Surf's up!: How rafting lemurs colonized Madagascar

Who doesn't love lemurs? The strepsirrhine primates, or wet-nosed cousins of ours, are favorite documentary subjects and extremely popular zoo attractions. And, in one of those bits of zoological trivia that everyone knows, lemurs only live on the island of Madagascar off Africa's southeastern coast. The question is how they got there.

Documenting the paths of animals during geological history is not an easy task. In the days before scientists understood plate tectonics, land bridges, now sunk beneath the ocean, were often used to explain the dispersal of organisms. While some land bridges did exist in the past, like the one that allowed mammoths to cross from modern-day Russia to North America, they were not nearly as widespread as had once been thought. Instead many scientists began to think about how organisms might float their way to new places by becoming accidental passengers on bits of vegetative detritus. As articulated by paleontologist G.G. Simpson, this was a kind of "sweepstakes" in which creatures would be cast out to sea on floating mats of plant matter and of those wayward animals a few might be washed up in a new habitat able to support them. From these few survivors of tropical storms entirely new ecologies could become established.

The trouble with this was that the currents surrounding Madagascar circulate in a way that would make it very difficult for any raft to make it to the island. Maybe the unique island fauna could be attributed to a land bridge, after all. Neither option seemed entirely satisfactory, but, in a paper just published in Nature, scientists Jason Ali and Matthew Huber took another look at the sweepstakes hypothesis. As it turns out, the currents surrounding Madagascar might have been more of a help than a hindrance in transporting rafting animals to the island.

During the Late Cretaceous Madagascar was home to dinosaurs, including the knobby-headed predator Majungasaurus, but 65 million years ago they died out along with the other non-avian dinosaurs in the rest of the world. At this time Madagascar was already separated from the rest of Africa, but this did not stop it from being colonized by mammals. Studies of the genetics of Madagascar's living inhabitants have indicated that the ancestors of its modern-day fauna, such as the primates that gave rise to lemurs, started to arrive soon after the extinction of the dinosaurs. The ancestors of lemurs were among the first to arrive, between 60-50 million years ago, followed by tenrecs between 42-25 million years ago, carnivorans between 26-19 million years ago, and rodents between 24-20 million years ago. Representatives of these mammal groups clearly did not walk over all at once, as might be the case with a land bridge, but instead arrived bit by bit over tens of millions of years.

Just because the land bridge hypothesis is not well-supported, however, does not mean that we can safely assume that the sweepstakes hypothesis is correct. As the authors of the paper note, critics of the rafting hypothesis have cited the present currents and winds that move south-southwest and thus would prevent rafts from making it to the eastward island. If the same situation was true in the past then it could be safely assumed that whatever early primates found themselves adrift would be deposited back along the African coast, if they returned to shore at all.

But there is no reason to believe that the winds and currents around Madagascar have remained constant during the 120 million years that the island has been separate from the African continent. During the past 60 million years alone multiple ocean gateways have opened and closed, and both Madagascar and the African mainland have moved over 1,650 km towards the equator. Given all these changes it must be considered whether the flow of water around Madagascar has been altered, and Ali and Huber have answered in the affirmative.

Unfortunately the intricacies of prehistoric ocean currents cannot be observed directly, so the authors of the new study simulated the behavior of past currents using computer modeling. After accounting for what is known about the position of continents and climate over the past 60 million years the authors found that the currents around Madagascar were quite different in the past. In particular, the model predicted that during the Eocene (about 56-34 million years ago) there was a "vigorous eddy" off the east coast of Madagascar that would have drawn whatever drifted off the coast eastwards towards the island rather than south along the channel as occurs today.

Yet the normal flow of these currents would not have been fast enough to transport living animals to Madagascar. Under normal conditions it simply would have taken too long. Instead, it appeared that there were periodically faster currents at certain times of the year that could allow rafts to cross the channel in 25-30 days, and if tropical storms formed in the area (as seems likely) rafts of vegetation might have been given an even faster journey. This could explain why the dispersal of mammals to Madagascar took so long. The survival of the animals on the rafts was contingent on peculiar conditions that only occurred for a few weeks every year.

This pattern was in place through the Eocene and into the Oligocene, but by the early Miocene (about 23 million years ago) Madagascar had shifted enough that the currents changed again. The pattern of currents became more like what we are now familiar with, and they cut off the island from colonization by rafters (though animals that could swim and fly could still arrive). The fauna of Madagascar could thus continue to evolve in near-isolation; a sort of evolutionary experiment that itself was contingent upon which groups won the dispersal sweepstakes. This makes the destruction of Madagascar's native fauna all the more tragic. Lemurs, like many other animals there, exist nowhere else. If they are wiped out we will have ripped apart one of the most fascinating evolutionary stories we have ever come to understand.

Ali, J., & Huber, M. (2010). Mammalian biodiversity on Madagascar controlled by ocean currents Nature DOI: 10.1038/nature08706