In 2005, scientists left one male and one female lizard on each of seven empty Bahamian islands; four years later, every colony had evolved shorter legs but still carried the body shape of its founding pair |

Scientists refer to this phenomenon as the “founder effect,” which occurs when a small number of founding individuals unintentionally carry characteristics that are not typical of the larger group they originated from. Since most real-world examples of founder events occurred long before anyone thought to measure them, it is challenging to test that theory effectively.Nature provided scientists with a unique opportunity to actually test this theory. A hurricane that hit a portion of the Bahamas in 2004 entirely destroyed the lizard populations on several tiny islands. Rather than let those islands simply be recolonised at random, a team led by biologist Jason Kolbe used the opportunity to run a controlled experiment. The results were published in Science in 2012. It showed how randomly selected founding pairs from the same large island source established brown anole populations on seven small islands, allowing researchers to track founder effects alongside subsequent natural selection.Building seven colonies from the ground upIn 2005, the crew released precisely one male and one female brown anole onto each of the seven recently vacated islands after capturing them from a single, sizable, densely forested island close to Great Abaco. Any early variations between the seven new colonies could only be explained by the random genetic makeup of that specific pair, not by external causes, since each founding pair originated from the same source population. In principle, the scrubby, narrow-limbed plants on the seven experimental islands should eventually promote shorter legs, while the source island favoured longer-legged lizards that were appropriate for its dense forest branches.Chance left its mark within a single yearDuring the first year, some island populations had noticeably longer hindlimbs, while others had shorter ones, and those differences were unrelated to the vegetation on the islands. Some islands already had noticeably longer-legged lizards in the first year, some shorter-legged ones, and this separation had nothing to do with which island had thicker or thinner vegetation. It was the kind of random signature predicted by the founder effect, reflecting which particular founding pair happened to establish each colony.

Selection drove all the same, but order never alteredResearchers tracked limb length and genetic markers in all seven experimental colonies and compared them with populations on nearby source islands over the next four years. All populations gained shorter hindlimbs slowly, as predicted, in response to the smaller island vegetation; natural selection was indeed acting on the lizards exactly as predicted. But the islands with the longest-legged founders at first still had the longest-legged lizards four years later, and the same was true at the shorter end of the scale. A 2012 summary in Nature Genetics noted that the team observed morphological adaptation and genetic variation in the island populations from 2005 to 2009, while founder effects persisted throughout the study. Chance determined where each colony began, and selection determined which way it moved from there, but neither force alone could explain what the lizards on those seven small islands finally became.