The Turks and Caicos anoleColin Donihue

The lizards didn’t see the hurricanes coming. Neither did Colin Donihue.

Last summer, Donihue, a researcher from Harvard University, traveled to the Caribbean islands of Turks and Caicos to study a local species of anole lizard. Conservationists were set to exterminate rats that had been introduced to the two islands to preserve their native wildlife, and Donihue wanted to see how the lizards might evolve once the rodents were gone. He and his colleagues captured dozens, and measured their bodies, legs, and toes. Then, in early September, they packed up and flew home, with a vague plan to return in a few years and measure the lizards again.

Four days later, Hurricane Irma arrived. It battered Turks and Caicos with 165-mph winds that created 20-foot waves, destroyed homes, snapped power lines, and killed at least 14 people. Two weeks later, Hurricane Maria delivered a second blow, only slightly less powerful than Irma’s.

Donihue and his team realized that they had a rare chance to see how natural disasters change the evolutionary fate of a group of animals. After all, they had been the last to observe the anoles before the hurricanes struck. So in October, they flew back to the islands.

“I wasn’t really sure what to expect,” Donihue says. He found that the islands had clearly taken a severe hit, but were also rebounding quickly. Many trees had been uprooted and stripped bare, but those that had survived were already putting out new leaves. And among the fresh greenery, there were anoles.

The lizards can be found throughout the Caribbean, clinging to twigs and tree trunks with their sticky toes. Whenever Donihue spotted the lizards, he would lasso them with loops of string at the end of modified fishing poles. (“It doesn’t hurt them, because they have very strong neck muscles,” he says. “It’s much easier than running up and grabbing them with your hands.”)

“To be honest, given how catastrophic hurricanes are, I thought it was plausible that survival would be random—that there wouldn’t be an advantage that would help [the lizards] survive,” he says. But when he compared the survivors’ measurements with those of the pre-hurricane population, he realized he was wrong.

He found that, on average, the post-hurricane lizards had toe pads that were 6 to 9 percent bigger than those of pre-hurricane individuals, and front legs that were 2 percent longer. This wasn’t because the bodies of specific lizards had changed; there’s no evidence that the toes of adult anoles can grow by that amount. Instead, the storms had simply wiped out all the lizards with small toe pads. By selecting for individuals that were better at clinging to surfaces—and presumably at withstanding high winds—the storms had changed the average proportions of the population.

These sound like small changes, but natural selection famously works on small physical variations, favoring some over others across many generations. “The changes were subtle, and we couldn’t have noticed them just by holding the lizards in our hands,” Donihue says. “But they were consistent between the two island populations, which makes us feel more confident that this wasn’t a fluke.” The variation in the toe and leg measurements narrowed after the storms, too, adding further evidence that the hurricanes had selected for lizards with particular kinds of bodies.

But one trend didn’t make sense: After the hurricanes, the average length of the lizards’ hind legs was 6 percent shorter than before—the opposite pattern from their front legs. “This was a real head-scratcher,” Donihue says.

He and his colleagues worked out what had happened by placing the lizards on small wooden posts and subjecting them to gusts from “the largest leaf blower we could find,” he says. The lizards would shimmy to the sheltered side of the posts, tuck their front legs close to their bodies, and cling for dear life. But because of the way their legs are structured, their hind thighs would always jut out. These exposed thighs caught the wind like sails, and would eventually rip the lizards from their secure footing (and into the safety nets the team had set up).

Once the team realized this, everything clicked into place. “With shorter thighs, you’re catching less wind,” Donihue says.

A lizard clings to an experimental perch in hurricane-force winds. (Colin Donihue)

“This is a striking case of rapid evolution, which, as we can see here, can proceed exceedingly fast, even within a generation,” says Carol Lee, who works at the Center for Rapid Evolution at the University of Wisconsin at Madison. “I expect there will be many more cases like this in the future, where catastrophic events impose strong selection on populations, and where populations will need to evolve or go extinct.”

For decades, Donihue’s colleagues, led by Jonathan Losos, have been documenting many examples of external threats quickly shaping the evolution of anoles. They’ve shown that some species evolved longer legs to better flee from invasive predators that were introduced to their islands. They’ve found that other anoles evolved larger toe pads to more effectively climb onto higher branches when new competitors drove them out of their usual low-lying habitats.

The team has also found examples of natural selection imposed by natural disasters. In 1998, they showed that over a 20-year period, anoles in the Bahamas tended to have longer legs in the months or years after hurricanes had hit. And just last year, they found that the extreme winter that hit the southern United States in late 2013 selected for Texan anoles that were more tolerant of the cold.

No one knows what will happen to the Texan anoles, or those in Turks and Caicos, in the long run. Now that the hurricanes have passed, the lizards are probably facing the same evolutionary pressures that they faced before, which might push their proportions back to baseline levels. Then again, another hurricane season looms. If these storms  mirror the severity of those in 2017, they might continue to direct the evolutionary fates of the anoles—and other Caribbean animals.

“I think we’ll find more and more of these studies coming out, since extreme climate events are becoming more frequent and more severe,” Donihue says. “Again and again, we’re seeing that these extreme events can have an evolutionary impact.”

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.