News this Week

Science  22 Jan 1999:
Vol. 283, Issue 5401, pp. 464
1. BIOETHICS

NIMH to Screen Studies for Science and Human Risks

1. Eliot Marshall

For more than a year, patient activists and ethicists have been calling for a revolution in the way clinical studies in psychiatry are monitored. They argue that studies of psychiatric disorders, especially trials that exacerbate symptoms or withdraw medication, expose subjects to needless risks. And they have been demanding a more critical, independent review of protocols. These ideas have circulated widely in the press and dominated the work of a national bioethics commission in 1998. Now, a surprising new voice has joined the chorus of reformers: Steven Hyman, director of the National Institute of Mental Health (NIMH), the nation's largest funder of clinical research in psychiatry.

At the NIMH advisory council meeting on 5 February, Hyman plans to seek approval of a new review panel that would screen “high risk” human studies before NIMH agrees to fund them. The goal, he says, is to reinforce the rule of “beneficence”: NIMH needs to remind everyone that “we have the privilege of doing clinical studies only when the outcome is good and important.” In addition, he would like to pare away some of the repetitious “me-too” studies now in the portfolio. Hyman has already been applying such criteria to NIMH's intramural research, conducted in the institute's labs in Bethesda, Maryland. On 5 January, he stunned many NIMH lab chiefs when he and NIMH scientific director Robert Desimone suspended enrollment in 29 of 108 clinical protocols and asked that more than 50 be rewritten to clarify scientific objectives or human subjects protections.

The intramural shake-out and the plan for a top-level review of new research—which is sparking concern among some NIMH grantees—are “separate,” according to Hyman. But they are connected, he says, by a desire to make sure that the science in NIMH studies is good enough to justify the use of human subjects. Protocols that rely on human volunteers should be designed with “questions that are crisp enough to give a result. … You should have a real question that you want to answer, set it up well, and you don't have to keep replicating it.”

In an interview with Science, Hyman said the proposed new panel—which would review both intramural projects and extramural grants—would include ethicists and other outsiders. Constituted as a subcommittee of his advisory council, it would review and approve the funding of clinical trials after a local institutional board has vetted them for safety, and shortly after an NIMH study section has ranked them for merit. This final check would focus on human studies that seem risky—a term that remains undefined but clearly would include the kind of work that got negative publicity in 1998: studies that halt mental patients' ongoing medication, replace it with a placebo, or “challenge” them by exposing them to chemicals that intensify their symptoms. It would balance scientific objectives with human risks.

As for the intramural shake-up, Hyman says he and Desimone made these “rather dramatic” decisions after reading the comments of an ad hoc panel that met at NIMH on 8 to 9 December to assess “every active intramural protocol.” Hyman says he convened the top-level review—the most sweeping ever done by NIMH—because the institute is under orders to bring its own protocols into line with standards applied to extramural research. The 20-member group of outsiders, co-chaired by psychiatric researchers Dennis Charney of Yale and Jeffrey Lieberman of the University of North Carolina, Chapel Hill, examined scientific summaries and clinical descriptions for every protocol submitted to the NIMH institutional review board.

“No one had been doing anything that would harm [patients],” Hyman says. But “in many cases, [the summaries] weren't written in such a way that you could either clearly state the scientific hypothesis or understand exactly what was going to happen to the human subjects.” In the past, investigators wrote in a “flexible” way to take advantage of new ideas as they appeared. Now investigators are being asked to redraft summaries to make their methods and goals clearer. But a few protocols, Hyman observes, “have not aged gracefully,” and these are not likely to be continued. These will not resume. Desimone expects most of the others will be back on track by spring.

Hyman concedes that part of the reason for this intense scrutiny of intramural and extramural NIMH studies is that “there is currently a lack of public trust in how things are happening.” Indeed, calls for increased scrutiny of psychiatric research began at least a decade ago, when relatives began complaining that research protocols were taking precedence over the needs of patients. For example, filmmaker Robert Aller sued the University of California, Los Angeles (UCLA) in 1992, alleging that a decision to end medication for his son, who was in a clinical trial, made the son's schizophrenia worse. UCLA denied the allegations, and Aller lost in court. But a federal inquiry ordered procedural changes at UCLA. And later, the media and an independent federal panel, the President's National Bioethics Advisory Commission (NBAC), looked into appeals for outside monitoring of such research and found many justified.

On the East Coast, an advocacy group led by Vera Hassner Sharav, Citizens for Responsible Care in Psychiatry and Research of New York City, launched a similar campaign. Hassner Sharav steered patient-critics to NBAC in September 1997, where they presented claims that clinical researchers were needlessly distressing patients in challenge trials. Later she asked NBAC to look into the use of ketamine, an anesthetic that has been given in small doses to hundreds of mental patients to provoke psychotic symptoms. Ketamine has short-lived effects like the hallucinogen PCP, causing weird auditory and visual disturbances. Based on its own review, NBAC recommended new protections for mental patients (Science, 27 November 1998, p. 1617).

Clinicians say the heaviest blow, however, came in late November, when the Boston Globe ran a devastating four-part series full of research horror stories. It concluded with an editorial asking the Justice Department to conduct a criminal investigation into challenge and drug-withdrawal studies. Research leaders were shaken. Within weeks, Hyman spoke publicly about the need for a new scientific panel to approve risky research.

Researchers doing the kind of work that has been criticized tend to see Hyman's move as a surrender to critics, and they don't like it. “I think it would be foolhardy,” says William Carpenter, director of the Maryland Psychiatric Research Center at the University of Maryland, Baltimore. Carpenter's colleagues were among those criticized for conducting ketamine studies. Singling out clinical psychiatry for an extra review would be wrong, Carpenter says, because it stigmatizes that area of biomedicine. “It will discourage the best young investigators,” he thinks, for “why would you go into a field that has a politicized review process, when others don't?” Like his peers, he insists that other medical researchers use probes that are at least as risky as those used in psychiatry.

Carpenter worries that once created, the new safety panel will slip out of NIMH's control. Critics “will see an innate conflict of interest” in allowing NIMH to watch over its own studies, he warns, and will take it over and “politicize” it. As a warning, he points to the attack on ketamine studies, which he defends as an important way of learning about the efficacy of schizophrenia drugs. He claims that through ketamine trials, people are finding that widely used antipsychotic drugs don't block the underlying pathological brain activity. As for side effects, “ketamine doesn't seem to cause much anxiety or distress,” he says. At a 1 December meeting at NIMH, Carpenter reported that data from about 60 patients in ketamine trials at NIMH, Yale, and Maryland reveal few signs of distress. The effects in the worst cases lasted no more than 2 days, he said, and most effects were over within 90 to 180 minutes. However, two patients were distressed enough that they dropped out of the research.

Donald Klein, a professor at Columbia University and a psychiatrist at the New York State Psychiatric Institute, also feels that a national safety review panel for psychiatry could stifle research. His own work for the past 2 decades has involved inducing panic in people with panic disorder by injecting them with sodium lactate. It has led him to a theory that many cases of panic disorder arise from an innate derangement of the suffocation alarm, a hypersensitivity to carbon dioxide. This research could never have been done, Klein says, without challenge studies, and he wonders whether it would have been permitted by a national safety panel.

Although a few senior clinicians like these are hostile to Hyman's proposal, others are keeping their powder dry. The president of the American College of Neuropharmacology, David Kupfer, chief of psychiatry at the University of Pittsburgh, says he's pleased that NIMH is trying to be “proactive,” but doubts that a national safety panel can do better than existing, local ones.

Hyman is aware of resistance within his community. But he believes NIMH must move ahead with the reforms. Given the attention being focused on the ethics of mental health research, he said at a meeting last December, the community needs “to get our house in order.”

2. BIOMEDICAL RESEARCH

Ruling May Free NIH to Fund Stem Cell Studies

1. Eliot Marshall

Scientists eager to begin studies on two new types of human stem cells got some good news this week: The National Institutes of Health (NIH) announced that, contrary to what many had feared, U.S. law does not bar federal support for this burgeoning field. Grant money could be approved as early as this fall, according to NIH staffers. Researchers hope to use the cells for studies ranging from basic research on early human development to the development of new technologies for tissue transplantation.

NIH director Harold Varmus announced on 19 January that, in the Administration's reading, “current law permits federal funds to be used for research using human pluripotent stem cells”—cells that have the potential to develop into a wide variety of human tissues. During a talk at a meeting of the National Bioethics Advisory Commission (NBAC) in Washington, D.C. Varmus released a memo on stem cell research by Harriet Rabb, general counsel of the Department of Health and Human Services. Rabb makes it clear that there is no legal reason why funding of stem cell research cannot begin now.

Rabb's ruling sets aside some of the concerns that arose in November, when researchers first announced that they had derived stem cell lines from human embryo and fetal tissue (Science, 6 November 1998, p. 1014). NIH officials were concerned that congressionally imposed rules on some types of embryo and fetal tissue research might restrict the use of the new stem cells to private labs. For example, a clause added to the 1999 NIH appropriations bill makes it unlawful to spend federal funds on the creation of embryos “for research purposes,” and it blocks support of research in which embryos are “destroyed, discarded or knowingly subjected to risk of injury or death. …” An earlier statute also restricts interstate transfer and the therapeutic use of fetal tissue.

In her memo, Rabb makes a distinction between federal support for the development and the use of stem cell lines. The congressional language would prohibit the development of cell lines from embryos but not necessarily from fetal tissue, she wrote. (Both stem cell lines announced in November were developed with private funds.) But the law doesn't apply to the use of stem cells from either source, she said. The law focuses on making a human “embryo” or “organism” for research, she notes, but stem cells are not organisms—or even precursor organisms, in her view—for they cannot develop into an embryo even if implanted in a woman's uterus.

Varmus discussed these detailed legal issues at the first of six meetings NBAC is planning for an ethics review of stem cell research. NBAC hopes to have a draft report by June, according to its chair, Harold Shapiro, the president of Princeton University. But Varmus said he would like to know by March what NBAC thinks of plans to start funding human stem cell research. After hearing from NBAC, researchers, and the public, he hopes to draw up “clear guidelines” describing what can and cannot be done under the law. He plans to distribute the guidelines by midyear, then set up a standing review committee to monitor compliance.

“I am delighted to hear NIH made this decision,” said Senator Arlen Specter (R-PA), chair of the subcommittee that approves NIH's appropriation bill, noting that it does not violate the intent of earlier legislation. Specter added that “the last 60 days have seen breakthrough developments on stem cell research,” promising “enormous advances” in many areas of disease research.

3. PALEOCLIMATE

El Niño Grew Strong As Cultures Were Born

1. Richard A. Kerr

El Niño's comings and goings—and the worldwide effects of this tropical Pacific warming—are by now so familiar that they seem to be a permanent fixture of Earth's climate. But in this issue of Science (p. 516), a group of researchers reports that a climate record cored from the bottom of a lake high in the Ecuadorian Andes suggests a much-weakened El Niño between 5000 and 12,000 years ago—or even none at all.

Because older climate records show that in even earlier epochs, El Niño operated much the same as today, the new lake record points to an El Niño that waxes and wanes over the millennia. “It's a really interesting record, an important result if correct,” says paleoclimatologist Konrad Hughen of Harvard University. Some researchers argue that the onset of the modern El Niño 5000 years ago may have helped shape the emergence of civilizations around the Pacific, and its vacillations may give clues to our climate future in the greenhouse world. But it will take more records like the one from the Ecuadorian lake to fully persuade the cautious paleoclimate community that El Niño sometimes takes a break.

Paleoclimatologist Donald Rodbell of Union College in Schenectady, New York, and his colleagues were actually searching for evidence about the end of the ice age, not El Niño, when they cored the bottom of Laguna Pallcacocha, a lake 4000 meters up in the Andes of southern Ecuador. Although the ice age record they hoped for didn't show up in the 9.2-meter-long core, hundreds of sedimentary “zebra-stripe” layers did.

After chemical and mineralogical analysis, Rodbell concluded that the alternating layers of light, organic-poor sediments and dark, organic-rich sediments record the comings and goings of torrential rains spawned by El Niños. In non-El Niño years, moderate rains apparently wash only a little sediment, heavily laden with the dark debris from vegetation, off the steep, bowl-shaped slopes surrounding the lake. In El Niño years, the unusually warm waters just offshore fuel powerful storms that wash large amounts of sediment unsullied by organic matter into the lake, leaving a lighter colored deposit.

A quick look showed a marked shift in the pattern of zebra stripes partway through the record. After Rodbell's team set a time scale for the record with carbon-14 dating and used medical imaging software to count the stripes, they were able to date and quantify the shift. During the past 5000 years or so, the lake recorded extreme rains every 2 to 8 years, following the same rhythm that El Niño has exhibited of late. But before 5000 years ago, that pattern fades away. Extreme rains recurred only at intervals of a couple of decades to 75 years. Only weak El Niños could have persisted in this period, says Rodbell, as they did not spark torrential rains near the lake. But before 12,000 years ago—earlier than could be seen confidently in the lake record—El Niño was going strong, according to new records from western Pacific corals and even Great Lakes sediments.

“It's an interesting story,” says Hughen. Still, he says, “this is one record from one lake basin.” Although the lake seems to have been a faithful recorder of heavy rains during historical times, changes in the vegetation around the lake, for example, might have obscured the record of earlier strong El Niños. Rodbell and his colleagues agree that more records from other areas are needed to confirm that there were no strong El Niño cycles 5000 to 12,000 years ago. One new record in coral—from the other end of the El Niño system in Papua New Guinea—does suggest a much subdued El Niño 8000 years ago, according to work presented last month at the meeting of the American Geophysical Union in San Francisco by paleoclimatologist Sandy Tudhope of Edinburgh University in the U.K.

El Niño's temporary absence also fits into an archaeological scenario for the emergence of complex cultures on the west coast of South America. Using evidence such as the species composition of mollusk shells, which reflects ocean temperature, archaeologist Daniel Sandweiss of the University of Maine, Orono, has long argued that El Niño was shut down between 5000 and at least 8000 years ago. The onset of crop-nourishing El Niño rains by 5000 years ago sparked population increases, temple construction, and more complex societies on the Pacific Coast, he suggests (see his Perspective on p. 499). Critics such as geologist Lisa Wells of Vanderbilt University in Nashville, Tennessee, disputed some of his evidence, but Wells now agrees that the lake core backs his claim about El Niño.

But if El Niño did switch off, no one is sure why—so no one is sure of El Niño's future. From 5000 to 8000 years ago, the world was warmer than today, during the so-called Altithermal regime. If warmth alone had suppressed or eliminated El Niño, then it might again fade with the greenhouse warming of the next century. But if something else was the trigger—such as the changing strength of the seasonal cycle, which was greater at that time thanks to changes in Earth's tilt and orbit—the analogy between the Altithermal and the greenhouse might not hold. “It makes a lot of sense to look back at a previous warm period,” says paleoclimate modeler John Kutzbach of the University of Wisconsin, Madison, who has started to model paleo-El Niño, “but it's incredibly tricky.” Humans may have been familiar with El Niño for thousands of years, but that doesn't mean we understand it yet.

4. PALEONTOLOGY

Stunning Fossil Shows Breath of a Dinosaur

1. Bernice Wuethrich*
1. Bernice Wuethrich is a writer at the National Museum of Natural History in Washington, D.C.

Hoping to see the innards of a 100-million-year-old dinosaur, respiratory physiologist John Ruben hauled an 80-watt ultraviolet (UV) lamp from Oregon to an archaeology office in Salerno, Italy. His trouble paid off: The UV light, which can coax out patterns invisible in ordinary light, conjured up the outlines of the juvenile dinosaur's intestines, liver, trachea, and muscles. Now Ruben and colleagues are using the arrangement of internal organs to bolster their idea that dinosaur lungs were structurally simple, most resembling those of living crocodiles. Their analysis, presented on page 514, would imply that these animals were basically cold-blooded.

That's an argument Ruben has made before (Science, 14 November 1997, p. 1267), but now he adds a new twist. He thinks these simple lungs were also able to power periods of high metabolism and intense activity. If so, the old question of whether dinosaurs were cold- or warm-blooded would have a hybrid answer. “This is almost better than warm-blooded,” he says.

Not everyone is convinced, but many researchers are intrigued. “If they're right, this could represent landmark work suggesting a whole new way to view dinosaur physiology—it could in some sense be bimodal,” says anatomist Lawrence Witmer at the Ohio University College of Osteopathic Medicine in Athens. Researchers have suspected dinosaurs of having some sort of hybrid metabolism, adds paleontologist James Farlow at Indiana University-Purdue University in Fort Wayne. The skeletons of theropod dinosaurs—meat eaters like Tyrannosaurus rex—suggest that they were highly active like warm-blooded mammals, but their bones lack the signatures of warm-bloodedness. Coupling an economical resting metabolism with a capacity for bursts of activity may have been the best of all possible metabolic worlds. “It's not surprising that they ruled Earth for over 100 million years,” Farlow says.

Ruben, of Oregon State University in Corvallis, says the specimen of Scipionyx samniticus, a raptor or small meat eater, offers even more dramatic support for his anatomical argument than did a specimen he examined before, a 120-million-year-old Chinese fossil. Under the UV lamp, his team could see Scipionyx's liver, which extended from the top to the bottom of the abdominal cavity, just behind the lung-heart cavity, as well as a muscle next to the pubis bone. In modern crocodiles this muscle runs from the pubis to the liver and helps move the liver back and forth like a piston, causing the lungs to expand and contract. An airtight layer of tissue, the diaphragm, separates liver and lungs.

Finding this arrangement, called a hepatic-piston diaphragm, in theropod dinosaurs rules out the possibility that they breathed with a sophisticated birdlike lung, the kind that supports birds' high metabolism, says Ruben. But because raptors like Scipionyx were among the most dynamic dinosaurs, Ruben began to question the assumption that a bird or mammal lung is needed for high metabolism. And recent work by other scientists (Science, 3 July 1998, p. 45) showed that a well-ventilated reptilian lung might be capable of unexpectedly high rates of gas exchange. Because most reptiles lack the power of a hepatic-piston diaphragm, relying only on the action of their ribs to ventilate their lungs, Ruben reconsidered the advantages of the diaphragm.

The problem with this logic, says Witmer, is that the only living animals with a hepatic-piston diaphragm are the sedentary crocodiles. But Ruben argues that crocs' sluggish, aquatic lifestyle is a secondary development. He suggests that ancestral crocodilians were dynamic, bipedal land-dwellers who used a hepatic-piston system for vigorous activity—as did dinosaurs.

Ruben further argues that the lack of a bird-type lung in dinosaurs casts doubt on the idea that they gave rise to birds. Some paleontologists disagree, because birds could have evolved their lung later. Indeed, questions remain about the reliability of the fossil itself. “You can't take a squashed specimen and interpret the position and shape of any soft organ inside,” says paleontologist Phil Currie of the Royal Tyrrell Museum in Drumheller, Alberta, Canada. Ruben counters that although the fossil is two-dimensional, “nothing is displaced. … All [organs] are preserved in relation to each other.”

In any case, there's no doubt that the idea of metabolically hybrid dinosaurs is an appealing middle ground. Says Witmer: “In some ways everyone could be right.”

5. LIGHT POLLUTION

Development Blocked Near Tucson Telescopes

1. Mark Muro*
1. Mark Muro writes from Tucson, Arizona.

TUCSON, ARIZONA—Astronomers are applauding a decision last week by county officials to reject a $900 million development that could have brightened skies and degraded viewing conditions at three major observatories nearby. By a 4-to-1 vote, the Pima County Board of Supervisors turned down a plan to build 6000 homes and a large commercial district on a former ranch at the foot of the telescope-studded Mount Hopkins, 60 kilometers south of Tucson. Instead of savoring their victory, however, scientists have pledged to work harder to preserve dark skies on the outskirts of this rapidly growing southwestern metropolis. “We won this time,” says Craig Foltz, director of the soon-to-be-reopened 6.5-meter Multiple Mirror Observatory (MMT), the world's fifth-largest telescope, on Mount Hopkins. “But we need to get out there and fight to protect the really good sites in the world.” Pima County Supervisor Sharon Bronson agrees. “The astronomers made the difference this time,” she says, “and I hope they will provide the impetus to amend our light ordinances to make them even more progressive and protective of the industry.” For 35 years, astronomy and the rapid urbanization of southern Arizona have coexisted peacefully. In 1972 the community created the first outdoor lighting code in a major city to reduce streetlight glare and restrict business and home lighting without damaging residents' standard of living. More recently, the International Dark-Sky Association (IDA), the world's first advocacy group of its kind, has worked to preserve good seeing conditions for Kitt Peak National Observatory, Whipple Observatory, and the University of Arizona binocular telescope being erected atop Mount Graham to the east. “Tucson is the pioneer; it practically invented light sensitivity,” observes Frederic Chaffee, director of the W. M. Keck Observatory in Kamuela, Hawaii, and a former director of the Mount Hopkins Observatory. “That's what's so distressing about this blowup.” The dispute burst into flames in December after smoldering for 3 years. Fairfield Homes of Green Valley, Arizona, had asked county officials to rezone 5700 acres of the undeveloped Canoa Ranch near the mountain's base for a residential and commercial development, including four times the number of homes permitted under current zoning plus offices, stores, and an airstrip. In October, a company-backed study requested by the county's Outdoor Lighting Code Committee said the development would produce less light pollution than would haphazard growth and a negligible increase in sky brightness. That estimate, however, was quickly challenged by astronomers. A study by Foltz and Chris Luginbuhl of the U.S. Naval Observatory in Flagstaff, Arizona, showed a potential 8% to 14% increase in sky brightness. Robert Kirshner, associate director of the Harvard-Smithsonian Center for Astrophysics in Cambridge, Massachusetts, which runs the Whipple Observatory, said such levels would “significantly compromise the usefulness of$220 million of taxpayers' investment” on Mount Hopkins. The developers' attorney, Frank Cassidy, dismissed the scientists' calculations and late last month talked about suing the Smithsonian for $900 million for “improperly” interceding in the zoning process. What followed were 2 weeks of heated public debate over how to reconcile rapid development in the Sun Belt with world-class astronomy. As astronomers circulated electronic petitions and faxed letters to county officials, Cassidy complained that the observatory's staff had avoided discussions of how to minimize the amount of light pollution and overstated the bright commercial development. Whipple's spokesperson, Dan Brocious, told a local newspaper that astronomers “have a duty to speak out” about potential actions that jeopardize telescopes like the MMT, now completing a$20 million expansion.

Now that the Canoa fight has died down, both sides seem to be taking the long view. “Nobody wants to hurt the observatory,” says David Williamson, president and chief executive officer of Fairfield Homes, who does not rule out resubmitting a scaled-down development plan. And astronomers vow to push for tougher light-emission standards later this year when public officials begin revising the area's pioneering lighting ordinance. “What this controversy revealed was that we need tighter controls in closer to the observatories,” says Dave Crawford, a founder of IDA and member of the lighting code committee. But some scientists suspect it's a losing battle. “With all this growth, you realize you can only stay so long in a place,” says Foltz. “And that's too bad.”

6. CREUTZFELDT-JAKOB DISEASE

Diagnostic Test Scores High Marks in Study

1. Michael Balter

Since 1996, when a new human brain disorder linked to eating beef from cattle infected with “mad cow disease,” or BSE, was first identified in the United Kingdom, health officials have been waiting for the other shoe to drop. Although only 34 cases of the disease—called variant Creutzfeldt-Jakob disease (vCJD)—have been confirmed so far in Britain, no one knows if these are isolated occurrences or the first signs of a major epidemic. Getting a handle on this crucial question has been made more difficult by the lack of a diagnostic test for vCJD. Now, help may be at hand: In the 16 January issue of The Lancet, a team of U.K. researchers reports that sensitive immunological and molecular tests can detect in tonsil biopsies an abnormal protein linked to vCJD.

Previously, the primary way physicians and researchers have diagnosed cases of vCJD has been through examination of brain biopsies taken from patients in advanced stages of the disease or, more commonly, after they had died. But a team led by neurologist John Collinge of St. Mary's Hospital in London has recently developed a different approach. Their tests seek to identify an abnormal form of a biomolecule called the prion protein (PrP) that is a signature of vCJD. The study shows that the tests can distinguish vCJD not only from normal controls but from other forms of CJD caused by different prion “strains” not linked with BSE, as well as from other prion-caused diseases.

With some reservations, the study is being welcomed by researchers as a first step toward a diagnostic test that could detect vCJD at earlier stages, as well as a tool for epidemiological studies. Britain's Medical Research Council and Department of Health are currently hatching plans to use the new test as part of a mass screening program of previously stored tonsils and appendixes, which might provide better estimates of how big an epidemic the country might be facing (Science, 4 September 1998, p. 1422).

The study was inspired by an earlier finding, reported by Collinge's group in 1997, of abnormal PrP in the tonsils of a patient who had died from vCJD. For the recent study, the team collected a large number of samples of lymphoid tissues—including tonsils, spleens, and lymph nodes—from a variety of sources. These included tonsil biopsies from 20 patients suspected of having some sort of prion disease—nine of whom were later shown to be suffering from vCJD. The rest were from tonsillectomies or stored autopsy materials taken from vCJD victims, normal controls, and sufferers of other neurological diseases, including “classic” forms of CJD not linked to BSE.

In search of PrP, Collinge's team then subjected samples to two types of laboratory test. The first, immunohistochemistry—in which a target protein is “stained” with antibodies that specifically recognize it—can tell if PrP is present but not whether it is abnormal. The team found that PrP was detectable in lymphoid tissue only among vCJD sufferers; they found no PrP in any of the other samples. The second test, known as Western blotting, detects proteins both by their molecular weight and their reactions with antibodies. Collinge has claimed that this test can distinguish different prion strains because they have different patterns of sugar residues on their surface and hence different molecular weights. The new results may support that claim: They confirmed that only vCJD sufferers had PrP in their lymphoid tissue and found that all the vCJD patients shared the same prion strain.

The study “looks very convincing,” says Oxford University epidemiologist Roy Anderson. And molecular biologist Chris Bostock, director of Britain's Institute for Animal Health, says that the new test “looks like a promising tool, along with others, to confirm diagnosis of vCJD.” But researchers caution that use of the test for wide-scale screening raises serious ethical questions. For example, with no cure for vCJD in sight, should people be told they are harboring the disease? “The situation is analogous to the early stages of the AIDS epidemic,” says Anderson. Health officials are therefore considering an anonymous screening program, for research purposes only.

Yet some researchers say that screening tonsil tissues could give rise to misleading data on the extent of the epidemic. Moira Bruce of the Institute for Animal Health's Neuropathogenesis Unit in Edinburgh points out that tonsils and appendixes are normally removed because they are inflamed and flooded with immune cells such as lymphocytes. “We know that expression of PrP on lymphocytes is elevated as part of the immune response,” Bruce says, so such a screening program could lead to “false positives” and overestimate the infected population.

On the other hand, some researchers believe abnormal PrP may be undetectable in the disease's early stages, and because nobody knows at what stage the protein moves from infected beef in the gut to lymphoid tissues, screening might underestimate the epidemic. Nevertheless, screening will be required if health officials are to know what they are up against. Says Collinge: “It would be irresponsible not to make use of this test. We might find evidence of a major problem, and we need to know sooner or later.”

7. DEVELOPMENT

Brain Stem Cells Show Their Potential

1. Evelyn Strauss

Brains memorize organic chemistry equations, control typing fingers, and integrate the sensory input needed to navigate snarled traffic—all feats of profound sophistication and versatility. Now the brain is proving itself even more of a renaissance organ, for some of its cells can perform the tasks of a completely different tissue.

On page 534, Angelo Vescovi, a neurobiologist at the National Neurological Institute Carlo Besta in Milan, Italy, and his colleagues report that neural stem cells, which give rise to the three main types of brain cells, can also become blood cells when transplanted into mice whose own blood-forming tissue, the bone marrow, has been mostly destroyed. It wasn't until the early 1990s that scientists found ways to isolate neural stem cells and grow them in the lab. “Now the brain's making blood,” Vescovi says.

“What's interesting is the idea that cells can shake their fates,” says Ron McKay, a neurobiologist at the National Institute of Neurological Disorders and Stroke in Bethesda, Maryland. The result provides a strong push to find other stem cell types with similar capabilities. And it opens the possibility of using neural stem cell transplants to treat human blood cell disorders such as aplastic anemia and severe combined immunodeficiency—an appealing idea, as bone marrow stem cells don't replenish themselves well in lab cultures.

Several observations had hinted at brain cells' versatility. During development, muscle cell types—which arise from a layer of embryonic cells distinct from that which generates the brain—appear in the central nervous system, says Vescovi. Furthermore, scientists sometimes see muscle tissue in a particular type of brain tumor. Because no one knows where the muscle cells come from, “we theorized that maybe there's a brain cell that possesses a much wider potential for differentiation than previously thought,” says Christopher Bjornson, a developmental biologist currently at the University of Washington, Seattle, and a co-author of the Science paper.

To find out, the team isolated neural stem cells from adult and embryonic mice and grew them singly in lab cultures. After irradiating mice to kill most of their bone marrow cells and create a vacancy that new cells might occupy, the researchers injected the neural stem cells into the animals. Because the donor mouse cells carried distinctive genetic markers, the researchers could trace their fate in the injected animals.

Five months later, the investigators found that the blood of the recipients contained cells that not only displayed the donor cell marker protein but also produced proteins that only mature blood cells make. They also showed that the animals' bone marrow carried immature blood precursor cells that were descended from the neural cells.

No one knows exactly what caused the neural cells to turn into blood cells. But Vescovi and his colleagues suspect that the neural cells might be responding to the same signals that normally stimulate the few remaining blood stem cells to reproduce and mature after irradiation wipes out most of the bone marrow. “The result suggests that there's something quite powerful in the mature adult blood system that can instruct cells from a different origin what to do,” says Arturo Alvarez-Buylla, a neurobiologist at The Rockefeller University in Manhattan.

Whatever that is, its effects appear to be long-lasting. The Vescovi team could detect the neural-derived blood cells a year after the injection. That means that the transplants may persist long enough to be clinically useful, the researchers say. If the Vescovi team's work can be replicated in humans, agrees Irving Weissman, a stem cell biologist at Stanford University School of Medicine, the neural cells “could become a source for blood stem cells.”

8. PARTICLE PHYSICS

Exploring the Proton Sea

1. Andrew Watson*
1. Andrew Watson is a writer in Norwich, U.K.

Each time physicists probe the teeming interior of the proton, aswarm with short-lived particles, they seem to turn up more surprises

In the Dr. Seuss story “Horton Hears a Who!,” Horton the elephant insists to his fellow animals, all deeply skeptical, that a speck of dust is teeming with life. With his sensitive ears, Horton can hear the chatter and buzz of its microscopic inhabitants—whole cities of them. Physicists studying the humble proton will understand his fascination. To most researchers, the proton is a workaday particle: the stuff that gives every atomic nucleus its positive charge, and the heart of the ubiquitous hydrogen atom. But recent studies probing deep into the proton are revealing a society as complex as the one on Horton's dust mote: a churning and bubbling sea of “virtual” particles that pop into existence for an instant, then disappear again, bathing more enduring components of the proton in a quantum flux.

The ephemeral nature of the sea's inhabitants, mass- and charge-carrying particles called quarks and force-carrying particles called gluons, belies their importance. “This virtual sea is responsible for many of the proton's properties, such as its mass, its structure, and its interaction with other particles and fields,” says Michael Leitch of the Los Alamos National Laboratory in New Mexico. Charting the sea is also important for future experiments: The world's most powerful particle accelerator, the Large Hadron Collider now being built at the CERN particle physics lab near Geneva, will slam protons together at enormous energies. One aim is to create the Higgs boson, the particle thought to endow all others with mass, which has been on physicists' “most wanted” list for 3 decades. Knowing what is in the proton is essential for calculating what will come out of those collisions. “If new physics is to be discovered, we need to understand the predictions from the old physics with some precision,” says Arie Bodek of the University of Rochester in New York.

Yet the normal theoretical apparatus used to describe the subatomic landscape can make few predictions at the energies found in the proton's interior. As a result, physicists found themselves in uncharted waters as they began exploring the interior of the proton by probing it with beams of other particles. Lately, a series of experiments at accelerators in Europe and the United States to measure the different types of quarks in the proton sea, compare the proportion of quarks to gluons, and identify differences in the quark sea of the proton and that of the proton's sister particle, the neutron, have delivered a string of surprises. As Anthony Thomas of the University of Adelaide in Australia puts it, “Every time we have tested a prejudice about the sea … it has proven to be wrong.”

When the proton was discovered by Ernest Rutherford in 1919, it was thought to be an indivisible basic building block of matter. But that fundamental status did not last long. Early proton-proton collision experiments in the 1930s revealed that the proton was more than an infinitesimally small “point-charge”: It had a finite size and presumably some kind of structure. Further experiments revealed a bewildering array of particles related to the proton, whose properties fell into patterns that cried out for an explanation in terms of more fundamental building blocks. A breakthrough came in the 1960s, when theoreticians Murray Gell-Mann of the California Institute of Technology in Pasadena and his ex-student George Zweig at CERN proposed that fundamental particles called quarks make up protons, neutrons, and the short-lived particles called mesons. Protons and neutrons contain three quarks each, and mesons a quark and an antiquark. In 1969, electron-proton collisions at the Stanford Linear Accelerator Center confirmed the existence of pointlike nuggets inside the proton, which had to be quarks.

The new picture painted by Gell-Mann and Zweig was simple: The proton is made of two so-called “up” quarks and a single “down” quark—its “valence” quarks. Each quark has a fractional electric charge, and the combination in the proton adds up to provide its single positive charge. In the neutron the numbers of ups and downs are reversed, giving a one-up, two-down combination that makes the neutron electrically neutral. The theory later found to govern these quarks and their interactions was dubbed quantum chromodynamics (QCD), now part of the Standard Model by which physicists understand the subnuclear world. QCD predicts that quarks carry a “color charge,” mimicking the familiar electrical charge, which is the source of the force binding them together and is carried by gluons, force particles analogous to the photons of electromagnetism.

But even this tidy model of three valence quarks and a buzz of gluons holding them together proved to be far from the whole story. Experiments at CERN in the early 1970s probing protons with ghostly particles called neutrinos revealed the presence of antiquarks along with the three valence quarks, and soon researchers' image of the proton began to change. A proton “is not a rigid thing with three balls in it all hooked up with springs or something,” says James Stirling of Britain's University of Durham. Instead, the gluons exchanged by the valence quarks have a tendency to split spontaneously into more gluons or quark-antiquark pairs, creating a lively “soup” made up of a seemingly infinite number of particles. “The valence quarks excite from the neighboring vacuum a dynamic ‘sea’ of short-lived gluons and quark-antiquark pairs,” says Leitch.

Physicists soon realized that to understand the proton they would have to understand the quark sea. “Past experiments have shown that a major part of the proton's momentum is carried by sea quarks and gluons,” says Dietrich Harrach of CERN. “[The sea] is going to tell us a lot of interesting detail about how QCD works on the scale of proton,” says Stirling. “We can't claim to understand QCD without understanding this.”

Getting an inside look

A window on the interior of the proton opened in 1992 with the inauguration of HERA, an electron-proton collider at Germany's DESY accelerator center that was especially designed to probe the proton's structure. Whereas earlier experiments explored the proton by firing electrons at a fixed proton target and examining the debris, HERA can accelerate both a beam of protons and a beam of electrons (or electrons' antimatter counterparts, positrons) and smash them together head-on. In such a collision, the probe electrons throw out photons that scatter off proton constituents, creating a freeze-frame image of the seething mass of quarks, antiquarks, and gluons in the proton.

Over the past 7 years, researchers have used HERA to take a census of the sea's inhabitants. One way to classify them is by the fraction of the whole proton's momentum they carry, revealed through the scattering angles and energies of the probe particles leaving the collision. Many measurements have confirmed that valence quarks carry a lot of momentum, while the number of virtual quarks and gluons mushrooms at smaller momentum fractions. That has come as no surprise. “The pileup of the gluons and … sea quarks at small momentum fractions is expected,” says Stirling. QCD predicts such a distribution because valence quarks naturally throw off gluons having a smaller momentum than themselves, and these gluons in turn spark quark-antiquark pairs having still smaller momenta, and so on down the line, explains John Dainton, a member of the team operating HERA's H1 detector.

Ongoing measurements at H1 and its sister detector ZEUS are also confirming another prediction of QCD: that at the smallest momentum fractions, virtual gluons greatly outnumber virtual quarks and antiquarks. “[Gluons] just prefer to split into each other,” says Stirling. With each split sharing the parent's momentum between the daughters, the result, according to QCD, is a burgeoning population of low-momentum gluons.

So far so good. But researchers got a surprise when they began to look at the types, or flavors, of quarks that inhabit the sea. “The obvious first guess would be that there is no flavor structure of the sea, it's just democratic,” says Stirling. The complication is that the different quarks have different masses, and heavier quarks will have a harder time popping into existence. This should mean that the three heaviest of the six quark flavors—charm, bottom, and top—should be rare within the proton. The lightest, up and down, should be present in equal amounts, and strange quarks, with a slightly higher mass, should be a bit less numerous.

HERA's electron beam has trouble mapping out these flavor distributions, however, because electrons are “flavor blind.” Hence a team at the Fermi National Accelerator Laboratory in Illinois is probing the nucleus using neutrinos, in an experiment called NuTeV. These wispy particles interact with quarks via a particle called the W boson, rather than a photon, in such a way that the debris of the collision reveals what kind of quark was involved. “What we have found,” says Michael Shaevitz of Columbia University in New York City, a member of the NuTeV team, “is that [the strange content] is much smaller than expected, about one-half the amount of the up or down quark sea.” He expects the result to be confirmed by further NuTeV measurements later this year.

Theorists have trouble explaining this because their trick for deriving predictions from QCD, called perturbation theory, cannot handle particles with a mass as small as the strange quark's. The more massive charm quark does, however, fall within the scope of perturbation theory, and estimates of the charm quark population inside protons compare favorably with recent measurements made by H1 and ZEUS at HERA, which have looked for charm quark-containing particles dislodged from the proton by collisions.

More depth to the sea

Leitch and his colleagues on another Fermilab experiment, nicknamed NuSea, last year uncovered an even more startling inconsistency: The number of antiup quarks in the proton sea is not the same as the number of antidowns. Physicists had always assumed that up and down quarks—and their antimatter partners—populate the sea equally. The NuSea team compared populations of these two antiquarks by firing a proton beam at two targets: a flask of hydrogen, which essentially contains nothing but protons, and a flask of deuterium, an isotope of hydrogen that contains equal numbers of neutrons and protons. Collisions between quarks in the proton beam and antiquarks in the target nuclei yield muons—heavy cousins of the electron—paired with antimuons, which a detector picks up. The ratio of the yields from the two different targets translates into the ratio of antidown to antiup quarks in the proton sea. The result, reported last year in the 27 April issue of Physical Review Letters, “establishes totally unambiguously that there are more antidown than antiup quarks in the proton,” says Thomas.

The case strengthened recently when experimenters at Hermes, HERA's third detector, collided positrons with both protons and deuterium and compared the numbers of positively and negatively charged pions that emerged. The experiment “is based on the fact that scattering [a positron] on an up antiquark most likely results in a negative pion, while hitting a down antiquark would lead to a positive pion,” says Antje Brüll of the Hermes team. The pion counts revealed an excess of down antiquarks.

The agreement between the Hermes results, reported in the 21 December 1998 Physical Review Letters, and the results from NuSea is “good,” says Brüll. The two experiments found that down antiquarks outnumber up antiquarks by as much as three to two, implying that the up and down quark populations of the sea are similarly unbalanced. This “relatively huge” difference is not really understood, says Stirling. “It is a fundamental property of perturbative QCD that the sea would have to be ‘flavor symmetric,’” says Thomas. The antidown-antiup imbalance “is telling us something vitally important about the … structure of the [proton]” beyond the range of perturbation theory, he says.

Thomas has proposed an explanation for this imbalance, called the meson cloud model, in which the proton fluctuates between being a pure proton and a mixture of a neutron plus a positively charged pi meson, and several other overlapping particle combinations allowed by quantum theory. Because a positive pi meson consists of an up quark and down antiquark, an experiment that “sees” the proton as a neutron plus a pion will record more down antiquarks than up antiquarks, according to Brüll. Both Hermes and the NuSea data “clearly favor the so-called meson cloud models,” she thinks.

Not everyone is so enthusiastic, however. “There are many high-energy people who refuse even to imagine that pions could contribute” within a proton or in such a simplistic way, says Thomas. Stirling feels that, because physicists cannot count quarks directly but must rely on the debris of collisions, there may be some bias in the way debris from certain collisions rearranges into observable particles.

And the proton keeps turning up new surprises. Two months ago, Thomas published a new analysis of Fermilab neutrino-nuclei data and a CERN experiment that scattered muons from protons and found what might be another anomaly of the quark sea. In the 9 November 1998 issue of Physical Review Letters, Thomas and his collaborators, Csaba Boros, also in Adelaide, and Tim Londergan at Indiana University, Bloomington, claim that for momentum fractions of less than about 10%, the number of up antiquarks in the proton does not equal the number of down antiquarks in the neutron.

If true, this would send shock waves through the particle physics community, as researchers have always assumed that protons and neutrons are related by a simple interchange of up and down quarks and antiquarks. “The effect, if it holds up experimentally, is huge,” says Thomas. “We know of no theoretical mechanism which could explain these data.” According to Shaevitz, however, forthcoming data from two Fermilab experiments may contest Thomas's ideas.

The challenge ahead will be to fit this swath of unexpected results into the framework of QCD. “While QCD is an amazingly beautiful theory, most of its consequences remain inaccessible to [theorists],” says Sarada Rajeev of the University of Rochester. But the clues that experimenters have been able to tease out of the proton have put some of the answers within reach. Forthcoming experiments may serve up more clues, and perhaps more surprises. Dr. Seuss's Horton finally gets the dust speck's inhabitants to shout loudly enough to be heard, convincing all the doubters. Perhaps diviners of the proton's sea could use some similar inside help.

9. NUCLEAR PHYSICS

Element 114 Lumbers Into View

1. Richard Stone

The race to capture one of the biggest prizes in nuclear physics—an exceptionally long-lived superheavy element—appears to be over. In a cautiously worded e-mail to a close-knit group vying for the trophy, scientists at the Joint Institute for Nuclear Research in Dubna, near Moscow, this month unveiled evidence for the creation of a nucleus with 114 protons—the heaviest element yet forged.

If confirmed, the sighting would mean far more than just another entry in the periodic table. Element 114 appears to last for 30 seconds before decaying, a longevity that would verify predictions of an “island of stability” beyond the lighter, less stable nuclei glimpsed earlier. “This is the most exciting event in our lives,” says Albert Ghiorso of Lawrence Berkeley National Laboratory (LBNL) in California, who has spent 35 years hoping his group would plant the flag on the fabled terrain. The finding, adds Sigurd Hofmann of the Institute for Heavy Ion Research (GSI) in Darmstadt, Germany, whose team many observers expected to get there first, “opens up a window to a quite new field of research.”

For a half-century physicists have used nuclear reactors and particle accelerators to forge new elements, beyond the 94 known to exist in nature. Like climbing taller and taller peaks, each successive effort has required vastly more energy and greater technological legerdemain. And for ever-more fleeting results: Although some transuranic isotopes last for years, an isotope of the last element created—number 112—is so unstable it sticks around a mere 280 microseconds. Theorists have predicted, however, that this trend toward instability would be reversed as additional protons and neutrons filled out nuclear shells. With a full shell of protons, element 114 should lie well within the stable island.

To make the element, the main contenders—GSI, LBNL, and Dubna in collaboration with Lawrence Livermore National Laboratory (LLNL) in California—plotted varying strategies (Science, 24 October 1997, p. 571). GSI went with cold fusion, a technique in which two medium-sized isotopes are fused in an accelerator—an approach that already secured their claim to bohrium (element 107), hassium (108), meitnerium (109), and the unnamed elements 111 and 112. Last spring Hofmann's GSI team tried to create element 113 but failed.

The Dubna-LLNL group took a different tack, heading straight for 114. Their hot-fusion approach involves smashing light elements into a heavy one like plutonium. For several weeks late last year, a team led by Dubna's Yuri Oganessian and Vladimir Utyonkov pounded a plutonium-244 target provided by LLNL with some 5 × 1018 atoms of a rare calcium isotope, calcium-48. Sifting the data from their detector, the team spotted what appears to be the unique signature of a decay chain starting with 289114, which hung around for 30 seconds before hiccuping an alpha particle to form an isotope of 112.

More work is needed to confirm the find, says Ghiorso, whose group will do follow-up studies. Says Dubna's Alexander Yeremin: “If at least one more event [is] found with similar characteristics, it will be good proof.” In a sad footnote, isotope pioneer Glenn Seaborg, 86, suffered a crippling stroke a few months ago and may not comprehend the news, says Ghiorso. Seaborg, whose name graces element 106, would be thrilled by a discovery that, if verified, would open a terra incognita for nuclear science.

10. SOCIETY FOR INTEGRATIVE BIOLOGY MEETING

Rocky Mountain Rendezvous

1. Elizabeth Pennisi

DENVER—Researchers from all walks of biology traded tales about snakes, whales, birds, and every other organism imaginable at the annual meeting of the Society for Integrative and Comparative Biology, held here from 6 to 10 January.

How Snakes May Have Lost Their Legs

For centuries, not just scientists but artists too have speculated about the limblessness of snakes. Michelangelo thought the loss occurred in the Garden of Eden. Now, two developmental biologists offer a less fanciful explanation—one involving genes and the proteins they produce—rather than divine intervention.

At the meeting, Marty Cohn of the University of Reading in the United Kingdom reported that he and Cheryll Tickle of the University of Dundee in Scotland have linked such important changes in snake evolution as the elongation of the thorax and the loss of the forelimbs to the altered activity of several HOX genes, which are involved in body patterning and limb formation. In addition, the failure to develop complete hindlimbs seems to be due to an inability of the embryonic tissue to respond to the normal developmental trigger.

“It's a beautiful piece of work” that demonstrates how developmental biology can help explain evolution, comments Michael Coates, a vertebrate paleontologist at University College London. “He's shown us the molecular details about a large-scale evolutionary change.”

Cohn decided to tackle snake evolution after he and others had demonstrated that certain HOX genes help prod limb development in chick and mouse embryos. He knew that HOX genes also control the formation of distinct neck, thoracic, lumbar, and caudal vertebrae. So, he used chemically tagged antibodies to find out which regions in the python embryo contain the proteins made from the HOXC6, HOXC8, or HOXB5 genes. He chose those three genes because HOXC6 and HOXC8 are normally expressed where the ribs develop, while in mice and chickens, HOXB5 helps define where the forelimbs sprout.

He found that the HOXC genes were turned on throughout the embryonic axial skeleton, thereby extending the thorax all the way from the top of the vertebral column to the hindlimb buds. And in the python embryos, HOXB5 was active throughout the cells of the lateral mesoderm, which lie along the sides of the embryo. In limbed embryos, this gene is active in just one part of this mesoderm, thereby helping to specify the starting point for the front legs. This change in HOXB5 expression, Cohn suggests, somehow contributes to a total loss of forelimbs.

Pythons have vestigial hindlimbs, and Cohn has linked their failure to develop normally to the fact that the embryonic hindlimb buds lack a leading edge of tissue called the apical ectodermal ridge, which prompts elongation of the leg. But no one knew whether the signal that would normally tell the ridge to develop was lacking or if the limb bud tissue just couldn't respond.

Cohn and Tickle's results point to this second possibility. For example, when the research-ers grafted limb bud cells from the snake into the appropriate spot in a chick embryo, the snake cells sent the right signal, inducing ridge formation. This suggests, Cohn reported, “the python ectoderm is not competent to respond to the signal,” although he doesn't yet know why.

If the python ectoderm could respond, however, it appears it could form hindlimbs. The researchers found that fibroblast growth factor 2, which stimulates limb growth in other organisms, did the same for the python embryo hindlimb, at least for the 24 hours they could study it. Thus, despite a few glitches, “all the other signaling networks [for limbs] are in place,” he reported.

Based on these results, Cohn suggested that the HOX gene changes came first in snake evolution, perhaps leading to a primitive snake resembling a fossil called Pachyrhachis problematicus, which had hindlimbs but no forelimbs. Next, other genetic mutations cut short the full development of hindlimbs and caused the trunk to elongate, resulting in the pythonlike snakes. Finally, further mutations caused the hindlimbs to disappear altogether as in more advanced snakes.

This molecular scenario is quite pleasing to Michael Caldwell, a paleontologist with the Canadian Museum of Nature in Ottawa. “He's done the development work and put it into a phylogenetic context,” Caldwell notes. “That's been very difficult to do.”

Coming to Grips With Sperm Whale Anatomy

Sometimes, all it takes is a little old-fashioned biology to solve a long-standing mystery. And sometimes it takes a bit of fancy technology. At the meeting, both approaches were used to better understand the world's largest toothed mammal, the sperm whale. On the low-tech end, simple dissection has helped explain how the creatures feed. In addition, high-tech computed tomography (CT) scanning of a sperm whale head has provided a new view of how the whales generate sounds and, perhaps, even how they seek mates.

Sperm whales feed mainly on deep-dwelling ocean animals like giant squid, and how they take in their prey has been something of a mystery. Because healthy sperm whales have narrow, pointed mouths with a lower jaw that hangs down, it seemed logical that they need it to stir up and grasp prey. But many museums contain whale skeletons whose jaws are either congenitally deformed or scarred by multiple breaks that occurred during the whale's life, alterations that would have prevented their feeding that way. At the meeting, Alexander Werth of Hampden-Sydney College in Virginia reported that such whales would have been able to survive because they don't use their jaws to grasp food. Instead, he favors an alternate idea, that sperm whales use their tongues to suck in prey.

A specialist in the feeding mechanisms of toothed whales, Werth dissected the tongue of a beached sperm whale and traced its muscular connections to the mouth. Most tongues, including those of humans, have a layer of built-in muscles that help them fold and twist so that they can manipulate food to be swallowed. But Werth found that these muscles were very reduced in the whale tongue he dissected. It is, however, well equipped with muscles connecting it to jaw and skull, so that it can quickly move backward and forward over the throat and create suction. This “is a very efficient way of getting stuff into the mouth,” comments Ted Cranford, a marine biologist who studies whales at California State University in San Diego. “Before this work, no one had even looked at the tongue to see if it could work this way.”

Although suction feeding has not been proven experimentally, the work is still impressive because with these rarely studied deep-sea denizens, “you have to look at whatever information you can gather,” points out D. Ann Pabst, a marine mammalogist at the University of North Carolina, Wilmington. “[Werth] is filling in a vacuum in knowledge.”

So is Cranford, who used a CT scanner designed to look for flaws in rocket engines to view the inside of the intact, frozen head of a juvenile sperm whale. Imaging the entire head, which, at 1 meter long, was quite small for a sperm whale, took 5 days. To keep the head from defrosting and decaying during that time, Cranford encased it in foam. Some colleagues “said it would be impossible,” he recalls. “But it worked perfectly.” Indeed, says marine mammalogist Sentiel Rommel of the Florida Department of Environmental Protection in St. Petersburg, the scans provide “a nice clean description” of the whale's head.

For one thing, they provide a clear view of the spermaceti organ, a large sac filled with waxy oil where the clicks, pings, whistles, and other sounds of the whale's vocabulary are thought to originate. By showing the densities of the various tissues of the head, the scan may eventually help reveal how sounds are channeled and amplified in their passageway to the exterior.

In addition, Cranford compared the structures in the whale's nose with their counterparts in dolphins, which he had previously studied using a normal CT scanner. He showed that the sperm whale nose is proportionally large for its body. He speculates that this swelled nose—and the magnitude and character of sound produced from it—may be one way a male sperm whale shows off to females, or to male rivals, who like stag deer are known to fight. By listening, Cranford suggested at the meeting, sperm whales may be able to size up suitors and enemies, even in the deepest, darkest depths of the ocean. That idea still requires further investigation, Pabst notes, but nevertheless, “it's a very interesting and provocative hypothesis.”

For some of the small, stocky Neotropical birds known as manakins, love knows few bounds. Many birds have evolved—in addition to melodic songs—brilliant coloration or elaborate tail feathers as part of an evolutionary race to win mates. But about half of the 40 or so manakin species have evolved an unusual ritual: The males use their wings to “sing” for their mates—clicking and rattling as the birds prance around during courtship displays. New results, described at the meeting by Kim Bostwick of the University of Kansas, Lawrence, now show that unlike most birds, which undergo rather superficial changes to appear sexier, these manakins have evolved structural modifications with potentially high costs.

Based on field and anatomical studies by herself and others, Bostwick, a grad student in ornithologist Richard Prum's lab at Kansas, has found that the adaptations range from changes in feather shape to dramatic alterations in wing structure. In some species, the wing bones and muscles have bulked up to the point where the birds' flying efficiency is likely to be compromised—a testimony to the strength of the evolutionary drive to reproduce. “Almost nowhere is there nearly as much change as we see in these manakins,” says Prum.

Bostwick's colleagues say her studies could provide a better understanding of how birds have evolved these extremes in their wing anatomy and how those traits relate to behavior. “She's finding some nice correlates” of anatomical change with behavior, says George Goslow Jr. an evolutionary biologist at Brown University in Providence, Rhode Island. “It's a very interesting study, and her approach is very logical.”

For the past several years, Bostwick has been doing field and anatomical studies of manakins in South America. At the meeting she described the shape and size of wing bones, feathers, and muscles of several species, including one representative from each of the three groups that have apparently evolved this noisemaking capability independently.

All the clicking birds had some changes in feathers, but the most unusual plumage belongs to the club-winged manakin, which also has the most unusual wing song. This bird hums, as well as clicks and rattles, and likely uses irregularly shaped inner flight feathers—twisted at the tips—in making the humming noise, Bostwick reported.

Also, the wing bones and musculature of noisemaking manakins had surprisingly striking variations. Because flight is so difficult, the proportions of the bones and muscles are virtually indistinguishable among most birds. Take the 300 species of flycatchers, a group closely related to manakins. They all have a thin ulna, one of the wing bones, for example. But the same bone in clicking manakins is thicker and is uniquely outfitted with knobs that help support the enlarged muscles attached to feathers. This setup may help brace the feathers for sound generation, Bostwick suggested.

Other wing changes may help manakins achieve the right noisemaking posture. “When club-winged manakins display, they lean forward and flip their wings back and then forward,” Bostwick explained. This requires that they rotate the elbow joint, a motion seen in few other birds. Whereas the muscle that moves that joint is small in most other birds, in the clicking manakins it is bulky and strong. In one species, the ends of the joint maneuvered by this muscle are also much more rounded, perhaps “translating into an increased mobility of that joint,” Bostwick pointed out.

She hasn't worked out how these changes would affect the manakins' flying ability. But she expects that their heavier wings would slow them down, compared to their agile flycatcher cousins. Bostwick plans to extend her work to include manakins that either have secondarily lost their clicking ability or have evolved to make the clicks with either their tail or their vocal cords. “It will be interesting to see if the loss of that ability has led to more changes in the morphology,” she notes.

11. JAPAN BUDGET

Science Spending Keeps Rising But May Miss 5-Year Targets

1. Dennis Normile

Japanese scientists appreciate the boost in government spending, even if it falls short of the goal. But money isn't everything, they say

TOKYO—Money talks, and for the past 3 years Japanese scientists have been listening closely to whether the government is keeping its promise to dramatically increase spending on science. They have been hearing some mixed signals.

The numbers are climbing, and science is expected to get another boost this week when legislators approve the fourth of five annual budgets covered by a key 5-year spending plan (see graph). That's a significant achievement for a country mired in an 8-year recession. But it's becoming clear that the ambitious spending goals won't be met and that institutions are lagging in adopting reforms to ensure that the money is put to good use. Some scientists also grumble that the government is favoring research with commercial potential. And many institutions marked for growth are in the awkward position of seeing their operating budgets shrink while they are building new facilities.

The 5-year spending plan grew out of a 1992 recommendation from the Council for Science and Technology, the nation's highest science advisory body. The council asked the government “as soon as possible” to double annual spending on research, then $18.6 billion. (All budget figures understate actual spending because they do not include salaries.) The resolution was embedded in a 1995 Basic Law for Science and Technology and a 5-year plan endorsed in 1996 by the Cabinet (Science, 28 June 1996, p. 1868). To meet their promise, officials agreed to spend 17 trillion yen ($150 billion) over the 1996–2000 period. But instead of incremental increases that would have lifted the regular 2000 R&D budget to twice the 1992 level, the government decided to count spending in supplemental budgets as well as in annual budgets. Supplemental budgets are an increasingly frequent mechanism to stimulate the economy with quick-spending schemes. The downside is that supplemental spending doesn't raise the base, and with 10% of the increase for science stuffed into three massive supplemental budgets, the regular science budget has grown much more slowly than anticipated. Even a whopping 17% increase next year, the amount needed to hit the 5-year target of 17 trillion yen, would result in a 2000 budget that's only 75% higher than the 1992 level of $18.6 billion. “Spending is still below the level we would like to see,” says Masayuki Shibata, director of the office of science policy at Monbusho, the Ministry of Education, Science, Sports, and Culture. Despite that shortfall, most scientists and administrators praise the government's overall track record. “We have to thank our politicians [for this support],” says Masao Ito, a neuroscientist at the Institute of Physical and Chemical Research (RIKEN), outside Tokyo. There is no question that funding for actual research has expanded dramatically. For example, Monbusho's Grants-in-Aid program for research at universities and Monbusho-affiliated institutes has jumped nearly 30% since 1996, to$1.16 billion. And both Monbusho and the Science and Technology Agency (STA) have initiated or expanded programs that handsomely reward the most active and innovative researchers and groups. “The variety of programs and their different approaches have really helped make the research environment more active,” says Hiroyuki Yoshikawa, an engineer who is president of the Science Council of Japan, the nation's leading private science association. That assessment is shared by Katsuhiko Sato, a cosmologist at the University of Tokyo who heads a nine-member group that in 1995 was among the first batch of successful applicants for a Monbusho Centers of Excellence program. The 5-year, $15 million award has enabled them to search for antiprotons, conduct particles work, and do theoretical studies on the early universe. But there have also been disappointments. Last year's regular budget actually cut operating funds for many big science projects at the same time the two supplemental budgets boosted spending on buildings and equipment (Science, 1 May 1998, p. 669). The cuts “went against the intent of the 5-year plan,” complains Akiyoshi Wada, a biophysicist who heads RIKEN's new Genomic Science Center. In addition, some believe that the spending has not benefited all fields. “I don't think the spending has been very balanced,” says Keiichi Kodaira, director-general of the National Astronomical Observatory, near Tokyo. Aside from the Centers of Excellence initiative, he says, the new programs typically fund fields deemed economically strategic, such as nanotechnology, information sciences, and molecular biology. And while the government has equipped new buildings with the latest instruments in neuroscience and genetics, the high-energy physics community recently had to combine plans for two accelerators (Science, 8 January, p. 155). “It seemed the only way we could get funding,” says Sakue Yamada, a director of the High Energy Accelerator Research Organization in Tsukuba. Although the 17-trillion-yen target has received most of the attention, the 5-year plan called for institutional reforms as well. Progress here has been mixed. Although practically all of the major institutes and university departments have heeded calls for independent reviews, most have been done by in-house committees. Still, Monbusho's Shibata says that this has helped shake up lethargic university researchers. “They are afraid of losing face,” he says. The going has been tougher for those trying to replace the current lifetime employment system for national employees with limited-term appointments or schemes that require faculty to earn tenure. Only 19 of 98 national universities have introduced modifications to lifetime employment, and typically only in very limited ways. “I'm not optimistic about [limited-term appointments] being adopted,” says Yoshikawa, a former president of the University of Tokyo. While researchers may debate the government's success in keeping its commitment, few dispute the overall positive effect of the 5-year plan. And most are optimistic that the upward spending trend will continue. “There is consensus among all the political parties on the importance of research-related spending,” says Yoshikawa. 12. MONGOLIA Science Hopes to Rebound in Post-Cold War Era 1. Andrew Lawler* 1. Andrew Lawler is a staff writer on fellowship at the Massachusetts Institute of Technology in Cambridge. Newly democratic, Mongolia hopes Western links will help it to overcome its isolation and regain its scientific prowess ULAAN BAATAR, MONGOLIA—A decade ago, this vast, isolated, and rugged country boasted a surprisingly strong research enterprise with 100 research institutes, 3000 researchers, and an annual influx of scientists from other parts of the East Bloc. “Our expertise and capacity was very high,” says B. Chadraa, president of the Mongolian Academy of Sciences and a Moscow-trained physicist. “We worked closely with Russia and [East] Germany.” Mongolia's geography helped: The Soviets built a series of seismological stations to monitor nuclear tests across the border in China, and they funded operations at a hilltop of telescopes to observe U.S. spy satellites through Mongolia's clear skies. In addition, the dinosaur graveyards of the Gobi desert were a big draw for paleontologists. But in 1991 Russia withdrew hundreds of thousands of its troops, and the generous subsidies for outside university education and research work disappeared. Today, the telescopes are shuttered by a lack of money for photographic plates, and the seismic stations are silent. So officials in this new democracy are looking West for help in building on modest initiatives in seismology and higher education and leveraging Mongolia's natural assets. Those efforts, the country's researchers note proudly, reach back 700 years, when Mongol emperor Kublai Khan organized the first international academy of sciences in Beijing. But creating those links won't be easy for a country that largely banned Westerners for half a century. “The situation was very difficult,” says Bazaryn Bekhtur, director of the Institute of Astronomy and Geophysics, sitting in the traditional round nomadic tent called a ger still favored by Mongolians. Bekhtur was visiting a group of Canadian astronomers camped out on a vast plain 50 miles south of the capital to monitor last fall's Leonid meteor shower. “We tried to set up some cooperation with Western countries,” he says, “but they had no good information on Mongolian science and technology. And we had no good information on them.” Communications were limited because the second language for most Mongolian researchers is Russian or German, not English. The government reacted to the crisis caused by the abrupt loss of Soviet support by reducing the number of scientific institutes to 20, with 11 devoted to basic research in the physical, biological, and social sciences. And while government spending on science and technology has held fairly steady at almost$3 million since 1991, the end of Soviet subsidies for oil and other essentials has triggered an inflationary spiral that has eaten heavily into purchasing power. “There is enough money to keep current programs going, but not for anything new,” says Chadraa. At least one-third of Mongolian researchers have abandoned science since the end of the Soviet era, he estimates. “The good people are leaving to go into business and politics,” says Bekhtur mournfully.

Bekhtur's mountaintop institute, on the outskirts of the capital, once was a beehive of activity. Soviet intelligence services came for a firsthand look at U.S. spy satellites and clues to their intended targets, while scientists conducted regular astronomical research. Today, Bekhtur's annual budget of about $75,000 has been only partially appropriated, and its bank of 10 telescopes, along with a large building for classrooms and offices, is largely empty. Astronomer Bayaraa Togookhuu waits in vain for Western scientists to show interest in a finely crafted 20-inch Schmidt telescope once used for variable star research. “A few thousand dollars is all that is needed to upgrade it,” says Martin Connors, an astronomer at Athabasca University in northern Alberta, Canada, who recently inspected the Schmidt telescope. The site's clear and dry air, its altitude and location on the opposite side of the globe from North America, and its political stability make it “potentially a good place for astronomy,” adds Bill Chang, who handles Mongolian-related research at the U.S. National Science Foundation (NSF). The outlook appears slightly brighter for Mongolian geologists and geophysicists. The Comprehensive Test Ban Treaty Organization in Vienna wants to place five seismic stations across Mongolia to keep a watch out for rogue nuclear tests. Chadraa says the$1.6 million contract, still under negotiation, would provide the ability to detect any atmospheric explosions and to measure for airborne radioactivity. Although the stations are designed for minimal maintenance, meaning few jobs for Mongolian scientists, Chadraa hopes they will lead to increased contacts between Mongolian and Western researchers.

Another possibility for greater contact is further exploration of the country's recent seismic history. “There have been several earthquakes in the last 50 years near magnitude 8,” says Jack Medlin, head of the Asian and Pacific geology section for the U.S. Geological Survey (USGS), which has sponsored four expeditions. The pattern of inner-continental quakes resembles activity in the U.S. midsection, and Mongolian fault lines are often exposed rather than buried under layers of rock.

USGS also is working with Mongolia and several other Asian nations to analyze the continent's mineral deposits. The next step, says Medlin, would be for Mongolia to conduct its own mineral assessment and environmental survey, at a cost of several million dollars. A USGS team will return to Ulaan Baatar in late spring to discuss the plan, which would require outside funding. In the meantime, a number of U.S. scientists have received NSF money to work with their Mongolian counterparts on everything from dinosaur fossils and grassland ecology to the pristine depths of Lake Chovsgol. A proposal from the Mongolian government to set aside vast tracts of land for conservation purposes could provide additional research opportunities.

Such cooperative efforts can only do so much to improve the country's science, however. In the long term, Mongolian administrators acknowledge that a better educated population will be essential. And that means supplementing the country's only major university. So in 1997 Chadraa converted a former Russian high-rise building into a campus, called the Ulaan Baatar University, that is run by the University of Colorado, Denver (UCD). The unusual arrangement, which UCD pioneered in Moscow and Beijing, gives the 60 Mongolian students now enrolled a chance to learn English, earn a U.S. degree, and apply for study in Denver or other U.S. universities. The academy and its U.S. partner share the cost of the \$4000 annual tuition. In addition, the Mongolian government subsidizes 30 graduate students at Denver and at other U.S. universities.

Chadraa, who also holds the position of university rector, acknowledges that the UCD relationship is a gamble. “It's hard for us—the textbooks, the tuition are very expensive, and we have spent a lot of money developing this.” But such a connection is a vital step toward raising a new generation of English-speaking researchers. For their part, UCD officials see the arrangement, which they hope will at least break even, as an opportunity to expand their presence in Asia.

Mongolia's efforts to build a peaceful democratic society and create a market economy win praise from foreigners, who contrast it with the chaos enveloping other parts of the former Soviet Bloc. Nevertheless, day-to-day life remains bleak. “Mongolia is a small country, and there is little support for science,” says one Mongolian researcher, noting that “science is at the bottom of the list” of programs funded by the country's Ministry of Enlightenment, which supports education, culture, and science. That is the harsh reality in a nation of few roads, schools, and exports, and whose airline is hard pressed to pay for maintenance on its single Airbus jet. But Mongolian researchers are betting they can extend the country's history of international contacts to bolster its scientific prowess.

13. AMERICAN ASTRONOMICAL SOCIETY MEETING

A Black Hole's Feeding Tube

1. James Glanz

AUSTIN, TEXAS—People who go to extremes often make the news, and the same goes for outlandish celestial objects: neutron stars and black holes. At the astronomy meeting here, clues to two mysteries emerged: how black holes fuel themselves and how some newborn neutron stars hide from view. (See last week's issue for earlier reports from the astronomy meeting.)

What feeds the cosmic manic-depressives called Seyfert galaxies? When these flickering galaxies are at their brightest, a tiny region at their core can outshine the entire Milky Way. Astrophysicists have theorized that a bar of gas, perhaps 100 light-years across, forms and acts as a “feeding tube” to squirt material into a central black hole, where the material gives off one last “Geronimo!” of brilliant electromagnetic radiation before vanishing into the gravitational maw. Eventually the bar disappears and gas settles more slowly to the center of the galaxy, which ceases its hyperkinetic ways and gives off a merely ordinary glow.

Until the American Astronomical Society meeting, however, no such bar had ever been reported. At the meeting, a team led by Almudena Alonso-Herrero of the University of Arizona, Tucson, and Roberto Maiolino of the Osservatorio di Arcetri in Florence, Italy, announced that they had used the Hubble Space Telescope (HST) and other instruments to pick out a small bar and trace the motion of its gas, which appears to be streaming toward the center of a Seyfert galaxy in full throat. “It could be that this is catching [a black hole] in the actual act of fueling,” says Michael Regan, an astronomer at the Carnegie Institution of Washington.

If so, the implications could go beyond Seyfert galaxies, because other so-called “active” galaxies—such as the much more brilliant quasars—may work in a similar way. Astronomers doubt that the full picture of how black holes are fed has emerged just yet, but Maiolino points out that Seyferts, with properties midway between ordinary galaxies and quasars, are good places to test the physics of the feeding. And Seyferts like the one his team studied, called Circinus, are much closer than quasars, making them easier to study. At a distance of 10 million light-years, Circinus “is a next-door neighbor compared to quasars,” says Maiolino.

Theories say that gas bars could form spontaneously, when gravity amplifies slight ripples in the disk of a galaxy. As gravity pulls in more and more gas, it would smash together in the bar, forming shock waves that could brake the spinning motion that the gas shares with the galaxy as a whole. No longer in the grip of centrifugal force, the material would quickly drain along the bar toward the black hole, like a roulette ball falling to the center of the wheel after it loses its spin. The final acceleration of the material around and into the black hole would throw off photons and produce the Seyfert galaxy's radiation, from an area the size of the solar system.

Ground-based telescopes can't resolve the fine detail needed to see such a bar at the center of another galaxy, so the team turned to the HST, working in infrared wavelengths that can penetrate dust at galactic centers. When the team took a close look at Circinus, one of the nearest Seyferts, the bar popped out of the HST images. Alonso-Herrero says that once they knew exactly where to look, they were then able to use data from the 3.9-meter Anglo-Australian Telescope in Australia to estimate the gas velocities from slight shifts in the wavelengths of the light that emerged.

“We were really amazed at how the observations resemble” the theory, says Alonso-Herrero. “It does appear to be forcing gas into the middle, in the way that models predict,” says Andrew Wilson, an astrophysicist at the University of Maryland, College Park, who visited the team's poster presentation here.

But not every Seyfert may dine the same way. In a poster right next to Alonso-Herrero's, Regan and the Carnegie Institution's John Mulchaey reported their own new HST observations of 104 galaxies of various types. Although the observations are mostly less detailed than the Circinus images, they suggest that such bars aren't always present in active Seyferts. The study also showed that the bars can turn up in ordinary galaxies, many of which are also thought to have black holes at their cores.

Whether Circinus is an isolated case now becomes the “tough question,” says Alonso-Herrero. By relying partly on visible rather than infrared light, Regan and Mulchaey could have missed bars in some now-active Seyfert galaxies, she thinks. Then again, a Seyfert might not need a bar for the full duration of its outburst. A short-lived bar might supply a black hole with enough gas to keep it radiating for hundreds of millions of years, then disperse. “If bars are the way [fueling] happens, they're not doing it all the time,” says Regan. “I think what we're coming to is that it's not the simple answer.”

14. AMERICAN ASTRONOMICAL SOCIETY MEETING

"Magnetars"--the Missing Pulsars?

1. Robert Irion*
1. Robert Irion is a science writer in Santa Cruz, California.

AUSTIN, TEXAS—People who go to extremes often make the news, and the same goes for outlandish celestial objects: neutron stars and black holes. At the astronomy meeting here, clues to two mysteries emerged: how black holes fuel themselves and how some newborn neutron stars hide from view. (See last week's issue for earlier reports from the astronomy meeting.)

Neutron stars, the ultradense cinders left behind when stars collapse in supernova explosions, inhabit a realm of weirdness almost as strange as that of black holes. Astrophysicists who struggle to grasp their bizarre properties have at least felt sure they understood how these stars begin: as pulsars, magnetized objects spinning rapidly after their violent birth and generating searchlight beams of radio waves. But now, recent findings of freshly forged neutron stars with radically different behaviors have put a new spin on that seemingly safe picture.

The culprits are “magnetars,” infant neutron stars apparently girdled by far stronger magnetic fields than pulsars have—so strong that the stars screech to a near standstill in a cosmic eye blink after their formation. Although magnetars were theoretical curiosities a year ago, astronomers are unveiling a growing number of them hurling x-rays and gamma rays, not radio waves, from the hearts of recent supernova blasts. Speakers at this month's meeting of the American Astronomical Society here ventured that magnetars may represent 10% of all young neutron stars, if not more. Given an anomaly of that size, “the previous pulsar paradigm is probably precluded,” Massachusetts Institute of Technology astronomer Victoria Kaspi proclaimed—provocatively.

Not all astronomers are willing to jump paradigms, after all. But many agree that magnetars could help solve what Eric Gotthelf of Columbia University calls the “missing pulsar problem.” Supernovae freckle the Milky Way with bright puffs of expanding debris, like the bursts of a July 4th fireworks display. Within each of these remnants, astronomers have assumed, a pulsar should lurk. Two stunning discoveries in 1968 seemed to cement this scenario: the famed Crab and Vela pulsars, broadcasting radio beams from within their respective supernova clouds. But since then, among hundreds of known remnants, researchers have found less than a dozen inhabited by radio pulsars. In the most recent washout, astronomers have yet to spot a whirling stellar corpse in the debris left by the great supernova of 1987.

Three factors help explain this puzzle, astronomers have believed. First, the radio beams from many pulsars may point away from Earth, making them undetectable. Second, supernovae can impart high-velocity “kicks” to pulsars and eject them from their remnants, which fade rapidly with time. Last, pulsars are notoriously faint and hard for radio surveys to spot.

Gotthelf now thinks another factor may make a bigger contribution to the pulsar deficit: Many neutron stars don't fit the classic Crab pulsar mold. At the meeting, he and colleague Gautam Vasisht of NASA's Jet Propulsion Laboratory in Pasadena, California, described several they have analyzed, including an x-ray-emitting neutron star that appears embedded in a 2000-year-old supernova remnant called Kes 73. Despite its youth, the neutron star spins just once every 12 seconds, the slowest rate ever seen. If the supernova left the collapsed core of the precursor star whirling 10 to 100 times per second at birth—as required by nearly all supernova models—then a magnetic field up to 1000 times stronger than an ordinary pulsar's must be applying the brakes, Gotthelf said. Such a strong field could stifle the neutron star's radio emissions but generate bursts of x-rays and gamma rays by literally cracking its exotic crust.

X-ray satellites have spotted a dozen other examples of isolated young neutron stars in supernova remnants, many of which have leisurely spins. That's enough to convince Gotthelf and Vasisht that they represent an important new class of objects. “Taken collectively, there are now more of these at the centers of supernova remnants than there are Crab-like pulsars,” Gotthelf says. “It's clear there's another evolutionary path.” Theorists aren't sure why some collapsing stars would spawn such magnetically fierce beasts while others would produce tame pulsars, but the evidence to date supports two different categories, Gotthelf believes. “As soon as Chandra [an x-ray satellite formerly called AXAF] goes up this year, we're going to find a lot more of these things,” he predicts.

Other pulsar experts are biding their time. “The claim that magnetars rather than ordinary radio pulsars are the more common products of supernova explosions seems to me to remain pretty speculative,” says astronomer Stephen Thorsett of Princeton University. Pulsar-hunter Andrew Lyne of the University of Manchester in England agrees: “The statistics are very sparse at the moment. Neutron stars with high magnetic fields simply may be an extension of the normal scheme with which we are familiar.”

Even if more magnetars turn up, astronomer Shri Kulkarni of the California Institute of Technology in Pasadena thinks they won't offer a catch-all explanation for the missing pulsars. “There's a bit of bandwagonism going on,” he says. “There is not just one class of objects we can call magnetars. My claim is that it's a zoo out there.”