Origins

Scientists use the “molecular clock”—an estimated rate of DNA mutation—to date key events such as migrations and the divergence of species. But just how accurately the clock keeps time has long been debated. A new study of living and ancient Antarctic penguins, like those on Ross Island at left, suggests that DNA mutates six times faster than predicted. That could mean that some species—such as chimps and humans—could have split off from each other much more recently in time than previously thought. The finding should help improve the dating of relatively recent events, including when people domesticated various crops and animals, and when major human migrations occurred.

To use the molecular clock, scientists estimate the rate of mutation in DNA, estimating that the mutations occur in a steady, clocklike manner. For example, if a gene accumulates changes at a rate of five every 1 million years, 25 mutations in a genetic sequence would mean that the sequences had diverged 5 million years ago. The technique has been used to estimate when humans separated from the other great apes, to estimate the arrival of people in the Americas, and to create evolutionary trees for many species. Molecular clocks are usually calibrated by using the age of a known species from the fossil record. But scientists disagree about the speed or rate at which mutations occur and under what circumstances the rate is influenced by natural selection or other factors.

To see just how accurate molecular dating is, David Lambert, an evolutionary biologist at Griffith University in Queensland, Australia, and colleagues looked at Adélie penguins. These Antarctic birds may be the best species yet for building an accurate clock, the team argues, because scientists can study the genetic sequences of both living and ancient members of the species. The penguins generally return each year to the same nesting ground; thus, each rookery can have layers of bones dating far back in time. Indeed, the birds have nested at some rookeries for 44,000 years. "You can take blood samples from the living penguins and then literally collect the bones of their ancestors" in the ground below, says Lambert, because the penguins usually return to their natal colony to mate. Other studies usually can only compare genes from organisms separated in time by millions of years. 

Using modern blood and ancient bone samples, the researchers extracted the entire mitochondrial genome from 12 modern and eight ancient penguins, including two that were dated to 44,000 years ago using radiocarbon methods. They then compared the mitochondrial DNA of the living penguins with the ancient ones to determine the number of mutations that had occurred. Because they had radiocarbon dates for the ages of the ancient penguins, the scientists could accurately measure the bird’s average mutation rate, ultimately calculating that its mitochondrial genome had evolved at a rate two-to-six times faster than previously estimated.

The team's findings, reported in this month’s issue of Trends in Genetics, support similar results for faster clocks in mitochondrial sequences in cattle. But in this new study, the researchers succeeded in calculating the rate of mutation within almost the entire mitochondrial genome, providing “more conclusive evidence,” for a rapidly ticking clock, says Dee Denver, an evolutionary biologist at Oregon State University in Corvallis and one of the paper’s co-authors. They also focused on a region of the genome that is known to not be influenced by natural selection, they write in the paper. Thus, they say that the resulting clock is not merely a reflection of penguin evolutionary history and can be applied to other species.

"It's novel and groundbreaking work," says Mark Hauber, an evolutionary biologist at Hunter College in New York City, who was not affiliated with the study. "It's a significant discovery," adds Elizabeth Matisoo-Smith, a biological anthropologist at Otago School of Medical Sciences in Dunedin, New Zealand, who expects it will help resolve several discrepancies between genetic data and the archaeological record, such as the peopling of the Pacific Islands and the Americas. However, the penguins’ rapid clock "should be confirmed on a wide diversity of species" before being adopted as the new standard, says Robert Wayne, an evolutionary geneticist at the University of California, Los Angeles.

Photo credit: Euan Young

by Elizabeth Pennisi

Charles Darwin worked hard to figure out how cooperation within a species—self-sacrifice among worker bees, for example—could have evolved. But he was stumped when it came to understanding cooperation between species. In his book, On the Origin of Species, he wrote, “Natural selection cannot possibly produce any modification in a species for the good of another species.” If that could be proved to have happened, “it would annihilate my theory, for such could not have been produced through natural selection.” And he scoffed at supposed examples perpetuated by some natural historians, such as a rattler using its rattle to warn prey. “I would almost as soon believe that the cat curls the end of its tail when preparing to spring, in order to warn the doomed mouse.” Instead, he argued that the rattle was meant to scare off birds and other potential predators.

Nonetheless, countless examples of cooperation between species exist—albeit many perhaps outside Darwin’s knowledge. Many long-standing partnerships are strengthened by specialized structures or traits in one species that benefit the other. Many of these relationships are dynamic, shifting back and forth over evolutionary time between exploitative and mutualistic.

Take the ant plants and their ants (see left). In tropical forests, certain types of trees make a home for ants that inhabit them, providing hollow stems or leaf pouches where the insects can roost and raise young. In return, the ants keep hungry herbivores at bay and sometimes kill off surrounding vegetation, creating a clearing around the trees.

Nineteenth century naturalists fiercely debated whether these ants forced their way into trees as parasites, wounding trees to make their nests, or whether they had a more benign relationship with the plants they lived in. In 1873, botanist Richard Spruce likened the ants to fleas on a dog—a nuisance. But others contended that some plants, acacias in particular, provided hollow thorns and food rewards to keep ants around for protection from herbivores. Many studies have supported this hypothesis over the past 4 decades. It seems the ants bite and poison surrounding vegetation, reducing the competition for space, water, and sunlight.

In the November issue of American Naturalist, David Edwards of the University of Leeds, U.K., and his colleagues describe how sometimes these ants get carried away. “I think most [researchers] believed that ants’ relationship with their host plants had been pretty well defined,” says John Tooker, a chemical ecologist at Pennsylvania State University, University Park. But in 1996, scientists observed a new, peculiar ant behavior. While exploring the jungles of southeastern Peru, a team of ethnobotanists came across a number of "devil's gardens," what many locals call the ant-made clearings around trees. Although they had seen such clearings before, the researchers were surprised by what the natives showed them next: Trees of other species on the outside of the clearing were scarred and swollen (see right) with networks of cavities filled with worker ants, queens, brood, and mealy bugs. "These galls made up a large percentage of the swollen trunk volume," says Edwards. Sometimes the internal excavations were so extensive that the tree had collapsed. The locals blame the scars on forest spirits.

The researchers think the ants are attacking these other trees because there aren’t enough ant plants to house ever-expanding colonies. “It suggests a level of ecosystem engineering [by the ants] not previously recognized,” says Tooker. And at this point, the relationship seems anything but mutual. “I expect this to be an antagonistic relationship because of the range of tree species that are galled by the ants,” Tooker notes. These are trees not typically associated with ants, and so there’s been little opportunity for a partnership to evolve. But if the ants patrol the area and ward off herbivores, then perhaps there is some payback by the ant to the tree, he adds.

The ants involved belong to the genus Myrmelachista, which typically nest in stems. Some species in the genus do not form associations to particular species, says John Longino, an entomologist at Evergreen State College in Olympia, Washington. He suggests that devil’s garden ants coevolved with new queens, gradually evolving a preference for certain plant species as the best nesting spots. Their targets eventually provided housing rather than chance having irregular holes chewed into their stems. Then, “perhaps it doesn’t take much to turn a mild-mannered and inconspicuous stem-nesting ant into a ferocious devil’s gardener. Maybe just the right kind of plant can encourage and manipulate those latent talents,” he says.

These enticements can lead to trouble, however, as Edwards's collaborator, Megan Frederickson of the University of Toronto in Canada, has discovered. Ever in need of more room, the ant Allomerus octoarticulatus takes a devious step to promote its host tree’s growth. It destroys any flower buds that the host produces. When Frederickson measured the growth rates of sterilized and reproductive plants, she found that the ant’s drastic maneuver did encourage more vegetative growth—and more living space for the ants. She reported those findings in the May issue of American Naturalist.

Darwin did not have the benefit of these experiments that show reciprocated benefits and the dynamic balance between giving and taking. But if he had, “he would have found ant-plant symbioses a real hoot,” says Mike Kaspari, an ecologist at the University of Oklahoma, Norman.

Photo credits: (ants) Megan Frederickson; (tree) Douglas Yu

by Michael Balter

LONDON—When Tim White of the University of California, Berkeley, agreed to speak at a human origins meeting^* at the Royal Society here, he sent no abstract and provided only a one-word title: “Ardipithecus.”

But that one word was enough to earn him a spot on the podium and a prominent role at the meeting. After 15 years of study, White and his colleagues have just published their massive, 108-page report on Ardipithecus ramidus, at 4.4 million years the oldest partial skeleton of a putative human ancestor (see the Science special issue of 2 October, cover at left). The 2-day meeting marked White’s first in-person presentation of Ardipithecus to a roomful of independent researchers; co-author anthropologist C. Owen Lovejoy of Kent State University in Ohio also made the trip. Their emphasis on Ardi’s non-apelike features, however, drew a tart response from some primatologists in the room.

The roughly 50-kilogram female, which the Ardipithecus team concludes walked upright although it also spent time in the trees, has a decidedly un-chimplike anatomy. White and colleagues therefore have asserted that living apes are not good models for understanding the last common ancestor (LCA) of humans and chimpanzees—a claim that has stung many primatologists. “I assumed … chimps might be helpful in tackling the challenges of human evolution,” said chimpanzee expert William McGrew of the University of Cambridge in the United Kingdom. But after Ardi, “all this chasing after chimpanzees was deemed to be irrelevant” to human evolution. McGrew, the first speaker, challenged a statement in one of the papers in Science that “no modern ape is a realistic proxy for characterizing early hominid evolution—whether social or locomotor.”

White, Lovejoy, and their colleagues base this conclusion on many features of Ardi’s postcranial anatomy. Although some researchers had assumed that the LCA had arms adapted for swinging in the trees and walked on its knuckles like chimps, A. ramidus lacks these and many other features possessed by modern apes such as chimps and gorillas. The team argues that Ardi’s anatomy suggests that its behavior and social structure also differed from that of modern chimps. For example, one of the papers, authored by Lovejoy, suggests that A. ramidus males were probably much less aggressive toward each other than chimp males are and that they pair-bonded with females.

McGrew told the attendees that although he agreed that “the LCA was not a chimpanzee,” the behavior of living chimps such as those at left can still inform hypotheses about the LCA’s behavior. McGrew listed the considerable evidence for chimp behaviors that possibly mirror those of extinct and modern humans, such as the use of complex tools, aiming and throwing objects, the construction of sleeping nests or platforms, and evidence for considerable spatial cognition such as the ability to remember the locations of thousands of trees in a forest. At least some of these behaviors, McGrew said, are shared by chimps and modern humans because they have deep evolutionary roots. “So to hypothetically credit the LCA with the ability of [chimps] is not unreasonable,” McGrew concluded. 

At the end of the day, when the meeting was thrown open for discussion among the roughly 200 attendees, White countered McGrew’s argument, pointing to what he saw as the dangers of using a chimp model for the LCA’s behavior. “If we try to model the LCA or even the earliest hominids based on living chimps, which have these adaptations to [swinging in the trees], to moving through that canopy so well and so quickly that they can take down a red colobus monkey, we could be very misled. Ardipithecus probably couldn’t do that, and the LCA probably couldn’t do that.” 

Yet White says he’s not trying to toss out all chimp research. “The fact that a few highly derived descendants managed to survive until today obviously enhances our appreciation of their evolution,” he told Science. “It also provides perspective on ours. The study and conservation of chimpanzees and bonobos are surely justifiable on their own merits, even though [early] hominid skeletal anatomy, behavior, and ecology do not match those seen now in these specialized persistent ape lineages.”

And Carol Ward, a paleoanthropologist from the University of Missouri School of Medicine in Columbia, points out that White and Lovejoy “aren’t saying that [studies of] chimps aren’t useful but that we didn’t evolve from chimps, so they are useful as referential models rather than actual representatives of what our ancestors were like.” Chris Stringer of the Natural History Museum in London, who co-organized the meeting with Alan Walker of Pennsylvania State University, University Park, also sees a role for chimp work in evolutionary studies. “There is such a lot of excellent and important data on ape behavior,”  Stringer told Science, “that we would be foolish to throw it out as providing potential models for the behavior of early hominins.”

^*The First 4 Million Years of Human Evolution, London, 19-20 October 2009.

Photo credits (top to bottom): Cover; Jupiterimages

by Greg Miller

Cephalopods—octopuses, squid, and their relatives—ruled the seas in the Cambrian era, some 500 million years ago. But their world changed in a big way with the Cambrian Explosion, a rapid diversification of life on Earth that included the origin of fish. Suddenly, cephalopods had new opportunities—delicious fish!—and their first serious competition and potential predators. They had to get smart in a hurry.

So it’s no wonder then that modern cephalopods have the most complex brains of any invertebrates. An octopus brain (lower, right) has 50 to 75 lobes and at least as many neurons (about 100 million) as a mouse brain (lower, left). And that’s not counting the smaller "brains" in each arm and the still smaller "brains" (ganglia, technically) associated with each sucker.

All this neural circuitry gives octopuses exquisite control over their bodies, including some nifty tricks for evading predators, and it has even prompted speculation about cephalopod consciousness.

Although the octopus brain rivals the size and complexity of many vertebrate brains, its architecture differs dramatically. “Short of martians showing up and offering themselves up to science, cephalopods are the only example outside of vertebrates of how to build a complex, clever brain,” says neuroscientist Cliff Ragsdale of the University of Chicago in Illinois. For that reason, Ragsdale says, these creatures have much to teach us about brain evolution.

Ragsdale and his postdoctoral fellow Shuichi Shigeno are investigating whether the neural circuits that control movement, memory, and other functions in an octopus brain work the same way as do the analogous circuits in other animals. The British zoologist J. Z. Young did early work on octopuses in the 1960s, mapping out their brain anatomy and ablating individual lobes to investigate their contributions to behavior. But as of yet, Ragsdale says, no one has brought modern molecular methods to bear on these questions.

That’s what Shigeno has begun doing. At the Society for Neuroscience annual meeting in Chicago earlier this week, he presented preliminary findings from a series of experiments that investigated gene expression in the developing brain of Octopus bimaculoides, a small octopus that inhabits mud flats along the southern California coast. Shigeno found that several genes with well-known roles in patterning “lower” brain regions such as the brain stem and spinal cord in mice are expressed in a similar pattern in the developing octopus brain.

Genes involved in the development of “higher” brain regions in mice, on the other hand, showed more diverse patterns in the octopus brain. Genes expressed in the mouse hypothalamus, an area that regulates sleep and appetite, among other functions, also showed up in a lobe of the octopus brain thought to perform analogous tasks. But genes expressed in the mouse hippocampus and cerebral cortex, although also expressed in the octopus brain, had a distribution that bore little resemblance to their distribution in mice. To Shigeno, the findings suggest that lower levels of the octopus brain may be wired much like the analogous circuits in the vertebrate brain, but higher levels may show more divergence from the vertebrate wiring diagram.

Ragsdale agrees, but he cautions that such studies of gene expression tell only part of the story. He and Shigeno already have other experiments in the works to map out neural circuits in the octopus brain more directly and to study the neurochemistry of neurons in areas such as the hypothalamus analog to see if they secrete the same signaling peptides as do hypothalamic neurons in other animals. “We have a lot to learn about these animals,” Ragsdale says.

Photo credits: Cliff Ragsdale and Suichi Shigeno

Reptiles look old school, and they have old school B cells that retain an ancient ability our B cells have lost, says a new study published today. Our B cells cannot engulf invading bacteria, but a turtle's can. The results help narrow down when the immune system's antibody factories stopped dining out.

The mammalian immune system divvies up the pathogen-fighting duties. Macrophages and similar cells perform phagocytosis, eating and destroying bacteria and other invaders. Instead of tangling with pathogens, B cells counterattack by pumping out antibodies that home in on intruders. According to immunologists, mammalian B cells aren't capable of phagocytosis.

Researchers infer that phagocytosis came first—single-celled organisms such as amoebas use the maneuver to capture food—and that B cells evolved from phagocytic cells (see the Origins essay on the evolution of the immune system). A 2006 study bolstered that hypothesis, showing that cells from fish and frogs have both abilities.

The question is when in vertebrate evolution some B cells lost their appetite for pathogens. Nobody had run a taste test on reptile cells, so a team from Illinois State University in Normal offered fluorescent beads to B cells from red-eared sliders, a kind of turtle (above). Some of the cells snarfed up the beads, indicating that they were capable of phagocytosis, report Laura Zimmerman and colleagues in Biology Letters.

These dining habits suggest that the B cells of mammals didn’t give up phagocytosis until after the group parted from reptiles. The researchers propose that B cells' double duty reinforces immunity in reptiles, whose antibodies aren't as potent or produced as quickly as those of mammals. To sharpen their picture of B cell evolution, researchers now need to determine whether bird B cells also abstain from phagocytosis.

Photo credit: Trisha M. Sears

“Ardi,” the oldest known skeleton of a hominin, or member of the human family, has grabbed headlines around the world since her unveiling in Science Thursday. Not surprisingly, the press coverage of the 4.4-million-year-old Ardipithecus ramidus has sometimes been sensational—and, in some cases, completely wrong.  Some newspapers and broadcasters have misinterpreted the authors’ finding that Ardi did not look like a chimpanzee or gorilla. Based on this anatomy, the authors proposed that Ardi shows that humans did not evolve from a “chimpanzee-like ape.” By that, they meant that Ardi evolved from an ancient ape that didn’t look like a chimpanzee or gorilla does today and that humans have retained some of those primitive traits.

But the word “chimpanzee-like” sometimes got lost in translation. Even the first version of a press release from Kent State University, where co-author C. Owen Lovejoy is on the faculty, said “Man Did Not Evolve From Apes.” And some media were clearly confused. The Torstar News Service in Canada wrote: “Man didn’t descend from apes. What is closer to the truth is that our knuckle-dragging cousins descended from us.”

A radio announcer in Baltimore, Maryland, asked me in an interview Monday if it was true that we were not apes—or even primates—and that we had our own, separate lineage that was more ancient. The same question came up in a Facebook chat with me and my editor at Science, Elizabeth Culotta, and has popped up in other media. 

Most disconcerting to the authors was the reporting on Ardi by the Arabic news network Al Jazeera, based in Doha, Qatar. A translation of the article written in Arabic starts with a headline that reads “Ardi Refutes Darwin’s Theory,” and the first sentence reads “American scientists have presented evidence that Darwin’s theory of evolution was wrong.” The article states that Ardi’s discovery “refutes the long-standing assumption that humans evolved from monkeys.”

Dr. Zaghloul El-Naggar, a professor of geology in several Arab universities (the article does not specify which ones), exclaims in the story that Westerners were beginning to “come to their senses after they used to deal with the origins of man from a materialistic perspective and by denying religions.” He goes on to claim that the age of Earth does not exceed 400,000 years, and that Ardi’s age of 4.4 million years is an exaggeration. 

For the record, all of this is plain wrong. Ardi is a primate descended from more ancient apes, as are all humans and human ancestors. Apes in turn are descended from monkeys. Chimpanzees are our closest living relatives— we share 96% of our DNA with them, and our lineages shared an ancestor sometime between 6 million and 8 million years ago, possibly earlier. The authors’ point is that the last common ancestor we shared with chimpanzees didn’t look like a chimp—which means that chimpanzees also have been evolving since the two lineages diverged. Finally, Ardi confirms rather than refutes Darwin’s prediction in 1871 that our progenitors lived on the African continent, as well as providing another link in the evolutionary chain from primitive apes to humans.

 by Elizabeth Pennisi

Why in tropical forests do tall broad-leaf trees tower over a layer of understory species? What dictates that shrubs and herbaceous plants pepper the ground below, creating an environment recognizable the world over as tropical forest.

Biologists have long wanted to know why forests and other ecological communities look the way they do. In the beginning of the 19th century, Prussian naturalist Alexander von Humboldt started to find out. He assembled the first comprehensive treatise on how vegetation varies with altitude, climate, soil, and other factors. The work was a groundbreaking exploration of the physical underpinnings of ecological structure: what determines the species that make up a community and their relative abundance.

More than a half-century later, Charles Darwin quietly conducted experiments in his garden at Down House that were even more seminal. Examining a patch of unkempt lawn as it went to seed, Darwin observed that the species changed through time and that competition led to the demise of less-vigorous ones.

Ever since, ecologists have wrestled with understanding what dictates the proportions of each plant in communities varying from meadows to montane forests. How these forces set up communities has "arguably been one of the most primary questions driving ecological science since its origins," says Brian Enquist of the University of Arizona, Tucson. Competition, predation, disturbance, and other factors have a heavy hand, and new research is showing the influential role of evolution, as well.

In this month's Origins essay, Erik Stokstad explores the thinking that has gone into understanding and interpreting community compositions. A combination of physical and biological forces organizes species into predictable communities such that a rainforest is recognizable no matter what part of the planet it grows in. He describes the role of these physical factors and the influence of different biological interactions. Some ecologists think competition is key; others contend the abundance and diversity of species in a community is determined mainly by random dispersal, speciation, and extinction. This latter idea, dubbed the "Unified Neutral Theory of Biodiversity," makes a radical assumption: It considers all organisms of the same trophic level (plants, say, or herbivores) as demographically identical; that is, each organism in a particular level has about the same chance of reproducing, dying, migrating, or giving rise to a new species. The neutral theory has had mixed support. Some researchers think that biological interactions and “neutral” factors work in concert. As challenging as sorting out the rules that govern ecological structure has been, the effort is worth it, say researchers, because of the potential conservation benefit.

Image: Katharine Sutliff

 by Elizabeth Pennisi

As social as humans are, their cooperative nature pales in comparison to that of ants, bees, wasps, and termites (see hill, left). Colonies of these insects can number in the millions and function seamlessly as “superorganisms.” In their book, The Superorganism: The Beauty, Elegance, and Strangeness of Insect Societies, Burt Hölldobler and E. O. Wilson point out that this way of life makes for very successful living. These insects represent a mere 2% of the insect species yet take up two-thirds of the insect biomass. In tropical rainforests, ants outweigh all the mammals and land vertebrates combined.

Yet the scores of entomologists and evolutionary biologists who have marveled at the efficiency of superorganisms have yet to sort out for sure how superorganisms evolved. True superorganisms are highly eusocial: Typically, one or a few queens lay all the eggs, which are tended to by nonreproductive workers. Their fecundity can be astonishing: In their more than 10-year life span, queens of Atta ants (see right) can produce 150 million daughters, for example. Multiple generations live together; and because workers are sterile, very few conflicts arise, and the colony runs quite efficiently.

Less extreme versions of this lifestyle exist, leading some to suggest that eusociality evolved in stages, starting with a female that set up communal nests with other females, with some forgoing reproduction to help provision and protect the young. A few have suggested that it’s not even that critical that the founding females be all that closely related.

In my Origins essay on cooperation, I  did not touch upon the origin of these truly social insects. But Jacobus Boomsma of the University of Copenhagen, Denmark, has thought extensively about this question and rejects the stepwise progression from cooperative breeding to eusociality, asserting that not just close kinship but also lifetime monogamy is critical to incipient eusociality. Early eusocial species “have a very special form of strict monogamy that has been unappreciated,” he says. This idea has been suggested before, but “Boomsma has performed a valuable service in reviving it and extending it,” says Andrew Bourke of the University of East Anglia, United Kingdom.

Termites mate for life, with a single queen and “king” producing generations of siblings, all equally related to one another. Once in their lifetime, wasps, bees, and ants leave the nest on a mating frenzy, with the queens returning with enough sperm to last the rest of their reproductive years. The consequence of having just one mate for life is that the many generations of offspring are all siblings that on average share half their genes. That number of genes in common is the same as they would have in common with their own offspring should they try to reproduce. Thus, if there is even a small survival advantage to group living, that advantage would be a strong enough selective force to encourage the evolution of sterile castes and true eusociality, Boomsma argues. “When a parent refrains from mating with any additional mates, their offspring are free to stop mating at all,” he explains. However, strict monogamy is rare, particularly over evolutionary time scales, and thus, so is eusociality.

In the late 1950s, Kansas University entomologist Charles Michener suggested that eusociality could arise in one of two ways. By the subsocial route, parents associated with offspring; by the parasocial or semisocial route, females joined forces with their peers in communal settings. Yet even today, that latter scenario lacks any hard evidence, says Boomsma. Such cooperative breeding setups never lead to permanently sterile helper castes; there are too many conflicts of interest. Those conflicts disappear where queens have a brief mating interval and then settle down for a life of reproduction using the sperm acquired during that one fling.

A 2008 phylogenetic analysis of mating systems in ants, bees, wasps, and termites supports Boomsma’s hypothesis. In 2008, William Hughes of the University of Leeds, U.K., and his colleagues looked deep into the past of 267 eusocial bees, wasps, and ants. They found ancestors of these lineages of eusocial insects were monogamous. Only later, once sterile castes had evolved, have some groups begun to mate more than once they reported.

To hear more about Boomsma’s hypothesis, listen to a talk he gave earlier this year at the Evolution of Society meeting sponsored by the Royal Society.

Credits: (termite hill) Summi; Atta ants: Adrian Pingstone

by Elizabeth Pennisi

In my essay, On the Origin of Cooperation, I describe experiments in which people are asked to play computer games that help reveal our cooperative tendencies, and I discuss other studies involving the use of microbes to get at the basic principles of working together. But Laurent Keller has gone a step further to work out details of social interactions. An evolutionary biologist at the University of Lausanne, Switzerland, Keller and his colleagues use evolving robots in experiments looking at the evolution of communication, an essential and complex ingredient of cooperation. The robots help him address questions that cannot be addressed through his studies of ants.

The 15-centimeter-diameter robots (see left) can move in all four directions, like tanks, and are rimmed with LEDs that can emit blue light either randomly or under control of a neural network, depending on the experiment. Each has an omnidirectional camera that sees red and blue light, as well as downward-facing infrared sensors that can distinguish gray from black. The robots are tasked with finding “food,” which emits red light. “Poison” also emits red light and can be distinguished only from up close, when the infrared sensors are able to detect a black paper circle under the poison—the food sits on a gray circle. As the experiments begin, robots flash blue when they find food, a signal others can tap into to locate the food as well

The robots have “brains” and “genomes.” The brains are the software running on onboard computers, with neural networks consisting of 10 “sensory neurons" that convey what the camera and sensors perceive to “activation neurons" that make the motors turn and that light up the LEDs. There are 33 connections in all, with different strengths, each determined by a “gene.”

The researchers tested 100 colonies, each with 10 robots, in a 300-square-centimeter arena with food at one end and poison at the other. (See movie.) The robots got points for detecting and staying by food or lost points for targeting poison. Not everyone could fit by the food, so there was some jostling for position.

Using that system, the researchers scored each robot’s fitness and took the average of the colony's members to assess the colony’s fitness. Depending on the specific test, the researchers selected a subset of the best robots or the best colonies, tweaked their "genes," and ran the experiments again—up to 500 generations' worth. In some cases, they created groups of “related” robots—all with the same “brains”—to assess the affect of kinship on the social dynamics. They wanted to know what behaviors evolved when competition for food was between relatives versus between nonrelatives.

Over time, robots evolved when and where they emitted blue light, and they came to associate high-blue-light areas with food. Foraging efficiency increased, particularly in the highly related robots chosen from the best colonies, Keller and his colleagues reported in 2007. “Reliable communication evolves either when individuals in a group are highly related or when there is increased competition between groups, and so reduced competition between individuals,” explains Keller’s collaborator Sara Mitri.

In a second series of experiments, the researchers focused on “unrelated” robots selected for their individual fitness. Cooperation theory suggests that unrelated individuals have less motivation to share information and thus “one would expect unreliable communication to evolve,” Mitri explains. After 52 generations, the robots were much less likely to emit blue light by food and instead tended to light up around the poison, Keller and colleagues reported 15 September in the Proceedings of the National Academy of Sciences.

But the researchers also discovered that some robots continued to light up at the food source, even after 500 generations. It seems that once the meaning of the blue light became ambiguous, few enough robots were attracted that competition for food diminished to the point that there was little selection against having the blue light on at the food source, they reported. The net result: individual variation in light production and responses, says Keller. That parallels what is seen in animals: variable communication strategies.

The robots are allowing us to address questions that cannot be answered with real organisms, says Keller. With respect to the evolution of communication, “there are no fossils allowing us to study how it evolved, and it is not really amenable to experimental evolution.”

Photo credit: Sara Mitri and Walter Karlen

Movie credit: Stéphane Magnenat, Matthieu Bontemps, and Kevin Frugier.

by Michael Balter

In 1984, a team led by Allan Wilson of the University of California, Berkeley, made scientific history: It published the first partial sequences of ancient mitochondrial DNA (mtDNA), from a museum specimen of the quagga, a zebralike animal that had gone extinct almost exactly 100 years before. Thus was born the field of paleogenetics, which celebrates its 25th anniversary this year. Over that quarter of a century, ancient DNA studies have opened new doors to our understanding of human evolution, tracked ancient diseases and the spread of farming, and unraveled the complex phylogenies of woolly mammoths and the bear family. But along with the triumphs have come setbacks and occasional disasters, as paleogeneticists have discovered to their chagrin how easily ancient DNA samples can become degraded and contaminated with modern DNA, giving rise to erroneous and misleading conclusions.

At her opening talk to an ancient DNA meeting this week in Paris,* Eva-Maria Geigl of the Jacques Monod Institute in Paris toured this tumultuous history. Geigl, who organizes an international meeting of ancient DNA experts in Paris every 3 years, pointed out that the field had a “tormented youth.” Just 3 years after extracting mtDNA from the quagga, Wilson’s team discovered that some of its sequences had undergone chemical alteration after the animal died, complicating attempts to figure out how closely related the extinct animal was to living horses and zebras. And in 1994, the claims of a research team to have sequenced dinosaur DNA—later discovered to be human contamination—nearly led to the premature death of paleogenetics. “DNA damage and DNA contamination almost killed the field,” Geigl said.

Happily, paleogenetics survived these early setbacks, as ancient DNA leaders began insisting on more rigorous standards. Among them were aseptic methods of excavation, as shown at left, to be sure that researchers did not contaminate samples with their own DNA, as well as laboratory controls to minimize the amplification of modern DNA. As a result, Geigl says, paleogenetics was able to regain its credibility, as shown by the successful extraction of DNA from medieval plague bacteria, extinct cave bears, early samples of domesticated cattle, pigs, maize, and wheat, not to mention Neandertals, one of the field’s greatest triumphs. After the first complete sequencing of mtDNA, from two species of moa (extinct flightless birds) in 2001, it was only a matter of time before ancient DNA researchers would turn to completely sequencing both the mtDNA and nuclear DNA of Neandertals and begin to solve that extinct hominin’s genetic relationship with modern humans once and for all.

Nevertheless, as researchers carry out ever more ambitious and sophisticated ancient DNA projects, the problems that plagued the field in its early days could threaten it again, Geigl said. This is especially true for paleogenetics studies involving poorly preserved DNA samples, such as those of the first domesticated cattle and pigs in the Near East, and where dry conditions result in poor DNA preservation. Geigl pointed out that many reagents used in molecular biology labs are prepared using products such as gelatin and bovine serum albumin, which derive from cattle and pigs and so contain traces of cattle and pig DNA. In a study carried out by her lab, Geigl reported, 13% of 1170 “blank” control samples were contaminated with bovine and porcine DNA. Another source of contamination came to light during forensic investigations into the 15-year career of an alleged serial killer called “The Phantom of Heilbronn,” who supposedly left her DNA at dozens of crime scenes across Europe. It turned out, Geigl reminded the audience, that the DNA actually came from a woman who worked in a cotton swab factory and had contaminated the swabs—used to collect DNA samples—with her own genetic material.

The answer to these problems, Geigl concluded, lies in new and improved techniques to decontaminate DNA samples before they are amplified and sequenced, some of which her own research team is currently working on. If they work out, paleogenetics can be sure of many more birthdays to come.

*Ancient DNA: From mitochondrial to nuclear DNA, from the evolution of populations to the selection of characters—25 years of paleogenetics, Paris, France, 14-16 September.

Photo credit: Courtesy of Eva-Maria Geigl