The Monkey's Voyage

by Alan de Queiroz

  Think again about the world as seen through the lens of vicariance. Begin with one landmass, filled with living species. If we start, say, in the Jurassic, those species might include mosses, ferns, and cycads; cockroaches and dragonflies; frogs and shrew-like mammals; pterosaurs and dinosaurs. Imagine that this landmass rifts apart and the two pieces begin to wander off in different directions. Now jump forward 150 million years or so to see what we have. In the vicariance view, we would expect to find related organisms on the two, now well-separated landmasses. All that is required is a kind of biological version of inertia—the lineages in question have to persist in both areas.

  This argument is persuasive. It appeals to our sense that a straightforward explanation is better than a convoluted one. It’s the reason there has been such a strong expectation among biogeographers that the Gondwanan continents and islands should still retain a clear biological signature of their common origins. The problem is that it ignores a hugely significant aspect of evolution, namely, extinction.

  Of course, every biologist knows that species and entire large groups go extinct, and no one thinks that all or even most of the lineages that existed when Gondwana began to break up are still around. Obviously, there are no living pterosaurs or dinosaurs (except for birds). Still, the implications of extinction haven’t always penetrated as deeply as they should. In particular, biogeographers haven’t given enough thought to the disappearance of lineages when contemplating the fate of old continental islands like the Falklands.

  The great metaphor of evolution is the tree of life. The branches are lineages, and the places where one branch grows out of another are where one species splits into two. Drawn as a living tree with a robust trunk and healthy green foliage, the metaphor suggests the exuberant proliferation of life. That image seems to agree with our present knowledge of the living world—the fact that a tropical rainforest might have hundreds of species of trees with hundreds of kinds of insects unique to each one, that new life forms are constantly being discovered in seemingly inhospitable places like boiling springs and sulphurous ocean vents, that nematodes are so numerous that their bodies alone could give a detailed outline of most of the landscape of the Earth. Life is exuberant. And yet, nearly all species that have ever existed have left no descendants that are alive today. Most of the exuberance has gone down a one-way path to oblivion.

  For most small islands, attrition by extinction is unavoidable. Specifically, if an island does not provide a species with an opportunity to speciate, that is, to form additional new species, then the set of lineages descended from the original inhabitants can only get smaller; species will inevitably go extinct and, without having given rise to any other species, that will mean the end of the lineage. The life span of a species is typically only a few million years, so the impact of extinction on the original lineages of such islands would be huge over the long life of Gondwanan islands such as the Falklands.

  To make matters worse, that impact will be exacerbated on islands by the simple fact that a species’ chance of going extinct is connected to the amount of area it occupies; the smaller the area, the greater the chance of extinction. Biologists have come up with several explanations for this relationship. One is that, if a species is found at only a few sites, there’s a greater chance that some event or events will make all the sites uninhabitable. Think of it as not having your eggs spread in enough baskets. For instance, consider the Devil’s Hole pupfish, a Mojave Desert species whose entire native range is a warm-spring pool with a surface area about the size of two bowling lanes. An earthquake or flood that disrupted the pupfish’s pool could wipe out the entire species. Another risk is that species with small ranges will generally consist of few individuals, and this makes them vulnerable to what ecologists call stochastic extinction. The idea here is that, even without any unusual environmental disruption, all populations experience random fluctuations in size (number of individuals), and the smaller the average size of the population, the more likely that one of its downward swings will hit zero. These random fluctuations are sometimes referred to as a drunkard’s walk; for small populations, you can picture the drunkard (representing population size) lurching back and forth close to a precipice, the precipice being extinction.

  Ecologists and conservation biologists continue to argue about which of these factors, or others I haven’t mentioned, are driving the connection between small area and high extinction risk, but no one doubts that the connection exists. This is why almost every species that has a very small geographic range is considered “at risk”—from the Kirtland’s Warbler, which breeds only in stands of young jack pines in a small part of northern Michigan, to the Devil’s Hole pupfish in its single spring pool, to many of the endemic species on Hawaii or the Galápagos.

  For large islands, the effect of small area on extinction will be less severe than it is for small islands, but the effect is probably still present. Even a very large island cannot harbor extremely widespread endemic species; the area of the entire island of Madagascar, for example, is far smaller than the range of continent-spanning species like the coyote or the American Robin. However, for large islands, or archipelagoes of smaller islands, speciation becomes important and can counter attrition by extinction; if the original species form new ones, then the extinction of a species does not necessarily translate to the loss of that particular lineage. Speciation, on islands and in general, can greatly prolong the life of taxa; in fact, almost all living species probably owe their existence to the evolutionary branching of their ancestors. One can imagine that, through this process of species proliferation, many of the original lineages on Gondwanan islands have managed to persist to the present. Maybe such lineages even dominate the biotas of some of these islands.

  However, paleontologists tell us unequivocally that there is more to extinction than just slow attrition. Sometimes, there are cataclysms, and there are reasons to believe that these events may be especially devastating for the biotas of islands.

  One stereotype of a paleontologist is a rough character in jeans, cowboy hat, and boots, somewhere out in the sagebrush-covered backcountry of Wyoming, trying to unearth a Tyrannosaurus with a pickax. The paleontologists I’m thinking of are not that guy; they tend instead to be cerebral, East-Coast types, more comfortable with programming languages than digging implements. They use the fossil record built up by the Tyrannosaurus hunters and other field paleontologists to see grand patterns in the history of life. Some of them have essentially spent their entire careers entering reams of data concerning fossils into computers, subjecting these data to arcane and ever more complex analyses, and seeing what the results tell us about the origin and demise of species and larger groups.

  The reams of data and arcane analyses have corroborated with hard numbers what was known before in a qualitative way: rates of extinction are highly variable through time. Graphs plotting the number of extinctions (of genera or families usually) through the geologic stages show a chaotic series of peaks and valleys of wildly different magnitude, with the especially prominent peaks representing mass extinctions, when many groups disappeared within relatively short spans of time. The great extinction that finished off the dinosaurs and many other taxa at the end of the Cretaceous is the best-known peak, one of the so-called Big Five, but it’s not the biggest. That honor goes to the end-Permian extinction—what Stephen Jay Gould called “The Great Dying”—which the number-crunching paleontologists tell us knocked off more than 90 percent of all species on the planet.

  Paleontologists argue about exactly what constitutes a mass extinction, but there is at least general agreement that the worst of them represent real catastrophes, times when something more than just “extinction as usual” was going on.49 These great pulses of extinction suggest some kind of rare event or set of circumstances, the impact of the comet or asteroid at the end of the Cretaceous being a prime example. We may be witnessing a mass extinction right now—one that’s
not at the level of the Big Five yet, but unfortunately has no end in sight—and it is certainly the result of an unusual set of events, namely, the origin of Homo sapiens and our series of technological “advances.”

  In discussing the slow attrition of lineages and the area effect on islands, we’ve essentially been dealing with extinction during periods when nothing extraordinary is happening. Now let us consider the cataclysmic. What effects might mass extinctions have had on the Gondwanan islands?

  Returning to the Falklands, we know that, after their separation from Africa, they experienced one of the Big Five, the end-Cretaceous mass extinction. Since that event was global (perhaps caused by the sun-­blocking dust cloud raised by the asteroid or comet impact), there is no reason to think the biota of the Falklands could have avoided it. In fact, it’s very possible that the proportion of species that went extinct was even greater on islands than on continents, simply because of the stochastic extinction numbers game; whatever caused the mass extinction in general might have reduced the already small populations of island species to the point where they were dangerously close to suffering extinction from random fluctuations in population size. In other words, during the end-Cretaceous event, a disproportionate number of island species might have taken the drunkard’s walk off the cliff of extinction.

  The Falklands also passed through several other minor global mass extinctions, including one at the end of the Eocene, about 34 million years ago, and another in the middle of the Miocene, some 14.5 million years ago. However, it seems very likely that the islands’ biota also suffered from what might be called “local mass extinctions,” events that wiped out many species, but only in a small region. In his paper, McDowall mentioned events of this kind, connected to the Pleistocene Ice Age cycles, to explain the disappearance of the original African biota of the Falklands. He suggested that these oscillations in climate, repeated several times, could have inflicted “a severe sorting of the Falklands biota.” What he meant was that these climatic shifts might have generated pulses of extinction, episodes during which many species got “sorted” either into the category of survivors or into the trashbin of history. On continents, the ice-age cycles apparently did not cause mass extinctions, because species that couldn’t handle the climate change could move. For instance, during the most recent ice age in North America, the ranges of many species, from voles to pine trees, shifted to the south. However, on islands like the Falklands, this would not have been an option. If the climate of the Falklands became too cold, or the growing season too short, for a particular species to survive, the members of that species would have nowhere to go. Terrestrial species can’t shift their ranges into the ocean.

  In a way, the Falklands, viewed over their long history, look like a perfect storm of extinction. To begin with, they’re small enough, at least in their current configuration, to have experienced a strong effect of area on extinction rate. Also, in their recent history, at least, the islands haven’t provided much opportunity for the formation of new species, the other variable in the attrition formula; nearly all of the endemic lineages consist of just one or a few species. Perhaps most importantly, they’ve had such a long history of isolation that they have almost undoubtedly been battered by multiple mass extinctions, including global ones, like the end-­Cretaceous event, and local ones, such as those connected to the Pleistocene Ice Age cycles. Maybe it’s no wonder that there’s not a single clear example of a living Falklands species whose ancestors were on board the microplate as it separated from Africa. The perfect storm has obliterated almost any trace of the islands’ Gondwanan origins.

  FROM THE BLACK ROBIN OF THE CHATHAMS TO THE LEMURS OF MADAGASCAR

  Compared to the other Gondwanan islands, the Falklands may be a somewhat unusual case. However, of the remaining islands, we can say one thing right off the bat: even with their perfect storm of extinction, the Falklands cannot possibly be more extreme than the Chatham Islands.

  The Chathams, which lie some 420 miles east of New Zealand, are a small archipelago with an area less than that of New York City. They’re cold and windy, but perhaps not quite as severe as the Falklands; the Chathams do have patches of native forest, for instance, and not so many peat bogs. As with many islands, the human history of the Chathams has been defined by remoteness. Maori from New Zealand colonized the archipelago at least five hundred years ago, but the descendants of those colonists, the Moriori, afterward remained isolated until the British reached the islands in 1791. By then the Moriori had developed a distinctive, peaceful culture—they had literally outlawed warfare—which unfortunately made them easy victims for the bellicose Maori, their distant brethren, who followed in the wake of the British. The Maori ended up killing or enslaving nearly all of the Moriori. Today, the population of the Chathams—a mix of European, Maori, and Moriori—is dwindling through other effects of remoteness, namely, the high cost of living and the desire to be connected to the larger human world.

  All of that, of course, happened within a thin slice of geologic history. In deep time, the Chathams haven’t always been isolated from other landmasses. Geologically speaking, they are part of the Chatham Rise, most of which is currently underwater and shows up on ocean maps as a pale blue swath of shallow water running back to New Zealand’s South Island. (Parts of the Rise are more than 3,000 feet below the ocean’s surface, but, compared to surrounding ocean regions, this is shallow.) However, at one point, more than 70 million years ago, most of the Chatham Rise was above water and formed part of Zealandia, and, a little further back in time, Zealandia made up a large chunk of the eastern edge of Gondwana.

  From this deep history, vicariance scientists have supposed that at least some species on the Chathams are Gondwanan relicts. That there was a Gondwanan biota on what is now the Chathams is undeniable: there are Late Cretaceous strata on the islands that contain the fossil remains of a typical Gondwanan flora, including southern beech trees, araucarian conifers, and podocarps. These beds even contain some unimpressive but unmistakable bones of theropod dinosaurs, relatives of Tyrannosaurus. But the big question, as with the Falklands, is, “Did any of those Gondwanan lineages persist to the present?”

  The rocks of the Chathams tell us the answer is, “Not likely.” Strata about 6 million years old are made up of sediments from lava flows and volcanic ash that indicate the emergence of a volcanic island. However, for most of the Cenozoic before that time, the area is only represented by marine sediments, such as limestones (made up of the hard parts of marine organisms) and underwater volcanics, suggesting that the Chathams were completely submerged.50 Thus, we have indications of the ultimate in local mass extinctions: if there were any Gondwanan relict species on the islands before their inundation, afterward there would certainly have been none.

  I’m reminded here that, when Tara and I were planning our trip to New Zealand, we briefly thought about taking the two-hour plane ride from Christchurch to the Chathams. That we decided against it was partly about time, partly about money, and, in a fairly direct way, partly about the fact that the whole place was probably underwater until recently. For me, the allure of the Chathams was their unique birds; for almost any serious birder, there’s a weird thrill in seeing a species you know can only be found on some small and remote piece of ground. I imagined spotting a little songbird called the Black Robin on the Chathams, the only place on Earth where that bird can be seen. However, browsing the Hand Guide to the Birds of New Zealand, I was struck by the obvious similarity of all the endemic birds of the Chathams to New Zealand species. The Black Robin looks a lot like the New Zealand robins and the Tomtit, the Chatham Island Warbler like the Grey Warbler of New Zealand, and the Chatham Island Oystercatcher like New Zealand’s Pied Oystercatcher (see Figure 10.4). The fact that none of the Chathams birds are as bizarrely unique as, say, a kiwi, or a kakapo, probably has everything to do with the islands’ submergence; under that scenario, the ancestors of all the endemic birds mu
st have arrived after the archipelago reemerged from the sea, and therefore they would have had little time to evolve into anything very distinctive. I’m not sure I could have resisted the lure of birds whose origins trace back tens of millions of years to the fragmentation of Gondwana, but species that hardly look different from their modern New Zealand relatives weren’t such a draw.

  10.4 Not such a draw for the birdwatcher: the Black Robin of the Chathams (left) is a close relative of the similar North Island Robin of New Zealand (right). Photos by Frances Schmechel (Black Robin) and Tony Wills (North Island Robin).

  The idea that species on the Chathams evolved only recently from ancestors that lived elsewhere has been confirmed by extensive molecular studies led by Steve Trewick, Adrian Paterson, and other scientists from New Zealand universities. These researchers have sequenced genes from a diverse collection of endemic Chathams plants and animals, including daisies, gentians, grasses, rails, parakeets, cockroaches, beetles, earwigs, and grasshopper relatives called wetas. They found exactly what one would expect for islands recently emerged from the sea: all the sampled Chathams species are very similar genetically to related species in New Zealand or elsewhere, supporting the idea that they arrived on the Chathams within the past few million years.

  In short, there is no evidence whatsoever of Gondwanan relict species on the Chathams. Instead, the geological and biological evidence dovetails neatly to indicate that, although the Chathams once had a Gondwanan biota complete with southern beech trees and dinosaurs, the entire modern flora and fauna reached the islands only very recently and had to cross ocean barriers to do it. The slate was wiped clean, and a new story, with an entirely new cast of characters, was written on it.

  In great contrast to the Chathams, New Caledonia, another piece of Zealandia, harbors an array of bizarre, seemingly ancient lineages that scream out for a Gondwanan interpretation. The most famous of these are perhaps Amborella trichopoda, a rainforest shrub that is likely the sister group to all other flowering plants, and the Kagu (Rhynochetos jubatus), a secretive forest bird that resembles a heron and is apparently most closely related to the Neotropical Sunbittern (which is not a bittern). New Caledonia also has the only known parasitic gymnosperm—a podocarp shrub that grows on the roots of another podocarp species—and, until people arrived and probably drove them to extinction, it was home to a large horned turtle and a small terrestrial crocodile that were the last surviving members of their respective genera. The island’s deep history is also suggested by substantial endemic radiations of skinks, geckos, and Araucaria conifers, among other groups.