News this Week

Science  31 Oct 2008:
Vol. 322, Issue 5902, pp. 658
1. U.S. ELECTION

Scientists Plant Grass-Roots Effort for Obama in Final Days of Contest

1. Eli Kintisch

As a graduate student at Iowa State University in Ames, physicist Bernice Durand worked for antiwar candidate Eugene McCarthy in 1968 in his failed bid for the Democratic nomination for U.S. president. Four decades later, the recently retired University of Wisconsin, Madison, professor and administrator has jumped back into the political fray on behalf of Senator Barack Obama. This time around, Durand is using a career's worth of contacts and organizational skills to build an unusual national grass-roots effort, focused on scientific issues, for the Democratic candidate.

Since September, Durand has worked with more than three dozen scientists who have placed articles or letters in 50-plus newspapers in 20 states, most of them considered still up for grabs. The scientists have also appeared on a handful of radio shows and been interviewed by reporters covering the campaign. “I feel I'm doing something that will make the country better,” says Durand. “On issues of science, on support for research, and on his interactions with the scientific community, there's no contest compared to [Senator John] McCain,” she says.

Political analysts say Obama has captured the lion's share of visible support among scientists. “It's an enthusiasm chasm,” says Michael Stebbins, president of the Scientists and Engineers for America (SEA) Action Fund, which set up a channel on YouTube for scientists of both political persuasions to explain their choice. As of press time, 22 videos have been posted—all by Obama supporters. “It's been frustrating. We want scientists to come out and say why they're voting for McCain,” says Stebbins, who has volunteered for the Obama campaign in addition to continuing his SEA efforts.

Durand traces her recent activism to conversations with colleagues this summer at the Aspen Center for Physics in Colorado. They passed along her name to physicist Donald Lamb of the University of Chicago in Illinois, who has helped organize the Obama campaign's scientific advisory committees (Science, 26 September, p. 1762). Lamb then invited Durand to launch the network.

Durand, who was eager to take up a new challenge after her retirement, estimates that she's spending 45 hours a week on the project. She's written more than 1000 e-mails aimed at building the network, recruiting authors, and providing suggestions on their draft articles or letters. The Obama campaign, which declined comment for this story, has kept the effort at arm's length. “It's a grass-roots thing,” says Lamb, who spent last week working to get out the vote in northern Virginia.

The pieces typically emphasize local concerns. In an op-ed published earlier this month in The Virginian-Pilot, Francis Collins, who in August stepped down as director of the National Human Genome Research Institute at the National Institutes of Health (NIH) in Bethesda, Maryland, described himself as “a citizen, a scientist, a physician, and a son of Virginia.” The author explains why he supports Obama's “science, technology and innovation agenda,” which includes doubling the NIH budget, and notes that NIH “supports most biomedical research in our premier universities—including many in Virginia.”

Many of the letter writers have also mentioned the list of U.S. Nobel Prize winners, now up to 70, who have endorsed Obama. “[I'm] not sure people up here would have heard about that otherwise,” says Brian Black, a biology professor at Bay de Noc Community College in Escanaba, Michigan, whose letter has been published in four papers located in the state's sparsely populated and Republican-leaning northern regions. “I'm sure there are those who think I'm a nut.”

Some scientists believe that their training prepares them to be effective political communicators. “A lot of interacting with voters involves studying the issues, developing a coherent and logical argument, and articulating one's ideas. And thinking on one's feet. And this is what scientists are trained to do,” says Daniel Holz, a physicist at Los Alamos National Laboratory in New Mexico. But Dartmouth College cosmologist Robert Caldwell was a little disappointed when four New Hampshire papers rejected his letters. He wonders if the “footnotes and references” he included to bolster his arguments hurt his chances. “They probably added to the word count,” he says.

Campaign donation records indicate that some scientists are supporting McCain, but Science could not find any evidence of grass-roots efforts by scientists on his behalf. (The McCain campaign did not reply to requests for comment.) One donor, mathematician Nakhle Asmar of the University of Missouri, Columbia, says “national security” was the reason he gave McCain $2300, the maximum allowed from an individual for the general election. Has he done anything more for the candidate? “I don't have time,” he says, adding that he believes “both [candidates] will be good for education and science.” Some who have volunteered for Obama say they would have preferred to remain nonpartisan but that the stakes are too high. Holz, who has written to local papers, recorded YouTube videos, and canvassed for Obama, worries that such activism could “compromise the scientific enterprise” by politicizing it. But the bigger problem, he says, is that “the scientific enterprise has already been compromised and politicized by Republicans.” White House science adviser John Marburger says he opposes “scientists using science to support their partisan views.” The problem, he says, is that they could “lose credibility with the public.” Durand doesn't pretend to know if her efforts have helped Obama. But she's got her fingers crossed. “I feel so much more optimistic now than I did in 1968,” she says. 2. SELF-EXPERIMENTATION Eat, Drink, and Be Wary: A Sugar's Sour Side 1. Jon Cohen In 2001, Ajit Varki's dream came true when he drank an extract from pig spit. Varki's strange culinary excursion was part of an experiment that he believed might help explain the unique susceptibility humans have to some infectious diseases, cancers, and heart ailments. A prominent researcher in the dual disciplines of sugar biology and evolutionary biology, Varki first purchased a few kilos of glands taken from the jaws of pigs. Next, his lab at the University of California, San Diego (UCSD), minced and homogenized them to extract the mucins. From these mucins, proteins secreted by mucosal surfaces, they plucked off a sugar called Neu5Gc for short. And that's what Varki's sweet tooth craved. Neu5Gc, which is also known as a sialic acid, is made by chimpanzees and many other mammals but not humans, and Varki's group earlier had found the genetic mutation that prevents us from making it. Varki suspected that when we are exposed to Neu5Gc, it incorporates into our cells, where it somehow makes humans more susceptible to a variety of diseases. The first step was to figure out how it entered the body—thus, the pig spit experiment. Varki reasoned that Neu5Gc could enter humans through food that contains it, such as red meat and milk products. To prove it, he proposed ingesting huge amounts of pig Neu5Gc and seeing where it went. But when Varki asked his institutional review board for permission to drink the Neu5Gc, some members balked. “I was at first told that self-experimentation was not allowed any more,” says Varki. He assured them that he would assess results using objective measures such as mass spectrometry to prove its presence—and he slyly asked if any members of the committee wanted to be the volunteers, he recalls. They gave him a green light. To establish a baseline, Varki restricted what he put in or on his body for 2 days before the experiment: no red meats, milk products, or lanolin shampoos, all of which contain Neu5Gc. Then he checked into a clinical research center at UCSD and drank 150 milligrams of the Neu5Gc dissolved in 100 milliliters of water. “It was slightly sweet and sour, slightly acidic,” says Varki, pig-spit connoisseur. He wasn't particularly worried that the Neu5Gc (full name, N-glycolylneuraminic acid) would make him sick. “It was like eating 14 pork steaks,” he says. “People do that on July 4th.” But to be safe, the clinical center kept him under observation all day, taking blood samples every 2 hours. No side effects surfaced. Urine, saliva, and hair trimmings over the next week all showed increased levels of Neu5Gc. Closer analyses showed that his cells had actually taken it up and incorporated it on their surfaces, as they do with other sialic acids in the synthesis of new glycans. Two of Varki's colleagues did the same self-experiment with similar results, which they published in the 14 October 2003 issue of the Proceedings of the National Academy of Sciences. “There's no other example I know of where you eat something foreign that outfoxes the biochemical systems and becomes part of you, no different from molecules made in your body,” Varki says. Varki has since taken that observation a step further. This week, Nature is publishing a new study online, led by Varki and two teams of researchers in Australia, that strongly ties Neu5Gc to a human disease and ingestion of red meat. “It's a very concrete example of how our susceptibility to disease might be governed by our diet,” says Carolyn Bertozzi, a carbohydrate chemist at UC Berkeley. “Ajit is an incredibly creative guy. Sometimes he's chasing strange meteors and comets, and sometimes he hits something. This is a really interesting story, and I'm very excited by it.” Varki has long wondered why chimpanzees and humans are genetically so similar but suffer from different diseases, and he sees Neu5Gc a s one key to solving that mystery. But so far, speculations have outnumbered evidence. “He's been looking for that direct link with disease, and it's been elusive,” says Bertozzi. That is, until the new Nature study, which she says is thoroughly convincing. Neu5Gc connected Varki and his Australian collaborators through a circuitous route that dates back to the death of several children in 1993 who ate tainted hamburgers from Jack in the Box restaurants in the United States. The culprit was later identified as a deadly strain of the gut bacteria Escherichia coli, known as 0157:H7. A toxin secreted by this E. coli, Shiga, can lead to hemolytic-uremic syndrome (HUS), which causes kidney failure. Molecular microbiologists James and Adrienne Paton, a husband-and-wife team at the University of Adelaide in Australia, subsequently discovered several other Shiga-producing E. coli that caused HUS outbreaks there, and one secreted a second toxin as well, subtilase cytotoxin (SubAB). Toxins must first bind to the surface of a cell to do their damage, which led the Patons to David Smith of Emory University School of Medicine in Atlanta, Georgia, who specializes in matchmaking ligands and receptors. Smith found that SubAB has a high affinity for Neu5Gc, and he told the Patons about Varki's work. Their subsequent collaborative studies make a compelling case that when humans eat meat or dairy products that have high levels of Neu5Gc, it becomes incorporated into their cell surfaces, and SubAB can bind to it. “It's the first time we've seen an example of a component in food being the preferred receptor for a bacterial toxin,” says James Paton. The researchers next showed precisely how Neu5Gc binds to the toxin, which included crystallizing SubAB, an intensive effort done in Jamie Rossjohn's lab at Monash University near Melbourne. Human cells fed Neu5Gc also became much more susceptible to SubAB, the team found. And mouse experiments further clarified the connections between SubAB, Neu5Gc, and disease. So consider the delicious irony. E. coli that produces the SubAB toxin contaminates red meat and milk products. Humans who ingest these foods incorporate Neu5Gc into their cells, making them hypersusceptible to SubAB—and much more likely to become seriously ill from the toxin. Although these insights have no immediate practical application, Varki hopes they may open a door that eventually helps explain and even thwart major diseases. Some forms of Neu5Gc are seen as foreign by the human immune system, and we sometimes create antibodies to it. Varki suspects that these antibodies may contribute to autoimmune diseases, cancers, and heart problems seen in humans but not in chimps. Pathogens can also directly bind to Neu5Gc on cell surfaces, and one strain of the malaria parasite does just that, readily causing disease in chimps but not humans. Paton suggests that the greatest impact of the new findings may be in sparking epidemiological studies of, say, vegans, that prove these links. Any way you look at it, Neu5Gc proves the point like never before: You are what you eat. 3. EVOLUTION Two Sets of Cave Bear DNA Uncover the Bear Facts 1. Michael Balter What kind of bear was Winnie-the-Pooh? Author A. A. Milne christened the fictional character after the teddy bear of his son, who in turn had borrowed the name from an American black bear in the London Zoo called Winnipeg. Yet for decades, researchers have argued about whether Winnipeg's scientific name should be Ursus americanus or Euarctos americanus. Indeed, although there are only eight species of living bears, scientists have come up with at least half a dozen versions of the bear family tree. Now a paper published by the Proceedings of the National Academy of Sciences (PNAS) online this week helps untangle bear phylogeny by presenting “the first mitochondrial genome” from the extinct cave bear, Ursus spelaeus. But another paper, published with little fanfare last July, also reported the complete mitochondrial DNA (mtDNA) of the cave bear, as well as that of the extinct American short-faced bear, Arctodus simus. The two teams are arguing about scientific priority. But for the bears, this means that two sets of data now illuminate their family tree, although the studies disagree about the timing of bear evolution. Both teams independently leaped a major technological hurdle, adds evolutionary biologist Hervé Bocherens of the University of Tübingen in Germany: They sequenced the first complete mitochondrial genomes from specimens that are tens of thousands of years old but not preserved in permafrost. “This opens the field of complete mitochondrial sequencing to a very wide range of extinct species,” says Bocherens. Researchers had already sequenced the mtDNA genome in all eight living species of bears and used the genetic differences among them to create family trees. But because the bears underwent a rapid and fairly recent radiation, those variations are not great. To have confidence in their trees, researchers needed data from extinct animals. They got it twice over. In this week's paper, a team led by biologist Jean-Marc Elalouf of the Atomic Energy Commission in Saclay, France, reported cave bear mtDNA from a celebrated source: a 32,000-year-old sternum bone from France's Chauvet Cave, site of the oldest known cave art (Science, 15 August, p. 904). In another paper back in July, evolutionary biologist Michael Hofreiter of the Max Planck Institute for Evolutionary Anthropology in Leipzig, Germany, reported in BMC Evolutionary Biology that his team had extracted mtDNA from a 44,000-year-old cave bear femur found in Austria; the team also reported the mtDNA genome of a 22,000-year-old American giant short-faced bear from Canada. Hofreiter's team sent its paper first to PNAS, on 11 December 2007, but it was rejected without review, say Hofreiter and PNAS Editor-in-Chief Randy Schekman. Elalouf 's paper, which Schekman says was handled by a different member of the PNAS editorial board, was submitted on 1 July 2008 and accepted the following month. Elalouf argues that although his team's paper is second, they submitted their sequence to GenBank—the National Institutes of Health's repository for DNA sequences—back on 4 December 2007. Hofreiter's team didn't submit its cave bear sequence to GenBank until 23 June 2008, shortly before its paper appeared. Schekman says that Elalouf has now agreed to add a “note in proof” to the print edition of PNAS, which will be published shortly, acknowledging the Hofreiter group's earlier paper; Hofreiter says that solution will satisfy his group's concerns. Whatever the priority, both groups agree on the outline of the bear family tree. They confirm that the giant panda was the first species to split off from the lineage leading to later bears, and both conclude that the cave bear shared a common ancestor with the brown bear and the polar bear, which turn out to be closely related to each other. Moreover, both teams slash the number of genera of living bears from seven in some schemes, to three for the Hofreiter group and four for the Elalouf group. They assign most species—including Winnipeg's—to the genus Ursus. Adding data from the two extinct bears provides a “robust” tree that is “important for understanding the evolutionary history of this mammalian family,” says Ya-ping Zhang, an evolutionary geneticist at the Kunming Institute of Zoology in China, who published the complete mtDNA genomes of five living bears in 2007. Yet when it comes to the timing of the recent bear radiation, the two groups part company. Elalouf concludes that it was only about 2 million to 3 million years ago, using a previous estimate of the giant panda's divergence at 12 million years ago as a chronological anchor point. Hofreiter's team anchors its tree with the much earlier divergence of the harbor seal and finds that the panda split off earlier, about 19 million years ago, and that the rest of the bears radiated about 5 million years ago. He notes that some aspects of climate changed dramatically about that time, when the Bering Strait opened and the Mediterranean Sea became drier. Other mammals also showed dramatic changes at this time, such as the split between the human and ape lineages. Researchers outside the fray are divided. “I would not take the divergence time too seriously in either paper,” says Xiaoming Wang, a paleontologist at the Natural History Museum of Los Angeles County in California. Wang adds that more research on the bears' nuclear genomes, which is still at an early stage, will be necessary to fine-tune the chronology. For now, at least, Winnie-the-Pooh should be happy to have his true scientific name at last. 4. BIOETHICS U.K. Approves New Embryo Law 1. Gretchen Vogel With the enthusiastic support of the scientific community, the British House of Commons has overwhelmingly approved a wide-ranging bill that expands the country's rules governing work with human embryos. The new standards, which have dismayed opponents of embryo research, spell out the kinds of research governed by the country's Human Fertilisation and Embryology Authority (HFEA). The 22 October vote in the Commons, which favored the bill 355 to 129, was considered its most significant hurdle, although the bill still needs final approval from the House of Lords. The bill updates the 1990 regulations establishing HFEA and a 2001 law governing nuclear transfer, as well as other regulations pertaining to reproductive technologies. It allows several kinds of research that were not covered previously, including interspecies nuclear transfer, in which scientists attempt to create an “admixed embryo” by fusing a human cell and an enucleated animal egg. Some scientists hope to use such embryos to derive embryonic stem (ES) cells. HFEA has already granted three licenses for such work, but opponents had challenged the licenses in court, charging that the agency had no legal authority to grant them. The new bill provides that authority. The bill also says that HFEA can grant licenses for research to create transgenic embryos carrying human and animal genes or to create chimeric embryos by mixing human and animal eggs or sperm. Opponents have claimed that the bill authorizes the creation of “humanzees.” But scientists “made a huge effort to allay fears that this was going to lead to real human hybrids,” says stem cell expert Stephen Minger of King's College London. (The bill forbids allowing any human-animal embryos to develop for longer than 14 days or implanting one in a human or animal womb.) Their lobbying paid off. “Many people said, 'I am naturally queasy about this. I would have voted against it, but you guys have made such a strong case I can't see any reason not to vote for it,'” Minger says. Developmental geneticist Robin Lovell-Badge of the National Institute for Medical Research in London says experiments mixing human sperm and, for example, transgenic mouse eggs can yield important insights into the process of fertilization. Researchers hoping to test new methods of storing human sperm or new contraceptives that target fertilization will also benefit. Some politicians had argued that induced pluripotent stem cells, which are ES-like cells that are reprogrammed using a cocktail of specific genes instead of an oocyte, render interspecies nuclear transfer unnecessary. Minger and Lovell-Badge cite important reasons to pursue the technique, for instance, to compare the ES cells that result from both processes. Studying nuclear transfer—without having to rely on scarce human oocytes—also offers the best chance for teasing apart exactly what happens to turn back the clock of an adult cell and allow it to direct the process of development again, Minger says. 5. SCIENCE IN CHINA You Say You Want a Revolution 1. Hao Xin* 1. With reporting by Richard Stone. Thirty years ago, Deng Xiaoping opened China to the world and brought scientists in from the cold. As researchers celebrate, some warn that the community still has major problems that need to be solved. Thirty years ago, Deng Xiaoping opened China to the world and brought scientists in from the cold. As researchers celebrate, some warn that the community still has major problems that need to be solved BEIJING—After Lu Yongxiang began a third term as president of the Chinese Academy of Sciences (CAS) last March, a TV talk show host asked the former varsity soccer player which question he would like to answer first: When can Chinese scientists win the Nobel Prize? Or when can the national men's soccer team win the World Cup? In both cases, Lu responded, the challenges are similar. China's scientific community and its much-derided men's national soccer team must build stronger foundations. For a China-based scientist to win a Nobel Prize or the soccer team to win the World Cup, Lu said, both need more money, more talent, and an environment that encourages innovation. In the 30 years since Deng Xiaoping and other leaders opened China to the outside world, China's science, like its economy, has grown immensely. According to the Organisation for Economic Co-Operation and Development's publication Main Science and Technology Indicators 2008, China's$87 billion R&D expenditure in 2006, in purchasing power parity dollars, was higher than all countries except the United States and Japan, and only the United States has more researchers—1,387,882 compared with China's 1,223,756. Officials with China's Ministry of Science and Technology (MOST) like to point out that China is now second only to the United States in the number of publications in international journals.

But in many ways, China punches below its weight in science. “Our country has not made contributions proportionate with its overall strength,” neuroscientist Rao Yi of Peking University and structural biologist Shi Yigong of Tsinghua University wrote in a recent editorial in the newspaper Huanqiu Shibao. They and others argue that China's rising R&D investments are being misspent on facilities and megaprojects that are driven by special interests, creating an illusion of grandeur rather than bringing China closer to the forefront of international research.

Also disturbing is that many Chinese scientists exhibit a surprising lack of curiosity, asserts Rao, who says he has endured “intellectual starvation” since returning to his homeland last year from Northwestern University's Feinberg School of Medicine in Chicago, Illinois. Although many scientists eagerly showcase their own work at conferences, Rao says, few discuss ideas informally or show up at seminars to listen to colleagues—interactions that inspire creativity in the West. “True collaborations are rare, and motivations for science are driven by temporary and relatively easy goals,” he says.

Another damning assessment comes from theoretical physicist and former president of CAS Zhou Guangzhao. In China, he says, “success is often scored by quantity rather than quality.” For that reason, Zhou contends, most Chinese scientists are content to follow well-trodden paths and churn out routine papers rather than strive for fundamental breakthroughs. Deference to status also makes it difficult for junior researchers to challenge academicians or science mandarins. That wasn't so in the 1950s and 1960s, when Zhou was working on China's atomic bomb project; then, he says, scientists treated one another as equals and worked collectively toward the goal of strengthening China. These days, many scientists say, there is greater freedom in society, but a market economy has made private interests the driving force of science, supplanting the idealism that inspired earlier generations of researchers.

Although public discussion about systemic problems in Chinese science runs up against censorship in state newspapers and TV—the government's voice—the blogosphere now provides a largely unfettered forum. “Blogs and the Internet as a whole are changing China's political discourse,” says Cao Cong, a senior research associate at the Neil D. Levin Graduate Institute of International Relations and Commerce in New York City. Cao, a blogger on ScienceNet.cn, hopes that “positive and constructive opinions raised in the blogosphere” will receive official attention. This seems to be happening. Earlier this month, Chinese media reported that Premier Wen Jiabao, alerted by a journalist's blog about a cover-up of a disaster in Shanxi Province—where a mudslide caused by dumping mining waste killed more than 40 people in August—sent a team to investigate. Zhao Yan, editor-in-chief of ScienceNet.cn, hopes the site's 1400-plus bloggers may spark a bottom-up reform not just in matters of public safety and governance but also in science and technology, about which frank talk among peers is sorely needed.

A time of revival

China's reforms and opening up followed the decade-long turmoil known as the Great Proletarian Cultural Revolution launched by Mao Zedong in 1966. Seeking to rid the country of what he labeled as feudal, bourgeois, and foreign influences, Mao closed universities and banished professors to the countryside to work as peasants. Research was halted, except in areas that served national needs such as defense.

Careers like that of Chen Jia'er, a young physicist in the 1960s, were thrown into reverse. After 3 years as an exchange researcher in nuclear physics at the University of Oxford in the U.K., Chen returned home in 1966, hoping to build a heavy-ion research program. Instead, he was branded a “reactionary academic authority” and sent for reeducation to a village in eastern China, where he laid railroad tracks, raised pigs, and worked odd jobs for almost a decade.

Mao's death in 1976 and the subsequent purge of the coterie led by Mao's widow released China from its ideological straitjacket. One of the first reform steps Deng took was to rehabilitate scientists from a class to be “won over, reeducated, and transformed” to vital members of society whose knowledge and expertise would help modernize the country. Scientists such as Chen were brought in from the cold.

In a keynote speech at the first National Science and Technology Convention in Beijing in March 1978, Deng declared that “science and technology is a productive force.” The “Spring of Science,” as the founding and then-CAS president Guo Moruo poetically pronounced, had arrived. Many scientists recall that time fondly. It was “the turning point of my life,” says Chen, who served as president of China's National Natural Science Foundation (NSFC) from 2000 to 2003.

Deng's call for modernization posed a daunting challenge. For some like Peking University biochemist Gu Xiaocheng, it meant racing to recoup lost ground. Gu had been one of the few professors permitted to remain in Beijing during the Cultural Revolution. She was part of a team that set out to synthesize insulin, a project considered in the national interest. They succeeded in 1965 and tried to determine insulin's crystal structure. But even for this elite group, “no international journals were available to us,” Gu says. After the Cultural Revolution in the late 1970s, “when I saw Science again after so long, I thought, 'They're speaking a different language.'We really didn't know how to catch up.”

Since then, China has worked to reform its R&D system, but these efforts have been top-down and often flawed, says He Zuoxiu of CAS's Institute of Theoretical Physics here. In the 1980s, then-Premier Zhao Ziyang wanted the marketplace to decide what research was needed and directed CAS to focus on applied research. The leadership “confused applied research with product development,” He says, and the resulting tendency to ignore basic research has weakened the country's ability to innovate. Chinese leaders abandoned Mao's idea of self-reliance and expected the country to acquire advanced technologies from multinational companies in exchange for giving them market access. Zhao infamously told Chinese scientists in 1985 to “go up hills and pick peaches,” reflecting his belief that China could simply reap the fruits of research done in other countries. But without accumulating one's own knowledge, says Chen, “it's impossible to have new ideas or really know how to apply them.”

A second major reform came in the late 1990s, when “indigenous innovation” became a buzzword. Former president of China Jiang Zemin called on scientists to focus their efforts on national needs, “to do [research in some areas] and not to do [it in others].” In response, in 1998 Lu launched CAS's Knowledge Innovation Project (KIP). Under that banner, Lu reduced the number of CAS institutes from more than 100 to about 80 and its 80,000 work force to 48,000 (Science, 23 February 2001, p. 1477). The streamlining, Lu says, made institutes “more active and dynamic.”

The innovation project achieved some positive results: “Our facilities were dilapidated before, but now many new buildings rival those at universities,” says Wang Zhizhen of CAS's Institute of Biophysics here. The work force is much younger than it was a decade ago, and most researchers have studied abroad. Some institutes, especially newer ones, such as the Institute of Neuroscience established in 1999, compete at an international level, says Lu.

But for many older institutes, the drastic work-force reduction only looks good on paper. These institutes must use money allocated to a smaller payroll to support retirees and staff members not counted as KIP personnel. To accomplish this, institutes collect “head taxes” from principal investigators with grants to augment PIs' salaries, based on their productivity, and to pay junior researchers and grad students. Although basic salaries for PIs are fixed at several thousand dollars a year, productivity-based supplements can boost annual incomes to well over $30,000. For grad students, the basic stipend is about$500 a year; those lucky enough to receive supplemental pay may get an additional $3000. Grants are golden because they provide the lion's share of productivity-based pay, even though many funders explicitly forbid using grant money this way. Institutes account for the payments as user fees, processing fees, or collaboration fees, according to several researchers who asked to remain anonymous to avoid retribution. These scientists estimate that about 10% of total grant money at well-funded institutions, and as much as 50% at poorer ones, is spent on salaries. As a result, some PIs go after grants beyond what's needed for research and outside their areas of expertise, says Zhao Zhongxian of the Institute of Physics here. To combat this problem, Zhou says salaries should be capped and the portion from grants should not exceed 3 months' worth of PI pay, as many U.S. research universities stipulate. A few Chinese institutes have adopted this approach. The Institute of Physics, Zhao says, has changed its formula for productivity-based pay such that a PI's salary does not increase linearly with the amount of grant money. Publications also contribute to a researcher's productivity-based pay. Institutes determine publication bonuses differently, but most take into account the impact factors of journals in which papers are published. CAS's Institute of Chemistry follows a typical formula in China: A paper in Science or Nature fetches$2500 or more; a paper published in journals such as Physical Review Letters (PRL) and the Journal of the American Chemical Society (JACS) brings about $1300; papers in journals with impact factors greater than three bring about$500; and papers in journals with impact factors under three are awarded less than \$200. Bonuses are divvied according to the authors' contributions. Universities also pay productivity-based salaries to professors.

A few institutes, including the Institute of Neuroscience, do not pay publication bonuses, whereas some, such as the Institute of Physics, have de-emphasized publication bonuses and only award several thousand dollars to papers published in four journals: Science, Nature, PRL, and JACS.

Quantity trumps quality

Both productivity-based pay and the way Chinese researchers are evaluated emphasize quantity over quality. This is partly because Chinese scientists are often fearful of giving offense if they critique a colleague's work truthfully, Zhao says. Instead, number-based evaluation is considered more objective and has gained popularity. To break the expectation of guaranteed employment regardless of performance—the “iron rice bowl”—Nanjing University in the early 1980s began to use the Science Citation Index (SCI) to measure the productivity of its professors. Since then, universities and research institutes have been ranked annually based on how many SCI papers they churn out. Science ministry grant applications often require PIs to state how many SCI papers they intend to publish, and researchers are promoted and occasionally demoted based on the number of their publications.

The top-performing one-third of CAS institutes has adopted a system of international review. For example, since 2003, the Institute of Genetics and Developmental Biology (IGDB) here has been using outside reviewers to evaluate PIs in its three main research areas. The institute invites a scientist from abroad to recruit a panel of reviewers for each area, says IGDB developmental biologist Zhang Jian. The panel anonymously reviews packages prepared by PIs and conducts a site visit to talk to scientists, research staff, and students. The visitors also give constructive comments to the lab under review. Reviews are conducted every 5 years; a few investigators have been forced out primarily because of the reviews, says Zhang.

Speaking out

To Xu Liangying, a retired science historian here, the root cause of the problem in China's scientific community is Deng's declaration 30 years ago of science and technology as a productive force, now an official mantra. Since then, the Chinese words for “science” and “technology” have been fused into “scitech,” which in common usage solely connotes technology. In China, science is expected to contribute directly to economic development and not to the pursuit of truth and knowledge, asserts Xu.

Xu has always spoken his mind, even though it has cost him dearly. In the late 1950s, he was branded a “rightist” and banished to his ancestral village in Anhui Province. For more than 2 decades, Xu toiled in the fields during the day and translated the collected works of Einstein into Chinese during his spare time. After being allowed back into CAS in 1978, he became China's foremost Einstein scholar. He also took up the cause of human rights in China. Xu was put under house arrest for a time in 1989 after writing an open letter, and collecting signatures for it, that called on the Central Committee of the Communist Party to release political prisoners and allow freedom of speech. Last April, the American Physical Society awarded Xu the Andrei Sakharov Prize for a “lifetime's advocacy of truth, democracy, and human rights.”

These days, speaking one's mind is not nearly as risky. Peking University's Rao was allowed to come back to China after he and others wrote an article in Nature's China Supplement in 2004 that advocated stripping MOST of its power to administer research funds and making the ministry an advisory body. Rao says China needs an impartial science and technology board to advise the State Council on policy and funding priorities. The board, he says, should be made up of people free of institutional conflicts of interest, replacing an existing science and technology group under the premier that consists of ministers who inevitably want more money for their own ministries.

Theoretical physicist He Zuoxiu of CAS agrees and says MOST failed to curb institutional interest when it led the formulation of China's mid- to long-term science and technology plan. “The plan does not represent true national interest; it is a balancing act among interest cliques,” says He. One of the plan's biggest flaws, he says, is its backing of “megaprojects” advocated by individual ministries and scientists (Science, 17 March 2006, p. 1548). Even though China needs to invest more heavily in renewable energy, an area critical for sustainable development, the plan hardly mentions it, He notes. Because the country's top leaders emphasize R&D for national needs, scientists often promote their own research as aligned with such needs, says Rao. Some use political clout and connections to designate their own projects or those of associates and friends as “top national priorities,” he says.

Many researchers discuss such issues openly on ScienceNet.cn, a Web site that has been running for fewer than 2 years and boasts tens of thousands of readers per day. The site's bloggers do not hide behind pseudonyms, which sets it apart from most Internet forums and blog sites in China. “Scientists have no problem with using real names,” says Zhao, “because they want to be responsible for what they say.”

These bloggers call for systemic reforms to curb special interests in setting research priorities, to make the funding system more transparent and fair, and to liberate scientists from meaningless evaluations imposed by administrators. These cries for reform offer a glimpse of what could happen in the future, as a new generation that has prospered in Deng's reformed China pushes its way into the ranks and pressures science leaders to live up to their expectations.

6. BIOMEDICAL RESEARCH

More Than Skin Deep

1. Lauren Cahoon

Scientists still don't know what causes scleroderma, a complex disease often marked by toughening skin and widespread internal fibrosis, but they're developing potential treatments nonetheless.

Scientists still don't know what causes scleroderma, a complex disease often marked by toughening skin and widespread internal fibrosis, but they're developing potential treatments nonetheless

At age 19, Barbara Lowe didn't think much of it when one of her fingers temporarily turned completely white during a lunch break at work. “My friends thought it was a party trick,” says Lowe. Twenty years later, however, the party trick had taken a serious turn. Most of her fingers and toes were suffering bouts of poor blood flow, and searing heartburn and lingering colds constantly hampered her. When she was taken to the hospital for pneumonia, doctors finally gave her symptoms a name: scleroderma. “I didn't know anyone who had it other than myself. My GP [general practitioner] couldn't tell me a lot of about it,” says Lowe. “You're kind of completely on your own with the disease. People just don't know what it is.”

Although scleroderma affects as many as 300,000 Americans and kills roughly 10,000 every year, this autoimmune disease remains an enigma and far from the public's radar. Its cause—or causes—remains murky. Genetic and environmental studies have yielded few clues, as the disease seems to strike almost at random.

Scleroderma was formally discovered in 1754 by Carlo Curzio, an Italian doctor who described treating a woman with thick, stiff skin—the symptom from which scleroderma gets its name (from Greek meaning “hard skin”). This skin condition, along with the circulation problem that Lowe experienced, known as Raynaud syndrome, are the classic signs of the disease—and many people with scleroderma suffer from only these disabling, not deadly, symptoms.

Another hallmark of scleroderma is that patients suffer a diverse array of symptoms, leading many physicians to consider the condition a constellation of disorders. In bad cases of systemic scleroderma, the most severe form of the disease, inflammation and fibrotic scar tissue flare up in multiple organs, blood vessels narrow and harden, and traitorous immune cells attack a person's own flesh. The dizzying complexity of scleroderma has kept scientists both frustrated and fascinated for decades. “The problem is we don't really understand what the primary basis is of the disease,” says Robert Lafyatis, a rheumatologist at Boston University (BU) School of Medicine. “We don't know if it's a vascular disease, a fibrotic disease, or an immune disease.”

At a recent meeting in Cambridge, U.K., Lafyatis and several hundred scleroderma investigators gathered* to compare notes and chart the field's progress. Despite continued bewilderment about what causes the disease, there was good news to report. In the 1980s, physicians began effectively treating the kidney problems that then killed most scleroderma patients. New insights into the disease's molecular underpinnings are helping to tackle other dangerous symptoms, too. Some researchers are even finding that “rebooting” a person's immune system with stem cell therapy may completely eliminate the systemic fibrosis that continues to kill many people. “There's been a change from nihilism to a bright light at the end of the tunnel,” says Alan Tyndall, a rheumatologist at the University of Basel in Switzerland. “As medical students [in the 1960s], we were told that scleroderma is a death sentence and there's no hope; now, that's changed.”

Many suspects, little proof

Since Curzio identified scleroderma, dozens of causes have been put forth but few have stood the test of time. Environmental factors such as organic solvents, asbestos, and even silicone breast implants have all been suspected triggers for the autoimmune reaction associated with scleroderma. However, the evidence for these factors has come from studies too small to be definitive; plenty of people get scleroderma without being exposed to any of these insults.

Fetal cells were another suspect. During pregnancy, such cells pass through the placental barrier and enter an expectant mother's circulation. Scientists found that these foreign cells could live for decades in a woman, and some theorized that the cells might trigger scleroderma. This theory could offer an explanation for one of scleroderma's puzzles: 80% of patients are women, most in their pospartum years. However, as more studies were done, some research suggested that fetal cells helped fight or prevent the disease in the mother, rather than cause it. The true role they play remains unresolved.

One solid scleroderma clue is the disease's link to genetics. Although the incidence in the U.S. general population is only about 14 people per million per year, the odds of a person developing the condition are more than 100 times greater if a family member does too, and more than 280 times greater if that member is your identical twin. Those rates indicate a genetic component to the disease, albeit a weak one because the great majority of family members or twins don't share the disease. “If you compare scleroderma to a traditional genetic disease, the genetic pattern in families is not that strong,” says Xiaodong Zhou, a clinician who studies scleroderma genetics at the University of Texas Health Science Center in Houston. “But if you compare it to the [prevalence in] the general population, … it's very high.”

The Choctaw Nation of Oklahoma, a Native American population, has a much higher prevalence of scleroderma—roughly three times higher than the general population, which has led to several studies surveying their DNA. Researchers pinpointed several DNA markers around the gene for fibrillin-1 that correlate with the Choctaw scleroderma cases—and the protein's presence in connective tissue makes sense, given the disease's symptoms. Still, fibrillin's exact role in the condition remains murky. Researchers have failed to link the fibrillin gene to other populations of scleroderma patients.

Other genetic studies of scleroderma have pointed to the major histocompatibility complex, an array of genes that controls immune cell function, but “it's a common region for autoimmune disease,” explains Zhou. “They've all been linked to that region.”

More recently, Michael Whitfield, a geneticist with Dartmouth Medical School, has used genome-wide microarrays to screen all the gene activity within scleroderma skin and tissue samples. In July, Whitfield and colleagues reported in PLoS ONE that 17 out of 22 scleroderma patients had a genetic fingerprint, a distinct pattern of gene activity, differing from controls. Most of the overactive genes were the usual suspects for an auto-immune disorder—those involved in immune cell activation, including T, B, and macrophage cells—but others with altered activity were genes involved in fibrosis and collagen growth. Whitfield also found a cell proliferation signature—a group of cell-cycle genes that are expressed only when cells are dividing, which hints that scleroderma tissue has higher rates of DNA replication.

Whitfield says the most surprising outcome from the study is that even tissue from scleroderma patients that looks normal still has the distinctive genetic fingerprint of the disease. His group plans to repeat the experiment with a larger sample to see if it continues to hold true. If so, Whitfield hopes the multigene fingerprint could be turned into diagnostic or predictive tools for clinicians. “The limiting factor in terms of understanding scleroderma is getting large enough patient cohorts to do studies,” notes Whitfield. “And it's the heterogeneity that's really plagued us, from the molecular level and from the genetic level.”

Whitfield says that his microarray approach isn't likely to yield one simple gene as an answer. “There's very likely to be multiple factors contributing to scleroderma,” he says. “We're not going to find a single mutation. It's almost certainly going to be a combination.”

Treating the symptoms

Although figuring out the root of scleroderma remains a major research goal, the number-one priority has been to find treatments that keep the symptoms at bay, a quest that has led to a game of cat and mouse between drugs and the disease. The number-one killer of scleroderma patients used to be kidney failure. Twenty percent of patients would die in middle age as their kidneys' arteries became clogged and constricted with smooth muscle cells. However, in 1979, scientists found a class of drugs, angiotensin-converting enzyme (ACE) inhibitors, which relax the disease's stranglehold on the kidney. Harrison Farber, director of the Pulmonary Hypertension Center at BU, calls ACE-inhibitor therapy one of “the biggest breakthroughs in scleroderma,” noting that people are no longer dying of renal failure.

Lungs then became the new battleground. Like the kidneys, the pulmonary arteries, the vessels that carry blood between the heart and the lungs, also tend to become occluded and constricted in scleroderma. But much to the frustration and puzzlement of doctors, ACE inhibitors don't help here. “Once people stopped dying of kidney failure, they now lived long enough to die from pulmonary problems,” says Francesco Del Galdo, associate director of the Scleroderma Center at Jefferson University Hospitals in Philadelphia, Pennsylvania. One in seven scleroderma patients developed pulmonary hypertension, and it was inevitably deadly. By the late 1990s, however, researchers began employing a variety of drugs, including Viagra, which help to widen and relax the lung's arteries.

Although these treatments are touted as helping patients live longer, some clinicians remain unconvinced. Six different drugs have received approval from the U.S. Food and Drug Administration for treatment of lung problems in scleroderma, but all were largely tested using a simple approach: the 6-minute walk test. If a patient testing a new drug could complete this exercise with greater ease, the drug was considered to be effective. However, only one clinical trial has ever shown the drugs to improve a scleroderma patient's survival against a placebo. “These drugs can help the symptoms,” says Lafyatis. “But it's not clear that it helps their mortality; … some patients benefit from it and some don't.”

Scleroderma poses a further deadly challenge: fibrosis. In severe cases, fibroblasts, the cells that create and maintain the extracellular matrix, seem to be irreversibly activated, inducing an unwanted scarring process. Chris Denton, an experimental rheumatologist at University College London Medical School, notes that now “it's the fibrosis that causes the mortality of disease, because it's happening simultaneously in multiple organs.” Yet the lungs remain the key battleground, as lung fibrosis can strike 70% of scleroderma patients, and scarring there leads to eventual suffocation.

Scientists at the Cambridge meeting agreed that eliminating the fibrosis problem should prevent scleroderma from being a killer and turn it into a more manageable chronic disease. How to thwart the process has been mostly a mystery, but there was new optimism at the conference. “We have a rich variety of potential molecular targets [for fibrosis], and for many of these, inhibitors are available and are being tested in clinical trials,” says Oliver Distler, head of the scleroderma clinic at the University of Zurich, Switzerland.

Most of these inhibitors aim at a longtime suspect for fibrosis: TGF-β a so-called cytokine with multifarious pathways. One track stimulates both collagen and scar-tissue formation, whereas other tracks stimulate cell death via apoptosis or cell differentiation. Researchers have found elevated levels of the receptors for TGF-β on fibroblasts from scleroderma patients, suggesting that the activated form of the protein plays a role in their dysfunction. (When TGF-β is initially secreted, it's bound by other molecules and inactive.) But what triggers the production and activation of TGF-β in scleroderma remains unknown.

Moreover, because TGF-β's stimulation of scar and collagen tissue is normal and necessary in many cases, completely disabling the cytokine pathway is not a good option for treating scleroderma. “If you can't do fibrosis, you're in big trouble,” says Tyndall.

Scientists have therefore been teasing out how to selectively block the TGF-β route that elicits overactive scar formation. Some believe that targeting the upstream proteins that activate TGF-β may be the key to halting scleroderma's fibrosis while allowing normal wound healing to continue. One such TGF-β-inhibiting drug is Gleevec, which has previously earned fame for its successes treating leukemia and other cancers. In recent years, physicians have reported cures or near-cures of late-stage scleroderma patients after trying Gleevec, which has motivated a number of phase II clinical trials of the drug for the condition. Luke Evnin, chair of the board of directors at the Scleroderma Research Foundation in San Francisco, California, cautions that the results so far are by no means definitive. “In larger case series, it is not clear that [the drug] is broadly applicable,” he says. “But still the enthusiasm from the initial cases persists.”

Many scleroderma scientists remain wary of overhyping Gleevec. “We had promising drugs 10 years ago that were very hopeful,” says Distler. “But the clinical studies actually failed.” Those drugs, he notes, were “rather unspecific immunosuppressive drugs.” They had appeared to work in phase II studies, says Distler, but then failed in larger randomized controlled trials.

Tyndall believes that stem cell transplants could solve scleroderma's fibrosis and perhaps all its other symptoms, too. The approach mirrors a strategy used to treat leukemia: Stem cells from bone marrow that can give rise to new immune cells are taken out from a patient, and then physicians use drugs to purposely obliterate the patient's entire immune system, a strategy called immunoablation. After that, the preharvested stem cells are infused back into the patient, where they can create fresh bone marrow and, it is hoped, a new functioning immune system. Tyndall says some of his scleroderma patients went back to normal after receiving this aggressive immune rebooting. A multicenter randomized phase III trial with 150 patients is under way to confirm these findings.

Nonetheless, those at the conference in Cambridge seemed to reach a consensus that scleroderma would need more than just one solution. “No two scleroderma patients are alike. It's really amazing to me,” says Farber. Each patient may need a tailor-made treatment: One requires an aggressive antifibrotic treatment whereas another demands an emphasis on fighting pulmonary hypertension, Farber and others suggested.

Still, Denton is increasingly optimistic. “I think we're in a stronger position now, because in the previous era, we were trying therapies with an unclear view of what the biology is,” he says. “Now we're in an era where treatments are having an impact on outcome, … and we're also starting to understand the very complex biology that links the different processes in scleroderma.”

• * 10th International Workshop on Scleroderma Research, 2-6 August.

7. BIOMEDICAL RESEARCH

Paul Klee, a Tragic Metamorphosis

1. Lauren Cahoon

Art experts believe scleroderma had a major impact on modern artist Paul Klee, as his later works shifted from vibrant to darker colors and emphasized themes such as mortality and suffering.

The famous modern artist Paul Klee is widely believed to have suffered from, and died of, scleroderma. Late in Klee's life, for example, the disease caused his fingers to curl so much he had trouble holding a paintbrush. “All his medical records were destroyed, so it's conjecture. But it's very hard to attribute [his symptoms] to anything else,” says rheumatologist John Varga of Northwestern University in Chicago, Illinois, who in 2004 wrote a review of Klee's works in relation to scleroderma. Art experts believe the disease had a major impact on Klee, as his later works shifted from vibrant to darker colors and emphasized themes such as mortality and suffering (above, Klee's Tragic Metamorphosis, 1939). “His style and technique really did change and evolve,” says Varga. In one late work, Captive, Klee painted a grotesque self-portrait that includes representations of cagelike bars. Scleroderma patients often say they feel “imprisoned within their own bodies,” says Varga.

8. SOCIETY OF VERTEBRATE PALEONTOLOGY 68TH ANNUAL MEETING

Skulls Show Dinos Blew Their Horns

At the Society of Vertebrate Paleontology meeting, a group presented the most sophisticated evidence yet that the nasal passages within the crests of lambeosaurs were used for vocalizing, not smelling.

SOCIETY OF VERTEBRATE PALEONTOLOGY 68TH ANNUAL MEETING, 15-18 OCTOBER, CLEVELAND, OHIO

Nothing gets a paleontologist's speculative juices flowing like a strange piece of anatomy. Case in point: lambeosaurs, duck-billed dinosaurs (hadrosaurs) whose skulls sported hollow, bony crests connected to the animals' noses. Scientists have argued that the weird headgear was good for fighting, snorkeling, smelling, cooling the brain, or signaling to other lambeosaurs with loud, resonant honks.

At the meeting, a group presented the most sophisticated evidence yet that the nasal passages within the crests were indeed used for vocalizing, not smelling. Computed tomography scans of lambeosaur skulls revealed that the brains weren't geared toward olfaction but that the inner ears were attuned to the frequencies the crests most likely produced. “Honking still survives” as a hypothesis, says David Weishampel of Johns Hopkins University in Baltimore, Maryland, who studied vocalization in the lambeosaur Parasaurolophus in the early 1980s.

At 10 meters long and weighing in at some 3 metric tons, lambeosaurs would have been some of the larger animals in the swampy floodplains of western North America, Asia, and Europe toward the end of the dinosaur era. They roamed around, mostly on their hind legs, grabbing vegetation with their toothless bills, then grinding it to a pulp with hundreds of small teeth in the back of their mouths. Fossil trackways suggest that flat-headed hadrosaurs lived in herds, and lambeosaurs may have, too.

Some early ideas about lambeosaur crests proved short-lived. The crests are too thin and brittle to have served as effective weapons, and the physics of breathing underwater through them turned out to be unworkable. Among the more plausible theories, vocalizing was first proposed in 1931 by a Swedish scientist who likened lambeosaurs to trumpeter swans. In the 1960s, paleontologist John Ostrom floated the ideas that the long, looping chambers inside the crests could have functioned as air-cooled radiators or heightened the animals' sense of smell. James Hopson of the University of Chicago in Illinois suggested in 1975 that the crest evolved its large size for visual display to attract mates.

David Evans of the Royal Ontario Museum in Toronto, Canada, studied the olfactory system of lambeosaurs and the nerves associated with it, looking at impressions of these nerve pathways that remain in skull bones. His findings, published in Paleobiology in January 2006, suggested that only a small part of the nasal cavity within the crest was used for smelling. But no one had actually looked at the entire brain of a lambeosaur. Working with Lawrence Witmer of Ohio University in Athens and others, Evans used computed tomography to scan the skulls of four species of lambeosaurids that lived about 75 million years ago.

The olfactory region turned out to make up less than 2% of the lambeosaurs' brains. In contrast, a crestless hadrosaur called Edmontosaurus had at least double that, whereas the predatory dinosaur Tarbosaurus devoted 9% of its brain to olfaction. “When you put it all together, the smell hypothesis can be rejected,” Evans said.

Still, the crest does seem to be important. The elaborate nasal ductwork of the lambeosaurs points to a “strong selective pressure” for evolutionary adaptation, the team concludes. The inner ear, as revealed by the scans, suggests what advantage the odd organs might have offered. The clue is part of the cochlea called the basilar papilla. In living birds, studies have shown that its length correlates with the range of frequencies an animal can best hear. If the same relationship held in lambeosaurs, Evans and colleagues conclude, their optimal frequency in adults was 400 hertz, about the mid-range of a modern cornet. That's close to the frequencies an acoustic computer model has generated from the crest of Parasaurolophus. Hadrosaurs could have used their calls to attract mates, help keep the herd together, or warn one another of approaching predators, the researchers say. Other evidence from the skull suggests that hadrosaurs might have been particularly social, smart animals: Their cerebral hemispheres make up about 43% of the entire brain—more than in any group of dinosaurs except the small birdlike dromeosaurs, thought to be the brightest, most behaviorally complex ancient dinosaurs. The findings are in press at The Anatomical Record.

9. SOCIETY OF VERTEBRATE PALEONTOLOGY 68TH ANNUAL MEETING

Two Legs Good

1. Elizabeth Culotta

At the Society of Vertebrate Paleontology meeting, researchers presented a single bone, the thighbone of an ancient australopithecine from Galili, Ethiopia, that may add an interesting piece to the puzzle of how our famed ancestor "Lucy's" two-legged gait evolved.

SOCIETY OF VERTEBRATE PALEONTOLOGY 68TH ANNUAL MEETING, 15-18 OCTOBER, CLEVELAND, OHIO

Our famed ancestor “Lucy” walked upright in the grasslands of Ethiopia 3.2 million years ago. But what of her ancestors? Researchers have glimpsed only bits and pieces of even older hominins, the group that includes humans and our ancestors. At the meeting, Bence Viola of the University of Vienna presented a single bone, the thighbone of an ancient australopithecine from Galili, Ethiopia, that may add an interesting piece to the puzzle of how Lucy's two-legged gait evolved. “It's a window into a time when key evolutionary changes are happening—it's exactly what you'd want,” says J. Michael Plavcan of the University of Arkansas, Fayetteville, who was not involved in the work.

Although the femur was broken off at its lower end, its size suggests an owner slightly larger than the roughly 1-meter-tall Lucy, Viola says. Argon-argon dates produced just a few weeks ago place the thighbone between 4.38 million and 3.92 million years ago—significantly earlier than the most ancient member of Lucy's species, Australopithecus afarensis, which dates to 3.6 million years ago.

In the view of Viola and his Vienna colleague Horst Seidler, the bone is more primitive than Lucy's femur and resembles that of a much earlier hominin, Orrorin tugenensis, thought to be about 6 million years old. They suspect that it came from Au. anamensis, a species that lived about 4 million years ago and is widely considered to be Lucy's ancestor.

Details of the femur's anatomy, such as a long neck of bone leading to a large femoral head (the “ball” of the hip's ball and socket joint), suggest that its owner—whatever its name—was bipedal, Viola said. But other clues imply that it may also have climbed trees, he added. For example, a thick layer of dense cortical bone is evenly distributed around the femoral neck. In upright walkers like us, that cortical bone is unevenly distributed. “Both Orrorin and this femur seem to show several traits which indicate bipedalism but also retain signs of arboreal behavior,” Viola says. That suggests that our ancestors' move out of the trees was a long process.

But others aren't so sure that this single bone shows tree-climbing behavior. “It's not compelling enough to convince me of arboreality,” says Yohannes Haile-Selassie of the Cleveland Museum of Natural History in Ohio. Carol Ward of the University of Missouri, Columbia, is more impressed: The cortical bone distribution is “pretty suggestive,” she says. But she, too, would like to see more specimens, and the other end of the femur, to be sure. Stay tuned: Haile-Selassie's team is working on an unpublished partial hominid skeleton, also about 4 million years old, that may shed further light on how Lucy's ancestors walked (Science, 11 March 2005, p. 1545).

10. SOCIETY OF VERTEBRATE PALEONTOLOGY 68TH ANNUAL MEETING

Snapshots From the Meeting

Snapshots from the Society of Vertebrate Paleontology meeting include the earliest tree-climber, a dinosaur-cruncher, and evidence that dinosaurs were struck down in their prime and that dinosaurs of at least one species were picky eaters as children.

SOCIETY OF VERTEBRATE PALEONTOLOGY 68TH ANNUAL MEETING, 15-18 OCTOBER, CLEVELAND, OHIO

Earliest tree-climbers. A 255-million-year-old creature called Suminia getmanovi caused a stir a few years ago when researchers examining its skull discovered it had the first known adaptations for highly efficient chewing of tough vegetation. Now a look at the rest of the skeleton reveals another record: the earliest evidence of living in trees. Jörg Fröbisch and Robert Reisz of the University of Toronto, Mississauga, in Canada studied a large block of stone that preserved 15 mostly complete skeletons. “This block provides a wealth of new information about Suminia,” Fröbisch said at the meeting. The 50-centimeter-long animals had very long hands and feet—about 40% of the length of the limbs—and digits adapted for grasping. The new skeletons may be the most complete early therapsid (mammal-like reptile) yet found, so they could shed more light into the early evolution of the group, which flourished in the Permian and Triassic periods between 265 million and 225 million years ago. Kenneth Angielczyk of the Field Museum in Chicago, Illinois, notes that apart from a few burrowing species, most Permian therapsids were unspecialized four-legged herbivores. “The fact that [Suminia] was going in another direction is pretty cool,” he says.

Dinosaur-cruncher. Any way you look at it, the extinct crocodilian Deinosuchus was a terror. Its 1.5-meter-long skull was double the length found in modern crocs, and the whole animal may have stretched more than 10 meters. To experts on the Late Cretaceous of North America, Deinosuchus practically screams “top predator.” Now François Therrien of the Royal Tyrrell Museum of Palaeontology in Drumheller, Canada, and colleagues have buttressed that claim by estimating the strength of its bite. Taking five specimens from the United States, they compared the bending force of the jaw with that of the American alligator. The result: Deinosuchus could bite 13 times as powerfully as a modern alligator, nearly on a par with Tyrannosaurus rex. “Most assuredly Deinosuchus would have been a top predator of coastal environments,” Therrien concluded. Paleontologist Hans-Dieter Sues of the Smithsonian's National Museum of Natural History in Washington, D.C., agrees. “Could it pull apart a hadrosaur? The answer was a definite yes.”

Steady on. How much damage did that end-Cretaceous asteroid inflict, anyway? Researchers have long debated whether dinosaurs were already in decline before it hit, or whether they were struck down in their prime. A new analysis by Matthew Carrano of the Smithsonian's National Museum of Natural History in Washington, D.C., suggests the latter. Carrano took a close look at the scientific literature from North America, where the fossil record of Cretaceous dinosaurs is richest, and corrected for biases such as errors in classification and differences in how intensively paleontologists have searched for fossils in various places and times. The results showed that dinosaur diversity stayed steadily healthy through the late Cretaceous period, in North America at least. John Alroy of the University of California, Santa Barbara, sees the matter as settled. “It's the nail in the coffin” for the declining-dino scenario, he says.

Acquired taste. New skulls of juvenile Diplodocus, described at the meeting, suggest that young sauropod dinosaurs may have been pickier eaters than adults were. Stretching about 30 meters long, fully grown Diplodocus—massive four-legged plant eaters closely related to Apatosaurus—had remarkably square jaws crammed with narrow-crowned teeth in the front (see illustration). One idea is that these teeth helped them quickly crop vegetation to sate an appetite that presumably matched their enormous bulk.

The new juvenile skulls look different. Their snouts are narrow and rounded, with a full mouthful of teeth. “We believe it records an important change in diet,” John Whitlock, a Ph.D. student at the University of Michigan, Ann Arbor, told the audience. Whitlock and his adviser, Jeffrey Wilson, argue that the shape of the snout provides clues to the feeding behavior. A rounder snout correlates with more selective browsing, rather than indiscriminate noshing on vegetation.

Peter Dodson of the School of Veterinary Medicine at the University of Pennsylvania isn't convinced that the juvenile belongs to the same species as the known adults. But Kristi Curry-Rogers of Macalester College in St. Paul says that even if it doesn't, the new skull still shows that the genus Diplodocus was more varied and interesting than scientists had realized. “The fact that this juvenile with a rounded skull occurs in the midst of so many other square-snouted adults introduces a new view of what young members of this larger group looked like,” she notes.