A new approach to cataloguing biodiversity, known as DNA bar-coding, has identified more than a dozen potentially new North American bird species previously overlooked by legions of ecologists and bird-watchers. The birds, which are largely indistinguishable from previously known species by look or sound, were identified by slight variations in a tiny piece of DNA.

Researchers ultimately hope to use the technique to create a database of almost all life on earth. “These findings set the groundwork to expand this initiative not just in birds but across animal life and beyond,” says Mark Stoeckle, a molecular biologist at Rockefeller University in New York City, who was involved in the project.

Taxonomists typically identify species on the basis of factors such as what the organism looks like, where it is found, and how it behaves. But those methods can be slow and require expertise in a narrow field, such as tropical ants or Pacific algae. DNA bar-coding, on the other hand, provides a quick and efficient way to identify many organisms.

First conceived of five years ago by Paul Hebert, an evolutionary biologist at the University of Guelph in Ontario, DNA bar-coding focuses on a specific piece of mitochondrial DNA, genetic material that resides outside cell nuclei and mutates relatively frequently. Previous studies have shown that this piece of DNA–the bar code–varies more between species than within species. In North American birds, for example, closely related species have 18 times as much difference as two members of the same species.

For the new study, published this week in the journal Molecular Ecology Notes, Hebert, Stoeckle, and collaborators analyzed specimens from museums and samples collected by birders–a total of 643 species, about 93 percent of all bird species that breed in or regularly visit the United States and Canada. They identified 15 groups of birds that were previously assumed to be members of existing species but had significantly different bar codes. “I was surprised,” says Stoeckle. “People have been looking at North American birds for a long time.” They also found evidence that some species thought to be separate, including eight species of gulls, have the same genetic bar code and therefore may actually be members of the same species.

The findings provide important validation for the bar-coding concept. The technique was viewed with skepticism early on by ecologists and taxonomists who doubted that one small piece of DNA could be used to reliably identify species. But the researchers found that bar-coding results corresponded to the traditional species classifications about 95 percent of the time. “Analysis shows that in the vast majority of cases bar-coding worked,” says Jim Edwards, director of the Global Biodiversity Information Facility in Denmark, an organization that promotes global sharing of biodiversity data. “I don’t know why there is divergence at the species level on this one little piece of DNA, but it works.”

Scientists caution that the bar-coding technique cannot definitively identify new species; genetic results must be used alongside more traditional taxonomic methods. In the bird study, for example, it’s not yet clear what to conclude about the gulls that share a genetic bar code. “Some of them may be the same species, or they may be young species that haven’t yet had time to evolve differences in this small region,” says Stoeckle.

Universities, museums, and other organizations around the world have created an international consortium to further develop the technique. Projects in progress are cataloguing the biodiversity of birds, fish, ants, and fungi. Hebert and collaborators have catalogued about 25,000 species to date, at a rate of about 2,000 specimens a week. They aim to analyze 500,000 species by 2014. “The goal is to register effectively every organism humans are likely to encounter,” he says.

The technology could eventually be used to identify crop pests or invasive species, or to monitor trade in endangered species. A fast way to catalogue species diversity could also be useful for conservation efforts. Scientists are now cataloguing various species in areas of Madagascar and the Amazonian rain forest, two areas undergoing rapid development. “Immense blocks of territory are under threat of development,” says Hebert. “They want to set aside regions of highest unique biodiversity, and we come in with a fast survey tool.”