Evolution not as unpredictable as thought, study says


Say you could hop into a DeLorean and travel back to when life on Earth began. Would fish migrate from water to land? Would the dinosaurs go extinct? At the end of our trip, would we still encounter life as we now know it?

Some scientists don’t think so. They argue that any number of chance events — storms and earthquakes, for example — would steer evolution down another course, making it impossible to predict.

But a study published Tuesday in Science has found that if we know the ecology of an area, we can predict the traits a species will evolve millions of years from now, despite all the chance events that could influence the outcome.


While most biologists agree that the natural environment can predict evolutionary changes over a few generations, whether it can do so over millions of years has remained “contentious,” said UC Davis population biologist and study leader Luke Mahler. Longer time spans mean more opportunities for chance events to occur.

But since “we can’t rewind the tape of life and replay it again” Mahler said, settling the controversy has been difficult. But Mahler approximated this time travel by studying evolution in anole lizards in the Greater Antilles islands -- Cuba, Hispaniola (which comprises the Dominican Republic and Haiti), Jamaica, and Puerto Rico -- which was like screening the tape at “four different theaters all at the same time,” he said. The islands are also ecologically similar.

Mahler and his colleagues found that nearly 90% of the time, anoles on different islands had evolved highly similar traits — a phenomenon called convergent evolution — and that their similar environments, not chance alone, can account for this similarity. That allowed them to conclude that, despite all the chance events that occurred over each species’ history, comparable ecological pressures had led them to comparable evolutionary outcomes.

Mahler decided to study the tiny, color-changing anoles because they’ve been evolving on the Greater Antilles for 40 million years, allowing him to examine their evolution over a long time scale. In addition, biologists had already constructed detailed anole evolutionary trees, making it easy to identify different species.

Scientists have observed convergent evolution in a few anole species in the Greater Antilles, but no one had done an analysis that considered every anole species, including those unique to different islands.

Strained diplomatic relations between the U.S. and Cuba made traveling to the island “ a hurdle” for scientists, Mahler said. In Haiti, safety concerns, especially after the 2010 earthquake, similarly stymied fieldwork there.


But eventually, Mahler was able to spend two years measuring an array of physical attributes in 100 out of 119 known anole species in the Greater Antilles, relying on both preserved museum specimens and animals in the wild. He used calipers to measure the lengths of their bodies, legs and tails, and counted the number of sticky pads on their toes.

Anoles are known to have adapted these physical traits specifically to survive in different ecological “niches.” For example, field anoles have long tails that help them cling to swaying grass blades, while treetop-dwelling anoles have a high number of sticky toe pads well-suited for climbing.

Mahler and his colleagues then developed several statistical models of evolution, testing to each to see which one best explained the physical characteristics they had observed. The best fit was one that relied on the idea of “adaptive peaks,” or particular combinations of features that natural selection would favor.

By that token, similar evnironments should have similar adaptive peaks -- collectively known as an “adaptive landscape” — but no one had tested the idea, until now.

The adaptive peak model “blows all those other models out of the water,” Mahler said. “It leaves them smoking.”

His analysis also revealed that the convergence he and his colleagues had observed wasn’t due just to chance. “You expect some convergence by chance,” he said. But in models with a high degree of chance, “you don’t get anything resembling the convergence that we see.”


Although the anoles inhabited different islands, the islands’ ecological similarity meant they faced the same adaptive landscape with the same niches to exploit. As the lizards diversified, their traits converged on the same peaks — ones that make them twig or leaf specialists, for example.

The only places where the anole species diverged from the handful of adaptive peaks were Cuba and Hispaniola — larger islands where anoles are more likely to find and exploit unique niches not found on the other islands.

Mahler’s findings are consistent with studies that have found similarities among cichlid fish species in different lakes in the African Rift. But the current study is “the first to show in a quantifiable way just how this [convergence] happened,” said Rosemary Gillespie, an evolutionary biologist at UC Berkeley who was not part of the study.

But Gillespie noted that, although the paper shows that ecology can predict evolution on small islands, “whether the same kind of thing happens on a continental scale we don’t know exactly.” Still, the paper shows this prediction is possible, she said.

Others have already begun testing Mahler’s ideas by unwittingly introducing anoles to Hawaii, a region ecologically similar to the Caribbean. (The anoles often hitch rides on ships carrying plants or produce.) If his conclusions hold up, the lizards should eventually resemble their Caribbean counterparts.

“I think it’s going to be an amazing experiment to come back in 3 million years and see how they turn out,” he said. “That’ll be the real test of our ideas.”


Return to Science Now.

Twitter: @mmpandika