# News this Week

Science  31 Jul 2009:
Vol. 325, Issue 5940, pp. 522
1. Particle Physics

# More Bad Connections May Limit LHC Energy or Delay Restart

MEYRIN, SWITZERLAND—Last September, the world's highest-energy particle smasher—the Large Hadron Collider, or LHC—mangled itself when a splice in a superconducting electrical line melted and set off a chain reaction of mechanical failures (Science, 26 September 2008, p. 1753). Since then, physicists here at the European particle physics laboratory, CERN, have installed exquisitely sensitive warning systems to monitor the delicate splices and head off a similar catastrophe (Science, 12 December 2008, p. 1620). Now, however, researchers have found flaws in different electrical connections that will limit the LHC's energy.

The bad connections will restrict the current that can course safely through the 1232 superconducting magnets that guide protons around the 27-kilometer subterranean ring. That restriction will limit the energy that the countercirculating bunches of protons will reach before they smash together in the hearts of four enormous particle detectors spaced around the ring. As Science went to press, researchers were testing the last of the 10,000 connections and planned to evaluate the data within a week, says Stephen Myers, director of accelerators and technology at CERN. “We will know the safe current we can run at without fixing anything else,” Myers says. “And that will tell us the maximum energy we can run at” for the first year.

If the problem is too bad, though, it could delay the restart of the gargantuan accelerator, which is now slated for November. Officials had hoped to power up in September, but that date has slid as workers patch a couple of leaks in the LHC's cryogenics system.

The faulty connections mark another bump in the road for the $5.5 billion collider, which was once supposed to turn on in 2005 but circulated its first particles only last year. The LHC aims to blast massive new particles into existence that could light the way to a deeper understanding of the universe. The current highest-energy collisions are produced by the Tevatron collider at the Fermi National Accelerator Laboratory in Batavia, Illinois, which fires protons into antiprotons at energies of 2 teraelectron-volts (TeV). The LHC is designed to reach 14 TeV, although CERN officials had already set 10 TeV as the maximum energy for this coming year. The new problems could limit the energy to 8 TeV or less. The new problems lie literally right next to the old ones. In last year's mishap, a soldered splice between two ribbonlike superconductors melted. The rupture caused the 9000 amps of current flowing through the line to “arc” to other nearby metal parts within the machine. In an instant, that lightning strike burned through the surrounding tube that kept the superconducting line bathed in frigid liquid helium. Boiling liquid and gas flooded the magnets' sealed casings, whose emergency relief valves were not designed to cope with such a deluge. As a result, a pressure wave shot through the machine, damaging 53 magnets and tearing some of the devices, which weigh up to 35 metric tons, from their moorings. The new worry involves not the superconducting splices but the ordinary copper that surrounds them. Within a magnet, the ribbonlike superconductor runs within a thick copper cladding. At the junction of two magnets, the ribbons lie exposed so that they can be soldered together, one on top of the other like the slices of bread in a grilled-cheese sandwich. The splice is then encased in a sleeve of “copper stabilizer,” which must be soldered to the cladding at either end of the junction. Since May, researchers have found that some of these copper-to-copper connections are not up to standards. The problem is not quite as serious as a flaw in the superconducting splices themselves. Ordinarily, all the current flows through the superconducting ribbon. It would flow through the copper stabilizer only if the superconducting wire got hot enough to “quench” and lose its ability to carry current without resistance. If a stabilizer were faulty, however, the current would continue to flow through the splice, which might melt like an overloaded fuse. The iffy stabilizers are also far more numerous than the bad splices. After the 19 September accident, CERN researchers tested most of the splices and found only two more faulty ones. In contrast, they've found about 80 bad copper stabilizers, including 46 in one octant of the machine alone, Myers says. Once accelerator physicists know how high in energy they can safely go, they will present that finding to the experimenters working with the particle detectors and ask them how to proceed, Myers says. If the energy is too low—significantly below 8 TeV—the experimenters may opt to delay the restart until the stabilizers can be fixed. But Myers says that between 8 and 10 TeV is likely and that experimenters probably will opt to run with that. “Both the machine people and the experimenters, we need to be running … to see what other problems might be out there waiting for us,” he says. That position resonates with experimenters working on the particle detectors. “It would still be very useful” to run at a lower energy, says Pauline Gagnon of Indiana University, Bloomington, who works on the ATLAS detector. “It will give us all the information that we need to calibrate our detector, and we will learn a lot about the LHC.” Even so, she adds, “nobody wants to run at 4 TeV for a year.” Meanwhile, physicists at the lab are privately debating whether the problem constitutes a design error. Similar connections at other machines are secured with clamps, and in retrospect, the LHC probably should have employed a similar design, Myers says. But Lyndon Evans, the CERN engineer who oversaw the construction of the LHC, notes that in 1999 the current design passed an external review and that the use of clamps was discouraged as it would have impeded the flow of the surrounding liquid helium. Either way, some physicists grumble that the problem should have been caught earlier. 2. Europe # Fix Funding Agency's 'Original Sin,' ERC Review Panel Demands 1. Martin Enserink Timidity is not a trait that psychologist and former Latvian president Vaira Vīķe-Freiberga, nicknamed the Baltic Iron Lady, is known for. So when she was asked to chair a review of the European Research Council (ERC), the 2-year-old funding agency that many believe will be central to the competitive future of European science, no one expected her to come up with another serving of bland Brussels policy-speak. Her report,* presented on 23 July to Janez Potočnik, the European Commissioner for Research, didn't disappoint. Despite a successful start—the ERC's grant calls have been flooded with applications, and scientists praise its focus on excellence as the sole criterion for picking proposals—the agency's structure is “a source of great frustration and ongoing low-level conflict” among its managers, opined Vīķe-Freiberga's panel. ERC's “overregulation, over-control, and over-steering” of grantees has “tremendous detrimental effects.” Scientists who have volunteered time to review grants are “surprised, not to say shocked,” by ERC's procedures. “It's written in a rather unusual style, isn't it,” says John Smith, deputy secretary general of the European University Association (EUA). “It's refreshing.” The review panel, which included former U.S. National Institutes of Health chief Elias Zerhouni, focused on the key question about ERC: Should scientists or bureaucrats run the show? With a € 7.5 billion budget for the first 7 years, ERC funds frontier research irrespective of national boundaries, a rarity in the European Union, where funding is typically subject to political and economical considerations. To preserve that freedom, early advocates of such a U.S.-style grant agency argued that ERC needed to be independently run by scientists, with a minimum of bureaucracy and at a safe distance from the European Commission. But when ERC got off the ground in early 2007, they got only half of what they wanted. ERC's scientific strategy is set by an independent Scientific Council, chaired by molecular biologist Fotis Kafatos of Imperial College London. Yet the management is in the hands of an Executive Agency, a body that is formally autonomous but is, for all practical purposes, controlled by the commission. The Scientific Council has a permanent representative in the Executive Agency, called the secretary-general. But the position, which Spanish economist Andreu Mas-Colell took over on 1 July, comes with no formal power. The review panel concludes that this “original sin” is at the source of many problems. The commission's recruiting system makes it hard to attract scientists to the agency, for instance, and E.U. rules designed to prevent fraud and waste clash with academic traditions. Grantees have to sign contracts that can keep them from switching research directions and force them to account for expenses or even keep time sheets; basic researchers need trust and flexibility, the panel says. Strict rules are also threatening some scientists' willingness to review grant applications. They need to mail in a copy of their passport, which the panel calls “completely abusive”; travel expenses paid out of pocket can take a long time to be reimbursed. “I review for a large number of international funding bodies, and this was the worst experience I have had in 30 years,” one scientist told the panel. “I would not agree to review again.” Unlike ERC's Scientific Council, which wants to wrestle control of the agency from the commission, Vīķe-Freiberga's panel says that ERC should first try to fix its problems within the current structure. It recommends a range of measures, including merging the roles of secretary-general and director of the Executive Agency into a single post to be filled by a scientist. “It should not be acceptable today in Europe that non-scientists […] run major European research programmes!” the panel exclaims. (On occasion, it hits the Caps Lock button to hammer home the message, for instance when it demands “a true PROFESSIONALIZATION both at the scientific and managerial level.”) Going the independent route, which Article 171 of the European Community Treaty makes possible, should be done only if a new review in 2 years shows that the many “irritants” endure, the panel says. It's a recommendation that both sides can live with. “It's a very fair and balanced report, and the recommendations are wise,” says Helga Nowotny, the Austrian vice-president of the Scientific Council. “It's a good, honest report that will help all of us a lot,” says Potočnik. Its strong language notwithstanding, the panel says ERC has done many things right. The agency was able to attract good reviewers—at least so far—and interest in its first grant calls was overwhelming. The panel's own survey also showed that most applicants and reviewers were happy, says Potočnik, who gets kudos himself for shielding ERC from political interference. An EUA consultation has shown that universities like the new agency, says Smith. ERC has not done as well at attracting non-European research talents, who are eligible for the funds if they move to Europe, notes Wilhelm Krull, director of the Volkswagen Foundation, a German grantmaking agency, but that may improve as ERC becomes better known. Nor is it giving young researchers the best shot at getting grants, says Mas-Colell. ERC Starter Grants are available for researchers with less than 10 years of experience after their Ph.D., but most of them have gone to scientists at the upper end of that spectrum. In the future, the Scientific Council wants to reserve a portion of the grants for true starters with less than 5 or 6 years of experience, Mas-Colell says. As ERC now knows all too well, getting a good start in life can be difficult. 3. ScienceNOW.org # From Science's Online Daily News Site Catching a Giant Wave Like a shadow of death racing across the ocean surface, a tsunami churns the air and darkens the water along its path. Researchers now report that they can detect this activity with radar, improving the chances that coastal populations can escape these killer waves. Can't Decide? Ask an Ant Classical philosophers called humans “the rational animal.” Clearly, they never looked closely at ants. A new study suggests that ant colonies avoid irrational decisions that people and other animals often make. Tree in a Time Warp A eucalyptus-like tree that grows in New Zealand is still defending itself from a giant bird that died out about 500 years ago. The lancewood tree changes its appearance twice in its lifetime—an adaptation, a new study suggests, that prevented it from being eaten by flightless moas. A Cloaking Device for Earthquakes Researchers say they have found a way to make buildings essentially invisible to earthquakes. If perfected, the technique could protect skyscrapers and homes alike from even the most devastating temblors. Microbe Evolution Gets a Push Researchers report that they've come up with a new way to modify the genomes of billions of microbes simultaneously and then finger the ones with the most interesting changes. Because the technique will likely work with most types of genomes, it could turbocharge efforts to engineer microbes to produce everything from novel therapeutic drugs to vast quantities of biofuels. Pump Up the Volume Anyone who has woken to a cacophony of squawks and chirps knows that birdsong, no matter how melodious, isn't always a welcome sound. Past research suggests that birds aren't keen on human din either. But a new study finds that not all birds think alike: Some species actually appear to seek out noisy environments. Read the full postings, comments, and more on sciencenow.sciencemag.org. 4. Restoration Ecology # Oysters Booming on New Reefs, But Can They Survive Disease? 1. Erik Stokstad Oysters were once so abundant in the Chesapeake Bay that they posed a navigation hazard. Perched on reefs made up of old shells, the filter-feeders had a sturdy spot to attach to and could avoid being covered by silt. But by the early 20th century, the reefs were nearly picked clean by humans, and the remaining oysters were ravaged by introduced diseases. Without a large population of oysters to keep them growing upward, the reefs sank and became topped with muck. Over the past 2 decades, restoration efforts to create new habitat have ramped up, but progress has been slow and patchy. Now a group of researchers is claiming an “unprecedented restoration” in the Great Wicomico River, a small tributary. Thanks to a large network of reefs created in 2004, a booming population of native oysters has taken hold, the group reports in a paper published online by Science (www.sciencemag.org/cgi/content/abstract/1176516) this week. Some researchers herald the accomplishment as a model for restoration, but others say that the new populations have yet to prove they can survive nowendemic diseases. “It's very early in the game to call this a success,” says Paula Jasinski of the National Oceanic and Atmospheric Administration's Chesapeake Bay Office. Most efforts to restore native oysters (Crassostrea virginica) in the bay have been relatively small, 0.5 hectares or so, largely due to the expense of building reefs. To create new habitat, state natural resource agencies and other partners have dredged up old oyster shells and scattered them on river bottoms. These thin layers tend to sink and be covered with mud within 3 to 5 years, degrading the habitat. David Schulte wanted to create something bigger and taller than the usual patches. A graduate student in marine biology at the Virginia Institute of Marine Science (VIMS) in Gloucester Point, Schulte also works for the U.S. Army Corps of Engineers. He helped designed the roughly$2 million project to build 35 hectares of subtidal reefs about 130 kilometers southeast of Washington, D.C. The Great Wicomico River is a good spot for restoration, in part because surviving oysters in it produce copious larvae.

In 2004, the corps created two kinds of reefs, one taller than the other. Researchers have built these so-called high-relief reefs elsewhere but not as expansively. Larvae from upstream settled on the new habitat, and within a year, young oysters were abundant. On average, they were five times more dense—up to 1000 oysters per square meter—on the taller reefs than on the shorter ones. The main reason, Schulte says: The extra height meant the reefs had less sediment and more exposed shell for larvae to settle on.

By March 2009, the estimated 185 million oysters constituted the largest known restored population of native oysters in the world, the authors say. And because there are already three generations of oysters present, the population is likely sustainable, says co-author Romuald Lipcius, an ecologist at VIMS.

With all the new oysters, the reefs themselves are accumulating shell debris, suggesting that they will grow in height on their own, as natural reefs used to. “It's very good news, especially for Chesapeake Bay,” says Sean Powers, a fisheries scientist at the University of South Alabama in Dauphin Island who studies oyster restoration. “For marine ecology, restoration ecology, it really is an advance.”

Most experts who work in the Chesapeake Bay are more cautious. They point out that the Great Wicomico River is a best-case scenario because of its plentiful larvae and its hydrology, which tends to prevent these larvae from drifting out to the bay. Jasinski says only a few other rivers would be as conducive to building up populations quickly. And so far, the new reefs don't have a large proportion of large oysters, longer than 75 millimeters, which are market-size and also the most fecund.

The biggest question is how well the new population will survive disease. Many oysters in the Chesapeake are afflicted with MSX, a condition caused by the protozoan Haplosporidium nelsoni, which was accidentally introduced in the bay in the 1960s. Young oysters catch the disease and are most likely to die after 3 to 4 years of infection, says James Wesson of the Virginia Marine Resources Commission in Newport News, who has been monitoring the population in the Great Wicomico River. There are already discouraging signs: Some adult oysters on the reefs, which now make up 80% of the new population, began to die last year. Based on past events, Wesson predicts 50% mortality among the adults by November.

Lipcius is more sanguine. “My firm belief is that the whole disease issue has been a red herring,” he says. Although some deaths are to be expected, he thinks that overall, a population of healthy adults living in good habitat will resist disease better than other populations have.

The corps is also building high-relief reefs in the Lynnhaven River, farther south, and planning for a project in another small river, the Piankatank. But good habitat doesn't come cheap. Larger rivers—not to mention the bay proper—will require even bigger reefs stocked with billions of hatchery oysters. Everyone agrees that oysters have a long way to go. But steps like those in the Great Wicomico are promising, says Tommy Leggett of the Chesapeake Bay Foundation, a nonprof it based in Annapolis: “I think we're headed in the right direction.”

5. Botany

# Plant Bar Code Soon to Become Reality

1. Claire Thomas

DNA bar-coding, the ambitious idea of using a short piece of DNA to tell every species in the world apart, is already a powerful tool for scientists studying animals. But for plant biologists, the idea has for the most part remained a pipe dream, stalling systematic studies of plants and efforts to conserve flora.

So far, every plant DNA sequence proposed as a marker has had problems, and the search for the “right” plant bar code has proven controversial (Science, 12 October 2007, p. 190). But at last, a solution appears within reach. A paper published this week in the Proceedings of the National Academy of Sciences (PNAS) proposes two genetic sequences, or loci, taken from chloroplast genes called matK and rbcL, as the official plant bar code. Although the new bar code works for some kinds of plants better than others, it identified 72% of all species on average and grouped 100% of plants into the correct genus.

The results comprise data from 25 major bar-coding labs, all part of the Plant Working Group (PWG), an organization set up in 2005 to help resolve the plant problem. It's now up to the Consortium for the Barcode of Life (CBOL) and a board of experts to endorse or reject the plant ID strategy within the next 2 months, says Peter Hollingsworth, chair of PWG. “I have a very strong hope that they will approve it, and we have worked very closely with CBOL throughout this process,” says Hollingsworth, a conservation geneticist at the Royal Botanic Garden Edinburgh.

Plant bar-coding centers are eager for that approval. They could then finally “tag” wide varieties of plant species and share that information online through the Barcode of Life Data Systems Web site. In fact, many optimistic researchers have begun bar-coding using matK and rbcL, after the two loci were informally proposed at a PWG meeting in September 2008.

But PWG admits that the two-locus bar code is far from perfect, and a minority within the group believes the search is not over. W. John Kress, a co-author of the PNAS paper, endorses the proposed bar code, but he says his lab will continue to add another DNA sequence called trnH-psbA to matK and rbcL. From these three, the best two-locus bar code would then be selected at a later date. “We're finding that we still get some pretty good results with the three-locus bar code, and personally I'm not ready to give that up yet,” says Kress, a botanist at the Smithsonian Institution in Washington, D.C.

The PNAS paper reports on seven candidate loci, including sections of genes and spacers (regions between genes), taken from work led by Kress; Ki-Joong Kim, a botanist at Korea University in Seoul; and Mark Chase at the Royal Botanic Gardens, Kew, in the United Kingdom. Each sequence was assessed on its ability to tell species apart and on factors such as its ability to be read rapidly by automated sequencers, says Robyn Cowan, a co-author and conservation geneticist at the Royal Botanic Gardens, Kew. Ultimately, four were ruled out, for reasons such as low discrimination rates or because the sequences were hard to read efficiently.

Of the remaining three, no individual sequence came up to scratch, so the loci were combined to bump up bar-code quality. The rbcL gene provides good-quality sequences and has effective primers (reagents used to “probe” or pick out the sequences) that work across a variety of plants, making it a good choice. Its ability to discriminate species hovers above the average at a rate of 61%. Both matK and trnH-psbA were better at telling species apart, with success rates of 66% and 69%, respectively. But trnH-psbA produced poorer quality sequences due to the presence of long mononucleotide repeats, in which one DNA base recurs excessively within the spacer. Such repeats can cause the sequence to be misread. On the other hand, matK doesn't have primers that allow it to be isolated in all plant groups. Although there is no immediate way to resolve the issues with trnH-psbA, the group expects the quality of matk primers to improve over time. Already, the current primers work well in angiosperms (flowering plants), being able to latch onto the loci in 90% of species, but are less successful with gymnosperms (cone-producing plants) at 83% and extremely poor with cryptogams (lower plants such as mosses and ferns) at 10%.

The proposed plant bar code remains less powerful than its animal counterparts, a gene called CO1, which averages upward of 95% discrimination. But PWG hopes to push up species discriminating power in the near future. One way of doing this is by using further supplementary bar codes in certain plant groups. For example, in protea, a group of flowering plants, botanists will use the ITS ribosomal spacer, says Chase. (See Perspective on challenges in plant bar-coding by Chase and Fay.)

Nevertheless, the proposed bar code provides a good “starting point and a reference for all subsequent plant DNA bar-coding work,” says Andrew Lowe, a plant conservation biologist at the University of Adelaide in Australia who was not involved with the paper. As the cost of sequencing drops, using several genetic loci for bar-coding will become a more manageable exercise, he adds.

If the bar code is approved, it will open the way for a variety of plant tagging projects, including large-scale systematic projects such as Tree-BOL, an effort to bar code all trees, and the international grass bar-coding project. While Tree-BOL has been collecting rbcL and matK bar-code loci for some time, the formal approval of this bar code would make funding applications and publishing results easier, says Damon Little, chair of Tree-BOL and a plant biologist at the New York Botanical Garden in New York City.

For now, the plant community eagerly awaits CBOL's decision. “We [could] at long last move forward with sequencing” on a larger scale, says Little.

6. Science Education

# Universities Begin to Rethink First-Year Biology Courses

1. Jeffrey Mervis

Introductory biology courses are often the last academic exposure nonscience majors at U.S. colleges have to science. Unfortunately, say science educators, the courses too often leave a bad taste in the mouths of students who spend more time in lectures than on experiential learning and in regurgitating facts rather than understanding the concepts behind them. As voters, those graduates apply their misconceptions of science to shape national policies on everything from evolution to stem cell research. So improving introductory biology is seen as a critical step toward raising the nation's scientific literacy.

In 2006, the National Science Foundation (NSF) began an initiative to do just that. This month, 500 researchers, educators, and policymakers met in Washington, D.C., at a conference sponsored by NSF and AAAS (which publishes Science) to assess how far they have come. The consensus: There's still a long way to go.

Correcting the problem, speakers agreed, will require changing a deep-rooted academic culture that values research over teaching and makes little provision for doing a good job in the classroom. “We hire new faculty with big start-up packages and expect them to set up their labs and get going on their research,” explained Joan Lorden, provost of the University of North Carolina, Charlotte. Then we say, ‘Here's your course load. And by the way, we'd like you to be an innovative instructor.’ But rarely do we give them the support they need to succeed.”

NSF's education directorate funds a slew of programs aimed at improving undergraduate instruction across disciplines, and in recent years the agency's biology directorate has begun to tap into those efforts. At the conference, biology chief James Collins described his plans to ramp up support for improving undergraduate biology. Assuming congressional approval of NSF's 2010 budget, Collins hopes to spend an additional $10 million next year on a variety of activities, including mentoring, curriculum development, and research experiences for students as well as faculty development. This month, for example, NSF announced a new competition for$50,000 “incubator” grants that would allow researchers and educators to work up a full-fledged proposal to carry out such improvements. The biology directorate is also hoping to join the education directorate in funding a university-based center to improve undergraduate retention and graduation rates among science majors, with special attention to underrepresented minorities.

Participants said they hoped to use the conference to shore up support for reforms already under way on their campuses. But they were realistic about the availability of resources. “I heard about one terrific program that could be a model for us. But it costs $5 million over 5 years,” notes Eric Stabb of the University of Georgia, Athens. “And I'm sure my dean doesn't have that kind of money lying around. But now we can use the conference to hammer the administration on the need to do more. And if we make a good case, maybe the state legislature can find the money.” Talk is cheap, agrees Collins, reminding participants that the burden is on them to sustain and scale up their reforms: “If anything is going to happen, it's because you will decide to do something.” 7. ScienceInsider # From the Science Policy Blog The U.S. House of Representatives dug into minutiae at the National Institutes of Health last week before passing a$31 billion NIH funding bill. At the behest of Darrell Issa (R-CA), members okayed an amendment to kill three peer-reviewed grants Issa doesn't like. They support efforts to understand the spread of HIV/AIDS by studying risky behavior among prostitutes in Thailand and China and alcoholics in Russia. Issa called them wasteful; biomedical groups protested. But rather than debate the issue, bill manager Representative David Obey (D–WI) accepted Issa's amendment. As with similar grant-killing measures a few years ago, many expect it to be stripped out in negotiations between the Senate and House on a final bill.

A stem cell paper that made international headlines earlier this month has been retracted following charges of plagiarism. The paper, which claimed to demonstrate how sperm could be made from human embryonic stem cells, was published online in Stem Cells and Development on 8 July with Karim Nayernia of Newcastle University in the U.K. as the corresponding author. Questions arose because of two paragraphs in the introduction that apparently were copied without attribution from a review article by another researcher. Graham Parker, editor-in-chief of Stem Cells and Development, told ScienceInsider that he decided to retract the paper on 21 July.

In other news, the U.S. Department of Homeland Security used a scientifically flawed study to justify its selection of Manhattan, Kansas, as the site for the proposed National Bio and Agro-Defense Facility, according to a draft federal report obtained by The Washington Post. And the Australian Stem Cell Centre hopes that a new business plan that shifts the center's emphasis from commercialization to research will help it regain momentum.

For more science policy news, go to blogs.sciencemag.org/scienceinsider/.

8. Science Education

# Reshuffling Graduate Training

1. Jeffrey Mervis

Nobelist Roald Hoffmann believes that taking graduate students off grants and giving them fellowships would be good for U.S. science. But others say such a radical change isn't in the cards.

Dressed in satin and sequins, Roald Hoffman has ridden atop the first science-themed float in Rio's famed Carnival. Once a month, he appears on stage at a New York City café to host a revue of science and the arts. It's all part of what Hoffmann, a 1981 Nobelist for his work on the theory of chemical reactions, calls his “extreme outreach to the community.”

“I think science should be fun,” Hoffmann said in May to the National Science Board, the oversight body for the National Science Foundation (NSF), when it awarded him its prestigious Public Service Medal. But after flashing pictures of himself at Carnival and on stage at the Cornelia Street Café in Greenwich Village, Hoffmann got down to business: “Now I want to shift gears and talk about something serious.”

What Hoffmann wanted to discuss is a proposal for changing how the U.S. government supports the training of graduate students in the sciences. Federal research agencies now funnel most of their money for graduate students through grants to faculty members. That's the case for nearly 90% of the 39,000 graduate students whom NSF supports each year and for about two-thirds of those getting money from the National Institutes of Health (NIH). The remaining students are funded via fellowships, awarded directly to them, or through traineeships, in which universities compete for a grant to support a certain number of students in a particular area for a fixed period of time.

The commingling of education and research has created a system that is the envy of the world in terms of research productivity. It's also not a bad deal for the student, who typically doesn't pay a penny to earn her Ph.D. The university picks up the tuition for her required courses, and her research is funded through a federal grant awarded to her adviser, who then hires her to work in his lab. In return, she'll probably teach some undergraduate classes during her first few semesters, after which her adviser will receive several years of skilled labor at below-market rates.

But that wildly successful system comes at a high cost to both students and the profession, says Hoffmann, who also made his case in an 8 May editorial in The Chronicle of Higher Education. And it's not sustainable, he argues, especially during tough economic times like these. A better approach, says Hoffmann, would be for the government to stop supporting graduate students on research grants—roughly 30% of a typical NSF chemistry grant pays for graduate students, for example—and use the money for competitive fellowships that students could use at the university of their choice.

That seemingly minor shift could have huge consequences for universities and for the entire U.S. research enterprise. Although they admit Hoffmann's proposal faces long odds, some community leaders say that such a change is long overdue and that his suggestion offers a promising road map. “The real power of an individual fellowship is that it empowers a young scientist to act in a more independent manner, on something creative and for which they have a passion,” says Thomas Cech, a Nobelist who recently returned to academia after a decade as head of the Howard Hughes Medical Institute (HHMI) in Chevy Chase, Maryland. “And that's what science is really about.” Under the current system, he says, “a graduate student is told, ‘Do experiment 2a because it's in our grant.’ That turns the student into a pair of hands. So I think a shift to fellowships would be an excellent idea.”

Shirley Tilghman, president of Princeton University and chair of a 1998 National Academies panel that offered advice on career paths in the life sciences, says a move away from supporting graduate students on research grants would also address two other major flaws. Although the current system has succeeded in maximizing the amount of research performed, she says, it has also degraded the quality of graduate training and led to an overproduction of Ph.D.s in some areas. Unhitching training from research grants would be a much-needed form of professional “birth control,” says Tilghman, who favors more federally funded traineeships. (Traineeships are grants awarded to institutions, which in turn promise to provide students with professional and career counseling as well as a chance to develop their scientific skills in specific areas.) Reducing the overall number of graduate students in the life sciences “is a price that I'd be willing to pay,” she says, in return for a better training environment and improved job prospects.

Fellowships already have a strong following. Building on a 2007 proposal from economist Richard Freeman of Harvard University, President Barack Obama has promised to triple by 2013 the annual number of NSF's prestigious Graduate Research Fellowships, which run for 3 years and cover all fields that NSF funds. And another newcomer to Washington, HHMI President Robert Tjian, hopes to revive a graduate fellowship program that the institute terminated in 2003 when money became tight. Tjian sees the program, which would be open to the most talented students from around the world who are studying in the United States, as an important investment in the next generation of academic researchers.

However, other academic leaders worry that Hoffmann's proposal risks killing the goose that laid the golden egg. “Any radical shift away from what we do now is risky because it would jeopardize a strong innovation system,” says Debra Stewart, president of the Council of Graduate Schools in Washington, D.C. Robert Berdahl, president of the Association of American Universities, also thinks that a wholesale shift to fellowships would be unwise because it would take the selection of graduate students out of the hands of investigators. “In effect, by making awards to individual researchers, we are asking faculty members to find the best students,” says Berdahl, a former chancellor of the University of California, Berkeley. “Presumably, there is a correlation between the quality of an individual [scientist] and the quality of the students in his or her lab.”

## A system out of balance

Hoffmann, a professor at Cornell University, says he began to think about the need for changing the current system during a series of recent departmental meetings on coping with the economic downturn. Most of the suggestions from faculty members, he concluded, would erode undergraduate instruction, about which he is passionate. Although Cornell officials say they are still working on a long-term plan, Hoffmann fears that a one-time, 5% cut in the chemistry department's operating budget starting this fall will be extended for 3 years and that the result will be larger classes, fewer instructors, and limits on enrollment in some courses. “We're firing some of our best teachers,” he says. In contrast, he adds, research programs are likely to be unaffected because they are funded by federal dollars that are beyond the university's control.

G. Peter Lepage, a physicist and dean of the College of Arts and Sciences at Cornell, says every university is struggling to educate undergraduates and maintain a strong research program in the face of shrinking endowments, reduced state subsidies, and pressure to hold down tuition increases. Lepage says he doesn't see how Hoffmann's suggestions would help undergraduates, and he worries that they could harm research. “I have to make sure I have enough money to cover our teaching responsibilities [to undergraduates],” he says. “And we'll figure out a way to do that. At the same time, our faculty need graduate students to do their research, and we need to admit enough of them to do the research as well as to teach the courses.”

The current recession could have an impact on undergraduates, Lepage acknowledges. Graduate science students typically spend their first 2 years as teaching assistants while they take courses and explore research options, he says, whereas their final 3 years are devoted to research. “If you have more research money and less money for TAs, maybe we'll have to rejigger that balance and go from four to three semesters of teaching [per graduate student] and from six to seven semesters of research,” he says. “But both missions will get done.”

Hoffmann readily admits that a shift to fellowships, which are now limited to U.S. citizens, would have one major unfortunate consequence: It would drain the graduate pool of most students from China, India, and other nations. Foreign students f ill a majority of the slots in many U.S. graduate programs in the natural sciences and engineering, but few could afford to come on their own dime. Hoffmann says he would regret losing those students but points to a silver lining. Having universities award fewer science Ph.D.s should force employers to pay higher salaries, he predicts, and attract more of the best U.S. students into science.

Tjian's plan would extend a helping hand to foreign students as well. (As a private philanthropy, Hughes doesn't have to answer to the political argument that U.S. tax dollars should be spent on Americans.) But the Hughes program will serve only a tiny fraction of the foreign graduate students now in the country.

Freeman, a labor economist who studies the dynamics of the scientific work force, sides with Cech and Hoffmann when it comes to the value of fellowships. However, Freeman thinks that Hoffmann's all-or-nothing plan ignores both economic and political realities. “We produce two things at our universities: education and science,” says Freeman. “That's what society wants from us. And students will still want to work in a lab.”

Freeman says Hoffmann's suggestions would result in “more expensive science, and that means fewer people doing it. That's not consistent with where most policymakers think we should be headed as a country. … I hate to reject something because it's radically different, but I think he needs to do a better job of modeling [the consequences].”

## Getting the work done

What would fewer graduate students mean for research? Tilghman says that many scientists reacted in horror to the suggestion in her 1998 report that a typical 10-member lab might shed one graduate student as a way to reduce the overproduction of Ph.D.s and improve the quality of their training. “The PIs [principal investigators] told us that the lab's productivity would go way down if they left,” she recalls.

Tilghman is dubious. “I think that's highly debatable, and in any case, it's never been rigorously tested,” she says. “Every scientist knows that graduate students often go through long periods in which they are totally unproductive.”

Barbara Baird, the chair of Hoffmann's department of chemistry and biochemistry, thinks that her colleague is ignoring a fundamental rule of academic research. “Federal agencies hand out money based on what the PI says is needed to do a particular project,” Baird explains. “If graduate students become a less dependable source of labor, then the tendency will be to simply hire postdocs. The work still needs to be done.”

At NIH, the bulk of the training programs are run by the National Institute of General Medical Sciences. Its director, Jeremy Berg, says he shares Hoffmann's concern about maintaining high-quality undergraduate and graduate programs in the face of mounting pressure from faculty members to maintain their research programs. “The biggest driver for the production of Ph.D.s is not the perception that there is an undersupply but rather that there's work that needs to be done,” says Berg. “However, even if they are cheap, I'd argue that students are also smart, committed, and hard-working labor.”

Hoffmann says he assumes that a system of competitive fellowships would widen the already large gap between the elite universities and the rest of the nation's system of higher education, pointing to the fact that the top-20 research universities historically have attracted a disproportionate share of NSF's graduate research fellowships. Increasing that imbalance would bother him, he admits, but not enough to torpedo the idea.

For other scientists, however, that outcome would be a showstopper. “My fear is that you'd be creating a narrower base for the country's research enterprise and lose the geographic diversity of science that now exists,” says Susan Gerbi, a biochemist at Brown University and an authority on training issues in the life sciences. “In addition, a school on the rise might not have access to the graduate students it needs.”

A fellowships-only system, Gerbi says, would also lead to “wild swings in enrollment from one year to the next.” On the other hand, say Gerbi and Tilghman, a shift to traineeships would reward universities that articulate a well-crafted approach to build up the talent pool in a particular area and also provide program stability.

Berg says that striking the right balance between types of graduate support is a perennial issue for NIH and that Hoffmann's proposal is “a blunt instrument for a subtle problem.” One complication, he says, is that federal research grants have become an increasingly important source of support for academic science “not just to support a research program but also to support institutional activities.” That's code for the overhead charges—about 50% to 60%—that universities add to federal grants. In comparison, training grants and fellowships typically have overhead rates of 10% or less.

Is that difference large enough to make fellowships unattractive to most universities? “I'd like to know” what administrators think about that, says Berg.

Cech thinks the different overhead rates do influence how universities view support for graduate students. But he says those reimbursement rates aren't carved in stone. “There's no law that you can only give 10% in indirect costs for a fellowship,” he argues. “You could make it 40%, on the grounds that they provide us with research results as well as training. Of course, that would cost more, so the money for training wouldn't go as far.”

Senior NSF officials actually considered a variation of Hoffmann's proposal several years ago, notes Esin Gulari, dean of science and engineering at Clemson University in South Carolina and a former head of engineering at NSF. The plan would have allowed researchers to request money for a certain number of traineeships as part of their grant application; at the same time, support for graduate students would be excluded from their grant. “But it never went further than that,” says Gulari, now a member of the science board and part of Hoffmann's target audience. “We were so focused on increasing the size of the stipends” for existing fellowships, she says, that the question of shifting the balance between various modes of support was never addressed.

Even those who agree with Hoffmann that changes are needed are not optimistic they will occur. Tilghman says the topic “is not high on the agenda” of most of her fellow university presidents. Instead, she's pinning her hopes on the heads of the various federal research agencies. But bringing about the changes Hoffmann has suggested, she adds, will require them to put the common good above the self-interest of their constituents, namely, individual scientists.

“We need to care most about the health of the overall scientific enterprise,” she says. “If your only perspective is attracting the labor to run your lab, then the status quo works very well.”

9. Ecology

# Saving a Venomous Ghost

1. Claire Thomas

Despite periodic predictions that it will go extinct, the Hispaniolan solenodon endures, and scientists now want to ensure the rare mammal's survival.

In 1907, explorer and naturalist Alpheus Hyatt Verrill went to the small island of Hispaniola in the West Indies to track down the elusive Hispaniolan solenodon (Solenodon paradoxus), one of the rare examples of a venomous mammal. Fellow scientists told Verrill the journey was “hopeless,” saying he was “as likely to secure specimens of ghosts.” However, the explorer managed to capture a pregnant female, which he sent back to the United States for examination by his father, zoologist Addison Emery Verrill. In a subsequent report, the father offered this pessimistic prediction: “Owing to the introduction of the mongoose and other causes this creature has become very rare and local. It is, without doubt, on the verge of extinction.”

Back then, things didn't look good for S. paradoxus, and little has changed 100 years on for this shrewlike animal, as sightings remain rare. Still, the solenodon is a survivor; scientists have found fossils of it tracing back 76 million years.

Now, U.K. scientists are teaming up to help ensure this little-studied creature survives even longer. Starting in October, Richard Young of the Durrell Wildlife Conservation Trust (DWCT) in Bath, U.K., will lead the most extensive solenodon surveys so far in Hispaniola—an island shared by two countries: Haiti and the Dominican Republic—to find out more about these creatures with the hope of coming up with a conservation plan. Young and Samuel Turvey, a paleontologist and conservation biologist at the Zoological Society of London (ZSL), will also examine the animal's genetics and evolutionary history. “It represents an animal which has not changed much since the time of the dinosaurs,” says Turvey.

In appearance, the Hispaniolan solenodon's tapered claws, stocky body, and long nose set it apart from most other modern mammals. It and the equally endangered Cuban solenodon—the landmasses of Cuba and Hispaniola were joined 25 million years ago—are also the only living mammals able to inject venom through grooved teeth, much the way a snake does, says Turvey.

In June and July of 2008, as part of a pilot study on Hispaniola to find out how best to catch or count the animals, Amy Hall, animal registrar of DWCT, set up movement-triggered cameras, looked for feces and hair, and deployed traps designed to ensnare live animals. Hall caught only one, protecting herself from the solenodon's poisonous bite with Kevlar gloves as she took pictures of its limbs and teeth and measured it. “It was fighting and letting out the most remarkable noise—a very loud, high-pitched cry—the loudest thing I've ever heard come out of an animal,” she recalls.

Before humans got to the island, Hispaniola was teeming with wildlife, with some 25 species of land mammals, including sloths and monkeys. But almost all of these have been driven to extinction, save the hardy solenodon and the hutia, a large-bodied rodent. “The region has experienced the world's highest level of mammal species extinctions, during both the historical era and the Holocene as a whole,” says Turvey, who has been studying the island's fossil record. How the solenodon survived is unclear, he notes, but its varied diet and ability to cope in a range of habitats likely helped.

That flexibility may have been key, as once Amerindians arrived on Hispaniola about 6000 years ago, mammal habitats began to disappear, particularly as agriculture increased and human populations rose. The island's subsequent colonization by the Spanish in the late 1400s meant deforestation and the introduction of cats, dogs, and mongooses, all solenodon predators.

It wasn't until 1833 that the solenodon was first described in scientific literature, by German zoologist Johann Friedrich von Brandt. Even then, it was rare. Another specimen was not reported until Verrill captured his lone specimen more than 70 years later.

In an April 2007 survey in Haiti, Turvey led a team that spent 11 days looking for signs of the creatures and interviewing locals: The remains of three dead solenodons were brought to the team, one of which had been partly eaten by subsistence farmers. That evidence, plus Hall's catch of a live animal in 2008, was enough to persuade the U.K.'s Department for Environment, Food and Rural Affairs to allocate £223,341 (roughly $369,000) from its Darwin Initiative for a 3-year study of solenodons and hutias. Starting this October, Young, Turvey, and others will investigate the abundance and distribution of solenodons across the island and evaluate what threatens the species. Local organizations, such as the Ornithological Society of Hispaniola, will help with field-work and raising public awareness. The ultimate goal is to develop a broad, long-term conservation plan for the solenodon. Alongside this work, DWCT and ZSL teams are examining the DNA of 10 to 20 specimens, many provided by local partners, to figure out the genetic variation among various isolated solenodon populations across the island. Preliminary analyses suggest there may be up to three branches of solenodon species on Hispaniola, says Young. This would fit with earlier observations in 2001 by biologist Jose Ottenwalder, who described a distinct southern subspecies, Solenodon paradoxus woodi, which was markedly smaller than its northern counterparts. Turvey hopes that in the end, studies of solenodons will provide insight into what primitive placental mammals were like. That's why he and other researchers believe saving the unusual species is so vital. “It's not really like anything else,” says Turvey. 10. Astronomy # A Quest for Cosmic Karma 1. Yudhijit Bhattacharjee Inexplicably hot bubbles in space have put researchers on the trail of feedback loops that slap weight limits on galaxies and stifle the birth of stars. In the heart of a galaxy, a violent cosmic drama is unfolding. Streams of gas fall into a massive black hole at the galactic center like wisps of smoke being swallowed by a vent. As the black hole feeds, the core of the galaxy shines with the brilliance of many million suns—what astronomers call an active galactic nucleus (AGN). Then the black hole shoots out two opposing jets of particles at nearly the speed of light. The shock from the jets drives gas away from the black hole, forming gigantic bubbles that send sound waves rippling through the galaxy's surroundings for millions of years. The process heats the gas farther out, preventing it from cooling and collapsing to form new stars. It also stops more gas from falling into the black hole, which, like a monster suffering from an extreme case of acid reflux, stops gaining mass and gravitational muscle. Both its growth and the growth of the galaxy come to a halt. Astronomers first glimpsed this cosmic shutoff valve in large galaxies in 2003 in high-resolution x-ray images of the Perseus cluster, a group of 500 galaxies about 250 million light-years from Earth. Since then, such “AGN feedback” has become one of the hottest topics in astrophysics. If it works the way some researchers think it does, AGN feedback is pivotal to understanding how galaxies evolve. It also appears to answer in one stroke two fundamental questions that have plagued researchers: Why are the masses of most galaxies so tightly connected to the masses of the enormous black holes at their centers? And why are very large and very small galaxies so much rarer than theory says they should be? A plausible answer to the first question, at least in midsize-to-large galaxies, is that AGN feedback acts like “a thermostat, a regulator to keep things matched,” says Meg Urry, an astronomer at Yale University. The mechanism also explains “why we don't see black holes bigger than 10 billion solar masses,” says Avi Loeb, a theoretical astrophysicist at Harvard University. “As you go to high masses, AGN feedback shuts off galaxy growth; hence we see few massive galaxies.” In the past 5 years, new observational and modeling studies have convinced astronomers that AGN feedback is universe-wide and has influenced galaxy evolution through much of cosmic history. A conclusive, detailed picture of how it works, however, remains a long way off. Although many astronomers believe that the expansion and movement of bubbles inflated by jets is what transfers energy into the gas within and around the galaxy, some argue that the gas is heated mainly by radiation emitted by the AGN. “Although mechanical heating by jets is the popular view, we simply do not know yet which mechanism is dominant,” says Loeb. ## Too darned hot When astronomer Andrew Fabian of the University of Cambridge in the United Kingdom took x-ray images of the Perseus cluster with the orbiting Chandra telescope, he expected to find its center awash in cold gas—a mere 10 million kelvin or below. The figure reflected how quickly the tenuous gas known as the intracluster medium was supposed to be cooling by emitting radiation. Fabian and his colleagues found that the gas in the core of the cluster, near the central galaxy, was tens of millions of degrees warmer than expected. Something must be heating it—but what? The pictures themselves provided an answer. They showed large cavities in the gas extending out from the central galaxy, right along the path of jets previously observed in radio images. (The particles in the jets are so energetic that they can be detected only by their radio emission.) The x-ray photos also showed ripples around the cavities, which the researchers saw as the unmistakable signatures of mechanical energy in motion. Other astronomers had suggested that jets shooting out from a galaxy might heat intracluster gas, but the idea had seemed as implausible as heating a house by shooting a laser up through the floorboards. The Perseus image, Fabian says, showed that “despite being radiatively inefficient, a jet could be extremely efficient in transferring kinetic energy.” Connecting the jets, bubbles, and waves into one picture, “we could actually see a chain of cause and effect,” he says. “If feedback didn't work, the central galaxy in Perseus would maybe have 10 times more stars than now observed.” This picture of mechanical feedback is now called the “radio mode,” because the jets emit strong radio waves. Researchers glimpsed AGN feedback in higher definition 3 years later, when Harvard astronomers William Forman, his wife, Christine Jones, and others used the Chandra telescope to observe M87—a massive galaxy in the middle of the Virgo cluster, just 1/7 as distant as Perseus. M87 also revealed bubbles and waves, including a circular ring around a bubble floating several light-years from the galactic core. Forman and his colleagues were able to calculate the temperature and pressure change across the wave fronts, providing a ball-park estimate of how much energy the waves were pumping into the gas. Using models developed with the help of Perseus and M87 observations, researchers have shown that the shock from the initial inflation by the jets, the propagation of waves, and the slow expansion of the bubbles into the intracluster medium can spread energy over vast distances to keep the gas from cooling. “The data imply that it's a relatively gentle process sustained over millions of years,” Fabian says. In such a “steady boil” scenario, calculations show that even occasional jets can keep the gas warm over long periods. Astronomers looking for more examples of feedback have seen several galaxies sporting ghost cavities or holes far from their centers. These bubbles probably formed when jets plowed through the intergalactic medium tens of millions of years ago and have since been slowly migrating through the gas. In a 2006 study, Fabian and colleagues spotted bubbles in 14 out of 20 galaxy clusters with unexpectedly warm gas at their centers. Other astronomers have followed up with radio observations to confirm that these bubbles—many of which are tiny compared with those seen in M87—were in fact produced by jets. Somak Raychaudhury of the University of Birmingham, U.K., and his colleagues have detected the faint signature of ancient jets leading into these bubbles using observations made through the Giant Metrewave Radio Telescope in Pune, India, an instrument designed for low radio frequencies. “As particles travel along the jets, they radiate energy; as they age, they emit lower and lower frequencies,” says Raychaudhury. “That allows us to reconstruct previous outbursts.” Even if a galaxy lacks jets now, he says, those long-vanished energy sources could still be keeping nearby space bubbling away merrily. ## The quick and the dead Although the broad outlines of the radio mode are not in dispute, the details are fuzzy and fraught with problems. One big stumbling block stems from geometry: The jets generally shoot out perpendicular to the galactic disk, sending the bubbles eddying away from that plane—not the most effective way to heat gas in and near the disk itself. “The bubbles are not going to do much to the immediate galaxy,” says Philip Hopkins, an astronomer at the University of California, Berkeley. Radio-mode effects might keep cluster gas from cooling and falling into the disk in the long run, Hopkins says. But to account for more-immediate processes—“things like shutting down the flow of material to the black hole or turning off the immediate star formation from gas that's already in the galaxy”—some other heating mechanism must be at work. Researchers think the key is a different kind of feedback called the quasar mode. It involves direct heating by radiation from the AGN or by explosive winds from near the black hole that blow gas out of the galaxy. Hopkins and many other theorists think this kind of feedback may have been dominant in the early universe, “quenching” quasars (essentially highly luminous AGNs) and transforming them into what are now massive “red and dead” galaxies, with very low levels of new star formation. Applied to more-recent galaxies, the quasar mode can account for AGN feedback in galaxies without jets. Some evidence for feedback on a large scale—possibly in the quasar mode—comes from a 2007 study by a group including Kevin Schawinski, now a postdoc at Yale, and Daniel Thomas of the University of Portsmouth in the U.K. By studying the spectra of some 16,000 galaxies with massive black holes at their centers, the researchers could infer which ones were still forming stars and which ones possessed the glowing core of an AGN. Most of the galaxies fell into one of two large bins: star-forming ones with no AGN activity and dead ones, also with no AGN activity. The remaining 10% of the galaxies bore the signatures of AGN activity and low levels of star formation. Estimating the average age of the three kinds of galaxies, the researchers found they fell into evolutionary sequence that suggested AGN feedback in action over hundreds of millions of years. “When the black holes are dormant, the galaxies form stars,” Thomas says. “Several hundred million years later, the black holes switch on AGN activity, suppressing star formation”—the 10% of galaxies in which the AGN had not yet completely stamped out the birth of new stars.The galaxies then move into the red-and-dead phase, the AGN eventually fizzling out. In a follow-up study of 20 galaxies along this evolutionary sequence, published in the February issue of The Astrophysical Journal, Thomas, Schawinski, and others found a more direct link between AGN activity and the halt to star formation. In galaxies in which the AGN had switched on, the researchers found evidence of little molecular gas—the fuel needed to make stars. “The AGN is destroying the molecular gas,” Thomas says: heating and blowing it out, as would be expected in the quasar mode. If AGN feedback really is as large-scale and far-ranging as Fabian and others suspect, it could solve a couple of major problems in astrophysics. One is the discrepancy between models of the universe and astronomical observations. The models accurately predict the cosmic distribution of dark matter—the invisible stuff that makes up the bulk of the universe. But they also predict that the galaxies that sit inside the dark matter halos ought to range in size much more widely than astronomers observe. The other problem is the surprising correlation between the mass of a galaxy and the mass of the black hole at its center. In the past decade, researchers have established that this ratio, known as the M-sigma relationship, is uniform for galaxies of different sizes throughout the universe. That regularity has led astronomers to wonder what physical process binds the destinies of the black hole and the galaxy. Cosmic limits to growth would explain those regularities of size and scale. As promising as AGN feedback appears, however, many more studies are needed to pin down the mechanisms by which it works and the time scales on which it operates, says Jeremiah Ostriker, a theorist at Princeton University. Most current models of both the radio and the quasar modes of feedback are too simplistic, Ostriker says. For example, “the calculations of the mechanical and thermal components of feedback through bubbles are highly uncertain.” Fabian agrees that the list of unresolved questions is long and expects the quest for answers to simmer on for a while. “I don't think we can write QED at the bottom of this,” he says. 11. China # Help Wanted: 2000 Leading Lights To Inject a Spirit of Innovation 1. Hao Xin* Reform-minded Chinese scientists hope a new program, Qianren Jihua, will reel in the sort of comrades who can help raise the nation's scientific game. Created just 5 years ago, the National Institute of Biological Sciences (NIBS) in Beijing is hailed—and envied—as a sanctuary where a select group of scientists is sheltered from China's patronage-driven funding system. So it was a major boost when officials announced earlier this month that biochemist Wang Xiaodong, a Howard Hughes Medical Institute investigator at the University of Texas Southwestern Medical Center at Dallas, not only would stay on for a second term at the helm of NIBS but also will close his lab in the United States and return to China for good. “There is no turning back,” Wang says. Wang is the biggest catch of a new program called Qianren Jihua, or Thousand-Person Plan, that aims to recruit up to 2000 top-notch scientists, entrepreneurs, and financial experts from abroad over the next 5 to 10 years. So far, Qianren status has been conferred on 96 “innovative talents,” who will work at universities and institutes, and 26 “entrepreneurial talents,” who will lead high-tech companies. The program was initiated by the Organization Department of the Central Committee of the Communist Party of China (Zhongzubu), which appoints and evaluates high-level cadres. Zhongzubu's foray into headhunting underlines Beijing's belief that China must instill a more innovative spirit into its work force to sustain economic development. Some scientists applaud Qianren as a timely effort to muster a critical mass of Western-trained researchers who, through strength in numbers, will gradually reform a scientific system rife with corruption, vested interests, and influence-peddling. “The plan aims to achieve two goals: to raise the level of research and to improve the academic environment,” says Qianren recipient Shi Yigong, a structural biologist who left Princeton University last year to lead the schools of medicine and of life sciences at Tsinghua University in Beijing. But the nascent program is under fire. Scientists who drafted and refined Qianren Jihua told Science the plan originally required recipients to work full-time in China; they were surprised, when the program rolled out last December, to see a minimum commitment set at 6 months a year for 3 years. Nor was the original plan limited to expats. “The idea was to establish government-endowed senior and junior faculty chair positions” to attract applications from qualified scholars from China or abroad, says Shi. Another bone of contention is money. The original plan called for steady research support for successful candidates, but Qianren offers recipients only a one-time$146,000 relocation subsidy from Zhongzubu. The education and science ministries had been expected to foot salaries—suggested by the proposal to be about 60% of what expats receive overseas—and start-up funds of up to $1.2 million over 5 to 7 years, depending on the field. Instead, universities have had to pony up salaries and start-up packages, at least for the first batch of recipients. Not surprisingly, says Qianren awardee Xu Huaxi, Xiamen University, which nominated him, “has become less enthusiastic.” Since 2003, Xu and fellow recipient Zhang Xiao-kun, both at the Burnham Institute for Medical Research in San Diego, California, have been volunteering to establish a biomedical research institute at the university along with three other overseas Xiamen alums. Xu says everything is still up in the air, from funding to how much time he will spend in Xiamen. Other Qianren recipients express similar uncertainty. “I am not sure how this will develop,” says Shao Zhifeng, who returned from the University of Virginia in Charlottesville last year to Shanghai Jiao Tong University to head its Institute of Systems Biology. ## Déjà vu For more than a decade, China has made a concerted effort to woo talented expats to return home. In 1998, the education ministry launched the Changjiang Scholars Program, partly sponsored by Hong Kong real estate tycoon Li Ka-shing to provide incentives—$14,600 supplemental salaries, high by standards then—to lure talent under age 45 to teach and do research at Chinese universities. When officials realized that most overseas tenured professors did not want to return full-time, the program expanded to accommodate senior part-timers (Science, 22 September 2006, p. 1721). Since 2003, the science ministry has also provided Changjiang Scholars who commit to 9 months a year in China with start-up funding of $293,000 over 3 years. Also in 1998, the Chinese Academy of Sciences (CAS) shifted the focus of its Bairen (Hundred-Person) Plan—launched in 1994 to support young researchers trained locally or abroad—to attract expats. CAS requires institutes that recruit Bairen awardees to provide$100,000 in start-up funds. After 1 year of full-time work, awardees are evaluated for merit grants of $253,000 over 3 years and a$40,000 housing subsidy from CAS.

Together, the two ongoing programs have recruited some 2000 scientists to return full time, most of them newly minted Ph.D.s and postdocs (see graph, p. 535). The programs have also enticed several hundred associate or full professors with tenure overseas to spend up to 3 months a year in China.

Qianren Jihua is the first such program administered by the central government and the first to target entrepreneurs as well as academics. It is the brainchild of Zhongzubu chief Li Yuanchao, who has made recruiting talented people a priority. Li has been pushing leaders at universities, CAS, and industry—almost all of whom are Party cadres—to be more daring and reel in talent faster.

After notice went out last December, organizations scrambled to find Qianren candidates, mostly by going after colleagues already moonlighting in China. Tsinghua signed up four part-timers as candidates, including Hu Xiaoping, a biomedical imaging expert at Emory University in Atlanta who has been helping Tsinghua establish a neuroscience imaging center. Hu is weighing his options: “The government wants us to decide quickly, but many policies are not clear,” he says. Universities, CAS institutes, and national labs nominated candidates to either the education or science ministries. Both ministries appointed review panels, composed of academicians and science mandarins, which invited candidates to Beijing to give 20-minute presentations.

Although Zhongzubu has not released recipients' names, a dozen or so universities have publicized their Qianren awardees. The majority of them had been recruited by previous incentive programs. Two of Beijing University's (Beida's) three recipients have been recruited home before this. She Zhensu, an applied mathematician at the University of California, Los Angeles (UCLA), was one of the first Changjiang scholars in 1999 (Science, 21 January 2000, p. 417). Last year, he finally gave up his tenured appointment at UCLA to work full-time at Beida. Another recipient and former Changjiang scholar is She's boss, Chen Shiyi, Beida's engineering dean. Chen continues to keep one foot in the United States—at Johns Hopkins University's engineering school—after nearly a decade of shuttling across the Pacific Ocean. According to Andrew Douglas, associate dean of academic affairs at the engineering school, Chen is on “partial leave, [which] would allow him to work 6 months of the year in China.” Chen declined and She did not respond to interview requests.

## Overcommitted

Holding down simultaneous appointments in China and overseas might be manageable for theoreticians such as Chen, but it is hard for experimentalists to run labs on two continents. Those who try may discover that “it will be very tiring, and in the end both labs will suffer,” says Hu. Wang Xiaodong thinks most U.S. universities would not allow tenured principal investigators in experimental sciences to work parttime for extended periods. “Qianren Jihua's 6-month requirement is not workable; it will create conflicts of commitment,” he says.

One Qianren awardee, Wang Peng, a carbohydrate researcher, is facing a review over his relationships with Chinese universities, according to documents obtained by Science. Wang has been an Ohio Eminent Scholar at the Ohio State University (OSU) in Columbus since 2003. Earlier that year, Shandong University in Jinan recruited Wang as a Changjiang-Scholar professor with a 9-month-per-year contract. Wang told Science that he “did not know what he was getting into” by accepting the Changjiang award; during his 5-year appointment, he says, he only spent short periods in Shandong. After the scholarship ended, Wang's name remained on the Shandong faculty roster as a Ph.D adviser. Wang says he resigned earlier this week from the Shandong position.

In 2007, Wang accepted an offer from Nankai University in Tianjin to set up a new school of pharmacy there and be its first dean. Nankai later nominated Wang for a Qianren award. Last March, Ohio State received a complaint about Wang's relationships with Chinese universities. Matthew Platz, interim dean of the colleges of biological sciences and of physical and mathematical sciences, says the matter is under review. In an April 2008 letter to OSU's biochemistry department chair, Wang explained that his activities at Nankai “are in the scope of consulting,” the title of dean is “symbolic,” and he would spend no more than 20% of the 9 months he is paid by OSU and 2 or 3 months in the summer when he is not paid by OSU or federal grants to consult at Nankai. As for his Qianren award, Wang says he has not decided what to do as Nankai has not made him an offer.

Unlike universities, CAS requires its Qianren recruits to work full-time in China, says Li Hefeng, director of CAS's Bureau for Human Resources and Education. Of the first batch of 122 Qianren recipients, 19 are affiliated with CAS, including five at the academy's University of Science and Technology of China in Hefei. CAS is providing $293,000 to each “innovative talent” and asks the recruiting institute to match the amount. No more than half of that sum may be used for salaries and fringe benefits for the recipient and lab members; the rest is for research overhead costs. A CAS Qianren recipient who previously received Bairen support will not get the$293,000, Li says.

A half-dozen university recruits told Science they will work full-time in China. Among them is mechanical engineer Robert Parker, who last September became associate dean for academic affairs at the University of Michigan-Shanghai Jiao Tong University Joint Institute in Shanghai. Parker says it wasn't the Qianren award that brought him to China but the challenge of building an institute from the ground up. “The goal,” he says, “is to create a school of engineering which in quality and scope reaches the level of a major engineering school in the United States.”

Other scientists who committed to working full-time in China—and who subsequently were given Qianren awards—echo that sentiment. Wang Xiaodong says he is returning to help like-minded colleagues steer Chinese academia toward a merit-based system. Wang, who was elected to the U.S. National Academy of Sciences in 2004 at age 41 for his research on mechanisms of cell death, says he is coming home to give it his all: “I do not want to leave any regrets.” China's push to recruit good scientists, it seems, is already paying off.

• * With reporting by Li Jiao, a writer in Beijing.

12. News

# Nursing China's Ailing Forests Back to Health

1. Richard Stone

A lauded effort to create mixed forest stands is giving villagers and loggers a chance to make a living while restoring ecosystem vitality.

PINGXIANG, CHINA—The swath of hilly terrain looked like a man's face after a poor shave. A few days earlier, villagers had clear-cut a fir forest and burned the stumps. Soon they will uproot the charred stubble and repopulate the barren land with eucalyptus, a fast-growing Australian import. “The timber is low quality,” says Lu Yuanchang, a forester at the Research Institute of Forest Resource Information Techniques of the Chinese Academy of Forestry (CAF) in Beijing. But after 5 years or so, the villagers who manage this community forest in southern China's Guangxi Zhuang Autonomous Region, near the border with Vietnam, will hack down the eucalyptus and pulp it for paper. “They should turn a quick profit,” Lu says.

Ecological expediency of this sort has been rampant in sections of southern China, as villagers and local governments conspire to transform vibrant forests into plantations for money-spinners such as eucalyptus, rubber, and oil palm. “In recent decades, no other country in the world has established more forest plantations than China,” says Heinrich Spiecker, director of the Institute for Forest Growth in Freiburg, Germany. Yet forests cover only about 18% of China's landmass, and timber yield and quality are lower than in many other places.

China's enthusiasm for monocultures has taken a heavy toll. “Converting large areas into single-species plantations is destroying the environment,” Lu says. “There has been large-scale degradation of China's forests.” Monocultures are often susceptible to pests and diseases, Spiecker notes. According to Lu, the most insidious effects stem from soil degradation, which destabilizes ecosystems. Monocultures, he says, were largely to blame for widespread forest losses inflicted by last year's ice storms (Science, 7 March 2008, p. 1318).

The problem is likely to get worse before it gets better. In 2003, the central government approved a regulation that grants Chinese citizens an extension of how long they can farm or manage a community forest—from 30 to 70 years. That enormous policy change effectively grants people land ownership for their lifetimes. “The policy is meant to stimulate families to take care of the land,” says Cai Daoxiong, director of CAF's Experimental Center of Tropical Forestry in Pingxiang. However, notes Bernhard von der Heyde, director of the sustainable forest management project of the German nonprofit development group GTZ in Beijing, “there is a big rift between national will and local implementation.” Seduced by easy money, many individuals do as they please with forest patches under their control. “Many people just concentrate on short-term economic benefits,” Cai says.

Lu hopes to restore China's forests before it's too late. Less than a kilometer from the budding eucalyptus plantation in Pingxiang is a CAF effort to engineer a “multipurpose” forest. In 2005, instead of clear-cutting, the Pingxiang center selectively harvested stands of Chinese fir (Cunninghamia lanceolata) and brought in seedlings of more than 20 hardwoods prized for their timber, including species of Castanopsis, a genus in the beech family, Ceylon ironwood (Mesua ferrea), and Michelia hedyosperma, a kind of magnolia. “It's really a simple concept,” Lu says. “We are transforming single-species stands into mixed stands for economic and ecological benefit.”

Near a Castanopsis sapling, Lu brushes away leaf litter and runs his fingers through rich humus. “In a good forest, you should not see the soil. You should only see humus,” he says. For this particular forest revival, Lu selected a complementary mix of pioneer and mid- to terminal-succession tree species that play different ecosystem roles, from enriching the soil to forming the kind of canopy and fruits that attract birds and other wildlife. In several years, CAF can harvest fir trees that reach a target girth and in a few decades start logging valuable hardwoods tree by tree. Selective logging, Lu says, is slightly more expensive in China than clear-cutting, but in time the greater profit from the hardwood timber will more than compensate for the higher costs. “And we save the ecosystem,” Lu says.

The science behind Lu's approach is not novel, but its impact on the restoration of China's forests could be revolutionary. And it's getting rave reviews. “Pingxiang is the best research model in China,” says Li Nu Yun, deputy director general of the afforestation department of China's State Forestry Administration in Beijing. Multipurpose forestry, Von der Heyde says, “will increase carbon stocks in the ground tremendously.” Lu and colleagues have launched multipurpose forest experiments in Guangxi, on the tropical island-province of Hainan, in the arid plains of Shanxi Province, and near Beijing. Experts say it should be possible to persuade government entities that oversee logging in state forests—more than 40% of China's forest cover—to set aside more and more land for multipurpose forestry. But persuading villagers to buy into the concept—and forgo quick profits—is more daunting. “It will take tremendous will and patience to implement this program,” says Von der Heyde.

## Nature's way

In the early 1980s, soon after China's devastating Cultural Revolution, researchers cast around for a more scientific approach to forest management. “We planted some high-value trees, and that was it,” Cai says. “But we really did not know how to practice sustainable forestry. Then Professor Lu came along.”

Lu had trained in Germany in the 1990s, where he had become steeped in the country's “close to nature” approach to forestry. Like other central European nations, Germany lost its old-growth forests to centuries of agriculture and industrial development. During his stint in Germany, Lu gained an appreciation for the country's methodical approach to managing secondary stands: selecting complementary tree species and minimizing the impact of timber harvests on ecosystem health.

The benefits of this approach are many and varied, Lu says. The better the mix of trees, the richer the humus and the greater the soil's capacity to retain water. Reducing runoff stabilizes not only the hydrology of the watershed but also the local climate, he says. And the ecosystems are more resistant to pests and disease.

When Lu returned to Beijing in 2000, he proposed that China adopt the close-to-nature approach. He encountered plenty of resistance: “People thought that ‘close to nature’ is only a luxury. It was the hardest thing in the world to change people's minds.” While a few centers like Pingxiang were tinkering with multifunctional forestry, the vast majority of state forests were managed according to the rotation system, in which stands are clear-cut and replanted, for example, with Pinus radiata pines, firs, or other fast-growing species. “That is the way it has been done for years,” Lu says. “China has had a very hard time coming around to the idea of multipurpose forestry.”

For 3 years after Lu's return to China, one grant proposal after another that he wrote was rejected. But his superiors tolerated his close-to-nature proselytizing, and in 2003 Lu finally won funding for a multipurpose forestry experiment near CAF's headquarters in Beijing's northwestern suburbs. The small plots are meant to be pleasant arboreal getaways for smog-choked Beijingers rather than a proof of principle that multipurpose forests make economic sense. To make that case, Lu needed to find colleagues who managed larger acreages and who could demonstrate that multipurpose forests, after a few decades, would pad the bottom line of timber concerns. When Lu paid a visit to Pingxiang in 2005, he realized that he and Cai “were of one mind.”

The duo immediately launched some multipurpose plots. “Lu introduced to us a scientific idea of forestry,” says Cai. “We now believe this management system is better than a rotation system.” Although the stands cover only about 100 hectares—a tiny fraction of the Pingxiang center's 20,000 hectares of land—“I am now convinced that all of our forest should be managed in a sustainable way. Gradually, we'll reduce the clear-cut area,” Cai says. “We want to make this forest center a model for all of China.”

## Reasonable facsimiles

Gently stirred by a light breeze, the canopy casts an ever-changing mottled pattern on a Castanopsis trunk in one Pingxiang stand. In 20 years or so this tree, like others of its kind here, will be ready for harvest. Like an adoring father, Lu gives it a pat. “We'll come back in a few years and check on it, see how it's doing,” he says.

Multipurpose forests in Guangxi and other regions will resemble, rather than imitate, original primary forests. It's hard to pin down what a forest in any given locale should look like: Its “potential natural vegetation” depends on local climate and therefore changes as the climate shifts. “Trying to determine past, present, and future potential natural vegetation is like a detective story,” says Lu. In Guangxi, he says, there may be no forests left with species assemblages that existed here 1000 years ago. Logging is not permitted near the Vietnam border, he says, but even those forests are at most a century old. “The fact is, forests are dynamic,” Lu says.

Rather than strive for an ideal that does not exist and may never have existed, Lu favors a more practical restoration target of achieving a healthy forest ecosystem that's a reasonable facsimile of potential natural vegetation. That doesn't require the wholesale abolition of plantations. Rather, it means convincing forest managers—and villagers—that multipurpose forests, by improving overall ecosystem health, will make many existing plantations viable. “A well-arranged mixture of forest types adapted to site conditions and local needs may best serve society,” says Spiecker. Li of the State Forestry Administration agrees. “Now local people can decide themselves what kind of species to plant over their lifetime,” she says. “They can think about what will be best for the next generation.”

13. News

# Restoring a ‘Biological Desert’ on Borneo

1. Dennis Normile

An economic incentive for area residents is a cornerstone of the Samboja Lestari project, an ambitious effort to transform a clear-cut site in Borneo into a mix of agroforestry plots and orangutan habitat.

For 30 years in Indonesia, Willie Smits observed how one conservation project after another would fail without local community support. Finally, Smits hit upon a solution: An economic incentive for area residents became a cornerstone of his Samboja Lestari project, an ambitious effort to transform a clear-cut site in Borneo into a mix of agroforestry plots and orangutan habitat.

“The extent and depth of his integration of ecological [restoration] with economic restoration is unique in my experience,” says Amory Lovins, an expert on sustainability at Rocky Mountain Institute, a think tank in Snowmass, Colorado. “It's the most important such project I know of in the tropics, if not in the world.”

Born in the Netherlands and now a citizen of Indonesia, Smits founded the Borneo Orangutan Survival Foundation in 1991 to rescue orphans. He says he realized that the underlying problem affecting orangutans and other wildlife is habitat loss. So in 2002, Smits purchased 2000 hectares of deforested land in eastern Borneo, near Balikpapan. The site was “a biological desert,” he says, and the district, with a 50% unemployment rate, was the province's poorest. Villagers who sold land to the project received a plot in a zone ringing the site where they planted acacia trees, which will provide timber, and sugar palms for sap to be processed into ethanol. Cash crops grown among the trees include ginger, papayas, cocoa, and chilies.

In the inner zone, teams of villagers were paid to plant fast-growing tree species to kick-start reforestation as well as slower-growing rainforest species. Enormous diversity—some 1600 tree species in all—has created a multi-layered canopy that makes the most of sunlight and maximizes biomass. Smits has sunk about $4.5 million, raised from a foundation he directs and from small contributions, into the project. So far, the reforested habitat has attracted 137 bird species and is now home to orangutans and sun bears, a species found in Southeast Asian rainforests. The greenery has lowered air temperatures by 3° to 5°C in the immediate vicinity and increased rainfall by 25%. Samboja Lestari's legal status as a nongovernmental project means it can avoid Indonesia's slow-moving bureaucracy and avoid compromises with commercial concerns eager to tap the region's coal deposits, Smits says. Community members know they, not the government, own the land, and the construction, replanting, and the Borneo Orangutan Survival Foundation's rescue and rehabilitation operations have generated 3000 jobs. Governance based on local traditions (Smits's wife is a tribal leader in North Sulawesi) means that the community, primarily through peer pressure, ensures that its own members do not fell trees and that they cooperate with project officials to keep poachers and loggers at bay. Although Smits hopes Samboja Lestari will set an example, he warns that there is no single recipe for forest restoration. “You have to do detailed design for each location,” he says. The key, he adds, is striking the right balance between suitable ecological methods and local needs. 14. News # Bringing Coral Reefs Back From the Living Dead 1. Dennis Normile A smattering of efforts are aiming to prove that degraded coral reefs can be restored to functionality, if not pristine beauty. ISHIGAKI, JAPAN—Last April on the northern edge of Sekisei Lagoon, Japan's largest coral reef, five divers placed dozens of stainless steel cages packed with custom-made ceramic disks on the sea floor near a patch of healthy coral in the ailing reef. All had to be ready before a mass spawning event expected during the new moon in May, in hopes that millions of coral larvae would settle on the disks. The high-tech operation went according to plan, and earlier this month the divers moved the cages to sheltered lagoon waters to enable the larvae to mature and form colonies. In another 18 months or so, the disks will be cemented into parts of the reef where a process known as bleaching, triggered by unusually warm waters, had killed the coral. In 30 years or so, researchers hope, the 27,000-hectare reef will be fully restored. One year earlier and 1000 kilometers to the southwest, the Philippine coastal community of Bolinao set out to restore its bleached and overfished reef. Residents decked out in homemade plywood flippers dove without air tanks from outrigger canoes. They broke off fragments of a robust coral from one part of the reef and wedged pieces into cracks in bleached sections. Six months later they returned, snapped off bits of the healthy transplanted coral, and repeated the process. The Bolinao reef restoration may be the most inexpensive in the world, whereas that in Sekisei Lagoon is probably the costliest. But both projects are important experiments in an urgent effort to stabilize the world's embattled coral reefs, which provide habitat for some 9 million species, including 4000 kinds of fish. Roughly 100 million people in developing countries depend on reefs for subsistence fishing and tourism, estimates the Global Environmental Facility (GEF). The “rainforests of the sea,” however, are threatened by human activity and natural disasters. About 19% of our planet's original global coral reef area has been destroyed; another 15% could be lost in the next 2 decades, according to the Global Coral Reef Monitoring Network's Status of Coral Reefs of the World: 2008. As losses mount, restoration projects are only just getting off the ground. “The science of reef restoration is in its infancy,” says Alasdair Edwards, a reef scientist at Newcastle University in New-castle upon Tyne, U.K. “There are tens of thousands of square kilometers of degraded reefs apparently out there, and almost all restoration trials are working at subhectare scales.” The immediate challenge, he and others say, is to show that promising techniques can be scaled up. That will require resources that are by no means guaranteed: A good share of recent restoration projects has been funded by GEF's Coral Reef Targeted Research program, which ends this year. Some experts are skeptical that restoration can make much of a difference. Healthy reefs lightly disturbed by humans typically recover from bleaching and natural disasters on their own, Edwards notes. But reefs pummeled by pollution, destructive fishing practices, or land reclamation are often pushed beyond recovery by bleaching or storms. “The only way to help is to reduce stresses that made [such reefs] degrade in the first place,” says David Fisk, a coral reef scientist and consultant in Geneva. ## Saving Sekisei The daunting challenges facing restoration scientists are evident at Sekisei Lagoon, cradled by Ishigaki and Iriomote, the two southernmost large islands of Okinawa prefecture. Sekisei's predicament has been documented by Mineo Okamoto, a specialist in marine assessment techniques at Tokyo University of Marine Science and Technology, who in 1993 started mapping coral in the lagoon using echo sounders and cameras to add detail to satellite images. Back then, Sekisei was in pristine condition, thanks to its remote location and status as a national park. But Okamoto sensed that change was coming from global warming, with its threat of more-frequent bleaching events, and rising ocean acidification, which is caused by the uptake of carbon dioxide from the atmosphere and weakens corals (Science, 4 May 2007, p. 678). Okamoto set out to document how such climate changes affect a coral reef. Okamoto completed his first coral map in 1998. That year an El Niño followed by a La Niña warmed the eastern Pacific to the point that zooxanthellae—algae that live symbiotically with coral and provide nutrients—deserted their hosts. Without zooxanthellae, corals blanch and unless the algae return within a few weeks, the colony starves and perishes. The 1998 bleaching event killed 16% of corals worldwide (Science, 27 October 2000, p. 682). Sekisei Lagoon's northern edge was hit hard. But the interior and southern rim were spared, which enabled coral to recolonize bleached sections without human intervention. Three bleaching events since 2001 devastated the interior and southern sections even as the northern rim recovered. But prevailing currents tend to sweep larvae produced on the northern edge away from the lagoon, impeding recolonization of recently damaged areas. To speed the reef 's recovery, Okamoto custom-made ceramic disks with grooves on the undersides that coral larvae can nestle in while avoiding algae that foul the disks' tops. Okamoto placed limited numbers of disks—which are small enough to fit in the palm of a hand—in the lagoon in 2002. Two years later, Japan's environment ministry made Sekisei Lagoon's restoration “a big public works project” and bestowed a budget of about$430,000 a year, says Okamoto, who now serves as a consultant. Teams have been placing cages filled with disks along the ocean side of the reef's 10-kilometer-long northern rim before the annual mass coral spawning. Months later, they move disks hosting juveniles to the lagoon's interior, where the corals mature. Okamoto estimates that the May 2008 spawning yielded about 10,000 colonies that will be placed on the lagoon's southern edge in December.

“We have set a long-term period of 30 years for this restoration project,” says Takanori Satoh, a ranger for the ministry's Coral Reef Research and Monitoring Center in Ishigaki. In 7 or 8 years, he says, researchers will be able to ascertain whether larvae from the trans planted corals have begun to recolonize the lagoon's interior.

## Economizing

Few projects can expect such unstinting support as Sekisei. “At the end of the day, [reef restoration] has got to be something low-cost and low-tech that involves the local communities,” says Edgardo Gomez, a marine biologist at the University of the Philippines in Diliman, who has worked with fishers in Bolinao, where the university has a marine lab, to restore reefs. Degraded by decades of blast fishing, Bolinao's reefs were devastated by the 1998 bleaching and have recovered only spottily. “The fish catch greatly diminished, and people had to resort to other ways of making a living,” Gomez says.

To restore the reef, Gomez needed to find a coral species that thrives even when handled roughly. “The villagers can't afford adhesives, and nobody in the town of Bolinao, save for the dive shop, has scuba equipment,” Gomez says. He settled on Porites cylindrica, a fast-growing branching coral. During their first training workshop in April 2008, Bolinao spear fishers planted about 400 square meters of reef. Six months later, 80% of transplanted coral was growing nicely and was big enough for bits to be broken off to replant another 400 square meters. The process was repeated in spring 2009. “If we keep doubling the areas we are restoring, in time it will become significant,” Gomez says.

Gomez is planning to introduce these methods elsewhere in the Philippines and in other countries. He admits that the patched-up reefs will lack the diversity of natural reefs. “What is most important is that you build back the [roughness] of the reef so fish and invertebrates have refuges,” he says. The promise of a healthy fishery gives locals an incentive to stick with an effort that may take years before they see a payoff.

Another approach to restoration on the cheap is what Baruch Rinkevich, a marine biologist at the National Institute of Oceanography in Haifa, Israel, calls “underwater silviculture.” Just as foresters rear trees in nurseries for transplantation, Rinkevich and colleagues have pioneered coral nurseries: mesh nets or even lengths of rope raised off the sea floor to avoid predators and sedimentation. Students, technicians, or local fishers collect donor corals, chop them into nubbins as small as 0.5 centimeters across, and glue these onto a substrate—anything from seashells to bits of plastic piping—that is then attached to the nursery nets. Rinkevich says up to 99% of nubbins survive, depending on the species, and 12 to 18 months later they are big enough to transplant onto a reef.

Rinkevich worked out the kinks of the technique in Eilat, on the Red Sea, starting in the mid-1990s. More recently, he has experimented in Jamaica, the Philippines, Singapore, Thailand, and Zanzibar. “The results are more than just promising; they are beautiful,” he says. Rinkevich claims that under optimal conditions in countries with cheap labor, he can produce a colony ready for transplantation for as little as 18 cents. So far, Rinkevich has built nurseries with 10,000 corals and transplanted up to 3000 colonies. He envisions nurseries with up to 200,000 corals and similarly supersized transplantations. “By doing that, I'm pretty sure we can change completely denuded reefs,” he says. Newcastle's Edwards agrees that nurseries are one of the most promising approaches. “You're minimizing the collateral damage [by cutting donor material into small pieces], getting transplants, and maximizing the effect you can have,” he says.

One drawback is that a reliance on nubbins from a single donor coral results in limited genetic diversity. To address that, Rinkevich and others are investigating a variation on the theme in which they capture eggs and sperm from the ocean during mass spawning events or from colonies taken into the lab just before spawning (Science, 14 December 2007, p. 1715). The gametes are mixed, and the resulting larvae settle onto substrates in tanks. The juvenile corals are moved into the nursery and transplanted to a reef a year or so later. Although this procedure boosts genetic diversity, it is more costly than nubbins and requires more expertise.

One recent trial in the Philippines led by James Guest, a marine biologist at the National University of Singapore, yielded about 1.6 million larvae from 19 colonies in tanks with 2000 concrete pins as substrates. Based on work in 2008, they expect about 1000 colonies to survive the first year. “We still have lots of work to do to improve these survival levels,” Guest says.

Another approach is more like animal husbandry than silviculture. Researchers have “got a bit of a handle on the process of sexual reproduction in corals,” which opens up the possibility of manipulating larvae, says Andrew Heyward, a coral reproductive specialist at the Australian Institute of Marine Science in Perth.

In a 2008 experiment in Palau, Heyward's team collected gametes from lab colonies and cultured larvae in improvised tanks: aboveground swimming pools. Each day they tested a few larvae to see if these were ready to settle and form a colony by exposing them to a reef chemical that triggers settlement in “competent” larvae. When half the batch was primed to settle, the researchers funneled larvae into an enclosure over an artificial reef. Six months later, they found five times as many juvenile corals on the treated rock as on nearby rocks that relied on natural recruitment. The equation is simple, says Heyward: More larvae in a given area mean better recruitment. “It takes almost no technology,” he says.

With the future of reefs hanging in the balance, coral scientists are eagerly following the nascent restoration efforts. The various techniques being deployed “are still basically unproven in the long term,” Edwards cautions. Fisk in Geneva is pessimistic about any restoration effort reaching a critical scale that would make a reef resilient. “You can't build big enough to withstand natural and manmade disturbances,” he says.

Others insist that reef restoration will ultimately prevail. Rinkevich compares reef rehab to terrestrial reforestation. Although most present-day forests in North America and Europe are human cultivated, they nonetheless provide habitat for wildlife and prevent soil erosion. “In similar ways, transplanted corals change the environment in an area around a reef: Fish arrive, and invertebrates arrive,” Rinkevich says. The detrimental impact of global warming and rising ocean acidification is “going to get worse before it gets better,” adds Gomez. But “waging a rear-guard action” through restoration should lessen the harm inflicted on reefs and on livelihoods.

Such debates might be settled “as soon as someone can come up with a large-scale restoration that has worked,” says Edwards. It may be up to Sekisei Lagoon and bootstrapping community efforts to prove that the world's battered coral reefs can be saved.

15. News

# Unleashing an Army to Repair Alien-Ravaged Ecosystems

1. Robert Koenig

South Africa has scored decisive victories in its grassroots assault on invasive species. But will the war ever end?

When fires devastated 10,000 hectares of the Cape Peninsula's Table Mountain range in 2000, South Africa faced a race against time to prevent fast-growing invasive plants from overwhelming native flora. Its solution: Ukuvuka, a campaign that deployed hundreds of workers to uproot aliens and rehabilitate and reseed sensitive tracts of the rugged terrain. Nine years later, native vegetation is flourishing.

Ukuvuka, from a Xhosa word meaning “to rise up,” is one of more than 300 projects initiated by a pioneering program called Working for Water (WfW). Since 1995, WfW has restored ecosystems by taking aim at some 200 alien plant species that clog waterways, degrade farmland, and heighten wildfire threat. Major efforts have included campaigns to control South American pompom weed, a meter-high plant that threatened to run rampant in KwaZulu-Natal province; clear invasive plants for ecosystem restoration along Kruger National Park's major rivers; and facilitate production of “eco-coffins” and school desks made from cleared invasives. At the same time, WfW has created tens of thousands of jobs in a country where one out of every four adults is unemployed.

After centuries of colonization and trade, South Africa is now home to “a witches' brew of alien species” from Australia, Europe, Asia, and the Americas, says ecologist Richard N. Mack of Washington State University, Pullman. As an antidote, WfW has cleared about 1 million hectares of invasive species in 15 years. “In terms of its reach, its intensity, longevity, and the number of invasive species it targets,” Mack says, “I can't think of another ongoing project quite as ambitious.”

## Shock troops

Alien species got a toehold in South Africa in the 1800s, when European settlers established sprawling plantations of Australian eucalyptus and European pine. Those species are thirstier than native trees and placed a heavier demand on the watershed, a problem compounded by the expanding agricultural sector. Many rivers and streams shriveled as banks became overgrown with invasive trees and weeds that arrived as imported seeds.

In 1994, the year that apartheid ended and Nelson Mandela became president, a panel of South African scientists and resource managers predicted a water crisis unless aggressive steps were taken to manage catchments and control invasive vegetation. Mandela's first minister for water affairs, human rights activist Kadar Asmal, hired ecologist Guy Preston, research chief at the University of Cape Town's Environmental Evaluation Unit, as an adviser. Together they launched WfW, which combined a public works campaign with water preservation and the fight against invasives. They initially focused on riverbank foliage, but the program soon expanded to a $100-million-per-year effort that tackles “the whole spectrum of alien plant invasions,” Preston says. To mount their campaign, Asmal and Preston enlisted an army: in the beginning, about 7500 part-timers whose ranks have since swelled to 29,000 per year. At a starting wage of$6 a day, the workers donned yellow WfW T-shirts and fanned out across the country, wielding chain saws, slash hooks, and herbicides against invasive plants. WfW scientists have also deployed more than 75 imported insects to eat their way through 45 invasive plant species, with generally good results so far. Harold Mooney, an ecologist at Stanford University in Palo Alto, California, who helped establish the Kenya-based Global Invasive Species Programme, calls WfW “an innovative program that addresses invasive species, water, and job creation in an integrated way.”

Preston says WfW has “made science relevant to poor people,” but that was only possible by making the program relevant to the new government. WfW has managed to win steady funding on the argument that ripping out water-slurping invasive plants costs less than building reservoirs to augment water supplies. “Invasive plants have a massive impact” on water levels, soaking up 7% of mean annual runoff water, Preston says. If left uncontrolled, research indicates, invasives could soak up 20%.

Buttressing that argument, scientists at South Africa's Council for Scientific and Industrial Research used field-mapping, biomass, and catchment studies to estimate that alien plants had invaded about 10 million hectares and consumed 3.3 billion cubic meters of water per year—about 75% of the volume of one of South Africa's largest river systems. “Alien plant control is expensive,” the council concluded in its 2000 report, but “control programs are cost-effective compared with alternative water-supply schemes.”

Weed-whacking writ large was only part of the solution. WfW also had to face down tree plantation companies. With prodding from WfW, the government, and the Forestry Stewardship Council, plantations were persuaded to remove some trees in riparian, wetland, and upper catchment areas and pay fees to offset the value of water their trees absorb.

## Invasive marathon

Restoring natural ecosystems after an alien invasion is “an ultralong-distance race,” Preston says. How that race is run depends on landscape, climate, the alien species, and the available native seed bank.

One common approach called “fell and remove” involves cutting down, weeding out, or burning alien species. Restorers assume that the indigenous seed bank will survive and the native plants will regrow. “In most cases, there are enough seeds of native species in the area to replace invasive plants that were removed,” Preston says. Sometimes, however, reseeding of native species is necessary. That was the case in restoring the native fynbos ecosystem, a highly diverse mix of heath and shrubland plants in the Table Mountain range where intense fires roasted indigenous seeds.

Even areas cleared of invasives tend to revert. Invasive alien plant surveys by the Agricultural Research Council's Institute for Soil, Climate and Water in Stellenbosch have noted progress in some regions but accelerating invasions in others, primarily where clearing has not taken place (see map). “The fact that 9 million to 10 million hectares are currently invaded and under 1 million has been cleared means that more resources need to be allocated,” says entomologist Dennis Rangi of the Commonwealth Agricultural Bureau International in Nairobi.

One challenge is that most invasive plant seed banks are healthy, allowing the aliens to grow back year after year. WfW sends out regular patrols to areas previously cleared by its shock troops to rip out return visitors. Another problem is that ridding an area of one invasive species can pave the way for another. After workers in 2004 cut down Eucalyptus grandis trees along the Sabie River, bugweed and a whole suite of other stubbier trees and shrubs sprouted in their place.

“Many invasives will quickly return unless the eradication effort is continued on a regular, or at least periodic, basis,” says Jeffrey McNeely, chief scientist at the International Union for Conservation of Nature in Gland, Switzerland. Adds plant ecologist Mark Lonsdale of Australia's Commonwealth Scientific and Industrial Research Organisation: “You have to be strategic because of long-lived seed banks, regrowth, and re-encroachment.”

That means WfW must grind on indefinitely to remain effective. “To the extent you can compel landowners to participate and shoulder the cost and generate revenue and indirect benefits from the project [as WfW has], you have something more sustainable,” says Jeffrey Waage, an invasive-plant expert who directs the London International Development Center.

Water savings is not the only benefit from WfW: Clearing invasive plants from threatened habitats has also helped save some native insect species. Dense canopies of woody invasive plants provide more shade than dragonflies and damselflies are accustomed to, says entomologist Michael Samways of the University of Stellenbosch. As a result, Preston says, “over half of our endemic species of dragonflies and damsel flies were facing extinction” before WfW began. Recently, two species written off as extinct—the Ceres Stream damselfly and the Cape Bluet dragonfly—have been spotted near ponds purged of invasive trees. Getting rid of alien flora has “absolutely” benefited these insects, Samways says.

WfW has expanded its resources by partnering with provincial and local governments: Cape Town's municipal authority has taken over the Ukuvuka fire-restoration campaign in the Table Mountains, and KwaZulu-Natal province's Invasive Alien Species Programme is running the campaigns against pompom and other invasives.

Trying to extend its success to related areas of ecological restoration, WfW has spawned a trio of programs in South Africa with a narrower focus: Working on Fire, Working for Wetlands, and Working for Energy (using biomass to produce energy). A key link between the Water and Fire initiatives is that “some invasives burn at an intensity of 10 times or more the indigenous vegetation they displace,” says Preston. He concedes that alien plants “are spreading and growing” in many regions where WfW has not had the resources to reach. Still, he says, “it's important to take action. It's like a cancer: What would happen if you do nothing?”

16. News

# Addicted to Rubber

1. Charles C. Mann*

A rising tide of rubber plantations is eating away at Southeast Asian ecosystems. Can an ancient forest crop help wean the region off its monoculture habit?

BAN NAMMA, LAOS—When the man from Huipeng Rubber visited, most of the village's men came to meet him. They hunkered down in sandals and worn T-shirts on the bare ground in front of the village headquarters, as the company agent, dressed in a sports coat, distributed cigarettes. Then the rep used a stick to sketch in the dirt the next step in the company's plan to remove 540 hectares (ha) of village forest and fields and replace them with rubber trees.

Ban Namma is at the edge of the Golden Triangle, the mountainous, thickly forested intersection of Laos, Myanmar (formerly Burma), and China. Long infamous for heroin, opium, and other poppy products, the region is now becoming known for another plant: Hevea brasiliensis, the Pará rubber tree. China has spread rubber across as much as 300,000 ha of Yunnan, replacing most of the southern province's lowland tropical forest. But now the nation is running out of land warm and wet enough for H. brasiliensis. In response, according to a 2008 report by economist Weiyi Shi for the German nonprofit development group GTZ, smallholders with Chinese connections and, more recently, Chinese companies have begun to clear large swaths of Laos for rubber. A single holding company, China-Lao Ruifeng Rubber, plans to cut and plant 300,000 ha; a second company, Yunnan Rubber, plans to convert 167,000 ha. As the tide of rubber sweeps from China into Laos, says Tang Jianwei, an ecologist at Yunnan's Xishuangbanna Tropical Botanical Garden (XTBG), the entire region is being transformed into an “organic factory”—with alarming environmental consequences.

Natural rubber is found suspended as minute particles in latex, a saplike substance made in hundreds of plants. Only a few, though, produce rubber suitable for human use, and the most important by far is the Amazonian tree H. brasiliensis. Rubbermakers tap the trees and—in a process reminiscent of making maple syrup—boil down the latex to draw out rubber. A problem, say ecologists, is that large-scale latex extraction is draining the water table in Yunnan's southern foothills. “Already, streams are running dry,” Tang says. “Villages are being forced to move because they've lost their water supply.”

Also alarming, rubber trees in China and Laos are mainly drawn from a small pool of parent stock, which means that their genetic diversity is low. In the 1930s, Henry Ford created huge rubber plantations in the lower Amazon, part of H. brasiliensis's home range, only to have them wiped out by a fungal disease, South American leaf blight (Microcyclus ulei). Studies have indicated that Asian rubber plantations, including those in China and Laos, are vulnerable to leaf blight. In the age of jet travel, the fungus should eventually find its way to Asia. “An outbreak could effectively strip the forest bare,” says Horst Weyerhaeuser of the National Agriculture and Forestry Research Institute in Vientiane, Laos. “It would take years—maybe decades—to recover.”

Putting the brakes on H. brasiliensis will not be easy. Southern Yunnan has vastly profited from its production; in a 2006 XTBG study, rubber sales increased one typical township's income almost 10-fold between 1988 and 2003. The region will continue to benefit; some forecasters believe that by 2020, global demand for natural rubber will outpace supply by as much 1.4 million metric tons (MT). (Total 2008 consumption was 9.9 million MT, according to the International Rubber Study Group.)

Rubber in the Golden Triangle has been a classic standoff between economics and ecology: Monocultural plantations are so much more profitable than any other lawful agricultural system in these hills that they have inevitably prevailed, no matter the environmental cost. But in a forthcoming article, Nicholas Menzies of the Asia Institute at the University of California, Los Angeles, argues that at least some smallholders have found out how to both make a living and restore forests to a healthier state. Surprisingly, their way forward is a return to one of the region's most ancient products: tea. Not only that, they are growing tea in a way that until recently was derided as backward and inefficient—in the forest, under the canopy of larger trees.

“Tea forests aren't pristine tropical ecosystems,” Menzies says. “But they are far more diverse—and probably far more stable—than gigantic monocultures of rubber.”

## Rubber bandwagon

Although synthetic rubber has existed since World War I, natural rubber is superior—and much cheaper—for high-stress purposes. Only natural rubber can be steam-cleaned in a medical sterilizer, then thrust into a freezer—and still adhere flexibly to glass and steel. Jet and truck tires are almost entirely natural rubber. Militaries are major rubber consumers—which is why the United States imposed a rubber blockade on China during the Korean War. The blockade helped persuade the Chinese to prioritize growing rubber in 1951. Among the few areas in the nation warm enough for this tropical species is Xishuangbanna prefecture, at the southern tip of Yunnan.

Xishuangbanna prefecture has long been China's most biologically diverse area. Although it comprises just 0.2% of the nation's landmass, it contains 25% of its higher plant species, 36% of its birds, and 22% of its mammals, report biologists at Yunnan's Kunming Institute of Ecology. In the 1950s and 1960s, the People's Liberation Army turned this richly forested area into a rubber haven. The plantations became, in effect, army bases; labor was provided by more than 100,000 workers, many of them urban students. As the Cultural Revolution ground on, student workers were awakened every day at 3:00 a.m. and sent to clear the forest, one recalled to anthropologist Judith Shapiro, author of Mao's War Against Nature: “Every day, we cut until 7:00 or 8:00 a.m., then ate a breakfast of rice gruel sent by the [Yunnan Army] Corps kitchen. We recited and studied Chairman Mao's ‘Three Articles’ and struggled against capitalism and revisionism. Then it was back to work until lunch break, then more work until 6:00. After we washed and ate, there were more hours of study and criticism meetings.”

Scoffing at botanists' admonitions as counterrevolutionary, the youths repeatedly planted rubber trees at altitudes where the trees were killed by storms and frost. Then they planted them again in the same locations—socialism would master nature, they insisted. The frenzy laid waste to parks, exacerbated erosion, and destroyed streams. But it didn't yield much rubber.

When China began its economic reforms in the late 1970s, the educated young people fled back to their home cities, precipitating a severe rural labor shortage. Yunnanese villagers were finally permitted to establish rubber farms. Between 1976 and 2003, the area devoted to rubber expanded 10-fold, shrinking tropical montane forest in that time from 50.8% of the prefecture to 10.3% as Xishuangbanna planters learned how to adapt to hostile conditions at the edge of rubber's geographic range. According to Hu Zhouyong of the Tropical Crops Research Institute in Jinghong, the prefecture capital, the relatively cold climate forces them to select for exceptionally robust trees. “Xishuangbanna is ahead of everywhere else in the world in terms of productivity,” Hu says.

Even as China became the world's biggest rubber consumer, its rubber producers were running out of space. They began to eye Laos, which has about 6 million people in an area the size of the United Kingdom. A few villages in northern Laos had begun planting on their own as early as 1994. But the real push didn't begin until the end of the decade, when China announced its “Go Out” (zou chu qu) strategy, which pushed Chinese companies to invest abroad. Beijing had already changed the old military farms into private enterprises—corporations with abundant clout. As part of Go Out, the central government announced that it would treat rubber-growing in Laos and Myanmar as an opium-replacement program, making the former Yunnan military farms eligible for subsidies: up to 80% of initial costs for companies to grow rubber across the border, as well as the interest on loans and exemption from most tariffs for imported rubber.

Companies and smallholders flooded across the border. Most northern Laotian villagers live in hamlets without electricity or running water; schools and hospitals are a distant dream. Seeing a chance to improve their material conditions, villagers jumped on the rubber bandwagon, cutting deals with Chinese companies and farms. “In China, they were as poor as us,” the village head of Ban Namma told Science. “Now they are rich—they have motorcycles and cars—because they planted rubber. We want to have the same.”

Partly backed by funds for opium eradication, Chinese investors provided Lao farmers with seeds, fertilizer, and training for growing cash crops. “The most important by far was rubber,” says Jefferson Fox of the East-West Center in Honolulu, who is working with colleagues to evaluate rubber's impact in Southeast Asia. “It just exploded.”

According to anthropologist Yayoi Fujita of the University of Chicago in Illinois, in 2003 rubber covered about 1 square kilometer in Laos, all near the border. It covered 45 km2 in 2006. The Laotian government has estimated that by 2010, rubber will cover 1800 km2 of the nation. Although the global economic crisis has slowed the pace of clearing, most researchers believe it will accelerate in the long run—along with the ill effects of that clearing.

## Rubber, hit the road?

Although the Golden Triangle receives as much as 254 cm of rain a year, three-quarters of it falls between May and October. The rest of the year the forest survives largely on dew from morning fog. “Back in the 1980s and 1990s, there was still fog at lunchtime,” says XTBG ecologist Tang. “Now it's gone by 11:00”—a symptom, he says, of a profoundly altered hydrological regime.

Tang and others blame rubber. In 2006, XTBG ecologist Wu Zhaolu and colleagues from Yunnan University and Xishuangbanna National Nature Reserve showed that converting tropical forest to rubber plantations increased surface water runoff by a factor of 3—which in turn jacked up soil erosion by a remarkable factor of 45.

The greatest effect may be underground. H. brasiliensis usually sheds its leaves in February, and new leaves begin budding in late March, at the peak of the dry season. The leaf loss means that the forest has lower albedo and fewer surfaces to retain dew. To propel growth, according to a 2007 study, the roots suck water from 1 to 2 meters below the surface. Tapping begins as new leaves appear and continues until they fall. To replace lost latex, the roots suck up still more water—annually, roughly 5000 kilograms per hectare, according to XTBG estimates. The result is to lower the region's water table. Rubber, Wu and colleagues argue, is making the forest both gather less moisture from the air and lose it—and soil nutrients—more quickly.

Beginning to heed ecologists' worries, Xishuangbanna effectively banned new rubber planting in 2006 by freezing all land rotation. The scheme is unlikely to have much effect, Shi notes, as it seems to violate China's newly reformed land laws. But even if Xishuangbanna farmers were to stop planting H. brasiliensis tomorrow, its area would keep rising as rubber trees invade remaining forest (Science, 21 March 2008, p. 1604).

Rubber's economic benefits may offset the ecological risks—but not if leaf blight arrives. Because rubber trees are grafted from high-yielding specimens, the great majority of plantation trees are clones. But as the area of rubber increases, it becomes an increasingly inviting target for M. ulei. For a century, isolation has spared rubber plantations, but a recently opened highway now links Singapore and Kunming, Yunnan's capital. If and when M. ulei appears, this corridor will provide transportation.

In the rubber tree's native Brazil, diversity protects the species from leaf blight; H. brasiliensis is usually widely dispersed throughout the forest, surrounded by other tree species. Ecologists have long argued that the best way to protect rubber plantations is to situate them in a larger, more diverse forest. That would require persuading some rubber planters to turn to something else.

Menzies argues for an alternative to monoculture plantations as a cash crop in these areas. In a forthcoming article in The Social Life of Forests, the proceedings of a May 2008 restoration conference, Menzies notes that Xishuangbanna was until recently best known for “big-leaf” tea (Camellia sinensis var. assamica), of which the most famous variety is pu-er. When the Communist era began, farmers were pushed into “people's communes” and instructed to create massive monocultures of short, clipped bushlike tea plants on terraces. Traditional methods were dismissed as primitive and backward.

Beginning in the 1990s, Menzies says, upscale tea fanciers, lured by tales of old-style pu-er tea, have sought out the remnants of ancient plantations of tea trees. Farmers have rebuilt them, creating a new market for what high-end merchant Peet's sells as “ancient trees organic pu-erh.” The price of shade-grown tea rose from about 20 yuan per kg in the early 1990s to 1200 yuan (about \$175) per kg in 2007, Menzies says.

Because forest tea is grown beneath a variety of other trees, it promotes more tree diversity than a typical tea or rubber plantation. And because it grows relatively tall, it can tolerate more understory plants than a monoculture, which is generally kept as free as possible of other plants.

Neither the central government nor Yunnan authorities have endorsed shade-grown tea as an alternative to monoculture plantations. Moreover, Menzies cautions, C. assamica cannot grow at very low altitudes, which means that tea could not be readily substituted for rubber across the lowlands. But in his view, “the rise of green markets and niche markets shows that traditional methods can economically compete with intensive, high-volume production.” Tea is not a perfect replacement for rubber, Tang agrees. Still, he says, “every tea plant in the forest is one less rubber tree.”

• * With reporting by Josh D'Aluisio-Guerrieri.