News this Week

Science  23 Mar 2007:
Vol. 315, Issue 5819, pp. 1646
1. 2008 BUDGET

Senators Offer Sympathetic Ear to Complaints on NIH's Fiscal Slide

1. Jennifer Couzin

Two powerful champions of biomedical research blasted the White House's proposal to cut funding for the National Institutes of Health (NIH) in 2008 and invited research leaders to vent their own frustrations at a Senate hearing this week. Senators Tom Harkin (D-IA) and Arlen Specter (R-PA), who head the subcommittee that handles NIH funding, grilled NIH Director Elias Zerhouni on 19 March about the impact of what would be the fifth consecutive year of subinflationary budgets for NIH. They heard senior scientists describe a bleak research climate in which the percentage of funded NIH grant applications has dropped from 30% to 20%. And the senators promised to press for more money for biomedical research in 2008.

None of this was unexpected; Harkin and Specter helped win NIH a 2% increase in 2007 that the White House didn't request (Science, 23 February, p. 1062). More surprising was an impassioned speech by Zerhouni about the need for federally funded human embryonic stem cell research. In response to a question, he diverged from Administration policy, asserting that statements that adult stem cells can perform the same tasks as the embryonic variety “do not hold scientific water.” He added that any attempt “to sideline NIH on an issue of such importance is shortsighted.”

President George W. Bush has ruled that only stem cell lines created before 9 August 2001 may be used in federally funded research. However, Zerhouni said, “it's very clear … that these cell lines will not be sufficient to do all the research that we need to do.” The Administration has declined to fund new human embryonic stem cell lines and said little about the extent to which federal dollars should back this research.

Zerhouni's call for boosting funding for human embryonic stem cell research comes as NIH is still struggling to apportion its $28.9 billion 2007 budget, enacted 5 weeks ago. Other agencies are struggling, too; the president's 2008 budget proposes “pretty much flat-funding everything other than defense,” says Jon Retzlaff, director of legislative relations for the Federation of American Societies for Experimental Biology in Bethesda, Maryland. Some agencies that fund research in the physical sciences, including the National Science Foundation, would, however, receive real increases under the Administration's budget proposals (Science, 9 February, p. 750). Ironically, even though Congress gave NIH a small increase in 2007, the agency is under a particular strain because it is coming off flush times in 1998 to 2003, when it saw its budget double. That prompted many universities to expand, construct new facilities, and recruit new investigators, says Retzlaff. Between 1998 and 2007, the number of standard investigator-initiated (R01) grants roughly doubled to about 50,000. Such expansion requires long-term commitments, researchers said, because the agency provides most researchers with at least a portion of their salary and covers overhead costs. “We bought in” to the doubling, “and now we're getting cut,” says Joan Brugge, chair of the cell biology department at Harvard Medical School, in an interview before she testified at the hearing. Brugge, who began to study cancer in college after her sister was diagnosed with a fatal brain tumor, says that the slowdown is especially frustrating given recent advances in understanding the basic biology of cancer. “This not only forestalls progress but creates an atmosphere of uncertainty and anxiety,” she told the senators. Scientists also spoke of undergraduates and graduate students turning away from biomedical research and senior investigators leaving the field after being unable to secure NIH funding. Robert Siliciano, who studies HIV at Johns Hopkins University in Baltimore, Maryland, said he used to spend 30% of his time applying for grants. Now, he told the senators, it's jumped to 60%. Universities are already mobilizing to lobby for more funding. Immediately after the Senate hearing, a coalition of nine institutions and 20 scientists, including the four who testified, released a glossy, 21-page report that describes recent strides in cancer, spinal cord injury, and other diseases, arguing that NIH grants are well spent and lamenting the effects of flat funding. But beyond the anecdotes, the researchers and university administrators offered up few hard figures on the harm flat budgets are causing. “I know this will not work out to be a mathematical formula,” said Specter, but he professed frustration at a lack of data that he might offer his more skeptical Senate colleagues. “What's going to happen to NIH if the budget is cut by$500 million?” he wanted to know. “It would be very helpful to know how many research projects you are undertaking and how many you're turning away.”

The hearing was the Senate's opening move in its consideration of NIH's 2008 budget, a process that is expected to take at least until the fall.

2. MATHEMATICS

Mapping the 248-Fold Way

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

For more than a century, mathematicians have struggled to comprehend a vast, 248-dimensional entity, known to them only as E8. They have described it as “magic” and “miraculous,” but until now, they could not really understand how it is put together.

This week, an international team of 18 mathematicians and computer scientists called the Atlas Project, headed by Jeffrey Adams of the University of Maryland, College Park, announced that a supercomputer called Sage has successfully “mapped” E8. In the words of Gregg Zuckerman, a mathematician at Yale University, “never before has the marriage of pure mathematics and supercomputing produced such a prized offspring.”

Launched in 2002, the Atlas Project has roots deep in mathematical history. The ancient Greeks were fascinated by crystals and polyhedra because of their rich symmetry. In a crystal, however, symmetry comes in discrete chunks and limited numbers. In the late 19th century, the Norwegian mathematician Sophus Lie (pronounced “lee”) started studying objects with smooth rotational symmetries. Such objects are rare in three-dimensional space: spheres, cylinders, doughnuts, and cones. But in higher dimensions, they are much more common, and their symmetries are expressed by Lie groups.

Even the simplest Lie group, the rotations of a sphere, has profound scientific importance. This group, called SO(3) or A1, controls the shape of electron orbitals. The next interesting Lie group, SU(3) or A2, describes the symmetries of quarks. Murray Gell-Mann nicknamed it the “Eightfold Way,” because it is an eight-dimensional group. Physicists' quest for grand unified theories has led to ever-larger groups—and E8 is the biggest, most exotically symmetric of them all. Its mathematical richness makes it a magnet for string theorists, among others. “There's a philosophical question of why nature would pick one group over another,” says John Baez, a mathematical physicist at the University of California, Riverside. “E8 is so awesome that nobody could quarrel with this choice.”

The German mathematician Wilhelm Killing posited the existence of E8 in 1887, in a paper that broke all possible Lie groups down into four infinite families (labeled A through D) plus five “exceptional groups” (G2, F4, E6, E7, and E8). His French colleague Elie Cartan described E8 in 1894. In essence, Killing and Cartan together supplied the taxonomy of Lie groups. The Atlas Project aimed to compute their genomes.

In mathematical terms, the team set out to map each group's “irreducible representations”: the set of different n-dimensional spaces on which the group's rotations can act. Earlier mathematicians, including Zuckerman (who was Adams's thesis adviser), had proved that the representations can be divided into families, each generated by its own formula. In theory, it should be possible to use the formulas to crank out irreducible representations of every group—including E8—in a finite time. But no one knew how long the computation might take, or even if it was feasible.

One person made it feasible. Fokko du Cloux was a Belgian mathematician and computer scientist with a gift for turning the abstract, and sometimes flawed, theorems of group theorists into working algorithms. “Mathematicians often take shortcuts because they know what has to be true, but the computer doesn't,” says David Vogan of the Massachusetts Institute of Technology, one of the participants in the project. “Fokko went back and found all the details that weren't quite right and made it all perfect.”

Du Cloux calculated that E8 has 453,060 families. (Vogan had expected about a billion.) That meant the “genome” of E8 would be a table with 453,060 rows and 453,060 columns, or more than 200 billion entries. Du Cloux conquered all the groups through E7, but he didn't have a big enough computer to touch E8.

Then disaster struck. In November 2005, du Cloux was diagnosed with amyotrophic lateral sclerosis. By May, he was bedridden and could breathe only with a respirator. “But he was completely engaged,” says Adams. “He would lie on his back in Lyons, with a video projector pointing at the ceiling. I would type here in Baltimore, and he would see on the ceiling what I was typing. We would use Skype to talk. All he wanted to talk about was mathematics—he never complained about this terrible thing that happened to him.”

Du Cloux finally streamlined the software enough that a supercomputer might be able to do the calculation. But he did not live to see its completion. He died on 10 November 2006.

On 8 January, Sage, a supercomputer at the University of Washington, computed the last entry in the table for E8. But the Atlas Project is not finished. Like the Human Genome Project, it has produced far too much data for mathematicians to assimilate overnight. Also, it includes a lot of representations that aren't “unitary,” which are the equivalent of junk DNA in the human genome. Nevertheless, Adams expects the work to have a practical impact very soon, especially for number theorists. “The typical thing is that I get a call from a number theorist who says ‘I have such and such a representation. Can you tell me if it is unitary?’” says Adams. “Until now, that's been a very painful procedure to figure out.” With the atlas, it could become a simple lookup. “If there were such an atlas, I'd buy it immediately,” says Peter Sarnak, a number theorist at Princeton University.

Mathematicians and physicists look likely to mine the database for years to come. Hermann Nicolai, a theoretical physicist at the University of Potsdam in Germany, says the fingerprints of E8 can be found all over heterotic string theory, the most popular version of quantum gravity. “If you ask me if this will be helpful tomorrow, I cannot say yet,” says Nicolai. “In the end, I think the symmetry of quantum gravity might be realized in a more subtle way than we understand yet. In that event, it will be very useful to have a guide or atlas.”

3. OPEN GOVERNMENT

House OKs Whistleblower Bill

1. Eli Kintisch

The U.S. House of Representatives has broadened protection for government scientists who claim that their bosses have undermined the scientific process or suppressed information. Last week, legislators passed The Whistleblower Protection Enhancement Act by a margin wide enough—331 to 94—to withstand a promised veto from President George W. Bush. But the measure faces an uphill road in the Senate.

The bill, H.R. 985, covers incidents involving the “dissemination of false or misleading” scientific information or actions that compromise “the validity” of federal research. It defines those actions as an “abuse of authority.” The legislation, which would also apply to instances in which government scientists are prevented from publishing data, includes provisions that open-government advocates say will give whistleblowers a better chance to prevail in federal court and to avoid retribution.

“It's very important,” says David Ross, a former Food and Drug Administration (FDA) staffer whose accusations that the agency ignored data on liver damage in patients before approving the antibiotic Ketek in 2004 have spawned a congressional investigation. Ross says the bill is needed because current protections are “worse than useless” at regulatory agencies such as the FDA and do not cover scientific disputes.

But others feel the bill would force the courts to address scientific questions outside their expertise. “If an agency or the Administration disagrees with the findings of a particular scientist, we should not be opening up our judicial system for those disagreements to be litigated as federal employee personnel issues,” said Representative Bill Sali (R-ID) during an unsuccessful effort on the floor to scrap the science provisions. Science policy expert Roger Pielke of the University of Colorado, Boulder, wonders how the provisions would be enforced, positing a situation in which an incorrect weather forecast could be labeled an abuse of authority because it contained false information.

A lobbyist for the Union of Concerned Scientists, which supports the legislation, says that such arguments could hold sway in the Senate, which presents a “much more challenging” environment.

4. MICROBIOLOGY

New Bacterial Defense Against Phage Invaders Identified

1. Jean Marx

Humans are not alone in having to fend off pathogens; even the simplest organisms are under a constant threat of invasion. Bacteria, for example, are awash in a sea of viruses known as bacteriophages. “Every 2 days, half the bacteria on Earth are killed [by bacteriophages],” says phage expert Vincent Fischetti of Rockefeller University in New York City. “It's a constant battle.” Researchers have now identified a new defense mechanism that helps bacteria hold their own in this battle.

On page 1709, a team led by Philippe Horvath and Rodolphe Barrangou of Danisco, a Danish company that produces bacterial cultures and other materials for the food-processing industry, reports that bacteria use a system, apparently akin to the RNA interference (RNAi) system of higher organisms, to block phage reproduction, thus making them resistant to infection.

The work could help the food and biotechnology industries, which use bacterial cultures to make products such as cheese and yogurt as well as proteins for human medicine. These industries, Fischetti says, “have a terrible problem with phage” ruining their cultures and could benefit from better phage-resistant bacterial strains.

The Danisco team's work also provides the first biological evidence for a function of so-called CRISPR sequences, which were identified in 2002 by Leo Schouls of the National Institute of Public Health in Bilthoven, the Netherlands, and his colleagues. These sequences, formally known by the descriptive name of “clustered regularly interspaced short palindromic repeats” because of the way they are arranged in the genome, are widely distributed in the genomes of both Bacteria and Archaea.

Accompanying the CRISPR sequences are a suite of perhaps four to 10 cas (CRISPR-associated) genes. Researchers have made a number of proposals about what these genes might do. For example, Eugene Koonin and Kira Makarova of the U.S. National Center for Biotechnology Information in Bethesda, Maryland, and their colleagues analyzed the cassequences and, based on those structures, suggested in 2002 that they might encode a new DNA repair system. But more recently, Koonin says, another idea emerged as several groups found that the spacer sequences within CRISPR regions resemble those of sequences in phage and also in plasmids, small extrachromosomal pieces of DNA that can be transmitted between bacterial species.

In a second analysis, published by Biology Direct on 16 March of last year, Makarova, Koonin, and colleagues proposed that the Cas proteins and CRISPR spacer sequences, which were presumably picked up by the bacteria during prior phage infections, together constitute a bacterial immune system that works by a mechanism similar to that of RNAi in higher organisms. The idea is that the spacers make short RNA sequences that can bind to complementary sequences in messenger RNAs made by invading phages. This would block their translation into proteins and mark them for degradation by Cas proteins, some of which resemble those known to be involved in RNAi.

The Danisco group has now provided direct evidence for that hypothesis. Working with the bacterium Streptococcus thermophilus, which is widely used to make yogurt and cheese, the researchers found that infection of the bacteria with phage leads to incorporation of phage-related spacer sequences within a CRISPR region. Such bacteria became resistant to further infection by the phage strains that contributed those sequences. But “if you take the spacers out, the resistance is lost,” says Horvath, who works at Danisco's lab in Dangé-Saint-Romain, France. The team also showed that at least one cas gene, which encodes a possible RNA-dicing nuclease, is necessary for the phage resistance. This shows, Fischetti says, that bacteria have “a very neat mechanism by which they are able to keep bacteriophage under control.”

Dennis Romero, a member of the Danisco team at the company's lab in Madison, Wisconsin, says that the CRISPR system may have a wider function as well. “In addition to matching phage, the spacers also match chromosomal and plasmid sequences,” he notes, and thus they might help control normal bacterial gene activity.

Whether or not that is the case, the findings open the door to using the CRISPR system to block specific gene activity in bacteria, just as RNAi is used in higher organisms. And then there is the possibility of producing more phage-resistant bacterial strains for industrial use. This could be accomplished by genetically engineering bacteria with appropriate CRISPR spacer sequences; Horvath says, however, that “Danisco has no plans to do that in light of consumer concerns about the use of GMO [genetically modified organisms], particularly in Europe.”

The researchers plan instead to simply expose bacteria to various phage strains and then select for those that are resistant. They can, however, use their knowledge of the CRISPR spacers to help screen for bacteria that carry the right spacers to confer the resistance they want. “Although we can genetically engineer,” Romero says, “we found that nature can do the work for us.”

5. GEOLOGY

A Trace of the Earliest Plate Tectonics Turns Up in Greenland

1. Richard A. Kerr

Geologists have discovered the earliest known remnants—by billions of years—of plate tectonics, the large-scale movement of Earth's crust. The rocks are preserved in plain sight among the intensely studied ancient rocks of southwest Greenland, a group of geologists reports on page 1704. These days, hot new sea floor forms from magma at mid-ocean ridges, spreads away as it cools, and eventually dives back into the deep interior. In its early days, Earth was still so hot throughout that researchers have wondered whether the planet might have been ridding itself of heat by some entirely different means. But the new discovery “indicates there was a modern-day plate tectonics operating shortly after formation of Earth,” says geologist Yildirim Dilek of Miami University in Oxford, Ohio.

Innumerable geologists have walked and flown over southwest Greenland's 12-kilometer-long stretch of baked, twisted, and tortured rock known as the Isua supracrustal belt. Dating from Earth's early adolescence 3.8 billion years ago, the Isua rocks hold clues to how the young planet worked, back when life might have gotten started. In fact, it was the search for microscopic signs of early life that brought geologist Harald Furnes of the University of Bergen, Norway, and colleagues to Isua in 2006. Furnes had long studied much younger scraps of ocean crust that had become stranded on land, called ophiolites, but that day he was looking for sea-floor lavas that might hold traces of ancient microbial borings.

Then Furnes and his colleagues came upon the sheeted dikes. These banded rocks are the hallmark of ophiolites and thus of sea-floor spreading. Built like a stack of cards, they are composed entirely of the thin sheets of once-molten rock injected into the crests of mid-ocean ridges as the newly formed plates spread away from the ridge. The Isua sheeted dikes are near previously identified components of ophiolites: distinctive “pillow” lavas extruded on the sea floor from underlying dikes, rock that solidified in magma chambers that fed the dikes, and never-melted mantle rock below that. “The major components [of an ophiolite] appear to be all there,” says geologist Kent Condie of the New Mexico Institute of Mining and Technology in Socorro. “I'm convinced.”

So Earth had sea-floor spreading almost 2 billion years earlier than previously known. What about the other end of the tectonic process? Today, old sea floor dives steeply into the deep interior on top of a relatively cold, rigid slab of tectonic plate, a process called subduction. But some geophysicists had suspected that old ocean plates might once have recycled themselves differently—say, by sinking straight into a hot magma mush as if it were quicksand.

Furnes thinks, but can't prove, that something like modern-day subduction was going on 3.8 billion years ago. Rocks adjacent to the Isua ophiolite geochemically resemble rocks called boninites. These rocks are cooked up only beneath island chains perched over subduction zones like those in today's western Pacific. If Isua has bona fide boninites, a magma mush would not work.

All the Isua rocks come from “a pretty well established subduction zone similar to what we have today,” concludes Dilek. “They're hard to explain in any other way.” Condie can't quite agree. “I don't think it's 100% definitive,” he says. “There's just enough ambiguity that it may or may not” have been entirely modern subduction. Enough ambiguity that Isua geologists will be heading back to the field with new eyes this summer.

Having a Blast, Wish You Were Here

The Large Hadron Collider at CERN will smash particles at unprecedented energy and may open new realms of discovery. It will secure Europe's ascendancy in particle physics for years to come

NEAR GENEVA, SWITZERLAND—Measuring 15 meters across and weighing 13,010 metric tons, the enormous disk of machinery dangles from bundles of cables like a gigantic yo-yo. Sectioned like an orange and festooned with electrical cables, the contraption could be mistaken for a flying saucer hoisted on edge. In fact, it's part of a huge barrel-shaped particle detector, the Compact Muon Solenoid (CMS), that will soon snare bits of matter from the new highest-energy particle smasher, the Large Hadron Collider (LHC) here at the European particle physics laboratory, CERN.

The colossus hovers a few centimeters above the concrete floor of a cavernous subterranean hall. All day, workers have lowered it down a shaft barely wide enough to take it. The 100-meter journey strains the nerves, says Hubert Gerwig, an engineer at CERN.

Now that it's almost over, Gerwig can relax a little. “Want to see it move?” he asks Archana Sharma, a physicist at CERN. Gerwig pushes the wall of metal. “If you get a feel for the resonant frequency, you can excite it” to oscillate, he says. Sure enough, the giant stirs. “Okay, okay, it's moving!” shouts Sharma, as millions of dollars' worth of delicate equipment sways ever so slightly across the grain of the concrete.

Gerwig isn't the only one here who is a little giddy with nervous excitement. In a few months, CERN researchers will have completed the 27-kilometer-long LHC, and in November, they hope to put the largest and most complex experimental device ever built through its warm-up laps. Smashing particles at energies seven times higher than the previous record, the LHC should blast out the one bit of matter missing from physicists' theory of the known particles. It could also spit out a slew of other particles and open a new era of discovery. The LHC will make CERN the world's center for particle physics.

Considering its size and technological complexity, the LHC “is the modern equivalent of the pyramids,” says Peter Limon of Fermi National Accelerator Laboratory (Fermilab) in Batavia, Illinois, who is working on the machine. But the LHC is more than a technological marvel. It embodies a broader movement in particle physics. For decades, the United States paced the field. Now, as the LHC eclipses Fermilab's Tevatron collider, Europe takes the lead. “It is certainly true that the center of gravity of physics has moved to CERN,” says Hans-Ake Gustafsson, an experimenter from Lund University in Sweden. “And the U.S. recognizes that because it's investing a lot of money in the experiments here.”

No one knows what the LHC will find. But this much is clear: The LHC has already created a revolution in particle physics.

LHC or -EST

It is difficult to describe the LHC without resorting to superlatives. Not only will the LHC smash particles at the highest energies, but it will also feed the largest and most complex particle detectors ever built for a collider. They will pump out the greatest torrent of data; in a year, each could fill a stack of DVDs 25 kilometers high. The LHC will consume a record 120 megawatts of power, enough to sustain every household in the canton of Geneva. At a cost of 4.7 billion Swiss francs ($3.8 billion), it's the most expensive collider ever built. The United States is chipping in$531 million, mostly for detectors.

Numbers alone cannot convey the immensity of the project, however. Step into the hall housing the ATLAS detector, and you find yourself face to face with a machine eight stories tall and half as long as a soccer field. The thing could fill the nave of a cathedral, but instead of the Holy Spirit, it's packed with particle trackers, light-emitting crystals, and enough other gizmos to fill 100 million data channels. And it's as precise as it is big, says CERN's Peter Jenni, spokesperson for the 1800-member ATLAS collaboration. It can measure the curving path of a particle called a muon to within 40 micrometers, half the width of a human hair.

Circling below the French countryside between Lake Geneva to the east and the Jura Mountains to the west, the accelerator itself looks a bit like a glorified sewer pipe. Visitors to the LHC's otherworldly tunnel must carry oxygen packs in case the machine's cryogenic system leaks suffocating helium; workers on bikes sneak up on the inattentive. The LHC lies along one side of the gently curving tunnel, an endless line of big blue cylinders connected end to end like sausages. These are the revolutionary magnets that steer the beams around the ring. Twice as strong as those at Fermilab's Tevatron, they in fact house two accelerators carrying protons in opposite directions.

The brawny collider aims foremost to discover one thing: a long-sought particle called the Higgs boson. The Higgs would complete the so-called standard model of the known particles, says Jonathan Ellis, a theorist at CERN. “You could consider the Higgs boson the period on the end of the standard-model sentence,” he says. But physicists hope the standard model is not the final word and that LHC will blast out other particles and surprises (see p. 1657). “We had Stephen Hawking here, and he told us that he wasn't so sure we'd find the Higgs boson and that he was more interested in mini-black holes,” Jenni says. “People have different ideas.”

Mighty ATLAS and CMS will race for those breakthroughs at the energy frontier. “It's going to come down to who is better prepared and whose detector is more complete” when the LHC starts running, says Tejinder Virdee of Imperial College London, spokesperson for the 2359-member CMS collaboration. “They can confirm [our discoveries], that's allowed.” The LHC could discover the particles predicted by a concept called supersymmetry after running for just a year, Virdee says. But the LHC will also feed two specialized detectors to stake claims to leadership in other areas as well.

A detector called LHCb will study the asymmetries between particles containing elementary bits of matter called bottom quarks and their antimatter foils. Physicists at specialized colliders in the United States and Japan have studied the subtle differences as the bottoms decay to other “flavors” of quark, in hopes of finding hints of new particles (Science, 13 October 2006, p. 248). Even if ATLAS and CMS see those particles directly, “you want to study how things couple to these new particles,” says CERN's Tatsuya Nakada. “And that's what you can do with flavor physics.

Seven kilometers away, a detector named ALICE will study a soup of particles called a quark-gluon plasma. The ultrahot plasma filled the infant universe, and physicists at Brookhaven National Laboratory in Upton, New York, have recreated it by smashing gold nuclei with their Relativistic Heavy Ion Collider (RHIC). For a few weeks a year, the LHC will smash lead nuclei at energies 28 times higher, letting ALICE peer deeper into the fleeting plasma, says CERN's Jürgen Schukraft. “The things you can look at here you can't look at with RHIC even if you run it for 50 years,” he says.

But first, researchers must complete the collider. After a decade of construction, they are on schedule to finish this year, says CERN's Lyndon Evans, who leads the effort. (Researchers expect to lower the last magnet into the tunnel in mid-April.) A Welshman with a sonorous voice and silver hair, Evans has the phlegmatic demeanor of one who has dealt with crises large and small. On his computer he pulls up graph after graph of progress on the LHC's myriad subsystems. On each a rising red “just in time” line stands out. “It has some magical properties,” Evans says. “Things tend to bounce off it” to stay on schedule.

In spite of the dash to the finish, the push for the LHC has been a marathon. Physicists dreamt the collider up more than 20 years ago even as CERN built another machine, the Large Electron-Positron Collider (LEP), which ran from 1989 to 2000. In fact, they planned to reuse LEP's tunnel and feed the LHC with existing accelerators. “Without that, it would have been impossible to build the machine on a constant [lab] budget,” says CERN Director General Robert Aymar. At the time, physicists in the United States were planning the 87-kilometer-long Superconducting Super Collider (SSC). The LHC couldn't match the SSC's energy, but it could smash more particles, says CERN chief scientist Jos Engelen. In 1993, the U.S. Congress killed the uncompleted SSC, leaving the field open for CERN, which gave the LHC the green light the following year.

Even recycling as much as they could, researchers had to push the limits of technology, Evans says: “Society is willing to pay a certain amount and no more, and to make the LHC possible, we had to be innovative.” Researchers have designed the strongest magnets by far for an accelerator, crammed two accelerators into one set of magnets, used wires of high-temperature superconductor to distribute power, and pioneered a type of radiation-hard electronics for their detectors. They even chill the liquid helium that cools the magnets to an extra-frigid 1.9 kelvin to make it a free-flowing superfluid, which is also an outstanding heat conductor.

To be sure, the LHC has hit some potholes along the way. In 2001, a review showed that the project was running behind schedule and 20% over budget, forcing the lab to scale back other projects and refinance the LHC (Science, 28 June 2002, p. 2317). In 2004, problems emerged with the cryogenic lines that transport the liquid helium to the magnets. Workers had to rip out, repair, and reinstall 3 kilometers of line, creating an enormous backup of the magnets they'd been installing as soon as they arrived. “I'd imagined storage for 50 magnets, and in the end I had to find room for 1000,” says Evans, who scattered them all over the lab.

Now, about 2 years behind their original schedule, researchers see the light—or, more correctly, the other end of the accelerator—at the end of the tunnel. Soon workers will put down their wrenches and welding torches, and researchers will begin to bring the machine to life. “We are very excited,” Engelen says, “and a bit worried because now we have to deliver.”

See you in Switzerland

Already, physicists are flocking to CERN in anticipation. Some 7500 of 111 different nationalities have registered to work on the site, as the LHC lures talent away from other experiments, such as CDF and D0 at Fermilab. “When I was hired, I started to work on D0,” says Adam Yurkewicz, a postdoc at Stony Brook University in New York. “But I told them I wanted to switch to ATLAS because it's the forefront; it's the place for discovery.”

In fact, CERN feels a bit like a resort for the nerdy set. Motley buildings nestle along streets named for Einstein, Feynman, and other famous physicists. Here and there lies equipment awaiting assembly. In the evenings, friends meet in the cafeteria to chat over a beer, a Danish Carlsberg or Czech Budweiser. “There is a different atmosphere than in the U.S.,” says David Silvermyr, a Swede from Oak Ridge National Laboratory in Tennessee. “If you go to lunch here, people are talking about, ‘We're excited about this,'or ‘We're going to build that.’ In the U.S., people talk about budgets.”

But if the LHC is changing the map of particle physics, it also marks a leap in the field's evolution toward ever-bigger projects. Since the 1960s, experimental collaborations have grown to include dozens, then hundreds, and now thousands of scientists. That explosive expansion has led some particle physicists to seek more intimate environs in other fields (Science, 5 January, p. 56). But it doesn't faze those who have chosen to work at the LHC. “I'd still have a sense of satisfaction no matter what was discovered and how big a role I had in it,” Yurkewicz says. “I'd know that I contributed.”

Nevertheless, researchers recognize the challenge of rising from such a crowd to a position of leadership. “It is a very competitive environment,” says Rosy Nikolaidou of CEA Saclay, France, who works on ATLAS. “Each day, you have to prove that you are the best and that you deserve your chance.” Nektarios Benekos, an ATLAS member from the Max Planck Institute for Physics in Munich, Germany, says young researchers must think strategically to avoid, for example, being pigeonholed. “For sure, you must not stick too much to a particular subsystem, because you lose touch with the entire detector,” he says.

Those who cannot move to Europe face the challenge of keeping contact with experiments thousands of kilometers away. That's a big problem for American physicists, who make up 20% of the ATLAS team and 30% of the CMS team. To address it, physicists are relying in part on a high-capacity computing network called the Grid to transmit data to key labs in other countries. Those “analysis support centers” will serve as gathering places that bring the LHC a little closer to home, says Michael Tuts of Columbia University, who manages the U.S.'s ATLAS research program. “They're places where you can go and get the water-cooler conversation,” he says.

In fact, even as CERN draws people, the Grid should help extend the reach of particle physics across the globe, says Harvey Newman of the California Institute of Technology in Pasadena, who chairs the board that oversees the U.S.'s CMS team. “There are countries that weren't in the field in a serious way, and now they are there,” he says. For example, physicists in Pakistan, India, and Brazil will have access to the LHC data in their home countries.

Waking the giant

The full torrent of terabytes may be a while in coming, however. Researchers plan to send protons around the ring in November and begin taking data next spring. Even then they will start at low energy—less than half the Tevatron's—and with low beam intensity, or “luminosity.” “If we can get up to a tenth of design [luminosity at full energy] after the first year, I think that would be miraculous,” says Michael Lamont, an accelerator physicist at CERN. “And I think the experimenters would be quite happy with that.”

Researchers must go slow because the LHC is the first collider powerful enough to destroy itself. Each of the LHC's beams packs a staggering 362 megajoules of energy, the equivalent of 90 kilograms of TNT and enough to melt 500 kilograms of copper. Should the machine accidentally steer a beam into its own innards, the protons could drill a hole tens of meters long, potentially taking the LHC out of action for months.

To prevent such a calamity, accelerator physicists have gone to extremes to protect the LHC from itself. More than 4000 super-fast beam-loss monitors will sense protons spraying out of the beam. Independently, beam-current monitors will infer losses by measuring the amount of circulating charge. And beam-position monitors will sense when the beams stray from their proper course. These systems can trigger magnets that can safely kick the beams out of the machine in the few hundred microseconds it takes to make two or three revolutions, less time than it takes a wobbly beam to veer off course entirely. “For the LHC, we've tried from the start to cover all the different possible failure scenarios so that we don't have an accident,” says CERN's Rüdiger Schmidt.

Even if nothing goes wrong, physicists must take extraordinary steps just to make the LHC run. The collider is designed to pack 1014 protons into each beam, and if just one 10-millionth of them flew into a magnet, they would heat it enough to temporarily kill its superconductivity, triggering a beam dump. To avoid such “quenches,” researchers have installed hundreds of adjustable constrictions called collimators that will catch the inevitable wayward particles. “Out of every 1000 particles [headed toward the collimators], not more than one should escape to reach the magnets downstream,” says CERN's Ralph Assmann. “Without this system, the LHC cannot run.”

Those are just the technical challenges. When the time finally arrives to power up the machine, the main challenge will be managing the people, Lamont says. “There's going to be a lot of people standing in the control room, maybe not twiddling the knobs but looking over your shoulder, and that's as it should be; this is as exciting as it gets,” Lamont says. “But what you really want is four guys sitting behind closed doors quietly figuring out how to make it work.” Those four will have to cope with the crowd. Who could blame anyone for wanting to be there when the LHC ushers in a new era in particle physics?

Stability, International Character Honed CERN's Competitive Edge

The qualities that helped the lab make the LHC a reality could put it a step ahead in the race for the next great particle smasher

In the 1980s, physicists hammered out plans for a gargantuan particle smasher that would reveal the key bit of matter that would complete their theory of the known particles. The behemoth would also blast out scads of new particles and open new vistas of inner space. It would be the hub about which the world of particle physics would turn for decades.

Meanwhile, a few researchers at the European lab, CERN, near Geneva, Switzerland, mused of building a smaller machine on the cheap. They called it the Large Hadron Collider (LHC).

Two decades later, the LHC is about to chase the discoveries never made by that other machine, the infamous Superconducting Super Collider (SSC). Designed to reach energies three times higher than those of the LHC, the SSC died uncompleted in 1993 when its budget ballooned from $4.6 billion to more than$8.3 billion and the U.S. Congress killed it.

Why did the SSC fail and the LHC succeed? Physicists can point to many stumbling blocks that tripped up the SSC (Science, 3 October 2003, p. 38). Instead of building at an existing lab, officials chose a remote site in Waxahachie, Texas; researchers made a small but expensive design change; the United States tried to go it alone and sought international partners only belatedly. The LHC succeeded for reasons equally concrete—and those factors could give CERN the edge in the competition for the next gigantic collider, the proposed 31-kilometer-long straight-shot International Linear Collider (ILC).

All agree that CERN's rock-steady budget was a key to its success in building the LHC. In keeping with the treaty that created the lab in 1954, each of CERN's now 20 member nations supports the lab in proportion to its gross domestic product. “The treaty creates stability because the member countries recognize that this isn't something you vote up or down every year,” says CERN Director General Robert Aymar. The arrangement sets the lab budget 5 years in advance and even allows officials to borrow against future income, as they did in 2002 when they found that the LHC was running 20% over budget. In contrast, the SSC was far more vulnerable. In the United States, Congress funds the national labs year by year, which means lab budgets fluctuate and projects such as the SSC face the ax repeatedly.

When building the LHC, CERN also benefited from moving continuously from one collider to the next. CERN researchers began designing the LHC even as they built another machine, the Large Electron-Positron Collider (LEP), which ran from 1989 to 2000. By using LEP's tunnel to house the LHC and existing accelerators to feed it, CERN officials saved billions of Swiss francs and built the LHC without an increase in the lab's budget.

Continuity has also helped the lab accrue talented personnel. “The most important thing to making a project like this work is the quality of the people working under you,” says CERN's Lyndon Evans, who leads LHC construction. “One of our advantages [in maintaining staff] is that we came off another project, LEP, which wasn't so long ago.”

Even before the LHC is up and running, physicists around the world are looking toward the next big accelerator and are trying to draw some lessons from the contrasting fates of the LHC and the SSC. They say they will need the ILC to study in detail the new particles the LHC should spot (Science, 21 February 2003, p. 1171). Researchers in the United States, Japan, and Europe all want to build the machine close to home, and “the U.S. and Japan had better look up and humbly learn from CERN's history and experience,” says Nobu Toge of the Japanese accelerator laboratory KEK in Tsukuba. Still, Toge adds, “everyone has a long way to go to learn how to make the ILC a successful global project before jumping over each other to see who is to host it.”

That hasn't stopped early jockeying, however. Europeans point to CERN's success with the LHC and its explicitly international character as big plusses. “Very probably, CERN has an advantage over any other place to host an even-more-international effort than already exists,” says CERN's Aymar.

But Japanese and American physicists say they have advantages of their own. For example, CERN's unwavering budget could actually be a liability in the competition for the ILC. Although CERN's funding doesn't dip unpredictably, it also doesn't climb quickly, as all 20 member nations must agree on any increase. In contrast, the U.S. government can rapidly ramp up funding for projects. “The U.S. system is more dynamic and can react to things more quickly,” says Pier Oddone, director of Fermi National Accelerator Laboratory in Batavia, Illinois. That may be important for the ILC, which will probably cost between $10 billion and$15 billion, with the host picking up half the tab.

Physicists in North America and Asia also note that, although CERN is an international laboratory, it does not embrace all countries equally. CERN's 20 member nations enjoy a special status compared to “guest” nations such as Japan and the United States, and many question whether the treaty structure is flexible enough to accommodate a truly global project.

Timing may be key. CERN will be paying off the LHC until 2011, and after that, the lab plans to upgrade the machine to produce even more collisions. So CERN will have its hands full until the middle of the next decade. In contrast, the United States will have no colliders for particle physcis in operation after 2009. Japan has a smaller collider that it may upgrade and will soon be finishing a large proton accelerator complex. “If we plan to build the ILC in the 2020s, CERN is a good candidate,” says KEK's Mitsuaki Nozaki. “However, if we wish to start construction soon after the first physics results come out of the LHC around 2010, then the U.S. and Japan are the only realistic candidates.”

Of course, whether the ILC gets built at all depends on whether the LHC discovers anything worthy of further study.

Physicists' Nightmare Scenario: The Higgs and Nothing Else

Many fear the LHC will cough up only the one particle they've sought for decades. Some would rather see nothing new at all

Suppose you are a particle physicist. A score of nations has given you several billion Swiss francs to build a machine that will probe the origins of mass, that ineffable something that keeps an object in steady motion unless shoved by a force. Your proposed explanation of mass requires a new particle, cryptically dubbed the Higgs boson, that your machine aims to espy. When, after 2 decades of preparation, you get ready to switch on your rig, you would fear nothing more than the possibility that you were wrong and the particle doesn't exist, right? Not exactly.

Many particle physicists say their greatest fear is that their grand new machine—the Large Hadron Collider (LHC) under construction at the European particle physics laboratory, CERN, near Geneva, Switzerland—will spot the Higgs boson and nothing else. If so, particle physics could grind to halt, they say. In fact, if the LHC doesn't reveal a plethora of new particles in addition to the Higgs, many say they would rather it see nothing new at all.

That may seem perverse, but put yourself again in the shoes of a particle physicist. In the 1960s and 1970s, researchers hammered out a theory called the standard model that, in spite of leaving out gravity and suffering from other shortcomings, has explained everything seen in collider experiments ever since and left physicists with few clues to a deeper theory. At the energies the LHC will reach, the standard model goes haywire, spitting out negative probabilities and other nonsense. So the collider has to cough up something new, researchers say. If it spits out only the Higgs, however, the new golden age of discovery could end as soon as it begins.

If the lone Higgs has just the right mass—about 190 times the mass of a proton—it would tie up the standard model's loose ends and leave physicists even more thoroughly stymied than before, says Jonathan Ellis, a theorist at CERN. “This would be the real five-star disaster,” he says, “because that would mean there wouldn't need to be any new physics all the way up to the Planck scale,” the mind-bogglingly high energy at which gravity pulls as hard as the other forces of nature. The Higgs alone could essentially mark a dissatisfying end to the ages-long quest into the essence of matter.

If, on the other hand, the LHC sees no new particles at all, then the very rules of quantum mechanics and even Einstein's special theory of relativity must be wrong. “It would mean that everything we thought we knew about everything falls apart,” says Harvey Newman, an experimenter at the California Institute of Technology in Pasadena. That would thrill many but is so unlikely that it would be “essentially impossible” for the LHC to see nothing new, Newman says. Others agree.

Physicists have no similar guarantee that the LHC will reveal not only the Higgs but also exotic new particles that would point to new physics and open a new era of discovery. So the LHC is a gamble, and many are pulling for the more exciting long shots.

Quack like a Higgs

Easily the most famous particle not yet discovered, the Higgs has even been crowned the “God particle” by one Nobel laureate. In reality, however, it is merely an ad hoc solution to an abstruse problem in the standard model: how to give particles mass.

The particular challenge is to give mass to particles called the W and Z bosons, which convey the weak nuclear force. According to the standard model, the weak force that causes a type of radioactive decay and the electromagnetic force that powers lightning and laptop computers are two facets of the same single thing. The two forces aren't precisely interchangeable: Electromagnetic forces can stretch between the stars, whereas the weak force doesn't even reach across the atomic nucleus. That range difference arises because photons, the quantum particles that make up an electromagnetic field, have no mass. In contrast, the particles that make up the weak force field, the W and Z bosons, are about 86 and 97 times as massive as the proton.

Unfortunately, the persnickety standard model falls apart if theorists simply assign masses to the W, Z, and other particles. So the masses must somehow arise from interactions of the otherwise massless particles themselves. In the 1960s, Peter Higgs, a theorist at Edinburgh University in the U.K., realized that empty space might be filled with a field, a bit like an electric field, that could drag on particles to give them inertia, the essence of mass. The field would consist of a new particle, the Higgs boson, lurking “virtually” in the vacuum.

Nature appears to follow this scheme. Using it, theorists predicted the masses of the W and Z. And at CERN in 1983, the two particles weighed in just as expected, in collisions energetic enough to pop them out of the vacuum.

Now, mounds of data point to the Higgs. For example, the lifetime and other properties of the Z depend on the cloud of virtual particles flitting around it like flies swarming a rotten ham sandwich. Precise studies of the Z suggest that a Higgs at most 200 times as hefty as the proton lurks in that cloud. Comparing the masses of the W and a particle called the top quark shows a similar thing, says Gordon Kane, a theorist at the University of Michigan, Ann Arbor. “These are two completely independent pieces of evidence that there is something that walks and talks and quacks like a Higgs,” Kane says. “The existence of the Higgs in the LHC range is essentially certain.”

Discovering the Higgs would complete the standard model. But finding only the Higgs would give physicists little to go on in their quest to answer deeper questions, such as whether the four forces of nature are somehow different aspects of the same thing, says Aldo Deandrea, a theorist at the University of Lyon I in France. “If you have just a Higgs that is consistent with the standard model, then you probably don't know what to do next,” he says. “What then?”

Good taste and extra dimensions

Most researchers say they'll never face that question because the LHC will discover plenty of other things. Many expect it to blast out particles predicted by a concept called supersymmetry (SUSY), which posits a heavier “superpartner” for every known particle. That may seem unduly complicated, but SUSY solves problems within the standard model, points toward a deeper theory, and may even explain the mysterious dark matter whose gravity holds the galaxies together. “SUSY is unique in that it does all these things automatically,” CERN's Ellis says.

Most concretely, SUSY solves a technical problem caused by the Higgs boson itself. The Higgs, too, must be shrouded in virtual particles, and they ought to send its mass skyrocketing. SUSY would explain why the Higgs is as light as it appears to be, because mathematically the effects of partner and superpartner on the Higgs mass tend to cancel each other. SUSY would also help explain the origin of the Higgs, which is just tacked onto the standard model but emerges naturally from the structure of SUSY.

SUSY could also help unify the four forces. The standard model accounts for three of them: the electromagnetic force, the weak force, and the strong nuclear force that binds particles called quarks into protons, neutrons, and other particles. The strengths of the three increase with the energy of collisions, and if the universe is supersymmetric, then all begin to tug equally hard at precisely the same energy somewhere below the Planck scale. That should make it easier to roll them and gravity together in one grand unified theory, says Frank Wilczek, a theorist at the Massachusetts Institute of Technology in Cambridge.

SUSY might even provide the dark matter that glues the galaxies together. Physicists believe that dark matter must consist of some stable particle that barely interacts with normal matter, and the least massive superpartner might just fit the bill. With all this evidence supporting it, SUSY is almost too beautiful to be wrong, some theorists say. “All these clues could be misleading,” Wilczek says, “but that would be a really cruel joke by Mother Nature—and in really bad taste on her part.”

The LHC might also reveal far wilder phenomena, such as inner parts to electrons and other supposedly indivisible bits of matter, tiny black holes, or even new dimensions of space that open only at very high energies. The spare room could explain, for example, why gravity is so much weaker than the other forces. “Something like extra dimensions I give a very small probability,” says Michael Tuts, a physicist at Columbia University. “But the potential is so big that it's very exciting.”

A sure bet

None of these more exotic possibilities is guaranteed. And particle physicists say that just discovering the Higgs would be a triumph. “If the Higgs is anything like theorists predict, we will find it,” says Peter Jenni, an experimenter at CERN. “We shouldn't be disappointed if we do.”

Physicists also admit that, regardless of the intellectual foment it would cause, finding nothing would create problems, at least with the governments that paid for the LHC. “Just imagine if we go to the CERN Council and say, ‘Thank you very much, we've just spent billions of Swiss francs, and there's nothing there,’” Ellis says. “I think they might be a tad disappointed.”

However, finding only the Higgs may make life nearly as difficult for physicists trying to persuade governments to build the next great particle smasher, the proposed International Linear Collider (ILC). Costing between $10 billion and$15 billion, the ILC would map out the conceptual terrain opened by the LHC (Science, 9 February, p. 746). By colliding indivisible electrons and positrons, the ILC would generate cleaner collisions that should reveal details of new particles that will be obscured by the messy proton-on-proton collision at LHC.

But if the ILC has only the Higgs to study, then it becomes “a very hard sell both scientifically and politically,” says David Cinabro, a particle-physicist-turned-astronomer at Wayne State University in Detroit, Michigan. “I think you'll have a really hard time arguing that's what you want $10 billion for,” he says. Others say such speculation is premature and pessimistic. “We are so used to discussing the new territory that we are going to enter that sometimes we think that we know what we are going to find,” says Jos Engelen, chief scientist at CERN. “Well, we don't, and I think it will be much more exciting than we expect.” That may be, but this much is certain already: Everyone hopes for more than just the Higgs. 9. TOXICOLOGY A Sluggish Response to Humanity's Biggest Mass Poisoning 1. Yudhijit Bhattacharjee Arsenic-laced water has sickened thousands in South Asia. After delays and false starts, India is addressing the problem with a$500 million safe-water initiative

CHANDALATHI, INDIA—Until the mid-1990s, the biggest foe of Gouchan and Renubala Ari and their extended family was poverty. Then a more insidious menace began to stalk the Ari home in Chandalathi, a cluster of mud huts on the edge of a yellow mustard field some 60 kilometers north of Kolkata. The first signs of trouble were brown spots on their hands and feet that, as the months passed, developed into thick calluses and lesions. It was several years later that doctors visiting the area recognized the hallmark symptoms of arsenic poisoning.

Tests confirmed that water from the well the Aris were using was laden with arsenic. Their oldest son and his wife were diagnosed with skin cancer, a disease linked with chronic low-level arsenic exposure. Gouchan sold his cow, goats, and ducks to pay for their treatment. The couple died anyway. Afraid of suffering the same fate, two younger sons moved to other parts of India. “Arsenic destroyed our home,” says Gouchan, a frail 76-year-old who walks with a limp because of arsenic lesions. “I'm tired of showing my calluses to strangers,” adds Renubala. “Who can understand our misery?”

Thousands of families in the state of West Bengal have been affected by this blight. More than 40 million people here live in areas with elevated levels of naturally occurring arsenic in the groundwater. Authorities estimate that 5 million in West Bengal drink water with arsenic concentrations above the government standard of 50 micrograms per liter. In neighboring Bangladesh, more than 82 million people live in contaminated areas. And the problem is widening: In recent years, researchers have found high levels of groundwater arsenic in several other Indian states, including Uttar Pradesh, Bihar, and Manipur.

Although there are no reliable statistics on arsenic victims in India and Bangladesh, one research group has counted at least 14,000 cases of arsenicosis in West Bengal alone. The arsenic scourge, says Allan Smith, an epidemiologist at the University of California, Berkeley, is the “largest poisoning of a population in history.”

It didn't have to turn out this way—certainly not in India, whose government frequently touts the country's burgeoning science and technology capacity. Here in West Bengal, officials have had a quarter-century to tackle the contamination. (Bangladesh learned of the threat a decade later.) Yet the government failed to investigate it adequately or provide alternative water resources to affected areas, critics charge. “For many years, government officials accused us of lying and exaggerating the problem,” says dermatologist Kshitish Saha, who uncovered the first cases of arsenicosis while at the School of Tropical Medicine in Kolkata.

Since the early 1990s, when Indian authorities began to respond more vigorously to the crisis, state and national governments have pumped tens of millions of dollars into solutions aimed at providing safe water. The results have been lackluster. A $7 million initiative to fit wells with arsenic filtration units failed because of improper maintenance. Another strategy—drilling deep wells that bypass arsenic-tainted aquifers—has produced mixed results. The most deplorable aspect of the tragedy, critics say, is that Indian officials have resisted educating villagers about the threat, partly out of concern that this could lead to societal unrest. This is unconscionable, says Dipankar Chakraborti, an environmental scientist at Jadavpur University (JU) in Kolkata. “If people are made to realize the dangers of drinking arsenic-contaminated water, they will take care of their own safety,” he says. West Bengal officials acknowledge that the state erred. But they say that a half-billion-dollar initiative now under way to install eight surface-water treatment plants and 360 high-capacity, deep wells fitted with arsenic-removal facilities will provide a long-term remedy to what ranks as one of the biggest public health disasters of the modern world. “Yes, there have been delays,” says D. N. Guha Majumdar, a gastroenterologist on West Bengal's arsenic task force. “But the government is acting now.” Shallow reactions Until the mid-1960s, much of West Bengal relied on untreated water from ponds, rivers, and open wells; as a result, cholera and other lethal waterborne diseases took a heavy toll. A savior arrived in the form of shallow tube wells, pipes bored into the ground with a hand pump at the top. When the technology for sinking this sort of well became affordable, the government and private citizens began installing them by the thousands. Deaths from infectious diseases fell sharply. But the tube wells spawned a new epidemic. After Saha first linked brown calluses to groundwater arsenic in 1982, the local government appointed a panel to examine the problem and find countermeasures. Over the next 5 years, teams from the School of Tropical Medicine, the All India Institute of Hygiene and Public Health, and other institutions documented evidence of chronic toxicity among hundreds of villagers in six districts of West Bengal. Although some experts early on blamed the illnesses on industrial pollution, it soon became clear that the culprit was arsenic in alluvial aquifers. Researchers studying the phenomenon—seen in many other countries, including China, Vietnam, Chile, and the United States—now believe that soil microbes liberate arsenic from harmless pyrites in the alluvium. Open wells, even in arsenic-rich areas, typically have low arsenic concentrations because when the water stands exposed to air for days, the metal binds to iron oxides and other compounds and precipitates out of the water column. This does not happen in an enclosed tube well. The crisis persuaded Chakraborti to give up a career in the United States and return to his Kolkata roots in 1988 to found the School of Environmental Studies at JU. Over the next 6 years, he and his group tested hundreds of tube wells, as well as skin, hair, nail, and urine samples. They found that arsenic contamination was widespread. In some areas, the government dug deep wells. But officials disputed the magnitude of the problem and ignored calls from Chakraborti and others to harness West Bengal's plentiful surface-water resources for a long-term solution. Frustrated, Chakraborti took off his gloves. (It is easy to mistake the pugnacious scientist for an activist. He once scolded the state's environment minister for smoking at a meeting; on another occasion, he advised a prominent arsenic researcher to take a course in water testing.) Chakraborti organized an international conference at JU in February 1995, at least in part, he says, to embarrass officials into action (Science, 11 October 1996, p. 174). He put victims front and center. “I had 19 arsenic patients sitting in the first row,” he says. Seventeen have since died. The 1995 conference made Chakraborti persona non grata to the state government: He says he has been shut out of government-sponsored meetings and accused of being a traitor. But the event had an impact. The next year, West Bengal officials requested$200 million from the national government for countermeasures, receiving less than half of the request. Part of the money expanded an initiative to sink deep tube wells in an effort to tap untainted water below the alluvial aquifers. And in 1998, West Bengal began equipping 2400 hand-pumped shallow tube wells with arsenic-removal units: $1500 adsorption towers packed with substances such as alumina or iron oxide granules. The state also began building a surface-water treatment plant in Malda, one of the worst-affected districts. For Chakraborti, this was not enough. Between 1999 and 2005, he and his colleagues evaluated the performance of nearly 600 of the hand-pumped arsenic-removal units. They uncovered numerous problems. Fifty units had been installed in areas with potable groundwater, while 73 were delivering water with arsenic concentrations above the permissible limit. And some 175 units allowed water through with unacceptable levels of iron. “Overall, the study showed that 82% of the [units] were not useful,” the researchers reported last year in the Water Quality Research Journal of Canada. They blamed it on a lack of maintenance, including a failure to periodically replace adsorption media. West Bengal officials dispute Chakraborti's analysis. Amiya Banerjee, chief engineer of the state's Public Health Engineering Directorate, claims that 70% of the installed arsenic-removal units are working fine. But he acknowledges that they are not being maintained well. Even before Chakraborti's study came out, the West Bengal government in early 2006 announced that it would no longer equip hand pumps with arsenic-removal units. “We realized that the government cannot oversee the maintenance of these units,” Banerjee says. Chakraborti's group has also assailed the government's strategy for boring deep tube wells indiscriminately. In the past 10 years, the researchers have found that at least 20 deep wells—100 to 150 meters deep—in West Bengal's North 24 Parganas and Murshidabad districts have gone from having virtually no arsenic in the water at the outset to concentrations ranging from 50 to 150 micrograms per liter (exceeding the 50-microgram limit) within 7 years. Even now, people seem blind to the risks: In interviews with Science, some villagers in North 24 Parganas district said they trusted that the water they drink from one of these wells is safe. Chakraborti argues that deep tube wells should be sunk only in areas where there is a thick clay barrier between the shallow, arsenic-contaminated aquifer and the deeper aquifer being tapped. “Subsurface geology should guide the strategy,” he says. Others disagree. Alexander van Geen, a geologist at Columbia University, is a strong advocate of deep tube wells as a short-term solution in Bangladesh. He says that mechanical failure of the wells, not arsenic leaching into deeper aquifers, is to blame for the handful of tainted wells. “Because of flawed construction, you end up drawing the shallow water,” says van Geen, whose team has documented four such failures out of 51 deep tube wells monitored over 5 years in the Araihazar region in Bangladesh. Although he agrees with Chakraborti's push for surface water as a long-term solution, van Geen says deep tube wells, regularly monitored, are “hard to beat” as a source of safe water over the next 10 years. Technical fixes may be debated, but nobody disputes the need for better public awareness. Years ago, the West Bengal government decided to paint tube wells with potable water blue or green and leave unsafe tube wells unmarked—rather than paint them red, as suggested by the government's arsenic task force. “The administration thought that would create unnecessary panic,” says Chandan Sengupta, a task force member who formerly managed a UNICEF project aimed at tackling the arsenic problem. When JU researchers in the mid-1990s took it upon themselves to paint unsafe wells red, Chakraborti says, “one legislator had the tube wells painted green and went around with a loudspeaker telling villagers that the water in them was fine.” Overcoming inertia The headquarters of West Bengal's Public Health Engineering Directorate is located on the sixth floor of a dull high-rise in the heart of Kolkata. Dim stairwells are splattered with red marks from people spitting paan, a popular snack consisting of fragrant condiments wrapped in a leaf. Hallways throng with vendors making tea on little kerosene stoves. In a large, open office room, desks are covered with mountains of paper, but many of the directorate's clerks and mid-ranking employees are nowhere to be seen. Across the hall, aides stand guard outside the private offices of senior officials, such as chief engineer Amiya Banerjee. Banerjee says that the government's handling of the arsenic crisis is now robust. “We knew all along that surface water had to be the long-term solution, but we needed a quick fix in the interim,” he says, referring to the deep tube wells and arsenic-removal units for hand-pumped wells. Banerjee also defends the decision not to paint unsafe tube wells red, explaining that doing so would likely have deterred villagers from drawing water for safe uses such as washing. In any case, Banerjee says, the government is now implementing long-term solutions. Groundwater will continue to be the mainstay. Under a$500 million initiative funded jointly by the state and national governments, West Bengal is sinking 360 large-diameter deep tube wells equipped with arsenic-treatment facilities to pipe water to 70% of the affected population. Government engineers will supervise maintenance of the treatment plants, Banerjee says.

Chakraborti contends that fitting each well with an arsenic-treatment plant is a waste of money, considering that only a small percentage of wells is likely to become contaminated. Van Geen too labels the plan as flawed. “Instead of setting up the treatment plants right away, it makes more sense to design a pumping system that could be connected to an arsenic-treatment module in the future if the need arises,” he says. If the government is determined to install large arsenic-treatment plants, he argues, it would be wiser to purify water from shallow, contaminated aquifers and conserve the deeper, arsenic-free aquifers for future use.

Also under the West Bengal initiative, five surface-water treatment plants are being built. Together with three plants already commissioned in North 24 Parganas, South 24 Parganas, and Malda, they will serve 30% of the affected population. To pay for operating costs, the government will charge families a connection fee and about \$1 a month. That's a risky strategy. Indian villagers typically don't pay for water, which officials acknowledge makes it difficult to get them to switch from a public tube well to a piped water connection. In the 3 years that the South 24 Parganas plant has been in operation, only 25,000 out of the 300,000 homes intended to be covered by the plant have taken a connection.

Other efforts under way include the distribution of cheap domestic filters and a drive to ensure that all 600,000 private tube wells are tested for arsenic. Majumdar of West Bengal's arsenic task force expects that everybody in West Bengal will have safe drinking water within 3 years.

Chakraborti is not as optimistic. And he wonders how many more people will suffer if awareness is not made an urgent priority. On a recent visit to Nadia district, he met a man with classic arsenicosis lesions who had never heard about arsenic before doctors diagnosed his disease last December. “This man drank contaminated water for years and then had to sell his land to find out what he was suffering from,” Chakraborti says. “What is happening here is a grave injustice.”

10. TOXICOLOGY

A Young Scientist Shaped by Adversity

1. Yudhijit Bhattacharjee

KOLKATA—First his uncle succumbed. Then his father, then his aunt. In 2004, when Kartik Biswas saw his mother, Dulali, develop the same lesions that presaged the deaths of his other family members, he took a 5-hour ride by bus and train from his village in Nadia district to Kolkata to seek out Dipankar Chakraborti, a crusading arsenic researcher at Jadavpur University. The encounter changed Biswas's life.

Chakraborti helped get Dulali admitted to a government hospital. Tests confirmed arsenicosis, and she underwent a skin graft on her palm. After bringing his mother home, Biswas spread the word on arsenic—for instance, by advising villagers which tube wells had tested positive for arsenic in Chakraborti's lab—and collected water samples for further testing. In 2005, Dulali got sick again and was diagnosed with skin cancer. Doctors amputated her arm, but the cancer had already spread, and she died last October.

This fall, Biswas, who is completing a master's degree in geography at a local college, will join Chakraborti's lab as a Ph.D. student to study countermeasures to groundwater contamination. Says Biswas, whose palms and feet bear the marks of mild arsenicosis, “Nobody should have to see their mother suffer like I did.”