Dull birds and bright ones beat so-so guys.

If you can't look brilliant, forget halfway-decent looks. Go for total loser.

That fashion tip works--in male lazuli buntings--because their blue plumage shows signs of a rarely documented evolutionary pattern called disruptive selection, contend Erick Greene of the University of Montana in Missoula and his colleagues in the Oct. 26 NATURE. Disruptive selection favors individuals with either of the opposite extremes of a trait and discourages moderation.

Among year-old males, "the really, really dull guys and the really, really bright guys do best," says coauthor Bruce E. Lyon of the University of California, Santa Cruz. Those extremes tend to win mates and good territories, whereas the males with moderately blue plumage lose out.

Lyon describes disruptive selection as one of three ways evolution shifts a trait. In the other cases, directional selection drives the trait toward one extreme, and stabilizing selection favors moderation.

To study western North America's lazuli buntings, Passerina amoena, researchers banded nearly 200 males vying for nesting territories in Montana. In a deluxe habitat, shrubs covered 92 percent of land, but in a bunting slum, only 3 percent.

Yearlings have the interest in and physiology for reproduction but vary widely in plumage splendor. Monitoring revealed that intermediately blue birds were the least likely to win a mate, hold a shrubby territory, and sire chicks. "It was a total shock," Lyon says.

The triumph of the brilliant youngsters made sense, he remembers. The researchers now argue that the unexpected success of drab birds comes from older males showing less aggression toward them than toward flashier studs. Thus, a dull youngster has a decent chance of winning territory near a bright adult.

Both birds benefit in such a neighborhood, the researchers contend. Paternity tests of chicks reveal that other males, possibly the tolerant neighbor, sneak visits to drab yearlings' mates. Even then, the biologists found, a drab yearling sires more chicks than a moderately blue one does.

Lyon says the next question for researchers is how much the plumage color depends on genes versus the environment.

"I think it's particularly interesting to find disruptive selection in plumage," says Thomas Smith of San Francisco State University, who described a famous disruptive-selection example in African finch beaks.

Feather color seems important in forming species, Smith notes. So, could disruptive selection have contributed to today's rainbow of birds?