In a very remote archipelago in the South Atlantic, an extinct volcano’s near-vertical cliffs protrude from the sea and meet to form a high plateau. The strange land mass, which looks like a birthday cake that’s been dropped into the ocean, is known as Inaccessible Island. No humans have lived there since 1873, but it is swarming with one species of small, strange, flightless bird. The question is: How did they get there?
Scientists have been asking this for a long time. A century ago, ornithologist Percy Lowe speculated that the ancestors of Atlantisia rogersi, the Inaccessible Island Rail, were also flightless and made it to the remote island by scuttling along now-submerged land bridges. But in the new Molecular Phylogenetics and Evolution study, an international team of scientists reveal a far more realistic — and impressive — backstory.
Martin Stervander, Ph.D. is an evolutionary biologist and lead author on the paper. “The answer we uncovered is that the ancestors of these rails colonized the island about 1.5 million years ago — which is not very long ago, in evolutionary times,” Stervander tells Inverse. “It seems like the birds flew about 2,174 miles from South America and then landed on Inaccessible Island, which was probably one of the first pieces of land that they saw.”
Stervander, now a postdoctoral researcher at the University of Oregon, studied the Inaccessible Island Rail during his time at Lund University. He journeyed to the island to study finches, but as Stervander recalls, he and his colleagues knew that “there were these incredibly weird, and odd, endemic flightless rails and nobody knew their evolutionary history.” So they decided to set up a net, down low on the ground. When they started playing the rail’s calls, two birds quickly ran into the net.
Catching the rails was a success in itself. Stervander says many species of rails — even those that can fly — are quite secretive, sticking close to fairly dense vegetation. When Stervander was on the island, it took five days before he even saw one. He then only saw four more. Despite their apparent scarcity, Inaccessible Island is densely inhabited by these birds: Stervander says that the latest estimate pegs the population to about 5,600 individuals.
“That basically means they are everywhere,” Stervander explains. “You can hear them all the time, calling and grunting. They scurry around in the vegetation like rodents.”
Catching the rails allowed the team to analyze their DNA and determine that their closest now-living relatives are the dot-winged crake in South America and the black rail found in South and North America. Once the common ancestor of these rails flew to Inaccessible Island, the species changed in a number of ways. Its bill became longer, its legs became sturdier, and its coloration slightly changed. But perhaps most fascinatingly, the Inaccessible Island Rail lost its ability to fly.
That’s because on Inaccessible Island there is really no need to fly. The birds can get food — flightless moths, berries, seeds, and worms — by walking around on the ground. There are no mammals or predators to the rails on the island, so there’s nothing to fly away from. Over time, the rails that invested less in flying mechanisms thrived because, on Inaccessible Island, natural selection doesn’t reward the ability to fly.
“It’s not like they lost their wings, but the wings are greatly reduced — they are short and their flight feathers are very short,” Stervander says. “Importantly, their flight muscles have been reduced dramatically. This reduction has happened repeatedly both for other rails and birds that end up on isolated islands because it’s quite energetically costly to keep these muscles going.”
And when Stervander says that Inaccessible Island is remote, he means remote. “I would argue it’s very aptly named,” Stervander laughs. “It really, really is quite inaccessible.”
He voyaged to the island — stuck in the middle of the ocean between South America and Africa — on a research vessel that took off from Cape Town, South Africa. The scientists journeyed for a week and then were flown by helicopter to the island itself. They were lucky to get this ride — this research vessel only travels to Inaccessible Island once a year. Researchers who miss the boat can take one of the approximately 15 postal or fishing boats that journey to Tristan da Cunha per year, then ask another fishing vessel to take them to Inaccessible Island itself.
That trip, Stevander explains, requires a “flexible” schedule. Peter Ryan, Ph.D., a co-author of this paper and a professor of ornithology at the University of Cape Town, was stuck on a fishing boat for five weeks before it could actually land on Inaccessible Island.
“It’s just so incredibly exposed to weather conditions,” Stervander says. “There are really only two or three places you could land in favorable conditions and really only one of them is on the right side.”
Yet, the remoteness of the island is a boon for these small, black birds. Stervander explains that if mammals, like rodents, were ever accidentally introduced to the islands, “it’s likely the rails would disappear very fast.” They’ve evolved to perfectly exist within very specific conditions, thriving in a world where they don’t need to fly.
Rails (Aves: Rallidae) are renowned for their extreme dispersal capability, which has given rise to numerous island lineages. Many insular species lost the ability to fly as a response to release from predator pressure—a feature causing rapid extinction when humans subsequently introduced mammals. The world’s smallest extant flightless bird, the Inaccessible Island Rail *Atlantisia rogersi, is endemic to Inaccessible Island, Tristan da Cunha archipelago, in the central South Atlantic Ocean. It is placed in a monotypic genus, but its taxonomic affinity, as well as geographic origin, are disputed. Contrary to its suggested Old World origin, we demonstrate that the Inaccessible Island Rail is nested within the mainly South American ‘Laterallus clade’ and that it colonized ≥3 million-year-old Inaccessible Island from South America c. 1.5 million years ago. The taxonomy of rails has traditionally been based on morphology, and convergent evolution has caused many cases of misclassification. We suggest a re-classification within the ‘Laterallus clade’ and call for extended coverage of taxon sampling for DNA sequencing.