# News this Week

Science  06 Jun 2003:
Vol. 300, Issue 5625, pp. 1484
1. MARINE RESEARCH

# Pew Panel Calls for Sea Change in U.S. Ocean Policy

1. David Malakoff

They hauled lobsters, tramped through marshes and canneries from coast to coast, and met with everyone from surfers to scientists. Now, after 3 years of deliberation, the 18 members of the Pew Oceans Commission have reached a verdict: The United States is guilty of neglecting its marine realm and must dramatically overhaul its approach to ocean policy to avoid expensive ecological catastrophes.

“U.S. ocean governance is in disarray [and] … the status quo is unacceptable,” the independent panel of policy and science heavyweights concluded in a report (pewoceans.org) released this week. Its proposed solutions include merging the government's fragmented ocean programs into a muscular new independent agency, creating a robust system of marine reserves, and doubling spending on marine research. The panel, led by former California congressman and Clinton White House chief of staff Leon Panetta, also asks federal legislators to create a new regional network of planning bodies and to set tighter standards for everything from coastal pollution to fish farming.

It's too early to tell if the current powers-that-be, including the Bush Administration and Congress, are interested in sailing into such treacherous waters. Although many of the recommendations “aren't new, reorganizing government is very tough; it's going to take unusual leadership,” says Carolyn Thoroughgood, dean of the College of Marine Studies at the University of Delaware in Newark. Still, marine advocates hope that the report, funded by the Pew Charitable Trusts in Philadelphia, Pennsylvania, will gain momentum from a similar study due out later this year from a government-appointed group headed by retired Admiral James Watkins, and that the two will produce a wave of change.

Both groups are attempting to follow in the footsteps of the influential Stratton Commission, which in 1969 issued a report that led to the creation of the National Oceanic and Atmospheric Administration (NOAA) and a host of important ocean-related policies. But those 1970s solutions, says the Pew panel, are no match for today's problems, which range from overfishing to coastal sprawl. NOAA's niche within the industry-oriented Department of Commerce has complicated efforts to protect fish stocks, for instance, and more than a dozen federal agencies now have a voice on marine issues, complicating coordination.

The panel, which included several current and former governors and mayors, a commercial fisher, and an ex-astronaut, suggested that Congress start by declaring the government's priority to be healthy oceans, not maximizing economic return. Next up would be eight new regional commissions, composed of public officials and responsible for writing long-term protection plans covering everything from “zoning” the ocean for various purposes to protecting coastal waters from harmful land uses. Congress should also double NOAA's budget, now about $3 billion, and make it an independent agency bulked up with ocean-related programs from other agencies, such as the Department of Interior's ocean minerals and marine mammal programs. Annual spending on research would also double, to about$1.5 billion, so that “science can provide the platform needed for ecosystem-based management that wasn't there 30 years ago,” says commissioner Charles Kennel, head of the Scripps Institution of Oceanography in La Jolla, California. Kennel and other panelists would also like to see round-the-clock marine monitoring systems and more marine restoration, social, and economic studies.

Ocean advocates don't expect an immediate response from Congress or the White House. But they'll be disappointed if the two reports don't at least kick off a serious debate about better ways to design and implement U.S. ocean policy.

2. HUMAN GENOME

# A Low Number Wins the GeneSweep Pool

1. Elizabeth Pennisi

COLD SPRING HARBOR, NEW YORK—The human genome has been sequenced, but calculating the number of genes it contains is taking more time. DNA experts have nonetheless decided they know who made the best prediction: Lee Rowen. The Seattle sequencer and two of her colleagues were declared winners this week of GeneSweep, an informal contest begun here 3 years ago at a genome meeting at the Cold Spring Harbor Laboratory (CSHL), in which researchers tried to guess just how many protein-coding sequences it takes to make a human.

When he organized the pool, Ewan Birney of the European Bioinformatics Institute near Cambridge, U.K., was convinced that the answer would be in hand by now. Most estimates at the time were high. Some gene counters were insisting humans had upward of 100,000 genes, and just a handful were hinting that the number might be half that or fewer.

The closer sequencers came to finishing the human genome, however, the lower the gene count went—and the more vexing it became to determine the actual number. For example, Birney estimates that with today's stringent criteria and improved gene-prediction programs, there might be 24,500 genes, of which 3000 might be “pseudogenes” that don't produce proteins and therefore don't count. But there's quite a bit of poorly explored DNA sequence—dubbed “dark matter”—that may contain more genes. Faced with this complexity, Birney thought he'd have to postpone picking a winner.

But then he decided that, no matter what, “we're coming up with a sub-30,000 gene count,” he reported at last week's CSHL genome symposium. And it turned out only a few bettors put their money on a number that low. Rowen, a sequencer at the Institute for Systems Biology in Seattle, Washington, was the closest—predicting 25,947 in 2001. So Birney decided that she should get the prize, half the $1200 pool. Paul Dear of the U.K. Medical Research Council and Olivier Jaillon of Genoscope in Evry, France, with guesses of 27,462 in 2000 and 26,500 in 2002, shared the other half of the prize. Rowen credits sequencer Jean Weissenbach of Genoscope with influencing her prediction; years ago, he suggested that the number would be low. “At the time, everyone nearly fell off their chair,” Rowen recalls. But now it seems undeniable that humans have protein-coding gene counts that are close to those of C. elegans. Along with her winnings, Rowen is waiting to collect an autographed copy of James Watson's book, The Double Helix. 3. GENOMIC MEDICINE # African-American Population Biobank Proposed 1. Jocelyn Kaiser Hoping to jump-start an era of personalized medicine for black Americans, researchers at Howard University in Washington, D.C., want to build the first large DNA and health database on people of African descent. The project, announced last week, aims to collect DNA samples from 25,000 volunteers over 5 years and use the data to probe how genetics and lifestyle factors contribute to common diseases. Experts say such biobanks will eventually help doctors tailor prescriptions to individuals and advise patients who are at elevated risk for certain diseases (Science, 8 November 2002, p. 1158). But although biobanks are planned or under way in Europe and at some U.S. medical institutions, Howard's Genomic Research in the African Diaspora (GRAD) biobank would be the first to look specifically at African Americans, who have higher rates of certain diseases than other U.S. populations. “This is a new front in biomedical research that we must be involved in,” says human geneticist Georgia Dunston of Howard University College of Medicine, who will direct the project. Adds population geneticist Mark Shriver of Pennsylvania State University, University Park, “This is definitely a major advance and something that's new for African Americans.” The role of race and ethnicity in medicine has been widely debated recently. On the one hand, genetic differences among individuals within a race are greater than differences between ethnic groups; moreover, African Americans have an estimated 17% to 20% Caucasian ancestry, blurring the biological meaning of “race.” But human geneticist Stephen Warren of Emory University in Atlanta says that a race-based data bank will still be useful because “gene frequencies tend to vary based on the continent of origin. It doesn't make a lot of sense to say race is irrelevant.” A large population database will also help disentangle genetics from socioeconomic and other environmental factors in understanding disease among African Americans, says Dunston. Dunston expects the project to study diseases such as prostate cancer, type II diabetes, hypertension, obesity, and asthma. Patients with these diseases and healthy volunteers will be recruited from Howard-affiliated clinics and physicians and, later, from elsewhere in the country and around the world. Howard genetic epidemiologist Charles Rotimi hopes the project will also help fill in “a complete picture” of how genes and the environment interact, since “Africa is the trunk of the human evolutionary tree.” The university hopes to build the biobank by enrolling patients from studies funded by the government and industry sources. Randall Maxey, president-elect of the National Medical Association, which represents African-American physicians, thinks that having a historically black university head the project will change attitudes about clinical trials among African Americans, who have been wary of participating since they learned of the notorious Tuskegee syphilis studies that started in the 1930s. “I think the trust factor is going to go up,” says Maxey, a Los Angeles nephrologist. To build the database, Howard is teaming with First Genetic Trust, a Chicago-based company that provides secure Web-based technology for storing clinical and genetic information (Science, 26 January 2001, p. 575). The GRAD biobank project will be one of its largest, says spokesperson Daniel Schmitt. The Howard team will meet with the company this month to begin planning a pilot study. 4. MARINE RESEARCH # Congress Fiddles With Law While Scientists Burn 1. Naomi Lubick When the U.S. Navy complained last year that a 1972 law to protect marine mammals was hampering its ability to test a new sonar system, many scientists were not unsympathetic. Although nobody disputes the need to protect the animals, researchers trying to understand the impact of human-produced sounds on marine mammals, as well as those exploring the world's oceans, have also found themselves ensnarled in the Marine Mammal Protection Act (MMPA). They were hoping that judicious changes in the law might reduce the amount of red tape delaying seemingly innocuous studies—and help them identify truly harmful sounds. Last month, at the Navy's request, the House of Representatives passed a bill that would indeed change the MMPA. But the language was a huge disappointment to scientists. A last-minute amendment loosening the restrictions applies only to the military, not to researchers. That means a continuation of what scientists say is a slow, cumbersome, and costly process. The measure, part of a broader bill that authorizes 2004 spending levels for the Department of Defense, is on a fast track through Congress. And unless legislators have a sudden change of heart, some advocates say that marine mammals could be worse off than under the current law. At the heart of the debate is the government's rules for issuing permits for “takings”: activities, from seismic research to weapons tests, that might affect marine mammals. The existing law requires permits for any activity with the “potential to injure … [or] disturb” a single animal or group. The provision has triggered 3 decades of fighting over what constitutes “harassment” and, thus, what actions require prior approval from government regulators. “The system is definitely broken with regard to the welfare of the animals,” says Scott Kraus of the New England Aquarium in Boston, who has waited 2 years for his permits to photograph and take fat samples of endangered right whales in the North Atlantic, in an effort to better protect them. The House bill addresses the military's concerns by adding “significant” to “potential” in the definition of harassment. The change “raises the bar,” says John Kunich of the Roger Williams University School of Law in Bristol, Rhode Island, by requiring direct evidence of harm—“or a bubbling gun, if you will.” He says the change, for example, would give the Navy greater latitude to test sonar because of the lack of conclusive scientific evidence that the technology is “demonstrably damaging.” But that still leaves researchers out in the cold. One scientist on the forefront of efforts to understand marine mammal acoustics is anatomist Darlene Ketten of Woods Hole Oceanographic Institution (WHOI) in Massachusetts, whose work showed the first strong link between the use of some types of military sonar and harm to certain kinds of whales (Science, 26 January 2001, p. 576). Last month Ketten told a House panel that researchers have been devastated by the current labyrinthine process of obtaining a permit and that “it would be very beneficial if the wording of the law were changed to be clearer.” In particular, many scientists would like the definition changed to focus on how a scientific study or other project would affect an entire population of animals rather than a single creature. That approach has been endorsed by a 2000 National Research Council panel that also recommended a new standard based on “biological significance.” Such a standard would exempt activities, such as the noise of a ship or low-frequency sonar, that would likely trigger only “insignificant changes in behavior” of individual animals, says Peter Tyack, a whale researcher at WHOI who recently was forced to return ashore from a planned research expedition because of a legal fight over whether he had satisfied another federal environmental rule. Last month the nonprofit Consortium for Oceanographic Research and Education also backed a revised definition of harassment while expressing concerns over the congressional language. But others are wary of modifying the act. Environmental groups worry that the military could abuse eased standards and that other groups—from oil drillers to shippers—will demand their own exemptions. And even Kraus says that he is “extremely cautious about changing” the MMPA, although changes “would probably help me.” At the same time, he thinks that the National Marine Fisheries Service (NMFS) should be doing a better job of administering the act. NMFS officials say that they are trying their best, and that most of the problems involving permitting stem from requirements under other laws. “The harassment definition is one change in a whole suite of things” the agency has been studying, says Donna Wieting, whose office oversees marine mammal conservation. NMFS is still reviewing the language pending before Congress. (The Senate has passed a similar Defense reauthorization bill that is silent on MMPA, and the two bodies must reconcile their differences.) But Wieting admits that “one word could make a huge difference.” NMFS is also worried that moving away from protecting individual animals could weaken the act. If the proposed changes are approved in the final bill, NMFS will write new regulations to interpret them. For scientists such as Kraus, Ketten, and Tyack, it would represent another step in their ongoing political education. 5. UNDERGROUND SCIENCE # Flooding Mine Could Doom Homestake Plan 1. David Malakoff Physicists hoping to convert a South Dakota gold mine into the world's deepest underground laboratory may soon see their dream sink without a trace. Barrick Gold Corp. of Toronto this week announced that it will soon turn off pumps that keep water out of its Homestake mine near Lead. Backers say that the flooding will doom efforts to persuade the National Science Foundation (NSF) to fund the$300 million project.

“I have a very hard time envisioning how we could sell [a proposal involving] a wet mine to NSF,” says physicist Wick Haxton of the University of Washington, Seattle, the proposal's lead investigator.

Ironically, Barrick's announcement that it would turn off the pumps on 10 June came just days after an NSF advisory panel concluded that Homestake was “by far the most favorable” of three candidate sites for a buried U.S. lab (Science, 28 March, p. 1965). The panel also urged the company to keep the mine dry. But executives at Barrick, which ceased mining the 2400-meter-deep complex in 2000, said the decision was based on safety and financial concerns. (Pumping costs hundreds of thousands of dollars a month.) Barrick and South Dakota officials also said the mine could be drained again if the lab proposal, which has been stalled by legal concerns over liability for possible pollution at the mine, ever moves ahead.

Haxton, however, estimates it could cost $40 million to dry the mine and possibly more to fix damaged support beams, elevators, and wiring. “I can't imagine NSF spending that kind of money just to get back to where we are now,” he says. NSF officials were not available for comment, but they had earlier planned to call a meeting to discuss the project's next steps. Haxton vows to work for a state-of-the-art underground laboratory “somewhere in the United States” even if Homestake dies. 6. INFECTIOUS DISEASES # Searching for a SARS Agenda 1. Martin Enserink BETHESDA, MARYLAND—With the resurgence of severe acute respiratory syndrome (SARS) in Toronto providing a sobering backdrop, a Who's Who in virology gathered here last week to help set a research agenda to fight this emerging threat. Old-timers such as smallpox legend D. A. Henderson and virus hunter C. J. Peters mingled with SARS celebrities such as Hong Kong virologist Malik Peiris and pharma and biotech executives, as well as Vanderbilt University's Mark Denison and other coronavirus veterans, who seemed delighted that their once-obscure field is now in demand. Besides grave concerns about a global pandemic that has killed more than 700 and sickened 8300, there was a palpable sense of excitement about the chance to work on a new and dangerous virus. Asked whether he was enjoying himself, the host of the meeting, U.S. National Institute of Allergy and Infectious Diseases (NIAID) Director Anthony Fauci, said, “Oh yes, very much so!” What's missing, however, is the money, and Fauci, who runs the biggest infectious-disease agency in the world, stayed mum on when he might have something to offer. Instead of a coherent research agenda, the participants came up with a plethora of questions about the outbreak and the virus that causes it. Epidemiologists, for instance, want to nail down the presumed animal reservoir for the SARS virus. Its presence in civet cats is intriguing, they say, but more work—including sampling a range of wild animals—is needed to make sure that the cats are not just bystanders or so-called intermediary hosts. It's also unclear whether some people can be infected with SARS—and infect others—without getting sick, how the occasional “superspreader” infects scores of others in a matter of hours, and why the disease so far has affected few children. Stopping the spread of the disease will be tough without quick and reliable diagnostic tests. A variety of methods are already in use, from testing for antibodies and viral genetic material to isolating the virus itself. But validating their accuracy will require taking a series of samples from many patients at various times during the course of the illness, researchers said—a major undertaking. Particularly problematic is the fact that most of the tests have failed to pick up the virus during the first few days of infection, when patients may feel well but start shedding the virus. Scientists at the U.S. National Institutes of Health have begun working on a vaccine made from a killed virus. Other proposed strategies include using a live, attenuated virus—often more potent but also more dangerous—as well as protein and DNA vaccines. Which weapon might work best is anyone's guess, however. Some researchers also worry that a vaccine might actually worsen SARS through an interaction with the immune system, as can a vaccine for another coronavirus disease, feline infectious peritonitis. Although researchers would like to understand better how the virus makes people sick and the role of the immune system, “we can't afford the luxury of sitting back and waiting for the answers,” said Robert Couch of Baylor College of Medicine in Houston. “We have to move forward.” But science costs money, and it's not clear who will pay. NIAID has opened up a few opportunities and added SARS to an existing grants program aimed at biodefense and emerging diseases. It has also tacked$420,000 onto an existing HIV vaccine award to GenVec for a SARS vaccine that uses an adenovirus as a vector. But most companies and academic researchers are still doing the SARS work on their own dime, hastily rearranging priorities and funding. For studies to expand, some researchers say, the government will have to pony up. If the market for a SARS vaccine turns out to be small, says Thomas Monath, chief scientific officer of the British-American vaccine company Acambis, industry may also need a guaranteed purchase of its product or some other sort of carrot to stay in the race.

Both Congress and President George W. Bush are acutely aware of the SARS threat, says Fauci, adding that “my job over the next couple of months” will be to obtain additional federal funding for SARS research. He hopes that the money won't come at the expense of the institute's vast bioterrorism program, which involves many of the same researchers and companies.

7. POSTWAR IRAQ

# Assyrian Gold Safe as Looters Threaten Southern Sites

1. Andrew Lawler

BAGHDAD—Five damp boxes containing the gold jewelry of ancient Assyrian queens are intact at an undisclosed safe location in Baghdad. Science has learned that the bulk of the Iraq National Museum's most precious artifacts are likewise secure in a secret storage area on the museum grounds. That's the good news, say U.S. and Iraqi officials, and it contrasts with reports of extensive losses during the initial days of the U.S. arrival in the capital. But the country's rich cultural heritage faces a new and terrible threat, say archaeologists: In recent weeks they have witnessed hundreds of looters pillaging sites throughout southern Iraq.

National Museum officials here are now laboriously conducting an inventory of their losses; the current count stands at a little over 1000 items. But U.S. Customs and Iraqi officials say that most objects once displayed in public galleries were moved to an air-raid shelter at an undisclosed location on the museum grounds before the first shots were fired. An Iraqi official confirmed the existence of the cache to Science, but museum officials decline to identify the location for security reasons. U.S. investigators say they have sworn testimony about the place from five museum employees but are keeping their distance.

Across the Tigris River, a delegation of Iraqi and U.S. officials this week opened the Central Bank vault, where the spectacular Nimrud grave goods have been stored since 1991. The hundreds of pieces of finely worked gold material have until now been seen by only a handful of archaeologists. The vault area flooded when the bank was bombed and looted this spring, and it took weeks to pump out the water. But the officials found the boxes with their seals intact, and as Science went to press they were planning to examine the contents after moving them to a secure location.

Outside Baghdad, however, the situation remains chaotic. The scale of the looting in the south is unprecedented, according to a team of U.S. officials, soldiers, and archaeologists who toured the area on 21 May in a U.S. military helicopter. The ancient site of Umm Al-Hafriyat “looks like a huge waffle” because of looting holes, says McGuire Gibson, a University of Chicago archaeologist on the flight. At the nearby mound of Adab, “there were 200 to 300 guys with shovels,” he says. The chopper landed at Umma, an important Sumerian city, and U.S. soldiers fired shots to chase off another band of 200 to 300 looters who had been busy digging trenches and tunnels. Likewise, a huge number of illicit diggers were chased from the ancient site of Isin. “It's phenomenal what's been done,” says Gibson. Although looting occurred at remote sites in the mid-1990s, U.S. and Iraqi researchers say the problem is much worse now.

Compared to the drama unfolding in the south, “the north is much better off,” says Jaber Khalil, chair of Iraq's State Board of Antiquities, who had just returned from a visit to that region with U.S. officials. The Mosul Museum storerooms were looted, although “we have no estimation yet of the losses,” he says. At nearby Nineveh and Nimrud, pieces of Assyrian stone friezes—potentially worth millions of dollars on the antiquities market—have been removed or damaged, Khalil added. A U.S. military unit near the Mosul Museum offers some protection there, and U.S. troops are posted at Nimrud around the clock. But guards remain only until sunset at Nineveh. U.S. troops also have secured Hatra, an important religious center northwest of Mosul that flourished around A.D. 200.

Halting the illicit digging is a formidable task, because U.S. troops are focused on preserving order in the cities and vehicles leaving the country are seldom searched. Small objects such as seals or cuneiform tablets are extremely valuable and easily transportable, and Iraqis say that antiquities markets are popping up everywhere in the south. The State Board staff members, meanwhile, have no cars, communications equipment, or money for guards at distant sites. Military officials are weighing the use of helicopter and satellite reconnaissance, according to U.S. Army Col. John Kessel in Baghdad, although their effectiveness will be limited without more frequent intervention.

In the meantime, officials expect the bleeding of archaeological objects to continue. Says Gibson: “It's going to be a very, very trying time.”

8. THEORETICAL PHYSICS

# Calculations Pop the Cork on Travel Through Spacetime Tunnels

1. Charles Seife

It's not hard to work your way out of a hole … a wormhole, that is. In science-fiction movies, wormholes—bridges between two regions of spacetime—are handy devices for traveling halfway across the universe in the blink of an eye. Scientists have long pondered whether they were physically plausible (probably, many concluded) and if so, whether spaceships would be able to cross the bridge (probably not). Now, a team of physicists concludes, it might be surprisingly easy to make a wormhole traversable.

Don't book your tickets to Andromeda yet. “We are not going to build a wormhole with current technology or even with presently foreseeable technology,” says Matt Visser, a physicist at Victoria University in Wellington, New Zealand, who took part in the work. But he says that studying wormholes allows physicists “to take theories we more or less understand [general relativity and quantum physics], put them together, and see what breaks.”

Relativity, oddly enough, stands up just fine. As outlandish as it might seem to move ships and people faster than the speed of light, Einstein's rules permit it because wormholes provide shortcuts across the fabric of space and time. Even so, theorists suspected that spacetime bridges would have to exact a heavy toll. That's because relativity also shows that any wormhole can be turned into a time machine. By passing through one, people might go back in time and kill their own ancestors before they were born, causing a paradox.

So physicists were not surprised to discover that wormholes, like black holes, have event horizons: regions beyond which not even light can escape. The event horizon, sitting squarely in the throat of the wormhole, would prevent a traveler from leaving the wormhole that he or she had entered. Nature seemed to have appointed a cosmic censor to prevent paradox.

In 1988, however, physicist Kip Thorne of the California Institute of Technology in Pasadena and his students discovered a way to get rid of the event horizon. By placing a chunk of “exotic matter” with negative energy—that is, less energy than an equivalent volume of empty space—in the throat of the wormhole, a spacetime traveler could banish the event horizon and make the wormhole passable. The rules of quantum mechanics require that such exotic matter is forever being created and annihilated on tiny scales. Unfortunately for sci-fi buffs, however, those quantum doses seemed inadequate to get rid of that pesky event horizon. “This made large wormholes prohibitively difficult to even contemplate,” says Visser.

Now it seems that the cosmic censor could be defeated. As they report in the current issue of Physical Review Letters, Visser and his colleagues investigated a wormhole that is very symmetric and chugged through extensive calculations to figure out how much exotic matter was needed to get rid of the event horizon. Contrary to conventional wormhole wisdom, they found that a quantum-sized smidgen of the stuff could make a black hole traversable.

Other wormhole researchers are cautious about interpreting Visser's calculations. “I'd be a little careful in saying you need only a small quantity of exotic matter” to destroy the event horizon, says Ulvi Yurtsever, a physicist at the Jet Propulsion Laboratory in Pasadena. Instead, he says, the equations allow that the job might require a large amount of exotic matter offset by a slightly smaller chunk of ordinary matter.

Even though nobody is rushing to the patent office to register their wormhole drives, theoreticians hope that traversable wormholes will at least see a bit of light at the end of the spacetime tunnel.

9. AFTER THE WAR

# Iraq's Shattered Universities

1. Andrew Lawler

Classes may be back in session, but Iraq's university system is largely wrecked after a decade of sanctions and repression, followed by war and mass looting

BAGHDAD—Najib Stipho, a chemical engineer and dean of Baghdad's Al-Mansour University College, awoke at dawn on 9 April to the sound of pounding on his front door. His cafeteria manager had come to warn him that looters were threatening the 2600-student private institution. “I told him to get as many guards as he could,” recalls Stipho, who then raced to take up battle stations with his two grown sons as U.S. tanks and troops rolled into the city. Two days and 4000 bullets later, Stipho's band of a dozen machine-gun-toting defenders had driven off the would-be looters. On 5 May the Christian university became the first in Iraq to reopen its doors.

But Al-Mansour's survival is the exception among Iraq's once-proud university system, already weakened by more than a decade of economic sanctions, brain drain, and political oppression. Scores of academic buildings around the country were bombed, looted, and burned during the U.S. invasion. Science departments were especially hard hit. Some were targeted by U.S. troops seeking evidence of complicity in developing weapons of mass destruction; others were plundered by mobs for their computers and scientific equipment.

The path to recovery will be arduous. Although classes for some 200,000 students at more than 40 public universities and colleges nominally resumed on 17 May, money is scarce, security is tenuous, and dozens of top administrators have been fired because of ties to Saddam Hussein's Baath Party. At the nation's premier technical institution, Baghdad's University of Technology, looters stole even the light bulbs before torching labs and lecture halls.

“Right now, Iraqi academics are struggling with a lack of everything, but in the long run I think they are hopeful,” says Stephen Curda, a higher education specialist working for the Department of Defense in Baghdad. So far, he says, the U.S. government has chipped in $270,000 to start rebuilding the technology university and allocated another$47,000 for repairs at Baghdad University, the country's largest higher education institution. Rebuilding expertise and morale, however, may prove more difficult.

## Mobbed

U.S. soldiers now guard the entrance to Baghdad University's well-landscaped campus, which during a recent visit was thronged with male and female students. Sami Al-Mudafar, a biochemist, was elected acting president by the faculty on 17 May under the direction of U.S. officials. But basic school necessities—from books to lab equipment—are sorely lacking, and sporadic electricity, the threat of street violence, and a severe gasoline shortage are keeping many students and professors at home.

And the country's flagship university was relatively lucky. Along with its medical school north of the city, it suffered only minor damage and looting, including the destruction of a medical library statue of Louis Pasteur after it was mistaken for Saddam Hussein. Al-Mustansiriya University, founded in the 13th century, was more seriously looted, but penitent thieves returned some of its furniture, books, and scientific equipment to a Shiite mosque in Sadr City, a poor neighborhood on Baghdad's outskirts.

Stipho's bold stand at Al-Mansour prevented the school's 10 science, engineering, and computing labs and 300 computers from falling into the hands of looters. “They were desperate to get inside,” says Stipho, adding that the looters also tried unsuccessfully to snatch university cars and buses from a protected compound with a stolen crane. “They came in mobs: A group of 50 would come, then would go, and another would come.”

Other, more sprawling, campuses were left undefended. At the University of Technology, which was visited frequently before the war by United Nations weapons inspectors, several Iraqi academics claimed that U.S. forces actively encouraged mobs. “American troops opened the doors, and the looters took it completely,” says one student who lives near that university, an account repeated by Mazui Kadhum, a non-Baathist and former University of Technology professor who is now dean of informatics at Al-Nahrain University. “U.S. tanks broke the gate” and urged looters to enter, he says. Curda acknowledges that U.S. tanks may have caused a large breach in the wall but says he has not received any complaints from university officials about the conduct of U.S. troops.

Kadhum's new institution, Al-Nahrain, formerly Saddam University, is an elite science and engineering school of some 1100 students that shares a campus with Baghdad University. Its generous salaries of up to $1000 month attracted some of the country's best talent. It also attracted the prewar attention of U.N. weapons inspectors. When Al-Nahrain professors returned to the campus a week after U.S. troops arrived in Baghdad, they found Marines bivouacked on the site—and the offices of the science, engineering, and informatics deans ransacked, along with the laser and electronics laboratories. “Motherboards with information that may be of use were taken,” he says. “But they did it in a very unprofessional way; there was equipment that had nothing to do with computers that they threw on the floor.” Kadhum believes that the damage was done by the invading forces rather than looters, citing as evidence dusty prints of U.S. military boots left in the labs and offices and an unscathed office of the political science dean. “The Marines entered every place that the inspectors went to; they knew where they were going,” he adds. A U.S. Defense Department official in Baghdad declined to comment on the allegation. Kadhum does not deny that university professors did consulting work for a host of ministries—including the defense and military industry ministries. But he insists that the university did not host any weapons work. “They had their own laboratories, their own techniques, their own setup” through the Ministry of Military Industry, he says. “It was a country of its own, directly linked to the president.” ## Rocketed science Another school hit hard by the war is Al-Kufa University, south of Baghdad, which was visited as recently as 18 January by U.N. inspectors. “There was fighting, bombs, explosions, looters, and burnings,” says Nabil Al-Rowi, an electrical engineer who until last month was president of the university, which has three campuses, nine colleges, and 10,000 students. The trouble began 26 March, says Al-Rowi, when U.S. Marines reached Kufa, site of one of the three campuses. Staff members left after the Marines “broke in the doors and were searching for computers” in the areas visited by U.N. inspectors, he says. Looters caused additional damage in the first weeks of April after the Marines left, he says. A rocket attack on 3 April severely damaged the central administration building in Najaf, which he says was located next to the chief Baathist commander of the region. Three days later the medicine, pharmacy, and education departments were looted. “The microbiology, physiology, and health physics labs were burned, completely destroyed,” he says. Looters also made off with some 100 ancient manuscripts, including some pre-Islamic documents of the Nestorian Church. After rejoining his family in Baghdad, Al-Rowi was fired from his job because of his membership in the Baath Party. Other institutions experienced varying amounts of damage. Eyewitnesses say Mosul University in the north suffered broken windows and trashed administrative areas, whereas Basra University is mostly gutted. Nasiriya University in the south was bombed because it had antiaircraft guns on campus, says psychology professor Fadhil Kzar. Although construction crews can repair the physical damage, Iraq's battered campuses have other needs that will be harder to fill. Importing science textbooks and a vast array of scientific equipment was illegal under U.N. sanctions imposed after the first Gulf War, and contact with the outside world was discouraged and often punished by the Saddam regime. Many of the country's scientific and technical elite fled abroad. Al-Rowi estimates that 1400 of 5400 people holding scientific and technical doctoral degrees emigrated between 1991 and 1998. At Al-Nahrain's engineering college, for example, Kadhum says that 18 of 45 professors left. This flight had a profound effect on the quality of education throughout the country; Iraqi students and faculty members talked privately of a system driven by bribes and party loyalty rather than educational standards. U.S. officials are promising a clean sweep to put the universities on firmer footing. Andrew Erdmann, the U.S. Department of State official in charge of higher education and science in Iraq, warned Baghdad University faculty members last month that academics associated with human-rights violations and weapons of mass destruction research would be purged and that high-ranking Baath Party members would not be allowed to hold the top tier of academic positions. Curda says party members not involved in nefarious activities can apply for an exemption, and many have done so. “Our goal is to bring back as many experienced administrators as we can,” he says, once they pass the necessary background checks. Many Iraqis say party membership was essential for climbing the academic ladder. “Most university professors and staff are members of the Baath Party,” says Mohammad Al-Awsi, Al-Nahrain's recently fired engineering dean and a party member since 1968. Even non-Baathists worry that the policy will undermine Iraq's already fragile technical and scientific foundation. “This will exclude good people,” says Abbas Abdul-Kader, president of the Iraqi Commission for Computers and Informatics and a non-Baathist. For the moment, university administrators say their priorities are security, reliable electricity, and operating funds. U.S. officials began paying some salaries last week, and Erdmann visited Persian Gulf states to drum up reconstruction money from wealthy emirs. The Department of State will fund a$20 million to \$30 million program to establish partnerships between U.S. and Iraqi universities, according to a department memo.

Erdmann calls the overall reintegration effort “a huge challenge.” But Curda says many university officials appear to welcome the U.S. presence: “They are excited to be free to travel, to contact overseas colleagues again, and to have basic academic freedom.” Adds Al-Awsi, who studied engineering at the University of Sheffield, U.K.: “We're not against the United States. We're trying to create good universities and set high standards for our students.” All sides agree that this goal won't be achieved without large doses of outside help and domestic tranquility.

10. HEART DISEASE

# How to Subdue a Swelling Heart

1. Jean Marx

Research on hereditary cardiomyopathies is turning up possible treatment targets for a condition that kills far more people—heart failure

Every time it happens, the news is shocking. A young athlete, seemingly at the peak of health, collapses on the playing field and dies. In more than one-third of such cases, the athlete turns out to have had an undiagnosed problem known as cardiomyopathy, in which the heart is abnormally enlarged, usually as a result of a genetic abnormality.

All told, cardiomyopathies claim more than 27,000 lives annually in the United States, and many of the victims had little or no warning that they were in danger. “You can be walking around with this and not know it. … Often the first symptom is sudden death,” says cardiologist Christine Seidman of Harvard Medical School in Boston. Now, researchers are getting a much better understanding of the genetic defects that cause cardiomyopathy—information that may also help in the fight against an equally dangerous but much more common condition, heart failure. It occurs when some stress, such as high blood pressure or damage from a heart attack, causes the heart muscle to progressively lose its ability to pump blood.

So far, researchers have identified nearly 20 genes that can, when mutated, cause cardiomyopathies. Most encode proteins, such as actin and myosin, that form the characteristic banded structures, or striations, of cardiac and skeletal muscle and generate muscle contractions. They're also getting a fix on just how those mutations lead to heart enlargement and ultimately to death.

The mutations found so far directly or indirectly disturb cardiac cells' ability to handle calcium—a critical problem, because calcium ions trigger muscle contractions and are also involved in cell regulatory activities. For example, calcium ion disruptions have been linked to altered patterns of gene expression that cause muscle cells to grow abnormally large, leading to a massive thickening of the heart walls. The disruptions may also trigger apoptosis, a form of cell suicide.

The identification of mutant genes may allow members of affected families to be screened. Those who carry the mutations could be identified before their hearts become visibly enlarged and could then take precautions, such as avoiding strenuous exercise.

Researchers have also identified a number of molecules that help bring about the calcium ion changes in response to the mutations. “We've reached a point where we have identified many key steps in signaling pathways leading to cardiomyopathy,” says Eric Olson of the University of Texas Southwestern Medical Center (UT Southwestern) in Dallas. Some of these molecules are potential targets for new therapies—and not just for cardiomyopathies.

Researchers are hopeful that what they are learning will lead to desperately needed treatments for what Kenneth Chien of the University of California, San Diego (UCSD), calls “garden-variety heart failure.” Heart failure deaths are on the rise, partly because cardiologists have gotten better at saving heart attack victims. Such patients are often left with damaged, failure-prone hearts. Every year in the United States alone, about 550,000 people are diagnosed with heart failure and more than a quarter of a million succumb to the condition. “Heart failure has a worse prognosis than cancer,” says Christine Seidman. “The life expectancy [after diagnosis] is about 7 years.”

## The muscle-bound heart

Cardiomyopathies come in two main forms, hypertrophic and dilated. In hypertrophic cardiomyopathy (HCM), the walls of the lower heart chambers, or ventricles, thicken—often dramatically. This happens because the cardiac muscle cells increase in size, not number; the cells of the mature heart don't divide. In dilated cardiomyopathy (DCM), the ventricle walls thicken somewhat, but most of the heart enlargement is due to expansion of the ventricle chambers.

Early on, these changes cause little harm and may even aid pumping of blood by ventricles. But eventually the heart deteriorates as cells die, perhaps by apoptosis, and are replaced by scar tissue. This can result in either sudden death, caused by a fatal heart arrhythmia, or in deterioration of the heart to the point where it can't pump blood effectively—much as happens in ordinary heart failure—and the patient needs a transplant. “In this case, bigger is definitely not better,” says Michael Schneider of Baylor College of Medicine in Houston, Texas.

In the late 1950s, recognition that sudden death occurred in multiple members of some families prompted the realization that the cardiomyopathies could be hereditary. The research was further aided when echocardiography, which uses ultrasound to image internal organs, came along in the 1970s. Because this technique is noninvasive, it could be applied even to nonsymptomatic family members, some of whom also turned out to have enlarged hearts.

Geneticists then turned their attention to these families to look for the gene defects that might be responsible. In the late 1980s, Christine Seidman teamed up with spouse Jonathan Seidman, a molecular geneticist also at Harvard Medical School, to mount such a search. Over the next few years, they and their colleagues found that HCM could be caused by mutations in the gene encoding the myosin heavy chain and in the genes for troponin T and α tropomyosin.

These proteins help form the sarcomere, the contractile unit of muscle cells. At the time, that was a big surprise, says Christine Seidman: “We never guessed that the defects would be in structural proteins.” They expected instead that the mutations would be in genes that control muscle cell growth. Since then, however, the Seidman team and others have found that HCM can be caused by mutations in genes for several more sarcomere proteins, including actin and myosin-binding protein C.

Some of these mutated proteins cause more cases of cardiomyopathy than others. In the 6 May issue of Circulation, a multi-institutional French team, including Pascale Richard and Ketty Schwartz of the Hôpital de la Salpêtrière in Paris, reported the results of a study in which they screened 197 HCM patients for mutations. The myosin heavy chain gene was affected in 40% of the patients with identifiable mutations and the gene for myosin-binding protein C was altered in another 42%. Because mutations in those genes together cause most hereditary HCM cases, they are good candidates for screening tests, the team proposes.

Tracking down the genes involved in hereditary DCM turned out to be more difficult, but researchers have now come up with several. Among them are many of the same sarcomere genes identified in HCM families. Although the mutations are different, the finding was surprising because DCM and HCM have been considered distinct clinical entities. In addition, DCM can be caused by mutations in a variety of other genes. These include some of the genes that cause the various types of muscular dystrophy, which encode proteins associated with the muscle cell membrane; the gene for a nuclear protein called lamin; and the phospholamban gene.

## Calcium rerouted

The genetic diversity of the heredity cardiomyopathies raises a big question: How do mutations in so many different proteins, many of which are the fundamental structural proteins of muscle cells, lead to enlarged hearts and other pathological changes? For the most part, the proteins are not lost as a result—such mutations would presumably be lethal. Instead, the mutations apparently alter the contractile performance of heart muscle cells. Researchers are still working out exactly how altered contractility brings about cardiomyopathies, but they have some intriguing clues.

For example, recent work from Chien's group indicates that certain DCM mutations cause heart muscle to stretch more than it should, thus causing the organ to dilate. These findings grew out of experiments performed about 6 years ago by the UCSD team in collaboration with Pico Caroni's group at the Friedrich Miescher Institute in Basel, Switzerland. The researchers produced a condition much like human DCM in mice by knocking out a protein called MLP that is needed to maintain the normal architecture of muscle cells. “The animals have all the features of human heart failure,” Chien says.

Because other investigators had found that calcium handling is disturbed in cardiac muscle cells in ordinary heart failure, the Chien team tested calcium's involvement in the heart failure of the MLP knockout mice. Muscle contractions are initiated when a nerve impulse or other stimulus triggers entry of a small quantity of calcium ions into the cells. This causes the release of a much larger number of calcium ions from the sarcoplasmic reticulum (SR), a membrane-bound compartment that laces around the sarcomeres, which extend through muscle cells in an orderly, linear fashion. At a critical concentration, these ions trigger a contraction by the sarcomeres. The contraction ends when the calcium ions are pumped back into the SR, readying the heart muscle for another contraction.

The calcium pump is regulated by the protein phospholamban, which in its unphosphorylated form sits on—and inhibits—the pump protein SERCA2. When phosphorylated, which happens in response to stimuli that bolster heart function, such as the hormone adrenalin, phospholamban unlatches from the pump, thereby allowing it to move calcium ions into the SR more quickly. As a result the heart can beat more effectively. “This protein regulates the calcium levels [in cardiac muscle cells] and therefore the pumping activity of the heart,” says Evangelia Kranias of the University of Cincinnati in Ohio, whose lab also studies phospholamban.

About 4 years ago, Chien and colleagues engineered mice that lack both a functional MLP gene and the phospholamban gene. As predicted, the hearts of the double knockout mice pumped better than those of mice lacking only MLP. What's more, they didn't develop DCM and go into heart failure. More recent results may explain why loss of phospholamban is protective in the mice.

Chien's group, working with colleagues in Europe and Japan, found that MLP is part of the muscle cell's stretch sensor, which detects when the cell is being overextended so that it can take corrective action. In the absence of MLP, the team reported in Cell in December 2002, the heart muscle becomes dilated. But when phospholamban is also missing or inhibited, Chien suggests, relaxation of the heart increases, thus blocking dilation. “Inhibiting phospholamban essentially shrinks the heart and prevents the overexpansion,” Chien says.

A similar mechanism may underlie human cardiomyopathies. The team added MLP to the list of genes linked to human DCM by identifying a mutation in the protein that causes the condition in nine German families. Defects in two other proteins, titin and T-cap, also lead to cardiomyopathy; they, too, are part of the stretch receptor.

The importance of phospholamban in DCM has just received further support from the Seidmans (Science, 28 February, p. 1410). Working with Kranias and David MacLennan of the University of Toronto, they identified a phospholamban mutation that causes DCM in humans. Here the mutant protein apparently inhibits the SR calcium pump by tying up the enzyme that adds phosphates to phospholamban to release it from the pump.

All this suggests that therapies aimed at preventing phospholamban from inhibiting the SR's calcium pump might be used to treat ordinary heart failure. The approach works in animal models. Chien and his colleagues have produced a modified phospholamban that prevents the normal protein from inhibiting the SR calcium pump. When the researchers introduced the gene for the modified phospholamban into hamsters with a genetic cardiomyopathy, it improved their heart function. The gene also prevented the heart failure normally produced in mice by tying off one of their coronary arteries.

Still, there could be a fly in the ointment. Recent results from Kranias and her colleagues suggest that blocking phospholamban action might not always be a good idea. The researchers reported in the March issue of the Journal of Clinical Investigation that they have identified another mutation that causes human DCM. In this case, instead of producing an abnormal form of the protein, the patients' hearts didn't appear to make phospholamban at all. If so, such patients would be equivalent to the phospholamban knockout mice, which are resistant to DCM. But people with the mutation get such a severe form of the disease that two of them required heart transplants at the ages of 16 and 27.

Despite the fact that the researchers weren't able to detect phospholamban in patients' heart tissue, which was available due to the transplants, the possibility remains that the cells produce a short-lived version of the protein that could inhibit the SR's calcium pump, much as the mutant discovered by the Seidmans does, or cause other disturbances in cardiac muscle function. Kranias and her colleagues are now testing these possibilities.

Whatever the outcome, Kranias says that there's “no way” she's ready to give up on phospholamban: “The Seidmans showed exactly what we expected. Phospholamban is a great target” for heart failure therapies. She notes, however, that it may be necessary to screen patients to find out exactly what is causing their condition before treating them with a phospholamban inhibitor.

Disturbances in calcium ion handling are also involved in the other type of cardiomyopathy, HCM. Some evidence for this comes from studies by the Seidman team of mice into which the researchers introduced a myosin gene bearing one of the mutations found in human HCM. The hearts of these animals grow thick walls much like those of human HCM patients. Even before that happens, however, calcium ions shift from the SR to the sarcomeres in the heart muscle cells of the altered mice. Exactly what causes this calcium ion shift is currently unclear, but it could have a number of consequences, including altering gene expression and triggering apoptosis, in addition to disturbing the heart's ability to contract.

## Regression in gene expression

When hearts go into failure, whether due to cardiomyopathy, heart attack, or other stresses, they undergo a marked change in gene expression. Their cells shift from producing normal adult proteins to producing those that are normally turned on only during the fetal period. They also activate genes such as myc and fos that help regulate cell growth, as well as generally turning up protein synthesis. Recent evidence suggests that these changes may be triggered by the altered calcium balance in the cells.

For example, the protein calmodulin plays a critical role in sensing changes in calcium concentrations in cells and transmitting this information to a variety of enzymes that bring about responses to those changes, including alterations in gene expression. Olson's group at UT Southwestern has recently shown that calmodulin-dependent enzymes lead to cardiac hypertrophy by two different gene transcription routes.

In one, calmodulin activates so-called kinase enzymes that add phosphates to an enzyme called HDAC that inhibits fetal gene expression in adult heart muscle cells. This phosphate addition causes HDAC to move out of the nucleus into the cytoplasm and allows a transcription factor known as MEF2 to turn on fetal and other genes. The Olson team confirmed HDAC's role by knocking out the gene in mice. Its absence results in “massive hypertrophy,” Olson says.

In the second path, calmodulin activates an enzyme called calcineurin, which removes phosphate groups from transcription factors called NFATs. As a result of that removal, the NFATs move from the cytoplasm to the nucleus, where they turn on a variety of genes, also leading to cardiac hypertrophy. Activation of gene expression by NFATs can be shut down by an enzyme called glycogen synthase kinase-3, which puts the phosphates back on the transcription factors.

Consistent with that action, the UT Southwestern researchers found that they could inhibit development of hypertrophy by overexpressing the kinase gene in the hearts of mice. “That kinase is cardioprotective,” Olson says. “The mice are resistant to cardiomyopathy and heart failure [induced by] a variety of stresses.”

Still another link between calcium and activation of gene transcription comes from Schneider's group at Baylor. The enzyme RNA polymerase II (Pol II) copies most of the cell's protein-coding genes into messenger RNA, which is the first step in making the proteins. A few years ago, Schneider and his colleagues noticed that in failing hearts, a particular region of Pol II, the so-called carboxyl terminal, becomes phosphorylated, a change that helps enhance gene transcription.

Following up on that observation, the researchers showed that a variety of stimuli that cause cardiac hypertrophy, including calcineurin overexpression, increase the activity of the kinase that adds phosphates to Pol II. Answering the question of exactly how the stimuli produce that increase proved to be more complicated. The levels of the enzyme didn't change, and Schneider says, “we were frankly stumped at that point.”

In 2001, however, other researchers identified a small nuclear RNA that sits on the kinase, inhibiting it. And last year, the Schneider team showed that “every hypertrophic signal we tested caused the RNA to fall off the kinase,” Schneider says. Further evidence that the loss of kinase inhibition and the resulting increase in Pol II activity leads to hypertrophic growth came when the researchers artificially boosted the kinase activity in the hearts of mice. The animals spontaneously developed hypertrophic hearts. All in all, the results suggest that calcium ion disturbances contribute to cardiac hypertrophy.

## Cells' demise

In addition to uncovering the pathways that cause hearts to dilate or become hypertrophic, researchers are beginning to get a handle on what ultimately kills many patients: the death of heart cells and their subsequent replacement with scar tissue. Recent evidence from Schneider and his colleagues suggests that fraying of the telomeres, the structures that cap the ends of the chromosomes, may be a signal that triggers this cell death.

In work reported in the 29 April issue of the Proceedings of the National Academy of Sciences, the Baylor group found that the telomeres were markedly shorter—by about 25%—in the hearts of patients who died of heart failure than in people who died of other causes. The telomere shortening was accompanied by a reduction in the concentration of TRF2, one of the proteins normally present in the telomeres, and by activation of a so-called checkpoint kinase that triggers apoptosis in response to DNA damage. The TRF2 reduction apparently triggered those changes. Schneider and his colleagues showed that they could produce telomere shortening and apoptosis in cultured heart cells by blocking production of TRF2, whereas adding TRF2 blocked those changes.

There may well be other routes to apoptosis in heart failure, however. Gerald Dorn and his colleagues at the University of Cincinnati have produced a mouse model of cardiac hypertrophy by overexpressing a protein called Gαq in the animals. This protein is part of the pathway by which the cardiac muscle responds to stimulation by hormones such as adrenalin. Overproduction of Gαq effectively overstresses the heart.

To find out what causes the apoptosis that eventually develops in the hearts of the modified animals, the Dorn team used DNA microarrays containing 90 apoptosis-related genes. Four of the genes turned out to be overexpressed in the hearts of modified mice compared to normal animals. The researchers focused on one, known as Nix, because it works through the mitochondria, releasing a cell-digesting enzyme called a caspase that helps carry out the final cell destruction in apoptosis.

In work described in the July 2002 issue of Nature Medicine, Dorn and his colleagues found that if they overexpressed Nix in the hearts of mice, the animals developed severe DCM and died shortly after birth. “The seeds of destruction are planted in the [failing] heart,” Dorn says. But, he adds, the result is “actually a little bit cool” because it also suggests that caspase inhibitors, which are being developed to treat a variety of degenerative diseases, might be useful for treating heart failure.

Indeed, the current work on cardiomyopathies and heart failure has turned up numerous potential targets for therapies. The occasional death of a young athlete will continue to shock, but the research on their hidden killers is now turning up some promising leads that could eventually benefit a broad range of heart patients.

11. PLANETARY SCIENCE

# Running Water Eroded a Frigid Early Mars

1. Richard A. Kerr

New data suggest that precipitation—rain or snow—and flowing water helped shape Mars in its first billion years, despite the subzero climate

When Mars was young, did its valleys resound with plashing brooks and falling rain, or was the planet always cold and dry? Therein lies a conundrum. For decades, spacecraft have been returning images of lacy networks that look like systems of river valleys—evidence, said geologists, that precipitation-fed streams and rivers had heavily eroded Mars in its first billion years. Yet climate models have long had early Mars locked in a deep freeze even colder than today's, unable to have snow beyond the poles, much less rain. “There's been a complete discordance between geomorphology screaming ‘rain!’ and the climate models saying ‘It's impossible!’” says Mars geologist Michael Carr of the U.S. Geological Survey in Menlo Park, California.

In recent months, geomorphology has been winning out. New analyses of the first direct measurements of martian topography have reinforced the case for running water. “It looks more and more as if you do need precipitation,” says Carr. Planetary scientist Oded Aharonson of the California Institute of Technology (Caltech) in Pasadena agrees that “there's been a shift [toward] a somewhat wetter Mars, implying a more hospitable planet” in its earliest days, when any martian life would have been getting its start.

Yet the conundrum remains: Although there's now agreement that there must have been some precipitation on early Mars, how could it have occurred on a planet that was cold, if not severely frigid? Geologists are coming up with a number of possibilities. Perhaps early Mars was more like today's Dry Valleys of Antarctica, some suggest, where a dusting of snow melts and runs down short-lived rivers in high summer. Or perhaps it rained only on a rare summer's day.

The puzzle began with the Viking mission in the 1970s. In images returned by Viking and later spacecraft, geologists could see some features that seemed to have been gouged out by running water—the so-called valley networks. But debate has been dragging on over whether they were carved by rain-fed streams and rivers, the products of a “warm and wet” climate, or by a process called sapping in which spring water eats away at the rock face at the head of a stream. Sapping seemed more likely under a frigid, dry climate like that of today. But new evidence from Mars Global Surveyor is changing that perception.

Surveyor, which arrived at Mars in 1997, carries an instrument called the Mars Orbiter Laser Altimeter (MOLA), which bounces a light beam off a spot on the surface and measures its height the way a radar measures distance. After several years and 671,121,600 laser shots, MOLA provided geologists with a stunning, imagelike topographic map of Mars accurate to a few meters of altitude—versus the previous 1 kilometer uncertainty. Such precise knowledge of the martian surface should hold clues to how it was shaped.

Planetary scientist Tomasz Stepinski of the Lunar and Planetary Institute in Houston and his colleagues put the MOLA topographic data in a model designed to separate wet weather from dry. In essence, this model drizzles “computational rain” on a terrain and watches the virtual wet stuff drain off. Surfaces already eroded by rain-fed runoff, such as most areas outside of glaciated terrain on Earth, show virtual draining patterns quite unlike those of surfaces that no water had ever touched, such as lunar terrain shaped by impact cratering.

By watching virtual rain runoff, “we can easily say whether a terrain is lunar or terrestrial,” says Stepinski. “Mars terrain is somewhere in between terrestrial and lunar. Maybe there was some rain on Mars, but to a small degree. The [terrain] roughness is not dominated by fluvial erosion; it's a mix of fluvial erosion and cratering.” There was never enough sustained rainfall, for example, to dominate the heavy cratering of 3.9 billion years ago and earlier, says Stepinski. And there has certainly been no sustained rain since then, despite suggested episodes of a warm and wet climate in later Mars history (Science, 12 February 1993, p. 910).

Aharonson of Caltech and colleagues at the Massachusetts Institute of Technology in Cambridge have also found signs in the MOLA data that flowing water caused limited erosion on early Mars. The shapes of the main channels of drainage basins, they found, are flatter and less concave than rain-fed channels on Earth and have frequent abrupt drops. Those are both signs that flowing water did not have a chance to thoroughly reshape the initial form of the land. At first, Aharonson thought the MOLA data pointed to the low flows of groundwater sapping. But several recent studies combining MOLA data and the latest images from Surveyor and Mars Odyssey have now persuaded Aharonson and others that the limited erosion they are seeing was to a significant degree the work of precipitation.

Atypical such study by planetary scientists Brian Hynek and Roger Phillips of Washington University in St. Louis, Missouri, will be published in Geology. They set out to see just how densely valley networks cover the surface. Previous tallies had found that the treelike patterns of martian valley networks were unreasonably sparse—with few branches on a single short trunk and few twigs on the branches. Such low-density, “immature” drainage systems are typical of the feeble erosion of sapping on Earth. Hynek and Phillips overlaid MOLA topography on the higher quality though still less-than-definitive camera images from recent missions and then mapped out valleys as far as they could.

Using the latest data made quite a difference. “The combination of MOLA and the best [camera] images really clarifies where valley networks are,” says Hynek. “Now we see roughly an order of magnitude more valley networks than previously estimated.” That makes the martian valley networks about as dense as the least dense runoff-fed drainage systems on Earth and generally more dense than systems carved by sapping alone. Planetary geologist Robert Craddock of the Smithsonian Institution's National Air and Space Museum in Washington, D.C., and his colleagues got similar results when they identified drainage networks in MOLA topography using a computer program and checked their reality with the latest images from Mars, as they reported at last March's Lunar and Planetary Science Conference (LPSC) in Houston.

Moreover, some of the newly apparent networks reach nearly up to the crests of ridges and crater rims. That would be unlikely with sapping, as Mars geologist Alan Howard of the University of Virginia in Charlottesville pointed out at the Joint Assembly of European and American earth science societies held in Nice, France, in April. There's precious little ground up there to store the water needed for sapping, he noted, but precipitation could simply fall on the crest and run down.

All these signs of runoff erosion have forced sapping proponents, such as Carr, to take precipitation seriously. But the limited amount of erosion suggests that it wasn't the result of a “warm and wet” early Mars. As Howard pointed out at a session on “The Mysteries of the Martian Rivers” organized at the fall meeting of the American Geophysical Union (AGU) in SanFrancisco, something like a few hundred meters of the landscape were removed in the first half-billion to 1 billion years of Mars history. That's the amount of erosion that rain and melting snow can cause in just 1 million to 10 million years on a terrestrial desert, Howard notes. However runoff shaped Mars, he says, it must have taken its time. “It looks like a warm early Mars is dead,” observed planetary scientist Arden Albee of Caltech after listening to the AGU session.

Researchers are coming up with a variety of ways in which runoff could have eroded the surface of early Mars in a leisurely fashion. Last December atmospheric scientist Teresa Segura of the University of Colorado, Boulder, and her colleagues suggested that the heat of huge asteroid impacts could have momentarily mobilized frozen water and driven heavy rains for decades (Science, 6 December 2002, p. 1866). Tens of millions of years of cold, dry climate would have prevailed between impacts, stretching out the erosion over long periods of time.

At the LPSC last March, planetary scientists Pascal Lee and Christopher McKay of NASA's Ames Research Center in Mountain View, California, pointed to possible terrestrial analogs of erosion on early Mars in the high Arctic and Antarctica's Dry Valleys, where it is “always cold, sometimes wet.” On Mars, just enough warmth to melt snowfall at the height of summer might have been summoned when the planet tilted far over on its side every few million years (Science, 11 April, p. 234). And planetary scientist Eric Gaidos of the University of Hawaii, Manoa, and geochemist Giles Marion of the Desert Research Institute in Reno, Nevada, are suggesting that water might have gushed onto the surface of a cold Mars from time to time if the long-term cooling of the planet's interior progressively froze some deep groundwater, which in turn could have squeezed shots of water onto the surface. So there may be more than one way to wet a planet.

12. AMERICAN ASTRONOMICAL SOCIETY MEETING

# When Do Planets Form? Inquiring Astronomers Want to Know

1. Robert Irion

NASHVILLE, TENNESSEE—Nearly 700 astronomers convened in Music City U.S.A. for some down-home cosmic reports from 25 to 29 May at the 202nd meeting of the American Astronomical Society.

Solar-system theorists are hotly debating how long it takes to make a planet like Jupiter. In the popular “bottom-up” scenario, such planets first must build solid cores by accumulating lots of pebble-sized and kilometer-sized bits. The cores' gravity then attracts shrouds of gas from the disk of matter around newborn stars. Models suggest that this recipe could take up to 10 million years. A heretical “top-down” alternative posits that gas giants collapse nearly instantly—cosmically speaking—by means of unstable, runaway clumping of gas in the rotating disk (Science, 29 November 2002, p. 1698).

Score one for the heretics in the latest round of this debate. During a spirited session at the meeting, teams of observers presented new data suggesting that protoplanetary disks vanish within a few million years. “Nature is telling us we'd better find a new way to make gas giants faster than the conventional wisdom [allows],” says theorist Alan Boss of the Carnegie Institution of Washington, who champions the top-down approach.

In one striking study, astronomers Elizabeth Lada of the University of Florida in Gainesville and Karl Haisch of the University of Michigan, Ann Arbor, examined infrared and radio-wave emissions streaming from warm dust in several clusters where stars are now forming. The clusters ranged from 0.3 million to 30 million years old. Many dusty disks speckled clusters that were less than 1 million years old, but Lada and Haisch saw precious few of them in a 3-million-year-old cluster and none at an age of 5 million years. If dust vanishes on this time scale—probably by accreting onto the star or getting blown into space by stellar winds—then the gas should vanish as well, Lada says.

An ongoing search for emission from carbon monoxide gas around active young stars supports Lada's conclusion. Astronomer Joan Najita of the National Optical Astronomy Observatory in Tucson, Arizona, reported that thus far, her team has been “shocked” to find little evidence of residual gas close to stars between 1 million and 10 million years old.

The evidence against slow planet formation is compelling, many in the audience agreed. “Because of this meeting, I'm now on the fence, where before I was in the orthodox camp,” says infrared astronomer Dana Backman of Franklin & Marshall College in Lancaster, Pennsylvania. And during her talk, Najita confessed to a potential “religious conversion” to the disk-instability scenario, pending her team's further work.

The traditionalists haven't backed down. Astrophysicist Olenka Hubickyj of the University of California, Santa Cruz, noted that the most successful bottom-up theories could account for Jupiter-sized planets in 3 million years with a simple adjustment. Lowering the amount of radiation absorbed by dust grains in the disk—a quantity called “opacity”—would cool the disk and force it to contract more quickly, leading to faster accretion of gas. As for the disk-instability models Boss espouses, Hubickyj notes, “his initial conditions [of temperatures in the disk] aren't realistic compared to what we find in nature.”

Still, Boss heard enough comments similar to Backman's and Najita's that he feels the tide may be turning. “There are lots of giant planets out there, so whatever process makes them has to be very efficient,” he says. “It's a thought that's now crystallizing: Nature makes them fast.”

13. AMERICAN ASTRONOMICAL SOCIETY MEETING

# Hubble Prepares to Go Really Deep

1. Robert Irion

NASHVILLE, TENNESSEE—Nearly 700 astronomers convened in Music City U.S.A. for some down-home cosmic reports from 25 to 29 May at the 202nd meeting of the American Astronomical Society.

Two images from the Hubble Space Telescope's 13 years in orbit have attained iconic status: eerie pillars of star-forming dust in the Eagle Nebula, and the “Hubble Deep Field” of thousands of galaxies in a nondescript patch of sky. But by early next year, a new image may become Hubble's star. The telescope will spend about 25 days staring at a single region to create the “Hubble Ultra Deep Field,” a view that should reveal an epoch when the precursors of modern galaxies were just ill-formed blobs of stars.

The project is the brainchild of astronomer Steven Beckwith, director of the Space Telescope Science Institute (STScI) in Baltimore, Maryland. The original 10-day Deep Field, completed in 1995, has had the most far-reaching research impact of any Hubble program, he reported at the meeting. Now that the telescope is outfitted with an improved instrument, the Advanced Camera for Surveys (Science, 22 February 2002, p. 1450), Beckwith felt the time was ripe to once again take advantage of “the benefits of going fanatically deep.”

Hubble will scrutinize a small slice of sky in the Southern Hemisphere where the Chandra X-ray Observatory also has focused for a long period. The Ultra Deep Field should expose objects five times fainter than those in its predecessor, and with five times the clarity. Beckwith expects the image to resolve details of protogalaxies and other objects that existed when the cosmos was less than 1 billion years old. “We could get back to a time when the universe looked strikingly different,” Beckwith says. “It's likely to be the deepest visual observation of the universe for at least a decade.”

Exposures will begin in September and will take 4 months to finish, in several stages. By February 2004, Beckwith's team will release the data to the entire community at once; not even STScI astronomers will get first crack at analyzing the image.

That universal access helped make the first Deep Field a success, recalls Robert Williams, STScI director at the time. However, many regarded his project as a huge risk. “It wasn't clear what we would see, and a lot of critics thought we would see nothing at all,” Williams says. That's not the case with the Ultra Deep Field, which he applauds: “I think we'll see the key building blocks of the universe. It's going to be a spectacular picture.”

14. AMERICAN ASTRONOMICAL SOCIETY MEETING

# Gamma Ray Bursts Get Magnetic

1. Robert Irion

NASHVILLE, TENNESSEE—Nearly 700 astronomers convened in Music City U.S.A. for some down-home cosmic reports from 25 to 29 May at the 202nd meeting of the American Astronomical Society.

Cherished notions about gamma ray bursts (GRBs), the brightest explosions in the universe, have been upended by an unlikely source: a satellite that studies the sun with nine solid state detectors, each just a bit bigger than a soda can. Highly organized radiation streaming from the heart of a recent GRB, picked up by the detectors as they aimed at the sun, strongly suggests that the burst consisted almost entirely of electromagnetic energy rather than fast-moving matter as most theorists have assumed.

Astrophysicists think GRBs arise when the cores of giant stars collapse into spinning black holes within about 10 seconds. According to computer models, a thick torus of whirling gas plunges into the black hole while jets blast out of the top and bottom of the torus, where the material is less dense. If a jet points toward Earth, we see a blast of gamma rays; otherwise, the event looks like a supernova. These models have not included strong magnetic fields, which are extremely difficult to simulate. However, no observations had shown that magnetic fields played a major role in controlling the outbursts of energy from GRBs.

That changed on 6 December 2002, when a burst popped off close to the sun's position on the sky—too close for most telescopes to monitor. However, the gamma rays triggered nearly 100,000 electronic blips in the germanium detectors of the Ramaty High Energy Solar Spectroscopic Imager (RHESSI), a NASA satellite that studies flares on the sun. Some gamma rays scattered from one detector into an adjacent one before being absorbed. By tracing the patterns of this scattering, the RHESSI team found that the gamma rays were almost completely polarized, or aligned with their electric fields in the same direction.

The best explanation is that strong magnetic fields—perhaps more powerful than any others in the universe—ordered the burst's energy, says physicist Steven Boggs of the University of California, Berkeley, who reported the work at the meeting and in the 22 May issue of Nature. Without such fields, the gamma rays would have sprayed into space with random orientations. No other instruments had ever seen polarization at gamma ray energies, says Boggs, who called it a “milestone” for the field. “RHESSI was not designed for this, but we knew that if we got lucky with a bright burst nearly on axis [of the detectors], we'd see it,” he added.

Theorists at the meeting, who had heard rumors of RHESSI's feat for months, were thrilled. “This is absolutely an astounding result,” says astrophysicist Donald Lamb of the University of Chicago. “It's a slam dunk that magnetic fields are dominant.” Lamb suspects that a jet of charged particles ejected from the region near the black hole could not stay organized enough to produce such strong alignment. Rather, he says, most energy from GRBs may consist of electromagnetic radiation.

Physicist Maxim Lyutikov of McGill University in Montreal, Quebec, and colleagues reached a similar conclusion in a recent online paper (arxiv.org/abs/astro-ph/0305410). In their view, GRBs are particle-free “cold fireballs,” launched from an ultramagnetic and furiously spinning neutron star rather than from a black hole.

Watch for more GRB news soon. A forthcoming analysis of energy patterns from dozens of bursts by Lamb and co-workers, described at the meeting, suggests that magnetic fields confine the jets to needlelike cones just half an angular degree wide—far narrower than theorists have thought.

15. AMERICAN ASTRONOMICAL SOCIETY MEETING

# Snapshots From the Meeting

1. Robert Irion

Earths under fire. The most Earth-like planets yet found beyond our solar system exist in the most hostile place imaginable. Maciej Konacki of the California Institute of Technology in Pasadena described a 13-year study of radio signals from PSR B1257+12, a rapidly spinning pulsar that hosts three known planets. Rhythmic changes in the timing of the pulses confirmed the masses of the two outer planets: 4.3 and 3.9 times that of Earth, by far the smallest verified elsewhere. The pulsar's intense radiation should sterilize the worlds.

Boom times. If you like things that blow up, colliding galaxies are your scene. Arp 299 (shown here), a galactic merger 140 million light-years away, hosts a “supernova factory” inside a furnace of gas stoked by the crash. In April 2002, astronomers used 11 radio telescopes spanning the United States to peer through the dust with exquisite resolution. They saw four blast waves from new supernovas less than 350 light-years apart; less than a year later, a fifth flared into view. Giant stars may explode inside this cauldron every other year, says Susan Neff of NASA's Goddard Space Flight Center in Greenbelt, Maryland—providing a rare nearby look at “starbursts” common during mergers in the distant universe.

Red light, blue light. Old galaxies display a tired-out reddish glow, whereas young galaxies emit the blue blaze of new stars. Now another distinction is clear: The two groups clump together in very different ways. Astronomers with the Sloan Digital Sky Survey examined the clustering tendencies of 2 million galaxies, the largest such analysis yet. Red ones form tighter knots than blue ones, but there's no smooth transition between the two groups, reported Tamás Budavári of Johns Hopkins University in Baltimore, Maryland. That surprising result should help cosmologists figure out how dark matter segregated the universe as galaxies evolved.