# News this Week

Science  18 Aug 2006:
Vol. 313, Issue 5789, pp. 898
1. U.S. SCIENCE POLICY

# Congress Quietly Tries to Craft Bill To Maintain U.S. Lead in Science

1. Jeffrey Mervis

In the dog days of August, while most members of Congress are back home campaigning for reelection or on holiday, a small group of staffers is at work in Washington, D.C., on legislation that could influence science spending for years to come. Their goal is to craft a broad bill aimed at bolstering U.S. competitiveness that Congress could pass before the November elections.

They face long odds. The White House has already expressed reservations about some aspects of the legislation, and the congressional calendar is short and already very crowded. Although Senate leaders say they are committed to the goal, House leaders appear less enthusiastic. But a powerful coalition of forces, including business leaders who can bend a member's ear, is keen for Congress to act. “Legislation would show the public that our nation's leaders have a long-range plan of action on U.S. competitiveness,” says Susan Traiman of the Business Roundtable, a consortium of 160 CEOs from across U.S. industry.

The legislation draws upon several efforts over the past year examining the status of U.S. science and technology, including the National Academies' Rising Above the Gathering Storm report and the National Summit on Competitiveness (Science, 21 October 2005, p. 423; 16 December 2005, p. 1752). In February, the Bush Administration proposed starting a 10-year doubling of basic research at the National Science Foundation (NSF), the Department of Energy's (DOE) Office of Science, and the National Institute of Standards and Technology's (NIST) core labs (Science, 17 February, p. 929) as part of its 2007 budget request. And the initial funding for what the Administration has dubbed the American Competitiveness Initiative (ACI) is working its way through the legislative process.

Science advocates can't say enough about the importance of ACI. But they believe even more is needed to improve math and science education and enhance U.S. innovation. Taking their cue from Gathering Storm and other reports, legislators from both parties introduced a fistful of bills earlier this year that would expand existing research and education activities at several agencies and set up new programs (see table).

Unlike annual appropriations bills, which determine how much each federal agency can spend in a given year, these authorization bills set desired funding levels over several years. Although they don't provide the cash, they can build political support for ongoing spending increases. Notes one university lobbyist: “You want Congress on record and the key committees behind an authorization bill, so that they can bail out appropriators when they hit rough seas.”

The goal of the quiet negotiations taking place this summer is a single bill. But the calls for increased spending are a sticking point for a Republican Party whose president, George W. Bush, has repeatedly pledged to reduce the federal deficit and whose congressional leaders hope to campaign this fall on their success in shrinking government. Several of the bills also expand NSF's role in science and math education, a position that clashes with the Administration's plans for the Department of Education to lead efforts to improve math and science education and manage all the ACI's education components.

Presidential science adviser Jack Marburger emphasized those points in hard-line letters this spring to the chairs of the committees as they prepared to vote out one of the Senate bills (S. 2802) and two House bills (HR 5356/5358). The Senate measure, Marburger warned Senator Ted Stevens (R-AK) on 17 May, “would undermine and delay” ongoing research at the three agencies, “duplicate or complicate existing education and technology programs,” and “compete with private investment” in both areas. The House bills, he told Representative Sherry Boehlert (R-NY) on 5 June, “would diminish the impact” of the requested increases for the three ACI agencies.

Boehlert says he was “quite disappointed” by Marburger's letter, noting the president's declaration in his January State of the Union address that the country “must continue to lead the world in human talent and creativity.” Boehlert added, “I thought that we had been working with OSTP on these issues,” referring to the White House Office of Science and Technology Policy that Marburger heads.

Three weeks after the House committee passed both bills, überstaffer Karl Rove, new domestic policy chief Karl Zinsmeister, and a score of high-tech industry and academic lobbyists met at the White House to discuss the pending legislation. Although nothing was resolved—some participants say Rove and Marburger scolded them for supporting the bills, whereas others say there was confusion over the various components—the White House told the lobbyists that its Office of Legislative Affairs, led by Candida Wolff, would be taking the lead in trying to craft an acceptable bill, pushing OSTP to the sidelines. In the Senate, lobbyists are heartened by the willingness of Senate Majority Leader Bill Frist (R-TN) to negotiate with the three chairs whose panels must sign off on the legislation—Stevens, Senator Pete Domenici (R-NM), who leads the Energy and National Resources Committee, and Senator Mike Enzi (R-WY), who heads the Health, Education, Labor, and Pensions Committee. Another important player, Senator Lamar Alexander (R-TN), acknowledged when he introduced a trio of bills in January that some of his colleagues “may wince at the price tag” of the legislation. But he cautioned that “maintaining America's brainpower advantage will not come on the cheap.”

Although none of the staffers involved would speak on the record, several confirmed that talks are taking place “on a regular basis.” They say Frist is determined to cobble together a single bill—with lower authorization levels and fewer new programs than in any of the pending versions—that the Senate could adopt during a 4-week window in September. Prospects in the House are less certain, although Boehlert says, “Hope springs eternal that we'll get an opportunity to go to the floor in September.”

Optimists, who hope that all sides will view a competitiveness bill as an asset heading into the November elections, dream of an Administration that accepts a competitiveness bill in return for getting its ACI education programs authorized. Pessimists worry that the House leadership will scuttle the effort by portraying the bills as a vehicle for “wasteful spending” and “a bloated bureaucracy.” And although nobody's betting that Congress will act this year, nobody has thrown in the towel.

2. AVIAN INFLUENZA

# Panel Confirms Report of Early H5N1 Human Case in China

1. Dennis Normile

An international panel of experts has confirmed that China's first human death from H5N1 avian influenza occurred in November 2003, and not 2 years later as Chinese authorities had previously reported. The finding raises as many questions as it settles.

The case was first reported in the 22 June issue of the New England Journal of Medicine (NEJM) by Wu-Chun Cao of the State Key Laboratory of Pathogens and Biosecurity, Beijing, and colleagues at institutions mainly affiliated with China's military. In a strange twist, someone claiming to be Cao tried to withdraw the letter, but the magazine had gone to press; Cao later told the editors the request did not come from him (Science, 30 June, p. 1855).

To verify the results, China's Ministry of Health retested tissue samples and assembled an international panel of flu experts to review the results with the cooperation of the World Health Organization (WHO). On 8 August, the ministry announced the panel's conclusion: The death of a 24-year-old male in November 2003, from what were then called unknown causes, was actually due to H5N1. The death occurred 3 months before China reported its first H5N1 outbreaks in poultry and 2 years before it reported any human cases. Chinese media reported that Vice Minister of Health Jiang Zuojun said at a 10 August press conference in Beijing that the case indicates the need for researchers “to improve communication and contact with disease prevention organizations.”

Flu experts widely believe H5N1 has been circulating in southern China at least since the virus was first identified in Hong Kong in 1997. It has never been clear if it was undetected or if local or national authorities were withholding information from the international community—or whether they were even aware of how serious a threat the virus posed.

Reached by phone, Cao said his team concluded the man died of H5N1 only shortly before submitting their letter to the NEJM. He said he is willing to discuss the results with scientists but not reporters. “It's a very sensitive issue,” he said, declining to take further questions.

Roy Wadia, a WHO spokesperson in Beijing, says that when two members of a Hong Kong family tested positive for H5N1 after a trip to the mainland's Fujian Province in February 2003, WHO asked Chinese authorities to investigate. Wadia says the agency was told that the H5N1 virus was not present in Fujian. Confirmation that the virus was in circulation earlier than reported “begs the question of whether more aggressive action might have made a difference in the (near worldwide) spread of this virus,” he says.

3. ASTRONOMY

# Judge Slaps Hawaii Over Mauna Kea Telescopes

1. Jeffrey Mervis

A Hawaii state judge has found fault with the process of approving new telescopes at the world's largest astronomical observatory. A 3 August ruling by Third Circuit Court Judge Glenn Hara could affect the pace of development atop the 4200-meter Mauna Kea, a mountain with special significance to native Hawaiians.

The case involves a 2004 permit issued to the University of Hawaii's Institute for Astronomy (IFA) to house a quartet of 1.8-meter telescopes that would have worked in conjunction with the twin 10-meter Keck telescopes on the summit to hunt for planets outside the solar system. The $70 million Outrigger interferometry project, which NASA canceled in February because of a tight budget after spending$20 million on the telescopes and domes, would have operated in a protected area that requires a special permit from the state's Board of Land and Natural Resources (BLNR). But Hara said the board should not have approved Outrigger in the absence of a “comprehensive management plan” for the summit, which already hosts 13 telescopes. “The resource that needs to be conserved, protected, and preserved is the summit, not just the area of the project,” Hara wrote in an eight-page decision (Civ. No. 04-1-397, Mauna Kea v. BLNR). Although only Outrigger has been canceled, the ruling will affect all future development on Mauna Kea.

“Astronomers always want the next best thing, and they don't want any restrictions placed on them,” says Lea Hong, a Honolulu attorney who represented the plaintiffs, which included local groups and the Sierra Club. “My clients aren't antiastronomy. But they do want meaningful community representation in a process that respects the environmental, cultural, and aesthetic aspects of the mountain.”

The judge's ruling “highlights an ambiguity” in the current procedures, admits Frederic Chaffee, director emeritus of the Keck Observatory. The university, which manages activities at the summit, adopted a Mauna Kea master development plan in 2000, Chaffee notes, but the state never approved it or any similarly comprehensive plan. That loophole allowed critics of the Outrigger telescope project to argue successfully that the state was ignoring its own rules for managing the summit.

“The board needs to adopt a master plan,” agrees BLNR chair Peter Young, adding that there's no t ime to waste because a U.S. government-funded panoramic survey telescope (Pan-STARRS 4) project is moving ahead quickly (Science, 12 May, p. 840) and the enormous Thirty-Meter Telescope, being planned by a public-private consortium, is waiting in the wings (http://www.tmt.org/). “We plan to work with the university to come up with something that would incorporate both those projects and others down the road.”

IFA Director Rolf Kudritzki says that, in retrospect, the process used for the Outrigger project “wasn't the best way to proceed.” But he says, “I don't see a problem for astronomy” in the wake of the judge's ruling. Chaffee estimates that “we've already done 80% of the work” on a comprehensive plan for the summit in preparing the Outrigger permit.

4. SCIENTIFIC EXCHANGES

# U.S. Loosens Policy on Ties to UNESCO

1. Yudhijit Bhattacharjee

The United States government has withdrawn restrictions it placed a year ago on contact between U.S. citizens and the United Nations Educational, Scientific, and Cultural Organization (UNESCO). U.S. scientific societies are relieved by the move, which they say should help restore free exchange between U.S. researchers and the international body.

In May 2005, Louise Oliver, the U.S. ambassador to UNESCO, sent a memo to UNESCO Director General Koichiro Matsuura asking the organization to consult U.S. officials before partnering with anybody in the United States or planning any U.S. events. Last month, in a memo to Matsuura, Oliver effectively retracted that directive by explaining that the U.S. government merely wants to stay informed about contacts between UNESCO and U.S. entities.

Last year's memo seems to have been “misinterpreted by some individuals within the UNESCO Secretariat,” Oliver says in her 25 July letter. “We understand that there have been instances where UNESCO staff informed U.S. individuals and nongovernmental organizations that they were required to obtain U.S. government approval before making contact with UNESCO or before entering into any contracts with UNESCO.”

Wendy White of the U.S. National Academies, which last year wrote to Oliver expressing its concern, says she hopes the new memo will repair any breaches between the U.S. scientific community and UNESCO caused by last year's memo. But Irving Lerch, of the American Physical Society and Americans for UNESCO, wonders if the status quo can be restored. “Some links between U.S. organizations and UNESCO have already snapped as a result of last year's directive,” he says, noting that a UNESCO staffer recently declined a meeting invitation that had not been routed through the U.S. government. Moreover, Lerch says, the U.S. government still wants UNESCO to give it advance notification of any contacts with U.S. organizations—a step that he says hinders free exchange.

5. CLINICAL RESEARCH

# Lessons From a Failed Drug Trial

1. Eliot Marshall

Doctors who treated the six young men who became desperately ill in a botched U.K. clinical trial last spring have released an in-depth record of the catastrophe. They confirm, for example, that the volunteers were given intravenous doses of a test drug in quick succession (10 minutes apart), even though the drug had never been given to humans before. The subjects began to show signs of illness within 50 to 90 minutes, according to the report. And within 12 to 16 hours, all six were transferred from a company research site, which couldn't handle the emergency, to the Northwick Park and St. Mark's Hospital in London, which rescued them. The men appear to have recovered. But even 30 days after the test, according to a paper released this week from the 7 September New England Journal of Medicine (NEJM), some had “short-term difficulties in finding words (particularly names).”

This detailed account—by Ganesh Suntharalingam and colleagues in Northwick Park's intensive care unit—fills gaps in an earlier report on the incident by an expert panel advising the U.K. government.* The NEJM paper confirms that the dr ug in this trial, a monoclonal antibody called TGN1412, caused a massive immune response that flooded the volunteers' blood with inflammatory agents, triggering systemic organ damage (Science, 24 March, p. 1688). The doctors do not attempt to explain why this happened. But they conclude that TGN1412 itself, not an impurity, caused the injuries. They also speculate that TGN1412, which was designed to activate T cells and regulatory T cells at the same time, also may have directly injured the immune system and focused inflammation in the lungs. Four patients had to receive oxygen by mask, and two had to be put on mechanical ventilators. All six experienced severe and “unexpected” depletion of lymphocytes, cells that are essential to the immune system. The likely long-term consequences are not known.

Others say that data in hand before the trial make it clear that TGN1412 should have been tested with more caution. Nirmala Bhogal, a molecular pharmacologist who has analyzed the trial for FRAME, a nonprofit in Nottingham, U.K., that advocates substitutes for animal testing, says that one preclinical study of TGN1412 in monkeys revealed a proinflammatory response that peaked at 2 hours. The drug company that owns TGN1412—TeGenero of Wurzburg, Germany—discounted this before the clinical trial as a minor effect. However, Bhogal says, in light of the monkey data, “it defies all logic” to dose human volunteers at intervals shorter than 2 hours. TeGenero filed for insolvency in July, and company officials could not be reached for comment.

In its draft report issued last month, the expert panel advising the government, chaired by Gordon Duff, a University of Sheffield specialist in genetics and the human inflammatory response, suggested that when drugs are given to human subjects for the first time, there should be a pause for “an appropriate period of observation” before the next person is dosed.

The Duff panel also offered specific ideas for improving dose-risk calculations. Its broadest proposal is that drug companies and regulators around the world should collect and share unpublished data on human drug reactions. The panel suggested creating a new, open access database for everyone's use. The panel is gathering comments on these and other ideas before issuing a final report to the U.K. government in September.

6. MATERIALS SCIENCE

# New 'Supercapacitor' Promises to Pack More Electrical Punch

1. Robert F. Service

When it comes to powering laptops and hybrid cars, batteries get most of the attention. But these gadgets and myriad others also contain devices known as capacitors that provide quick bursts of energy. Capacitors can't store as much power as batteries, but the latest “supercapacitors” have started to close the gap. Now, their storage capabilities may be about to take another big jump.

In a report published online this week by Science (www.sciencemag.org/cgi/content/abstract/1132195), researchers from the United States and France report that by carefully controlling the nanoscale structure of a carbon-based supercapacitor, they've managed to increase the amount of electrical charges it can hold by about 50%. “It looks like they've got something significant there,” says John Miller, a physicist who runs JME Inc., a supercapacitor materials evaluation company in Shaker Heights, Ohio. If this performance translates to commercial devices, it could help manufacturers create smaller and cheaper power packs for everything from cameras to cars, Miller says. First, however, researchers need to learn more about how it works.

Typically, a capacitor contains a pair of electrodes surrounded by an electrolyte. When a voltage is applied between the electrodes, oppositely charged ions in the electrolyte snuggle up to each electrode and remain there even when the applied voltage is turned off. When the two electrodes are connected by a wire, electrons flow from the negative electrode to balance the charges in the positive electrode and do work en route.

For many years, carbon has been the electrode material of choice for supercapacitors because it conducts electricity, is light, and can be formed into a meshlike structure that sops up ions like a sponge. The smaller the pores in the material, the larger its surface area and the more charge the capacitor can hold—at least up to a point. When ions move through an electrolyte, other molecules attracted to their charge normally encircle them like groupies mobbing a rock star. Researchers have long thought that if the pores in a carbon supercapacitor got too small—below about 1 billionth of a meter, or nanometer—the ion would not be able to squeeze through with its entourage, and thus the material's overall ability to store charge would drop. But because they had no way to carefully control the pore size throughout a large capacitor, they couldn't test this notion.

Yury Gogotsi and his colleagues at Drexel University in Philadelphia, Pennsylvania, however, came up with a new way to do just that. They started with one of several commercially available compounds called a metal carbide, a mixture of a metal such as titanium and carbon. They then heated their material in a furnace while exposing it to chlorine gas. The gas reacted with the metal, forming volatile compounds that could easily be separated from the mixture, leaving behind carbon shot through with a continuous mesh of voids. By controlling the temperature and other conditions in their reactor, the researchers found they could tailor the holes in their carbon mesh to be a uniform size, between 0.6 and 2.25 nanometers across.

When Gogotsi and his students measured the charge-storing capabilities of the material, they got a shock. “We thought it would be useless” to study the smallest pores, Gogotsi says. But in powdered samples, their carbon with the 0.6-nanometer pores held 50% more charge than powders of standard supercapacitors. Gogotsi's group later teamed up with Patrice Simon, a leading supercapacitor expert at the University of Paul Sabatier in Toulouse, France, whose lab confirmed the results.

On a molecular level, it appeared that ions must be wiggling into the tiny pores, by either squeezing their entourage ions or perhaps abandoning them altogether. But how that could happen remains a puzzle, Miller says. In normal carbon supercapacitors, ions nestling up to an electrode form a layer about 1 nanometer thick. So if there is less space than that in the pores of the new material, it's not clear how they can get in. “That will be a bit controversial,” Miller says. But both he and Gogotsi point out that thanks to the newfound control over pore size, researchers should quickly be able to figure out just what is going on.

7. ASTRONOMY

# Nearby Cluster Shows Extremes of Stardom

1. Yudhijit Bhattacharjee

In the life of every small star, there comes a moment of reckoning when it stands on the edge between burnout and enduring brilliance. If the star's mass lies below a certain value, it runs out of nuclear fuel and begins to fade into a husk known as a brown dwarf. If its mass exceeds that value, the center of the star becomes hot enough to achieve a state of self-sustaining fusion, allowing it to burn merrily for trillions of years.

The critical mass, known as the brown dwarf limit, has been a fundamental prediction of stellar evolutionary theory. Now, for the first time, researchers have identified and measured this threshold in reality. On page 936, Harvey Richer of the University of British Columbia in Vancouver, Canada, and colleagues report the brown dwarf limit for stars in the nearby NGC 6397 globular cluster (above) and show that it matches the predicted value of 0.083 times the sun's mass. The researchers also report that the cluster's faintest white dwarfs—burnt-out remains of massive stars that grow dim as they cool over time—confirm another theoretical prediction, that white dwarfs turn bluer as they age.

The team's observation of the brown dwarf limit is “of prime importance” in helping theorists confirm their account of how stars evolve, says Gilles Chabrier, an astrophysicist at the École Normale Supérieure in Lyons, France. And by identifying the coolest white dwarfs in the population, Chabrier says, the researchers have taken “a key step toward determining the age of the cluster.”

For their study, Richer and his colleagues trained the Hubble Space Telescope on a section of NGC 6397 for 5 days at a stretch. “This was a very long exposure, so we could see fainter objects than had been seen before, even with this instrument,” says Brad Hansen, an astronomer at the University of California, Los Angeles, and a co-author of the paper. From a computer analysis of the images, the researchers were able to spot stars that were barely alight. Hubble could have detected stars that were fainter still, but the researchers didn't see any. That convinced them that they had identified the smallest stars capable of stably burning hydrogen in their cores.

Using a similar analysis, the researchers identified the faintest—and hence the coldest—white dwarfs. The observation bears out a prediction Hansen made in 1998: As spent stars get cooler, they emit radiation of longer and longer wavelengths, appearing redder in the process. But once a star cools to below 4000 kelvin, its atmosphere forms hydrogen molecules that absorb the redder wavelengths of radiation emanating from its core. As a result, the star's spectrum shifts from red to blue and gets bluer as the temperature falls. The white dwarfs in the study showed exactly that trend. “It's a wonderful illustration of quantum physics taking place in the atmosphere of stars,” says Chabrier.

The white-dwarf results open the door to establishing the age of the cluster, Hansen says: “It's like identifying the time of death of a corpse from the body temperature.” And because the NGC 6397 cluster is one of the oldest in the galaxy—as determined from the rarity of metals in its composition—learning its history would provide valuable insights into the early formation of the Milky Way, he says.

8. MEDICINE

# Desperate Measures

1. Jennifer Couzin

Twenty-five years after the first fetal surgery was performed, doctors and ethicists are trying to learn whether and when the drastic procedures work—and whether they're worth the frightening risks

ON 23 JANUARY 2002, SURGEONS CUT A 30-centimeter incision in Lorie Barber's abdomen, peeling away layers of tissue to reach her 23-week-old fetus. Delicately removing the uterus and slitting it open, the doctors at Vanderbilt University Medical Center in Nashville, Tennessee, stitched closed a gaping hole at the base of the fetus's spine. That opening was the signature left by spina bifida, which can cause paralysis, hydrocephalus, and other lifelong disabilities.

Thirteen days after the surgery, Nicole Eva Barber was born, more than 3 months early and weighing in at 1 pound and 10 ounces (740 grams). Nearly all fetal surgeries, the Barbers had been warned, carry a risk of premature birth. That hadn't deterred them.

Lorie Barber and her husband had come to Vanderbilt from their home in Ohio, desperate and devastated. Weeks earlier, a genetic counselor had discussed the diagnosis and presented two options: terminate the pregnancy or have the baby. The Barbers reached for a third choice they'd learned of over the Internet: fetal surgery that might offer their child a better life.

But for the Barbers, as for hundreds of other couples who have endured fetal surgery for a variety of conditions, there were no guarantees that the benefits of this treatment would outweigh its risks to both mother and fetus. Although roughly a dozen medical centers worldwide now offer fetal surgery, it remains highly experimental. Few fetal surgeries have been tested systematically in clinical trials, and for those that have, the results are decidedly mixed—suggesting anything from no advantage to robust benefit.

Part of the problem is that fetal surgeries are maddeningly difficult to evaluate in clinical trials. That's true of surgical interventions generally, and many enter mainstream practice without rigorous testing. But as diagnostic imaging advances, making it possible to visualize still more fetal anomalies potentially amenable to surgery, a growing number of physicians and ethicists are calling for trials to measure fetal surgery's worth against standard postnatal care. Perhaps more than anything, they fear that fetal surgeries, once confined to the most dismal cases, are becoming routine before their safety and effectiveness can be rigorously tested.

“Oftentimes, these therapies kind of take on a life of their own,” says Timothy Crombleholme, a pediatric surgeon and director of the Fetal Care Center at Cincinnati Children's Hospital Medical Center in Ohio, “and the window to evaluate them … goes away.” To keep that window open, fetal surgery centers banded together last spring to form a clinical trials network that they hope will speed testing of various fetal treatments, before they become entrenched.

## First breaths

Fetal surgery began in 1981 at the University of California, San Francisco (UCSF), as a last-ditch effort to save otherwise doomed fetuses. The hope was that by correcting a life-threatening defect early, surgeons could prevent further damage and save the fetus's life.

The first successful surgery, to repair a urinary obstruction that triggers kidney and lung failure after birth, resulted in a boy born alive who recently celebrated his 25th birthday. In the hands of UCSF pediatric surgeon Michael Harrison and his colleagues, rare conditions considered fatal sometimes proved no longer so. By following the natural history of certain diseases—in other words, how babies fared with standard, postnatal care—the physicians felt they could gauge fetal surgery's effectiveness.

As word spread about what the UCSF team was doing, “people would present [us] with a problem, often in the form of a patient, and say, ‘Do something,’” says Mitchell Golbus, an obstetrician and geneticist, now retired, who helped develop the UCSF program. In this way, the surgeries gradually spread to other life-threatening conditions. Among them was twin-twin transfusion syndrome, an often fatal circulatory disorder that strikes twins.

Another, congenital diaphragmatic hernia (CDH), occurs when abdominal organs migrate through a hole in the diaphragm to the chest in utero, compressing lung development and leaving newborns with inadequate lung capacity. The disease afflicts about 1 in 2500 babies worldwide, and all require surgery early in life. Intervening during fetal development, it was thought, might leave babies with larger, healthier lungs at birth and thus a much better chance of survival. When pediatric surgeons first began exploring fetal surgeries for CDH, about 30% of infants born with the condition survived.

In 1989, after 5 years of failed attempts in fetuses who died from the disease, UCSF performed the first successful CDH fetal surgery, closing the hole in the diaphragm. Although buoyed by their victories, even the most enthusiastic recognized that although they might be saving some very sick fetuses, the early surgeries had unsettling downsides. Some fetuses died from surgery itself, and others were born extremely prematurely. Moreover, some of the healthy women who underwent fetal surgery ended up in intensive care, hit dangerously hard by side effects from drugs given to prevent early labor.

“You have to make sure you have very good justification” for these surgeries, says Golbus, “because you're taking a healthy mother and running the risk of making her unhealthy.”

With that in mind, Harrison pushed for and led the first-ever clinical trial of fetal surgery. Begun in the early 1990s, the trial was designed to test so-called open surgery for CDH, the surgical approach that Lorie Barber endured for spina bifida. In CDH cases, the mother's womb is opened and the fetus partially removed for the operation.

Behind the scenes, the trial was a nightmare. Uneasy about the treatment's novelty, officials at the National Institutes of Health (NIH), which funded the trial, and UCSF's human subjects oversight committee took 2 years to approve it. Soon after the trial began, it was abruptly halted amid reports that women inside and outside the study who had undergone fetal surgery suffered pulmonary edema. The cause was traced to nitroglycerin, given experimentally to prevent early labor. The study restarted, finally ending 8 years after Harrison first proposed it. The randomized trial eventually compared the survival of four fetuses who had open surgery with seven who did not.

Logistics aside, Harrison and colleagues were convinced going in that CDH surgery would prove beneficial. “We thought for sure the randomized trial couldn't fail,” says Russell Jennings, then a fellow with Harrison and now head of the Advanced Fetal Care Center at Children's Hospital Boston. The truth was less kind. In a paper published in 1997 in The Journal of Pediatric Surgery, Harrison and colleagues reported that survival rates were 75% in the treated group and 86% in the control group, a difference that was not statistically significant, given the small numbers involved. One baby in each group died.

But around that time, two teams working with fetal lambs—one led by Harrison and the other by Jay Wilson of Children's Hospital Boston—found that they could correct the defect by blocking the trachea. This less invasive mechanical fix had a dramatic effect, keeping fluid pressure in the lungs high and forcing them to grow more rapidly. (The herniated diaphragm, surgeons found, could be repaired after birth.) “Right after our first sheep, we said, ‘This is it; we have cured diaphragmatic hernia,’” recalls Jennings.

A second trial testing this endoscopic technique, however, met with disappointment. Published in 2003 in The New England Journal of Medicine, that study found that 8 of 11 treated fetuses survived. But so did 10 of 13 in the control group. The treated babies who lived did have larger, healthier lungs, as the sheep studies had predicted, but those benefits were often muted by prematurity.

Although they didn't show survival advantages from fetal surgery, the trials did underscore risks to the fetus and the mother. Both open and endoscopic surgery greatly boosted the chance of premature birth. Babies in the open-surgery trial were born at 32 weeks, on average, and at 31 weeks in the endoscopic trial, roughly 6 weeks earlier than babies in both control groups. Since then, other risks have surfaced. Roughly 5% to 15% of women undergoing endoscopic fetal surgery experience a rupture in their uterine membrane, which puts the mother at risk of infection and may force early delivery of the baby—a complication that can also strike subsequent pregnancies. The risk is lower, about 4%, in open surgeries, says Scott Adzick, who runs the Center for Fetal Diagnosis and Treatment at the Children's Hospital of Philadelphia (CHOP).

Another reason for the confounding CDH results is that while the trials chugged along, survival odds were trending upward for CDH babies given postnatal respiratory support and surgery. At major pediatric centers such as Children's Hospital Boston, 95% of babies with CDH now survive, says Wilson, although many suffer long-term gastrointestinal and respiratory complications.

At least some surgeons still believe CDH fetal surgery offers the best hope for a healthy life. A variant on the endoscopic surgery tested in the second trial is now practiced regularly in Europe. More than 90 fetuses have been operated on so far, says Jan Deprest, a gynecologist at University Hospital Gasthuisberg Leuven in Belgium. Deprest says that his technique improves survival by 50%—but in Europe, say U.S. surgeons, CDH survival rates are lower than in North America (an assertion Deprest disputes). The surgeries, which focus on fetuses with the worst prognoses, have generated controversy, in part given the failure of other CDH fetal surgeries to show benefit. Both Deprest and Harrison are eager for yet another trial, to, as Deprest puts it, “have this discussion finished.”

Regardless, as postnatal medical care advances, many babies with CDH and other conditions who once perished now pull through. But often, they're left with lingering disabilities. That has physicians considering fetal surgery's power to enhance life's quality.

## A better life?

As the CDH trials continued, fetal surgery was stretching to accommodate its first non-life-threatening defect, spina bifida. “It really shifted … fetal treatment into another realm,” recasting the benefit-risk balance irrevocably, says Nancy Chescheir, an obstetrician at Vanderbilt.

Spina bifida arises very early in pregnancy, when the fetus's spinal cord fails to close. Children with the disability rarely die from it, but they often need shunts to drain fluid from their brains and suffer mobility, learning, and bladder and bowel problems.

Experiments on fetal lambs in the 1980s and early 1990s suggested that closing the wound in utero could reduce these complications. Fetal surgeons believe that sealing the opening may protect the spinal cord from continuing damage, perhaps by preventing exposure to amniotic fluid and normalizing fluid dynamics in the fetus's brain. Physicians at Vanderbilt proclaimed the first open fetal surgery for spina bifida in 1997, and hundreds of families streamed into Nashville.

The University of North Carolina, Chapel Hill, also offered the surgery. Sue Estroff, an anthropologist who chairs the university's maternal-fetal intervention advisory group, says that families typically came determined to proceed and weren't swayed by discussions of risks and benefits. “Our concepts of [informed] consent didn't fit what we saw,” she says. “People brought ideas about what it means to be a good parent.”

One was Lorie Barber. She imagined eschewing the surgery, only to have her child later grow up to say, “Mom, you knew about this, and it was available back then. Why didn't you try, why didn't you go for it?”

Barber was also encouraged by preliminary data from Vanderbilt suggesting that surgery might lessen the need for a shunt. In a paper published 2 years ago (although details were shared with families earlier as they accrued), Joseph Bruner, who oversaw spina bifida fetal surgeries at Vanderbilt and now works in Tennessee, and his colleagues reported that of 116 fetuses who had the surgery, 54% required a shunt before 1 year of age. The shunt rate for children who don't have fetal surgery has been estimated at as high as 85%, although it's thought to be drifting downward as neurosurgeons shunt more conservatively.

Even so, substantial questions about the surgery's benefit remained. For one, “spina bifida in humans happens at 8 weeks' gestation,” says Harrison. “We cannot work [on a fetus] at 8 weeks.” It's possible that by the time technology permits surgeons to operate—at about 20 weeks—the bulk of the damage has already occurred, making the drastic surgery largely futile. At the same time, because babies with spina bifida have an excellent chance of survival, life-threatening fetal surgery was creating ethically tenuous scenarios. At CHOP, says Crombleholme, who trained there before moving to Cincinnati, three babies with spina bifida who underwent fetal surgery were born so prematurely that they died. “These are three patients who would have survived,” he says.

Many physicians and ethicists became convinced that the only way to assess this surgery was in a clinical trial with a control group. With that in mind, NIH launched a trial in 2003, based at Vanderbilt, CHOP, and UCSF. Originally slated to end next year, the trial randomly assigns 100 mothers to surgery and 100 more to standard care. Success is measured by survival, the need for a shunt in the baby's first year, and neurologic function at 30 months.

By the time the trial began, physicians had performed more than 200 spina bifida fetal surgeries, and demand showed no signs of abating. To ensure that women would sign up for the trial, all hospitals halted spina bifida fetal surgeries outside the study.

Despite these efforts, recruitment has been sluggish. The trial was supposed to have begun a full year ago, but so far just 99 women have signed on. Explanations include a reluctance to be randomly assigned to either fetal surgery or a control group and a mother's unwillingness to remain at the surgical center until birth, as the trial mandates. But one thing is apparent: Continuing to recruit at this pace, “we're up to a 10-year trial,” double the time anticipated, says Chescheir. So far, NIH has alloted more than $14 million to it. Some surgeons quietly question whether spina bifida fetal surgery will survive. “The surgery itself is dying a slow death because of the length of the trial,” says Bruner. Unlike a drug, “surgery is a living, evolving entity,” he says. Doing a trial means “you have to freeze it in time,” halting the subtle enhancements surgeons routinely make. Safer, endoscopic approaches have not yet been effective at repairing spina bifida lesions, for example. As a result, in Europe, where open fetal surgeries are considered too aggressive to the mother, the procedure is not offered. ## The heart of the matter Some physicians are converging instead on another new frontier: fetal heart surgery. In the operating suites at Children's Hospital Boston, doctors now regularly perform procedures on fetuses with heart defects, 77 and counting since 2000. Although Boston is the only center in the world to have done more than a handful of these surgeries, says Wayne Tworetzky, a cardiologist there, other centers are considering whether to follow suit. Heart defects are ideal candidates for fetal surgery, Tworetzky and others say. High-tech fetal imaging has made diagnosis easier, and heart defects are a common and serious scourge in babies. The surgery to address them involves inserting a needle into the mother's abdomen and guiding it via ultrasound into the fetus's heart. But as with other fetal surgeries, the cardiac procedures are raising difficult questions of their own—in particular, whether cardiologists understand enough about the defects they're trying to fix in utero. Nor is it clear that they can identify the fetuses most likely to benefit. Take hypoplastic left heart syndrome (HLHS), the defect that the Boston team most commonly targets. Babies with HLHS are born lacking a functioning left ventricle, which leaves them with “only one pumping chamber,” says Tworetzky. Soon after birth, the infants turn ashen and struggle to breathe and feed normally. HLHS is not curable, and although most children can be treated with a series of operations or a heart transplant, their long-term prognosis is still shaky. Strategies to fix HLHS in utero, however, are complicated by questions about what's driving the disease. In some babies, HLHS seems to begin with a problem that is straightforward enough to fix: a blocked heart valve. Sophisticated tests on a pregnant woman can determine whether her fetus has this blocked valve. The surgery targets this obstruction in the hope that clearing it gives the left ventricle time to develop. However, although a blocked valve is certainly associated with the heart defect, it's not yet clear that it's the key culprit. Fixing it, then, might be less likely to help than it would be if the blocked valve were causative. “It's possible that these lesions which we consider primary … could be secondary, [and] relieving those would not necessarily improve muscle growth,” says Abraham Rudolph, a former chief of pediatric cardiology at UCSF who spent decades studying fetal circulation. And there's a second catch. Only a subset of fetuses with the blocked valve develop HLHS. Others are born with just the blockage, which can be corrected postnatally. Physicians at Children's Hospital Boston such as Tworetzky and cardiologist James Lock have done a number of studies to try to identify which fetuses with blocked heart valves go on to develop HLHS, because the risks of fetal surgery cannot currently be justified for the others. “We have strict criteria; you have to have this and this and not that,” says Tworetzky. In March, he and his colleagues published a paper in Circulation suggesting that certain types of blood flow in fetal hearts can predict HLHS—and thus which mothers and their fetuses are best suited for surgery. But other hospitals are hanging back. “There's logic to it, it makes sense, but it hasn't been rigorously tested,” says Jack Rychik, the head of CHOP's Fetal Heart Center, of the work in Boston. Last month, CHOP performed its first fetal heart procedure—but that fetus had multiple heart defects and an especially poor prognosis. Rychik wants firmer guarantees that he can pick the right mothers and fetuses for surgery and for now is not comfortable operating on all the same classes of women and fetuses treated in Boston. Instead, Rychik is working to bring together eight centers, including Boston, to create a registry of fetuses with various heart defects who would be followed until birth. “The Boston experience has given us a kick in the pants” to examine the natural history of HLHS and other defects before fetal heart surgery becomes routine, says Rychik, who adds that the therapy may soon merit a clinical trial. In April, 17 centers in North America launched the North American Fetal Therapy Network to create a single voice to advocate for and help develop fetal treatment trials. It hopes its endorsement of certain trial proposals will encourage NIH and other funders to supply the millions of dollars these studies can cost. Rychik and others are treading cautiously in part because families seek fetal surgery wherever possible. Even the Barbers, whose daughter spent 6 weeks on a ventilator and 103 days in a neonatal intensive care unit, say the price of surgery was worth it. Now 4 years old, Nicole's moderate hydrocephalus has not required a shunt. She's a strong-willed, talkative little girl who walks unassisted, attends a typical preschool, and enjoys bringing in the mail. Her mother has no regrets. 9. MENTAL HEALTH # Widening the Attack on Combat-Related Mental Health Problems 1. Greg Miller The lessons of Vietnam have prompted U.S. military leaders to do more to protect the mental health of troops in Iraq and Afghanistan. But will these efforts be effective? Airman C couldn't shake the image of the young Iraqi boy. Days earlier, a car bomb went off as his convoy drove down a busy city street, injuring three fellow soldiers and killing perhaps 10 Iraqi civilians. The boy sat at the edge of a street near a light pole. The explosion had torn off the right side of his jaw and opened his neck, exposing his esophagus. He reached up to Airman C, as if to ask for help, and called out “American, American, American.” But the word had just come in that the convoy was in danger of a second attack, and Airman C had to move on. After the attack, the image of the boy kept replaying in Airman C's mind throughout the day and at night, in frequent nightmares, says Alan Peterson, a clinical psychologist who treated the 22-year-old airman in the field for symptoms of posttraumatic stress disorder (PTSD). Peterson, who served with the U.S. Air Force as the chief psychologist at Balad Airbase near Baghdad until January 2005, knows firsthand how the tremendous stress and horrific situations faced by the troops can affect their mental health. And increasingly, he and others say, U.S. military leaders are showing a much greater willingness to acknowledge—and address—these problems than in the past. More psychologists and other mental health professionals have, like Peterson, been deployed to the frontlines of Iraq and Afghanistan. The Department of Defense (DOD) has instituted a universal screening program to monitor the health, including the mental health, of troops returning from combat. And several innovative programs aim to ease their transition to civilian life. Although it's too early to gauge the effectiveness of these interventions—and the screening program, in particular, has its detractors—many observers see them as a welcome sign of progress. “The leadership has taken a very proactive stance toward mental health issues,” says Charles Hoge, a psychiatrist at Walter Reed Army Medical Center in Washington, D.C. “We're well aware this time that there's an expected psychological cost of war.” ## Lingering effects War has always taken a toll on soldiers' psyches. After the Vietnam War, the chronic difficulties suffered by many returning troops—including flashbacks, nightmares, and feelings of detachment—inspired a new psychiatric diagnosis, PTSD, which officially entered the psychiatric lexicon in 1980 as an entry in the Diagnostic and Statistical Manual of Mental Disorders, third edition. From the start, however, gauging the prevalence of PTSD in combat veterans has been difficult and controversial. A study of Vietnam vets by the Centers for Disease Control and Prevention in Atlanta, Georgia, in 1988 put the prevalence of PTSD at 15%. But a short time later, the National Vietnam Veterans Readjustment Study (NVVRS), commissioned by Congress, doubled that estimate—concluding that 31% of Vietnam vets had suffered from PTSD at some point in their lives. Many researchers suspected that the NVVRS number was too high: Critics pointed out that the proportion of vets who'd suffered PTSD was twice the proportion who'd served in combat roles. A reanalysis of the NVVRS, which attempts to exclude vets whose symptoms weren't severe enough to interfere with daily living, now puts the figure at about 19% (see Report on p. 979 and Perspective on p. 923). But even that figure translates to more than 500,000 cases among Vietnam vets. Experts say it's impossible to know how many of the troops serving in Iraq or Afghanistan will develop PTSD or other combat-related mental problems, but preliminary data hint at a prevalence similar to the new figure for Vietnam. Soldiers in all three conflicts have faced similar stressors: a constant threat of ambush, a high casualty rate among both soldiers and civilians, and a difficulty distinguishing friend from foe. More frequent and more intense combat experiences raise the risk of PTSD, and there is no shortage of these, especially in Iraq. In the 1 July 2004 issue of The New England Journal of Medicine (NEJM), Hoge and colleagues reported that among 894 members of an Army combat unit that had recently returned from Iraq, 89% reported being attacked or ambushed, and 95% reported seeing dead bodies or human remains. This NEJM paper also provided one of the first looks at how soldiers are holding up under such conditions. In an anonymous survey, 17% of members of the same Army unit reported symptoms of PTSD, generalized anxiety, or depression. The researchers found a somewhat lower prevalence of such symptoms, about 11%, in 1962 soldiers deployed to Afghanistan, presumably reflecting their lower levels of reported combat experience. A second study by Hoge and colleagues, based on an abbreviated survey but a much larger sample of 238,938 Army and Marine personnel returning from Iraq and Afghanistan, produced similar findings. In the 1 March 2006 issue of The Journal of the American Medical Association (JAMA), they reported that 19% of those serving in Iraq and 11% of those serving in Afghanistan reported symptoms of PTSD or other mental health problems (see table, below). Although the figure for Iraq troops seems similar on the surface to the new figure for Vietnam, there's no way to make a definitive comparison, says Bruce Dohrenwend, a psychiatric epidemiologist at Columbia University who led the new reevaluation of the NVVRS. Whereas the raw data for NVVRS was collected in one-on-one clinical interviews lasting an hour or more, the data in Hoge's studies come from soldiers ticking off symptoms on a questionnaire. “I don't know how symptom scales compare to a rigorous diagnosis by a trained clinician,” Dohrenwend says. Concerns about the reliability of using surveys to detect mental problems have led some experts to question the effectiveness of DOD's new postdeployment health assessment program, instituted at the onset of the Iraq War. Every service member must now complete a three-page questionnaire that includes about a half-page of questions on mental health, either immediately before or within 2 weeks of returning home. Those who screen positive for a mental health problem get a follow-up interview with a clinician. The goal is to catch soldiers who need help early on, says Hoge, and get them treatment before their symptoms develop into a full-blown disorder or become compounded by family, alcohol, or drug problems, as happened to many soldiers after Vietnam. The 2006 JAMA paper was the first report based on data collected by the health assessment program. DOD recently announced plans to repeat the assessment at 3 and 6 months postdeployment. That's a good idea, Hoge and others say, because little is known about the time course of PTSD and because some problems might be masked by soldiers' initial euphoria over returning home. But Simon Wessely, a psychiatrist at Kings College London and civilian adviser to the British Army, dismisses the DOD screening effort as scientifically unfounded and a likely waste of resources. The main problem, he and others say, is that such surveys haven't proven effective at predicting which individuals will need mental health help. On one hand, the screen may tend to exaggerate problems by only tallying symptoms, says Richard McNally, a psychologist at Harvard University. It's possible to have a couple of symptoms of PTSD but not a full-blown disorder that requires intervention, he says: “Almost everybody is changed by the experience of fighting in a war. You have to draw a distinction between normal human emotions that are evoked by a horrible experience and things that impair daily life.” The follow-up interviews should help cut down on false positives, however, McNally says. A greater concern, in his view, is that the survey may fail to identify some veterans who need assistance. Peterson agrees. “The word on the street is, ‘Don't tell them you have symptoms, or you'll have to see a shrink,’” he says. Soldiers who take the survey before returning stateside are often tempted to hide symptoms to avoid delaying their return, and worries about the confidentiality are also widespread, Peterson says. Many soldiers may not check the box because they fear negative ramifications on their careers. The recent JAMA paper seems to bolster concerns that the screening isn't very accurate at identifying individuals who need help. Hoge's team found, for example, that fewer than 8% of servicemen seeking mental health care in the first year after their return had been referred by the screening program, suggesting that the vast majority of those who sought help weren't flagged by the program. At the same time, fewer than 20% of those who did report mental health concerns on the survey were referred to a mental health professional for further evaluation. In a May 2006 report, the U.S. Government Accountability Office chastised DOD for failing to get veterans mental health care when they need it. “The drive to do widespread screening is a laudable ambition, but it's driven largely by politics and the desire to be seen to be doing something and not by any evidence I've seen that it's doing any good,” says Wessely. ## In their own terms Wessely says he's advised the U.K. government that money and effort would be better spent on expanding mental health services and making them more amenable to veterans than on screening. He cites a U.S. DOD program called Battlemind, started by Hoge, as “a promising way forward.” In presentations to returning combat units and in videos and other materials available online for soldiers and their families, Battlemind explains how the combat-zone mindset can lead to problems at home. For example, in a combat zone, constant awareness of the surroundings is crucial for survival. But once soldiers return home, such heightened attention can leave them anxious and easily startled—a PTSD symptom called hypervigilance. The program appeals to soldiers because it uses language they understand and steers clear of mental health jargon, says Hoge: “We don't talk about hypervigilance; we talk about tactical awareness.” Meanwhile, Peterson, who has retired from the military and is now at the University of Texas Health Science Center in San Antonio, is developing a training program for Air Force mental health providers to teach them how to treat PTSD symptoms in the field. Airman C's symptoms were severe, and in previous conflicts, he would likely have been evacuated. But Peterson used a type of exposure therapy with him that has also proven useful with civilian rape and accident victims. In four sessions over the course of several weeks, Airman C talked through his ordeal in excruciating detail with Peterson, who made an audiotape of the session and asked him to listen to it at least once in the coming week. It's a painful process, but it helps desensitize people to the traumatic event, Peterson says. With time, Airman C's symptoms dissipated, and he began to feel more like himself, Peterson reported in a case study published last year. He now has funding from DOD to evaluate the therapy in a larger trial. “We think part of the key is early intervention,” he says. 10. RADIO ASTRONOMY # Candidate Sites for World's Largest Telescope Face First Big Hurdle 1. Robert Koenig For more than a decade, the Square Kilometer Array has been a paper project, an instrument for astronomers to dream about. Now it's time to start getting real CAPE TOWN, SOUTH AFRICA—One site is a barren tract of South Africa's Karoo semi-desert, so quiet that the only sound is a hawk's cry. Another, 12,000 kilometers away, is an arid plain that stretches to western Australia's horizon. Yet another site is across the Pacific Ocean, in Argentina's high, dry plateaus; the fourth is nestled in natural bowls between the angular karst hills of southeastern China. The challenge for a select group of radio astronomers this summer is to recommend which site would potentially make a suitable home for the largest astronomical instrument ever built: the International Square Kilometer Array (SKA). Planned as a network of some 4000 radio dishes spread over an area several thousand kilometers across, SKA—whose name refers to the total collecting area of the planned instruments—will be 100 times as sensitive as today's best radio telescopes. In choosing which locations will work, SKA's international steering committee is looking for a stable ionosphere, predictable weather, and good infrastructure. But radio silence, perhaps above all else, is golden, as is the host government's ability to maintain it. At a meeting in Germany at the end of this month, the committee is to whittle down, on scientific and technical grounds, the current four to a short list of acceptable SKA sites. Then begins the delicate political dance of persuading funders to bankroll it and agreeing on the final site. With a price tag likely to be at least$1 billion, the SKA collaboration is treading carefully. The project has already been working for more than a decade; a reference design for the telescope was completed only this year, and the array itself is unlikely to be finished before 2020. But the team doesn't want to force the pace and fall into the traps that have beset other recent international collaborations such as the International Thermonuclear Experimental Reactor (Science, 1 July 2005, p. 28) and the Atacama Large Millimeter Array (ALMA), a telescope under construction in Chile (Science, 19 May, p. 990).

From the start, SKA has been a grassroots project, and with scientists at more than 50 institutes in 17 countries involved, there is a strong emphasis on collaboration. Bo Peng of China's National Astronomical Observatories in Beijing says that SKA “has been a very successful international project for a decade, on the basis of cooperation, not so much competition for SKA science, technology, and even site ranking.”

In the competing regions, the site selection has energized radio astronomy and engineering with new instruments, more government backing, and scientific networking. “It is having a tremendous focusing impact on scientists,” says Netherlands-based astronomer Richard Schilizzi, director of the SKA project.

“The collaboration is fantastic,” agrees Australia's SKA planning office chief, astrophysicist Michelle Storey, “not only within Australia's radio community but also among astronomers from all the countries involved in the SKA.” Astronomer Justin Jonas, South Africa's SKA project scientist, foresees “a tremendous boost to astronomy and to science in general in southern Africa.” Jin Chengjin of the National Astronomical Observatories says his nation's radio-astronomy community is being opened up by the SKA effort because “the competition helps us in communicating with international astronomical and technological communities.” Ricardo Morras of the Argentine Radio Astronomy Institute says winning the SKA site would be “a major breakthrough in the history of radio astronomy in Latin America.”

## Probing dark matter and energy

SKA's main aim is to search for faint radio signals from the most distant reaches of the universe, helping scientists examine clues to what existed before the first stars were born and to probe the nature of dark matter and dark energy.

George Miley of Leiden University in the Netherlands says the instrument has the potential to be “a giant step forward for radio astronomy.” By exploiting new technologies such as steering the observing direction electronically instead of by moving the dish, unprecedented supercomputing power, and multibeaming—observing several regions of the sky simultaneously—SKA will attain orders-of-magnitude improvement in frequency resolution and the area of sky that can be observed at any given time.

Astronomer James Cordes of Cornell University, head of the U.S. SKA team, agrees that the telescope “will be a fantastic discovery instrument.” Schilizzi points out that “radio astronomy over the years has resulted in many unexpected discoveries, from the cosmic microwave background to pulsars to quasars to dark matter in galaxies.” By peering into the early universe, SKA “will give us a handle on the effects of dark energy and its evolution as the universe expands,” he adds.

Miley says data from SKA would help scientists study astrophysical phenomena that are impossible to probe using optical telescopes or millimeter arrays such as ALMA. Examples include neutral hydrogen, a diagnostic of the early universe in an era before the first galaxies formed, and synchrotron emission, radiation given off when electrons are accelerated—a unique probe of magnetic fields throughout the universe.

## Site competition, science cooperation

But there is much still to do before observations can begin. Officials of the four national site planning offices have been competing hotly behind the scenes. On 4 July, a parade of these officials presented their proposals separately at a meeting in Cambridge, U.K.

The Australians argue that their core site in Mileura and their proposed remote sites have the lowest radio frequency interference (RFI) and that all are located in Australia, with the option of extra stations in New Zealand. Australia has a “very strong tradition in radio astronomy,” Storey says, and several of its leading universities already are intensely involved in SKA research and development.

The South Africans contend that their site, northwest of the town of Carnarvon in the arid Karoo region, offers “very good radio-quiet status, excellent ionospheric and tropospheric conditions, strong government support and infrastructure,” says former astronomer Bernie Fanaroff, South Africa's SKA chief. Remote array stations would be located in seven other African countries, bolstering astronomy and engineering across the southern part of the continent.

The Chinese say their bid is strong because of the site's relatively quiet electromagnetic environment and because the karst depressions offer the possibility of building a smaller number of huge static dishes, which would “ease the correlation process and help in calibrating the network effectively,” Jin says. He adds that the valleys offer “good local shielding against radio interference from outside.”

The Argentineans maintain that their site, in a high, arid plain about 1100 kilometers west of Buenos Aires, offers the best combination of key factors. Morras says the proximity of existing or planned “frontline 21st century astronomical facilities,” such as ALMA, the European Southern Observatory's Very Large Telescope, and the Magellan 6-meter telescopes, all sited in Chile, “will allow simultaneous observations with a great number of telescopes operating on a wide range of different frequencies.”

Each site also has its drawbacks. Some astronomers worry that the ionosphere above the proposed Argentine location is less stable than that over the others. Morras concedes that “there are signal fluctuations, known as ionospheric scintillation, in that region.” But he says the Argentinean site's core “is near the southern limit” of the scintillation phenomenon.

The China site, meanwhile, is in the Northern Hemisphere, which would limit SKA's ability to observe our Milky Way galaxy, whose center is visible in the Southern Hemisphere. Another possible disadvantage, Jin concedes, is that the site's humid climate “is not suitable for observations at frequencies higher than 10 gigahertz.”

Regional politics may also become an issue. South Africa's roping in of Namibia, Botswana, Mozambique, Kenya, Madagascar, Ghana, and Mauritius has the advantage of boosting the bid's political clout but could also increase the risk that politics or economic problems might affect the remote stations. The Australian site, meanwhile, does not share any part of the sky with Europe and may involve higher construction costs.

In an effort to keep down levels of RFI from humanmade sources such as cell phones, TV transmitters, car ignitions, and power lines, Australia, South Africa, and Argentina are establishing “radio-astronomy reserves” around their candidate sites. Proximity to major cities could be another big RFI headache; one analysis suggests that the SKA core site would need to be at least 500 kilometers away from major urban centers. This could be another problem for the Argentinean site, which has two cities located about 100 kilometers away.

## Power and money

Although the details of the final design cannot be fixed until the site is known, SKA will likely include a network of 4000 or more small dishes (each about 10 meters in diameter) operating at frequencies from 1 to 25 GHz and aperture arrays (flat collections of detectors, looking a bit like solar panels, that can see many parts of the sky at once) operating from 100 MHz to 1 GHz, all connected together in a giant interferometer. Half of the total collecting area will be concentrated in the 5-kilometer-wide core, with the rest spread over several thousand kilometers.

SKA's current timeline calls for system design to begin in 2008—by which time the final site should have been chosen—and for construction to start on the first 10% of the collecting area in 2011. SKA would start collecting data in 2014 and would be fully operational by 2020. That scope and schedule depends on funding. So far, the largest contributions in the current Pathfinder phase have come from Europe (€10.4 million from the European Union and €28 million from national funding agencies) and from in-kind R&D contributions from South Africa and Australia related to their demonstrator projects (see sidebar, p. 911).

The biggest funding question mark is the U.S. National Science Foundation (NSF), which so far has avoided making a firm commitment. “We foresee that this will be an issue for the next Decade Survey, which is likely to start in about a year from now,” says the director of NSF's astronomical sciences division, Wayne Van Citters. He adds that, given NSF's commitment to complete ALMA and other major projects “that are more advanced in planning” than is SKA, “we are realistically quite a few years away from any consideration of a construction project of this magnitude.”

Over the past 4 years, NSF has provided about $1.8 million for SKA “technology development” in grants to Cordes's team at Cornell. The U.S. SKA team has also been kept busy with university funding and money from the likes of Microsoft Corp. co-founder Paul Allen. Allen has part-funded the Allen Telescope Array (ATA), formerly called the One Hectare Telescope, a joint effort by the SETI Institute and the University of California, Berkeley. The ATA project has involved developing many components relevant to SKA, such as relatively inexpensive antennas and mounts, broadband feeds, and some signal-processing hardware. Schilizzi says the American work on the Allen instrument “will be of crucial importance to the technology of the SKA.” NSF's initial reluctance to commit large sums to SKA was a major reason why the U.S. SKA consortium opted last year not to submit a site proposal, Cordes says. But Schilizzi says he is “still hopeful” that NSF eventually will allocate substantial money to the project. “It may be that funding will come in a phased way, with one region dominating the early funding and another region contributing more later,” he says. Such staggered funding would not be at odds with the slow but steady approach that SKA researchers have adopted. Despite the uncertainties, team members remain confident. Miley says that funding for large telescopes always seems to be up in the air: “In my view, the question is not whether SKA will be funded but rather when this will occur.” 11. RADIO ASTRONOMY # From KAT to FAST, Telescope Project Sprouts Test Beds 1. Robert Koenig High-tech radio-astronomy dishes are popping up in several remote areas of the world as demonstrators for the International Square Kilometer Array (SKA). Although each will be only a tiny fraction of the size of the planned SKA, the instruments will be capable of conducting cutting-edge science. Scientists in three of the candidate countries—Australia, China, and South Africa—have persuaded their governments to back the design and initial construction of prototype projects that are roughly 1% of SKA's dish area or smaller but that would demonstrate key technologies near the same core sites that are proposed for SKA itself. Australia's Extended New Technology Demonstrator (xNTD) is planned as a full-system prototype, making use of an innovative phased focal-plane array, a detector made of a patchwork of antenna elements that can steer its field of view electronically. The demonstrator, operating over the frequency range of 0.8 to 1.8 gigahertz, will be built at the Mileura site, with completion targeted for 2009. Others are in the works too: the SKA Molonglo Prototype from the University of Sydney for low-frequency radio astronomy, and the Mileura Widefield Array-Low Frequency Demonstrator, built in partnership with the Massachusetts Institute of Technology's Haystack Observatory. Meanwhile, South Africa's planned Karoo Array Telescope (KAT) is expected to have 20 dishes, each 15 meters in diameter, with innovative “smart” feeds in the focal plane. The South African government has committed the equivalent of about$50 million so far, and the prototype dish will be ready next year, with first light in 2009.

China's planned demonstrator instrument, the 500-meter-diameter Aperture Spherical Radio Telescope (FAST), would have a cable-supported reflector made up of 1800 hexagons—the world's largest single dish. Built in one of the karst depressions in Guizhou Province that are envisioned to be part of SKA, FAST “may be seen as a forerunner or prototype of the Chinese SKA concept,” says Jin Chengjin of the National Astronomical Observatories in Beijing.

Although FAST is a somewhat different design, astrophysicist Michelle Storey, head of Australia's SKA planning office, says KAT and xNTD “are similar in many ways.” That's a reason to be cautious, says Renee C. Kraan-Korteweg, who heads the University of Cape Town's astronomy department. It is important for the SKA competitors to make sure their demonstrator instruments and scientific goals complement one another. “Both the KAT and xNTD are interesting scientific instruments. But we need to communicate and avoid doing the same things,” Kraan-Korteweg says.

No matter where SKA itself is eventually located, director Richard Schilizzi says, “all of these projects will generate great science, and technically they are taking innovative approaches, for example in terms of utilizing focal-plane arrays.” They are also fostering a sense of community, says Kraan-Korteweg: “There is competition at one level, but there also is a great deal of collaboration, especially with the Australians and the groups at ASTRON [in the Netherlands] and in England. The Australian and South African groups have been sharing information in a cooperative way.”