# News this Week

Science  14 May 1999:
Vol. 284, Issue 5417, pp. 1094
1. ASTRONOMY

# China Doubles Budget to Create Astronomy Megainstitute

1. Hui Li*
1. Writes for China Features in Beijing.

BEIJING—China has embarked on a fundamental overhaul of the way it runs its astronomy programs. It will double spending on astronomy research, consolidate the management of the country's telescopes under one roof, and slash payrolls by more than half as part of a broader effort to strengthen the country's capacity in basic research.

The new initiative will be led by the National Astronomical Observatory Center (NAOC), one of a dozen megainstitutes being created by the Chinese Academy of Sciences (CAS) under its Knowledge Innovation Program (Science, 8 January, p. 150). “This will be the biggest reform in 50 years for China's astronomical research and should be marked as a grand holiday,” says the center's deputy director, Wang Jingxiu, a professor at the Beijing Astronomical Observatory.

NAOC, established on 23 April, links four major observatories in Beijing, Nanjing, Shanghai, and Yunnan as well as three satellite operations around the country and two planned telescopes, one already under construction. It will receive $6 million a year over 3 years in operating funds, twice the combined budgets of the existing facilities. At the same time, the new center will have a staff of 406, less than one-third the current level of 1429 employees, as some scientists and many administrators and support staff will lose their jobs. “The innovation program has provided an opportunity and financial means for us to deal with the problems of redundant institutions, irrational distribution of resources, and low efficiency,” says Ai Guoxing, head of the Beijing Astronomical Observatory and director of the new center. The separate observatories and stations will continue to exist for a time, caring for their staffs' nonresearch needs and helping to relocate those whose jobs were eliminated by the reorganization. Ai, a 60-year-old solar physicist, has long been a major force in Chinese astronomy. “He's not very democratic, but he has the vision and the determination to get things done,” says Wang Haimin of the New Jersey Institute of Technology in Newark, a former student and current collaborator of Ai's. “I would expect him to focus on a few important projects.” One key aspect of the new organization is a competitive process that will fund the best scientists, regardless of age and current position, based on peer review by outside panels. A network of 35 research groups headed by principal scientists will be set up to conduct research in nine priority areas,* with the scientists being given a big voice in hiring from outside their institution and in managing their research funds. Twenty-eight such groups have already been set up, each with five to 10 times the research budget of existing teams as well as higher individual salaries. The idea of putting sufficient resources in the hands of the most capable scientists is already paying off, says 33-year-old Luo Guoquan, deputy director of the Yunnan Observatory and principal scientist of a research group. Luo says a Chinese colleague, trained at Cambridge University, scrapped plans to return to England after hearing about the new center. Now the 34-year-old has been made a principal professor, too. “We young people are pleased to see the improvements,” says Luo, “and we feel we have brighter prospects now.” The new center will also assume responsibility for assigning time on the major facilities, which include a 2.16-meter optical telescope in Xinglong station in Hebei province northeast of Beijing, a multichannel solar magnetic telescope on the outskirts of Beijing, two 25-meter radio telescopes in Shanghai and Xinjiang, and a 13.7-meter millimeter-wave telescope in Qinghai. Proposals from Chinese scientists outside the center and foreign scientists will compete for two-thirds of the allocation, says Ai, who adds that observatories and stations have traditionally resisted opening their facilities to outsiders, including CAS scientists. “It has been a waste of resources and low efficiency,” he says. For example, Wang notes that the Qinghai millimeter telescope was used by only three or four groups a year because of its remote location in western China and a tiny research budget. To make it more attractive to researchers, NAOC has created 18 positions and is soliciting applications from young astronomers with doctoral degrees. Another important task for the center will be to plan and carry out new government-funded megascience projects. A 4-meter field telescope called LAMOST (Large Area Multi-Object Spectroscopy Telescope) is currently under construction at Xinglong station (Science, 8 November 1996, p. 915), while a space solar telescope and a 500-meter-wide radio telescope called FAST are still under development (Science, 7 August 1998, p. 771). CAS is also moving ahead with plans for a new optical telescope, in the 3- to 4-meter range, for the Yunnan Observatory. It would replace a 1-meter telescope near Kunming whose capability has been degraded by light pollution. NAOC's arrival on the scene should help cut through red tape on domestic projects, says Wang, as well as offering one-stop shopping to potential collaborators from abroad. That streamlined management is also expected to foster partnerships with Chinese universities that operate strong programs in basic astronomical research. Three subcenters will be created in Beijing University, Nanjing University, and the China Science and Technology University in Hefei, with money mainly coming from the center. If all goes smoothly, NAOC will evolve into the National Astronomical Observatory by the end of 2000 and the separate facilities will lose their individual identities. The triple attraction of new telescopes, more money, and an improved working environment is expected to bolster Chinese astronomy in the coming decade, says Wang. “It is impossible to make conditions as comfortable as our foreign counterparts,” he says, “but we will try our best.” • *The nine fields are large-scale structures, galaxy formation and evolution, high-energy and catastrophic processes, star formation and evolution, solar magnetic activities and solar environments, space geodynamics, dynamics of solar system and celestial bodies, space-based observations and exploration, and methods and techniques in astronomy. 2. MANTLE DYNAMICS # Iceland's Fires Tap the Heart of the Planet 1. Richard A. Kerr On Earth, what goes up must come down. But inside the planet, just how things work hasn't been so obvious. Researchers probing Earth's interior have traced great sheets of rock—former sea floor—plunging deep into Earth's mantle, hard against the molten iron core (Science, 31 January 1997, p. 613). But do such slabs—or any other rock from these great depths—ever come up again? Many researchers have thought that mantle plumes—narrow columns of rising hot rock that feed volcanoes at hot spots like Hawaii and Iceland—tap the deep mantle, but they had no hard evidence, and others argued that plumes have shallower roots. Now, with a sharper view of the lower mantle, a pair of seismologists is extending Iceland's plume to the very base of the mantle, 2900 kilometers down. By using new tricks to process earthquake waves into a seismic image, seismologists Harmen Bijwaard and Wim Spakman of Utrecht University in the Netherlands have produced an image of a plume rising from near the bottom of the mantle all theway up to Iceland's surface, as they report in a recent issue of Earth and Planetary Science Letters. “It's really cool—you can see the plume going down,” says seismologist Richard Allen of Princeton University. “I would now say there's good evidence—although not yet proof—that the Iceland plume originates from the core-mantle boundary.” Seismologists have tried to map the plume by combining the travel times of waves passing under Iceland from distant earthquakes to form a kind of CT (computerized tomography) scan. Hot rock slows seismic waves, so researchers could check such images for a hot, seismically slow column of rock. But with conventional methods, the limited seismic data give such fuzzy views of the mantle that a complete plume can't be recognized, although researchers have found parts of it. Using seismic waves recorded on Iceland itself, for example, seismologist Cicely Wolfe of the Woods Hole Oceanographic Institution (WHOI) in Massachusetts and her colleagues imaged a 350-kilometer-wide plume extending to 400 kilometers beneath the island; Allen recently narrowed the width of that plume to 200 kilometers. Two years ago, Yang Shen of WHOI and his colleagues showed that at greater depths, something hot and narrow beneath Iceland is apparently raising the traditional 660-kilometer boundary between upper and lower mantle by 20 kilometers (Science, 6 February 1998, p. 806). And last fall Donald Helmberger and his colleagues at the California Institute of Technology in Pasadena seismically detected a 250-kilometer-wide dome of partially molten rock at the bottom of the mantle beneath Iceland—just the kind of structure that has been proposed as a likely source for plumes (Science, 31 January 1997, p. 614). To fill in the rest of the picture, Bijwaard and Spakman brought the lower mantle into better focus by varying the image's resolution from place to place, sharpening it where many wave paths happen to pierce a particular spot, rather than using a uniform but lower resolution, as is typically done. The pair also calculated the 7.6 million seismic wave paths individually rather than averaging many wave paths together, as is often done to save computing time. The end result is what they call “the first rather detailed image of a whole mantle plume.” The pictured plume is no textbook example—it's still as much as 500 kilometers wide versus the expected couple of hundred kilometers, twisted a bit, and even seems to branch at one point. And because few seismic waves passed through the top and very bottom of the mantle, the image is not very reliable there. But “for most of the lower mantle,” says Bijwaard, “vertical resolution is very good, implying the continuous structure seen there probably really is continuous.” Allen agrees, and Rob van der Hilst of the Massachusetts Institute of Technology—whose own global images from the same raw data hadn't shown a distinct plume—also accepts that the new work “is indeed evidence for a continuous plume” in the lower mantle. If Iceland does build itself with rock from the very heart of the planet, as now seems likely, then the circle will be closed: Earth's surface not only sinks into the depths but deep rock feeds the surface, offering scientists another window into the planet's deepest depths. 3. TRANSGENIC ANIMALS # Fertility Therapy May Aid Gene Transfer 1. Michael Hagmann The premillennium frenzy about cloned drug-secreting sheep and cows, or pigs that have been given human genes in hopes of using them as organ donors, tends to gloss over the fact that introducing foreign genes into animals other than mice is still very difficult. Because current techniques—which primarily involve injecting DNA directly into fertilized eggs—have only a modest success rate, costs can soar to more than$300,000 for a single cow carrying a foreign gene. Now, genome tinkers may have a new tool for beefing up animal genomes with exotic DNA: sperm.

On page 1180, embryologists Anthony Perry, Teruhiko Wakayama, and Ryuzo Yanagimachi of the University of Hawaii School of Medicine in Honolulu and their colleagues report that they have used sperm to transfer a foreign gene into mice. The technique is a modification of a method originally developed by Yanagimachi for injecting sperm into eggs that is now standard for in vitro fertilization of human eggs. About 20% of the mouse pups born in the group's experiments carried the transgene—which is “definitely on the high side of what's done routinely,” says George Seidel, a reproductive physiologist at Colorado State University in Fort Collins. Adds embryo physiologist Gary Anderson of the University of California, Davis, “If this works in other species, people will jump on it like a banshee.”

For some this is a giant “if,” however. Reproductive biologist James Robl of the University of Massachusetts, Amherst, describes the technique as “very interesting.” But he adds, “I'm not sure how widespread its applicability will be.” He and others note that before the mouse work, the sperm injection method, known as ICSI for intracytoplasmic sperm injection, had been shown to work well only in humans.

Researchers have been trying to use sperm to create transgenic animals for about 10 years. In early experiments, they simply mixed the DNA to be transferred with sperm and used the mixture for in vitro fertilization. Although the technique initially appeared to work, Anderson says, “even leading people in the field haven't been able to repeat the original result.” Today, there seems to be general agreement that such sperm-mediated gene transfer succeeds, but with highly variable efficiency.

About 2 years ago, Perry decided to take a quick shot at seeing whether Yanagimachi's ICSI method would do better. The researchers first removed the propellant tails from sperm and subjected the sperm heads to freezing or detergents to disrupt their cell membrane. They then mixed the sperm heads with a gene encoding green fluorescent protein (GFP). To inject this mixture, the team used a so-called piezoelectric device, which drives the tiny injection pipette very fast and precisely into mouse eggs.

Compared to the manual injection devices used in human fertility clinics, piezoinjection seems to be far less disruptive for the egg. “The mouse egg is the most fragile of [all species]. This study would have been impossible without the piezodevice,” says Robert Wall, a geneticist at the U.S. Department of Agriculture's Agricultural Research Service in Beltsville, Maryland.

When the researchers injected the GFP gene along with untreated sperm, only 26% of the embryos carried the transgene. But it was present in up to 87% of the early embryos produced by the detergent-treated or frozen sperm—as indicated by the embryos glowing green under an ultraviolet lamp. Ultimately, however, only about 20% of the newborn pups that developed from the injected eggs carried the GFP gene. Perry suspects that GFP has a deleterious effect on embryonic development, so the transgenic fetuses tend to be selectively aborted. Whatever happens, a majority of the animals that end up with the transgene transmit it to their offspring.

Experts in the field of transgenic animals welcome the new study for adding yet another gimmick to their arsenal of techniques. “It's always nice to have a large collection,” Robl says. And the longer the list, he adds, the easier it might become to make transgenics in other species. “This study will create a stir of activity,” Seidel says. “I'm sure a lot of people will be trying it.”

Whether the expected activity will pay off is, however, a matter of debate. “It will probably not supersede anything that is out there because the efficiency is not that much better than [DNA] microinjection,” says Wall. (Current DNA microinjection has a success rate of about 10% in mice.) But Perry says that comparing his new technique to the much more advanced DNA microinjection is unfair. He notes that the efficiency of the latter has increased by three- to fourfold since the first experiments. “If we have a similar increase, almost every animal will be transgenic,” Perry concludes.

But the skeptics contend that even if that can be accomplished for mice, using ICSI to transfer genes into other species still might not work. “ICSI is technically quite challenging; it's not as simple as DNA microinjection,” says Wall. Perry disagrees, saying that it can't be that hard because “three of the four people who performed the injections had never made a transgenic animal in their lives. It'll take a little time [to perfect the technique]; we're only at the start.”

And he should have help in the effort. Despite his reservations, Wall recently purchased one of the $10,000 piezodevices to see for himself whether ICSI lives up to the claims. 4. NEUROBIOLOGY # Time Cues Help the Brain See Objects 1. Marcia Barinaga One of the most critical jobs our visual system performs is to group the myriad features of a visual scene into the discrete objects that form the scene. As New York University neuroscientist Anthony Movshon explains, “It is very hard to analyze an image until you have broken it into the objects it contains.” Fortunately our brains are virtuosos at this task, called binding, linking even the disjointed parts of an object that is heavily obscured, such as a figure seen through dense woods. Work described in this issue reveals one of the brain's tricks for sorting through scenes so efficiently. Using psychophysics experiments, in which human volunteers solve visual puzzles, neurobiologists had already shown that cues such as color, continuity, texture, and movement help the brain assemble objects from their parts. On page 1165, Randolph Blake and graduate student Sang-Hun Lee of Vanderbilt University in Nashville, Tennessee, provide the most convincing demonstration yet that the visual system can also exploit timing information, even in the absence of any other cues, grouping features of the scene that change at the same time into coherent objects. This is “perhaps the most interesting new work in visual psychophysics to come out in the past 10 years,” says neuroscientist William Newsome of Stanford University School of Medicine. “It gives us even more appreciation than we had before for the clever approaches that biological visual systems take to make sense of the visual world.” In addition, the work has spurred a vigorous debate among neuroscientists about whether it provides support for a controversial hypothesis postulating that neurons in the visual system that are responding to different parts of the same object become bound together by firing in synchrony. For their experiments, Blake and Lee tested people's ability to see shapes in abstract patterns of small circular patches on a computer screen. As anyone knows who has watched a well-drilled marching band spell out the school's letters by having some members march in a different direction than the rest, a common direction of motion can make shapes jump out of such a pattern. But Blake and Lee wanted to test whether the brain could see a shape based not on common motion, but simply on common timing of visual changes. Several recent studies had suggested that the brain could use timing information for binding, but Blake notes that those tests—based on flickering patterns of dots—contained other cues as well. He and Lee wanted a test in which timing was the only cue to the form hidden in the scene. To do this they used an array of patches that resembles a completely chaotic marching band, whose members move about the field in a random way. In a rectangular region at the center of the field, however, all of the patches repeatedly changed their direction of movement simultaneously, at irregular intervals. People viewing the test could see the rectangle well enough to tell whether it was oriented vertically or horizontally. (Demonstrations of these displays can be seen online.) “They have very carefully removed all the spatial information,” says Movshon, and “demonstrated that … you can drive a binding process purely with temporal stimuli.” It is reasonable that timing would be a binding cue, says Movshon, because when a real object moves, all its parts generally begin to move at the same time. What's more, in some natural situations, timing alone, rather than a shared direction of motion, might be the main cue. “Imagine a disturbance in a forest, created by a predator moving around in a tree,” he says. A synchronous change in the movement of leaves on that branch may be the only clue to the predator's presence. Lee and Blake's experiment doesn't identify the brain neurons responsible for seeing the rectangle, but Newsome notes that movement-sensitive neurons reacting to the synchronous change in motion are likely to fire synchronously. And that notion sounds familiar, says Stanford neuroscientist David Heeger. It “makes you think immediately,” he says, about the hypothesis advanced by neuroscientist Wolf Singer of the Max Planck Institute for Brain Research in Frankfurt, Germany, and others, that synchronous firing binds together neurons perceiving the same object (Science, 24 August 1990, p. 856). Heeger notes, however, that the subjects in Blake and Lee's experiments are responding to synchronous timing in the visual image itself. Singer, on the other hand, holds that neurons responding to different parts of an object will synchronize even if there is no timing signal coming from the object. “I don't think [this new experiment] bears at all on the question of whether the brain uses temporal synchrony to signal binding for other kinds of patterns,” says Movshon, and Newsome, Heeger, and others adamantly agree. But Singer does not. He is elated, he says, to see “direct evidence that synchrony is interpreted by the cortex as a signature of relatedness when it is induced externally. It would be strange,” he adds, “if internally generated synchrony … were interpreted in a different way.” While that debate rages on, researchers are jumping on other avenues opened by the Vanderbilt team's findings. They “tell us something new and important about what the visual system can do,” says Newsome. And that, adds Heeger, “opens up the opportunity for trying to measure and understand the underlying neural basis. Immediately you think, ‘What is it that the neurons are doing; what is the neural code for this?’” A number of labs, he says, are sure to design experiments to search for that neural code. 5. PLANETARY SCIENCE # Asteroids Form Rocky Relationships 1. Richard A. Kerr A run-in with a huge asteroid is bad enough, as movies like Deep Impact have made all too graphic. Now there's a scenario for the next round of threat-to-humanity movies: double impacts. Sightings of asteroids with companions—the latest of them just reported on the Web—are convincing astronomers that such pairs are far from rare. Indirect evidence, such as paired impact craters on Earth, had already hinted that asteroids sometimes come in pairs. In the late 1970s, some astronomers watching stars blink out as asteroids passed in front of them reported extra flickerings that might have been caused by companions—although colleagues remained skeptical (Science, 17 July 1987, p. 250). The first direct proof that asteroids can have moons of their own came when the Galileo spacecraft flew by 56-kilometer Ida in 1993 and photographed tiny Dactyl, a 1.5-kilometer body orbiting about 100 kilometers away. And now astronomers observing from the ground have detected a much heftier companion around the 214-kilometer asteroid Eugenia. In a meeting abstract newly posted to the Web (scorpio.tn.cornell.edu/ACM/web_abs.html), astronomer William Merline of Southwest Research Institute in Boulder, Colorado, and his colleagues report that they spotted a 15-kilometer satellite orbiting about 1200 kilometers from Eugenia. Eugenia and its satellite are a single fuzzy spot of light in an ordinary telescope, but late last year, in the course of a 200-asteroid search for satellites, Merline's group was able to separate them with the 3.6-meter Canada-France-Hawaii Telescope (CFHT) on Mauna Kea, Hawaii. The CFHT was equipped with an adaptive optics system that precisely undoes the blurring effects of atmospheric distortion (Science, 27 June 1997, p. 1994). More candidates for binary asteroids are emerging from observations of the pulsating brightness of asteroids that pass near Earth. Most asteroids reflect varying amounts of sunlight as they rotate because of their irregular shapes, but a half-dozen so-called near-Earth asteroids (NEAs) observed by Petr Pravec of Ondrejov Observatory near Prague and his colleagues and by Stefano Mottola of the DLR in Berlin flicker as if one body is periodically passing in front of or behind another perhaps twice its size. Although the Eugenia observations are “pretty hard evidence” for a satellite, says astronomer Alan Harris of the Jet Propulsion Laboratory in Pasadena, California, the light variations of at least a couple of the NEAs are “highly suggestive.” As satellites of asteroids have proliferated, theoreticians have been trying to explain how they formed. After Dactyl was spotted near Ida, some suggested that the pair came together after a collision shattered a precursor body into a swarm of smaller asteroids, and a larger fragment managed to capture a smaller one gravitationally. But no one has tested this idea with detailed calculations. Noting that smaller collisions may have turned many other asteroids into rubble piles, William Bottke of Cornell University and Jay Melosh of the University of Arizona, Tucson, suggested another scenario in 1996: Earth's gravity, they said, could split a rubble-pile asteroid in two if it passed nearby. Eugenia's satellite is more perplexing, says Melosh. Although Merline calculates from the satellite's orbit that Eugenia has a low density and therefore is probably a rubble pile, it is in the main belt of asteroids between Mars and Jupiter and never would have passed close by a planet. Observes Melosh: “That will certainly give us theoreticians something to chew on.” 6. HIV # French-Led Therapy Fund Kicks Off in Africa 1. Michael Balter PARIS—Potent antiviral drugs have begun to cut the death rate from HIV infection in developed countries, allowing many infected people to live longer and relatively normal lives. But in the developing world, where 90% of the world's estimated 35 million HIV-infected people live, the high cost of these drugs makes them virtually unobtainable, and death rates continue to climb (see next story). In December 1997, French health minister Bernard Kouchner, supported by France's president, Jacques Chirac, launched a campaign to reduce this global inequity: He proposed the creation of an international fund to subsidize anti-HIV therapies in the developing world. Last month, after nearly a year and a half of often frustrating toil by Kouchner and his aides to raise money, the fund announced its first projects. The Fund for International Therapeutic Solidarity (FSTI), as it is now called, will provide about$1.7 million over the next 4 years to support therapy and prevention programs for selected groups of patients in the West African nation of Côte d'Ivoire, with emphasis on preventing the transmission of HIV from infected mothers to their infants. In addition to FSTI's contribution, a charitable foundation set up by the French arm of the Glaxo Wellcome drug company will kick in about $250,000 and the Côte d'Ivoire government will supply$1 million. Other partners—including the U.S. Centers for Disease Control and Prevention in Atlanta and UNAIDS, the Geneva-based United Nations AIDS program—will lend logistical support and expertise.

Kouchner has had a tough time raising money for his fund: The sole contributor so far is the French government, which has provided $4 million in start-up money. “It has been a terrible battle” to get the FSTI started, says Eric Chevallier, Kouchner's senior adviser in charge of the effort. “Obviously we have been rather frustrated.” On the other hand, Chevallier says, more recently there has been a “noticeable evolution” of positive attitudes from potential donors. For example, an additional$3.3 million may soon be forthcoming from the European Commission, after a vote by the European Parliament last October directing the commission to make a donation. And Chevallier says Kouchner's staff is currently talking with the World Bank about participating in the fund.

Because of the limited funds, the pilot project in Côte d'Ivoire will target a small fraction of the at least 700,000 people estimated to be HIV-positive in that country. It will subsidize “bitherapy”—two drugs that inhibit the key HIV enzyme reverse transcriptase (RT)—for just 500 patients, chosen among infected AIDS prevention activists and patients who have already participated in clinical trials for antiviral agents. A second, larger effort will support testing and therapeutic follow-up for 20,000 pregnant women and their families in Abidjan, the nation's capital. Those who test positive will be offered a “short course” of antiviral drugs to prevent mother-child transmission, and mothers in the advanced stages of HIV infection will be offered triple therapy—generally two RT inhibitors and a protease inhibitor directed at another HIV enzyme.

Although they welcome the program, health officials agree that it will have only a small impact on the explosive AIDS epidemic. Makan Coulibaly, coordinator of the Côte d'Ivoire government's HIV treatment access program, says that with about 90,000 women giving birth each year in Abidjan, the program will catch only a fraction of the potentially infected population. “It is very easy to be paralyzed by the magnitude of the problem,” says Joseph Saba, who is in charge of UNAIDS's drug-access initiatives. “But do we wait until everything is perfect and everyone has access, or should we go on a step-by-step basis?”

Despite Kouchner's limited success at raising money for the FSTI, a dozen other countries have lined up to request money from the fund. With the money remaining, new programs in Uganda and Morocco will be starting up soon, and Chevallier says that once the European Commission begins to contribute, the number of recipients should expand considerably. Says Saba: “This initiative has one great merit—it is trying to accomplish something.”

7. HIV

# AIDS Now World's Fourth Biggest Killer

1. Michael Balter

PARIS—AIDS is now the fourth leading cause of death in the world, and the number one killer in Africa, according to figures released this week by the World Health Organization (WHO). The disease has moved up several notches from last year's ranking as seventh leading killer worldwide, according to WHO's latest World Health Report. Only ischemic heart disease, cerebrovascular disease, and acute lower respiratory infections outrank AIDS on the international death list. In Africa, AIDS caused an estimated 1,830,000 mortalities in 1998, twice as many as due to malaria, which is now relegated to the number two spot on the continent's roster of lethal diseases.

View this table:

Bernhard Schwartländer, senior epidemiologist for UNAIDS, the United Nations AIDS program, says that some of the change in disease ranking is due to new and improved methodologies for estimating disease mortality, which have revised estimates of some diseases downward while AIDS cases have been skyrocketing. Nevertheless, the new figures dramatically vindicate warnings late last year by UNAIDS that the epidemic is still raging out of control (Science, 4 December 1998, p. 1790). UNAIDS estimates that new infections by HIV, the virus that causes the disease, are increasing by at least 6 million each year. But a UNAIDS study released last month indicated that donations to international AIDS programs have failed to keep up with the growth of the epidemic.

“AIDS is now the [single] leading infectious disease killer in the world, and the number one killer of Africans,” Peter Piot, UNAIDS's executive director, told Science. “It's an outrage that the international community is only investing $150 million each year to stem the spread of HIV in Africa.” 8. SPACE STATION # A$100 Billion Orbiting Lab Takes Shape. What Will It Do?

1. David Malakoff

Plans have become reality for the international space station now being assembled in orbit. But only a tiny share of the research portfolio has been selected, and the scientific challenges remain daunting

Sometime next spring, if all goes as planned, space shuttle astronauts will deliver a full-sized plastic model of a human head and torso—outfitted with real teeth and bones—to a 10-meter-long orbiting laboratory shaped like a soda can. Known to its creators at NASA's Johnson Space Center in Houston, Texas, as Fred, the mummylike mannequin is already a veteran space traveler; an earlier version flew last year aboard the space shuttle. Nevertheless, Fred's latest trip into space will make history: He'll be the star of the first major scientific exercise aboard the international space station (ISS).

Much of Fred's expected supporting cast, however, hasn't even made it to tryouts yet. Although engineers and scientists have spent nearly 2 decades planning the station, and 5 months have passed since spacewalkers mated its first two segments, fewer than 100 experiments—a tiny fraction of the work scientists hope to perform during the station's planned 10-year life—have been fully approved for launch. And the pace isn't likely to pick up anytime soon. Funding cuts have delayed the construction of hardware needed to take advantage of the station's low-gravity environment, and the demands of assembling the station over the next 5 years will leave astronauts with little time for science. In addition, the megaproject's 16 partners are still hammering out a process for collaboratively choosing experiments. Interested researchers also must overcome a thicket of technical obstacles, from vibrations that could wreck sensitive studies to a cloud of contamination floating around the station that may coat and possibly blind sensitive instruments. “There are a whole lot of unanswered questions,” concedes Kathryn Clark, NASA's chief scientist for the project.

A debate over the scientific value of the station, which could cost $100 billion to build and operate, has been boiling ever since the idea was first floated (see sidebar on p. 1106). Now that the station is becoming a reality, however, the discussion has shifted to its scientific capabilities, the rationale behind the experiments to be flown, and the nature of the results to be gleaned. And it's not a moment too soon: Next week, the shuttle Discovery is scheduled to carry supplies to the orbiting construction site, and its first science segment—the U.S.-built laboratory Destiny—is due to be bolted on next spring. “For better or worse, this mission is now under way,” says astrophysicist Claude Canizares, chair of the National Academy of Sciences' (NAS's) Space Studies advisory board. COMING ATTRACTIONS Although researchers have lots of ideas for station science, most are still years away from reality. And although most early experiments have been selected nation by nation, the plan is to employ international peer review. So far, however, only life scientists have participated in a global competition. View this table: ## Potential payoff Although it's still far too early to assess the quality of the research scheduled to fly on the station, Fred is as good a place as any to start looking for answers about the station's potential scientific payoff. NASA radiation researcher Gautam Badhwar will be spending the next few months calibrating his equipment as he methodically prepares Fred for flight. Once Fred is aboard, hundreds of sensors encased in the dummy's mock organs, along with several related instruments nearby, will help researchers determine how much harmful cosmic radiation is penetrating the bodies of the station's crew mem-bers. Eventually, scientists hope the information will lead to better safeguards for space travelers to Mars and elsewhere. Fred illustrates both what is right and wrong with station science, which government officials repeatedly cite as the chief rationale for building it. Proponents say that Fred's scheduled 14-week run makes use of one of the station's most valuable assets: time. Whereas researchers using the space shuttle have at best a fortnight to collect data, those using the station can plan experiments lasting months or years. The longer durations will allow Fred, for instance, “to collect far more data than was possible when [he] flew on the shuttle,” says Badhwar. The station also gives scientists a chance to compile larger, more statistically valid data sets by repeating experiments. “You won't have to wait five frustrating years to repeat or retry an experiment,” says crystallographer Lawrence DeLucas of the University of Alabama, Birmingham, a veteran of growing protein crystals aboard the shuttle. Researchers in other fields are also looking forward to extended experimental time aboard the station's six pressurized research modules and dozens of external payload sites. Combustion and materials scientists, for instance, are planning to tinker with burning droplets and solidifying metals in an effort to uncover basic properties, while life scientists will study how people, plants, and even insects react to life in microgravity. Outside the station, astronomers want to hang instruments that will monitor everything from the sun to x-rays, while earth scientists analyze the planet's atmosphere and land forms. There will also be long-duration tests of new materials and technologies, such as laser-communication systems, that might eventually fly on craft bound for deep space. And although corporate interest is low, station officials are also counting on a host of commercial research payloads, such as Earth-monitoring cameras. But those who believe ISS science will produce few useful results see Fred as a poster child for the station's limitations as a scientific platform. For example, the station's low orbit means that the mannequin will not record data from further out in space, where Earth's atmosphere provides no protection from radiation. That could make the findings of little help to interplanetary voyagers. The low orbit also makes the station useless to many astronomers, as Earth's upper atmosphere blocks many forms of light and radiation and prevents some kinds of instruments from getting an unobstructed view of distant objects. Altitude is only one of many technical challenges facing researchers. Contrary to its sedate image, for instance, the station will flex and roll like a rubber raft bobbing on a long swell because of atmospheric drag. That movement will make it difficult to operate detectors that need to lock onto a particular patch of Earth or sky. To compensate, the Europeans are building a highly accurate pointing device that will initially be used to allow SAGE III, an instrument analyzing atmospheric chemistry, to stay on target. The station will also require periodic lifts into a safer orbit to counteract its continuous sinking toward Earth. Those rocket burns, along with frequent shuttle and supply rocket dockings, will cause the station to shake and vibrate, disrupting sensitive experiments, perhaps even those mounted in special vibration-resistant racks. Some experiments will also be vulnerable to “g-jitter,” the constant variation in gravitational force in different parts of the station. As a result, researchers will have to consider whether their experiments will work outside the station's “gravitational sweet spot,” says Clark. Instruments hung outside the station, and a special Earth-observing window in the U.S. lab, may face another difficulty: a cloud of contamination that is expected to hover around the station, coating exposed instruments with a potentially troublesome patina. Some of the gunk will come from the station's structure, which will produce gases when it is exposed to the vacuum of space. Another source is the station's 33 exhaust vents, which will also spew gases and water vapor. Visiting rockets will also leave behind a trail of particulates. “It could be like living in Pig Pen's cloud,” says one NASA engineer, alluding to a cartoon character perpetually surrounded by a storm of dust. Nonetheless, he predicts such problems “won't be insurmountable.” Astronauts, for instance, will be able to periodically remove a shield protecting the window and return it to Earth for cleaning. ## Time and money Other challenges involve human, not engineering, issues. It's not yet clear, for instance, whether Russian cosmonauts will fully cooperate in research needing human subjects; some have said they want to be paid extra. It is also unknown how much the astronauts—who will be busy juggling everything from repairs to daily chores—will have leeway to say “no” to ground-based scientists seeking extra help with their experiments. To head off conflicts that sometimes marred researcher-astronaut relations aboard Mir, launched by the Soviet Union in 1986 as the world's first long-duration space station, Clark has jokingly suggested creating a “People for the Ethical Treatment of Astronauts” group. “Researchers can't expect the crew to answer to their beck and call,” she says. Clark and other NASA officials also worry that there will not be enough training time to educate the seven-member station crews in the array of experiments they will be operating. Many scientists hope to operate their experiments by remote control from Earth to get around the labor bottleneck. But it's an open question whether the station's communications systems will be up to the task. “Ideally, you automate so that an astronaut doesn't have to be involved,” says Clark. Another way to attack the problem, some station planners say, is to run station science in “campaign mode,” with each crew specializing in experiments in a particular field, such as fluid physics or materials science. That approach could also help stretch the station's limited supply of electricity by temporarily focusing it on investigations, such as combustion experiments, that require lots of power. The biggest obstacle to a high-quality scientific payoff, however, may be money. At NASA, science programs have been slowed by construction overruns that have forced agency officials to repeatedly “borrow” money from the station's research accounts. That, in turn, has delayed the completion of science hardware, including fluid physics and animal study facilities. NASA Administrator Dan Goldin has put a positive spin on the cuts, saying the delays will keep the science program “more in phase with” the pace of station construction. Unless Congress adds money to NASA's budget, however, White House forecasts call for the United States' ISS research fund to get$363 million less over the next 6 years than once planned. The cuts mean the science account will grow more slowly than envisioned, from about $350 million this year to$550 million in 2004.

To stretch the dollars, NASA officials are emphasizing experiments that—like Fred—recycle equipment that has already flown and studies using generic facilities that support more than one experimenter. NASA's combustion science program, for instance, is “encouraging proposals that fit existing hardware,” says manager Merrill King. The agency's need to stretch funds—and thus maintain an active corps of researchers—has even led it to tilt temporarily toward funding more ground-based projects, which are cheaper to carry out. The agency's Microgravity Research Program, for instance, now funds seven Earth-bound experiments for every one destined for flight, up from a 3:1 ratio in 1991.

Inadequate funding is a much more serious problem in Russia, where economic uncertainty has paralyzed many scientists hoping to place experiments aboard two Russian science modules, tentatively scheduled to be launched in 2004 (Science, 20 November 1998, p. 1391). “We have lots of plans but very little money,” says earth scientist Vladimir Kuznetsov, deputy director of the Russian Academy of Sciences' Institute of Terrestrial Magnetism in Troitsk. Russia's problems have provided an unexpected bonus to U.S. researchers, however: Last year, in a move to provide the teetering Russian Space Agency with cash, NASA bought hundreds of hours of cosmonaut time to carry out science during the assembly phase (Science, 9 October 1998, p. 206).

NASA officials hope Russia's funding crunch will foster greater teamwork by forcing it into the arms of several international working groups trying to coordinate station science. Although leading Russian scientists have urged their government to participate in the joint agenda-setting, the Russian modules have so far remained “a separate world,” says Clark. Other NASA officials worry that the war in Kosovo could further fray an already testy relationship.

Outside Russia, however, station planners have been pleasantly surprised by the number of scientists seeking to win funds for station-related research. Japanese scientists have submitted more than 750 proposals to two early, Japan-only funding rounds, with about 50 projects still in the race for a launch spot. In 1997, European researchers offered nearly 100 entries in a continent-wide competition for instruments to hang outside the station. At the same time, the first international call for life science experiments attracted more than 500 proposals from the United States, Canada, Japan, and Europe. Peer reviewers eventually deemed 27 worthy of flight, with more than half of the projects coming from outside the United States.

With plans to use international peer review as the norm for selecting station research, however, scientists are wondering how funding will work. For instance, some ask what will happen if NASA or the European Space Agency (ESA) is unwilling to fund a top-rated project suggested by its scientists. Will lower quality science from a richer agency take its place? Or will the partners create a common fund to pay for the best experiments, regardless of origin? Some U.S. scientists also wonder if their station projects will be competitive with proposals for cheaper, unmanned platforms to be flown on different spacecraft. To avoid that problem, some would like NASA to set aside special funds. But NASA's W. Vernon Jones, who oversees the agency's space science research programs, is against the idea. Earmarking funds for the ISS, he says, “would send the signal that this is lower quality science.”

Station partners are also working out how to divvy up the station's space. The United States controls the bulk of its power and attached payload sites and, in exchange for launching the European and Japanese science modules, NASA has claimed title to almost half of the space in each. In contrast, Russia has kept 100% control of its two science modules. But the allocations are constantly shifting, as the partners barter space and equipment in what Clark calls “the dance of trading science.” ESA, for instance, is building the pointing device, a lab freezer, and other equipment in exchange for the right to place two astronauts and a few experiments aboard the station before its own Columbus Orbiting Facility arrives in 2004.

## Human research first

Such dickering is expected to continue even after the station's science program begins in earnest next spring with the launch of the Human Research Facility (HRF). The HRF, the U.S. hardware that will house Fred, will occupy two of the lab's refrigerator-sized experiment racks. It will bristle with more than a dozen instruments, including an ultrasound imager and a high-tech bathroom scale that can measure a body's mass in microgravity. Astronauts will be able to use the facility to monitor their health, while Earth-bound researchers will collect data on one of space travel's most pressing questions: Why does life in low gravity disrupt perception, promote bone loss, and cause other health problems?

“The human studies are in many ways the most defensible science planned for the station, assuming you believe in manned exploration,” says one biomedical researcher involved in an NAS review of the program. Other findings that could eventually aid astronauts, he notes, might come from studies of how latent viruses carried by the crew members respond to the stress of life in orbit, and how genes involved in growth and sleep cycles behave in the absence of normal gravity.

Some life science research planned for the station is more controversial, however. In particular, NASA's long-running effort to grow protein crystals in space—slated to get major attention aboard the station starting next year—has generated intense opposition. Last summer, a committee convened by the American Society for Cell Biology (ASCB) called on the space agency to kill the space-based portion of the program, concluding it had made “no serious contributions to knowledge of protein structure or to drug discovery” (Science, 24 July 1998, p. 497).

In March, the program attracted more criticism after NASA issued press releases claiming that structural data from space-grown crystals had helped an international team of researchers based at the University of Alabama, Birmingham (UAB), and nearby BioCryst Pharmaceuticals to develop a promising flu drug. The claim infuriated one of the researchers involved, biochemist W. Graeme Laver of the Australian National University in Canberra. He says his one-time funder inflated the importance of its space-based work in “another pathetic attempt” to boost the crystal program's image. In fact, Laver says, the single space-produced crystal involved in the project was grown aboard Mir without NASA's help. “And it had nothing to do with the drug's development. BioCryst's findings came from crystals I grew on Earth,” he adds.

The twin attacks have put UAB crystallographer DeLucas, a former chief scientist for the station and now head of a NASA-funded research center, on the defensive. He says the ASCB report is “dead wrong” and that low gravity has allowed researchers to grow several dozen kinds of crystals that are larger and purer than those produced on Earth, making it easier for crystallographers to deduce their structure. He also notes that the program has been extensively peer reviewed and is currently under the microscope of another NAS review panel, which will deliver its verdict later this year. As for Laver's complaint, DeLucas believes his former colleague has overreacted to an unfortunate bit of NASA hype. In retrospect, agency officials “could have toned down the [press release],” DeLucas says. “You would have to say there was a little overenthusiasm from NASA,” adds Charles Bugg, BioCryst's president.

A former NASA scientist who helped get the crystal program going in the 1980s, Bugg says this isn't the first time the agency has hyped the science it supports. Other researchers confirm that a similar controversy once beset a second type of crystal-growing experiment planned for the station. Two decades ago, NASA officials drew heavy flak from scientists for touting its preliminary attempts to grow metallic semiconductor crystals aboard Skylab as the beginning of a new age of space-based manufacturing. “It took decades to stop the talk of factories in space” and to bring the rhetoric more in line with reality, recalls NASA's Michael Wargo, who manages its materials science program. Such hype, however, continues to create “a lot of resentment,” says Bugg.

Ironically, some researchers now consider studies of how metals and other materials solidify in microgravity to be one of the station's most intriguing offerings. In particular, there is broad interest in a suite of planned experiments to examine how the complex, fern-shaped metallic crystals called dendrites freeze into shape. The tests, researchers say, will help reveal the basic mechanisms that drive dendrite growth—fundamental information that is obscured on Earth by the tug of gravity. Materials scientists may have to wait years, however, to get into space: A U.S.- and European-built materials science facility—which will include an array of furnaces and freezers, including one that will suspend samples in a magnetic field—will be ready no earlier than 2002.

Researchers interested in using the station as a platform for Earth and space studies face a similar wait. U.S. Japanese, and European researchers have plans for at least a dozen large instruments that will be bolted onto the station's exterior. They include several x-ray observers, a superaccurate atomic clock, a trio of detectors designed to monitor the sun, and a device to scan Earth's surface for forest fires. Although some could orbit on free-flying satellites, researchers say others are too bulky to be launched aboard rockets, or require human tending that only the station can provide.

Perhaps the best known attached payload is the $50 million Alpha Magnetic Spectrometer (AMS), an antimatter hunter proposed by Nobel laureate physicist Sam Ting of the Massachusetts Institute of Technology in Cambridge. NASA eagerly embraced the controversial idea 5 years ago, and an early version of the AMS, funded by the Department of Energy, has already flown aboard the shuttle. An improved model is scheduled to arrive at the station in 2002 and spend 4 years sifting through the rain of cosmic rays for evidence that some are antimatter particles. Although many physicists are skeptical of the project, station proponents say it exemplifies the station's ability to shoulder bulky payloads. The boxy AMS, for instance, depends on having a large surface area exposed for long periods in the hope of catching rare particles. The same is true of several other instruments proposed to study cosmic rays of different energy levels, including the Extremely Heavy Cosmic-Ray Composition Observatory (ECCO), which could reach the station as early as 2003. In ECCO's case, the station also offers the opportunity to retrieve data-collecting arrays for analysis on Earth, notes physicist Thomas Gaisser of the University of Delaware, Newark, who heads the NAS's Committee on Cosmic-Ray Physics. Still, Gaisser says there is “an admittedly political aspect” to decisions to place some of the instruments aboard the station. Although some payloads could fly as independent satellites, he compares the station's users to Charles Darwin aboard The Beagle: “You seize opportunities as they arise.” That philosophy appears to have taken hold in some researchers. At a recent international conference on space station science,* scientists released a variety of trial balloons. They included one scheme to launch minisatellites that would orbit the station and warn astronauts of potentially dangerous changes in space weather, and another to use it as a base camp to assemble a huge neutrino detector from dozens of flowerlike petals. Launching this giant detector—which would look for high-energy particles difficult to spot on Earth—would be “nearly impossible in a single rocket launch scheme,” says Yoshiyuki Takahashi of the University of Alabama, Huntsville. But the station, he says, offers a chance “to think grand thoughts.” It's far too early to know whether such grand ideas, or any of the other science planned for the station, will pan out. But having provided the money, politicians are now looking to scientists to make the best use of the investment. • *Conference on International Space Station Utilization, Space Technology & Applications International Forum '99, 31 January-4 February. 9. SPACE STATION # Making a Deal With the Devil 1. Andrew Lawler NASA Administrator Jim Beggs knew he faced long odds as he journeyed up a Colorado mountain to a meeting 17 years ago. His job was to explain his vision of a human base in orbit to a dozen eminent researchers from the National Research Council's (NRC's) space sciences board. The astronomers and astrophysicists who dominated the NRC panel were no fans of putting people into space. “We were quite skeptical,” confirms Thomas Donahue, then board chair and now professor emeritus of earth and space sciences at the University of Michigan, Ann Arbor. “We didn't want to touch it.” Beggs, a wily former aerospace executive with a folksy manner and a penchant for quoting Shakespeare, had made an international space station his top priority. He had been crisscrossing the globe to assemble a powerful coalition of supporters from government, industry, and academia. The Snowmass, Colorado, meeting in the summer of 1982 was his first major attempt to woo the scientific community. Despite their misgivings, panel members promised Beggs that they would study the potential scientific benefits of a station. But a few months later, a two-page report made it clear that they hadn't changed their minds. There is “no need for a space station to support missions addressing high priority science missions for the next two decades,” declared the report, “Space Science in a Space Station Era.” In particular, the astronomers and earth scientists were not impressed with Beggs's plans to enhance research on the station by building robotic servicing facilities and viewing platforms. However, the scientists did not wholly condemn the project. The report noted that “a manned space station could eventually provide significant opportunities for a number of disciplines in space science” if there were adequate funding. In particular, it noted the “special relationship” between the station and the life sciences, adding that studying humans, animals, and plants in zero gravity would be a prerequisite for long-term space missions. That opportunity excited a handful of researchers in the nascent fields of space biology and microgravity materials who longed for a permanently inhabited space platform. In May 1983, the board's space biology and medicine committee gave Beggs a four-page summary of the space station's potential that envisioned scientist-astronauts conducting a host of physiological experiments along with fundamental biological research. “We were very much against stretching the station to use it for astronomy or earth sciences,” recalls Lou Lanzerotti, a Lucent Technologies physicist and engineer who is a former NRC board chair. “But conducting life and microgravity sciences seemed appropriate.” At the same time, board members chose not to go public with their doubts. “We saw the handwriting on the wall,” says Donahue, referring to the growing coalition that Beggs was building among politicians and industrial leaders. A vocal opposition, he says, could have created enemies while denying scientists potential research opportunities. Beggs agrees: “I don't know what good it would have done for [scientists] to have made a big fuss.” That reticence did not extend to the entire community, however, some of whom worked behind the scenes to discredit the idea. The quiet lobbying campaign included former President Ronald Reagan's science adviser, George “Jay” Keyworth, as well as some officers of the National Academy of Sciences (NAS), the NRC's parent body. The resistance proved futile, however, as Reagan announced his support for the effort in January 1984. And his underlying reasons—a desire to counteract Soviet advances in space—made irrelevant any debate over its scientific merit, according to NASA and NAS officials. From that moment, says former NAS President Frank Press, it was tough “for scientists to shoot it down [because] the primary issue was not the science but the national goals.” Beggs wasted no time in acting on the president's decision. Within a month he had formed a task force of outside researchers to define the scientific uses of the station. Led by Peter Banks, an earth scientist who is now president of Erim Inc. a remote-sensing company in Michigan, the panel concluded in the summer that the orbiting base would benefit a number of disciplines and that it should include a centrifuge for life sciences research. Although Donahue says disdainfully that the panel “carried water for NASA,” Banks believes that the report began the long and grueling process of inculcating scientific values into what was primarily an engineering effort. Looking to further broaden his base of support among scientists, Beggs invited Donahue to his office shortly after the Banks panel was formed to cement an alliance with the scientific community. The administrator promised to allocate 20% of NASA's overall R&D budget to space science. Beggs also agreed to arrange for “high-level” scientific input into the program, with the Goddard Space Flight Center in Greenbelt, Maryland, and Pasadena, California's Jet Propulsion Laboratory—the two space science-dominated NASA centers—playing a central role in designing the station. In return, Beggs reached a tacit understanding with Donahue not to actively oppose the station. Beggs's gambit succeeded. The two men shook hands on the deal and later put it in writing. The arrangement was embraced by Beggs's successor, James Fletcher, who wrote Donahue a few years later that “without a healthy science budget at NASA, the reason for the station's existence is compromised.” The compact has proved remarkably durable, surviving more than a decade of stormy negotiations, lengthy studies, and bitter budget battles in Congress that included attacks by a few scientific societies. But was it the right thing for scientists to do? No, says Margaret Geller, an astronomer at the Harvard-Smithsonian Observatory in Cambridge, Massachusetts, who quit the Banks committee in protest. To her, Banks's committee and Donahue had traded their scientific principles for a slice of station pie. “You don't make Faustian bargains,” she says now. “It was obvious to me the station had nothing to do with science. People were jumping on board to get money.” In response, Donahue and others, including Lanzerotti and Banks, say the decision to work with rather than against the space station has paid off. Space science funding grew through the 1980s along with NASA's overall budget, creating a flotilla of robotic spacecraft and a flood of new data. And life and microgravity researchers are gearing up to conduct experiments on the orbiting base now being assembled. Even so, Donahue acknowledges that the critics have a point and that backing the space station wasn't the scientific community's finest hour. “I made a pact with the devil,” he admits. Although space science has benefited, he adds, “the whole [station] program has been a botched mess.” 10. PHYSICAL ANTHROPOLOGY MEETING # Anthropologists Probe Bones, Stones, and Molecules 1. Elizabeth Culotta COLUMBUS, OHIO—More than 500 presentations at the annual meeting of the American Association for Physical Anthropology here covered all aspects of human nature, from genetics to stone tools. Two highlights explored brain shape in hominids and the original Asian homeland of the first Americans. ## Early Changes in Brain Shape Around 2.5 million to 3 million years ago, the genus Homo had yet to appear, and the small hominids believed to be our close relatives had brains hardly bigger than a chimpanzee's. But controversial reconstructions presented at the meeting suggest that by then, the brains of some of these creatures were already reorganizing toward the human configuration. By inferring brain shape from the contours of fossil skulls, paleoanthropologist Dean Falk of the University at Albany, New York, and her colleagues conclude that in the hominid Australopithecus africanus, brain regions thought to be used in abstract thinking and language were beginning to show some resemblances to those of modern humans. “Brain reorganization started to happen early,” says Falk. The work also showed that the robust australopithecines, heavy-jawed creatures that lie off the main line to humans, lacked these more modern features and had smaller brains than had been thought. These bold conclusions sparked much interest at the meeting, but many of those who attended Falk's packed presentation aren't ready to accept all her interpretations. “Something like this is always exciting,” says paleoanthropologist Leslie Aiello of University College, London. “But I'm not completely convinced that we can take the incipient Homo traits back that far.” Falk and her colleagues examined more than a dozen australopithecine brain endocasts—internal casts of skulls that preserve brain shape. They found several key differences between the three robust species and A. africanus. For example, in the lower part of the frontal lobe behind and above the eyes, africanus brains were gently rounded and expanded in size, somewhat like those of modern humans—albeit overall much smaller—but robust brains were pointy and beak-shaped, like those of apes. In modern humans, says Falk, this area contains a small region called Brodmann's area 10, thought to be involved in abstract thinking and planning. And the front part of africanus temporal lobes was expanded; that's the same brain region active in modern humans when they recognize and name a face, says Falk. She speculates that “the robusts were apelike in their cognitive abilities,” while “for africanus, I wonder if there's something going on with planning and the first glimmers of naming?” Falk says her finding about brain shape could strengthen the case for A. africanus as an ancestor of modern humans. Other researchers note, however, that candidate ancestors such as A. afarensis and the newly discovered A. garhi may also have these features, which would mean there is no special link between A. africanus and Homo. However, Falk's analysis does take the root-chewing robust forms down another peg, by shrinking their brain sizes. Using their new findings about brain shape as guides, Falk and graduate student John Guyer, who is also a sculptor, reconstructed half a dozen australopithecine skulls. The A. africanus size estimates, including that of a specimen called Sts 71, which Falk had suggested was wrongly estimated (Science, 12 June 1998, p. 1714), all matched published numbers. But for four robust skulls, cranial capacity dropped by 30 to 68 cubic centimeters (cc). Falk concludes that the average size of all known robust crania is not close to 500 cc, as previously reported, but 450 cc—the same as that of A. africanus. (A modern human brain is about 1350 cc.) If, as some researchers suspect, the robusts were a bit larger than A. africanus, that gives africanus a proportionately larger brain. But not everyone is willing to buy these new numbers. Ralph Holloway of Columbia University, who did many of the previous reconstructions, sticks by his earlier estimates, in which he used both A. africanus and living apes as models for robust brains. Holloway is also not prepared to accept Falk's “Homo-like” features in A. africanus. He notes that only one or two africanus endocasts are complete enough to show the temporal and frontal lobe features, which could simply be natural variations. “I do not accept their conclusions. … These areas are quite variable in chimpanzee and gorilla, and particularly so in modern humans, and I am distrustful that somehow australopithecine variability was any less,” he says. Even if the distinctions are real, adds paleoanthropologist Bill Kimbel of the Institute of Human Origins at Arizona State University in Tempe, structural differences in the robust skull could force the brain to take a different shape from that of africanus—in which case the differences might have no bearing on the hominids' thinking ability. Still, researchers say, Falk has pointed out brain features to search for in new hominid finds. “Qualitative differences between africanus and robusts are important,” says Aiello. “What remains to be seen is … if they are truly cognitive differences.” ## The New World's Founding Fathers The adventures of the first Americans took them 10,000 kilometers, from the Bering Strait to Alaska, British Columbia, and eventually all the way to the tip of Tierra del Fuego. But where did their journey begin? At a special symposium on the peopling of the New World, two independent presentations of genetic data pointed to roughly the same region of Southern Siberia as the original homeland of the men, if not the women, who eventually colonized the New World. Multiple genetic trails left in the Y chromosomes of living men all lead to a region near Lake Baikal, according to talks by geneticist Mike Hammer of the University of Arizona, Tucson, and molecular anthropologist Theodore Schurr of the Southwest Foundation for Biomedical Research in San Antonio. Markers from the mitochondrial DNA (mtDNA), which is passed down through the mother and so reveals women's movements, paint a more complicated picture. But even so, “we've got a place we can point to on a map now,” says Hammer, “a place for archaeologists to start thinking about connections.” And indeed, archaeologists report that about 20,000 years ago the Baikal region was home to a mysterious people called the Mal'ta, who have been suggested as ancestral stock for New World peoples. Hammer's team tracked the ancient Asian homeland of Native American founding fathers by sampling the Y chromosome—which is found only in males—from 2198 men from 60 populations worldwide, including 19 Native American and 15 indigenous North Asian groups. They sought sites where the Y chromosomes from different populations tend to have different DNA bases. Earlier work had noted that many Native Americans have one particular set of such mutations, called a haplotype (Science, 5 March, p. 1439). Hammer also found this major haplotype, which he calls 1G, in half of all Native Americans. But his large sample yielded five additional New World haplotypes. For example, 25% of all Native Americans carry the set of mutations Hammer calls 1C, and 5% carry 1F; three other haplotypes are found at lower frequencies. To trace these genetic variations back to their source, Hammer sampled more than 1000 men from across Asia. He found that all six New World haplotypes are now concentrated in two centers, northwestern and northeastern Siberia, but the indigenous peoples now in those regions are thought to have migrated from around Lake Baikal. “If you step back and look at the big picture, you're seeing a big generalized region around Lake Baikal,” Hammer says. Those findings fit with independent data presented by Schurr. In a sample of more than 300 ethnic Siberians and 280 Native Americans, he and his colleagues see two primary ancestral Native American patrilineages, which may include one or more haplotypes. One lineage turns up commonly in peoples west of Baikal, like Hammer's 1C. Schurr also sees a sublineage that apparently arose further east in Asia, perhaps near the Amur River, and then spread both west into Siberia and further north toward the Bering Strait and eventually the New World. Schurr says that mtDNA markers hint at several Asian source areas, including one in Mongolia, perhaps indicating different Asian roots for the men and women who first populated the New World. And all this complexity suggests multiple migrations from Asia, say Schurr and Hammer. Other geneticists have suggested a single migration, but the diversity of markers in Native Americans makes that unlikely, argues Stephen Zegura of the University of Arizona, Tucson, a co-author on Hammer's talk. “A single population that includes all the Y chromosome and mtDNA variants we're seeing would have to be very, very large. It's hard to explain it all with a single migration.” But no matter how many trips, many of the males in the party apparently started with the same peripatetic population in Siberia. Archaeologists have previously noted a potential source culture around Lake Baikal, dated 25,000 to 20,000 years ago: the Mal'ta, a mammoth-hunting people known for blade and biface tools that researchers have speculated might be the precursors of the Clovis points early Americans made 12,000 years ago. Some archaeologists had been skeptical of the link; as Ted Goebel of the University of Nevada, Las Vegas, notes, “there's a huge gap” between those dates, and later Siberian technologies don't look like anything in the New World. That puzzle remains, says Goebel, but he is quick to add that the new genetic data will spur archaeologists like himself to focus even more intently on the Baikal region. Says Goebel: “For sure the answers, yea or nay, lie somewhere up there on the mammoth steppe.” 11. SCIENTIFIC COMMUNITY # Mathematics Gets Institutionalized--Again 1. Barry Cipra The NSF is expanding its program of mathematics institutes to bring visiting researchers from second-tier universities into the mainstream Mathematics is often a solitary pursuit, but it's showing signs of becoming markedly more social. Since 1982 the National Science Foundation (NSF) has funded two mathematics research institutes, where mathematicians from different institutions work in collaboration, often with scientists from other disciplines. Now NSF is set to expand the effort. This month, the NSF announced plans to fund three institutes, the winners of a competition that drew between 10 and 20 entries. Two of the winners are the existing institutes at the University of California, Berkeley, and the University of Minnesota. The third is a new institute at the University of California, Los Angeles (UCLA). The proposed grants total approximately$8 million per year for 5 years, roughly 8% of the NSF budget for the mathematical sciences and a \$2.5 million annual increase in spending.

NSF sees the institutes as a way to help research mathematicians at “second-tier” universities stay in touch with the mainstream, says Donald Lewis, program director for the Division of Mathematical Sciences at NSF. He points out that few of these mathematicians receive any NSF support: “Institutes and conference centers, if sufficient in number, would give nonfunded researchers an opportunity to keep abreast of the latest developments,” he says. NSF also wants to create links between mathematics and other disciplines, a primary goal of the new UCLA institute.

Mathematicians applaud the move. “I think the institutes are an excellent idea,” says William Jaco, a mathematician at Oklahoma State University in Stillwater and former executive director of the American Mathematical Society. “Having venues for this type of long-term collaboration is as valuable for mathematicians as a laboratory is for laboratory scientists.”

The two existing institutes, the Berkeley-based Mathematical Sciences Research Institute (MSRI) and the Institute for Mathematics and Its Applications (IMA) at the University of Minnesota, opened shop in 1982. Each hosts upward of 100 visitors at any given time, including students and postdocs as well as scientists from other fields. The institutes run semester- or yearlong programs for collaborative research and teaching on broad topics, with shorter sessions on specific subjects. Slated for 1999–2000 at MSRI, for example, is a yearlong program on noncommutative algebra (a branch of mathematics that is especially important, for example, in quantum mechanics) and semester- is especially important, for example, in quantum mechanics) and semester-length programs on inverse Galois problems and numerical methods.

Perched on a mountainside east of the Berkeley campus, overlooking the San Francisco Bay, MSRI was long viewed as a bastion of pure mathematics but has adopted a more interdisciplinary stance in recent years. It now has a network of corporate affiliates, including Hewlett-Packard and Pfizer Corp. and has held workshops on topics ranging from materials science to mathematics and the Human Genome Project. It has also begun sponsoring joint postdocs with some of the corporate affiliates. Says MSRI deputy director Hugo Rossi, “We really want to represent mathematics in the broadest possible way.”

The focus at IMA, as its name implies, has always been on applications, notes IMA director Willard Miller: “We've helped a lot in showing people the importance of mathematics in other fields.” The current yearlong program has been in the area of mathematical biology. On tap for 2000–01 is mathematics in multimedia, including speech recognition and natural language modeling, computer security and privacy issues, and geometric design for three-dimensional graphics. The institute also plans to set up “industrial consortia” to work on mathematical aspects of specific problems from industry, such as optical devices. Miller says he hopes these consortia will live on as online collaborations after the IMA program is over: “Once they're set up, we will bid them bon voyage and set up some more.”

The refunding of MSRI and IMA surprised no one, although the competition was fierce—“It was not just a pro forma application,” remarks Miller. The two institutes' approach has generally been perceived as a huge success at creating collaborations among mathematicians and introducing them to other disciplines. Indeed, “it's a model that the rest of the world is copying,” observes Gil Strang, a mathematician at the Massachusetts Institute of Technology who will serve on the advisory board of the new UCLA institute, pointing to a proliferation of mathematics institutes in locations from England to Singapore.

Lewis notes that the United States still lags Western Europe, which now has 10 institutes for roughly the same number of mathematicians, and even Canada, which has three institutes for a mathematical community 1/6 the size of the U.S. community. The United States doesn't have enough math institutes to do justice to the range of possible topics, says Lewis: “It's a strong argument why we should have far more institutes than we now have.”

The new kid on the block, the Institute for Pure and Applied Mathematics at UCLA, won't open until the fall of 2000, but co-director Eitan Tadmor promises IPAM will be even more extensively interdisciplinary than MSRI or IMA. IPAM will aim at a roughly 50:50 mix of mathematicians and scientists from other fields, with three major programs each year. “The basic objective is to encourage cooperation between mathematics and other scientific disciplines,” Tadmor says.

That thinking was crucial to IPAM's successful bid. “We have an enormous need right now for institutes that bridge mathematics to the other sciences,” notes Lewis. “I think our idea was fresh enough,” adds Tadmor, explaining that IPAM will aim to draw in scientists who don't already collaborate with mathematicians. “We wanted to have this interdisciplinary interaction, and we wanted to put together not the usual suspects.”

No programs have been scheduled as yet, but among the topics mentioned in IPAM's proposal to the NSF are programs in computational chemistry and geometric-based motion (which covers ground from crystal growth to image processing). IPAM also envisions holding “reunion” conferences for participants 1 and 2 years after their program. “We realize that to make real contributions in math and sciences takes time,” says Tony Chan, chair of the UCLA math department (and a principal investigator on the IPAM proposal). “Our role is to facilitate that” by helping potential collaborators stay in touch. He thinks the setting will help: IPAM plans to use UCLA's Lake Arrowhead conference center, a lakeside resort in the San Bernardino mountains.

12. SCIENTIFIC COMMUNITY

# Private Money Adds Two Institutes

1. Barry Cipra

The National Science Foundation isn't alone in opening its wallet to support new mathematics institutes. Two new institutes are also getting under way with private funding. The American Institute of Mathematics (AIM) has begun operations in Palo Alto, California, and the Clay Mathematics Institute (CMI) is gearing up in Cambridge, Massachusetts.

The institutes are the brainchildren of two wealthy businessmen: John Fry, who owns a chain of electronics stores in California, and Landon Clay, former chair of the board of Eaton Vance Corp. a Boston-based mutual fund management firm. AIM's director, Brian Conrey, who is on extended leave from Oklahoma State University in Stillwater, says Fry has spent nearly a million dollars in the last 3 years to get AIM up and running. The institute currently operates out of office space in downtown Palo Alto, next to one of Fry's stores. There are ambitious plans to open a conference center in Morgan Hills, with an extensive mathematics library that would include a large collection of rare manuscripts that Fry plans to acquire.

For now, AIM is funding a number of “high-level collaborations on focused projects” in pure mathematics, Conrey explains, targeting big puzzles that take concentrated brainpower. The institute has brought together a group of researchers to work on mathematics related to the Riemann hypothesis, an important conjecture in number theory that has surprising connections with quantum chaos (Science, 20 December 1996, p. 2014). Off site, it is funding a group at Princeton University working on the three-dimensional Euler equation, which stems from the study of fluid dynamics. Another group, based at Oklahoma State, is tackling a problem in topology known as the Lopez conjecture.

CMI's goal will be to encourage creative and original mathematical thinking, says director Arthur Jaffe, a mathematician at Harvard University, where he holds a chair endowed by Clay. Clay, who says the endowment will be in the “eight figures,” says, “We're seeking to support individuals of promise.”

CMI has no plans for a building, Jaffe notes. (It currently runs from the attic of his house.) Instead, it will operate more along the lines of a foundation, supporting research through grants. CMI has a number of projects in mind, including a joint project with the American Mathematical Society to support mathematicians at the Independent Moscow University in Russia. However, Jaffe says, “it's a new organization, so our plans are constantly evolving.”

13. ASTRONOMY

# Labs Hold the Key to the 21-Micrometer Mystery

1. Alexander Hellemans*
1. Alexander Hellemans is a Writer in Naples, Italy.

Some substance not usually associated with stars is glowing around 12 red giants. Researchers hope to find it at the lab bench

For more than a decade, a strange infrared glow coming from certain red giant stars has perplexed astronomers. Centered on a wavelength of 21 micrometers, the emission forms a wide band in the infrared spectrum, which implies that it comes from a large complex molecule or a solid and not from the atoms or simple molecules normally found around stars. Now an intense e-mail debate is raging over the nature of the source, with researchers proposing substances never before detected in space, including polymers, ball-shaped fullerenes, and even nanodiamonds.

The debate was touched off when the first detailed analysis of the feature, based on observations by the European Space Agency's Infrared Space Observatory (ISO), was published earlier this year. And this has spurred on astronomers who are working with infrared spectroscopists to try to find a substance that produces a matching spectrum. “It is a mystery,” says Sun Kwok of the University of Calgary in Canada.

Kwok and his colleague Kevin Volk, along with Bruce Hrivnak of Valparaiso University in Indiana, announced the first four stars exhibiting this feature in 1988 using the Infrared Astronomical Satellite (IRAS). But the low resolution of IRAS's spectrometers made it hard to discern details of the emission feature. After the launch of ISO and its high-resolution spectrometers in 1995, however, the number of stars showing the feature grew to 12, and their nature became clearer. “The feature can only be observed in a very precise evolutionary stage, in the short transition between the red giant phase and the planetary nebula stage,” says Pedro Garcia-Lario of the ISO Science Operations Centre near Madrid. Stars at that stage of their life smoke like an old lantern, blowing off dust that is rich in carbon compounds.

In March of this year, Garcia-Lario and his colleagues published a detailed study of the infrared spectrum of one of these stars, IRAS 16594-4656, in the Astrophysical Journal. This report, along with another high-resolution study of several similar stars published in this week's Astrophysical Journal by Kwok and his colleagues, has sparked a host of theories on the identity of this mysterious compound. “The feature is so strong that it has to come from a common element,” says Kwok, and in such stars “carbon and hydrogen are the obvious ingredients.” The ISO observations have shown that the profile of the emission feature—identical for all 12 stars—is almost 4 micrometers wide, ruling out small molecules as the source because they produce narrow emission lines. “It has to be either a mixture of similar kinds of molecules, like hydrogenated fullerenes, or a very large molecule complex, or a polymer, or even a solid,” says Kwok.

Now astronomers, spectroscopists, and theorists are joining forces to find a substance that produces the same infrared spectrum. And, as in all the best detective stories, suspects abound. Theoretical physicist Adrian Webster of the University of Edinburgh in the U.K. has proposed that the mysterious compound may be hydrogenated fullerenes. The basic fullerene consists of 60 carbon atoms arranged in a sphere, and anywhere from one to 60 hydrogen atoms can festoon it. “I calculated the general spectrum of a mixture of hydrites, and it turned out to be a broad feature centered on 21 microns,” Webster says. Garcia-Lario says that fullerenes could well be produced by the decomposition of large hydrogenated amorphous carbon grains, together with so-called polycyclic aromatic hydrocarbons, ring-shaped molecules known to exist in red giant atmospheres.

However, some astronomers have put forward another, equally exotic candidate: nanodiamonds. Scientists had thought that diamonds could only form on the solid surface of a planet or other body, not in the atmosphere of a star. But Jo Nuth of Goddard Space Flight Center in Greenbelt, Maryland, cites work over the past few years by materials scientists Rustum Roy of Pennsylvania State University in University Park, Andrew Phelps of the University of Dayton in Ohio, and others, which showed that nanodiamonds could form straight from a carbon vapor without a substrate. “This broke the taboo that says that vapor deposition processes weren't directly applicable to circumstellar shells,” says Nuth.

This proposal is supported by Hugh Hill, also at Goddard, who along with researchers from France studied nitrogen-doped nanodiamonds 1 to 3 nanometers in size, which they had extracted from a meteorite. When they studied the diamonds in the lab, they detected infrared emission at 21 micrometers. “We found a fascinating resemblance,” says Hill. “We feel that the 21-micron feature represents the best evidence so far that nanodiamonds can be detected in the interstellar medium.” However, Thomas Henning of the University of Jena in Germany is not convinced. The 21-micrometer feature in laboratory diamonds is too weak, he says. “You would have to have large amounts of nanodiamonds doped with nitrogen in these [circumstellar] environments” to produce the observed spectra, he says. Henning also believes that the technique used to extract the diamonds from the meteorites may contribute to the 21-micrometer feature.

Henning has his own candidate: silicon disulfide. “In silicon disulfide, there is a clear 21-micron feature which fits relatively well,” he says, but he admits that, just as with nanodiamond, explaining the strength of the observed emission would require a lot more silicon disulfide than is usually found around stars. Nuth also points out that silicon disulfide has other emission features, for example near 10 micrometers, that are not observed in the 12 stars.

ISO's useful life ended in April last year, and no infrared observatory has yet been launched that could provide more detail about the 21-micrometer feature. So researchers are pushing forward with laboratory studies, testing candidate sources, and scrutinizing the ISO data for more clues: other emission lines in the infrared spectrum that are only found in the 12 stars. The quest won't be easy, says Kwok. “We are most likely looking at something we don't even have on Earth.”