Major evolutionary study rewrites bird ‘tree of life’

June 26, 2008

GAINESVILLE, Fla. — The flamingo looks like it should be closely related to the stork or crane, but its closest relative may actually be the diminutive, modest grebe.

Falcons would seem to have evolved together with hawks and eagles, but the species are not close kin, appearing to have acquired their hawk-like characteristics completely independently from their look-alikes.

The largest group of birds, the passerines — which include such common North American birds as mockingbirds, blue jays and wrens — may be most closely related to a group not common at all in our latitudes: parrots.

These are among the surprising findings of the largest-ever study of the evolution of birds described in an article set to appear Thursday in Science. The study — conducted in large part by five institutions, including the University of Florida — challenges current bird classifications, alters understanding of avian evolution and provides a valuable resource for studying the evolutionary history of birds.

“We basically tried to make the next big step in understanding how different types of birds are related to each other,” said Rebecca Kimball, an associate professor of zoology at UF and one of three lead authors of the Science paper. “There has still been a lot of controversy about this issue. Some people have suggested that it is an unsolvable problem.”

Birds are among the most studied and loved animals, and much of what is known about animal biology — from natural history to ecology, speciation and reproduction — is based on birds. Nevertheless, the avian tree of life has remained elusive — until now.

For more than five years, the Early Bird Project has been examining DNA from all major living groups of birds. So far, scientists have built a dataset of more than 32 kilobases — a kilobase is a DNA fragment containing 1,000 base pairs — from 19 different locations on the DNA of each of 169 bird species. The result is equivalent to a small genome project.

“Our study and the remarkable new understanding of the evolutionary relationships of birds that it affords was possible only because of the technological advances of the last few years that have enabled us to sample larger portions of genomes,” said co-lead author Shannon Hackett of The Field Museum in Chicago. “Our study yielded robust results and illustrates the power of collecting genome-scale data to reconstructed difficult evolutionary trees.”

The study’s results are so broad that the scientific names of dozens of birds will have to be changed, and biology textbooks and birdwatchers’ field guides will have to be revised. The study resulted in a number of new or expanded findings about bird evolution:

• Birds adapted to diverse environments several distinct times because many birds that now live on water did not evolve from a different waterbird group, and many birds that now live on land did not evolve from a different landbird group.
• Distinctive bird lifestyles, such as birds that are nocturnal or raptorial, evolved several times. For example, contrary to conventional thinking, colorful, daytime hummingbirds evolved from drab nocturnal nightjars.
• Shorebirds are not an early evolutionary group, which refutes the widely held view that shorebirds gave rise to all modern birds.

“We have come a long way, but we haven’t answered every question,” Kimball said, adding that the study both confirmed and rejected traditional ideas about bird evolution. For example, she said, researchers confirmed the results of several other studies showing that chickens and turkeys are closely related to ducks and geese. But the study’s finding that passerines are close relatives of parrots is new and surprising, because parrots are anatomically very different from the birds in the passerine group.

Kimball said labs at each of the major institutions involved in the study took charge of sequencing separate genes from the same bird species. Because of the enormous amount of data involved in the sequencing work, the UF researchers, led by zoology assistant professor Edward Braun, sought to predict how many genes were needed for a satisfactory result.

“We actually tried to simulate the evolutionary process,” she said. “If we hypothesize that birds evolved in this rapid evolution, how much data would we need to have a chance to start to find those bits of information that would help us piece together these relationships? We used that simulation as our guide in how we set up the project.”

The UF portion of the research was funded with a $455,000 grant from the National Science Foundationthrough its Assembling the Tree-of-Life Research Program. Other UF authors were Braun and David Steadman, a UF ornithologist, both co-principal investigators on the project with Kimball. Graduate students Andrew Cox, Kin-Lan Han and postdoctoral researcher Tamaki Yuri also served as co-authors