotten started. Once the water was gone—into the subsurface or outer space—erosion over the eons, perhaps by the wind, could have begun to expose the layers.

    The notion of pervasive standing water on Mars is sure to get a mixed reception. “They're terrific images,” says planetary geologist Alfred McEwen of the University of Arizona in Tucson, a MOC team member, but “the interpretations are going to be controversial. They do look sedimentary, and [in MOC images] they look similar from place to place. It's a little dangerous, but it's reasonable to conclude they formed by a similar process and from similar materials. But there are lots of processes that produce sedimentary deposits,” from water and wind erosion to erupting volcanoes and cratering.

    The debate over the origin (or origins) of martian layered sediments starts with the source of the sediments. Malin and Edgett are inclined toward erosion of the highlands as the predominant source, while Head favors a role for volcanic ash, and planetary geologist Nathalie Cabrol of NASA's Ames Research Center at Moffett Field, California, points to impact ejecta as another likely player. Carr takes a more ecumenical approach. Seeing no obvious path taken by the sediments, such as a river channel, he concludes that “water pools in the craters and acts as a trap for material blowing around.” (Malin and Edgett also offer a “substantially more exotic,” but in their view less likely, alternative: a martian atmosphere whose mass—and thus the ability of its winds to pick up and carry volcanic or impact dust—waxes and wanes so as to produce the layering.)

    The central question underlying all these scenarios is how much surface water the planet ever enjoyed. In Head's latest analysis of MGS altimetry data, a shorelinelike profile appears around the northern lowlands, but no one, including Head, sees the case for an ocean as closed yet. As to lakes, this year Cabrol published her latest analyses of Viking images showing what she believes to be craters filled with typical lake sediments, including deltas where water had flowed in. And in a number of meeting abstracts, Parker has documented a flow path from sediment-filled Argyre crater along a trough connecting a series of one-time crater lakes that runs downhill 4000 kilometers to the northern lowlands.

    Even if they accept that standing water trapped at least some of the sediments, scientists disagree about just when the layers formed. Malin and Edgett believe the layers are likely Noachian by analogy with the layered sediments in Valles Marineris. Rather than relatively young lake deposits laid down long after the chasms of Valles Marineris split open the ancient rock there, they believe these layers are ancient rock protruding from the walls of Valles Marineris. “That's unconvincing,” says McEwen. He sees the Valles Marineris deposits as later, younger additions. “All we can say is they're younger than 3 [billion] to 4 billion years.”

    Understanding the nature of sediment layering on Mars is going to take time. “Doing fieldwork would be terrific,” notes McEwen. Barring that, “we need some way of getting age dating of the materials and compositional information [from orbit] at high resolution.” Orbital missions in this decade would return the compositional analyses through high-resolution spectroscopy. Dating would require rovers with instruments that have not yet been designed. The story of Mars, water, and life will be a while in the telling.


    Has Leishmaniasis Become Endemic in the U.S.?

    1. Martin Enserink

    Believed to be all but absent from the U.S., the Leishmania parasite has infected more than 1000 hunting dogs

    Since its surprise attack in New York City last year, the West Nile virus has become the most infamous new pathogen in the United States. But a handful of researchers and public health authorities are now worrying about another new bug on the block. Since last January, they have discovered that hunting dogs in 21 U.S. states and the Canadian province of Ontario are infected with the parasite that causes visceral leishmaniasis—a protozoan thought to be all but absent from the United States. Visceral leishmaniasis is an infection of the liver, spleen, and bone marrow; it can affect humans as well as dogs and can be lethal when untreated.

    Nobody is claiming that the disease is about to run riot among the U.S. population. But the widespread outbreak in dogs has experts wondering whether visceral leishmaniasis—which sickens over half a million people yearly in South America, Africa, the Mediterranean, and India—has become an endemic disease in North America. If so, scattered human cases are possible, says Peter Schantz, a researcher at the Centers for Disease Control and Prevention (CDC) in Atlanta: “It's certainly something to be concerned about.” To gauge that risk, Schantz and his colleagues are trying to find out how the parasite—identified as Leishmania infantum by the Institute of Public Health in Rome, Italy—spreads in North America. And they're still flummoxed by how the parasite could have become so widespread in the first place.

    Dogs are Leishmania's main reservoir, although most of them can be infected for years without severe symptoms. The parasite is spread to humans and other mammals by sandflies of the Phlebotomidae family, blood-feeding insects about a fourth the size of a mosquito with a nasty bite. Travelers occasionally return from the tropics with an infection. And sporadic cases of Leishmania-infected dogs have occurred in the United States, says Schantz; most were pets brought here by military families returning from a base in the Mediterranean. In the 1980s and early 1990s, a few foxhounds that had never been in an endemic area were also diagnosed with the disease in the southern United States. Nobody could determine how they became infected, but few bothered to look very thoroughly.

    That changed in late 1999, when foxhounds at an exclusive hunt club in Millbrook, 130 kilometers north of New York City, came down with symptoms such as bleeding, wasting, seizures, hair loss, and kidney failure. Almost two dozen died, and club members spared no expense to find out what was killing their companions. Because the area is infested with ticks, the club's veterinarian called in Edward Breitschwerdt of North Carolina State University (NCSU) in Raleigh, an expert in tick-borne infections. But Breitschwerdt and his colleagues couldn't find anything in the dogs' blood. When the club finally drove two ailing dogs down to Raleigh for a thorough checkup, however, the researchers found Leishmania parasites in the animals' joint fluid.

    Since then, researchers from NCSU, CDC, and the Walter Reed Army Institute of Research in Silver Spring, Maryland, with the aid of the Masters of Foxhounds Association of America, have tested almost all of the 11,000 foxhounds in U.S. hunt packs. About 12% of them, from the eastern United States, have antibodies to Leishmania—even though the vast majority are asymptomatic. Researchers have also tested almost 50 stray dogs from around Millbrook; none had the disease, says Schantz. In addition, CDC researchers have examined old blood samples from 450 dogs from several states; they too were negative.

    Those findings leave a big riddle: How is the infection transmitted? At least four species of sandflies inhabit the United States, says Edgar Rowton, a medical entomologist at Walter Reed. But they have never been found farther north than New Jersey, hundreds of kilometers south of Millbrook. So how could the disease have ventured as far north as Michigan and Canada? And why didn't other dogs, or even humans, become infected?

    Schantz posits an unusual mode of transmission. Foxhounds live close together in hunt club kennels, he points out, and they may infect each other without the help of sandflies—for instance, through sexual contact, or through wounds sustained when they tear through the forest.

    Other experts disagree. Although direct transmission between dogs has been shown to occur rarely, concedes Robert Killick-Kendrick, a leishmaniasis researcher at Imperial College in London, he doesn't believe that could explain the high prevalence among U.S. foxhounds. He thinks sandflies are the main vector, just as they are everywhere else. Foxhounds, he points out, often travel for joint hunts, training, or dog shows with other clubs. On one of these visits south, the northern dogs may have been bitten by sandflies, he says. (The New York pack indeed were frequent travelers.) A closer look may reveal that other dogs have been bitten and infected in the south as well, he says.

    Another possibility is that there are sandflies in the north that nobody has ever seen, says Rowton, because nobody has looked carefully. Rowton, for one, has studied sandflies from endemic areas extensively but is only now beginning to look at North American varieties. Yet another possibility is that other insect vectors exist.

    The debate is not just academic. If sandflies are spreading the disease, there's a chance they may also infect humans; in fact, that may already have happened, says NCSU's Breitschwerdt. Because a few foxhound cases occurred in the 1980s, “we probably uncovered a smoldering epidemic that has been in this country for 20 years or more,” he says. Doctors could have easily missed a few scattered human cases of visceral leishmaniasis, whose symptoms include fever, malaise, and weight loss. But there is no reason to panic, says Killick-Kendrick. The Leishmania infantum strain isolated from the Millbrook dogs is the same one found throughout the Mediterranean; there, about 20% of all dogs are infected, yet human cases are rare, and an effective treatment exists. Most at risk are people with a weakened immune system. “From what I know from about the Mediterranean, the risk is very low indeed,” says Killick-Kendrick. “I would be very upset if a scare erupted about this.”

    The route of transmission will also determine whether the disease can ever be wiped off the continent. If Leishmania is transmitted directly among foxhounds, it might be eradicated by simply culling all infected dogs. In fact, some hunt clubs have already started putting down their sick animals. But if sandflies are involved, they have likely already infected other dogs, foxes, or coyotes, making it impossible to stamp out Leishmania. “That's the fear,” says Breitschwerdt. “If the disease has become endemic in the United States, then we can't eliminate it, and we'll have to live with it.”


    Taking the Measure of the Wildest Dance on Earth

    1. Dana Mackenzie*
    1. Dana Mackenzie is a writer in Santa Cruz, California.

    By exploiting the symmetry of randomness, three mathematicians have revealed the geometric underpinnings of Brownian motion

    If you could watch an individual air molecule, you would see a dance that puts the wildest mosh pit to shame. Slamming into its neighbors, rebounding, ricocheting without letup, each humble particle traces out a path so jittery that nothing can tame it. The slowest slow-motion camera, the most powerful zoom lens, would only bring quicker and smaller lurches into view.

    Now, a trio of American and French mathematicians has proved that the frenetic random dance called Brownian motion has geometric properties that can be calculated as exactly as the circumference of a circle. The methods they used to prove that counterintuitive notion seem likely to apply to other random processes, some as familiar as the flow of water through a filter. The proof, presented at the recent Current Developments in Mathematics 2000 conference* sponsored by Harvard University and the Massachusetts Institute of Technology, is drawing rave reviews. Says Yuval Peres, a mathematician at the University of California, Berkeley, “I feel their work is one of the finest achievements in probability theory in the last 20 years.”

    The proof by Gregory Lawler of Duke University, Oded Schramm of Microsoft Research, and Wendelin Werner of the Université de Paris-Sud describes the probability that two or more neighboring air molecules, trapped in a plane, will escape to a large distance apart without crossing one another's tracks. In theory, the molecules could travel in straight lines, avoiding collisions with other particles; in practice, however, it is infinitely more likely that they will get jostled into tangled fractal paths.

    Whether those paths cross has little physical significance: “Particles in the real world aren't worrying about where they've been,” Lawler notes, and they usually are not confined to a plane. But the numerical parameters that describe the likelihood of crossing, called intersection exponents, interest physicists intensely, as they model a variety of systems near a phase transition. In the study of magnetic materials, for example, similar “critical exponents” describe how short-range correlations between electrical spins produce long-range order.

    Faced with the infinite complexity of fractal Brownian motion, mathematicians and physicists usually prefer to simplify it by restricting particles to a grid that lets them move in only two directions—up and down or side to side—like the stylus in an Etch A Sketch toy drawing screen. They also require the particles to move only in discrete steps. The finer the grid is, the more closely the Etch A Sketch squiggle resembles true Brownian motion.

    Unfortunately, such simplified “finite-lattice models” have not led to a rigorous derivation of the long-sought intersection exponents. “[Brownian motion] problems have been studied to death on the lattice using combinatorial methods, and no exact solution is in sight,” says John Cardy, a theoretical physicist at Oxford University. Lattice models also lack some crucial characteristics of true Brownian motion. For example, in the lattice version, a strong enough magnifying glass would reveal the underlying graininess of the motion. Real Brownian motion when magnified still looks like Brownian motion—even if the magnification varies from point to point, as in a funhouse mirror.

    That extremely strong symmetry property, called “conformal invariance,” may actually make the fractal Brownian paths easier to work with than their lattice imitations. In 1999, Lawler and Werner showed that the intersection exponents for Brownian motion are determined by its symmetry properties alone, regardless of what physical process produces the motion. Any other random, conformally invariant process that doesn't get distorted by edge effects (a condition called “locality”) would have the same intersection exponents. Such a non-Brownian random process might prove a mathematical godsend to stymied researchers. But did it even exist? Lawler and Werner had no idea.

    Then, independently, Schramm found it. Using an ingenious combination of 20th-century probability theory and 19th-century conformal mapping theory, he discovered a wholly new process, which he called stochastic Loewner evolution (SLE). Although SLE looks two-dimensional, Schramm discovered a mathematical trick for reducing it to one dimension—as if the two knobs of an Etch A Sketch toy were secretly controlled by a single master dial. That made the intersection exponents for SLE much simpler to compute. Werner, Lawler, and Schramm then showed that SLE was also conformally invariant and local, thus confirming that its exponents were the same as the exponents for two-dimensional Brownian motion. Their proof is now available on the Web ( as a series of preprints totaling over 100 pages, the first of which has been accepted by the journal Acta Mathematica.

    The exponents settle a variety of related problems about Brownian motion. They show, for example, that the outer edge or “frontier” of a Brownian motion is a fractal with dimension 4/3. In other words, just as the circumference of a circle is proportional to its diameter, the size of a Brownian path's frontier is proportional to the 4/3 power of its diameter (the longest distance across the frontier). When Benoit Mandelbrot proposed that neat relationship in his 1982 book, The Fractal Geometry of Nature, mathematical colleagues shrugged it off as speculation, Lawler recalls. But 18 years later, Mandelbrot has been vindicated.

    Most tantalizingly for physicists, the SLE process may describe a number of other random phenomena. The best candidate appears to be “critical percolation,” a way of describing how water and other liquids flow through a porous barrier. To model it in two dimensions, physicists start with a blank filter ruled like a honeycomb with hexagonal cells, then randomly assign each cell to be either permeable or impermeable. By flowing through clusters of permeable cells, water can percolate across the honeycomb. If the cells of the honeycomb are made vanishingly small, Schramm believes, the boundaries of those clusters become random curves identical with the ones the SLE process produces.

    “It's fantastic that the process that is conjectured to be important for percolation is rigorously proved to be connected to Brownian motion,” Peres says. As Rick Durrett, a probability theorist at Cornell University, explains, “Physicists like to think various models are in the same universality class. This may be one of the first examples where you can prove one model is equivalent to a second.”

    • *17-18 November, Cambridge, Massachusetts.


    X-rays Hit the Spot for Astrophysicists

    1. Robert Irion

    HONOLULU—About 500 astronomers flocked to Waikiki Beach from 6 to 10 November for a meeting of the American Astronomical Society's High-Energy Astrophysics Division. Looking splendid in their complimentary aloha shirts, speakers told tales of intense radiation from deep space, including x-rays from quasars and baby stars.

    Hot Times for Baby Stars

    A dark nest of dust seems like a cool place for a baby star to fledge. However, astronomers have learned that long before some new stars ignite their nuclear furnaces, they unleash powerful flares of x-rays that reach temperatures of 100 million degrees Celsius. Now, two reports at the meeting indicate that that surprising process is common in the infancies of all types of stars.

    Stars hatch within dense clouds of gas and dust, some of which forms whirling protoplanetary disks, or “proplyds.” The Orion Nebula, the closest major stellar nursery, contains many such knots. Thick dust prevents optical light from escaping, but astronomers can use x-rays to detect the protostars within. Previous low-resolution studies of emerging stars with x-ray observatories, including the German ROSAT satellite and the Japanese ASCA, suggested that a few nearby stars could emit hot x-rays at a tender age. However, resolving x-rays from scores of stars in Orion became possible only last year, when the Chandra X-ray Observatory was launched.

    Two new Chandra studies show that cracklingly hot stellar childhoods are common. First, astronomer Norbert Schulz of the Massachusetts Institute of Technology and his colleagues probed Orion's heart, a close-packed cluster of stars and protostars called the Trapezium. Energetic x-rays streamed from all of the Trapezium's stars, regardless of their masses. High-mass stars raged at up to 80 million degrees, three times hotter than previously measured. That rules out a scenario that some researchers have used to explain x-rays from infant stars: strong stellar winds that plow into the surrounding gas, creating fierce shocks. “We cannot explain the highest temperatures we see with shocks,” Schulz says.

    Blazing babies.

    Newborn stars in the Orion Nebula emit surprisingly hot bursts of x-rays (blue, with white contours).


    Even low-mass proplyds got into the act, emitting x-rays that pointed to steady temperatures of 60 million to 80 million degrees —far beyond the occasional hot bursts seen in earlier observations. “These are stars like our sun that are only about 300,000 years old, so they haven't even started burning yet,” Schulz notes.

    Deepening the mystery, a Japanese team of astronomers led by Katsuji Koyama of Kyoto University and Yohko Tsuboi of Pennsylvania State University examined low-mass objects in a different part of Orion and in another dense cloud, called ρ Ophiuchi. Those protostars were even younger, merely 10,000 to 100,000 years old. Still, Chandra perceived torrents of x-rays from flares that sometimes approached 100 million degrees. Koyama contrasts this with gas and dust temperatures of a few tens of degrees at the cores of the clouds. “No one expected that stars could produce such x-ray activity at such an early stage,” Koyama says.

    The two teams can think of just one explanation: intensely twisted coils of magnetic field that lace through the rapidly rotating protostars and their disks. Such tangled fields should short-circuit and snap, says Penn State astronomer Eric Feigelson, flash-heating the surrounding gas to tens of millions of degrees. The resulting flares might unleash hundreds of thousands of times more energy than flares on our sun today.

    Tsuboi's detection of x-rays from the youngest protostars ever observed is especially noteworthy, Feigelson believes. “The greatest implication will be to help us learn how early the x-rays turn on and whether they affect the star-formation process,” he says. “It's possible, but controversial, that x-rays will sufficiently ionize the gas in the proplyd and surrounding environment to lock it to the magnetic field. If that were true, we might have to alter our entire picture of early star and planet formation.” Current theories focus on the self-gravitation of the collapsing cloud and the hydrodynamics of neutral molecules, he notes, rather than magnetic interactions among charged molecules.

    The genesis of strong magnetic fields within proplyds is “very puzzling,” says astronomer Andrea Dupree of the Harvard-Smithsonian Center for Astrophysics. “There are a lot of options on the table,” including fields generated at the roiling interfaces between the growing stars and the disks of gas around them. “The geometry of the field is not known,” Feigelson agrees. “It's an open question.”

    Quixotic Quasar May Yield Cosmic Yardstick

    Flickering x-rays from a mirage in the depths of space may refine estimates of how quickly the universe is expanding. The x-rays stream from a quasar, whose light splinters along several paths as it passes another galaxy along the line of sight to Earth. Slight variations in the time it takes for x-rays to navigate each path promise to reveal the distance to the intervening galaxy, according to research presented at the meeting. That, in turn, could give astronomers their longest yardstick yet for gauging the growth rate of space itself.

    The quasar, a beacon called RXJ 0911.4+0551, shines near the fringes of the observable universe. It probably draws its power from matter swirling into a huge black hole at the core of a young galaxy. Gravity from a galaxy between Earth and the quasar shears the quasar's light into four closely spaced beams. This optical fracturing makes the quasar appear in a telescope as four spots, an illusion called a gravitational lens.

    Cosmologists admire more than the eerie beauty of such lenses. For example, lenses probe the curvature of the universe, thanks to precise predictions of the light paths from Einstein's general theory of relativity. In 1991, astrophysicist Ramesh Narayan of the Harvard-Smithsonian Center for Astrophysics in Cambridge, Massachusetts, also showed that one could derive an exact distance to a lensing galaxy by observing the difference in light's travel times along each path, a quantity called “time delay.” The delay is evident when one of the quasar's images flares up, followed some time later by similar flares in the other spots.

    A firm distance, combined with the known speed at which the lensing galaxy is receding from Earth, leads directly to a measure of the Hubble constant—the rate at which the universe is growing. “The beauty of this technique is that it's a one-shot direct measurement of distance,” Narayan says. “We don't have anything else like it.” Other methods of gauging distance require astronomers to build a “distance ladder” from nearby objects out to remote ones, he notes—a process rife with uncertainty.

    Astronomers have seen time delays in lensed quasars by monitoring peaks and valleys in their radio or optical emissions. However, it takes months or years for quasars to vary at those wavelengths, and the changes are subtle. Now, astronomers George Chartas of Pennsylvania State University, University Park, and Marshall Bautz of the Massachusetts Institute of Technology have overcome both of those hurdles by finding x-ray emissions from RXJ 0911.4+0551 that fluctuate dramatically in a few hours.

    The team used the Chandra X-ray Observatory on 2 November 1999 to document a sharp, 40-minute flare in one of the quasar's four spots. Calculations suggest that if Chandra had focused on the quasar continuously for a half-day before that, the same flare would have brightened at least one of the other three spots. Chartas believes that longer observations of RXJ 0911.4+0551 and a dozen other tightly spaced quasar lenses will nail down their time delays to an accuracy of 1%. The team has won further time on Chandra and the European XMM-Newton x-ray satellite to pursue their flickering quarries.

    Even if they can nail down the time delays to that accuracy, calculating the distance to the lensing galaxy would still involve a lot of uncertainty. Einstein's equations require a model for how mass is distributed in the intervening galaxy. That's a tough chore, says astronomer Wendy Freedman of the Carnegie Observatories in Pasadena, California. “Astrophysical lenses are messy,” she observes, as clusters of galaxies and hidden clumps of dark matter can alter the light paths.

    Indeed, Narayan notes that the extreme closeness of the lensed images in RXJ 0911.4+0551 suggests that some distortion in the lensing galaxy—or even a small satellite galaxy—induces the mirage. That departure from a smooth blob of mass “would seriously compromise the measurements, and you wouldn't even realize it,” he says.

    Still, both Narayan and Freedman welcome the technique as a check on estimates of the Hubble constant. Freedman leads a team that will soon publish a slightly revised value of 72 kilometers per second per megaparsec (which roughly translates to an age of 13 billion years for the universe). That figure, based on a distance-ladder approach with the Hubble Space Telescope, has a 10% margin of error, Freedman says. Chartas and his colleagues, she believes, will have to scrutinize x-ray flares from many quasar lenses to reach the same plateau.