# News this Week

Science  07 May 2004:
Vol. 304, Issue 5672, pp. 804
1. EPIDEMIOLOGY

# Public Enemy Number One: Tobacco or Obesity?

1. Eliot Marshall

Sloth combined with bad diet may soon displace tobacco as the biggest cause of avoidable death in the United States, according to a recent report by scientists at the U.S. Centers for Disease Control and Prevention (CDC) in Atlanta. But some researchers, including a few at CDC, dismiss this prediction, saying the underlying data are weak. They argue that the paper's compatibility with a new antiobesity theme in government public health pronouncements—rather than sound analysis—propelled it into print. The authors deny this, saying they relied on the best available data and methods, which were extensively reviewed before publication.

The study, published in the 10 March issue of the Journal of the American Medical Association (JAMA), estimates that in the United States in 2000, there were 435,000 deaths associated with tobacco use compared with 400,000 deaths from “poor diet and physical inactivity” (see table). The authors—four CDC epidemiologists, including agency chief Julie Gerberding—blame the category that includes obesity for causing the “largest increase” in deaths since a similar study 10 years ago. (The other authors are Ali Mokdad, James Marks, and Donna Stroup.)

View this table:

That conclusion galls anti-tobacco activist Stanton Glantz, an expert in heart disease at the University of California, San Francisco, who has made a career of battling cigarette companies. A tireless advocate of smoking controls, Glantz argues that the evidence on tobacco is well tested, whereas the new numbers on obesity are weak—or as one critic in CDC says, “loosey-goosey.” The General Accounting Office last year investigated CDC's tobacco-mortality estimates—at the behest of a legislator from a tobacco state—and gave them high marks.

Specifically, Glantz and others grumble that the CDC authors use inconsistent methods for calculating relative risks associated with tobacco and bad diet. According to Glantz, this study bases its obesity risk factors on studies of people who were more youthful on average than the U.S. profile. Death is more likely to be blamed on obesity among young people than among old people. This small bias, if projected onto the whole nation, can overstate obesity's importance. In contrast, risks for tobacco were calculated in an age-specific way and summed, taking out the age bias, critics claim. For this reason alone, Glantz argues, the paper should be “withdrawn.”

Several epidemiologists at CDC and the National Institutes of Health (NIH) echoed Glantz's concerns but declined to speak on the record. “I don't want to lose my job,” said one CDC staffer who does research in this area. Critics also object that the authors added an arbitrary number of deaths from poor nutrition (15,000) to the obesity category. A CDC scientist says internal discussions on these issues got “very contentious” months before publication and left some feeling that the conclusions were not debatable.

Not so, says Stroup, a mathematical statistician: The paper passed through an “extensive review by colleagues in the field, all the way up the chain, including by folks in the office of the director,” as well as a review at JAMA. This is not an original project, she explains, but an improvement on a similar report 10 years ago. That study of 1990 data, by former CDC chief William Foege and epidemiologist Michael McGinnis, put 300,000 deaths in the obesity category. It also was criticized for its sketchy description of sources and methods; the authors of the new version say one objective is to clear the fog. Public health leaders had been using the 1990 paper recently in a campaign against obesity, Stroup says; when Senator Bill Frist (R-TN) requested an update, CDC agreed.

The methods used to reach the 2000 findings are “well accepted in epidemiology texts and courses,” Stroup says. Co-author Mokdad thinks there are a lot of “misconceptions” about the paper. The stipulated 15,000 deaths from poor nutrition in the obesity section, for example, represent a “conservative estimate” obtained by tripling the number (4242) of death certificates citing this cause in 2000.

The CDC authors say that JAMA's space limitations prevented them from publishing the background information that some readers want. They hope to provide all of this—and more—on a CDC Web site in “about 1 month,” says Mokdad. His hope is to make the analysis as reader-friendly as possible, so that public health agencies across the country can use the same approach to calculate local threats from tobacco, obesity, and other killers.

2. GLOBAL CHANGE

# Getting Warmer, However You Measure It

1. Richard A. Kerr

Could thermometers be lying about global warming? They show that Earth's surface has been warming of late by a substantial 0.17°C per decade. That's about what climate models predict under the growing greenhouse. But at least by some analyses, orbiting satellite sensors have detected a distinctly slower warming in the lower atmosphere or even none at all. This gap between the comfortingly familiar thermometer and the globally comprehensive satellite has long been Exhibit A for contrarians arguing that the greenhouse effect is feeble.

Now, in this week's issue of Nature, a group of researchers describes a way to correct a long-standing problem in the satellite readings caused by the ongoing cooling of the stratosphere. Once adjusted, satellite temperatures show much the same warming in the lower atmosphere as do thermometers at the surface. “Within the uncertainty of reasonable methods of analysis of the satellite data, this paper essentially reconciles the satellite surface temperature trend debate,” says meteorologist Thomas Vonder Haar of Colorado State University in Fort Collins. “Our disparate observing systems are in agreement.”

For more than a decade, atmospheric scientists have argued about how to analyze weather satellite measurements of the atmosphere's microwave emissions (Science, 25 September 1998, p. 1948). The intensity of the emissions depends on the temperature of the layer of the atmosphere doing the emitting. But scientists disagree about how to analyze emissions from the lower atmosphere, a few kilometers above the surface, measured by a series of satellite-borne Microwave Sounding Units (MSUs) 850 kilometers up. Last year, three groups published MSU-based warming trends that ranged from 0.02°C to 0.10°C to 0.24°C per decade, bracketing the surface warming.

There is agreement, however, that all three of these reported trends fail to take into account the fact that the emitting layer in question extends into the lower stratosphere, where strong cooling has occurred due largely to the loss of ozone. In the Nature paper, meteorologist Qiang Fu of the University of Washington, Seattle, and his colleagues offer a way to correct for the stratosphere's effect on the calculated lower atmosphere temperature by using stratospheric MSU measurements and balloon temperature measurements. When applied to the middle of the three recent trend estimates, the correction raises the trend from 0.10°C per decade to 0.18°C per decade, versus the surface trend of 0.17°C per decade. Even the low estimate rises to 0.10°C per decade.

Fu and colleagues' approach “is a fairly simple but illuminating way of looking at the problem,” says meteorologist David Karoly of the University of Oklahoma in Norman. “I don't see any fundamental problem with the analysis.” That would leave researchers with the uncertainties evident in the spread of the three published estimates, but for now satellite temperatures can no longer be used to portray a feeble greenhouse effect.

3. U.S. RESEARCH POLICY

# NSF Warned Off Smithsonian Pact

1. Jeffrey Mervis

Responding to objections from an influential senator, the National Science Foundation has suspended plans to let scientists at the Smithsonian Institution compete equally with academic researchers for NSF grants (Science, 2 April, p. 26). University-based organizations are also fighting the planned policy change, because in their view it will increase the competition for limited NSF dollars.

“This is a very dangerous precedent … that will open up NSF's programs to all federal researchers,” Senator Kit Bond (R-MO) wrote on 23 April to Warren Washington, the chair of NSF's oversight body, the National Science Board. Bond, who chairs the Senate spending panel that sets NSF's budget, urged the board to reconsider. Four days later, Washington assured Bond in a letter that “I have asked the NSF acting director to discontinue negotiations [with the Smithsonian Institution]. … The board shares your concerns.”

The board triggered the controversy on 25 March when it voted to provide an exception for the Smithsonian's 500 scientists because of the institution's “unique status … and its special contributions to science research and education.” The board asked Acting NSF Director Arden Bement to negotiate an agreement with the Smithsonian and submit it for review. However, some members warned Bement to make sure that the agreement did not send NSF down a “slippery slope.”

Bond's opposition creates a mixed message from Congress. Leaders in the House of Representatives had advocated the policy change as a way to clarify NSF's inconsistent handling of applications from Smithsonian scientists, some of whom had been allowed to compete equally with university scientists while others were summarily rejected. Congress had urged NSF to clear up the confusion in a report accompanying NSF's 2003 spending bill, but one science lobbyist speculates that Senate aides may have signed off on the language “at 3 a.m., and probably nobody realized the implications at the time.”

The university community didn't object at the time but has now jumped on the bandwagon. On 28 April the Association of American Universities, which has posted the correspondence online (aau.edu), wrote the science board echoing Bond's fears of “a dangerous precedent.” Then it pressed home its point: “We believe that … researchers supported from [federal] agencies should not be able to supplement appropriated funds with research grants from other agencies.”

Bement says his talks with David Evans, the Smithsonian's undersecretary for science, are “dead in the water” pending guidance from Congress and the Administration, which this winter ordered a review of federal guidelines governing support for scientists at other agencies. “I don't think there's anything unique about the Smithsonian,” says Kathie Olsen of the White House's Office of Science and Technology Policy, who expects that the report, due this month, will prompt an interagency discussion of the issue.

4. FISHERIES SCIENCE

# Plan to Count Hatchery Salmon Criticized

1. Jocelyn Kaiser

In a decision with sweeping ramifications for conservation on the West Coast, the Bush Administration plans to count hatchery-raised fish along with wild fish in determing whether a Pacific salmon run is endangered. The move could leave some of 27 imperiled populations of salmon and steelhead trout without federal protection and ease development restrictions on millions of hectares where salmon spawn.

The draft policy, reported last week by The Washington Post, has pleased landowner groups but outraged environmental groups and some scientists. “This is not scientifically valid,” fumes Ransom Myers of Dalhousie University in Halifax, Nova Scotia. Myers and five other ecologists recently argued that hatchery fish develop traits and behaviors ill-suited to the wild and that including them in salmon counts could allow wild fish to die out (Science, 26 March, p. 1980). But National Oceanic and Atmospheric Administration (NOAA) Fisheries official James Lecky says that hatcheries “can be managed in a way that contributes to the conservation of endangered species” and that some hatchery fish should be counted.

The government has long dumped fish raised in hatcheries into rivers to maintain commercially valuable stocks harmed by dam-building and other activities. In 2001, a federal judge said NOAA treated hatchery fish inconsistently in its policymaking and ordered it to consider counting them when deciding whether a population is endangered (Science, 30 November 2001, p. 1806).

Last year, a NOAA advisory panel on which Myers served concluded that hatcheries don't help sustain a population over the long term and that no hatchery fish should be included in “distinct” wild populations. NOAA disagrees. Its draft policy says that “hatchery fish that are genetically no more than moderately divergent from a natural population … will be considered” in deciding whether to list a specific population.

Counting some hatchery fish is just one element of the policy, Lecky noted. More important, he says, the new policy upholds the importance of conserving habitat and calls for further studies. “We're looking to improve the science of hatcheries,” he says.

But outside observers remain dubious. Fish biologist Anne Kapuscinski of the University of Minnesota, Twin Cities, says that having scientific disagreement on whether hatcheries are always harmful is different from counting hatchery fish in listing decisions. “That's going too far,” says Kapuscinski, asserting that two National Academy of Sciences panels on salmon conservation on which she served “would not agree with this policy.” Chris Wood of Trout Unlimited says “the absurdity of the decision” becomes clear if it's extrapolated to other animals. By the same logic, he says, elk raised on game farms should be counted in deciding whether wild elk are endangered.

A federal court has ordered NOAA by 28 May to review the status of eight stocks now listed as endangered or threatened. Lecky, who hopes to include all 27, says there will not be a “wholesale” delisting.

5. NATIONAL SCIENCE FOUNDATION

# State Profiles Join Global Review of Science

1. Jeffrey Mervis

There's a saying that all politics is local. And this year the National Science Foundation (NSF) has applied that wisdom to its biennial compendium of facts about the global scientific enterprise with a first-ever analysis of how the 50 U.S. states stack up on some two dozen metrics.

The new chapter of Science and Engineering Indicators is destined to be a must-read among U.S. politicians eager to know whether their constituents are keeping up with the Joneses, scientifically speaking. Rather than simply documenting the hegemony of the biggest and wealthiest states, the measures offer insights into which states are capitalizing on their intellectual capital and which ones may be facing a brain drain. The chapter replaces one on information technology, perhaps a signal that the IT field has moved from curious phenomenon to essential fuel for each nation's scientific engine.

As always, Indicators is filled with intriguing international comparisons. Its figures and tables range from which countries lead the market for high-technology products and services to which nation's scientists have picked up the pace in the worldwide race to publish in the best journals. And best of all, it's available online (nsf.gov/sbe/srs/seind04) and on a CD-ROM.

6. BIODEFENSE

# Smallpox Vaccines: Looking Beyond the Next Generation

1. Martin Enserink

Smallpox may be gone, but poxvirus research—spurred by the specter of bioterrorism—is thriving. In the May issue of the Journal of Virology, a team of U.S. Army researchers reports that a new DNA vaccine can protect monkeys from what would otherwise be a lethal dose of monkeypox. The results are “very encouraging,” says Mariano Esteban, a poxvirus researcher at the Universidad Autónoma in Madrid, Spain, adding that they could prelude a safer generation of smallpox vaccines for humans to replace the old standbys from the smallpox eradication era. But DNA vaccines could be preempted: By the time they are developed, a much safer derivative of the current vaccine, called modified vaccinia Ankara (MVA), may be in production.

The standard smallpox vaccine, a virus called vaccinia, is amazingly effective, which explains how the disease was eradicated in the late 1970s. But it can cause rare but severe side effects—such as encephalitis and serious skin infections—in certain risk groups. And recently, a new side effect has shown up: When the U.S. government started vaccinating civilians and the military with a vaccinia strain called DryVax last year, a small number of recipients developed a heart inflammation, which further dampened enthusiasm for the ambitious campaign.

A modern-day version of DryVax, of which the U.S. government has purchased more than 200 million doses (Science, 15 August 2003, p. 912), faces similar problems. The company that produces it, Acambis in Cambridge, Massachusetts, announced on 13 April that it has halted its phase III clinical trial after three vaccinees developed myopericarditis.

Searching for alternatives, Jay Hooper and his colleagues at the U.S. Army Medical Research Institute of Infectious Diseases (USAMRIID) in Fort Detrick, Maryland, developed a vaccine composed of the genes for four vaccinia proteins that were thought to be important in eliciting immunity. In a previous study, they had shown that the quartet protected mice from vaccinia, which usually kills them; the same combination, dubbed 4pox, also elicited antibodies in rhesus macaques.

In the new study, the Army team put three of those animals to the test, boosting their immunity with another 4pox shot and then challenging them with a monkeypox dose that killed nonvaccinated monkeys. All three survived, with only mild to moderate disease. Two other monkeys that were given DryVax did better, as expected, developing no disease at all.

The study shows that the DNA approach has potential, says vaccine researcher David Weiner of the University of Pennsylvania in Philadelphia, although he cautions that 4pox might be less effective in humans. “Mice are liars, and nonhuman primates exaggerate,” he says. Still, says Weiner, there are ways to enhance efficacy, for instance by trying more or different genes or by using a so-called prime-boost strategy, in which the DNA vaccine is followed by a shot with the protein subunits themselves. “We can't say we have a new smallpox vaccine yet, but it's an incredibly important step,” he says.

Where that step will lead will depend, among other things, on the fate of the MVA vaccine now in development. Produced by passaging vaccinia 574 times in chicken embryo fibroblasts and widely used in Germany in the 1970s, MVA is known to be much safer than DryVax. The U.S. National Institute of Allergy and Infectious Diseases (NIAID) is supporting its development by Acambis and the Danish-German biotech Bavarian-Nordic and plans to buy more than 50 million doses—enough to vaccinate those in the United States who can't take DryVax.

But as a live vaccine, MVA also carries some risk, says Hooper; it may have unpleasant surprises in store. NIAID Director Anthony Fauci agrees. “MVA is clearly marching along towards licensure,” he says, “but it never hurts to have a couple of irons in the fire.”

7. BIRD EVOLUTION

# Surprise Hummingbird Fossil Sets Experts Abuzz

If it's hummingbirds you're after, the New World is the only place to be. Of the 300-plus species of the hovering, nectar-sipping birds, almost all live in Central and South America. Experts agree that all species of modern hummingbirds evolved there and later spread to North America, but it appeared they had never set wing in Eurasia.

Now, fragile bones in 30-million-year-old rocks from southern Germany show that hummingbirds were much farther-flung than anyone expected. “The amazing thing about this fossil is that it's essentially a modern hummingbird,” says Margaret Rubega of the University of Connecticut, Storrs. “My mind is a little blown.” The discovery, which ornithologist Gerald Mayr of the Forschungsinstitut Senckenberg in Frankfurt, Germany, describes on page 861, raises questions about where early hummingbirds evolved and why the European ones became extinct.

Hummingbird history has long been shrouded in mystery, chiefly because the delicate-boned creatures have left so few fossils. None at all have been found in the Western Hemisphere. Hints of Old World origins appeared when a possible primitive insect-eating hummingbird, Parargornis messelensis, turned up in 49-million-year-old rocks in Messel, Germany. The only other fossil hummingbirds are the 30-million-year-old Argornis caucasicus and Jungornis tesselatus, both incomplete, from the Caucasus. They appear to have been able to hover, but it's not clear whether they had modern-style beaks. Last year Mayr classified all three as “stem taxa,” extinct relatives that share a common ancestor with modern hummingbirds, but not all experts were convinced.

The new fossil, called Eurotrochilus inexpectatus, is the first fossil of a modern-looking hummingbird and the closest to modern ones. When Mayr came across two partially prepared specimens of the creature in a museum drawer in Stuttgart, “I didn't have a real idea what it was,” he says. But closer inspection revealed evidence that the specimens were hummingbirds. One of them sports a long, slender beak adapted for feeding on nectar. The clincher was the short, stocky humerus with a bony knob that probably allows the wing to rotate during hovering flight. Although the tip of the beak is not preserved, Mayr estimates that the bird measured 4 to 5 cm from head to tail, as big as a medium-sized modern hummingbird.

Not everyone accepts the identification. “The similarities with modern hummingbirds are pretty superficial,” says Joel Cracraft of the American Museum of Natural History in New York City. Other experts, however, say they are convinced.

Eurotrochilus demonstrates that in the Old World, hummingbird ancestors had evolved the main features of living hummingbirds by 30 million years ago, Mayr says. That might explain why a handful of European flowers appear adapted for hovering birds, he adds. It could be that these plants first evolved with hummingbirds and were pollinated by them. This conclusion makes sense to Ethan Temeles, an ecologist at Amherst College in Massachusetts who studies the coevolution of plants and pollinators. Finding a fossil pollinator, as Mayr has done, can help explain the evolutionary history of both the plant and pollinator, he says.

Why did Old World hummingbirds become extinct? One possibility, Mayr says, is that songbirds outcompeted them. Where the whole hovering tribe came from, meanwhile, remains up in the air. The four European fossils could suggest that stem taxa first appeared in the Old World and spread to the Americas across the Bering Strait, but the fossil record is so sparse that it's just speculation. Answering those questions will take more discoveries. For paleontologists scouting for clues to hummingbird history, the Old World may become the new place to be.

# Harvard Joins Reform Movement

1. Andrew Lawler

BOSTON—Harvard University has joined a growing number of elite schools attempting to revamp undergraduate science education. The effort, which is part of a larger rethinking of Harvard's entire undergraduate program, could double the number of science courses required of nonscience majors, provide a more interdisciplinary approach to life and physical sciences, and encourage students to conduct research abroad.

Last week, a panel of students, faculty, and administrators delivered a 69-page report* that proposes new introductory science courses and urges the university to give its undergraduates “a genuine view of the excitement of research science.” Scientists across the country say that changes at Harvard College, to be spelled out over the next year, are sure to spark increased interest among other universities in overhauling undergraduate science courses.

The reform movement is driven by concern that many undergraduates are turned off by their science courses and leave school without an appreciation for research. To address that problem, this fall Columbia University in New York City will require that all undergraduates take a general course called “Frontiers in Science” (Science, 18 October 2002, p. 531). Taught by star professors, the course will include small weekly seminars. Other schools, such as the University of Michigan, Ann Arbor, are experimenting with greater undergraduate involvement in research projects. And the University of Wisconsin, Madison, has restructured its tenure and merit-pay criteria to encourage better science teaching.

The Harvard report examined the needs of undergraduates in three categories—science majors, those preparing for medical school, and those whose interests lie outside science. It recommends that all undergraduates take one interdisciplinary course in life sciences and one in physical sciences, rather than the current system of choosing an introductory course in one of the two fields. “This is a significant new emphasis on educating all students,” says Harvard biologist Richard Losick, who was involved in the study. “This will be real science teaching, not the history of science or science for poets.”

Those majoring in science, says the report, should be given more opportunity to experience how science is conducted by working in a research lab. The report's emphasis on international experience, says Losick, should be extended to opportunities for doing lab science around the globe.

Benedict Gross, a mathematician and dean of Harvard College who co-chaired the study, says Harvard this summer will appoint a science working group to come up with a detailed science curriculum. The details could be ready for discussion by the entire Harvard community by the end of the next academic year, he adds. The last major change to Harvard's undergraduate curriculum took place in 1978.

Harvard's freshman class of 1650 is tiny compared to the enrollments of many state universities. But outside scientists and administrators say having Harvard on board should further their reform campaigns. “This is terrific,” says Peter Bruns, vice president for grants and special projects at the Howard Hughes Medical Institute in Chevy Chase, Maryland, which has funded efforts to improve undergraduate education. “And it is about time, since we live in an increasingly scientific world.”

9. NEUROSCIENCE

# Locating a New Step in Pain's Pathway

1. Jean Marx

As pain sufferers can attest, there's room for improvement in painkilling medications. Many of the current ones can cause side effects, such as stomach ulcers, particularly in people who have to take them over long periods of time for conditions such as arthritis. Work described in this issue now points to what may be a good new target for analgesic drugs; it also sheds light on an inflammatory pain sensitization, which causes patients to feel intense pain even in response to normally innocuous stimuli such as a light touch.

On page 884, an international team led by Ulrike Müller of the Max Planck Institute (MPI) for Brain Research in Frankfurt, Germany, reports having identified the α3 form of the receptor for the neurotransmitter glycine as a key intermediate in transmitting pain signals from the spinal cord to the brain. The work shows that the receptor is needed for pain sensitization—the first time a function has been identified for this particular receptor.

Previously, pain sensitization, or hyperalgesia, was thought to be due predominantly to changes at the inflamed sites themselves, but recent evidence suggests that alterations in the spinal cord are even more important. The new findings now confirm that; they show that the glycine receptor is the target of prostaglandin E2 (PGE2), a pain-sensitizing signaling molecule. “They identified a specific component that shows how this [pain sensitization] system works,” says Anthony Yaksh, whose group at the University of California, San Diego, also has evidence that PGE2 acts in the spinal cord.

When Müller began her work several years ago, her goal was to identify a function for the α3 type of glycine receptor. Robert Harvey, a former MPI colleague now at the School of Pharmacy in London, U.K., produced antibodies that enabled the team to locate the receptor in the spinal cord. In contrast to other glycine receptors, which are found throughout the cord, the α3 receptor occurs only in a particular layer where pain neurons from the peripheral tissues terminate. That suggested it might somehow be involved in transmitting pain signals.

At that point, the Frankfurt team joined forces with Hanns Ulrich Zeilhofer and his colleagues at the University of Erlangen-Nürnberg, whose earlier work had hinted at a possible role in pain for a glycine receptor. It showed that PGE2 inhibits glycine receptor activity in the same spinal cord layer where the MPI group had found the α3 receptor. Because glycine receptors suppress neuronal firing, inhibiting them with PGE2 could facilitate transmission of pain signals to the brain. At the time, however, there were no tools with which to prove that. But now Müller and Harvey wondered whether the Zeilhofer group had encountered the α3 receptor. Further studies by the Zeilhofer group confirmed that hypothesis. Recordings of electrical signals from spinal cord neurons showed that PGE2 inhibition was totally absent in tissue from mice in which the α3 receptor gene had been knocked out.

To pin down the receptor's role, the researchers compared pain responses in the knockout mice with those in normal animals. The two groups responded the same way to acute pain stimuli. But when the researchers first induced inflammation in the animals' paws by injecting an irritant, they found that the prolonged sensitization to further pain stimuli seen in normal animals did not occur in the knockouts. “Central prostaglandins work only through this one receptor” to promote pain sensitization, says pain expert Clifford Woolf of Harvard's Massachusetts General Hospital in Boston.

That also means that aspirin and other so-called nonsteroidal anti- inflammatory drugs, which produce analgesia by blocking prostaglandin production, exert their effects in the spinal cord, and not just in peripheral tissues as was once thought. And as Yaksh points out, the α3 glycine receptor itself “potentially represents an important target for drugs in chronic pain states.” If researchers find drugs that stimulate the receptor, they may be able to expand the arsenal of analgesics.

10. STANDARDS

# A Most Unbearable Weight

1. Gabrielle Walker*
1. Gabrielle Walker, a writer based in London, is the author of Snowball Earth.

The kilogram is the only unit of measure still defined by a physical object. Now, a marathon effort to tie the kilo to a constant of nature is nearing the finish line

LONDON—”You're in luck!” says Stephen Downes. An unlikely looking treasure is making a rare foray into open air. Perched on the pan of a balance in the U.K.'s National Physical Laboratory is a squat cylinder of metal, some 4 centimeters tall, with a faint number 18 engraved on its dull gray side. This unprepossessing lump is so precious that it spends most of its life locked away in a bombproof subterranean vault.

That's because “kilogram-18,” as it is known, is the ultimate arbiter of every weighing scale in Britain. Like its clutch of siblings around the world (the U.S. equivalent, kilogram-20, is converted into pounds), number 18 is ruled over by a chunk of metal in Paris. Dubbed “Le Grand K,” this is the granddaddy of weights the world over, the literal embodiment of the kilogram, and it has been removed from its heavily guarded chamber just three times in 120 years.

The care with which these objects must be handled is a constant exasperation to physicists. “It's a dinosaur of a process,” says Downes. Whereas other units of measurement are defined by constants of nature such as light speed or atomic vibrations, the kilogram alone in the scientific lexicon is still tied to a physical object. “The kilogram is a black spot on the white jacket of the entire measuring system,” says Peter Becker of Germany's national metrology institute, the Physikalisch-Technische Bundesanstalt (PTB) in Braunschweig.

Researchers around the world have been trying to eradicate this spot for decades, using approaches ranging from counting the number of atoms in a kilogram to converting the mass into a kind of electric force. Now at last the electric approach is pulling ahead of its rivals, and results due later this year or early next year could finally set in motion the redefining of the kilogram. The atom counters, however, have not yet given up hope. Last month an international consortium began a last-ditch attempt to bring its troubled project back into play.

The problem is more than aesthetic: Unlike physical constants, an artifact can change over time. Although kilogram-18 and the other national standards have differed very little from the one true kilogram in Paris over the past century, they could still be changing in unison. All were forged in the 1880s in a heavily polluted London, and if molecules of pollutants have been slipping back out of the metal matrix, the standards may have been losing mass ever since. Or they may have gained mass. The atmosphere contains more mercury than it did 100 years ago, and the kilograms are made from an alloy containing platinum, which soaks up mercury.

“That's the scary part,” says Richard Steiner of the U.S. National Institute of Standards and Technology (NIST) in Gaithersburg, Maryland. “We believe that they're not changing very badly or noticeably, but because there's no absolute way to measure we don't know for sure.”

## A fine balancing act

Elsewhere in the National Physical Laboratory, Downes's colleague Ian Robinson is tweaking the apparatus that he hopes will put kilogram-18 and kin out of business. Known as a “Watt balance” after the unit of electrical power, it stands some 2 meters tall in an air-conditioned inner sanctum. It looks at first like a sophisticated version of any standard set of weighing scales, its centerpiece a hefty aluminum arm bearing metal balance plates on either end. But the right side is simply a counterweight. On the left, the action end, a standard gold-plated kilogram (a copy of the original) dangles above a spiderlike Pyrex cage wrapped with copper coils. The cage dips into a magnet so strong that even the stray field outside its casing can stop analog watches in their tracks, and any wrenches in the room must be carefully hooked on bolts to prevent them from crashing into its walls.

As the wire coil sits in its magnetic field, a current passing through it generates an upward electromagnetic force. Robinson adjusts the current until this force exactly balances gravity's downward pull on the kilogram. Next, he removes the kilogram and measures the strength of the magnetic field by moving the coil down through the magnet and monitoring the voltage generated. A laser pointing up through the barrel of the magnet measures the coil's movement, and the time it takes is judged against an ultraprecise reference signal piped in by cable.

Elsewhere in the room, machines are busily measuring other parts of the equation. A six-legged gravimeter monitors the pull of Earth's gravitational field, and a large stainless steel Dewar contains devices that measure the wire's voltage with extraordinary precision, thanks to quantum effects that force electrons to adopt exact energy levels, like rungs on a ladder.

Robinson's goal is to tie the kilogram to fundamental constants of nature with an accuracy of one part in 100 million, which is a little less than the amount by which Le Grand K and its siblings have drifted apart over the past century. If he succeeds, he will be able to redefine the unit of mass in terms of length, time, gravity, and a number called Planck's constant, which is related to the spacing of energy levels in the quantum ladder.

To achieve this, the machine must be staggeringly sensitive. Even a faint thumbprint on the test mass would send the results way out of whack, as would any stray vibrations. Experiments are run at night when the building is otherwise unoccupied; during runs, a red tube on the wall lights up, advising researchers to “MOVE GENTLY.”

Robinson has flirted with measurements at a level of parts per 108, but he has been chary of publishing because the values drift frustratingly after just a few weeks. He thinks he knows where the problem lies. The measurements must all be made under vacuum to eliminate the buoyancy of air, and Robinson suspects that when he pumps out the air, the delicate machinery's alignment shifts slightly. He is now exploring how to realign the machine after the air is removed.

Meanwhile, a Watt balance at NIST also looks promising. In 1998 Steiner and his colleagues published results accurate to one part in 10 million. With a revamped machine, they are obtaining values at a few parts in 108, although they too are plagued by drift. Steiner thinks that electrical noise is skewing the results: “The smaller the effect, the more causes you possibly have, and that's what makes it hard.”

Still, the success of these two Watt balances has brought new machines springing up like mayflowers. The Swiss have a prototype in operation, and one is being built in Paris. “It's a very powerful technique,” Robinson says. “Though they all use the same basic approach, this machine, the one at NIST, and the Swiss one are physically completely different. If they all produced the same value independently, it would be a very strong indication that the technique is working. And that's the sort of assurance you need before you can start talking about redefining.”

## Atomic number crunching

Just as things are looking good for the Watt balance teams, their chief rivals, the atom counters, have hit a severe problem. This approach is more intuitive than the complex Watt balance method is: It simply involves counting the number of atoms in a known mass. Adding up something like 1023 atoms individually would take far too long, so several groups have been trying to find clever ways of speeding up the process. Michael Gläser of Germany's PTB has been sending a beam of gold ions into a collector and using the current they carry to tally them as they accumulate. The trouble is that ions can bounce. Any ions that deliver their charge but then spring back out of the box send the counting hopelessly askew.

More promising is an international collaboration named after the Avogadro constant, the total number of carbon-12 atoms in 0.012 kilograms. The Avogadro project involves grapefruit-sized, ultrapure silicon spheres, manufactured so that each is a single crystal with no internal voids. Measuring the diameter of the spheres and the spacing between their atoms should reveal exactly how many atoms they contain.

The problem is that even the purest silicon from the semiconductor industry comes in an array of isotopes, each with a slightly different mass. So the Avogadro researchers must know the precise combination of isotopes in any one sample. And it is in measuring these ratios that they have hit a roadblock. Nearly a decade ago, they came up with an answer to a few parts in 10 million, but try as they might, they have gone no further.

Hoping to get back on open road, the PTB last month signed a contract with former nuclear technologists in Russia to make highly enriched silicon in which 99.99% of the atoms are Si-28. Labs around the world are eagerly awaiting this material, which should be ready in 2006.

However, because the material for a single 1-kilogram sphere will cost more than €1 million, there will be only one to go around. “The game's not over, but it's taken an expensive turn for the Avogadro project,” says Downes. And even that single, pristine sphere may contain hidden blemishes. Downes has discovered that a surface film of silicon oxide or organic contaminants could distort the diameter measurement. He acknowledges that these complications put the Avogadro project on the back foot. “The Watt balance does seem to be getting more of an edge,” he says.

Other researchers are murmuring that the enriched sphere is beginning to sound like tying the kilogram to yet another physical object. “You can make and measure one perfect silicon sphere, but then that's it,” says NIST's Steiner. “You're basically creating a new absolute artifact.” Becker disagrees but concedes that, given the astronomical expense involved, this is likely to be a one-off experiment rather than a reliable new recipe for the kilogram. Even so, he says, the enriched silicon should provide a vital test of the Watt balance results. “When you want to change the definition of the kilogram, you need at least two independent methods,” says Becker. “If we provide a check on the Watt balance method, we'll have done our job.”

And even though confidence is building in the Watt balance labs, researchers there all declare that they and the Avogadro labs are engaged in a collaboration rather than a race. “Nobody will act on one lab's results,” says Robinson. “It's going to take a lot of results from a lot of places before anyone is prepared to move away from the kilogram. Even if you come in with your results first, you still have to wait around for everybody else to arrive with theirs.”

The verdict from the Watt balances should arrive within the next few years. The final step will be persuading the General Conference on Weights and Measures, the international committee that serves as the guardian of measurement systems, to take the new definition on board.

The Watt and Avogadro scientists, at least, can't wait to consign kilogram-18 and its compatriots to the scrap heap. “We've been doing this the same way since the 18th century, and it would be nice to move on,” says Robinson. “A system based on nature will always be better than one based on artifacts, because it's likely that nature will last a little longer.”

# Reclaiming Iraq's Past: Life on the Front Lines

1. Andrew Lawler

Three foreign scholars in Iraq are helping repair, protect, and rejuvenate the country's shattered artifacts, ancient sites, and institutions

When the Bush Administration was preparing to invade Iraq in March 2003, many archaeologists had a personal stake in trying to avert the conflict. They feared for the safety of their Iraqi colleagues and worried about the impact of modern warfare on the innumerable ancient sites that dot Mesopotamia's landscape. The subsequent looting 1 year ago of cultural institutions such as the Iraq Museum and the continuing pillage of sites around the country confirmed their worst fears. Profoundly shocked, many scholars who study this region denounced the U.S.-led coalition forces for failing to control the looters.

But after Saddam Hussein was toppled, archaeologists had to decide whether to make common cause with the very coalition forces they had criticized. Three researchers, including two U.S. archaeologists and one raised in Iraq but living in London, chose to plunge into the maelstrom of occupied Iraq. Each was skeptical about the rationale for the war, though all were critical of Saddam's repressive regime. Now, with the aid of U.S. government funds, they are playing pivotal roles in rebuilding Iraq's damaged universities and museums and are reconnecting their near-destitute colleagues with the international scientific community.

It is no simple task. Going to work requires an armed guard, and their nights are punctuated by gunfire. But the three are betting that their efforts will benefit the next generation of Iraqi and foreign scholars.

## Lamia Al-Gailani:Insider With a Sense of Urgency

When soldiers brought back the famed Besetki statue to the Iraq Museum last fall, it was covered in excrement. The looters who took the 4300-year-old massive bronze sculpture—one of the museum's most prized objects—hid it in a cesspool outside Baghdad. But Iraqi-born Lamia Al-Gailani, who met the soldiers at the museum door, wasn't repelled by the strong odor. “It is an absolutely super piece; it has long been my favorite,” she says with enthusiasm.

That moment was one of the few happy ones in the past year. As a consultant to the Iraq Reconstruction and Development Council, which is funded by the ruling coalition, Al-Gailani's job is to help museum employees reorganize both the hundreds of thousands of artifacts and the institution itself. Beyond the orgy of looting in April 2003, the museum also suffers from years of neglect. “By last April, there was no official photographer, no specialist in Sumerian art, and no one who knew anything about Greek, Roman, or Sassanian coins,” she told a packed auditorium of archaeologists at a meeting last month in Berlin.

Most foreign archaeologists are reluctant to criticize their hosts publicly. But not Al-Gailani, whose Baghdadi family roots go back to a famed medieval philosopher. In fact, she was the first Iraqi woman to participate in excavations, and she worked for the museum for 8 years before leaving in 1970 to earn a Ph.D. at University College London, where she still lives. An independent researcher, she returns to Iraq nearly every year for research on Babylonian cuneiform seals, many of which were stolen in last year's looting. “I'm not counted as an outsider,” she says.

Al-Gailani feels a great sense of urgency. “It's not an exaggeration to say that thousands of objects need work, from metal ones to pottery,” she says. And that urgency goes back to the first Gulf War. In 2000, Al-Gailani was there when museum officials opened boxes of artifacts hastily packed away before U.S. bombs arrived in 1991. “Quite a lot of the ivories”—many from the 9th century B.C.E. Assyrian capital of Nimrud—“had completely crumbled, and the beautiful Halaf pottery was covered in black mold,” she recalls. “It is awful to remember.”

Today, hundreds of the museum's most precious objects hidden since before the 2003 war remain in an undisclosed location and in an unknown state of preservation. They'll stay there until the museum finishes upgrading its security and climate systems. Italian conservationists have refurbished the museum lab that was vandalized during the looting and are working with a handful of Iraqi specialists to restore objects smashed or damaged during the looting and during the decade of neglect. But Al-Gailani says a much more intensive effort is needed to save the large number of objects that are in a precarious state of preservation.

She hopes “to leave the museum in the 21st century” when she departs this summer. She's also trying to revive Iraq's only archaeology journal, Sumer, which was published sporadically during the 1990s. And she is starting to raise money for exhibitions, despite the lack of immediate plans to reopen the museum to the public. “The begging bowl,” she notes, “is out.”

“It's give-back time,” says Elizabeth Stone, an archaeologist at Stony Brook University in New York. A vocal opponent of the U.S. invasion, she has shelved her academic research into ancient Mesopotamia to help Iraqis at the universities of Baghdad and Mosul rebuild their shattered archaeology and environmental health departments. Directing a $10.9 million, 3-year grant from the U.S. Agency for International Development (USAID), Stone is shuttling back and forth to Iraq to set up educational programs and computer systems. This summer she plans to bring home a half-dozen students who hope to get their Ph.D.s and become the vanguard of a new generation of Iraqi archaeologists. The peripatetic 55-year-old speaks as quickly as she moves. In one recent 2-week trek she traveled from New York to Baghdad to Oxford to Paris to Berlin and back home, a pace that leaves little time to worry about safety. During the 1980s, while researching cuneiform tablets housed at the Iraq Museum, she woke occasionally to the sound of Scud missiles from Iran plummeting into the capital. “They made a hell of a noise,” she recalls. “And I always seemed to be sleeping next to a plate-glass window.” As part of the USAID contract, Stone will conduct two 5-week intensive sessions covering everything from micromorphology to computer mapping to cuneiform. Some 45 Iraqis, from professors to students, are slated to attend. But simply finding a venue is a challenge. She suggested the northern Iraq city of Erbil, a more secure city than Baghdad but still a domestic trip. “Many don't even have passports,” she notes. But many Iraqis fear that being seen with foreigners will label them as collaborators. Meanwhile, Stone is trying to acquire high-speed Internet access, computer software, Global Positioning System equipment, and cameras for the shattered archaeology faculties at Baghdad and Mosul universities. They will also need training, as many faculty members and students have little experience using such modern archaeological tools. “A lot of good Iraqis want to rebuild the country, but we are all handcuffed,” says Yasin Ali, the only member of Mosul University's archaeology/philology department with a Ph.D. “We have to work without electricity at temperatures above 50°C—we sweat as we teach,” he says. Last month Stone and other team members interviewed 26 candidates for a Ph.D. program at Stony Brook that USAID will support. Eight were selected, but Stone says she was “devastated” by the erosion of English language skills since the 1980s. “Only one passed the English exam,” she said. The students will first receive intensive English instruction—if they can obtain visas. “If we can train six people, that will have a tremendous impact on the future,” says Stone. ## John Russell:Fighting Against Impossible Odds One year ago, John Russell was decrying the havoc wreaked on Baghdad's cultural institutions in the midst of the U.S. invasion. Now Russell, from Boston's Massachusetts College of Art, is finishing a sabbatical during which he has tried to reassemble some of that heritage. For the past 9 months, Russell has lived in a trailer beside Saddam's former palace in central Baghdad, protected by U.S. soldiers and an American-issued bulletproof vest while serving as an adviser to Iraq's Ministry of Culture. When he was asked by U.S. officials to take the job, Russell hesitated before deciding that he couldn't leave his Iraqi friends and colleagues in the lurch. Now he is wrapping up a most unusual sabbatical year. “I'm on the floor,” he apologized during a recent phone interview. “We just got bombed; every few evenings we get a hit.” Russell's earlier work documenting the remaining carved friezes at King Sennacherib's palace at Nineveh near Mosul led to the return of at least one part of a slab from a British collector who had benefited from the looting that became widespread after the first Gulf War. He has been outspoken in criticizing the antiquities trade. But reining in that trade in today's Iraq is proving a tough task; devastating looting continues in the Iraqi countryside. Russell is also helping refurbish the Iraq Museum by drumming up funds, working closely with the museum staff, and finding and managing the necessary contractors. As he prepares to return home, he's not optimistic about the chances of slowing the destruction of the country's heritage. “No one wants to hear this, but it is going to be impossible to win this battle the way things are now,” he says. Russell took a helicopter tour of southern sites in January to document the looting, which has left many ancient sites resembling a slice of Swiss cheese. But that devastation, he notes, began during the mid-1990s, when the lure of digging for antiquities—particularly in the vast and unruly south—proved too tempting for a nation in an economic tailspin. Carabinieri in charge of the southern region now patrol “aggressively” using helicopters and conducting raids, Russell says. But the economic ruin makes looting a profitable business, and looters simply avoid guarded sites and go elsewhere. “Now they are using floodlights to dig at night,” he adds. “It's a problem that the coalition can't reasonably be expected to turn around in 1 year.” Russell prays for more legal excavations and more prosperity in the southern region. But neither seems very likely. Legal digging is difficult and dangerous, although a team of Austrian archaeologists quietly continued working as recently as last fall at Borsippa, an ancient city southwest of Baghdad. Only a handful of southern sites—where the bulk of Iraq's mounds exist—are secured. An American military base surrounds the ancient city of Ur, and a German and an American research team continue to pay for guards at, respectively, Uruk and Nippur, important Sumerian cities that flourished in the third millennium B.C.E. But the vast majority of sites—most of which have never been legally excavated—are vulnerable, and few archaeologists plan to visit them anytime soon. Russell hopes that a few hundred archaeological site guards now being trained will go at least partway to restoring order. Some of his Iraqi colleagues propose more drastic measures, such as bombing the trucks and buses that take looters to sites every day. Russell demurs, saying, “Americans don't do that.” But he wouldn't mind confiscating such vehicles. In the meantime, he savors small victories, such as a new protective roof over Sennacherib's palace in Nineveh and armed guards to patrol the beloved site. 12. HUNTINGTON'S DISEASE # Unorthodox Clinical Trials Meld Science and Care 1. Jennifer Couzin A husband-and-wife team touched by the fatal disease will test six compounds in Huntington's sufferers. Patients are thrilled, but neurologists are skeptical After her husband was diagnosed with Huntington's disease in 1997 at age 54, Gayle Tinnerman began to steep herself in medical lingo. She scoured the Internet for experimental treatments, anything that might slow the horrific neurological decline that assaults Huntington's sufferers: involuntary movements that make even reading a book impossible; depression and emotional outbursts; a gradual inability to swallow food, to talk, and to think; and, inescapably, death. Tinnerman, who lives in San Diego, California, is frustrated by what she reads. Mice with a version of Huntington's would receive a new drug and their condition would improve, but the drug would not be quickly tested in people. “It's getting a little old, bringing these mice articles home,” she says. “My husband said, ‘The mice keep getting better. What about me?’” So, when Tinnerman heard about an unorthodox drug-testing regimen, she jumped at the chance to sign up. The organizers are a husband-and-wife team: physician LaVonne Veatch Goodman, whose first husband died of Huntington's and whose children are at risk for the disease, and computer scientist Nathan Goodman, who works at the Institute for Systems Biology in Seattle. Exasperated by scientists' reliance on traditional clinical trials and the lack of proven therapies, the Goodmans are enlisting dozens of Huntington's patients to test a combination of one drug and five supplements, all of them already on the market. Their goal is twofold: quickly distribute compounds that could help patients and learn something about the effects of these compounds on the disease. They readily admit that the two goals sometimes collide. Many neurologists are skeptical about linking these objectives. Although some quietly applaud the Goodmans for bending standard protocols to try to improve clinical studies, they and others worry that the pair is compromising science. They question whether the program's uncharacteristic design of “individualized” trials, in which patients may be on different combinations of compounds and are assessed by a home computer program, rather than in a neurologist's office, will yield any useful information. There's also the chance that the project may diminish the pool of subjects available for rigorous clinical trials—the kind most scientists are comfortable with. Huntington's families, in contrast, have fiercely embraced the Goodmans' plan, called Huntington's Disease Drug Works (HDDW). Dozens are clamoring to sign up. They cite not just the perceived sluggish pace of traditional research but also the excessive focus on finding a Huntington's cure, slighting less glamorous but potentially valuable therapies. “Anything that offers hope, we'll grab it,” says Peggy Maceo of Austin, Texas, whose sister suffers from the disease. One thing skeptics and supporters agree on, however, is what's driving HDDW: 11 years after discovering the defective gene, the field is tantalizingly close to finding therapies for a disease that currently has none. “Once you see light at the end of the tunnel,” says Martha Nance, who oversees the Huntington's program at Hennepin County Medical Center in Minneapolis, Minnesota, “people are going to start disagreeing” about how to get there. ## Pushing the envelope Veatch Goodman's motives are personal: She has two grown children from her first marriage, and each has a 50-50 chance of developing Huntington's. Like 95% of people in this position, her children have declined gene testing. But they're approaching the age when, if Huntington's is etched in their genes, they'll begin showing symptoms. Veatch Goodman, too, is frustrated that few compounds that show promise in mice ever advance to human studies. Roadblocks arise because clinical trials are costly and Huntington's patients are few—just 30,000 live in the United States—so drug companies have little incentive to pursue treatments. At least one new Huntington's research group, the High Q Foundation, believes the field is ripe, and it plans to coax companies to come into it. But Veatch Goodman doubts that the conventional approach will suffice. The sheer volume of compounds that show hope in mice makes methodical, one-by-one clinical testing unwieldy. When the Goodmans decided to set up their own organization, they sought ideas from advocates for other diseases who'd stepped outside mainstream medical research. One is James Heywood. A young engineer, Heywood was spurred to action in 1999 after his 29-year-old brother Stephen was diagnosed with amyotrophic lateral sclerosis (ALS). Heywood quit his job at the Neuroscience Institute in La Jolla, California, raised millions of dollars, and launched a vast mouse-screening factory for potential ALS drugs. Today, Heywood's ALS Therapy Development Foundation in Cambridge, Massachusetts, runs four mouse-drug studies for ALS every month, posting all the results—positive and, a rarity in science, negative—on the foundation's Web site. Earlier this spring, a book by Jonathan Weiner chronicling Heywood's efforts, His Brother's Keeper, hit bookstores. Although Heywood inspired the Goodmans, they were deterred by the high cost of his approach—a single mouse study runs upward of$50,000. Besides, they reasoned, plenty of Huntington's mouse work was already under way in academic labs.

Instead, the pair chose to do human trials of compounds already on the market—either nutritional supplements or drugs approved for other diseases—that had been tried at least once in Huntington's mouse models and had a strong safety profile. They also decided to combine them—a heresy in drug testing because this makes it much harder to assess the safety and effectiveness of each drug, something the Goodmans weren't interested in. Only a handful of clinical trials employs drug combinations; many are at M. D. Anderson Cancer Center in Houston, Texas, where an overhaul in the hospital's biostatistics department spurred innovative trial designs (Science, 6 February, p. 784). The Goodmans' approach differs from Anderson's, though: It is as much therapeutic as experimental. “We are making it much easier to access the drugs,” explains Veatch Goodman.

The Goodmans selected two prescription drugs and five supplements; they presented the plan to an independent review board in Olympia, Washington, that normally reviews industry-sponsored studies. Its members balked at the high dose of one drug, a liver disease therapy called ursodiol, which was then set aside. But, in April, it approved the other substances and gave the Goodmans the go-ahead to enroll 50 patients in HDDW's first year and 50 in each of the 2 years following.

The lineup includes one drug, cysteamine, used to treat an extremely rare metabolic disorder in children called cystinosis, and five supplements: creatine, traditionally used to boost athletic performance; coenzyme Q10, claimed to remedy and prevent a number of diseases; omega-3 fatty acids; a sugar substitute called trehalose; and blueberry extract. (Although blueberry extract has not been tested in Huntington's mouse models, it looked promising in a peer-reviewed study treating rats with a version of Alzheimer's disease.)

A top priority is to get the compounds out fast—and free—to families. Neither cysteamine nor trehalose is widely available, but the Goodmans bought stocks of both from companies. Indeed, companies supplied two of the supplements free and the other three at a reduced cost. (HDDW is run on a shoestring budget, largely with a \$100,000 gift from an anonymous donor.)

## Inviting doubt

Many mainstream researchers are skeptical about HDDW's methods, saying they don't stack up against traditional clinical trials. One question people ask is, How will these studies get a handle on a patient's baseline health, critical to gauging a therapy's effects? The Goodmans are requesting—but not mandating—that participants gather 3 to 6 months of data before beginning, using a computer program designed by Nat Goodman and Jerry Lampson, another Huntington's advocate. The computer program measures motor and cognitive abilities through simple tasks such as finger tapping and word matching. But clinicians question whether a brief, self-administered computer test is accurate enough.

A member of HDDW's scientific advisory board, Charles Guttmann, who runs the Center for Neurological Imaging at Brigham and Women's Hospital in Boston, has suggested that all HDDW participants receive a baseline magnetic resonance imaging scan and subsequent MRIs over time to help see whether the compounds are having a measurable effect on brain structures involved in Huntington's. But although the Goodmans don't rule out using MRIs, they say that adding imaging to the mix would require additional funds and time they'd rather not spare before handing out their compounds. In another break from the norm, the Goodmans are also enrolling some patients who are already taking some of the compounds—even though that could confound efforts to tease apart their effects.

Veatch Goodman expects that many patients will end up taking all six compounds. A patient's progression will be tracked once a week by the computer, with the results sent to the Goodmans and the patient's physician. Nat Goodman says a year's data are needed to begin assessing the compounds' impact.

But Steven Hersch, a neurologist at Massachusetts General Hospital who is heavily involved in the Huntington's Study Group, which oversees clinical trials, questions whether any effects will be discernable. “It's just not clear to me that you would get usable information” about the compounds, he says. The Goodmans' desire to offer potential therapies to patients, he says, “outweighs considerations of what's necessarily good science.”

Other researchers, meanwhile, worry that they may be competing for Huntington's patients with the Goodmans, because the combination supplied by HDDW may render patients ineligible for some clinical trials—particularly those testing the same compounds. “That is a concern,” says Ira Shoulson, a neurologist at the University of Rochester, New York, and head of the Huntington's Study Group. Academic studies are planned or under way for cysteamine, coenzyme Q10, and creatine.

Bernard Ravina, the program director in clinical trials at the National Institute of Neurological Disorders and Stroke in Bethesda, Maryland, worries that HDDW could “lose a lot of valuable information” about its participants by not assessing them more methodically. Around the time that HDDW plans began circulating, his institute launched what could be seen as a more rigorous version of the project, the Systematic Evaluation of Treatments for Huntington's Disease. He invited nominations for potential compounds that could help patients; the list now tops 200. That number is being whittled down to fewer than five.

View this table:

Nat Goodman doesn't wholly disagree with his critics. He thinks the computer program will miss subtle drug effects but pick up dramatic ones, including an unlikely success in halting disease progression. But he admits that the HDDW approach—combining six unproven compounds, relying on computers to assess patients, and lacking a control group—is so unusual that “all of our data can be challenged. We're quite aware of that.”

But the Goodmans also believe that there has to be a better way to test drugs. They're not alone. “I reject the knee-jerk reaction that any change is bad,” says Heywood. Although he has some doubts about HDDW's promise, he also believes that “science is way too wedded to conservative approaches,” and he applauds the Goodmans for pushing the envelope.

Nancy Wexler, a neuropsychologist at Columbia University in New York City, has also thought long and hard about what could be done to enhance trials. A high-profile Huntington's researcher, Wexler has also been touched by the disease: Her mother died of it. “You always struggle with yourself: How do you make this more efficient? How do you go faster?” she says. The Huntington's Study Group, she says, follows the gold standard of placebo-controlled studies. “I think there are other ways also, more creative trials,” says Wexler. What those might be—and whether HDDW is one—she's doesn't yet know.

13. AMERICAN ASSOCIATION OF PHYSICAL ANTHROPOLOGISTS MEETING

# Tracking the Evolutionary History of a "Warrior" Gene

1. Ann Gibbons

TAMPA, FLORIDA—About 1200 researchers attended the 73rd Annual Meeting of the American Association of Physical Anthropologists here from 14 to 17 April to hear talks on primate genes, behavior, and fossils.

For males, a bit of aggression and risk-taking can earn rewards—just ask real-estate magnate Donald Trump. But inappropriate aggression can lead to violence, addiction, early death, and, the worst fate of all in evolutionary terms, no offspring.

Now, researchers have found signs of this balancing act in the genes of our primate cousins. At the meeting, a team of geneticists traced one genetic variant, an allele that predisposes men to aggressive, impulsive, and even violent behavior, to chimpanzees, gorillas, and other primates. They conclude that this and similar variants arose at least 25 million years ago in a monkey ancestor.

In order to be retained for so long, these variants must have conferred some selective advantage on the monkeys—and humans—who carried them, says author Tim Newman, a biological anthropologist at the National Institute on Alcohol Abuse and Alcoholism (NIAAA) in Rockville, Maryland. What we see today as dangerously inappropriate behavior could be “simply out of context,” says Newman. “Bold, aggressive males might have been quicker to catch prey or detect threats.” Others agree: “If this [allele] has been around that long, then it must be maintained by balancing selection,” says biological anthropologist Henry Harpending of the University of Utah in Salt Lake City.

The gene, found on the X chromosome, codes for an enzyme called monoamine oxidase A (MAOA), which breaks down several neurotransmitters in the brain, such as dopamine and serotonin, thus preventing excess neurotransmitters from interfering with communication among neurons. But the gene is polymorphic: A repeat sequence of 30 base pairs has been inserted from three to five times into the promoter region. Fewer repeats mean that less MAOA enzyme is produced and fewer neurotransmitters are removed.

The MAOA gene's effects have been linked to aggression. Lab mice that lack the enzyme are more aggressive, and one human family whose members do not produce the enzyme at all has been linked with violent behavior (Science, 18 June 1993, p. 1722). Men who carry the short allele, and so presumably produce a limited amount of enzyme, have been shown to be more likely to be aggressive, impulsive, and even violent if they were abused as children or drink alcohol. Men who had the short variant and were mistreated as boys were four times more likely than other men to have committed violent crimes such as rape, robbery, and assault, according to one study that tracked boys from birth in New Zealand (Science, 2 August 2002, p. 851). (Women also inherit the allele, but the effects are easier to study in men, who have only one X chromosome.)

These findings intrigued psychiatrist Klaus-Peter Lesch of the University of Würzburg in Germany, who works with the NIAAA group. His team first found, in macaques, a similar 18-base-pair repeat that also modulates MAOA enzyme activity. And macaques with less enzyme were more aggressive than other macaques when competing for food, says Lesch.

Newman then sampled all apes and many monkeys—almost 600 primates in all—and found the same 30-base-pair repeat seen in humans or the shorter 18-base-pair repeat, among other forms. He noted that apes and Old World (Asian and African) monkeys carried these alleles, whereas New World (South American) monkeys did not. That suggests that the allele arose after New World and Old World monkeys split, but before apes and Old World monkeys diverged about 25 million years ago.

During those 25 million years, aggressive and risk-taking behavior must have had reproductive payoffs for some males, says Newman. But the gene didn't sweep through populations, because if a male was too violent, he probably died before reproducing. Newman suggests that the MAOA gene may offer a rare example of so-called balancing selection, in which selection favors two or more forms of a gene and maintains all the forms in a population. “The human social environment required the development of all kinds of emotional and cognitive capabilities, and [it] demanded variation in impulsivity in humans,” agrees David Goldman, a member of the NIAAA team. “It's what I call the warrior vs. the worrier.” In other words, primate politics has long favored more than one route to success.

14. AMERICAN ASSOCIATION OF PHYSICAL ANTHROPOLOGISTS MEETING

# Chimpanzee Gang Warfare

1. Ann Gibbons

TAMPA, FLORIDA—About 1200 researchers attended the 73rd Annual Meeting of the American Association of Physical Anthropologists here from 14 to 17 April to hear talks on primate genes, behavior, and fossils.

Primatologists have long known that chimpanzees can be demonic: Bands of males routinely head to the borders of their territory to seek, and sometimes destroy, foreign chimpanzees. But what triggers these patrols, and why do males of the troop—who compete fiercely with one another most of the time—seem to cooperate while on patrol? The answer, it seems, may be a mob mentality.

In a study of a group of 150 chimpanzees at Ngogo in the Kibale National Park in Uganda, researchers found that chimpanzees went on patrol only after they had assembled enough members to have overwhelming force. Patrols require “safety in numbers” because attacking a foreign chimpanzee is dangerous, explains primatologist John Mitani of the University of Michigan, Ann Arbor, co-author of the study with primatologist David Watts of Yale University.

Once a patrol formed, its members exhibited frequent displays of male bonding. “Cooperation among males is rare among animals,” says Watts. “It is conspicuous that closely related chimpanzees and humans deindividualize to engage in this coalitional aggression against outsiders.”

Mitani and Watts, whose team has studied chimpanzees at Ngogo continuously for 9 years, observed patrols by shadowing males through the rainforest for 24 months during a 5-year period. When they analyzed the data, they found that chimpanzees went on patrol erratically—sometimes nine times a month, sometimes not at all. Previous observations had suggested that patrols increased when fruit was abundant or in response to intruder chimpanzees who might steal females, kill infants, or take feeding grounds. There had also been speculation that patrols declined when females were fertile. But the data showed that intruders, fruit, and females had little effect on the number of patrols.

Instead, patrols often started when a large group of 18 or so males gathered together. First the males began grooming each other, switching partners, and clustering together. Then, as if they had counted heads, they headed out in tight single file and were uncharacteristically quiet. More than once, the males passed right by a group of red colobus monkeys, which they hunt enthusiastically at other times.

Sometimes patrolling males stopped to quickly groom each other, occasionally hugging and rubbing their genitals together. In another talk, Watts and Mitani reported that males share meat with other males more than they do with females for sex, further underscoring the importance of male bonds. “I think all of this is their way of reinforcing their commitment to each other, as if they're saying ‘I'm with you,’” says Watts.

Once the males ventured into foreign terrain, if they came across a chimp alone, they would overwhelm it with sheer force. In one instance, so many chimpanzees piled on top of the victim that it was impossible to see it, says Mitani; in that case, as on four other occasions, the victim died, whereas most of the patrolling chimps walked away unharmed. “There really is a chimpanzee mob psychology,” says Watts.

Conversely, a group of 19 males beat a hasty retreat after they stalked one male but then heard other males nearby. “We saw them ignominiously turn and silently run home,” recalls Watts.

All this is “uncannily reminiscent” of reports of human males in traditional, small-scale societies who wait to assemble enough males before they raid neighboring tribes, says primatologist Bill McGrew of Miami University in Oxford, Ohio. “What strikes me is the same males a chimp competes with back home for a female are the guys who could save his life when he goes out in these patrols,” says McGrew. “It makes us realize what the functional significance is for male-male bonds.”

15. AMERICAN ASSOCIATION OF PHYSICAL ANTHROPOLOGISTS MEETING

# Seeing What an Extinct Monkey Saw

1. Ann Gibbons

TAMPA, FLORIDA—About 1200 researchers attended the 73rd Annual Meeting of the American Association of Physical Anthropologists here from 14 to 17 April to hear talks on primate genes, behavior, and fossils.

Paleontologists studying an extinct animal are often tempted to imagine what it saw millions of years ago. Now, with the clever use of imaging tools, some researchers actually are getting inside the heads of a few extinct primates, looking for clues to how they perceived their long-lost worlds.

At the meeting, Duke University paleontologist Richard Kay and colleagues reported using computerized tomography (CT) scans on the skull of a 20-million-year-old monkey long thought to have been nocturnal because it has large eye sockets—and concluded that it lacked night vision. The find supports the notion that although the ancestor of all primates was nocturnal—many of the earliest primate fossils also have large eye sockets—night vision was lost early in the lineage leading to monkeys, apes, and humans, says paleoanthropologist Robert Martin of the Field Museum in Chicago.

Paleontologists have suspected that the lineage of anthropoids—monkeys, apes, and humans—became diurnal early on and certainly before 34 million years ago, which is the date that fossils indicate for the divergence between New World and Old World monkeys. But two South American species throw a monkey wrench (so to speak) into that hypothesis: a living nocturnal owl monkey, Aotus, and a 20- million-year-old fossil monkey, Tremacebus harringtoni. Two examples of nocturnality raise the possibility that this trait may have persisted in New World monkeys—or else it evolved independently several times, something often considered unlikely. “If Tremacebus was nocturnal, some might even suggest that the earliest anthropoids had retained nocturnal habits from the ancestral primates,” says Martin.

In Aotus, certain details suggest that nocturnality may have re-evolved independently. As for Tremacebus, a probable fruit eater only a third of a meter long, paleontologists have thought it was nocturnal ever since it was discovered in Argentina in the 1930s because it too seems to have large eye sockets. But Kay and paleontologist Chris Kirk of the University of Texas, Austin, showed that when plotted relative to skull size, Tremacebus's eyes weren't that large; they fall into a zone where the socket sizes of nocturnal and diurnal primates overlap.

Then Kay came up with another clever test to separate nocturnal and diurnal species. Primates active at night rely heavily on smell to navigate, and they have large olfactory bulbs—the part of the brain responsible for smell—relative to skull size. “Olfaction might be the sense that makes sense for nocturnality,” says Kay. Although the olfactory bulb has long since decayed in Tremacebus, Kay proposed that the size of the bulb in monkeys corresponds with the width of a hole between their brows that houses the bulb, the olfactory fossa. CT scans of living monkeys showed this correlation.

When Kay examined the CT scan of Tremacebus, he found that its olfactory fossa was small, implying that its bulb also was small relative to skull size. With relatively small orbits and olfactory bulb, Kay concludes, Tremacebus was no night owl. If he's right, the large-looking orbits of this species might be just a remnant of an old late-night lifestyle.