Realizing that the distribution of species on islands provides conclusive proof of evolution was one of the greatest pieces of sleuthing in the history of biology. This too was the work of Darwin, whose ideas still loom mightily over the field of biogeography. In chapter 12 of The Origin, Darwin reports fact after fact, painstakingly gathered over years of observation and correspondence, building his case like a brilliant lawyer. When I teach the evidence for evolution to my students, this is my favorite lecture. It's an hourlong mystery story, an accumulation of seemingly disparate data that in the end resolves into an airtight case for evolution.

But before we get to the evidence, we need to distinguish two types of islands. The first are continental islands: those islands once connected to a continent but later separated either by rising sea levels that flooded former land bridges or by moving continental plates. These islands include, among many others, the British Isles, Japan, Sri Lanka, Tasmania, and Madagascar. Some are old (Madagascar parted from Africa about 160 million years ago), others much younger (Great Britain separated from Europe around 300,000 years ago, probably during a catastrophic flood spilling from a large, dammed-up lake to the north). Oceanic islands, on the other hand, are those that were never connected to a continent; they arose from the sea floor, initially bereft of life, as growing volcanoes or coral reefs. These include the Hawaiian Islands, the Galápagos archipelago, St. Helena, and the Juan Fernández group described at the beginning of this chapter.

The "island" argument for evolution starts with the following observation: oceanic islands are missing many types of native species that we see on both continents and continental islands. Take Hawaii, a tropical archipelago whose islands occupy about 6,400 square miles, only slightly smaller than the state of Massachusetts. While the islands are well stocked with native birds, plants, and insects, they completely lack native freshwater fish, amphibians, reptiles, and land mammals. Napoleon's island of St. Helena and the archipelago of Juan Fernández lack these same groups, but still have plenty of endemic plants, birds, and insects. The Galápagos Islands do have a few native reptiles (land and marine iguanas, as well as the famous giant tortoises), but they too are missing native mammals, amphibians, and freshwater fish. Over and over again, on the oceanic islands that dot the Pacific, the South Atlantic, and the Indian Ocean, one sees a pattern of missing groups—more to the point, the same missing groups.

At first blush, these absences seem bizarre. If you look at even a very small patch of a tropical continent or a continental island, say in Peru, New Guinea, or Japan, you'll find plenty of native fish, amphibians, reptiles, and mammals.

As Darwin noted, this disparity is hard to explain under a creationist scenario: "He who admits the doctrine of creation of each separate species, will have to admit, that a sufficient number of the best adapted plants and animals have not been created on oceanic islands."

But how do we know that mammals, amphibians, freshwater fish, and reptiles really are suited to oceanic islands? Maybe the creator didn't put them there because they wouldn't do well. One obvious reply is that continental islands do have these animals, so why would a creator put different types of animals on continental versus oceanic islands? How the island was formed shouldn't make a difference. But Darwin ends the sentence given above with an ever better response: "... for man has unintentionally stocked them from various sources far more fully and perfectly than nature."

In other words, mammals, amphibians, freshwater fish, and reptiles often do very well when humans introduce them to oceanic islands. In fact, they often take over, wiping out native species. Introduced pigs and goats have overrun Hawaii, making meals of native plants. Introduced rats and mongooses have destroyed or endangered many of Hawaii's spectacular birds. The cane toad, a huge poisonous amphibian native to tropical America, was introduced to Hawaii in 1932 to control beetles on sugarcane. The toads are now a pest, breeding prolifically and killing cats and dogs who mistake them for a meal. The Galápagos Islands have no native amphibians, but an Ecuadorian tree frog, introduced in 1998, has established itself on three islands. On Sao Tomé, the volcanic island off the west coast of Africa where I collect fruit flies for my own research, black cobras have been introduced—perhaps accidentally— from the African mainland. They have done so well that we simply won't work in certain areas of the island, as the cobras are so numerous that we can encounter several dozen of these deadly and aggressive snakes in a single day. Land mammals do well on islands too—introduced goats helped Alexander Selkirk stay alive on Juan Fernández, and they also thrive on St. Helena. Throughout the world the story is the same: humans introduce species to oceanic islands where they did not exist, and these species displace or destroy native forms. So much for the argument that oceanic islands are somehow unsuitable for mammals, amphibians, reptiles, and fish.

The next step of the argument is this: although oceanic islands lack many basic kinds of animals, the types that are found there are often present in profusion, comprising many similar species. Take the Galápagos. Among its thirteen islands there are twenty-eight species of birds found nowhere else. And of these twenty-eight, fourteen belong to a single group of closely related birds: the famous Galápagos finches. No continent or continental island has a bird fauna so heavily dominated by finches. Yet despite their shared finch-like traits, the Galápagos group is ecologically quite diverse, with different species specializing on foods as different as insects, seeds, and the eggs of other species. The "carpenter finch" is one of those rare species that uses tools—in this case a cactus spine or twig—to pry insects from trees. Carpenter finches fill the ecological role of woodpeckers, which don't live in the Galápagos. And there's even a "vampire finch" that pecks wounds on the rear ends of seabirds and then laps up the blood.

Hawaii has an even more spectacular radiation of birds, the honey-creepers. When the Polynesians arrived in Hawaii about 1,500 years ago, they found about 140 species of native birds (we know this from studies of bird "subfossils": bones preserved in ancient waste dumps and lava tubes). Around sixty of these species—nearly half the bird fauna—were honeycreepers, all descended from a single ancestral finch that arrived on the islands about four million years ago. Sadly, only twenty species of honeycreeper remain, all of them endangered. The rest were destroyed by hunting, habitat loss, and human-introduced predators like rats and mongooses. But even the few remaining honeycreepers show a fantastic diversity of ecological roles, as shown in figure 22. The bill of a bird can tell us a lot about its diet. Some species have curved bills for sipping nectar from flowers, others stout, parrot-like bills for cracking hard seeds or crushing twigs, still others thin pointy bills for picking insects from foliage, and some even have hooked bills for prying insects from trees, filling the role of a woodpecker. As on the Galápagos, we see one group that is overrepresented, with species filling niches occupied by very different species on continents or continental islands.

Oceanic islands also harbor radiations of plants and insects. St. Helena, though lacking many groups of insects, is home to dozens of species of small, flightless beetles, especially wood weevils. On Hawaii,



Pjrrotbill pali1a figure 22. An adaptive radiation: some related species of Hawaiian honeycreepers that evolved after their finchlike ancestor colonized the islands. Each finch has a bill that enables it to use different food. The 'i'iwi's slender bill helps it sip nectar from long tubular flowers, the 'akepa has a slightly crossed bill that allows it to pry open buds to search for insects and spiders, the Maui parrotbill has a massive bill for prying up bark and splitting twigs to find beetle larvae, and the palila's short but strong bill helps it open seed pods and extract the seeds.

the group that I study—fruit flies of the genus Drosophila—is positively luxuriant. Although the Hawaiian Islands make up only 0.004 percent of Earth's land, they contain nearly half of the world's 2,000 species of Drosophila. And then there are the remarkable radiations of plants in the sunflower family on Juan Fernández and St. Helena, some of which have become small woody trees. Only on oceanic islands can small flowering plants, freed from competition with larger shrubs and trees, evolve into trees themselves.

So far we've learned two sets of facts about oceanic islands: they are missing many groups of species that live on continents and continental islands, and yet the groups that are found on oceanic islands are replete with many similar species. Together these observations show that, compared to other areas of the world, life on oceanic islands is unbalanced. Any theory of biogeography worth its salt has to explain this contrast.

But there's something more here too. Take a look at the following list of the groups that are often native to oceanic islands and those that are usually missing (Juan Fernández is just one group of islands that conforms to the list):



Plants Birds

Insects and other

Land mammals


Amphibians arthropods (e.g., spiders) Freshwater fish

What's the difference between the two columns? A moment's thought gives the answer. Species in the first column can colonize an oceanic island through long-distance dispersal; species in the second column lack this ability. Birds are capable of flying great distances over the sea, carrying with them not only their own eggs but also seeds of plants they've eaten (which can germinate from their droppings), parasites in their feathers, and small organisms sticking to mud on their feet. Plants can get to islands as seeds, floating across expanses of sea. Seeds with barbs or sticky coverings can hitchhike to islands on the feathers of birds. The light spores of ferns, fungi, and mosses can be carried huge distances by the wind. Insects, too, can fly to islands or be taken by winds.

In contrast, animals in the second column have great difficulty crossing expanses of sea. Land mammals and reptiles are heavy and can't swim very far. And most amphibians and freshwater fish simply can't survive in saltwater.

So the kinds of species that we find on oceanic islands are precisely those that can arrive across the sea from distant lands. But what is the evidence that they do so? Every ornithologist knows of occasional "visitor" birds found thousands of miles from their normal habitat, the victim of winds or faulty navigation. Some birds have even established breeding colonies on oceanic islands in historical time. The purple gallinule, long an occasional visitor to the remote island of Tristan da Cunha in the South Atlantic, finally started breeding there in the 1950s.

Darwin himself did some simple yet elegant experiments showing that seeds from some plant species could still germinate after prolonged immersion in seawater. Seeds from the West Indies have been found on the distant shores of Scotland, obviously carried by the Gulf Stream, and "drift seeds" from continents or other islands are also found on the shores of South Pacific islands. Caged birds can retain plant seeds in their digestive tracts for a week or more, showing the likelihood of long-distance transport. And there have been many successful attempts to sample insects in the air using traps attached to airplanes or ships far from land. Among the species collected have been locusts, moths, butterflies, flies, aphids, and beetles. Charles Lindbergh, on a 1933 trip across the Atlantic, exposed glass microscope slides to the air, capturing numerous microorganisms and insect parts. Many spiders disperse as juveniles by "ballooning" with parachutes of silk; these wanderers have been found several hundred miles from land.

Animals and plants can also hitch rides to islands on "rafts"—logs or masses of vegetation that float away from continents, usually from the mouths of rivers. In 1995 one of these large rafts, probably blown by a hurricane, deposited a cargo of fifteen green iguanas on the Caribbean island of Anguilla, where they had not previously existed, from a source 200 miles away. Logs of Douglas fir from North America have been found on Hawaii, and logs from South America have made it to Tasmania. Rafting like this explains the presence of the occasional endemic reptile on oceanic islands, such as the Galápagos iguanas and tortoises.

Further, when you look at the type of insects and plants native to oceanic islands, they are from groups that are the best colonizers. Most of the insects are small, precisely those that would be easily picked up by wind. Compared to weedy plants, trees are relatively rare on oceanic islands, almost certainly because many trees have heavy seeds that neither float nor are eaten by birds. (The coconut palm, with its large buoyant seeds, is a notable exception, occurring on almost all Pacific and Indian Ocean islands). The relative rarity of trees, in fact, explains why many plants that are short weeds on continents have evolved into woody tree-like forms on islands.

Terrestrial mammals are not good colonizers, and that's why oceanic islands lack them. But they don't lack all mammals. This brings up two exceptions that prove the rule. The first was noted by Darwin:

Although terrestrial mammals do not occur on oceanic islands, aerial mammals do occur on almost every island. New Zealand possesses two bats found nowhere else in the world: Norfolk Island, the Viti Archipelago, the Bonin Islands, the Caroline and Marianne [Mariana] Archipelagoes, and Mauritius, all possess their peculiar bats. Why, it may be asked, has the supposed creative force produced bats and no other mammals on remote islands? On my view this question can easily be answered; for no terrestrial mammal can be transported across a wide space of sea, but bats can fly across.

And there are also aquatic mammals on islands. Hawaii has one, the endemic monk seal, and Juan Fernández has a native fur seal. If native mammals on oceanic islands were not created, but descended from colonists, you'd predict that those ancestral colonists must have been able to fly or swim.

Now, it's clear that long-distance dispersal of a given species to a distant island can't be a frequent event. The chance that an insect or bird could not only traverse vast expanses of sea to land on an island, but also establish a breeding population once it got there (this requires either an already fertilized female or at least two individuals of opposite sex), must be very low. And if dispersal were common, life on oceanic islands would be quite similar to that of continents and continental islands. Nevertheless, most oceanic islands have been around for millions of years, long enough to permit some colonization. As the zoologist George Gaylord Simpson remarked, "Any event that is not absolutely impossi-ble...becomes probable if enough time passes." To take a hypothetical example, suppose that a given species has only one chance in a million of colonizing an island each year. It's easy to show that after a million years have passed, there is a large probability that the island would have been colonized at least once: 63 percent, to be exact.

One final observation closes the chain of logic that secures the case for evolution on islands. And that is this: with few exceptions, the animals and plants on oceanic islands are most similar to species found on the nearest mainland. This is true, for example, of the Galápagos Islands, whose species resemble those from the west coast of South America. The similarity can't be explained by the argument that the islands and South America have similar habitats for divinely created species, because the Galápagos are dry, treeless, and volcanic—quite different from the lush tropics that dominate the Americas. Darwin was especially eloquent on this point:

The naturalist, looking at the inhabitants of these volcanic islands in the Pacific, distant several hundred miles from the continent, feels that he is standing on American land. Why should this be so? Why should the species which are supposed to have been created in the Galápagos Archipelago, and nowhere else, bear so plainly the stamp of affinity to those created in America? There is nothing in the conditions of life, in the geological nature of the islands, in their height or climate, or in the proportions in which the several classes are associated together, which closely resemble the conditions of the South American coast: in fact, there is a considerable dissimilarity in all these respects____Facts such as these admit of no sort of explanation on the ordinary view of independent creation; whereas on the view here maintained, it is obvious that the Galápagos Islands would be likely to receive colonists from America, whether by occasional means of transport or (though I do not believe in this doctrine) by formerly continuous land... such colonists would be liable to modification,—the principle of inheritance still betraying their original birthplace.

What is true of the Galápagos is also true of other oceanic islands. The closest relatives of the endemic plants and animals on Juan Fernández come from the temperate forests of southern South America, the closest continent. Most of the species on Hawaii are similar (but not identical) to those from the nearby Indo-Pacific region—Indonesia, New Guinea, Fiji, Samoa, and Tahiti—or from the Americas. Now, given the vagaries of winds and the direction of ocean currents, we don't expect every island colonist to come from the closest source. Four percent of Hawaiian plant species, for example, have their closest relatives in Siberia or Alaska. Still, the similarity of island species to those on the nearest mainland demands explanation.

To sum up, oceanic islands have features that distinguish them from either continents or continental islands. Oceanic islands have unbalanced biotas—they are missing major groups of organisms, and the same ones are missing on different islands. But the types of organisms that are there often comprise many similar species—a radiation—and they are the types of species, like birds and insects, that can disperse most easily over large stretches of ocean. And the species most similar to those inhabiting oceanic islands are usually found on the nearest mainland, even though their habitats are different.

How do these observations fit together? They make sense under a simple evolutionary explanation: the inhabitants of oceanic islands descended from earlier species that colonized the islands, usually from nearby continents, in rare events of long-distance dispersal. Once there, accidental colonists were able to form many species because oceanic islands offer lots of empty habitats that lack competitors and predators. This explains why speciation and natural selection go wild on islands, producing "adaptive radiations" like that of the Hawaiian honeycreepers. Everything fits together if you add accidental dispersal, which is known to occur, to the Darwinian processes of selection, evolution, common ancestry, and speciation. In short, oceanic islands demonstrate every tenet of evolutionary theory.

It's important to remember that these patterns do not generally hold for continental islands (we'll come to an exception in a second), which share species with the continents to which they once were joined. The plants and animals of Great Britain, for example, form a much more balanced ecosystem, having species largely identical to those of mainland Europe. Unlike oceanic islands, continental islands were cut adrift with most of their species already in place.

Now try to think of a theory that explains the patterns we've discussed by invoking the special creation of species on oceanic islands and continents. Why would a creator happen to leave amphibians, mammals, fish, and reptiles off oceanic islands, but not continental ones? Why did a creator produce radiations of similar species on oceanic islands, but not continental ones? And why were species on oceanic islands created to resemble those from the nearest mainland? There are no good answers—unless, of course, you presume that the goal of a creator was to make species look as though they evolved on islands. Nobody is keen to embrace that answer, which explains why creationists simply shy away from island biogeography.

We can now make one final prediction. Very old continental islands, which separated from the mainland eons ago, should show evolutionary patterns that fall between those of young continental islands and oceanic islands. Old continental islands such as Madagascar and New Zealand, cut off from their continents 160 million and eighty-five million years ago respectively, will have been isolated before many groups like primates and modern plants had evolved. Once these islands parted from the mainland, some of their ecological niches remained unfilled. This opens the door for some later-evolving species to successfully colonize and establish themselves. We can predict, then, that these old continental islands should have a somewhat unbalanced flora and fauna, showing some of the biogeographic peculiarities of true oceanic islands.

And indeed, this is just what we find. Madagascar is famous for its unusual fauna and flora, including many native plants and, of course, its unique lemurs—the most primitive of the primates—whose ancestors, after arriving in Madagascar some sixty million years ago, radiated into more than seventy-five endemic species. New Zealand, too, has many natives, the most well-known being flightless birds: the giant moa, a 13-foot-tall monster hunted to extinction by about 1500, the kiwi, and that fat, ground-dwelling parrot, the kakapo. New Zealand also shows some of the "imbalance" of oceanic islands: it has only a few endemic reptiles, only one species of amphibian, and two native mammals, both bats (though a small fossil mammal was recently found). It too had a radiation—there were eleven species of moas, all now gone. And, like oceanic islands, the species on Madagascar and New Zealand are related to those found on the nearest mainland: Africa and Australia respectively.

Was this article helpful?

0 0

Post a comment