# News this Week

Science  19 Dec 1997:
Vol. 278, Issue 5346, pp. 2038
1. BREAKTHROUGH OF THE YEAR

# CLONING: The Lamb That Roared

1. Elizabeth Pennisi

A LAMB CLONED FROM A SINGLE CELL OF AN ADULT SHEEP DEMONSTRATED THE POWER OF CLONING TECHNOLOGY, SURPRISING BOTH RESEARCHERS AND THE PUBLIC, AND IGNITING A FIERCE DEBATE ABOUT ETHICS

A year ago, few researchers would have guessed that science's most stunning achievement in 1997 would come from a barn. But in late February, a bleating, white-nosed lamb swept into the public eye: 7-month-old Dolly, the first animal cloned from an adult cell. She electrified both the research community and the general public, for although animals had been cloned before, creating a sheep from a single cell of a 6-year-old ewe was a stunning technological feat that many had thought impossible.

Cloning is of practical import, as it can be used to quickly create herds of identical animals that churn out medically useful proteins; the first such animals—a handful of transgenic sheep clones—are described on page 2130 of this issue. But the implications of cloning technology go much further, opening up new avenues of research in cancer, development, and even aging. Indeed, Dolly forces a reexamination of what it means to grow old, for although she is now 18 months old, her DNA, taken from the donor cell, may be almost 8 years old.

According to conventional wisdom, adult cells cannot give rise to new, mature organisms. So after Dolly's debut, researchers scrambled to understand how she was created. Scientific societies convened their own impromptu meetings to discuss both the scientific and ethical implications of the work, and companies specializing in transgenic animals saw their stock value jump overnight.

But despite Dolly's soft brown eyes, some feared that she was a wolf in sheep's clothing, come to steal humankind's individuality and autonomy. She sparked calls for a ban on human cloning in the United States, Switzerland, China, and other nations, and to some, she raised the sci-fi specter of cookie-cutter clones grown for spare parts. But whether welcomed or feared, cloning in 1997 forced scientists and the public alike to rethink their basic ideas about life, and to confront the implications of our growing ability to manipulate life's blueprint.

As is true for many breakthroughs, cloning represents the convergence of advances in several disciplines over several decades. Painstaking progress in sheep reproductive biology, genetic manipulation, and cell culture all paved the way for Dolly. But the critical technique is nuclear transfer, in which the intact nucleus of one cell is absorbed into an egg whose own nucleus has been removed.

Researchers seeking to unlock the secrets of embryonic development had been working to perfect this technique for 40 years, starting with experiments in frogs in 1952. They transferred the nuclei of embryonic or tadpole cells into frog eggs and succeeded in raising cloned tadpoles and even adult frogs. But the older the frog cell donating the DNA, the less likely was the resulting clone to develop normally. When donor cells from an adult were used, no frog clone ever developed beyond the tadpole stage. And in mice, the typical mammalian model organism, results were even more discouraging. At the time, researchers couldn't get viable young from anything but nuclei taken from very early embryos—the two- to four-cell stage. So most biologists came to accept that mature cells could not give rise to entire organisms, especially in mice. Only an egg cell possessed that mysterious power, called totipotency.

But those working with cows and sheep were not quite persuaded. A team of researchers at the Roslin Institute outside Edinburgh, Scotland, for example, suspected that previous failures were caused by donor DNA that was in a different stage of the cell cycle than the recipient egg cell. They used nuclear transfer to clone sheep from embryonic cells, and in 1996 announced the birth of two cloned lambs. Next, they cloned sheep from fetal fibroblast cells. And in partnership with a local biotechnology company, they attempted what everyone had said was impossible: to clone a sheep from adult cells.

To do this, the team used cultured udder cells, taken from a 6-year-old ewe, and then starved them, forcing most of their genes to enter an inactive phase that the researchers hoped would match the cell-cycle stage of the recipient eggs. Once the udder-cell nuclei were transferred into the eggs, still-unknown factors coaxed that “inactivated” 6-year-old DNA to go back in time, so to speak, and apparently become totipotent once more, directing the eggs to develop into lambs. Out of 277 such eggs, only one produced a healthy living animal: Dolly.

To a startled public, Dolly made the horrors of science fiction clones seem all too possible. If she could be cloned from an udder cell, people wondered, then why not a dictator from his nose, as was attempted in the movie Sleeper, or a spare self as a reservoir of replacement body parts? Such things are safely in the realm of fiction, of course, but many people, scientists included, became concerned that cloning people would dehumanize our species and spoke out against it.

Yet upon reflection it's clear that just as identical twins grow up to be individuals, clones would never be truly identical. Even Dolly is not an exact replica of the ewe used to clone her, because she did not develop in that ewe's uterus nor receive its genes in the cellular organelles called mitochondria.

For now, Dolly stands alone. No one, not even the Roslin team, has made a second animal from an adult cell. Of course, most biomedical researchers work with mice—and mouse nuclear transfer results are still dismal. So attention is focused on the handful of labs worldwide working on cloning in livestock. Most are starting with fetal cells, whose DNA can more easily be made totipotent. So far, several firms say they too have cloned either sheep or cows from fetal cells, and one group has cloned monkeys from embryonic cells.

Nuclear transfer experiments are under way in other species too, ranging from zebrafish to rabbits. Among basic researchers, the Scottish group's success has inspired new experiments looking at how DNA changes as a cell matures. Clarifying the nature of totipotency may spark insight into what makes cells and organisms age, and how cell growth can go awry, as in cancer. Researchers are watching Dolly closely, for although so far she seems the 18-month-old she's supposed to be, her DNA may make her age prematurely.

Cloning experts point out that the true identity of Dolly's progenitor cell is not known for sure—it's possible that it was a stem cell, known to be able to develop into several kinds of tissues. Even if that's true, the ability to restore totipotency to easily harvested adult cells would offer a potentially simple method to replace lost or damaged cells.

Meanwhile, the Roslin team has taken the next step toward making cloning economically useful by cloning sheep carrying foreign genes. Three sheep carry a marker gene, and two have both the marker gene and the gene for the human factor IX protein, which some hemophiliacs take to aid blood clotting. These sheep were cloned from transgenic fetal fibroblast cells, not adult cells, so they are most remarkable not as clones, but because they developed successfully despite having undergone genetic manipulation.

On the drawing board are flocks of sheep that make factor IX and other useful proteins in their milk. Other scientists are developing nuclear transfer techniques to create other types of genetically tailored livestock, opening the door to better animal models of genetic diseases, animals as organ donors, and possibly leaner, faster growing livestock.

Indeed, as with all breakthroughs, it's not possible yet to foretell exactly where cloning will lead. Although initial reactions were universally against all human cloning, there have been whispers that such cloning may one day have a place in giving infertile couples genetic offspring. Whatever direction the research takes, however, the public is likely to demand a say in how cloning is applied. Biologists, ethicists, and others will be wrestling with the implications of this birth in a barn for years to come.

Research:

1. I. Wilmut et al., “Viable offspring derived from fetal and adult mammalian cells,” Nature 385, 810–813 (27 February 1997).

2. K.H.S. Campbell et al., “Sheep cloned by nuclear transfer from a cultured cell line,” Nature 380, 64–66 (7 March 1996).

3. E. Pennisi and N. Williams, “Will Dolly send in the clones?” Science 275 (5305), 1415 (7 March 1997).

Ethics:

1. H. T. Shapiro, “Policy Forum: Ethical and policy issues of human cloning,” Science 277 (5323), 195 (11 July 1997).

2. E. Marshall, “Mammalian cloning debate heats up,” Science 275 (5307), 1733 (21 March 1997).

3. E. Marshall, “Clinton urges outlawing human cloning,” Science 276 (5319), 1640 (13 June 1997).

2. BREAKTHROUGH OF THE YEAR

# The Runners-Up

## The News and Editorial Staffs

When it comes to Mars exploration, getting there has always been half the battle. From the start of the space age to the beginning of 1997, 19 missions set out—and more than half failed. No mission had touched down on martian soil since 1976. So on the 4th of July this year, when the Mars Pathfinder lander sent back images of the first robot to roam the surface of another planet, jubilation was the order of the day. But Pathfinder was more than just a stunning technological achievement: It returned a bounty of scientific information from an intriguing part of the Red Planet and did so on the cheap. As the first of NASA's “faster, cheaper, better” Discovery missions, Pathfinder broke through a history of martian jinxes and high costs. The landing captivated audiences worldwide and vindicated NASA's gamble on a new, smaller scale approach to solar system exploration.

Pathfinder's technological victory was all the more impressive given its Rube Goldberg approach to landing on Mars. Previous landers have relied on an expensive rocket to first get into orbit around their targets, but Pathfinder blazed straight into the martian atmosphere behind a simple heat shield, popped open a parachute while still at supersonic speeds, and then began searching for the surface using onboard radar. At literally the last second, the craft fired its three small retrorockets and cut the parachute loose from the lander. Encased in airbags, the lander plummeted the last 30 meters in a free fall, finally bouncing to a stop—right side up.

When the rover started returning data, color images of the martian landscape—released in real time on CNN and in almost real time on the World Wide Web—entranced the public. To geologists, the images also revealed traces of the great flood that a billion years earlier had sculpted a distant hill, stacked nearby boulders, and left meter-scale ripples in the ground. Compositional measurements made by Sojourner on rocks such as Barnacle Bill and Yogi revealed that they contain a surprising amount of silica, suggesting that geological forces had reprocessed the rocks sometime in martian history. And Sojourner's discovery of rounded pebbles that may have been water-worn in flowing streams added evidence that Mars's surface was warmer and wetter in its earliest days—when life might have gotten a start.

Mission engineers at the Jet Propulsion Laboratory in Pasadena, California, managed all this with less than $270 million and within a 3-year development schedule. Such economy runs counter to the old bigger-is-better culture of spacecraft design, which produced, for example, the$3.3 billion Cassini mission to Saturn. And Pathfinder's achievements are the first in what NASA hopes will be a series of blockbusters from the Discovery program. Next up, on 5 January, Lunar Prospector lifts off for its mapping orbit around the moon.

## Synchrotrons Shine New Light

Beams of x-rays and ultraviolet light have long been used to reveal the atomic structure of materials. But by all accounts 1997 was a banner year for a new generation of stadium-sized machines known as synchrotrons, which produce the brightest beams yet and can illuminate the structural secrets of matter—both living and inert—down to individual atoms. This year saw the commissioning of SPring-8, the world's most powerful synchrotron, in Nishi-Harima, Japan, and marked the first full year of operation of the second most powerful machine, the Advanced Photon Source in Argonne, Illinois. And synchrotrons around the globe yielded some striking breakthroughs in the structure of materials.

Among the highlights: An international team used the European Synchrotron Radiation Facility (ESRF) in Grenoble, France, to produce an atomic-scale map of the nucleosome core particle, thus gaining new insights into how this DNA-protein complex manages to coil meters of DNA inside each cell. A Swiss and French team used the ESRF beam to solve the structure of bacteriorhodopsin, a membrane protein whose small crystals had defeated previous attempts to determine its structure. And a group at Oxford University solved the largest x-ray crystal structure to date, that of the bluetongue virus, made up of more than 1000 separate proteins.

Despite such smashing results, budget woes threaten a new ultraviolet synchrotron, the Advanced Light Source, in the United States. But worldwide, synchrotrons show no signs of slowing—another 26 are in the works.

## Keeping Time

As the days ticked by in 1997, those studying organisms' internal clocks marked time with periodic bursts of discovery. Researchers isolated several new genes that help keep daily rhythms, including the first two mammalian clock genes. And one fruit fly clock gene was found to be active throughout the fly, suggesting that many cells, not just those of the brain, can keep time.

Before this year, only three clock genes had been identified, two in the fruit fly (per and tim) and one in bread mold, called frequency (frq); these genes code for proteins whose concentrations rise and fall on a cycle set to 24 hours by sunlight. Then, last May, researchers reported two new genes in bread mold, white collar-1 and -2, which turn on the transcription of frq, thus participating in a feedback mechanism that keeps the clock ticking. Later that month, another team isolated the first timekeeping gene from a mammal—a mouse gene called Clock—which also appears to regulate circadian rhythms.

In September, two teams independently discovered a gene that resembles per in mice and humans. Having similar genes in such divergent species as humans and flies implies that the genes that make up the clock's gears and springs may have been conserved since the earliest days of biological time. Finally, in the closing days of 1997, researchers found that per is active not only in the brains of flies, but in many other tissues as well, suggesting that many independent clocks are ticking away in the body—and that the brain is only one of many timekeepers.

## Violence at a Distance

They are the most violent events in the universe, and for 30 years they ranked high on the cosmic mystery top 10. Gamma ray bursts, sudden explosions of high-energy radiation occurring almost daily at random positions in the sky, were first detected in the early 1960s. But gamma ray detectors couldn't accurately determine their location or distance from Earth. Some astronomers thought the bursts were far-off events, others that they occur in our own galaxy.

Then on 28 February, the Italian-Dutch satellite BeppoSAX simultaneously detected a burst in both gamma ray and x-ray wavelengths, using two separate detectors. An x-ray telescope aboard the satellite was also able to catch the afterglow of the burst and pinpoint its position. Alerted via the Internet, Dutch astronomers at an observatory in the Canary Islands found a dimming optical light source at the burst position, coinciding with what appeared to be a distant galaxy.

Ten weeks later, on 8 May 1997, it happened again—a second BeppoSAX detection was linked with an optical source. This time, American astronomers using a Hawaii-based telescope were able to study the light of the burst in great detail and to peg its distance at several billion light-years. It seems that gamma ray bursts occur in the far reaches of the universe, making them by far the most energetic events in the cosmos, exceeded only by the big bang itself. But what distant cataclysms cause these bursts? They might be the collision of two dense neutron stars, but their true nature remains a mystery to be solved another year.

## A Glimpse of Neandertal DNA

Ever since the first skeleton of a Neandertal was discovered in Germany's Neander Valley in 1856, anthropologists have wondered whether this burly human was ancestral to living humans or an evolutionary dead end. Many more Neandertals have turned up since, but the bones alone haven't settled the mystery.

Then in July a team in Munich announced that it had recovered and analyzed a snippet of DNA from the arm bone of that original Neandertal. The researchers had pieced together a 379-base pair sequence from the mitochondrial DNA (mtDNA) in the cell's energy-producing organelles. This Neandertal mtDNA spelled out a sequence very different from that in living humans, supporting the view that Neandertals were not our ancestors, but a separate species that went extinct.

Only one small part of the genome was reconstructed, but at 30,000 to 100,000 years old this is the oldest DNA extracted from a human. And it is a triumph for the besieged field of ancient DNA. Earlier, spectacular claims of analyzing DNA from insects in amber have not been replicated, and many had almost given up on the idea of extracting useful DNA from ancient fossils. But this work was replicated in a U.S. lab, convincing even skeptics that it is the real thing.

## Nanotubes on a Roll

Since their discovery in 1991, nanometer-sized tubes of carbon have been seen as the right stuff for everything from future electronic devices to ultrastrong materials. In 1996, researchers developed a laser-based method to produce high yields of single-walled nanotubes (SWNTs), and in 1997 they brought the tubes closer to their potential by testing, tweaking, and filling them.

Cousins of the spherical buckminsterfullerene (C60, Molecule of the Year in 1991), nanotubes are sheets of graphite—carbon atoms arrayed in adjoining hexagons—that are rolled up and capped at the ends. Those made of only a single wall of carbon are prized for their regular structures and predictable behavior. For example, theorists predicted early on that depending on their architecture, SWNTs should be semiconductors or metals, key building blocks for electronic devices. Scientists confirmed both predictions in 1997. Individual nanotubes were found to be excellent conductors, a property that could be enhanced by doping their outer surfaces. And a slightly different bonding arrangement between a single pair of polygons was shown to turn a nanotube into a simple semiconducting electronic device. Other researchers demonstrated their ability to manipulate the tubes by stuffing them with gas or with gallium nitride rods.

Also this year, a French and U.S. team came up with a cheaper way to produce SWNTs using a simple electric arc discharge, a feat that's likely to make these cylinders easier to come by and therefore study. But no known method can produce the tons of SWNTs needed for a commodity material, so the push to make larger batches of tiny tubes will continue.

## The Other Ocean

This year planetary scientists gathered solid evidence that our ocean is not alone. The Galileo spacecraft orbiting Jupiter returned images of the surface of the jovian moon Europa, revealing what looks like an icy crust floating on a watery ocean. Absolute proof of a deep sea might be beyond Galileo's abilities, but the newly credible case for liquid water—the crucial ingredient for life—raises the prospect of alien stirrings beneath Europa's icy surface.

The signs of a europan ocean were varied. Faults, rifts, and jumbled crustal blocks suggest that the surface layer of ice was thin when last disturbed; in one spot, iceberglike blocks seem to have floated in a now-frozen sea. And some large impact craters appear to have punched through thin ice, leaving flat blemishes rather than rimmed craters.

These and other clues suggest that when the europan surface was last disrupted, the ice was only 10 to 20 kilometers thick, leaving 100 kilometers or more below for liquid water. And large areas nearly unblemished by the drizzle of small impactors suggest that all this disruption was recent or even ongoing. An intensive campaign by Galileo during the next 14 months may turn up more evidence, but look for proposals to send a spacecraft to orbit this moon to probe for the final answer.

## Genomes Galore

The growing tower of microbial genetic data was buttressed by two more cornerstones this year, and geneticists also pushed closer to what once seemed a pie-in-the-sky goal—analyzing whole genomes.

Researchers sequenced the entire genetic codes of two well-studied microbes, the common gut microbe Escherichia coli and the soil bacterium Bacillus subtilis. These bacteria—each with a genome more than 4 million bases long—have been laboratory workhorses for generations of biologists. Now, scientists can link decades of physiological and biochemical work to the genes involved.

This was also the year when whole-genome sequencing took off. Once a maverick approach, this shotgun method has become commonplace in organisms with fewer than 2 million bases. This year, it yielded genomes of three archaea (primitive microbes often adapted to extreme environments) and several pathogens, such as Helicobacter pylori, infamous as the cause of ulcers, and Borrelia burgdorferi, the spirochete behind Lyme disease. More than 40 other microbial genome efforts are under way.

This output has been both empowering and humbling. With whole genomes to compare, researchers can classify genes into functional or ancestral families. From these studies, biochemists are tracking down new proteins in key metabolic pathways, providing insight into such questions as how pathogens gain access to their hosts. Yet even in familiar E. coli, about a third of the putative new genes are of unknown function, which stakes out a new challenge for the years ahead.

## Neurons in the News

In 1997, researchers homed in on clues to the workings of the central nervous system, clues that may one day lead to new treatments for ailments ranging from Parkinson's disease to spinal cord injuries.

This year, scientists fingered the first genetic cause of Parkinson's, tying a mutation in a gene called α-synuclein to a heritable form of the disease in a large Italian family. The work not only put to rest a long-standing debate over whether genes play a role in Parkinson's, but also pointed to a possible mechanism for the disease, involving abnormal processing of the protein.

Other research shed light on the biology of the dopamine-producing brain cells that die in Parkinson's. Work from a Swedish lab using mice showed that a receptor protein called Nurr1 is needed for both proper development of dopamine-producing cells and the synthesis of a healthy amount of dopamine.

There were advances in Alzheimer's disease too: Researchers identified a potential new player, novel brain lesions called “AMY plaques,” that may contribute to the disease. And this year also raised hopes that injured spinal cords may one day be rewired. Groups from London and San Diego linked the sprouting of nerve fibers in the severed cords of adult rats with some return of function—a feat long thought impossible.

## References:

1. 1.
2. 2.
3. 1.
4. 2.
5. 3.
6. 1.
7. 2.
8. 3.
9. 4.
10. 1.
11. 2.
12. 3.
13. 4.
14. 5.
15. 6.
16. 7.
17. 8.
18. 9.
19. 1.
20. 2.
21. 3.
22. 1.
23. 2.
24. 3.
25. 4.
26. 1.
27. 1.
28. 2.
29. 3.
30. 1.
31. 2.
32. 3.
3. SCIENCE AND SOCIETY

# Bumper Crop for Pop Science

Most scientists think the general public pays scant attention to research, but in 1997 several science stories became hits in popular culture, as the mass media—particularly television and the Internet—discovered that the process of discovery itself can be a rich source of entertainment as well as information.

Space science was perhaps the most popular story, as a string of marvels, mishaps, and milestones drew eyes upward in numbers probably not matched since the days of the Apollo moon project. In March and April, there was Hale-Bopp—for once, a comet whose brightness exceeded its ballyhoo. The comet turned even urbanites into backyard astronomers, and nearly doubled previous records for visibility and endurance. Lasting even longer, however, were the travails of Russia's dilapidated Mir space station. This space-based soap opera repeatedly stopped the hearts of Mir watchers, as a rotating cast of astronauts battled fires, computer breakdowns, and other calamities.

Then on the 4th of July, the Mars Pathfinder lander bounced down on the Red Planet, announced on CNN with much fanfare and 24-hour live coverage from NASA's mission control room. Pathfinder's intrepid Sojourner rover wasn't equipped to search for signs of life, but Sojourner and its lively band of handlers at mission control charmed TV and World Wide Web audiences anyway. The “hits” pummeling Pathfinder Web sites peaked at millions per hour, with the grand total now nearing 1 billion. As with many hit TV shows, there was a toy tie-in: Mattel's Hot Wheels version of the rover went flying, not crawling, off store shelves.

Space exploration may have mimicked high-adventure science fiction, but reproductive biologists' exploits in a petri dish in Scotland evoked, for many, the genre's cautionary side. Dolly—the ewe cloned from an adult mammary-gland cell—sparked hasty calls from politicians for a moratorium on human cloning, forced ordinary citizens to reconsider the meaning of individuality, and became a staple of late-night TV comedians. Like headless tadpoles in Britain, genetically engineered soybeans in Germany, and the new film Alien Resurrection, the cloning story tapped public fears of genetic technologies. Scientists may wish that ordinary citizens knew more about science than they do—but they can hardly deny the strength of its grip on the public imagination.

4. ENVIRONMENT

# Politicians Sweat Over Global Warming

Scientific uncertainty is nothing new to policy-makers. But this year it occupied center stage as the global warming issue, long smoldering on the scientific sidelines, spread to the political arena. Earlier this month, politicians, scientists, and lobbyists gathered in Kyoto, Japan, to face up to a burning question: If human emissions of carbon dioxide and other greenhouse gases will warm the world, what should be done about it? While the answer depends in part on science, 10 days of arduous debate at Kyoto—as delegates hammered out a historic accord—made it plain that the science is only the start.

Well before Kyoto, scientists had already aired their views in a report by the Intergovernmental Panel on Climate Change (IPCC)—the most comprehensive international assessment ever on an environmental question. But while the effort was impressive, it was hardly conclusive. The report said warming due to a doubling of greenhouse gases, expected sometime late in the next century, could range from 1.5 to 4.5 degrees Celsius—from moderate to outright catastrophic. And scientists still can't be sure just what the effects of climate change will be on various regions of the world.

Given that uncertainty, most policy-makers agreed that emissions need to be restricted—but how much, and who will bear the cost? Kyoto provided a few answers, including emission reductions ranging from 6% to 8% below 1990 levels by Japan, the United States, and Europe, as well as the inclusion of all six major greenhouse gases in the tally. Of special interest to scientists, a clean development fund would channel new energy-saving technologies to developing countries. Such a fund—intended as the first step toward participation by poorer nations—may thrust science back into the center of the debate. But it won't provide any quick fixes for negotiators when they reconvene in November 1998 in Buenos Aires to tackle several unresolved issues, including whether and how developing nations will participate in the treaty.

## References:

5. RESEARCH TRENDS

# New Research Horizons

What fields will rise to prominence in 1998? Science surveyed the research world and found six likely prospects.

Forecasting future shocks. Climate researchers made a sharp call on 1997's massive El Niño, but the burgeoning field of long-range climate prediction has its reputation on the line once more: New predictions for this winter have been posted, from warmth in Minnesota to drought in southern Africa. Such seasonal forecasts pale before the new frontier of decadal predictions, based on understanding the slow mood swings of the oceans.

The expanding universe. Views of a handful of distant stellar explosions, taken this year by the Hubble Space Telescope, suggest that the ballooning of the universe has slowed so little over time that it may expand forever, rather than going to blazes in a final collapse. As more observations come in, expect these data to have a chilling effect on mainstream theories of the universe's birth.

Personalized prescriptions. The surge of interest in genotyping technologies—and in the firms developing them—is reviving pharmacogenetics, a decades-old vision of tailoring drugs to a patient's genetic makeup. Thanks to DNA chips and arrays, rapid DNA analyses may soon be able to reveal genetic variations affecting drug metabolism and side effects. These technologies are already changing how clinical trials are conducted and may eventually revolutionize how medications are prescribed.

Ribosomal inspection. At long last, researchers are beginning to get a detailed look at the inner machinery of the ribosome, the cell's protein factory. Advances in the techniques of structural biology are revealing the intricate dynamics of this large complex of RNA and proteins. Higher resolution images of the ribosome are likely to yield the exact nature of some of the many steps involved in protein production.

Diversity debate. Experiments with artificial ecosystems have bolstered the traditional view that biodiversity improves ecosystem functions. But recent work on natural ecosystems suggests that biodiversity may not be crucial to such functions as how nutrients cycle, while research in microbial communities suggests that high biodiversity makes ecosystems more predictable. As humans continue to wipe out species, expect more research on the science of why biodiversity matters.

Designer crops. The United States leads in transgenic crops, but Europe is fast catching up, with more than 100 trials—ranging from maize to strawberries—approved in Britain. A few transgenic crops have earned European Union approval and will soon be on the market. But will reluctant European consumers bite? Expect a battle for acceptance, as lobbyists press for new labeling rules.

1. 2.
2. 3.

1. 4.
2. 5.

1. 6.
2. 7.
3. 8.

1. 9.

## Transgenic crops:

1. 10.
6. RESEARCH TRENDS

# Scorecard '96

Last year, Science's editors gazed into the future to predict the hot fields of the next 12 months. Here's how our favorites fared this year, showing whether our crystal ball was cloudy or clear.

The cure for cancer. Promising new results boosted bold new therapies, such as blocking the growth of blood vessels that feed tumors, or designing a virus to kill cancer cells. But clinical trials are proceeding slowly, and it may be years before such tactics pan out.

Advanced Photon Source. Several beamlines saw first light this year, while other groups used this x-ray synchrotron to probe everything from protein structure to the behavior of catalysts in water (see second runner-up, p. 2040).

Computer security. Although worried Web users kept a wary eye on information security, there were no major new breaches of the encryption codes used to safeguard business data.

Synthetic carbohydrates. New clinical trials were launched to see if these sugar-based molecules may work as a cancer vaccine, coaxing the immune system into attacking the natural carbos that decorate tumor cells.

Quantum error correction. New schemes for correcting errors in unimaginably fast quantum computers leaped forward, as theorists moved beyond correcting mistakes in memory to protecting logic operations. And the first experimental realizations of quantum error correction are in the publication pipeline.

Supersymmetric particles. A fire at CERN slowed the search for the elusive signs of supersymmetry. Optimists still see hints of the particles in the modest amount of data gathered and predict that their signatures will turn up next year at CERN.

7. HUMAN GENOME PROJECT

# 'Playing Chicken' Over Gene Markers

1. Eliot Marshall

IN A RACE WITH INDUSTRY, HUMAN GENOME PROJECT LEADERS ARE SCRAMBLING TO COLLECT A DIVERSE SET OF SINGLE-BASE HUMAN DNA VARIATIONS BEFORE THEY ARE PATENTED

The U.S. government is getting set to launch a big addition to its human genome project this winter, and it is doing so with “some urgency,” says geneticist Francis Collins, the chief architect of the new initiative. Details of the venture—which involves sequencing snippets of DNA from hundreds of individuals of different racial backgrounds and putting them in a public repository—were discussed at a meeting at the National Institutes of Health (NIH) last week, and a strategic plan will be put together over Christmas. NIH hopes to issue a funding announcement in December or January, winners will be selected by summer, and investigators are supposed to begin putting out data in less than a year. “Some urgency” seems to be an understatement.

Unofficially, it's known as the SNPs project (for single nucleotide polymorphisms, pronounced “snips”). The basic strategy, decided upon at last week's meeting, is to collect at least 100,000 single-base variations in human DNA donated by 100 to 500 people in four major population categories: African, Asian, European, and Native American. These variable sites will serve as landmarks for creating a new, very fine-grained map of the genome, says Collins, director of the National Human Genome Research Institute (NHGRI), which will help investigators track down elusive genes that cannot be found using family studies. The collection would also provide some data on variable forms of known genes, and it may breathe life into the beleaguered human diversity project, which was supposed to survey DNA variation in populations around the world, but has become bogged down in politics (Science, 24 October, p. 568). Most important, according to Collins and other advocates, the project should encourage researchers to adopt a common format and quality-assurance methods when collecting and reporting DNA variations, which should make the data more useful to other scientists.

After organizers identify a reliable way of finding new SNPs, they hope to move quickly into production mode, collecting thousands of SNPs before they are locked up in a hodgepodge of other DNA collection schemes—many of them commercial (see sidebar). The effort is likely to cost $20 million to$30 million over the next 3 years—and probably tens of millions more after that. Collins has already received the green light from NHGRI's advisory council, and he says that 18 NIH institute directors have promised to share the costs. NIH is now providing seed money with funds from its current budget.

Funding may not be the biggest problem, however. Concerns over how to protect confidentiality and ensure that DNA donors have given informed consent for the use of their genetic data could present a far more difficult set of obstacles. Indeed, there were some tense moments at last week's meeting—a gathering of an ad hoc working group of scientists, bioethicists, and government officials who are advising Collins on the project—over a proposal to use some existing DNA collections to kick-start the venture. Some felt that it would be improper to use these collections without going back to the original donors for their consent, yet without a ready-made source of DNA, the new project will be slow to get started.

## Why hurry?

Collins first began promoting this project about 3 months ago (Science, 19 September, p. 1752), and it is already on the verge of getting under way. The main reason it is moving at warp speed, say scientist-advisers on the project, is that big academic labs and companies have jumped into genomic data collection in the past year, and U.S. genome project leaders want to make sure that they don't patent most SNPs before the smaller labs have a chance to use them as low-cost genomic mapping tools. Or, as one geneticist said: “It's a game of chicken between Francis and the companies.”

Aravinda Chakravarti, a geneticist at Case Western Reserve University in Cleveland and the most prominent academic champion of the SNP project, co-authored with Collins a Policy Forum in Science last month where they warned that if this effort doesn't get public support, much of the SNP data will be socked away in “private collections” (28 November, p. 1580). The information might then be subject to “a tangled web of restrictive intellectual property attachments … inhibiting many researchers from using these powerful tools.”

Besides, says Chakravarti, the “cottage-industry style” of gathering such data is coming to an end. Today, much information about genetic variation is acquired haphazardly, as researchers collect information about polymorphisms in families or explore clinical data about a particular gene. For example, the intense focus on two genes linked to breast cancer (BRCA1 and BRCA2) has turned up hundreds of alleles, scores of which have been patented. But new technology may make it possible to sift thousands of small variations out of a collection of DNA in minutes, simply by using a mutation-sensing “chip” to scan a person's genome for anomalies. Such devices could enable companies to scoop up massive amounts of data on DNA variation. “When you consider the magnitude of what's coming down the pike,” says Chakravarti, “we will lose information if we don't combine it all in one place.”

Companies that are generating their own SNP collections don't buy the argument for a big public investment in a database, however. “There's a lot of me-too-ism in this field right now,” says Gualberto Ruaño, a Yale University geneticist and founder of Genaissance Therapeutics Inc. in New Haven, Connecticut. His company is attempting to develop what Ruaño calls “personalized medicine” by identifying variant forms of important human genes—such as those that code for the estrogen receptor—patenting them, and designing drugs that conform to the particular molecular structures associated with the most common types of genes.

Ruaño says it might be wiser for the public agency to “keep its focus” on completing the full sequence of the human genome. Later, he said, it can add variation data. Fred Ledley, president of another small company that's exploiting human DNA variation for drug development—Variagenics Inc. of Cambridge, Massachusetts—says, “We welcome efforts by the NIH to systematize sequence variation in a public database.” But, like Ruaño and other industry people, he says that to secure private investment he must continue to patent human genetic variations.

That argument bothers Kenneth Weiss, an anthropologist at Pennsylvania State University in University Park, an adviser to Collins on the SNP project and a strong supporter of it. “For the good of science, [this information] should be made available as widely as possible so as many scientists as possible can think about it,” he says. Weiss personally opposes patenting any human genetic sequences.

Although many academic scientists agree with that sentiment, some question the need to move ahead so rapidly with the SNP project. “This is incredibly hasty,” says one researcher who asked not to be identified. “Why can't they slow down and improve the science?” he asks, arguing that polymorphism data would be far more valuable if it were linked to detailed biomedical information about the donors. Such information (and permission to use it) would take more time to obtain.

But Collins and his NHGRI staff say that they want quick access to the best DNA samples currently available, because this may prove to be the biggest hurdle to getting started. The grant money for analyzing the DNA won't begin flowing until October 1998. By then, NHGRI needs to have access to a large source of DNA from individuals of diverse backgrounds, ensure that donors have given proper consent, and place the DNA into immortalized cell lines. These issues occupied most of the working group's meeting last week.

## DNA on tap

The strategy session, chaired by Chakravarti, came up with a scheme for sampling four major population groups, leaving it to a technical subcommittee to devise numerical targets. For example, evolutionary geneticists have shown that there is far greater genetic variation within African populations than in non-African populations, and also that certain alleles considered rare in European DNA samples are common in African samples. This reflects the greater age and diversity of the African gene pool. Yet publicly available DNA collections contain little African material; Native American and Asian contributions are similarly scant. As a result, say NHGRI staffers, it will be essential to collect DNA from a racially structured set of donors. Once the DNA has been sampled, however, all personal and racial data will have to be removed to protect privacy—diminishing the scientific value of the project, but bolstering its ethical foundation.

Where those samples would come from—and how to ensure that donors have given appropriate consent and that their privacy is safeguarded—prompted the most intense debates. NHGRI staffers had set their hopes on getting a set of ready-made cell lines containing DNA from a broad sample of the U.S. population, created by the National Center for Health Statistics (NCHS). From 1989 to 1994, that agency's National Health and Nutrition Examination Survey (NHANES) collected blood from more than 17,000 representative individuals around the country to obtain a snapshot of the health of the U.S. population. Karen Steinberg of the Centers for Disease Control and Prevention later converted more than 8000 of these samples into immortalized cell lines. But Chakravarti warned that “it is not a done deal” that genome researchers would be allowed use this material. The reason: Because DNA studies had not been foreseen, NHANES staff did not ask the donors for permission to put their DNA in a database.

Managers of the NHANES data asked their human subject research panel for guidance on this issue more than a year ago. In September, the panel said it would be acceptable to run some health studies on the cell lines, but only if the samples were made completely anonymous by stripping them of all identifiers (Science, 21 November, p. 1389). Edward Sondik, director of the NCHS, has interpreted this ruling to mean that DNA data from these samples cannot be put in a database—even with anonymous identifiers—without consent, because a third party who knows the name and genes of an individual might still be able to ferret out unique genetic information.

But the prospect of asking for fresh consent from donors concerned NHANES officials. As Diane Wagener of NCHS explained, they worry that if people receive a letter informing them that their DNA has already been immortalized in cell lines and requesting approval for hard-to-understand genetic studies, they might say, “I don't want to have anything to do with [NHANES].”

As the cell lines seemed to become less accessible by the hour, an annoyed Collins declared that, after months of discussion, “I am very troubled to learn that there still doesn't seem to be a clear answer” about whether they can be used. After a coffee break, Sondik announced that 600 DNA samples that are not part of the primary NHANES set will be made available for the SNP project, and possibly a small fraction of the primary set, as well. Consent will have to be obtained from the donors, only 75% of whom are expected to be reachable through old addresses. But even with a strong response, the NHANES contribution will not be enough. For example, it is short on Asian and Native American DNA. NHGRI will therefore have to find other donors, perhaps among patients in ongoing NIH projects.

The NHANES samples will, at least, allow the SNP project to get started. Now NHGRI staffers must lay out the technical parameters, the sequencing goals, deadlines, and cost limitations. They hope to complete all of this by January. Then, if the whole scheme doesn't run into a wall, the genome community will witness a brand-new competition in gene discovery.

8. HUMAN GENOME PROJECT

# The Hunting of the SNP

1. Eliot Marshall

The government's plan to create a genomewide catalog of human DNA variation (see main text) represents a twist on the way big research projects like this usually get started. Often, government-sponsored research galvanizes industry, but this time the impetus is going the other way. The academics are worried that private companies wielding new technologies for scanning the genome will snap up the SNPs (single nucleotide polymorphisms) and patent them.

• ECOLOGY

# Counting Creatures of the Serengeti, Great and Small

1. Virginia Morell

Serengeti National Park, TanzaniaIt's just after dawn, and in the dense woodlands along the Grumeti River the birds are beginning their morning chorus. This is prime data-collecting time for ecologists A.R.E. Sinclair and Simon Mduma of the University of British Columbia (UBC) in Vancouver, and they bend their heads and cup their ears to better hear the bird calls. “That's two gonoleks singing their duet; the cooing is a green pigeon; and that harsh cry is a gray-headed kingfisher,” says Sinclair. Mduma and Sinclair's wife, Anne, jot down the species on their data sheets. Crouching low, Sinclair listens again, then creeps forward, following a hippo's trail through the dark stands of fig and African olive trees. Silently, he points to an oval circle of mashed grass and ferns where the hippo had rested, then raises his eyebrows in mock alarm at the bus-length crocodile stretched on a sandy beach across the river. But it's the birds Sinclair is after, and a few moments later when he first hears and then sees an olive-green bulbul, he smiles broadly. “They are rare and hard to see,” he whispers, “and so are exactly the kind of bird we're trying to find by creeping around like this. It's the kind of species most likely to vanish if we lose these forests.”

Counting olive-green bulbuls in a forest may seem odd in the Serengeti, an ecosystem best known for vast herds of antelopes and large carnivores racing across wide-open grasslands. Indeed, most scientific research in these famed East African plains has focused on grassland plants and the “glamour” animals—lions, rhinos, buffaloes, and cheetahs. Sinclair himself, who directs UBC's Centre for Biodiversity Research, has done several such studies since he began working in the Serengeti in 1965.

But even the man whom other researchers sometimes refer to as “Mr. Serengeti” says that his past work has not been comprehensive enough, and that it's time to pay attention to the other creatures of the park—the smaller mammals, birds, and insects—and their diverse habitats, such as this little-known riverine forest. With a small grant from the Natural Sciences and Engineering Research Council of Canada, Sinclair has just launched the first systematic biodiversity study of the park. “We need that baseline data: the variety of species, their numbers, habitats, and how all these are interrelated,” he says. “Without it, we have no way of knowing what generates the Serengeti's diversity, how the park is changing, or how people are affecting the environment outside the park.” For example, riverine forests such as this one along the Grumeti are shrinking, but no one knows why—whether it's part of a natural cycle, or due to human influence.

Armed with new—if meager—funds from government research agencies, scientists worldwide are setting out on similar surveying missions in key areas. “There's still no part of the planet where we know every species,” says Joel Martin, program director for the U.S. National Science Foundation's (NSF's) biotic surveys and inventories. “And we're losing species and habitat at an unprecedented rate, which makes these kinds of surveys imperative.” But there are no set rules about the best way to tackle this enormous task. Some surveys, such as those funded by Conservation International, are quick-hit projects, with a team of scientists sampling as much as they can in a region in a 4-week period. Others target all the plants of a country, or a specific group of animals in a region and their parasites. A few projects try to identify every organism in a given area, although such all-taxa surveys are costly and difficult to coordinate. (See Science, 9 May, p. 893, on the breakup of one such planned survey.)

For Sinclair, a simple approach is best. With two men and two Land-Rovers—plus a lifetime of natural history observations—he hopes to understand, if not count precisely, the densities, habitats, and cycles of the Serengeti's flora and fauna. Although some note that the shoestring approach may miss crucial information about, for example, all the insects, Sinclair and Mduma argue that the Serengeti Biodiversity Project is a more realistic model for the developing world than all-taxa surveys. “It's a tremendously exciting approach and just what we need more of,” says Peter Raven, director of the Missouri Botanical Garden in St. Louis, “because he's looking for cycles of change” and “setting this up with Tanzanians.”

Money for such work wasn't available even in the Serengeti until recently, say Sinclair and others. “Everybody's idea of the Serengeti is a big acacia tree with a leopard hanging in it,” says Peter Arcese, an ecologist at the University of Wisconsin, Madison, who has also worked extensively in the Serengeti. “So that's where the grants went.” The emphasis on the big glamour pusses only began to shift around 1990, says Sinclair, when researchers began to hammer home the fact that Earth is now in the grip of a sixth great extinction, following the five recorded by fossils (Science, 21 July 1995, p. 347). Suddenly, documenting the remaining species became an urgent task, and funding agencies began putting up money to make it happen. The NSF's inventory program, for example, was launched in 1991; this year's budget is about $2.4 million. Conservation International had begun its Rapid Assessment Program 2 years earlier. Often, these projects focused on areas that had not been studied at all. But now, researchers are beginning to discover the scientific value of long-studied national parks such as the Serengeti, Kruger Park in South Africa, and Yellowstone, says Sinclair: “They're the places that hold a record of our natural world—and in about as pristine a condition as we can now find.” But even though they are protected, they continue to suffer decline—which makes documenting what's in them even more important, adds Keith Langdon, a biogeographer at Great Smoky Mountains National Park in Tennessee, who's coordinating a proposed survey there—the first ever all-taxa inventory of any U.S. park (Science, 12 December, p. 1871). Tackling a project of this size seems at first glance a logistical nightmare—particularly with a staff of two and only$60,000 a year. In the 14,673-square-kilometer Serengeti, habitats range from grasslands to various kinds of acacia forests to stony outcrops, and organisms range from a minimum of 28 acacia species to untold numbers of beetles. So instead of visiting each square kilometer, Sinclair and Mduma are studying samples of each habitat. And instead of collecting thousands of specimens and then enlisting dozens of taxonomic specialists to identify them, as is done in the all-taxa surveys, the pair is simply “starting with the species we know or can easily identify,” Sinclair says. Mduma is a small-mammal specialist, and ecologist Sinclair's first task as an undergraduate researcher in 1965 was to learn all 517 known bird species in the Serengeti by sight and song. As for the butterflies, the team is identifying them like any old-fashioned natural historian—by using a just-published field guide.

Sinclair acknowledges that this approach won't document every organism from microbe to mammal, but it will give the big picture—and how that picture is changing (see sidebar). “A list [of species] by itself isn't that helpful for conservation purposes,” says Sinclair. “What we want to know is where do you find the species, what habitats are they living in, and how do these habitats change over time? The Serengeti isn't like a museum. It has cycles; it does change.”

He and Mduma sample along transects—the main roads—at least once a year, spotting large birds such as shrikes, starlings, and doves. Smaller birds, such as warblers and finches, are identified by both sight and sound. Later in the survey, they will also set mistnets in the dense forests to catch more secretive bird species. For the small mammals, they will put out live traps and cover animal trails with wet sand to record tracks.

Sinclair is also drawing on his 32 years of working in the park and making informal observations. He has a habit of jotting down sightings of particular species in a pocket notebook and photographing key areas annually, giving him a huge data bank to draw on. Anne Sinclair is even combing the letters the couple wrote home over the past 30 years, pulling out notes on animal and plant sightings. “It's not very high-tech,” says Sinclair, “but if you're diligent about it, in time, you can pick up patterns that you'd otherwise miss.”

Among those patterns are the densities of species in their particular habitats and how these change over time. “Thirty to 40% of the park has changed its vegetation community in the last 25 years,” says Sinclair, “and that change should bring an accompanying change in the fauna.” For example, he and Mduma hope to identify healthy stands of forest with abundant animals and compare them to more fragmented, disturbed forests. “That may give us some idea of how much forest is necessary to maintain the species, and an inkling about why this fragmentation is occurring,” says Sinclair.

Once they have detailed data on certain habitats, they can spot-check other examples of the same habitat to see if the diversity patterns hold true. Then they can get a sense of how the whole park works by putting it all together.

The transect results, coupled with Sinclair's notebooks, are already turning up new long-term biodiversity patterns. For example, certain shrikes and thrushes have moved into the park, says Sinclair—ones “that I know for certain weren't here 30 years ago, because they are so visible. We don't yet know why this has happened.” Similarly, the visit to the forest along the Grumeti River turned up a small population of black-and-white colobus monkeys, the farthest west these monkeys have ever been seen. Again, at the moment, no one knows why.

But species are disappearing, too, particularly in the riverine forests. One site had seemingly healthy populations of trogons and large-casqued hornbills in 1965. Now it has lost much of its tree cover, as well as many species of birds, including these two. “It's quite clear that some [bird] species will exist only in intact forests, ones with a complete canopy cover,” says Sinclair. “We want to measure how much canopy cover is necessary for maintaining species like those.” So far it seems that the canopy needn't be wide—perhaps only 50 to 100 meters—but it must extend for some distance along the river.

Eventually, the duo plans to extend the faunal inventory to the park's insects, reptiles, and fish with the aid of specialists in those fields. But because of limited funds, Sinclair imagines that only a handful of people will be involved, so many organisms would still be left out.

Tropical ecologist Daniel Janzen of the University of Pennsylvania in Philadelphia, who pioneered the idea of all-taxa inventories, cautions that to achieve his goals, Sinclair may need to identify more species. That means more people and much more money. “If you're going to invade Normandy, you're going to have to pile on the resources,” says Janzen. “It's OK to start out small like they're doing, but to do the full inventory requires a massive attack. And it's going to be expensive.” Without that full inventory, including, for example, the “gut flora of the buffalo,” Sinclair won't have the “full Yellow Pages” of the Serengeti, adds Janzen.

Other experts, including James L. Patton, an evolutionary biologist at the University of California, Berkeley, who has begun the first small-mammal survey of the Amazon Basin, think Sinclair's more scattershot approach is feasible. “To get down to the soil microorganisms isn't always necessary,” says Patton. Besides, he adds, Sinclair's focus on the Serengeti's dynamism “as the key to its biodiversity is what makes this project so neat.” Raven, who is investigating the possibility of having his institution team up with Sinclair, adds that “Sinclair is gathering baseline data that will help people manage that park; you don't need every microbial organism for that.”

The debate doesn't trouble Sinclair, who thinks that their count will take a minimum of 10 years and “in some ways, given the park's cycles, it will never be complete.” Nor does this open-ended aspect of his research worry him: “It's a record of what we see here today in the Serengeti. It's what I wish the first European explorers in East Africa had recorded. If they'd done this, we'd have a much better understanding of how the Serengeti changes over time—and a far better idea of how to preserve it.”

• ECOLOGY

# Return of the Forest

1. Virginia Morell

Serengeti National Park, TanzaniaBack in 1980, when the acacia and bush forests of the Serengeti National Park were shrinking, ecologist A.R.E. Sinclair tried to take notes on the “last tree in the Serengeti.” “I remember climbing to the top of that ridge,” he says, pointing to a steep hill. “There was then one acacia growing at the very top.” The forest decline had been going on since before Sinclair started working in the park in 1965—and elephants were held responsible. “Everyone blamed the elephants, which was easy to do since you could see them eating the trees,” he says.

But now, in the late 1990s, Sinclair has a very different perspective on the comings and goings of Serengeti species. His long-term monitoring of the park's ecosystem (see main text) has persuaded him that rather than having a single cause, such changes are driven by complex interactions among such factors as the life-span of acacia trees, the numbers of wildebeest, and the influence of humans. The loss of forest—and its recent return—is a case in point.

By the mid-1970s, some conservationists feared that the Serengeti was doomed to become a desert unless elephants were culled. “People thought the forests and bush were the way ‘pristine Africa’ was supposed to be,” recalls Sinclair. In fact, the “pristine” woodlands of the 1930s and '40s were new, themselves part of a larger trend. They had grown up after cattle belonging to the local people had been devastated by an outbreak of rinderpest disease in the 1890s—leading to mass human starvation.

People burn grasslands, in part to create fresh pasture for cattle, and fewer people meant fewer fires. As a result, the acacia seedlings grew up, ultimately creating forests in the midst of the grasslands. But by the 1920s and '30s, the human population had begun to recover, although the Serengeti's vast wildebeest herds—also devastated by rinderpest—did not. Without the wildebeest munching their way across the plains, the grass grew tall, and people—from Masai herders to park rangers—once again began setting fires. Now, however, they set more and hotter fires to control the grasses.

In the 1960s, several factors came together: A wet climate favored the grasses, so even more fires were lit, scorching new tree seedlings. The older acacias, which live only 60 to 70 years, began to die. And although elephants also fed on the young trees, Sinclair and his then-graduate student Holly Dublin did field experiments in the 1980s showing that fire—not elephants—killed off most of the seedlings. The result was a nearly treeless Serengeti.

By the mid-1970s, the ungulates had at last recovered, following the establishment of a successful rinderpest control program in 1963. Because wildebeest were again cropping the grass, people slowly began setting fewer fires. Today, the park is dramatically different from what Sinclair saw when he first arrived 32 years ago. Numbers of buffalo and elephants are far lower due to heavy poaching (although elephants have been increasing since the 1990 ivory ban). The wildebeest population has soared to about 1 million; human-set fires are down to about a quarter of what they were—and the acacias have returned.

Dense young stands of forest cover most of the park's hillsides and flank its grasslands. “That shows just how easy it is to get these things wrong,” says Sinclair. “The Serengeti didn't turn into a desert, but a forest.” And the “last tree on the Serengeti”? It's surrounded by 3-meter-tall acacias, standing limb to limb from the top of the ridge to its base.

• PHYSICS

# Cool Sounds at 200 Decibels

1. Dana Mackenzie
1. Dana Mackenzie is a science and mathematics writer in Santa Cruz, California.

The loudest controlled sounds ever made by humans were produced earlier this month—not by a rock band, but by a physicist. At the Acoustical Society of America meeting in San Diego, Timothy Lucas of MacroSonix Corp. in Richmond, Virginia, demonstrated a new “acoustic compressor” that uses ultraintense sound waves to do the work of a mechanical pump. The technology may soon be used in everyday appliances such as refrigerators and air conditioners.

The idea of the compressor is simple: You shake a can back and forth to create vibrations in the air inside. Just as a child can produce huge waves in a bathtub by sloshing back and forth at just the right rate (a phenomenon called resonance), the air vibrations become especially intense if the can is agitated at a certain frequency. But the water in the child's bathtub will splash out if the waves start to crest. For acoustical engineers, the analogous problem is shock waves, which dissipate the sound energy as heat. By making his compressor just the right shape—essentially that of a bowling pin—Lucas was able to keep the shock waves from forming, even as the can vibrated at about 600 times a second.

How loud are the resulting sounds? The pain threshold is about 120 decibels, and a jet engine produces 150 decibels. If you stand next to a sound of 165 decibels, it will ignite your hair. The sound waves inside Lucas's compressor are about 3000 times more powerful, or about 200 decibels. But because the can's own vibrations are much smaller than the vibrations of the air, on the outside it sounds just like an ordinary compressor.

The intense sound waves oscillate between low and high pressure in certain regions; with the help of valves that open and close at the right moments, these pressure differences can suck gas into the compressor and shoot it out at high pressure. Lucas's compressor could be especially useful for refrigerators and air conditioners, which work by compressing a refrigerant—traditionally a chlorofluorocarbon. Steve Garrett, a physicist at Pennsylvania State University in University Park, explains that some of the ozone-sparing refrigerants now being used break down in the oil that lubricates a conventional compressor. But Lucas's compressor has no moving parts inside and therefore requires no lubrication. MacroSonix has already signed a licensing agreement with an appliance manufacturer.

Other specialists in acoustics call Lucas's compressor a breakthrough. “What Timothy Lucas has done is shift the debate from whether acoustic compression can be done to who can do it better,” says Garrett. Lucas himself thinks his sound waves will ultimately find many other roles. “Electromagnetic waves have been commercialized for over 100 years,” he says, “but the commercial application of sound waves has only scratched the surface.”