# News this Week

Science  19 Jan 2007:
Vol. 315, Issue 5810, pp. 312
1. NEUROBIOLOGY

# A Surprising Connection Between Memory and Imagination

1. Greg Miller

People with amnesia struggle to remember their past. They may also struggle to envision their future, according to a new study. Researchers have found that people with amnesia caused by damage to the hippocampus, a brain region intimately tied to memory, have difficulty envisioning commonplace scenarios they might reasonably expect to encounter in the future.

The findings challenge long-held views about the function of the hippocampus and the nature of memory, says Lynn Nadel, a cognitive neuroscientist at the University of Arizona in Tucson. “The claim here is that the same system we use to remember the past we also use to construct possible futures,” says Nadel.

In the new study, published online this week by the Proceedings of the National Academy of Sciences, cognitive neuroscientist Eleanor Maguire of University College London and colleagues examined five amnesic patients. All of them had severe memory deficits caused by damage to the hippocampus; they had great difficulty forming new memories and recalling events that happened after their injuries. Ten healthy individuals who matched the patients' ages and education levels participated in the study as controls.

Maguire's team asked each subject to imagine and describe several ordinary experiences, such as meeting a friend or visiting a beach, a pub, or a market. The healthy subjects provided rich descriptions, remarking for example on the curve of a beach, the sound of waves hitting the shore, and the feel of burning hot sand. The amnesic patients were able to follow the researchers' instructions, but their descriptions were far less vivid. They described fewer objects, fewer sensory details such as sounds and smells, and fewer thoughts or emotions that might be evoked in the imagined scenario. The patients' responses on a questionnaire indicated that what they saw in their mind's eye were fragmented collections of images rather than coherent scenes.

The work suggests that the hippocampus may have a broader role in cognition than many researchers have thought, says Morris Moscovitch, a cognitive neuroscientist at the University of Toronto in Canada. The textbook view is that the main function of the hippocampus is to encode new memories, creating an initial memory trace that is eventually filed away to the cortex for long-term storage. In this view, the hippocampus is not needed to maintain or retrieve memories once they've been stored in the cortex. If this view were correct, Maguire says, the patients in her study, who did not have substantial damage to the cortex, should have been able to construct imaginary experiences by drawing on memories formed before their injuries. But their inability to integrate those memories into a coherent imagined scene suggests that the hippocampus does more than simply record current events.

Nadel, Moscovitch, and others have argued in recent years that the traditional view of the hippocampus's role is too narrow. Work from Moscovitch and colleagues, for example, suggests that the hippocampus binds together elements of remembered scenes to create vivid and coherent memories. Maguire's findings point to a similar role in constructing imagined scenes, Moscovitch says. “In order to have vivid constructions of the past, the future, or of imaginary events, you always need the hippocampus,” he says.

The idea that thinking about the past has much in common with thinking about the future has ancient roots, says Yadin Dudai, a neuroscientist at the Weizmann Institute of Science in Rehovot, Israel: “In prescientific times, many people thought that the role of memory is not necessarily to remember the past but to enable you to imagine the future.” In modern times, Dudai says, the notion was resurrected by the memory researcher Endel Tulving, who speculated that the ability to predict the future was a major driving force in the evolution of memory. Even so, Dudai says, only very recently have studies like Maguire's begun to provide experimental evidence that memory and imagination may share neural circuitry.

More evidence comes from a functional magnetic resonance imaging study now in press at Neuropsychologia. Cognitive neuroscientist Donna Addis and psychologist Daniel Schacter of Harvard University scanned the brains of healthy volunteers who had been asked either to recall a vivid memory or to envision a future experience. Both situations activated a similar network of brain regions, including the hippocampus, the researchers found.

If the hippocampus does turn out to be as important for imagination as it is for memory, that could have interesting implications for aging, Addis says. The hippocampus is one of the first brain regions to show signs of deterioration as we get older, and Addis has recently found evidence that the ability to envision future experiences declines in parallel with memory as people age. Meanwhile, Moscovitch is examining the work of famous artists and novelists to see whether the detail of their work declined in their later years. The picture of the hippocampus that's emerging suggests yet another compromise facing us in old age, he suggests: “Age will contribute wisdom because you can draw on a lot of past experience, but that experience may not be quite as rich as it used to be.”

2. SCIENTIFIC PUBLISHING

# Fossil Dealers Launch Research Journal

One of the biggest taboos in paleontology is publishing papers about fossils owned by private collectors. The problem is that these fossils are traded, and analyses can't be vetted if a new owner slams the door shut. Last week, a trade association of commercial fossil dealers launched an online journal to provide assurances of access to and documentation of the fossils. Predictably, the announcement has sent ripples through the academic community.

“This self-publishing of fossils in private hands will further undermine our science,” says Mark Goodwin of the University of California, Berkeley, one of many scientists who oppose the new venture. But other scientists are hoping that their collaboration will bring important skeletons out of the closet and raise the standards of commercial dealers.

The crux of the issue is access to specimens. All researchers agree that important, rare fossils must be available for study in perpetuity. A fossil in a private collection may end up lost to science if it deteriorates or is sold to a secretive buyer. Many academics also dislike commercial collecting because sloppy fieldwork can damage fossils and omit important contextual information.

For those reasons, the editorial policy of the Society of Vertebrate Paleontology (SVP) commits members to publishing only on fossils that are cataloged in public institutions. Nevertheless, researchers sometime find themselves casting a furtive eye over important but privately owned specimens. “We know about this stuff, but we can't say anything,” say Thomas Holtz of the University of Maryland, College Park, one of several academics who studied a Tyrannosaurus rex skull owned by a British millionaire while it was temporarily at the Carnegie Museum of Natural History in Pittsburgh, Pennsylvania.

The founders of the new publication, the Journal of Paleontological Sciences (JPS), say they want to help bring these fossils into the daylight. “It's a shame to ignore a fossil just because it is not in a public repository,” says commercial collector Mike Triebold, president of the Association of Applied Paleontological Sciences (AAPS), the trade group that is publishing JPS. Some specialized journals, mostly in Europe, already publish analyses of privately held specimens, but AAPS has laid down explicit guidelines to address the concerns of the scientific community and to encourage more responsible behavior among dealers and collectors. “They really are trying hard to do it right,” says Kenneth Carpenter of the Denver Museum of Nature and Science in Colorado, one of two academics on the nine-member editorial board.

The free journal will be published online quarterly, with plans to sell an annual print edition. Editorial board member Walter Stein, a dealer in Belle Fourche, South Dakota, says the goal is a self-sustaining journal with advertisements—but none for fossils. A pool of 20 academics will review submissions in which the owner has agreed to make the fossil freely available to researchers. The exact location of the excavation site must be registered with the journal, although that information will remain a trade secret for 25 years. Other documentation, such as photographs and replicas, must also be supplied.

Opponents say that's not good enough. The pledge of access by the current owner is no guarantee of future availability, says Goodwin: “The fact remains that the owner retains all rights and controls access to the specimen.” Photographs are no substitute, because they can't be measured accurately and can be faked, notes Hans-Dieter Sues of the National Museum of Natural History in Washington, D.C. “If you can't look at [a fossil] yourself, it might as well not exist.”

Some scientists worry that publishing research on privately owned specimens could remove the incentive for collectors to donate their fossils in return for having them studied and named after the discoverers. “In the worst case, the journal might be just a free propaganda platform for private specimens … and thus for raising their price,” says Reinhold Leinfelder, director of the Berlin Museum of Natural History. High prices lead to more thefts and illegal collecting, says Goodwin, adding that any involvement in the journal by academics “undermines the science of paleontology and borders on unethical conduct.”

Carpenter and others disagree, saying that the benefits of collaboration outweigh the risks. “Better to try to work with people and help guide them toward doing the right thing for science rather than shove them away and say, ‘We won't work with you unless you do everything we want,’” Holtz says. “That hasn't worked.”

SVP isn't likely to budge on permanent, guaranteed access, says society vice president Blaire Van Valkenburgh of the University of California, Los Angeles. “You have to stick tightly to your guns when you're doing science.”

3. MOLECULAR BIOLOGY

# Trafficking Protein Suspected in Alzheimer's Disease

1. Jean Marx

Traffic control is every bit as important for our cells as it is for our cities. A protein that ends up in the wrong cellular location can cause as much trouble as, say, a car crash in midtown Manhattan. Indeed, growing evidence suggests that improper protein trafficking contributes to the development of Alzheimer's disease (AD) by fostering the production of abnormal amyloid deposits, a key pathological feature of the mind-robbing disease. A new genetic study now gives a boost to the idea that misdirected protein transport contributes to the development of AD, particularly in older people.

In a Nature Genetics paper published online on 14 January, a multi-institutional team linked a gene called SORL1 to the late-onset form of AD. “This gene is robustly associated with an increased risk of Alzheimer's disease in several groups,” the group's co-leader Peter St. George-Hyslop of the University of Toronto in Canada said at a press conference last week. (The other co-leaders are Lindsay Farrer of Boston University School of Medicine, Richard Mayeux of Columbia University's College of Physicians and Surgeons in New York City, and Steven Younkin of the Mayo Clinic in Jacksonville, Florida.)

The protein made by the gene, also known as SORLA or LR11, is thought to be involved in regulating protein movements through the cell. The Nature Genetics results, combined with recent findings from other groups, suggest that mutations in SORL1 lead to a decrease in its protein product, which in turn increases the risk of developing the disorder. When the SORL1 protein is lacking, St. George-Hyslop says, the so-called amyloid precursor protein (APP) is trafficked off to compartments in the cell that contain enzymes that snip out and release the small and highly toxic protein fragment known as β amyloid. Accumulation of this protein is thought to underlie brain neuron degeneration. If confirmed, the new findings should clarify the causes of AD and point to better ways of identifying—and possibly treating—people at risk of developing the disease.

Other genes have been linked to AD. Mutations in three of these, APP itself and presenilin 1 and -2, cause the early-onset form of the disease that strikes people in their 40s, 50s, and 60s. Most AD cases occur later in life, after age 70, however, and getting a handle on the genes that contribute to these late-onset cases has been difficult. Although there are many candidate genes, only one has been firmly established, the epsi lon 4 variant of the APOE gene, involved in perhaps 30% of the cases.

To try to find other genes for late-onset AD, about 4 years ago, St. George-Hyslop, Mayeux, Farrer, Younkin, and their colleagues embarked on a large collaborative study involving some 41 co-authors located at 14 institutions. Because most β amyloid is produced in endosomes, membrane-bound vacuoles that transport proteins from the outer cell membrane where APP is normally located into the cell interior, the researchers decided to focus on seven genes, including SORL1, involved in protein shuttling in the endosomes.

They looked for associations between AD and gene variations called single-nucleotide polymorphisms (SNPs) in people of varying ethnic backgrounds—about 6800 total, divided between AD patients and unaffected controls. Only SNPs in SORL1 associated with the disease. These are only markers, however. The true disease-causing mutations have not yet been identified. Until they are, Mayeux says, the researchers can't get a good fix on how much they contribute to AD, but he estimates they may be involved in 10% to 20% of the cases.

Researchers do have some good ideas, however, about how SORL1 alterations could lead to AD. About 2 years ago, James Lah and colleagues at Emory University in Atlanta, Georgia, found that brain tissue from AD patients contains less SORL1 protein than do brains of unaffected individuals. Since then, the Emory team and also that of Thomas Willnow at the Max Delbrück Center for Molecular Medicine in Berlin, Germany, have been building a case that SORL1 protects against AD by directing APP away from the late endosomes, where it will be clipped into β amyloid.

They've found, for example, that increasing SORL1 expression in neurons decreases β-amyloid production, but decreasing the gene's expression increases β-amyloid production. The SORL1 genetics group reports similar findings in their Nature Genetics paper. “The biology is pretty compelling” in favor of the idea that a SORL1 deficiency can increase AD susceptibility, says AD geneticist Rudolph Tanzi of Harvard's Massachusetts General Hospital in Boston.

A genetic linkage between the SORL1 pathway and AD would nail down those findings. “That would definitely implicate the pathway,” says another AD researcher, Sam Sisodia of the University of Chicago in Illinois. Still, he and others, including St. George-Hyslop and the other team members, say that the results need to be confirmed. In unpublished work described at meetings, Tanzi and his colleagues found a weak link between SORL1 and early-onset AD but no link to the late-onset form.

Such discrepancies are common in attempts to find genes linked to diseases that have multiple causes. But as Farrer pointed out at the press conference, finding the linkage in four distinct ethnic groups—persons of European descent, African Americans, Caribbean Hispanics, and Israeli Arabs—gives them confidence. That, Farrer says, “was unexpected and highly unlikely to be due to chance.” Tanzi, too, is optimistic that the result will hold up. “When everything is said and done,” he predicts, “I think [SORL1] will be added to the list” of AD genes.

4. STEM CELL DEBATE

# Scientists Protest 'Misrepresentation' as Senate Vote Looms

1. Constance Holden

Several leading scientists are charging that the White House has misrepresented their research in an attempt to influence the ongoing stem cell debate in Congress.

On 11 January, the U.S. House of Representatives voted overwhelmingly to expand the number of human embryonic stem (ES) cell lines available to federally funded researchers. The bill, designated H.R. 3 and considered a top priority in the new Democrat-controlled Congress, passed 253 to 174—a significant jump in support from May 2005 when the same bill passed by 238 to 194. But it still falls more than 30 votes short of the two-thirds needed to override a presidential veto.

With the Senate expected to vote on the same measure next month, the stage is set for a replay of the veto dealt by President George W. Bush last July. The White House has rebuffed attempts by the bill's sponsors to meet with the president, and it's fighting hard to cast the debate in its own terms.

As part of that effort, the White House Domestic Policy Council issued a new report on 10 January to promote methods of getting stem cells that don't harm embryos. The report, Advancing Stem Cell Science Without Destroying Human Life, suggests that a variety of “non-embryo-destructive” approaches may prove capable of creating cell lines with all the potential of ES cell lines. It repeatedly mentions a new study by scientists at Wake Forest University in Winston-Salem, North Carolina, who reported in the January issue of Nature Biotechnology that stem cells found in human amniotic fluid have many of the same qualities as ES cells (Science, 12 January, p. 170). The White House report also touts adult stem cells, saying some “may be pluripotent,” and suggests that these could be adequate to treat Parkinson's disease and other conditions for which ES cells have been held up as the great hope. Also put forth are potential alternative sources for pluripotent ES-like cells, advanced in the past by the President's Council on Bioethics, which include the possibility of getting viable cells from dead embryos or through somatic cell “de-differentiation.”

Scientists have already reacted strongly to some of the material in the report. Three Harvard stem cell researchers—Kevin Eggan, Chad Cowan, and Douglas Melton—wrote to the sponsors of H.R. 3 complaining of the “clear misrepresentation of our work” in the document, which heavily cites Eggan and Cowan. “We are surprised to see our work on reprogramming adult stem cells used to support arguments that research involving human embryonic stem cells is unnecessary,” they wrote. “Our work directly involves the use of human embryonic stem cells … [and] is precisely the type of research that is currently being harmed by” the president's policy.

Anthony Atala, lead author of the amniotic cell paper, also feels that his work is being misinterpreted. Opponents of H.R. 3 have seized upon the report, which appeared on the eve of the House debate, and are citing it as further evidence that noncontroversial cell types can substitute for ES cells. Atala wrote a letter to the bill's sponsors emphasizing that that is not the case.

More friction may be in store: According to The Wall Street Journal, presidential aides are drafting a possible executive order favoring alternative sources, although a White House spokesperson says they have nothing to announce at present.

The focus of the debate now turns to the Senate, which passed the same bill (now labeled S. 5) in the last Congress by a vote of 63 to 37. Many estimates put the count at 66 in favor—one vote short of a veto override. But the bill's advocates think there might be a chance of avoiding a veto, because Senate rules will allow for amendments. Certain changes could make the bill more palatable to the president—such as adding provisions for more ethical oversight; a program to promote embryo adoption; or even a new, later deadline for cells that are eligible for federal funding. (Currently, cells have to have been derived by 9 August 2001 to qualify.)

Still, many see another veto as the likely outcome. But stem cell advocates are convinced that public opinion is increasingly on their side. “This is an issue that will not go away,” says one of the bill's sponsors, Representative Diana DeGette (D-CO). Until it becomes law, “we intend to introduce it over and over.”

5. REGULATORY SCIENCE

# Panel Pans Proposed Change in U.S. Risk Assessment

1. Daniel Charles*
1. Daniel Charles is a freelance writer in Washington, D.C.

Government regulators and toxicologists with private industry are assessing the impact of an unusual rebuke last week by an expert panel of a White House proposal to change how the U.S. government practices risk assessment. The expert panel, which called the proposal “fundamentally flawed,” urged the White House to focus on “goals and general principles” and to leave the details of risk assessment to agencies with more expertise.

The proposal, issued in January 2006 by the White House Office of Management and Budget (OMB), laid out technical guidelines for estimating risks posed by anything from toxic chemicals to large engineering projects. Among other things, the guidelines called on agencies to calculate and disclose the uncertainties surrounding their risk estimates. Agencies shouldn't just create a “worst-case” estimate, it said, but also prepare a “central or expected estimate.”

That approach is often unrealistic, says a panel of the National Academies' National Research Council (NRC) asked to review the proposed guidelines. A major reason it's not feasible, say public health officials, is because the data required to calculate such values may not exist. “The OMB is looking for a Cadillac of risk assessment,” says Gary Ginsberg, a toxicologist with the Connecticut Department of Public Health. “A lot of the time, we're lucky if we have something that moves.”

The dispute highlights one of the most difficult issues in risk assessment: what to do when there are no data showing, for instance, how many people became ill after exposure to a particular chemical. “For a few chemicals, you can do this,” says Joseph Rodricks, a former U.S. Food and Drug Administration official now with the consulting firm ENVIRON International Corp. in Arlington, Virginia. “But most of the time, you can't.”

Rodricks points to the example of acrylamide, a chemical used to manufacture dyes. Animals exposed to this chemical have developed cancer, but studies of workers at factories where acrylamide is used haven't found similar effects. With no solid data from humans to go on, the Environmental Protection Agency (EPA) assumes that animal studies do reflect human biology and, as a result, regulates acrylamide as a substance that can cause cancer.

Such assumptions play a critical role in risk assessment, says Rodricks, a member of the NRC review panel. “You have to do something rather than nothing,” he says. The OMB's proposed guidelines, however, didn't provide any guidance to agencies on how to cope with an absence of data.

Many critics, including industry groups, have faulted EPA for sometimes relying on worst-case assumptions in reviewing substances such as perchlorate, a chemical used in rocket fuel that's been found in drinking water supplies. “I think the OMB was trying to push agencies in the direction of being perhaps less conservative, or more realistic, about risks,” says Rodricks.

But the actual result of implementing OMB's demand for more and better data, according to some critics of the proposal, would be regulatory paralysis. “It would give industry lots of tools for slowing down the process,” says Renee Sharp, an analyst in the Oakland, California, off ice of the Environmental Working Group.

In response to the NRC's report, the White House announced last week that it will reconsider its proposal. “We will not finalize it in its current form,” said Steven Aitken, acting administrator of the OMB office that issued the guidelines.

In the future, many hope that new scientific advances will begin to fill in the data gaps about effects on human health. Among the new techniques are detailed studies of the biology of human cells, or examination of chemical effects on arrays of DNA. NRC has convened another expert panel to evaluate how these and other techniques might improve EPA's risk analysis.

Such new approaches can raise as many questions as they answer, Rodricks admits. “You can find all kinds of effects to chemical exposures” once you start looking for them, he says. The hard part is translating these observations into useful guidance on the chemical's potential harm to humans.

6. SCIENTIFIC MISCONDUCT

# Former Hwang Colleague Faked Monkey Data, U.S. Says

1. Constance Holden

Another Korean cloning repercussion sounded last week: A former member of Woo Suk Hwang's research team has been sanctioned by the Off ice of Research Integrity (ORI) of the U.S. Public Health Service for faking figures in a paper on monkey cloning being prepared at the University of Pittsburgh in Pennsylvania.

Jong Hyuk Park, who was a postdoc in the lab of Pittsburgh researcher Gerald Schatten, has been barred for 3 years from any relationships with U.S. agencies. The university began investigating the monkey work in January 2006, after Schatten alerted officials to possible irregularities, and the paper was never submitted.

Schatten has not talked to the press for more than a year, but university spokesperson Lisa Rossi says that, despite the fraud, the research team is confident that it succeeded in cloning rhesus monkey embryos and generating pluripotent embryonic stem (ES) cell lines from them. The team is repeating the experiments before submitting the paper. Don Wolf of the Oregon National Primate Research Center in Beaverton says he has “little doubt that it is possible” that Schatten has derived ES cells from cloned monkey embryos, as Schatten reported at a meeting in Toronto last March, but says he will “withhold judgment.”

Park was a postdoc at Magee-Women's Research Institute in Pittsburgh from August 2004 to February 2006. The university concluded last April that he had committed misconduct and referred the case to ORI, which published its finding in the Federal Register on 9 January. According to the ORI notice, Park “repeatedly misrepresented” the accuracy of one of the figures in the paper to the Pittsburgh investigative panel, “presented false figures as true,” and “falsified the record of revisions of the figures by deleting all prior versions from the laboratory server.” ORI says the research, funded by the National Institutes of Health, was to be submitted to Nature in a paper entitled “Rhesus Embryonic Stem Cells Established by Nuclear Transfer: Tetraploid ESCs Differ from Fertilized Ones.”

John Dahlberg, director of ORI's Division of Investigative Oversight, says the Pittsburgh investigation verified that three pluripotent lines of rhesus ES cells were in fact generated. But Park used photographs from one of them as representing all three. “This is not a huge case of misconduct, to be honest; much of the damage was in the cover-up,” says Dahlberg.

Park, before coming to Pittsburgh, was on Hwang's team in Seoul, where he was a co-author of two papers, published in Science in 2004 and 2005, that were later retracted. A 9 January statement from Pittsburgh says “other papers co-authored by Dr. Park also have been retracted,” but no information was available on what they were.

The episode is another blow to Schatten, whose collaboration with Hwang led to his being found guilty of “research misbehavior” by his university in February 2006. Although the panel found no evidence that Schatten falsified data or was aware of any fraud, it said he failed to exercise “a sufficiently critical perspective” in ensuring that the 2005 Science paper was sound (Science, 17 February 2006, p. 928).

Although Korean prosecutors said Park was involved in fabrication of data for the 2004 Hwang paper, he was not one of the six researchers indicted on misconduct charges last May in Seoul. ORI does not know the current whereabouts of Park, who returned to Seoul to talk to prosecutors last February.

7. ASTROBIOLOGY

# Astrobiology Fights for Its Life

1. Andrew Lawler

A decade after NASA pledged to create a robust program to find and understand life in the universe, researchers face a debilitating budget crunch and skepticism within their own agency

These should be heady times for astrobiologists. Reports of recent liquid water on Mars and organic matter in the far reaches of the solar system signal that the fledgling discipline, which seeks to understand the nature of life in the universe, is coming of age. Add an expanding roster of newly discovered extrasolar planets and examples of life flourishing in extreme environments on Earth—amid the high ultraviolet of the Andes, in Australian radioactive springs, and in granite formations deep underground—and the research challenges seem boundless. “The field is not only promising,” says Steven D'Hondt, an oceanographer at the University of Rhode Island, Kingston, who studies microbial life deep in ocean sediment. “It is productive and extremely successful.”

But don't ask D'Hondt how astrobiology is faring in his lab. He is turning away prospective graduate students because his support from NASA has dried up. D'Hondt's colleagues have similar tales to tell. They are scrambling to find funds from other sources to cope with a 50% cut over the past 2 years in NASA's support for astrobiology.

“We're in dark times now for astrobiology,” says Michael Meyer, a former senior scientist for astrobiology and now NASA's lead scientist on Mars exploration. Researchers are afraid that the field may go the way of the agency's life and materials science effort, a once-robust $1 billion program now virtually extinct as more pressing needs in the human space flight program have siphoned off funds. Those fears grew stronger last summer when NASA Administrator Michael Griffin told the Mars Society that astrobiology is marginal to the agency's mission. The fiscal downturn has meant staff cuts at the program's centerpiece, the decade-old NASA Astrobiology Institute (NAI) at NASA's Ames Research Center, on the edge of Silicon Valley in Mountain View, California, and less money for the outside scientists it supports. New, politically savvy leaders in place at Ames, NAI, and NASA headquarters have high hopes that they can make a better case for astrobiology within the space agency and outside. And the new Congress, which includes a more powerful California delegation, is expected to go to bat for the field in upcoming budget battles with the White House. “We're going to emerge from this in an even stronger position,” insists Carl Pilcher, the new NAI director. But others aren't so sanguine. “I feel a pang in my stomach,” says Kenneth Nealson, a biologist at the University of Wisconsin, Madison. “Survival is going to be tough.” ## Life mission NASA has spent a half-century looking for life beyond Earth. For most of that time, however, the exercise was an afterthought to the agency's main focus on space exploration. That modest effort underwent a dramatic change in the mid-1990s, the same time NASA's sprawling Ames center—founded on the eve of World War II to promote aeronautical research—appeared to be on the verge of closure. A team of senior NASA officials proposed a makeover for Ames that would draw upon its existing small programs in exobiology, the life sciences, and computing to focus on two core missions: computing and what was termed astrobiology. The idea wowed then-NASA chief Daniel Goldin, who was eager to link up with the exploding biological revolution. The makeover got a boost when scientists announced in 1996 that they had detected evidence of fossilized life in a martian meteorite found in Antarctica. With the backing of then-Vice President Al Gore, Goldin folded NAI and astrobiology funding as a whole into a larger package that included several ambitious Mars missions. He forecast a$100-million-a-year budget line for astrobiology that would help biologists, astronomers, geologists, chemists, and other researchers probe the nature of life on Earth and throughout the universe. His vision had its critics—particularly among biologists—who groused that NASA was using the hype over the martian meteorite to jump on the biology bandwagon. But those concerns were given little credence by a White House, Congress, and NASA leadership intent on pursuing a field that had captured the public's imagination.

Befitting its nontraditional subject, the new NAI was designed to be a nontraditional institute. Its research, done by collaborative teams from universities outside the institute, would focus on the existence of habitable planets and moons within the solar system, the origins of life on early Earth, the limits of terrestrial life, and signatures of life on extrasolar planets. Instead of having a large staff housed in one location, NAI would function as a “virtual” institute, employing a few civil servants who would grow a cadre of experienced scientists working arm in arm with NASA engineers to plan future missions. “We knew that if we didn't keep a connection to missions, this thing would look like NSF [the National Science Foundation], and someone would challenge its existence,” says physicist G. Scott Hubbard, NAI's founding director and later chief of Ames itself. He and others envisioned NAI pursuing basic research while also taking a lead role in proposing astrobiology instruments that could fly on spacecraft.

In 1998, after a stiff competition, NAI picked 11 university teams to receive approximately $1 million annually for 5 years. The next year, Hubbard was replaced by Baruch Blumberg, a Nobelist in medicine from Fox Chase Cancer Center in Philadelphia, Pennsylvania, who played a critical role in developing a vaccine for hepatitis B. The appointment satisfied Goldin's demand to Hubbard to find “a King Kong biologist” who would provide NAI with instant biological gravitas. But that gravitas came at a price: Blumberg had no experience with NASA or space projects. “My understanding was that this was to be a basic science institute, and the teams were selected and funded on that basis,” Blumberg recalls. “Scientists were given broad direction” to pursue a host of topics. And so they did, with projects on everything from chemical evolution in the interstellar medium to biosynthetic pathways in living cells. The work was intended to lay the foundation for seeking signatures of life, says Blumberg, an approach that he calls “very mission-oriented.” When Blumberg returned to Fox Chase in 2002, astrobiology at NASA appeared to be thriving. The institute's original$4 million annual budget had grown to $25 million, the number of NAI teams stood at 16, and some 15% of the 150 senior scientists on those teams were members of the U.S. National Academy of Sciences. NASA was also spending considerable sums on technologies to monitor life on other planets and for traditional exobiology, which focuses on prebiotic conditions for life in the universe. But astrobiology's apparent good health proved illusory. The gulf between NAI and the engineers who traditionally run NASA began to widen as the institute's work diverged from the agency's mission pipeline. “Engineers know what they're going to build, while basic scientists don't know what they are going to find,” acknowledges Blumberg. Although many scientists associated with the institute worked on missions to Mars and other astrobiology-related flight projects, NAI and its team members did not have a seat at the table when the relevant instruments were chosen. Without their participation, NAI had little direct impact on planning missions. And missions are at the heart of NASA's reason for being. Several university scientists and NASA insiders give Blumberg credit for establishing an excellent research program, but they believe that his failure to pursue a solid role in future missions became the institute's Achilles' heel. Others note that Blumberg and his successor labored under tough constraints. Spacecraft projects can take a decade or two to complete, and the cost and technology needed to build specific instruments far exceed the means of a$60-million-a-year institute. Such instruments typically require the technical and scientific muscle of a NASA center, aerospace corporation, or large research university. Having its roster of scientists on competitive, 5-year grants hampers long-range planning. And government regulations preclude the teams from bidding on new NASA projects because they are already collaborating with the agency. Their diversity would also make it virtually impossible to settle on a single instrument, notes Bruce Runnegar, a University of California, Los Angeles, paleontologist who served as a team member and then succeeded Blumberg.

Over time, NAI's portfolio veered noticeably toward the study of extremophiles. That unintended shift—driven perhaps by the high quality of proposals it received in that area, say scientists—was a boon to researchers, who canvassed the globe to examine life that could live off radioactivity from rocks in deep mines, metabolize in subzero temperatures, or thrive in highly alkaline or highly acidic environments. “If you are looking for life on Mars, Europa, or the outer planets, you have to look at other kinds of energy sources,” says Andrew Knoll, a Harvard University paleobiologist who led one of the teams whose work was not renewed. That makes work on organisms in mines on Earth relevant to the search for subsurface life on Mars, he notes.

But the emphasis on extremophiles also widened the gap between the institute and NASA's core mission. NASA continued to develop, launch, and gather data on missions to Mars, asteroids, Jupiter, and Saturn that provided exciting data on the existence of water and other conditions that might be favorable to life elsewhere. At the same time, astronomers using both space-based and ground-based telescopes detected extrasolar planets with increasing frequency.

However, none of these missions—most of which were well under way when the institute was formed—include specific instruments designed to test for life. That makes it hard to judge NAI's impact over the past decade. “What credit can the NAI take? I don't have a good quantitative answer,” says Bruce Jakosky, a planetary scientist at the University of Colorado, Boulder, and longtime advocate of the field. And with the exception of a contribution to the future Mars Science Laboratory slated for a 2009 launch, the NASA astrobiology effort is not directly involved in upcoming missions. The Terrestrial Planet Finder, a good candidate for picking up biological signals from extrasolar planets, has been put on indefinite hold, as has a proposed Astrobiology Field Lab to Mars that could probe beneath the planet's surface for hidden microbes.

Its ambiguous contributions make astrobiology tremendously vulnerable as NASA attempts to finish the space station, build a new launcher, and set up a base for humans on the moon—all without significant budget increases. Whereas space and earth sciences have formidable political allies, astrobiology so far has proved too small, too scattered, and too new to fight off budget threats. Griffin has proposed cutting astrobiology funding in 2007 to half of its 2005 level, and NAI has repeatedly delayed its next team competition. Without a new round of winners, there will be no teams left by 2008.

That decline runs counter to the conclusions of a May report from the National Academies' National Research Council that called astrobiology “an outstanding example of the development of a successful new interdisciplinary area” and recommended continued robust funding. However, Griffin says that it's not his job to nurture a fledgling field that won't help him put humans on the moon. Asked at an August meeting of the Mars Society about the impact of the cuts on astrobiology students, he retorted that “if they want to work for government money, they must look at what the government wants—not what they think it should want.”

## New life

Despite Griffin's skepticism, some scientists expect astrobiology to survive and prosper. Last fall, John Rummel took over as astrobiology chief at NASA headquarters. A biologist with a strong affinity for the human space program, Rummel is a respected agency insider. At the same time, Pilcher, a longtime NASA headquarters official, took over as the institute's fourth director. And his boss is Simon P. “Pete” Worden (see following story), who has big plans for Ames.

Rummel and Pilcher confront a worried batch of researchers as well as a shrinking pool of graduate students. “Plenty of people are getting fed up with the lack of proposals funded,” says Kevin Hand, a graduate student at Stanford University in Palo Alto, California. “People are doing other things ancillary to astrobiology,” he notes, while they wait for NASA to pump more money back into the effort. And some researchers such as Nealson are skeptical that the program can be redirected to make it more relevant to exploration-focused NASA.

There is a chance Congress may come to the rescue. Whereas Republican legislators regularly defended U.S. President George W. Bush's push for a new launcher and human exploration of the moon, Democrats have spoken out against raiding the science budget to pay for those projects. And some members of the overwhelmingly Democratic California delegation—including Representative Anna Eshoo (D-CA), who represents the area around Ames and is a close ally of new House Speaker Nancy Pelosi (D-CA)—are aware and concerned about the fate of astrobiology.

In the meantime, scientists soldier on. Thanks to an NSF grant, D'Hondt traveled to the South Pacific last month to study deep-sea microbes. But he is worried that the NASA cuts may inflict long-term damage to the field. “We won't be able to produce the scientists needed for future space missions,” he warns.

Not everyone is so pessimistic. Even if the institute becomes a victim of the current budgetary storm, many scientists think that the field will survive. “Given the incredible nature of the questions posed by astrobiology,” says Hand, “I'd be doing this if I had to pick up dimes from the street.”

8. ASTROBIOLOGY

# It Rains in Spain and Wilts in Australia

1. Andrew Lawler

In a gleaming steel-and-glass building in Madrid surrounded by manicured lawns, dozens of Spanish researchers are probing the hows, wheres, and whys of life in the universe. With an annual budget approaching $3 million, Spain's Centro de Astrobiología is thriving. Interdisciplinary teams of scientists, engineers, and lab technicians are working on a sophisticated laser that could sample elements in martian soil as early as 2013, developing a drill system for detecting organisms under the Red Planet's surface later in that decade, and simulating in their labs Earth's early conditions. Each project includes European or American researchers from a host of institutions. The center's happy buzz of activity stands in stark contrast to the angst felt by its U.S. cousin, the NASA Astrobiology Institute (NAI) (see main text). Ironically, the creation of NAI spurred Spanish researchers to start their own institute in 1999. Construction grants from the European Union and the Spanish government provided a$20 million showcase for astrobiology, and the defense and education and science ministries—along with the regional government of Madrid—supply a steady stream of operating funds. “We are independent and have sufficient funds,” says Director Juan-Pérez Mercader, ticking off a long list of ongoing projects. Although its budget and staff are a fraction of the size of NAI and its university teams, the Madrid center is linked both to basic research in the lab and to specific European Space Agency missions, such as the ExoMars orbiter and rover planned for launch in 2013.

The same cannot be said for the 4-year-old Australian Centre for Astrobiology at Macquarie University in Sydney. The Australian institute also was a beneficiary of the U.S. decision to develop the field. “We had a lot of encouragement from NAI and [its director] Baruch Blumberg,” says Director Malcolm Walter. With five scientists, 10 graduate students, and an annual budget of $1 million, the Australian center focuses primarily on extremophile research. But a proposed 40% cut in government funding, which Walter sees as a direct result of the NASA cuts, is likely to mean lay-offs. “When the U.S. sneezes, we get a cold,” he says. Some U.S. scientists worry that the Europeans are moving into astrobiology's driver seat. “The center of gravity will shift to Europe, and we'll lose leadership,” predicts Lynn Rothschild, a biologist at NASA Ames Research Center in Mountain View, California. Indeed, younger researchers may want to book a flight to Madrid. Spain's center has just announced plans for a graduate student program, and as Mercader makes clear, “it is open to anyone who wants to come.” 9. ASTROBIOLOGY # Pete Worden 'Ames' for the Moon and Beyond 1. Andrew Lawler The new head of Ames Research Center hopes to sell NASA on some of his unconventional ideas To Simon P. “Pete” Worden, NASA's Ames Research Center in Silicon Valley seemed like the perfect beachhead from which to launch a retrograde campaign for a new generation of smaller, cheaper, faster scientific spacecraft. But the maverick astronomer and retired U.S. Air Force general had barely arrived as the center's director last May when he encountered unexpected fire. The first blow was the transfer of responsibility for developing lunar robotic orbiters and landers—the center's key piece in U.S. President George W. Bush's human exploration effort announced 3 years ago—from Ames to Marshall Space Flight Center in Huntsville, Alabama. Weeks later, Ames lost another project when NASA headquarters decided that the rival Dryden Flight Center in southern California was better able to hold down the cost of readying the Stratospheric Observatory for Infrared Astronomy for flight later this decade. By the end of the summer, Worden's superiors shot down his bold proposal to incorporate smaller and cheaper probes into the fleet set to explore the lunar surface early in the next decade. Those three early setbacks haven't deterred Worden, a self-proclaimed NASA basher who jokes that the agency's initials stand for “Never a Straight Answer.” Instead, Worden remains bent on radical changes for the troubled lab. Ames and its famous neighbor, Google, last month agreed to an innovative technology-sharing deal that will make NASA's enormous archives of Earth and space data accessible to the public. The deal could pave the way for Google Moon to join Google Earth and Google Mars. And Worden hasn't given up on smaller, faster, and cheaper: He has wrangled$10 million from his bosses to begin thinking about small spacecraft that could journey to asteroids and the outer solar system as well as the moon. He hopes to scale up the program once there's more money for such activities. In the meantime, he's pursuing contracts from other federal agencies to help the center's 2500-strong workforce weather the current NASA budget crisis.

True mavericks are rare among the government's colorless cadres of generals and civilian bureaucrats. But the 57-year-old Worden, who earned an astronomy doctorate from the University of Arizona, Tucson, has a history of bucking conventional wisdom regardless of its effect on his career. In the 1980s, he was an early advocate of President Ronald Reagan's Strategic Defense Initiative, an unpopular stance that earned him the sobriquet of Darth Vader in space circles. As a White House staffer under Reagan's successor, President George H. W. Bush, Worden helped engineer the ouster of NASA chief Richard Truly and his replacement by Daniel Goldin, who touted the smaller, faster, cheaper approach. He then led a tight-knit group of Defense Department officials that applied the philosophy to the successful 1994 Clementine mission to the moon, finding hints of ice at the lunar poles and thoroughly embarrassing NASA and its fleet of large, costly spacecraft.

After the 9/11 terrorist attacks, Worden did a brief and controversial stint as chief of the Pentagon's Office of Strategic Influence, set up to place stories favorable to the United States in foreign media and on the Internet. But then-Defense Secretary Donald Rumsfeld shut it down after the office came to be seen as simply a propaganda vehicle for the Bush Administration.

Worden's unconventional ideas often make his superiors nervous—he served for more than a decade as a full colonel before winning his f irst star. The debacle with Rumsfeld squashed further chances for promotions, so after working briefly for Senator Sam Brownback (R-KS), Worden left the military in 2004 to join the University of Arizona as a research professor. He lost out to his less-controversial civilian friend Michael Griffin when Sean O'Keefe stepped down as NASA administrator. As for the Ames appointment, mutual acquaintances say Griffin is eager for Worden's help in promoting the president's new exploration vision but chose to keep him far outside the fishbowl of Washington politics.

Worden spoke recently with Science about his setbacks, plans, and vision for the center.

## On budget cuts to life sciences and astrobiology

The agency has been given certain priorities and missions by Congress and the president. Astrobiology—not that it isn't superb science—has a lower priority. But there is non-NASA funding—the private sector, other government agencies—and we are aggressively pursuing those options. Is it easy? No. It is much like what happens at a university. I spent the last 2 years as research faculty at Arizona. I didn't have a tenured position, and you did the work you needed to do. I'm a scientist. If I were king, I'd double the science budgets. I think scientific exploration of the solar system and the universe is really exciting and a key area of our future. I'd love to spend two-thirds of the defense budget on science if I could get away with it.

## On tension between Griffin and the science community

It's unfortunate there's a perceived problem. There are clearly a lot of incensed people. Mike's position—which I support—is that an agency has a set of customers, first and foremost the Congress and the White House. They set priorities. If they want to change those priorities, they can. There has been a tendency [for astronomers] to regard what NASA does as a sinecure.

## On how scientists can help

I'm an advocate of small, fast missions that could do 80% of the capability for 10% of the cost. What would be useful is for the scientific community to prioritize missions within the budget we've got, so we can get more science, better science, by doing more smaller missions and fewer bigger ones.

On exploration versus science

10. ASTROBIOLOGY

# Robot Seeks New Life--and New Funding--in the Abyss of Zacatón

1. Kevin Krajick*
1. Kevin Krajick is a writer in New York City.

With missions to other worlds in mind, explorers ready an ambitious robot to plumb a deadly sinkhole, looking for new life—or at least the bottom

Thirteen years ago on a sunny spring morning, two divers prepared to descend into what could be the world's deepest water-filled pit: northeastern Mexico's El Zacatón, a 180-meterwide limestone sinkhole filled by hydrothermal springs. The water is 30°C, teeming with strange microbes, and pitch-black below the first 30 meters. One diver was Sheck Exley, then holder of the world's scuba depth record; the other was his friend Jim Bowden, a top underwater caver. They wished each other luck, adjusted their masks, and began free-falling down separate safety lines. Ten hours later, Bowden surfaced with a new world record—925 feet (282 meters)—without ever finding the bottom. Exley did not surface. Three days later, his body was pulled out, tangled in the line. No one knows what killed him.

The sinkhole's depth remains unknown; sonars work poorly in narrow spaces, so readings peter out at about 330 meters. But this week another team is preparing to replumb the mysteries of Zacatón—this time, with an audacious new robot made to probe both its geology and biology. The NASA-funded Deep Phreatic Thermal Explorer (DEPTHX) is designed to navigate and map deep underwater tunnels, spot living things, grab them, and bring them back—all without direction from the surface. If it survives its first voyage in March, DEPTHX will be a major advance in robotics and exploration of extreme environments. If it survives NASA budget cuts, it could be a model for probing Jupiter's moon Europa, where Zacatón-like cracks or holes in the icy surface may offer the best chance of finding extraterrestrial life.

Compared to other autonomous underwater vehicles (AUVs), DEPTHX is “well ahead,” says Gwyn Griffiths, head of the National Oceanography Centre underwater lab in Southampton, U.K. But like other NASA-funded astrobiology projects, the robot's future is uncertain. Its funding is about to run out, and a follow-up project may be a long shot as NASA cuts back support for such efforts (see p. 318). “Robotic exploration of our planet and the universe has been wildly successful and cheap,” says Dana Yoerger, an AUV guru and cheerleader for DEPTHX at Woods Hole Oceanographic Institution in Massachusetts. “To cut back on stuff like that for manned exploration is going to give the taxpayers very poor value.”

The ringleader of DEPTHX is extreme engineer and cave explorer Bill Stone, who in 1989 made it to NASA's semifinal astronaut-selection round but was nixed as being too independent. During the past 3 decades, he has worked on space and military projects for the National Institute of Standards and Technology and on the side explored some of Earth's most dangerous caverns. Traveling a kilometer or more under the surface, he has stayed down for weeks at a time in air-filled caves. Underwater, he has often dived through twisty, silt-choked passages, re-emerging alive to appear in National Geographic. Finding standard scuba tanks too bulky, he invented a compact rebreather that recycles gases, now used by divers worldwide. In 1998, he made the first high-resolution maps of an underwater cavern, Florida's Wakulla Spring, by inventing a torpedo-like personal propulsion vehicle studded with sonars—the precursor to DEPTHX. He and colleagues drove the devices through 6.4 kilometers of inky-black passages to create three-dimensional (3D) images of the invisible walls. “Deep cave systems are the last terrestrial frontier; they push the limits of human endeavor, technology, and psychology,” says Stone.

They are also dangerous. Stone has lost 16 friends to exploration accidents and has dragged out the bodies of seven himself. Exley was his cave-diving mentor. “I've come to the conclusion that there are places where humans cannot travel safely,” says Stone, now 54. “We need a surrogate.”

At Zacatón, Stone is working with Marcus Gary, a University of Texas, Austin, Ph.D. student who assisted at the fatal 1994 dive and became obsessed with the sinkhole. In a Geological Society of America paper last year, Gary reported that the system owes its vastness to volcanism that adds heat and gases to water running into the limestone. This hastens chemical dissolution of the rock as well as making things cozy for unusual bacteria. In 2003, Stone and Gary joined with a cast of luminaries in space, robotics, and microbiology to win a $5 million, 3-year grant from NASA's Astrobiology Science and Technology for Exploring Planets (ASTEP) program to use Zacatón as a proving ground for a prototype robot that could explore Europa. A side benefit would be exploring Zacatón itself. Another team member is Richard Greenberg, a planetary scientist at the University of Arizona in Tucson who helped show that Europa, about the size of Earth's moon, has a hidden ocean covered with an icy crust. Tides crack and puncture the ice from below, creating sinkholelike features on the surface. Like Zacatón, Europa's ocean is also probably heated by volcanism—ideal for the development of life. Many scientists think a robot might have to melt through some 10 kilometers of ice to reach liquid and thus life, but Greenberg says organisms—probably strictly microbes—may also lie in the surficial slushy cracks and holes. “The beauty of this robot is that it would have the smarts to get in there and look itself,” he says. (A separate craft would probably melt its way to the bottom of the ice and release one or more DEPTHX-like robots to search the liquid ocean.) So far, robots have made only baby steps toward this goal. Deep-sea research still depends heavily on remotely operated vehicles powered by tethers from mother ships. Even the Mars rovers receive radioed instructions from Earth and power from the sun. Robots sent under ground or ice can receive neither, because these block radio and light waves, and tethers would become tangled. New AUVs hold promise, but so far most operate in open waters, merely recording temperature, depth, and salinity. “The fully 3D environment and true autonomy are things robotics is only beginning to address,” says David Wettergreen, a robotics engineer at Carnegie Mellon University in Pittsburgh, Pennsylvania, and DEPTHX team member. Early drawings of the 1500-kilogram DEPTHX robot had it looking like an outboard motor, but in 2005, the team switched to a flattened egg shape to help make it slippery and all-seeing in tight spaces. Wettergreen's team has girdled the surface with 56 transducers that bounce narrow sonar beams in all directions. These hook to a newly elaborated technology called simultaneous localization and mapping (SLAM). As DEPTHX moves—slowly, about 0.1 meters per second—SLAM computers integrate the signals into real-time maps of walls, ceilings, and floors. Theoretically, the craft should hover within less than an arm's length of these features and traverse almost any passage it can fit into. As it travels, it will store the maps and look both forward and back; it is supposed to recognize where it is, and, when it is time to go home, follow the maps back. In spring 2005, the team lowered a stripped-down version of the sonar array into Zacatón and retrieved exquisite 3D images of the sinkhole down to 290 meters, the first even partial glimpse of its shape. Below its wide, circular top, it narrows into something like a tornado spout. Gary says the bottom could lie as deep as a kilometer—the probable limit of water-soluble limestone—and labyrinths of horizontal tunnels could run many kilometers, possibly connecting to other sinkholes nearby. As for navigation, the team is still working out bugs. During a shakedown cruise at a water-filled quarry in Austin, Texas, last month, the robot smacked the muddy bottom, then surfaced unexpectedly under the team's rowboat, smashing a$5000 Wi-Fi antenna and rattling researchers. “The risk of losing this vehicle down there is non-negligible,” admits Carnegie Mellon roboticist George Kantor.

The greatest challenge at Zacatón may be finding and sampling organisms. It's no problem on top: Along with little fish, water moccasins and other snakes up to 2.7 meters long cut the surface faster than humans can swim. But below about half a meter, the hot, chemical-laced water lacks both oxygen and conventional aquatic life. Divers have found a shallow tunnel connecting Zacatón to a nearby river that holds the bones of countless turtles; like Exley, they may have dived too far or too long. On the other hand, hydrogen sulfide and other volcanic gases feed thriving communities of extremophile microbes. Each morning the water is clear, but by noon it turns milky gray, probably from elemental sulfur precipitated out by photosynthetic sulfide-eating bacteria. Further down, the walls are lined with spongy red and purple microbe mats, says the team's microbiologist, John Spear of the Colorado School of Mines in Golden. In the first 82 meters—as far as human divers dare sample—Spear has spotted 27 divisions of bacteria, including six that may be new, along with archaea and planktonic diatoms. “The diversity is astounding. I think that if we get down further, there will be even more,” he says. He expects only microbes but does not rule out bigger life forms. “We could run into tubeworms, or crabs, or something else. We really have no idea.”

To search for such life, DEPTHX is equipped with sensors and software that will allow it to follow plumes of heat, sulfide, nitrates, or turbidity—likely emanations of the volcanic vents that almost certainly lie below and potential hot spots for life. “It's the old ‘getting hot, getting cold'game, built into a robotic brain,” says co-investigator Daniel Durda, a planetary scientist and cave diver at Southwest Research Institute (SwRI) in Boulder, Colorado (Science, 6 September 2002, p. 1640). The robot also continually sips water through a microscope designed to pick out living cells by spotting motion. If cell numbers spike, the robot may follow that trail and suck in a water sample to carry back. Once the robot reaches a likely spot, cameras are programmed to look for changes in colors, textures, or shapes that could set bacterial mats apart from bare rock or open water. “We're not quite sure what we're looking for yet—just something different,” says SwRI engineer Ernest Franke, head of the science-package team. DEPTHX has an arm with a coring device primed to stick itself into a prospective life form and pluck out a sample about the size of one's thumb-end, Little Jack Horner-style.

Skeptics may think all this unlikely, but from 2002 to 2005, a NASA-funded robot that Wettergreen worked on crisscrossed Chile's near-lifeless Atacama Desert seeking patches of photosynthetic algae by its telltale fluorescence. The robot successfully scooped up samples and applied dyes to detect amino acids and lipids—the stuff of life.

However, NASA has slashed exobiology budgets, and this could prematurely end DEPTHX and related ventures. ASTEP, the main funder of exotic robots, has gone from $15 million in 2005 to a planned$4 million in 2007. DEPTHX's current funding ends after its March deployment. Stone and colleagues had next hoped to develop a smaller, smarter DEPTHX to slip into Antarctic subglacial lakes, and then to design a robot to land on Europa by 2020. Already, another cutting-edge DEPTHX-like AUV slated to sample volcanic vents under the ice-covered Arctic Ocean this summer has run over budget, and ASTEP has not rescued it. But John Rummell, NASA's senior scientist for astrobiology, notes that there is still some money in 2007. “It is my fervent hope that we'll be able to fund the next stage of DEPTHX,” he says.

This month, the team will test the robot in the shallower sinkholes near Zacatón. Electrical-resistivity tests around several suggest that their bottoms may be false floors of travertine hiding much deeper watery voids below. These may be sealed environments like the Antarctic subglacial lakes or like Europa's hidden ocean. Then, DEPTHX will swim down into Zacatón itself and meet whatever might be living there. “Whether we actually get that far, we'll see,” says Stone. “No guts, no glory, right?”

11. BIOMECHANICS

# Crab's Downfall Reveals a Hole in Biomechanics Studies

1. Elizabeth Pennisi

The melding of materials and movement to better understand locomotion gets a boost from physicists studying the properties of granular materials

PHOENIX, ARIZONA—When it comes to running on sand, the ghost crab is an Olympic champion. With legs that are a blur to the naked eye, Ocypode quadrata scoots up to 2 meters per second on hard-packed sand. But soften up the sand a bit, and the gold medal instead goes to the zebra-tailed lizard, an animal that spends little time on the grainy material. This surprising observation, reported earlier this month here at the annual meeting of the Society for Integrative and Comparative Biology, comes courtesy of physicist Daniel Goldman of the Georgia Institute of Technology in Atlanta.

Goldman has jumped into the field of biomechanics by employing a device physicists have long used to examine granular materials. That's allowed him to study how animals move over different kinds of surfaces, an approach that Goldman and others feel has been neglected to a large extent. “It's nice to see practical and theoretical applications of granular physics applied to an organismal biomechanics problem,” says Andrew Biewener, a biomechanicist at Harvard University. “It creates an entirely new field of investigation,” which will advance both basic biology and robot engineering.

Until now, most researchers have studied how animals walk, run, trot, and otherwise move using hard, nonskid platforms. “When we studied forces, the last thing we wanted was to have slippery surfaces,” says Catherine Loudon, a biomechanicist at the University of California, Irvine. And this approach has proved useful, as researchers have made progress analyzing how muscles and tendons make different gaits possible (Science, 21 January 2005, p. 346).

But in the wild, organisms must contend with mud, gravel, and ground littered with debris. Sand can be particularly challenging, as its grains give way briefly underfoot, transforming the surface from a solid to a virtual liquid. Goldman wants to understand how organisms deal with this complexity. “We can't predict how animals will move until we understand the substrate,” he says.

At the University of California, Berkeley, Goldman and Wyatt Korff, now at the California Institute of Technology in Pasadena, built a “fluidized” bed, a box of glass beads that were stand-ins for sand. The bed's underside has a porous membrane, and by pumping air at different speeds up through the membrane, Goldman can change how tightly packed the beads are, thereby controlling the properties of the “sand.” More air results in looser packing and, eventually, a surface much like quicksand. Aerated enough, the bed turns into a fluid. The method is “extremely brilliant,” says Frank Fish, a biomechanicist at West Chester University in Pennsylvania.

Goldman and his colleagues chased ghost crabs, geckos, and various lizard species down a sand-filled track and across the bed, filming the animals as they traversed hard, soft, and “liquid” sand. In addition, he and Korff dropped wires attached to rods into the sand to determine the mechanical requirements for locomotion in sand of different consistencies.

As expected, the ghost crab zoomed across the hard-packed sand. But in soft sand, its eight legs sank in, and the crab trudged along at about 40 centimeters per second. That's about the speed of the gecko, which is adapted for living in trees, not on beaches. “We didn't think there would be such a big difference,” Goldman says. The Mohave fringe-toed lizard, another sand dweller, also got bogged down: Its speed dropped by 10%. “Being specialized for sand doesn't necessarily mean better performance” on all forms of sand, Goldman reported.

The big winner on the softer sands was the zebra-tailed lizard. It left the ghost crab in the dust, maintaining at least a 1.5-meter-per-second pace, even in quicksand. This species lives in a varied environment, traveling through brush and on rocks, gravel, and, occasionally, sand; therefore, Goldman expected that it would lack any special adaptation that would enable it to excel on any one surface. But the zebra-tailed lizard didn't sink, and “it seems to use feet as a buffer against the substrate,” Goldman said. The lizard has extremely long, gangly toes, and Goldman discovered that it spreads the toes wide as they hit the sand and then curls them up as it lifts the foot. He suspects that sand caught between the toes causes the sand to stop flowing such that it supports the lizard's weight and allows the animal to push off into the next step.

Fish is not convinced that long toes are the secret to this lizard's success. “I don't think they understand enough about the dimensions of the feet and how they interact with the sandy environment,” he says.

Goldman is addressing those interactions. He and Korff have designed an artificial “foot”: a rod with crossed wires attached perpendicularly at the end. They vary the angle between the wires and drop the “foot” into the sand, measuring how far it sinks. “The penetration depth depends on the angle” between the individual wires, Goldman reported. “It shows geometry can be important in your foot.”

Understanding the differences between how the ghost crab and zebra-tailed lizard move could help engineers make better robots, which for the most part stop dead in sand. “You want to have robots that can run around on all surfaces,” says Loudon. For that reason, “it's of great importance to understand how animals can [handle] such different surfaces.”

12. WILLIAM K. HARTMANN PROFILE

# Renaissance Man of the Solar System

1. Richard A. Kerr

Scientific innovator from the moon to Mars, author of textbooks and novels, and space artist, William Hartmann is the independent scientist writ large

TUCSON, ARIZONA—Whether it's his office, studio, or home, William Hartmann packs the pictures in. Everywhere the walls are covered with paintings. And they are paintings of everywhere: an abstract of Paris by writer Henry Miller; Carmel Beach, California, by Hartmann's space artist hero, the late Chesley Bonestell; his own depiction of a European café; a Swiss landscape by his grandfather.

And then there are the chockablock paintings of the Great Out There from Hartmann's 35-year career as a space artist. The one still on the easel in his backyard studio depicts icy geysering on Saturn's moon Enceladus. Another shows how a giant impact 4.5 billion years ago could have splashed the makings of the moon off Earth, an offbeat hypothesis he co-originated, which is now the scientific consensus on how the moon formed. Hartmann's imagined views of space blend with his words in his fifth-edition planetary science textbook, his half-dozen popular books, and even his first novel.

“He's one of the most productive, innovative scientists in the field,” even while pursuing his painting and writing, says planetary geologist Ross Irwin of the National Air and Space Museum in Washington, D.C. “He's made as many discoveries as anyone could hope to.”

And he's done it without tenure, without an academic position or even a salaried one the past 36 years. He did help found the Planetary Science Institute in Tucson, where he—like all other PSI staffers—has done his science on nothing but soft money. A nonprofit, PSI has lately been a model for a growing number of planetary scientists looking for workplaces “run by scientists for the benefit of science,” as PSI's founders put it.

## A backyard start

Hartmann, 67, approached planetary science at an auspicious moment asking the right question. “What are the planets like?” he wondered as a 14-year-old. Through the telescope that he built with the help of his engineer father, he could make out the cratered lunar surface well enough, but from his home in suburban Pittsburgh, mysterious Mars was a tiny disk of shifting smudges.

At the time, professional astronomers couldn't do much better. As a graduate student at the University of Arizona (UA) at the dawn of the space age, the early 1960s, Hartmann found himself studying existing images of the moon taken from Earth. But his adviser Gerard Kuiper, a founding father of planetary science, showed him a new way to look at the moon, one that would shape much of his scientific career.

From Earth, astronomers could see only the near side of the moon. Around the edges, they could see it only at terrain distorting low angles. So Hartmann projected telescopic photographs of the moon on a plain white 1-meter sphere and photographed this lunar globe from the side, simulating an overhead perspective not available until the Apollo missions. His darkroom experience would eventually involve him in a major UFO study for the U.S. Air Force and image analysis for the U.S. House of Representatives Select Committee on Assassinations. The scientific payoff, however, was the discovery of the Orientale basin, a huge impact scar on the extreme west limb of the moon, the first of its kind to be recognized.

Orientale's discovery would typify much of Hartmann's science. “A lot of my career has been the big-picture stuff,” he says. While most of his contemporaries worked to become the world's expert on the topics of their dissertations and to turn out data to five significant figures, Hartmann headed the other way. “The first-order things linking different planets has always appealed to me,” he says.

In that spirit, he next used impact craters on the moon and on Earth to gauge the age of the great lunar lava plains that shape the man in the moon. Assuming impactors steadily rain in from the asteroid belt like sand through an hourglass, the number of craters on a surface measure the age of that surface. Comparing lunar crater counts and craters on dated surfaces on Earth, Hartmann calculated the age of the lunar lava plains to be about 3.6 billion years. The late Eugene Shoemaker, the leading cratering expert of the time, put the age at 0.1 billion years. Five years later, lab dating of Apollo rocks proved that Hartmann's estimate was right on.

In the early 1970s, after serving on the science team for Mariner 9—the first artificial satellite of Mars—Hartmann applied his crater-counting idea to the Red Planet. Mars may have looked geologically decrepit, but he found some lava plains whose low crater counts implied that they could be a mere 100 million years old, born last year if Mars were an octogenarian. In the 1980s, dating of meteorites from Mars confirmed the youthfulness of at least some martian lavas. By the late 1990s, Hartmann had co-authored the gold-standard cratering chronology for the inner solar system.

## Freelancers unite

So in the late 1960s, Hartmann was at the start of a roll. He even had an assistant professor's position at UA. But by 1970, Kuiper—concerned that Hartmann develop some professional independence—was nudging him out of the nest when there were still few places to go in planetary science. As it happened, Hartmann had company. In 1968, UA graduate Alan Binder had talked his Chicago-based employer into opening a Tucson office. By 1971, the office included not only Hartmann but also UA Ph.D. Donald Davis and UA student and Massachusetts Institute of Technology Ph.D. Clark Chapman, now at the Southwest Research Institute in Boulder, Colorado.

From the start, the Tucson group was on a mission of its own. It would pass through a half-dozen leased offices and work its way through three parent organizations, always searching for a free hand and a lower overhead. It was a nonprofit division before going independent in the late 1990s, but the philosophy remained the same. “We tried to design PSI to be good for the individual researcher,” says Hartmann. “We designed it around how people wanted to live and work. We didn't have faculty meetings or deans, but we didn't have assured money either.”

That meant gathering a few young researchers less interested in teaching than in doing hands-on science in overlapping fields, and then bringing them up to speed on the fine art of winning NASA grants. By the mid-1970s, PSI was a group of five researchers with all the expertise necessary to tease out the secrets of how a disk of dust and gas had clumped into balls of ice and rock that banged into each other to form planets, moons, asteroids, and comets. Hartmann briefly served as manager of the group until, as a short history of PSI by Davis, Hartmann, and a colleague puts it, “his natural inabilities were recognized.”

PSI has ballooned in the past 5 years, as have several other nonprofit planetary institutes. It has gone from a staff of 13 and annual revenue of three-quarters of a million dollars to a staff of 55 (half of them women, with only four or five administrators) and an annual revenue of \$3.5 million. Many were attracted not just by the organizational simplicity but also by the geographical flexibility to work where they wish or where their spouses work. The result is a “virtual institute.”

The PSI synergy soon led Hartmann to the biggest find of his career. From his own cratering work, Russian theoretical studies, and lab experiments, he and PSI colleagues realized that bodies violently colliding in the still-forming solar system came in a range of sizes. There were far more small ones than large ones. The object that hit the moon to form the Orientale basin was about 150 kilometers across. What was the biggest body that could have hit the nascent Earth? wondered Hartmann.

If the biggest impact were big enough to blast some of Earth's iron-poor mantle into orbit to form the moon, “it seemed to me that would explain a lot,” says Hartmann. Apollo astronauts had just brought back lunar rocks for geochemists to analyze. “I was in awe of the geochemists because they worked at such high levels of accuracy,” he says. Yet they couldn't explain even the grossest of the moon's properties, such as its dearth of iron.

So Hartmann teamed with Davis—a dynamicist—to develop the giant-impact theory of the moon's origin. Pretty much ignored after its 1975 publication, it became the surprise leading contender at a workshop in 1984 and has been pulling away ever since. The success of the giant-impact hypothesis gave Hartmann renewed faith in making inferences from a few fundamental properties rather than a welter of data. Hartmann has missed the misspelling of his own name on papers, Chapman says, but “he sees things in a subjective way that can be more effective” than fighting through all the details.

## Another side of science

At least part of Hartmann's more intuitive approach was nearly left behind with the trappings of amateur astronomy. He had grown up with a sketchbook in his hand and his grandfather's paintings all over the walls of his home. But by grad school, he had absorbed the message that he'd need physics, not graphics, to understand the planets. That changed in 1970 when a publisher asked him to write an introductory textbook for planetary science. The prospects for illustrating solar system bodies other than the moon and Mars were bleak. So he solicited paintings from his growing contacts in the space-art community and took up his own acrylics and brush.

Hartmann sees a productive interaction between “the painter's eye versus the scientist's eye.” For instance, sometimes he understands planetary photometry—the interaction of light and surface—“experientially” by noting how the color of a terrestrial dune he's painting changes with changing sun angle. And trying to paint an alien scene not yet visited by a lander “forces you to ask what we actually know.”

As a young scientist, says Hartmann, he was considered something of a dilettante because of his painting. But other scientists eventually saw the rewards of someone translating their data into a form accessible to the public. In 1997, Hartmann even won the first Carl Sagan Medal for communication of science to the public from the American Astronomical Society, in part for his art.

The other part of his public communication has been writing, starting with a textbook and a half-dozen popular science books. Since the Sagan award, he has also published two novels. Mars Underground combines science fiction and mystery on a scientifically realistic Mars, while Cities of Gold switches back and forth between 1989 Tucson and the early days of Spanish incursion there. “I like having a dialogue with interesting people who take you to different places or times,” says Hartmann. “Novelists can be the scientists of the human psyche. You can talk about everything.”

Eyeing retirement, Hartmann has been cutting back his science in recent years to favor painting and writing, although he still feels an obligation to contribute to PSI and its effort to groom new planetary scientists. Whether PSI is primed to produce another renaissance man, or woman, is hard to tell. “A lot of the questions I wanted answered at 14 have been answered,” says Hartmann. And gigabytes of those pesky details he tended to avoid have been returned from spacecraft, with terabytes more to come.