Most people don't get to use the tree-climbing skills they perfected as children once they're adults. But for Jochen Wolf, an evolutionary biologist at Uppsala University in Sweden, climbing trees is an essential part of his job. He regularly shimmies 60 feet up into the treetops, where he gingerly plucks fledgling crows from their nests and lowers them to his team below.

Wolf's climbing exploits have focused on two species of birds — carrion crows, which predominate in western Germany, and the closely related hooded crows that prevail further to the east, in Sweden and Poland. The two groups can mate with each other, but they look very different — carrion crows are black, and hooded crows have black-and-gray bodies and the birds strongly prefer mates of their own kind. For a long as anyone can remember, the two groups have remained distinct, save for a narrow band of habitat stretching from Denmark through eastern Germany to northern Italy where they sometimes intermingle.

The crows present a puzzling question to biologists, which gets to the heart of what it means to be a species: Given that hooded and carrion crows can mate and swap genes, how do the two groups maintain their individual identities? It's as if you mixed red and yellow paint in a bucket but the two colors stubbornly refused to make orange.

To understand how two similar crow species, carrion crows (top) and hooded crows (bottom), remain distinct, scientists collect blood samples and other data.

Christen Bossu (top)

To understand how two similar crow species, carrion crows (top) and hooded crows (bottom), remain distinct, scientists collect blood samples and other data.

In new research published in June in the journal Science, Wolf's team has found that a surprisingly small chunk of DNA may hold the answer. A comparison of the carrion and hooded-crow genomes showed that the sequences are almost identical. Differences in just 82 DNA letters, out of a total of about 1.2 billion, appear to separate the two groups. Almost all of them are clustered in a small part of one chromosome.  "Maybe just a few genes make a species what they are," said Chris Jiggins, a biologist at the University of Cambridge in England, who was not involved in the study. "Maybe the rest of genome can flow, so species are much more fluid than we imagined before."

The findings are striking because they suggest that just a few genes can keep two populations apart. Something within that segment of DNA stops black crows from mating with gray ones and vice versa, creating a tenuous mating barrier that could represent one of the earliest steps in the formation of new species. "They look very different and prefer to mate with their own kind, and all of that must be controlled by these narrow regions," Jiggins said.

Crows aren't alone in their behavior. A deluge of genetic data in recent years suggests that interbreeding between species is more widespread than scientists ever imagined. "I think people will be surprised and the view of species will be challenged as more data comes along," Jiggins said. "I think it will lead to a fundamental shift in how they view what a species is."

Unknown Species

The traditional way to define two related organisms as distinct species is by their inability to mate. The Swedish naturalist Carl Linnaeus, who wandered the halls of Uppsala University more than 250 years ago, employed this definition when he created the classification system we still use today. But scientists have been arguing over what makes a species for more than a century.

Charles Darwin himself declined to define the concept in his landmark book "On the Origin of Species." "Darwin, when he proved that species evolved, also proved there was no such thing as species," said James Mallet, an evolutionary biologist at Harvard University. If organisms are constantly evolving, then drawing a precise dividing line between two different species will necessarily be difficult.

Cydno Longwing and Common Postman butterflies.

Courtesy of Chris Jiggins

Despite their different appearance, the cydno longwing butterfly (top, bottom right) and the common postman butterfly (bottom left) can mate and produce viable hybrids.

Indeed, evolutionary biologists tend to take a more pragmatic approach to defining species, one that depends on their avenue of study. A distinction could be based on morphological or genetic differences, for example. "When we start speaking of species, it's in the eye of the beholder," Wolf said.

The more interesting and important question, biologists say, is what drives two populations to diverge, a process known as speciation. That question has heated up over the last five years with the rapid advance of genomic technology. Until recently, the study of speciation has focused on ecology and behavior out in the field, as well as mating experiments, but scientists now find themselves able to analyze the genomes of a menagerie of wild creatures, including closely related ones. "Just a few years ago, it wasn't possible to sequence the genome of wild organisms," Wolf said. "Now we can, and that's fantastic."

The results — from studies of crows, butterflies, mosquitoes, fish and other organisms — suggest that the concept of species is even more muddled than we thought, and that genetic changes don't always align with more visible ones, such as appearance. "In some cases, species have big morphological and behavioral changes with only a few genetic changes, and in other cases, there is lots of genetic change with few visible results," said Matthew Hahn, a biologist at Indiana University.

Birds of a Feather

When Wolf pulls himself up a branch toward a nest, the young birds aren't particularly startled to see him. Instead, "they open their mouths, waiting to be fed," Wolf said. Their parents feel differently, however, calling out from nearby treetops. "They always come back to them," said Christen Bossu, a postdoctoral researcher in Wolf's lab.

Wolf's team measures the young birds' wing length and color and collects blood samples for genomic studies before returning them to the nest. In their recent paper, the researchers not only looked at the genetic code, but also studied how gene activity varied between the two populations. They found the biggest difference in the genes that make pigment, which are active in the skin tissue and control feather color. Many of these genes lie within the DNA segment that differs between carrion and hooded crows, suggesting that somehow the pigment genes that give the two groups their unique appearance are also keeping the species separate. But how?

The most obvious explanation is that genes within this region also influence how the birds choose their mates. So-called assortative mating, in which animals that look similar are more likely to mate with each other, is one of the causes of new species development. Simple imprinting is one way to drive this phenomenon; if you were raised by a gray crow, you might prefer a gray crow as a mate.

Mosquito Mania

The malaria-spreading mosquito, Anopheles gambiae, is in some ways the polar opposite of carrion and hooded crows. Two strains of the insect, known as S and M, are indistinguishable to the naked eye, live almost on top of each other, and can interbreed, yet their genomes are strikingly different. Last year, taxonomists pronounced the strains to be two distinct species. The question they raise for biologists, however, is very similar to the one being asked about the crows: "How do you get two species from an ancestral population in which everyone is packed together so closely?" Hahn said.

One possibility is that the mosquitoes have adapted to slightly different habitats. Even though the two strains live side by side, they seem to choose different niches — one tends to live inside a dwelling and the other outside, one lays eggs in temporary puddles and the other chooses more stable bodies of water. Researchers don't yet know what genes drive these decisions, but Hahn said they have an interesting list of candidates, such as genes that would help eggs mature faster, an important trait when breeding in temporary puddles.

A second possibility ties together mate choice, color palette and vision. Maybe black crows can see other black crows more easily than they can see hooded crows and are thus more likely to mate with them, Wolf said. If the genes related to color and the genes involved in this aspect of vision sit near each other on the genome, they are more likely to be inherited together. (The further apart two genes lie on the genome, the more probable it is that they will be separated when they are passed down.) Two neighboring genes with this kind of synergistic effect on mate choice could easily drive the separation of the species. Indeed, researchers have found a gene in the region that is likely linked to vision. They believe it influences how well the birds perceive contrast, a hypothesis they are now testing in captive crows.

Adult crows are too clever to be captured, so in May, just before his paper appeared in Science, Wolf embarked on another tree-climbing excursion. In addition to blood and feathers, he collected about a dozen baby birds. They are now being raised in a new aviary in Sweden, where they are eating scientists out of house and home. (Cattle hearts are one of their favorite meals.) Researchers will train the birds to respond to visual cues, such as flashing lights, and then figure out whether hooded and carrion crows can detect different visual contrasts. According to Wolf, it's possible that black crows detect strong visual contrasts differently from hooded crows, which could explain why they seek out other black crows as mates.

Butterfly Effect

Wolf's crows aren't the only set of interbreeding species that maintain their distinct identity. Across the Atlantic, two species of heliconius butterfly — the cydno longwing (Heliconius cydno) and the postman butterfly (H. melpomene) — reside in overlapping locales in South America and can mate with each other despite their different appearance, though it happens rarely. The cydno longwing is black with white or yellow markings, and the postman is black with red and yellow markings. Each has evolved to mimic the wing pattern of a different poisonous butterfly, which helps protect them from predation. But like Wolf's crows, the cydno longwing and postman prefer to mate with their own kind.

Genome analysis suggests that the two species are swapping genes at a surprising rate. But each species has genome segments unique to its own kind, which seem to persist despite the mixing of the rest of the genome. It's as if these parts of the genome were made of oil and the rest of water; the water easily mixes but the oil remains in distinct droplets.

Chris Jiggins

Courtesy of Chris Jiggins

Chris Jiggins, a biologist at the University of Cambridge, collects butterflies near Puyo, Ecuador.

Scientists have dubbed such regions of the genome "islands of speciation." The persistence of such islands is a phenomenon that has been observed in a variety of organisms. Natural selection appears to put evolutionary pressure on these regions, which keeps both the genes and their corresponding traits distinct even in the face of interbreeding, while the rest of the genome can mix. Scientists theorize that these areas do the bulk of the work in maintaining individual species, perhaps by preserving different color patterns or mating behavior. Jiggins and others are now trying to figure out what kinds of genes reside within these islands, and how they drive two populations apart. "When you do start to diverge, what types of genes diverge first? Which genes drive speciation? What are the first things that become differentiated?" Hahn asked.

A major driver of this process may be genes that control multiple traits. "There often seem to be a few genes in the genome that have large effects, often on multiple things," said Ole Seehausen, an evolutionary ecologist at the University of Bern in Switzerland. "A gene that affects how well an individual does in one environment or the other might affect how they see each other and how they mate with each other." Genes that lie close together on the genome (such as the crows' pigment and vision genes) may have the same effect, since they tend to be inherited together.

The behavior of heliconius butterflies supports this idea. After scientists observed years of mating in the lab, they pinpointed a gene linked to wing patterning, which differs between the two species. A neighboring gene is linked to mating preference, though scientists have not yet identified the specific gene.

Taken together, the research is beginning to create a picture of the process of speciation. It might start with a small region of the genome, likely housing genes linked to mating, as seems to be the case with crows. Then that region expands, and new islands harboring other divergent genes emerge, creating islands of speciation across the genome.

Meeting Halfway

The crow hybrid zone — the narrow strip of land where the carrion and hooded crows intermingle — isn't dramatic in any way. No mountains set it apart, blocking one species from another. The landscape to the east and to the west is similar, with both species inhabiting the same type of forests. Exactly how the two groups carved out their territories is not yet clear.

The groups probably split during the ice ages, when glaciers repeatedly covered northern Europe. Crows and other animals moved south, likely taking refuge in two different locales. When the glaciers receded, the two populations moved north, meeting in the hybrid zone. But scientists don't yet know if this happened in the latest ice age, only about 10,000 to 20,000 years ago, or in an earlier one, as far back as two million years ago.

This uncertainty highlights one of the challenges of studying speciation. Sometimes two very different possible histories can produce the same genetic pattern. The shared regions of the genome surrounding islands of speciation, for example, may have other explanations, such as shared ancestry, as Hahn argued in a paper published in July. Two species might have similar genomes not only because they recently swapped genes, but because they share a distant parent species. "People went overboard with interpreting islands of speciation," Hahn said.

Carrion and hooded crows could be quite old species whose genomes became similar through interbreeding. Or they could be quite young, having split off from a common ancestor relatively recently, with the small chunk of divergence the first sign of speciation. Wolf's group favors the latter interpretation but hopes to address the question directly with further genetic analysis.

So what does all this mean for the definition of species? Scientists still don't have a definitive answer. Simply defining species based on genetics doesn't solve the problem. As Wolf and others have shown, the answer depends on where in the genome you look. "It's really hard to draw a boundary," Wolf said. "Different parts of the genome tell you different things."