… there’s a special providence in the fall of a sparrow.
—WILLIAM SHAKESPEARE,
Hamlet
And for myself I am fully convinced that there does exist, in Nature, means of Selection, always in action & of which the perfection cannot be exaggerated.
—CHARLES DARWIN,
Natural Selection
Before he leaves Daphne’s north rim, Peter Grant stoops and scans the dirt by the path. With his floppy-brimmed sun hat and gray beard he looks both cheerful and grave, reading the dust. He is hunting for a Tribulus seed.
“There was a time when you’d just say, ‘There’s one, there’s one, there’s one.’ Now you’ve got to search for them,” he says.
He kneels beside a Tribulus plant, or what is left of it. After almost four years of drought, the plant has withdrawn to the roots. It looks like a black claw hiding from the sun. All around it the lava is coated with layers of old guano, and the glare on this white paint makes Peter squint, even though the morning sky is still overcast. He brushes aside a pebble or two.
“Here’s a Trib,” he says at last, holding it up in his palm. The Tribulus plant dropped this mericarp during the last wet season, to wait for the next wet season. Now the plant is a withered stump, and the mericarp, still waiting, has bleached to the color of driftwood.
Though it is guarded by two long sharp spines, this mericarp has been chipped open at one end. Within the broken place, Peter can see two dark pod holes, side by side, like tiny eye sockets, both empty. “Just two seeds taken out,” he says.
At this moment, up and down the volcano, four hundred Darwin’s finches are doing what Peter has just done. They are turning over pebbles, inspecting the lava, raking the cindery dust with their claws, sometimes poking their heads down into dark crevices, looking for the last bleached seeds. To open new ground, one of them will sometimes brace its head against a big rock and roll over another rock with its feet. A finch that weighed less than 30 grams was once seen rolling over a rock that weighed almost 400. That is like a man rolling a boulder that weighs one ton. It is the labor of Sisyphus, and unlike Sisyphus, Darwin’s finches cannot keep it up forever. They are wearing down the horny sheaths of their beaks. Some of them have scraped their feathered crowns almost bald. Their occasional reward is a treasure like the object in Peter’s palm, one more tough vitamin capsule, a husk with a few kernels no one else has eaten.
In the Grants’ first four years on this island, they never saw the struggle for existence get this intense. Those were good years for Darwin’s finches. By the end of the Grants’ first season, for instance, there were about fifteen hundred fortis on Daphne Major. Nine out of ten of those fortis were still alive in December, just before the next rains came. There were also about three hundred cactus finches on the island that first April, and nineteen out of twenty of them survived the dry season and made it through to December.
Their fourth year, in 1976, was especially wet and green. There were great bouts of rain in January and February, and light showers in April and May, a total of 137 millimeters of rain, which is a good year for Darwin’s finches.
The fifth year of the study, 1977, began well too. Rain fell right on schedule in the first week of January. Within days, green leaves unfolded and flower buds opened all over Daphne Major, with here and there a few caterpillars crawling on the buds, fast food for Darwin’s finches. There were more than one thousand fortis and almost three hundred cactus finches on the island.
By this time the Grants’ first pair of colleagues in the islands, Ian and Lynette Abbott, had gone back to Australia, and Jamie Smith had gone back to Canada. The Daphne watch had been taken over by Peter Boag and Laurene Ratcliffe. Boag was eager for the island’s fortis to start laying eggs, because he needed more parents and offspring for his study. He had also received special permission from the staff of the National Park Service of Ecuador to perform his egg-switching experiment. After the fortis eggs hatched, he was planning to band the foster chicks. The next season he would measure their beaks’ length, depth, width, and finish his Ph.D. with a bang.
After the first rain, a few pairs of cactus finches mated. (Cactus finches often breed before much rain has fallen, perhaps because they make most of their living from cactus.) The birds laid their eggs in nests they built in the cactus trees, and the eggs hatched fine.
Fortis do not breed until a little more rain has fallen. Peter and Laurene waited for the next good cloudburst, the one that would trigger the fortis to mate. But after the first week of January the sky above Daphne Major was like the sky that hangs over the Grants this morning, gray, low, and gloomily quiet. There was one more shower, a very light one, then nothing but clouds and heat.
The small rain that fell in the first week of the year did not settle into the soil. There is no place on Daphne’s slopes for water to pool, and there is not much dirt to soak it up. The rain ran down the sides of the volcano as if pouring off a roof, and trickled away into the sea. Whatever was left was baked away by the sun or dried in the sea breezes on the rim and in the eddying winds that circle within the great bowl of the crater, stirred as if by a fire in the crater bowl each morning, as the rising sun begins to heat the lava.
Peter and Laurene checked on the nests when the cactus finch chicks were seven days old. Cactus finches build domed nests deep inside the cactus bushes, where they are well protected from owls. The chicks were peeping away with as much energy as they do every year. Peter and Laurene reached into each cactus bush, lifted the chicks out, set them down in a hat, and measured them one by one. In a normal year, flies and moths would have been flitting around the cactus trees by now, and the chicks’ mothers and fathers would have been bringing bugs to the chicks. The air should have been so thick with bugs that three sweeps of a net through the air would come up with hundreds. But at the end of that January the island was still so dry that there were few flowers to bring out the insects. Three sweeps of the net came up with only a couple of bugs.
When Boag and Ratcliffe sampled the contents of the chicks’ crops with an eyedropper they saw that most were almost empty. They found a little pollen, a few bits of flowers, or a seed kernel, sometimes a small spider.
All over Daphne, leaves shriveled, flowers wilted. Boag and Ratcliffe had not planned to stay on the island this long. The way things usually went, they would go early in the year and band the fledglings in their nests on the seventh day. Then they would leave the island and come back later on to see how all the young birds were doing. But this year, Peter and Laurene were stuck. They could not leave the island until the fortis had their fledglings.
From the upper rim of the crater, Boag watched the horizon. The wind usually comes from the south, and it is blocked by the much larger and taller island of Santa Cruz, which lies 8 kilometers to the windward of Daphne Major. Storms drop most of their rain on the south side of Santa Cruz, so that the north side of the big island, and all of Daphne Major, lies in a rain shadow. Boag could see regular downpours along the coast of Santa Cruz. The island of Santiago, 30 kilometers to the northwest, where Darwin passed a fortnight, got soaked too that spring.
“We panted, drank water, and read books,” says Laurene. “I read The Agony and the Ecstasy. Peter read every single book on World War II.”
“That’s the place to read The Rise and Fall of the Third Reich,” says Peter.
A few scattered showers fell on a few scattered days, 24 millimeters of rain in all. This was not enough to move the fortis to pair off and mate, and it was not enough to fill the air with moths and flies. Two out of three of the cactus finches’ chicks died in the nest, and those fledglings that did make it out stayed close to their parents for twice as long as usual, some of them for more than a month. They hopped beside their mothers and fathers and begged with piercing cries and much shaking of their wings.
During the previous June, when the island was wet and green, there had been more than 10 grams of seeds in an average square meter of lava. The finches had already eaten their way through many of those seeds during the dry season of 1976. Even if the rains had fallen as late as March or April of 1977 the seed supply would soon have rebounded, and the fortis would have begun to pair off and breed and lay eggs for Boag’s experiment. But the weeks went by, and the rains did not come, and the birds did not pair off. Day after day they went on pecking over the same square meters for the same diminishing supply of seeds. By June of that year there were only 6 grams of seeds per square meter. By December there would be only 3 grams.
As they always do in dry times, the birds went on looking for the easiest seeds. But now they were sharing the last of the last of the pistachio nuts. They were down to the bottom of the bowl. In June of the previous year, four out of five seeds that a finch picked up were easy, scoring less than 1 on the Struggle Index. But as the small, soft, easy seeds of Heliotropium and other plants disappeared, the rating climbed and climbed, peaking above 6. The birds were forced to struggle with the big, tough seeds of the Palo Santo, and the cactus, and Tribulus, symbol of the struggle for existence, a seed sheathed in swords.
Back in 1973, it had been quite rare to see a fortis try to crack one of the iron stars of the Tribulus, and when one of the birds did try, it needed an average of almost fifteen seconds to crack the mericarp. Boag and Ratcliffe got out the stopwatch again. A fortis could now get the side of its beak across a corner of a mericarp and twist, twist, until the capsule chipped in less than six seconds. The birds were getting a lot of practice with Tribulus—that is, those birds that could handle it at all.
Some of the very smallest fortis on the island, the ones whose beaks were too small for caltrop, were poking around the Chamaesyce instead. The herb Chamaesyce has small soft seeds, but it also has a milky, sticky latex when its leaves are wounded and its stems are broken. These little fortis, along with the very smallest birds on the island, the immigrant fuliginosa, began hunting for seeds in the Chamaesyce in spite of its latex. The feathers on the crowns of their heads got so matted, gummy, and sticky that they rubbed off afterward as the birds raked the cinders and gravel looking for more seeds. Their bare scalps were exposed to the sun all day. Boag and Ratcliffe began to find little bald finches lying dead on the lava.
They kept up the routine of capturing and measuring finches, dangling them in the weighing cup and recording the numbers in their waterproof notebooks (that year they did not need them to be waterproof). By June, many of the birds’ weights were down as much as a quarter from the June before. A large number of these finches had failed to molt, although by now they badly needed a new set of feathers. Some of their contour feathers were so worn that the down underneath was exposed, as Boag would report later in his now-famous paper on the drought of 1977. He and Laurene found dead fortis lying on the lava with feathers so disheveled they looked as if they had been combed the wrong way. The fraying of feathers was hardest on the smallest species, the fuliginosa, which when not eating the small, soft seeds of Chamaesyce were poking about for bugs in the lichen on the torchwood trees, Boag says, or hawking for bugs from a naked branch. They needed all their contour feathers for flight.
Even in good years, finch watchers have to be careful never to leave a bucket of water standing open in camp, or Darwin’s finches will jump in and drown. On one island another Galápagos biologist, an iguana watcher, did leave a jerry can open once, and in the morning it was full of finches. The bucket is like an oasis—it draws thirsty animals from far and wide. Once at the Charles Darwin Research Station a centipede a foot long crawled into an open bucket, and as grasshoppers hopped in, the centipede ate them one by one.
This year the whole camp on Daphne Major became an oasis of sorts. A flock of finches—mostly juveniles, birds born the green year before—hung around the finch watchers’ tent and picked up crumbs. Peter and Laurene grew particularly attached to one female finch—“Number 1750 or something like that,” Boag says. “She would follow us around the camp. She didn’t make it through the drought, unfortunately.”
Down on the crater floor the blue-footed boobies shifted their weight from one leg to the other to cool off their webbed feet. Boag stuck a thermometer into the ground, in the tortured shadow of a cactus, and the soil was hotter than 50° C (122° F). Even where there were seeds lying out in the open, the heat was keeping the finches from foraging there between the hours of 11 a.m. and 3 p.m. Meanwhile at night Peter and Laurene would be shivering in their tent if temperatures got down to 75° F, because their bodies were so used to the heat. Boag lay awake and wondered how his flocks were doing, just as biblical shepherds once did in a far-distant desert: “In the day the drought consumed me, and the frost by night.”
Now and then a frigatebird harried a blue-footed booby out of its kill of fish. If the fish dropped on the island, as many as ten or twenty finches would flock around it. They also scavenged broken eggs and fresh booby guano. They hung close when the boobies fed their young and fought for the fish scraps, and when owls left something of their kill, finches fought over that too.
In other years the finches had ignored the lava lizards that scuttle about the rocks. But once that year, Peter and Laurene saw a female cactus finch eating a black lizard tail, and nearby they spotted a female lizard with a freshly broken stump. Some days later, they saw the same bird chase after another female lava lizard, pecking at its tail. She might have started a new entry on the finch menu—but that was the end of the episode. Another time they saw a blue-footed booby wounded by a frigatebird. A fortis stood beneath the wounded booby and drank the blood as it dripped on a rock.
“We just sat there, month after month,” Boag says. “At the time, we were depressed. We were losing the breeding season, so we wouldn’t get a generation. Plus, all these birds were disappearing. We kept up doing the normal checks and censuses. But our feeling was not the thrill of seeing evolution in action, as one might conclude from reading the subsequent papers, but the moderate despair of doing a research project and seeing your birds dying.”
All of the cactus finch fledglings died before they were three months old. Not a single fortis laid an egg or built a nest. Of course, this was the whole design of the study: to watch and see—“on spec,” says Boag—if there were going to be any minor selective episodes. But now that a selective episode was definitely in progress, Boag was wretched. The egg switch would have been a wonderful experiment, and he was sure that watching natural selection in action would never make him Dr. Boag. At best, the events that he and Laurene were documenting this year would make a page of someone else’s thesis a long way down the road. “We thought it very unlikely that we’d be able to measure it at all,” Boag says. “We thought, watch ten years, and then maybe. So I didn’t recognize what was going on—the magnitude of the effect.”
THEY WENT HOME at last and went through their data. There were only five or six months between field excursions, so they had very little time before they had to go back into the field for another round: just enough to get the data out of their notebooks and type them up while they could still interpret their scrawls and scribbles. Peter and Laurene got their data out of the field books and saw who was left alive and who had died. But there wasn’t much time.
Boag pondered the shambles of his thesis. To get a large enough statistical sample of family resemblances he had to measure many hundreds of adults and their young. Even in a normal year only a fraction of the adults that he measured would breed. Only a fraction of those that bred would build a nest that he and Laurene would find. He would band all of the nestlings he found, and about half of those would die. He might measure two thousand birds, and he might end up with only one hundred of their offspring. “They’re very precious birds, if you see what I mean,” he says now. “Each dead bird is a lost data point. So that was my main concern. I was not thinking of each bird as a data point for selection. I was thinking of it as a lost data point for heritability.”
When he and Laurene went back in January 1978, they watched the bright yellow flowers of the cactus trees open up all over the island, as they do at the turn of each year. All of Darwin’s finches converged on the blossoms, gorging themselves on the pollen and drinking the nectar.
Boag and Ratcliffe did the usual census. They found fewer than two hundred finches alive on the island. Just one finch in seven had made it through the drought. They measured these survivors, and they also measured the mummified carcasses of the dead finches they found lying on the lava, banded and unbanded. The island is large enough that the finches that disappear each year usually disappear without a trace. They never found the body of their camp finch or most of the others. But they collected all the numbers they could, and at the end of that season, they went home again and they typed them into the computer for analysis.
Today when they lecture on the selection event of 1977, Peter Grant and Peter Boag plot the effects of the drought in three curves. The curves start in March 1976, when the island of Daphne Major was still green and lush. They end in December 1977, when the cactus flowered and the worst of the drought was over.
All through the drought the total mass of seeds on the island went down, down, down. The average size and hardness of the remaining seeds went up, up, up. The total number of finches on the island fell with the food supply: 1,400 in March 1976, 1,300 in January 1977, fewer than 300 in December.
Next they take the finches species by species. At the start of 1977 there were about 1,200 fortis on Daphne. By the end of the year there were 180, a loss of 85 percent.
At the start of the year there were exactly 280 cactus finches on the island. By the end of it there were 110, a loss of 60 percent.
Of the smallest ground finches, fuliginosa, there were a dozen on the island at the start of 1977, and only one of them survived the year.
They also plot the age of the survivors. Many of the survivors were the oldest birds on the island, and had been banded by the Grant team in 1973. Not a single fortis was born that year on the island, and only a single one of the fortis that had been born the year before survived the drought. Only one of the young cactus finches born the year before survived. The drought practically wiped out the cohort that was born the year before that too. That generation became rare, and rarer with each passing year, like steel pennies minted in a war.
At last, Grant and Boag look at the beaks of the survivors. They know how variable the beaks are. They know how much the variations matter. They know how the plants were doing, what the weather was doing, how life on the island was squeezing the finches. They know all these figures with unprecedented precision, as well as the dimensions of the finches that made it through the drought and the finches that did not.
Among fortis, they already knew that the biggest birds with the deepest beaks had the best equipment for big tough seeds like Tribulus; and when they totted up the statistics, they saw that during the drought, when big tough seeds were all a bird could find, these big-bodied, big-beaked birds had come through the best. The surviving fortis were an average of 5 to 6 percent larger than the dead. The average fortis beak before the drought was 10.68 millimeters long and 9.42 deep. The average beak of the fortis that survived the drought was 11.07 millimeters long and 9.96 deep. Variations too small to see with the naked eye had helped make the difference between life and death. The mills of God grind exceeding small.
Not only had they seen natural selection in action. It was the most intense episode of natural selection ever documented in nature. One result was a bizarre tilt of the sex ratios on the island. At the start of the drought there were about 600 males and 600 females. By the end of the drought more than 150 of the males were still alive, but only a pitiful remnant of the females. Males are typically larger than females by 5 percent, with proportionately bigger beaks, so the males generally had an edge.
In other words, among the males the biggest survived, and among females the biggest survived, but many more males survived than females. And what made the difference between life and death was often “the slightest variation,” an imperceptible difference in the size of the beak, just as Darwin’s theory predicts.
Many people—even biologists, even today—find the power of slight variations hard to believe. “Once, just as I was beginning a lecture,” says Peter Grant, “a biologist in the audience interrupted me: ‘How much difference do you claim to see,’ he asked me, ‘between the beak of a finch that survives and the beak of a finch that dies?’
“ ‘One half of a millimeter, on average,’ I told him.
“ ‘I don’t believe it!’ the man said. ‘I don’t believe a half of a millimeter really matters so much.’
“ ‘Well, that’s the fact,’ I said. ‘Watch my data and then ask questions.’ And he asked no questions.”
“None,” Rosemary agrees. “And he sat there scowling, fidgeting and talking the whole time.”
NATURAL SELECTION BY ITSELF is not evolution. It is only a mechanism that, according to Darwin, can lead to evolution. As Peter and Rosemary Grant put it, natural selection takes place within a generation, but evolution takes place across generations.
In the drought of 1977, they had seen and documented natural selection in action. The decimation of the finches by selection had been as ruthless as the aristocratic breeder of bulldogs in Darwin’s day who said, “I breed many, and hang many.”
But the finch watchers did not yet know if the episode would translate into an evolutionary change. They only knew that, according to theory, it was possible, since the beak variations are heritable: the changes that are wrought upon one generation can be passed on to the next, becoming muffled and compressed or stretched and warped, over the years, as they pass down the line of the generations and onward into the future.
This is a step the founders of Darwinian theory considered logically inevitable, but which many of those who came after Darwin have doubted. Raymond Pearl again: “In the minds of an astonishingly large number of people, which number includes some rather great names in the world of science, it is precisely the same thing to show that something logically must be so, as it is to show that it is so. If the formal rules of logic are satisfied, truth seems to them to be thereby established. No further evidence is demanded. As everyone knows, this attitude led practically to the intellectual bankruptcy of the whole evolution theory.…”
On January 9, 1978, the clouds rained at last on Daphne Major. More than 50 millimeters fell that day. The rain fell on nothing green, only rocks and dead-looking trees and withered weeds. The rain streamed down the sides of the mountain.
All over the island, the male finches, the drought’s survivors, did what they do each year in the first bout of rain. They flew to the highest points in their territories: to the crown of a tree that rises from a fissure toward the sky, or the crazy steeple of a cactus on the summit of a rockfall. Perched on these wet command posts, looking as skeletal and tattered as they had ever been in their lives, each cock opened his famous beak, like a rooster in a barnyard in the first light of day, and began singing in the rain.
The rain transformed the island. Within a week there were leaves and flowers on the torchwood. Green stems shot up before the finch watchers’ eyes. “Merremia seedlings were 5 centimeters tall,” Peter Boag reported afterward; “Portulaca was in complete leaf, and Amaranthus was 2 centimeters tall.” Soon Tribulus and a dozen other plants had green fruits or seed heads, and the buds of the Portulaca were crawling with bugs. From the sea the dusty sides of the old volcano turned from its dull predawn shade of brown to a noon-green emerald, a tropical paradise.
Not a single pair of finches on the island had survived intact. The drought had taken the life of at least one member of every couple. But as the rains fell, many of the females’ beaks began turning dark brown, and the males’ beaks turned black, signs that the birds were ready to mate once again.
The males built nests in the cactus and sang for days on end from the highest cactus top they owned. The females hopped from territory to territory, inspecting the nests and presumably the singers.
Of course the skewed sex ratio put a spin on that breeding season. Among fortis there were now six males for every female. Each female could choose among many males, but only one male in every six could win a female.
The males flew after the females that visited their territories in what the finch watchers call the “sex chase.” The females hopped around and flew around to visit nest after nest, and took part in chase after chase, before one by one they each settled down with a single male.
Again the finch watchers watched and measured. They found that the males the females had picked were not a random sample, any more than the ones the drought had spared were a random sample. The successful males tended to be the largest of the large. They were the males with the very blackest, most mature plumage and the ones with the deepest beaks.
Because of the crazy sex ratio, most of the males were left out; only a very small subsample of survivors had a chance to mate. But every single one of the eligible females was able to pair off. One female cactus finch set a record for the island, breeding five times and producing thirteen young.
Now it became of great significance that variations of body and beak are passed on from one generation to the next with fidelity. As a result, the males’ unequal luck in love helped to perpetuate the effects of the drought. The male and female fortis that survived in 1978 were already significantly bigger birds than the average fortis had been before the drought. Of this group the males that became fathers were bigger than the rest. And the young birds that hatched and grew up that year turned out to be big too, and their beaks were deep. The average fortis beak of the new generation was 4 or 5 percent deeper than the beak of their ancestors before the drought.
In the drought of 1977 the Finch Unit had seen natural selection in action. Now in its aftermath they saw evolution in action, in the dimensions of the birds’ beaks and in many other dimensions too.
After that, the watchers on Daphne Major had to keep watching. They had to keep coming back. Not only is Darwin’s process in action among Darwin’s finches, not only can natural selection lead to evolution among their flocks, but it leads there much more swiftly than Darwin supposed possible. The finch watchers had to find out what would happen next.
But even if they had quit there, what they had seen on Daphne Major from 1973 through 1978 would be enough to fill an old and rather embarrassing lacuna in Darwinism. In 1909, at the centenary of Darwin’s birth, during a scientific meeting at Cambridge University, the German biologist Weismann asked whether natural selection can really explain the first small steps of evolutionary change. “To this question even one who, like myself, has been for many years a convinced adherent of the theory of selection, can only reply: ‘We must assume so, but we cannot prove it in any case.’ ”
Several human generations later, Darwin’s descendants no longer have to assume. They can supply what Weismann wanted with case after case and, now, with Darwin’s finches.
One of Peter Grant’s graduate students, Trevor Price, has reviewed those early years using powerful mathematical tools that did not exist in the 1970s. These tools allow investigators to disentangle which among all the changing features of a bird or a fish or a fern is most strongly selected during a selective episode. That is, they help show investigators which changes in the living form were essential and which were simply following along, which parts of a living form were the targets of selection. This technique (it is known as partial regression analysis) was developed in 1983 by the evolutionary theorists Russ Lande and Steve Arnold. As soon as Lande and Arnold published the technique, Price applied it to Boag’s drought. This reanalysis brought the evolutionary event into even sharper focus.
Price knew that the survivors and their offspring were larger in weight, wing length, tarsus length, and also in beak length, depth, and width. However, partial regression analysis shows that not all of those were selected by the drought with equal emphasis. During that terrible drought on Daphne Major, among fortis, nature was selecting most powerfully for bigger body size and deeper beaks. Nature was not selecting for longer beaks; a fortis with a long beak had no special advantage in the drought. And nature was rejecting the birds with wider beaks. So it was big birds with deep but relatively narrow beaks that were favored: perhaps, Peter Grant writes, “because a narrow yet deep bill was the best instrument for performing the difficult task of tearing, twisting, and biting the mericarps of Tribulus to expose the seeds.”
So the birds were not simply magnified by the drought: they were reformed and revised. They were changed by their dead. Their beaks were carved by their losses.
In most places on this planet, the sight of a dead bird is so rare that it shocks us, even scares us. We recoil as if something has gone wrong in the cosmos, as if a shutter has creaked open that should have been kept closed, exposing a shadow world beyond our world, a place we were not meant to see.
But on the desert island of Daphne Major, dead birds are commonplace. They are everywhere. The lava is always littered with wishbones and beaked skulls. Whole seabirds lie outstretched here and there as if still in flight, odorless and mummified like feathered pharaons in the dry and desiccating heat. Each generation lies where it falls, and the next generation builds on the ruins of the one before. They hatch in a morgue, breed in a crypt, and lie down with their ancestors, as if here not only life but death too is asking to be watched.
Evolution discloses a meaning in death, although the meaning is like some of the berries that Darwin tasted in the Galápagos, “acid & Austere.” There is a special providence in the fall of a sparrow. Even drought bears fruit. Even death is a seed.