Evolution Icon Evolution

Rafting Stormy Waters: When Biogeography Contradicts Common Ancestry


The orderly pattern of biogeographic distribution of plants and animals was one of the lines of evidence that Charles Darwin mentioned in support of his theory of common descent with modification. Likewise, modern presentations of evidence in favor of evolution almost never fail to mention biogeography. Indeed it is a neat fact that many organisms that are endemic on islands are most similar to species of the adjacent mainland, or that fossil kangaroos have been found exclusively in Australia, which happens to be their modern area of distribution. This orderly pattern of biogeographic distribution is usually explained by reference to either dispersal or vicariance (continental drift).

However, it is far from true that biogeography unambiguously supports common ancestry, or that patterns of biogeographic distribution always align well with the pattern of reconstructed phylogenetic branching or the supposed age of origin. Indeed, there are many tenacious problems of biogeography and paleobiogeography that do not square well with the evolutionary paradigm of common descent. Those problems include disjunct distributions of organisms that do not occur in adjacent regions (or at least regions that are thought to have separated by continental drift). Examples include freshwater crabs (Sternberg & Cumberlidge 2001); the colocolo opossum from Chiloe, more closely related to Australian marsupials than to other American opossums; the lichen genera Heteroplacidium and Ramboldia that are found in Australia, New Zealand, South Africa, and the Falkland Islands but also in the Mediterranean region, e.g., Sardinia (Australian National Botanic Gardens 2012); or the New World planthopper group Plesiodelphacinae in Japan (Asche et al. 2016). Sometimes it has been the discovery of fossils that created biogeographical puzzles, like the discovery of a fossil platypus from South America (Pascual et al. 1992), or a fossil hummingbird from Europe (Louchart et al. 2008), which destroyed the previously undisputed evolutionary stories about the endemic origins of these groups.

Ratite Birds

An iconic example of disjunct distribution is that of ratite birds. These include very large flightless birds like the African ostriches, the extinct elephant birds from Madagascar, the South American rheas, the Australian emus, the cassowaries from Australasia, and kiwis and the extinct moas from New Zealand. Most biologists assumed that ratites are monophyletic and originated from a common flightless ancestor, and that their disjunct distribution is explained by the breakup of the ancient southern supercontinent Gondwana. Unfortunately, a few problems spoil this delightful just-so story:

  1. Moas do not seem to be more closely related to the New Zealand kiwis (Cooper et al. 2001, Haddrath & Baker 2001).
  2. Instead, the elephant birds of Madagascar are claimed to be the closest relatives of the New Zealand kiwis (Mitchell et al. 2014).
  3. The breakup of Gondwana is much too long ago to explain the distribution of ratite birds by continental drift (vicariance biogeography), which cannot have originated before the Paleogene according to fossil and molecular clock evidence (Prum et al. 2015).

But wait, there is help from cladistics: Just shake the tree and reshuffle the taxa, so that all the ostrich-like birds could have evolved independently by multiple convergence from flying ancestors, and their biogeographic distribution could thus be due to normal aerial dispersal. This was suggested by the genomic study of Harshman et al. (2008), who found the flying tinamous to be nested within flightless ratite birds. This is, of course, contradicted by many morphological characters that unite all ratites, as well as by molecular data, and Harshman even mentions five recent studies that strongly supported ratite monophyly. These conflicting data were ignored in Harshman’s DNA study that was crafted by not fewer than 19 co-authors. Certainly the new molecular tree must be so robust that the new molecular evidence outweighs the conflicting evidence. Except that that is not the case: indeed Harshman et al. (2008) presented two very different trees, one in which tinamous are closer to rheas, and one in which they are closer to cassowaries, emus, and kiwis. But who cares? At least we got rid of an inconvenient biogeographical problem. 

Unfortunately, there is more: In a study, Haddrath & Baker (2012) confirmed that tinamous nest within ratites, but now they are more closesy related to the extinct moas, while rheas are closer to kiwis, cassowaries, and emus. Certainly, now we’ve got it right, at least till the next study with other genes, like that of Smith et al. (2013), who again got two different results for the position of tinamous like Harshman et al. (2008). Meanwhile a new study by Baker et al. (2014) vindicated the result of Haddrath & Baker (2012) (note that Scherz 2013 disputed all the alleged evidence for a position of tinamous within ratites). Isn’t phylogenetic tree reconstruction a wonderful science?

Freshwater Snails on South Pacific Islands

Zielske et al. (2016) described the enigmatic pattern of long-distance dispersal of minute freshwater gastropods of the family Tateidae across the South Pacific, presumably during the Paleogene by birds as vectors. They found that the more remote archipelagos harboring the genus Fluviopupa were colonized from New Zealand with a complex westward dispersal, so that “geographical distance was not an appropriate predicator of phylogenetic relationship.” Aha! Let me translate this into plain English: the biogeographic pattern does not support common ancestry, but has to be explained away with ad hoc hypotheses.

The Rafting Hypothesis for Oceanic Dispersal

Anyway, some of the biggest biogeographic problems are posed by organisms that must be assumed by evolutionists to have dispersed across oceans via rafting to other continents (many examples are listed by Luskin, and de Queiroz 2005). This rafting hypothesis was first suggested by Alfred Russel Wallace, and elaborated by modern evolutionists, who have even proposed floating islands as a mode for long-distance dispersal of vertebrates across oceans (Houle 1998). Nevertheless, even the famed paleontologist George Gaylord Simpson (1940) acknowledged that “this sort of adventitious migration is dragged in when necessary to explain away any facts that contradict the main thesis.” Well said!

Trapdoor Spiders

Among the very diverse spider fauna of Australia there is one species that stands out: the tree trapdoor spider Moggridgea rainbowi, which only occurs on Kangaroo Island, but has its closest relatives of the same genus living in Africa. A phylogenomic study suggested that the Australian species separated from its African con-generic sister species about 2-16 million years ago, which is much later than the separation of the Australian and African continents around 95 million years ago. Therefore, it was recently proposed by Harrison et al. (2017) that large tarantulas rafted 6,000 miles across the wild Indian Ocean from Africa all the way to Australia (PLOS 2017), and not to the long coast of Western Australia but to a small island near Adelaide in South Australia. Professor Andrew Austin, the PhD supervisor of the publication’s lead author, Sophie Harrison, said in an interview, “At first thought, this does seem incredible” (University of Adelaide 2017). At first thought? I give it a second and third thought and still find it incredible.


Worm-lizards (Amphisbaenia) are a distinct (mostly legless) subgroup of squamate reptilians. They are bizarre and cryptic predators with a burrowing way of life, which raises the problem how to explain their disjunct distribution in South America, Africa, the Middle East, and parts of North America and Europe. A phylogenetic study suggested that the South American and African forms only separated about 40 million years ago, when both continents were already widely separated by the south Atlantic Ocean. Therefore, biologists are forced to assume that these subterranean animals rafted across the ocean. Actually, they have to assume not one but at least three (maybe even five) trans-oceanic dispersals (Longrich et al. 2015): from North America to Europe, from North America to Africa, and from Africa to South America. To address the fact that something like this is not only highly unlikely but indeed has never been observed, scientists invoke millions of years like a magic wand to allow for ridiculously improbable explanations within geological timescales (Panciroli 2016). 

Iguanas and Boine Snakes on Pacific Islands

Iguanas and boine snakes are mostly found in America, except for their enigmatic occurrence on Madagascar and on the two Pacific islands of Fiji and Tonga, far away from the American continent. Again scientists have preferred an explanation suggesting that these animals arrived on these islands by rafting on floating mats of vegetation, which would have taken about six months for the 5,000-mile journey (University of Chicago Press 2010). However, this leading explanation was recently disputed in favor of terrestrial dispersal via hypothetical land bridge connections (Noonan et al. 2010). Of course, this new hypothesis has its own problems and needs several ad hoc hypotheses, because we have no independent evidence for a land connection to Asia and/or Australia at the crucial time in Earth history, and fossil and subfossil iguanids are absent from Australia and other (in-between) Pacific islands. Noonan et al. frankly admit that they could “not conclusively demonstrate[e] the path to the Pacific for boine snakes and iguanid lizards,” but nevertheless these scientists preferred their new hypothesis over “the greatest vertebrate rafting event ever proposed.” Obviously they considered such an event as too unlikely to be true. But why is that, if millions of years are supposed to make even the most unlikely stuff easily possible, multiple times over? 

Concerning the occurrence of iguanas and boine snakes on Madagascar, a colonization via the Indian subcontinent has been proposed, because fossil representatives of these two animal groups are known from Asia but not from Africa (Vences 2004). The nagging problem is that the Malagasy iguanid genera have been found to be nested within South American species. This is another biogeographical enigma that does not fit well with the standard evolutionary narrative.

New World Monkeys

The most famous example of assumed rafting dispersal is the case of New World monkeys (Platyrrhini). The oldest fossil record of this primate group is Bransinella boliviana from a 26-million-year-old late Oligocene locality in Bolivia, and Perupithecus ucayaliensis from the Late Eocene (around 41 million years ago) of Peru, which agrees with an evolutionary age of New World monkeys dated by molecular clock at about 37-40 million years. The closest proposed relatives are Talahpithecus (Oligopithecidae) and Proteopithecus (Proteopithecidae) from the Eocene of North Africa, so that most primatologists think that New World monkeys evolved in Africa before crossing the Atlantic Ocean (Bond et al. 2015). Neither fossil nor living representatives are known from America north of Mexico, so that colonization via an Asian-North American land-bridge seems very unlikely, especially as South America was separated from North America from 80-3.5 million years ago. South America separated from Africa by continental drift about 90-120 million years ago, which is much too old for the ancestors of New World monkeys to have travelled with the drifting continent, as these ancestors only appear in the Eocene. Nobody has any plausible idea how the ancestors of New World monkeys could have managed to cross the whole Atlantic Ocean from Africa to South America, because single animals rafting on trees seem to be an absurd explanation (Fleagle & Gilbert 2006), for which there is not a single modern observation. This especially holds because you would need a viable population of at least a few conspecific male and female animals at the same place and time. Estimates vary between 10-100 conspecific individuals that would have been required as a founder population for New World monkeys. That would have required a big raft for sure, and the Atlantic was at least 1,400 km wide in the Eocene. The problem of survive such a rafting journey across the ocean is complicated by the fact that even small mammals from arid regions (e.g., degus) cannot survive more than two weeks without fresh water, but the journey would have taken at least 60 days when relying on currents and still at least 14 days with “sailing” (Gabbatiss 2016). Sailing? No problem, you just need floating islands with vertical trees acting as sails. Wow, the imagination of evolutionists is nearly unlimited. One wonders if they have ever been on a boat on the ocean in stormy weather.

But as Donald Prothero has confirmed (Prothero 2015), “monkeys were not the only colonists to reach South America by rafting from Africa. It turns out that there are lots of animals that did the same thing: geckos, skinks, tortoises, the blind burrowing reptiles known as amphisbaenids, and even the peculiar birds known as hoatzins. Most impressive of all were the caviomorph rodents.” Yes, even rodents and birds are believed to have crossed the ocean on rafts (Poux et al. 2006, Naish 2011). It looks like there was some very busy ocean travel going on in those times, which suddenly stopped as soon as humans could have observed and recorded it.


Yes, it is a fatal problem for the fantastic rafting hypotheses that in the entire history of human seafaring there exists not a single documented case where larger terrestrial animals were actually observed rafting in the middle of a large ocean. The only empirical observation for rafting dispersal is a group of 15 Anolis lizards found in 1995 washed ashore on a Caribbean island, having apparently drifted after a hurricane 200 miles from the island of Guadeloupe to Anguilla, which both belong to the Leeward Islands of the Lesser Antilles in the Caribbean (Censky et al. 1998). Another case seems to be an Aldabra giant tortoise washed ashore on the coast of East Africa, probably having been drifting the 740 km distance for about three weeks (Gerlach et al. 2006). Charles Darwin hoped to solve such problems with the claim that given enough time, many things that are unlikely can happen, and “thus, neo-Darwinian evolutionists are forced to appeal to ‘unlikely’ or ‘unexpected’ migration of organisms, in some cases requiring the crossing of oceans to account for the biogeographical data. This kind of data may not necessarily absolutely falsify Darwinism, but at the least it challenges the simplistic argument that biogeography supports universal common descent through congruence between migration pathways and evolutionary history. In many cases, the congruence is simply not there” (Luskin 2015). There is a further problem though: we meanwhile have substantial paleobiogeographical evidence that such dispersal by rafting simply did not happen in either direction even in cases of much smaller distances and long periods of time (Krause 2001, Clyde et al. 2003), as for example during the Cretaceous between Africa and Madagascar (even though this is complicated by the issue of paleocurrents; Ali & Huber 2010), or between India and Asia when the Indian subcontinent was close but not yet attached to Asia.

In all these examples a polyphyletic view much better agrees with the empirical evidence than does universal common descent. The latter explanation creates almost unresolvable problems of impossible routes of dispersal or a much too early dating of vicariance events that conflicts with the actual fossil record or the geological data on continental drift. At least the existence of such conflicting evidence should be acknowledged by evolutionists, but of course this does not happen and instead we are confronted with an endless flood of ad hoc hypotheses that try to explain away the conflicting evidence and even claim it as support for evolution.


Photo credit: Guy Kawasaki [CC BY-SA 4.0], from Wikimedia Commons.