Australasian Science: Australia's authority on science since 1938

Escape to Madagascar

Propithecus diadema, the diademed sifaka.  Credit: Mitchell Irwin

Propithecus diadema, the diademed sifaka. Credit: Mitchell Irwin

By Karen Samonds

Madagascar’s bizarre assemblage of fauna didn’t evolve from the fossils found on the island, so how did they get there?

‘‘For naturalists Madagascar is the true Promised Land. Nature seems to have withdrawn there into a private sanctuary, to work on models different from any she has used elsewhere. There you meet the most unusual and marvelous forms at every step . . . What an admirable country, this Madagascar!’’ Joseph-Philibert Commerson, 1771

The sun was hot, and my mind was off wandering again. Tired, and covered with sand and sunscreen, I looked again at the 40-million-year-old fossil bone protruding out of the side of the ancient cliff. Could it be the side of a crocodile vertebra? Maybe a dugong rib? Or, simply another fragment of non-identifiable turtle shell?

Slowly I scraped away at the surrounding rock hoping that it would reveal itself to be something truly spectacular. In fact, there are lots of tantalising things it could be as our current understanding of this critical time period in Madagascar – the past 80 million years – is virtually a blank slate. That’s a problem, because the early fossils we have found do not seem to have much to do with the bizarre assemblage of animals living on the island today.

So how did the current residents get there if they didn’t evolve from the fossils currently known? Did they walk or fly or float from the nearest mainland? That’s the question we set ourselves, and from our calculations based on the different animals, when they arrived, where the island was, and what the winds and currents were doing, we think that most of them probably rafted their way across to Madagascar from Africa.

For centuries, naturalists have been drawn to Madagascar’s bizarre plants and animals. Long considered a biodiversity hotspot, Madagascar is the fourth largest island in the world. While it currently lies close to Africa, it shares a long history with Australia, Antarctica, South America and India as part of the Gondwana supercontinent 200 million years ago, separating from these other continents 85–90 million years ago. Despite all of the well-deserved attention the island’s living animals receive, the details of how these strange creatures made it to Madagascar remains largely a puzzle.

The fossil record is usually the best source of direct evidence we can use to answer these types of questions, but this record is unusually sparse in Madagascar. In fact, there are only two periods when Madagascar’s fossil record is very rich: the Late Cretaceous (~80 million years ago during “the Age of the Dinosaurs”) and a tiny slice in the recent past extending back only tens of thousands of years.

On the ancient side of this gap we are fortunate to have a rich fossil record including a menagerie of bizarre forms, such as the giant “frog from hell” (Beelzebufo, the largest frog ever known to have lived), a completely herbivorous crocodile and a dinosaur with teeth specialised for eating fish. Unfortunately, many of these species are evolutionary “dead ends” that went extinct at the end of the Cretaceous.

Even for groups that do have modern relatives, like crocodiles, none of them are closely related to the modern forms – except for one side-necked turtle – meaning that this rich fossil assemblage doesn’t tell us much about the living animals’ evolutionary ancestors.

The little bit that we do know about the evolutionary history of Madagascar’s living groups is based on “subfossils” (referring to their geologically young age), with our deepest glimpse at a mere ~80,000 years ago. A fascinating collection of subfossils has been described from Madagascar, including a now-extinct megafauna comprising giant gorilla-sized lemurs, pygmy hippopotami and massive elephant birds weighing up to 450 kg. Many of these fossils appear to be closely related to the island’s modern animals, but they are so close to each other in age that this glimpse just simply isn’t deep enough – they represent lost cousins rather than ancestors.

So, despite this wealth of fossils, they haven’t got us any closer to understanding how, when and from where Madagascar’s modern animal groups arrived. The 80 million year “gap” in the fossil record between these two collections is precisely the time when most of its modern groups are thought to have not only evolved but to have also arrived, somehow, on the island.

This irreconcilable situation has left many researchers scratching their heads. If these animals were not “stranded” when Madagascar was separated, then how did it acquire its unusual collection of animals and plants, especially those with close relatives in distant lands?

Some have suggested that the only rational explanation is that sea levels must have dropped, exposing a path for the island’s ancestors to walk across. Despite the attractive simplicity of this scenario, when you look at the striking differences between Madagascar’s animals and what we consider typical “African” forms, it begins to lose its attractiveness.

Madagascar lacks many of the major animal groups found in Africa (e.g. giraffes, zebras, big cats, antelopes etc.), and in fact all of the island’s endemic mammals represent only five groups: bats, primates, rodents, tenrecs and carnivores. This disparity between regions makes it difficult to accept past land bridges as a migration route, as one would expect much more similarity across the Mozambique Channel if exposed land enabled animals to cross back and forth.

Floating Islands
Despite being the more difficult scenario to imagine, currently the most widely accepted hypothesis is that most of Madagascar’s groups arrived by flying, rafting or swimming across an oceanic barrier, largely from Africa. While this seems plausible for flying groups, such as birds and bats, this explanation also requires the ancestors of land-bound animals to have miraculously crossed a vast marine barrier more than 400 km wide – no easy feat.

Rafting has long been suggested as an explanation for Madagascar’s fauna, but many have argued that these types of “freak” dispersal events are too implausible and too rare to have made an impact on the modern forms. However, other researchers have played “devil’s advocate” and argued that even exceptionally rare events will happen if they are given sufficient amounts of time.

Advocates for rafting have also supported their claims using anecdotes from tantalising historic sightings of animals on floating clumps of vegetation, usually produced after major storm events. These “floating islands”, some as large as 100 metres across, have been witnessed drifting aimlessly at sea, some containing large fruiting trees and small pockets of fresh water. Animals have also been witnessed catching rides on these rafts, including iguanas, coconut-throwing monkeys and even a human baby. All of this information suggests that this type of transport happens, at least in extraordinary situations.

Lost at Sea?
In order to examine whether rafting potentially played an important role in shaping Madagascar’s modern groups, we tested hypotheses about what one would predict to see if rafting really worked on an intercontinental scale. We compiled a database of more than 80 of Madagascar’s animal groups – everything from cichlid fishes, chameleons and crocodiles to lemurs, parrots and flying foxes. For each group we first used phylogenetic and bio­geographic relationships to examine when they were reconstructed to arrive and from where, and looked for what patterns emerged.

We also made predictions about rafting. For example, we predicted more arrivals from Africa (the “closest source”) than a more distant place like Asia. We also supposed that if an animal had a “dispersal advantage”, such as flying or swimming, that this would increase its chances of arrival.

Finally we also examined whether the direction of past ocean currents had an effect on the arrival rates of what we deemed “obligatory rafters” – animals thought to have arrived after Madagascar was isolated, and did not have the capacity to swim or fly. Worldwide, prevailing ocean currents are stable and predictable, and depend largely on the positions of the major landmasses. But approximately 15 million years ago, during the mid-Miocene, the northward drift of Madagascar caused a reversal in how the ocean currents flowed. Before the change they were largely eastward (from Africa to Madagascar, and therefore favourable for transport in that direction), while afterwards they shifted westward (from Madagascar to Africa).

If ocean currents played an important role in rafting, we thought we might see differences in “who arrived when, and from where” before and after this shift occurred.

Wind, Waves and Wings
Our results are consistent with the hypothesis that rafting played a critical role in establishing Madagascar’s bizarre animal groups. We saw that swimming and rafting from Africa was most successful when winds and ocean currents were conducive to that direction of dispersal (e.g. before the mid-Miocene) and that, after ocean currents reversed, the number of successful rafters from Africa decreased markedly.

We also demonstrated that flightless animals arrived virtually exclusively from Africa (which was the closest available landmass), and that after the shift in ocean currents, when rafts would have encountered prevailing ocean currents in the wrong direction, the overwhelming majority of groups arriving were able to fly.

It is important to remember that dispersal by rafting is not something that needs to occur often. When you consider the tens of millions of years we’re talking about to establish the modern forms, our models postulate only two successful colonisations to Madagascar every million years.

In fact, for most groups it takes only one rare event to fully explain their presence on Madagascar. For example, DNA evidence indicates that one ancestral primate species made it across 40–50 million years ago, and gave rise to the 101 descendent lemur species you can find living on Madagascar today.

Perhaps most interestingly, our work implies that it took 81 successful colonisations of Madagascar to create the diversity of bizarre modern forms, but there were also likely many more groups that arrived and later went extinct without leaving a trace of their presence for us to see today. This is because the overall “apparent” arrival rate decreases as you get closer to the present, probably because some older arrivals went extinct and were replaced by newer arrivals. If this assertion is correct, our continued exploration for fossils within this time period could lead to very significant and unexpected future discoveries.

With renewed enthusiasm I scraped away again at the rock surrounding the corner of bone sticking out. The rock itself was harder than at most fossil sites, meaning that exposing fossils was quite a slow process. A close inspection of the sediments shows that these rocks were clearly formed by shallow marine waters, probably close to the original coastline. The baking sun and the action of the tides further cemented the rocks, often making them as hard as concrete.

While the fossils we have discovered thus far are the remains of ocean creatures, in other parts of the world, fossils of land-dwelling animals have been discovered rarely (and usually after years of hard work) in deposits like these after their bones or teeth had washed into the ocean from a nearby stream or river, perhaps during a heavy rain or tropical storm.

I moved my eyes closer to the buff-coloured rocks in the hope of seeing such a treasure – perhaps a glint of enamel reflected in the blazing sun might lead me to a tooth of the ancestral lemur, tenrec or carnivore. After searching until my eyes started to hurt, I glanced up at the ocean waves and couldn’t help but imagine one of these epic trans-oceanic journeys of the past – a raft from Africa completing its voyage by washing up on a beach much like this one. A few unwitting travellers, bewildered, climbing onto shore and setting out into the “promised land”, beginning a legacy whose history we are still unravelling.

Karen Samonds is a palaeontologist and Senior Lecturer in the School of Biomedical Sciences at the University of Queensland.