Cook reached as far as 60 degrees south on his first passage round the Horn. But, either way, once they had made it into the Pacific, navigators found it almost impossible to advance. They could make no headway in these latitudes against the ferocious westerly winds. Southward lay the ice and snow—they were nearly to the Antarctic Circle—east was the coast of South America, and the only direction open to them was north.
It is this particular set of circumstances—winds, distances, continental obstacles, and sailing capacity—that explains a curious fact about early European encounters in the Pacific, which is that, even with the whole, wide ocean before them, almost all the early navigators followed variants of the same route. With one or two exceptions, they crossed the South Pacific on a long northwesterly diagonal, or, more properly, a dogleg, sailing north and then turning west once they picked up the trades. They did this not because they thought it was the path most likely to yield important discoveries—as the historian J. C. Beaglehole drily observed, “sailing on some variant of the great north-west line, of necessity a ship made through a vast deal of empty ocean”—but because it was the path dictated by the currents and the winds. As a consequence, many important islands that lie off this route, like Hawai‘i, were not encountered for centuries, while others, some of them minuscule, like the tiny atoll of Puka Puka in the Tuamotu Archipelago, were discovered over and over again.
THE TUAMOTUS, ALSO known as the Low or Dangerous Archipelago, feature in almost every early European account of the Pacific, for the simple reason that they lie directly across most variants of the great northwest line. A screen of some seventy-eight “low islands,” or atolls, the Tuamotus stretch for eight hundred miles along a northwest–southeast axis about halfway between the Marquesas and Tahiti. Most of these atolls are comparatively small, on average perhaps ten to twenty miles wide, but their key feature, at least from a navigator’s point of view, is their height. None of these islands reach an elevation of more than twenty feet; most are barely twelve feet above the tide line at their highest point. They are, as Stevenson put it, “as flat as a plate upon the sea.” What this means for sailors is that they are invisible until one is all but upon them, and later navigators, who knew more about what they were getting into, tended to avoid this maze of reefs and islands that was also sometimes known as the Labyrinth.
From the air, the Tuamotus are a dazzling sight: bright circlets of green and white floating like diadems in a sapphire sea. But, as the early explorers quickly discovered, up close there is not much to an atoll. Barely an island at all, it is really a necklace of islets, or motu, to use the Polynesian word, strung along a circle of reef. The motu are composed entirely of coral: sand, cobbles, coral blocks, and a kind of conglomerate known as beachrock. Verdant from a distance, they in fact have only the thinnest layer of topsoil and can support just a few salt-tolerant species of shrubs and trees. There are no natural sources of fresh water apart from rain, though there is an interesting phenomenon known as a Ghyben-Herzberg lens. This is a layer of fresh water which floats on top of the seawater that infiltrates the porous coral rock. Under the right conditions—the island cannot be too small, it cannot be in a state of drought, the well cannot be dug too deep—it is possible to extract fresh water from a pit dug into the sand, as a group of seventeenth-century Dutch sailors accidentally discovered on an atoll they named Waterlandt.
It was Charles Darwin who first articulated the theory of how coral atolls are formed. On his way across the Pacific in the Beagle, Darwin sailed through the Tuamotu Archipelago, recording his first impression of an atoll as seen from the top of the ship’s mast. “A long and brilliantly white beach,” he wrote, “is capped by a margin of green vegetation; and the strip, looking either way, rapidly narrows away in the distance, and sinks beneath the horizon. From the mast-head a wide expanse of smooth water can be seen within the ring.” It was already understood in Darwin’s day that corals were creatures—“animalcules,” as one writer put it—and that they could grow only in comparatively shallow water. And yet, here they were in the middle of the ocean, in a place where the water was so deep it could not be measured by any conventional means. (The Dutch named a second atoll Sonder Grondt, i.e., “Bottomless,” because they could find no place to anchor.) The obvious question concerned their foundation, or, as Darwin put it, “On what have the reef-building corals based their great structures?”
One theory popular at the time was the idea that atolls grew up on the rims of submerged volcanic craters. There were good reasons for associating them with vulcanism—high islands and low islands are found in close proximity throughout the Pacific. But there were also problems with the crater idea: some large atolls exceed the size of any known volcanic craters; some small atolls exist in clusters that cannot easily be envisioned as craters; and many volcanic islands are closely surrounded by coral reefs, which, if the theory were correct, would make them volcanoes within volcanoes—an explanation that seems unlikely at best.
Darwin’s notion, still the most widely accepted view, was that there is an association between atolls and volcanic islands, but that atolls begin life not on the rims of extinct craters but in the shallow waters of an island’s shores. Like many of Darwin’s ideas, his theory of coral atoll formation had the virtue of explaining not just how atolls are formed but how that process is linked to other kinds of coral formations, thus neatly accounting for all instances of what is essentially the same thing. He recognized that fringing reefs (on the shores of islands), barrier reefs (surrounding islands at some distance from the shore), and atolls (rings of coral without any island at all) are, in fact, a series of stages. The key to connecting them was the concept of subsidence—the idea that an island gradually sinks while the coral encircling it continues to grow. Thus, in the course of time, a fringing reef would become a barrier reef, and a barrier reef would eventually become an atoll.
AN ATOLL IS a very natural habitat for anything that swims or flies through the air. Atolls are home to more than a quarter of the world’s marine fish species, a mind-boggling array of angelfish, clown fish, batfish, parrotfish, snappers, puffers, emperors, jacks, rays, wrasses, barracudas, and sharks. And that’s without even mentioning all the other sea creatures—the turtles, lobsters, porpoises, squid, snails, clams, crabs, urchins, oysters, and the whole exotic understory of the corals themselves. Atolls are also an obvious haven for birds, both those that range over the ocean by day and return to the islands at night and those that migrate thousands of miles, summering in places like Alaska and wintering over in the tropics.
For terrestrial life, however, it is quite a different matter. A typical atoll in the Tuamotus might support thirty indigenous species of plants and trees—as compared with the more than four hundred native plant species that might be found on a high island like Tahiti, or the many thousands that grow on a large continental island like New Zealand—and, among land animals, only lizards and crabs. While there are places on an atoll where one might, for a moment, imagine oneself to be surrounded by land—places where one’s line of sight is blocked by trees or shrubs—a few minutes’ walk in any direction will quickly dispel the illusion. Strolling the length of even a largish motu, you eventually come to a place where you can see water on both sides. At such moments it becomes breathtakingly clear that the ground beneath your feet is not really land in the way that most people understand it, but rather the tip of an undersea world that has temporarily emerged from the ocean. The real action, the real landscape, is all of water: the great rollers that boom and crash on the reef, the rush and suck of the tide through the passes, the breathtaking hues of the lagoon.
And yet, when Europeans first reached the Pacific, they found virtually all the larger atolls inhabited. Even those that were clearly too small to support a permanent population often showed signs of human activity. On one tiny, uninhabited atoll, an early explorer found an abandoned canoe and piles of coconuts at the foot of a tree; on another there was the puzzling presence of unaccompanied dogs. Even the pit in which the Dutch sailors found water had almost certainly been dug by someone else. What all this appeared to suggest was that even the most insignificant and isolated specks of land were being visited by people who could come and go.
There are not many good early descriptions of these people. The Tuamotus offered almost none of the things that European sailors needed—namely, food, water, and safe ports—and their complex network of reefs was dangerous to ships. With so little to be gained and so much to be lost, Europeans tended