New Species Arise More Quickly Than Previously Believed, UMass Amherst Researcher Finds

AMHERST, Mass. - The splitting of a species into two new species may occur in far fewer generations than scientists previously believed, according to a study led by University of Massachusetts postdoctoral researcher Andrew Hendry. Hendry, an evolutionary ecologist, conducted his study on two populations of sockeye salmon in the Pacific Northwest. The findings are published in the Oct. 20 issue of the journal, Science. His co-authors are Paul Bentzen and Thomas Quinn, both of the University of Washington, John Wenburg of the University of Montana, and Eric Volk of the Washington State Department of Fish and Game.

"There is a widely-held perception that when one population splits into two different environments, traits evolve quickly and, as a result, the two new populations become less likely to interbreed. That is, they become reproductively isolated. This process, called ecological speciation, may be one of the easiest and fastest ways that new species arise. Our results suggest that this perception may not only be correct, but in spades," said Hendry. "The classic examples of ecological speciation are for groups that have existed for 10,000 years. Even the fastest examples are for some insects over 200-400 generations. In these cases, we know reproductive isolation evolved sometime in the past, but we don''t know how quickly."

In contrast, Hendry''s team found evolutionary adaptations and reproductive isolation in salmon after only 12-14 generations: some 60-70 years. Specifically, scientists studied salmon introduced into Lake Washington, in Washington State, during the 1930s and 1940s. Soon after the initial introductions, two populations became established, one spawning in a river and one along a lake beach. "Sockeye salmon bury their eggs and spawn in different kinds of locations, and in a variety of environments, even in a small system such as this," Hendry explained. "When new populations become established at different sites, you''d expect them to evolve different adaptations, and that''s in fact what happened."

Some of the differences between river salmon and beach salmon included:

* Body depth of males - male beach salmon are "deeper" from back to belly than their river counterparts. This trait influences mating success, Hendry says. "Deeper-bodied males gain access to more females during mating but, in a river with a strong current, a deep-bodied male would be inefficient hydrodynamically, and would have a more difficult time, so river fish have adapted by becoming slimmer, more streamlined."

* Size of females at breeding age - river females are larger than their beach counterparts. Hendry points out that the river female''s larger size enables her to bury her eggs deeper in gravel, decreasing the chances that the eggs will be disturbed or destroyed by high water flows.

"Rapid evolution has also been documented for other organisms," Hendry points out. "The unique twist to our study is that we were able to demonstrate these differences resulted in reproductive isolation." Scientists examined tiny ear bones known as otoliths, which have a sort of bar-code identifying each fish as to where it had been born, the river or the beach. "We can look at breeding adults and know who is a river resident, who is a beach resident, and who is a beach immigrant - a fish born in the river but now breeding at the beach," said Hendry. "We found that a large percentage of the adults spawning in the beach came from the river - almost 39 percent each generation," he said. "If those fish were successful in producing offspring, the two populations would homogenize."

Genetic analysis of "beach immigrants" relative to "residents" revealed just the opposite: immigrants are considerably less successful at producing offspring. This difference may have arisen because immigrants have reduced mating success or, if a beach immigrant does breed, the resulting hybrid offspring will be less likely to survive, Hendry said, because they''re less than ideally suited to either environment. This evidence demonstrates for the first time how rapidly adaptation can lead to reproductive isolation, and it is about 10 times faster than the previously accepted maximum. "This should really make us rethink the importance of natural selection and adaptation to the rapid generation of new species and the generation of biological diversity."

Hendry does offer a word of caution: "Despite our findings of rapid adaptation and reproductive isolation, I don''t necessarily presume these two salmon populations will evolve into what would be recognized as separate species. We have simply used new populations to demonstrate the same processes that lead to new species."

Hendry is a Darwin postdoctoral fellow, and conducts research in the biology department''s Organismic and Evolutionary Biology program. He conducted his graduate work at the University of Washington in Seattle.

NOTE: Andrew Hendry can be reached at 413/545-3565 or

Paul Bentzen can be reached at 206/685-9994 or

Thomas Quinn can be reached at 206/543-9042 or

John Wenburg can be reached at 406/243-5503 or

Eric Volk can be reached at 360/902-2759 or