Life Finds a Way
Page 6
Unfortunately, try as they might, nature’s children are sometimes forced into incestuous relationships. Whenever a population is decimated by a cataclysmic storm, hunted to near-extinction, or swept onto a small and remote island, only a few individuals may be left to keep the population alive and the gene pool from drying up. Among all such disasters that can befall a population, exile on a small island is especially fascinating to biologists. That’s not only because sandy beaches and unspoiled forests make great destinations for field research. Islands are like isolated laboratories where one can observe evolution in action, and this action holds a key to understanding the creativity of nature—and of much more than nature, as we shall see in later chapters.
Imagine four unrelated people, two men and two women, who are shipwrecked on a tiny Pacific atoll that does not provide space, water, or nourishment for many more. Not having much choice in the dating game, they form two couples and have children. The children of the two couples are genetically unrelated, and if each of these children, once grown, has children only with a member of the other couple, their children—the second generation—will also be unrelated. But when these grandchildren are ready to have children themselves, inbreeding will become unavoidable. The simple reason: all members of the second generation share the same grandparents, or about one quarter of their genes. From that generation onward, all individuals on the island will be part of the same family.
The same thing happens—everybody eventually becomes genetically related—even if the starting population is a bit larger. It just takes longer. As a rule of thumb, it takes some ten generations for ten initial individuals, or one hundred generations for one hundred individuals, to congeal into a single big, but perhaps not genetically happy, family. The more individuals, the longer it takes, but eventually a population of any size will have the same fate—everybody will be everybody else’s relative, although some individuals will be more inbred than others.11
Whether enforced by circumstances or by a breeder, inbreeding has the same consequence: genetic trouble. Manx cats from the Isle of Man, named after the Celtic natives of this tiny, 570-square-kilometer island in the Irish Sea, could tell you a story or two about this trouble. They do not have a club foot like King Tut, but their skeleton has a defect no less obvious, born from multiple generations of inbreeding: it lacks a tail. That defect impairs a cat’s balance and precludes some of the acrobatic feats that cats are known for. But it is not the worst problem of this race. Many kittens have deformed spines, or are even stillborn, and litter sizes are generally small—low fecundity. Fortunately, these disadvantages are offset by some advantages, among them the great hunting skills that made Manx cats popular as mousers on farms, and a playful, almost doglike disposition cherished by some owners.
The rough lives of inbred Soay sheep on the island of Hirta tell a similar story. Hirta is part of the windswept, treeless, and rugged St. Kilda archipelago off the Scottish west coast. The archipelago has been uninhabited since its last thirty-six human inhabitants evacuated to the mainland in 1930 once they’d had enough of their harsh life. The Soay sheep—Soay means island of sheep in old Norse—lacked this option, and maybe they did not mind, because they have lived on the islands in small numbers since time immemorial. Yet their life is no bed of roses, because many of them are plagued by intestinal worms that siphon off the scarce nutrients the sheep ingest. And when the most inbred sheep die during the winter—up to 70 percent of the whole population perishes in bad times—it is really the overwhelming number of worms that kills them, because inbreeding has weakened their immune systems.12
The same pattern can be found on Mandarte Island, a truly tiny Canadian island of almost six hectares off the coast of British Columbia, home to some one hundred song sparrows whose numbers fluctuate over the years. When a severe winter storm in 1989 left only eleven surviving birds, it had killed off the most inbred ones.13
Club-footed King Tut, worm-plagued sheep, and numerous other cases of inbreeding underscore that natural selection is not all-powerful; its relentless uphill pull in nature’s adaptive landscapes can be stymied. To understand when and why that happens, we need to understand a phenomenon that has affected life since its origins, even though it has only been appreciated since the early twentieth century. That’s when biologists like Sewall Wright first saw that studying entire populations and not just individuals is crucial to understanding evolution. Today, the phenomenon is called genetic drift, but for some time it was called the Sewall Wright effect, in honor of Wright’s contributions to its discovery.14
Wright was among the first to view a population not just as a collection of organisms but as a pool of genes or alleles. That shift in perspective is key to understanding genetic drift. Small populations have a small gene pool, and large populations have a large one. Let’s imagine a very small population with a pool of only four alleles that influence some trait, such as a person’s eye color. Imagine that our small gene pool contains two alleles for brown eyes and two for blue eyes. To understand how genetic drift affects the fate of these alleles over many generations, it is useful to think of this gene pool as a bowl containing four marbles of two different colors—such as the black and gray circles on the left-hand side of Figure 3.1.15
In the early twentieth century, geneticists already realized that passing on genes from one generation to the next—creating the next generation’s gene pool—works just like a series of blind draws from such a bowl. So, to create a new gene pool, let’s first blindly draw a marble and note its color—brown or blue—put it back in the bowl, then draw another marble (still without looking), note its color, put it back, and repeat the drawing and replacing a third and fourth time to create a new pool of four genes. (Putting the marble back before drawing again captures an important feature of inheritance: different sperm or egg cells can inherit a copy of the same allele from one of their parents.)
Figure 3.1.
At the end of this repeated drawing, we have picked four marbles. If you tried this at home, you would see that these marbles will not necessarily show the same color combination—two brown and two blue—as the parental gene pool. Sometimes we may get three brown and one blue marble, four brown ones, or, conversely, three blue and one brown marble, or four blue ones (Figure 3.1). The number of marbles of a given color corresponds to the frequency of the alleles of a different type. The main point is this: their number varies randomly from one generation to the next.
To see how such a gene pool would evolve over multiple generations, we can draw four marbles from the new pool in the exact same way to form a third-generation gene pool, then repeat that drawing to generate a fourth generation, and so on, ad infinitum.
Early twentieth-century population geneticists used the mathematics of probability theory to find out how this gene pool will change in the long run. The math is complex, but its main lessons are simple. First, the number of marbles of any one color—alleles of any one type—continues to change randomly and unpredictably until alleles of only one type—brown or blue—are left. When that happens, one of the two alleles has become extinct, whereas the other, in population genetic jargon, has become fixed in the population. The gene pool will remain unchanged from that point onward unless a DNA mutation recreates the other allele, and the chances of that are minute.16
The math also shows that while one allele will eventually become fixed—with certainty—which allele will be the lucky one is as unpredictable as the outcome of a coin toss. In 50 percent of populations, the blue allele will become fixed, and all individuals will end up having blue eyes. In the other 50 percent, all individuals will have brown eyes.
Even though human genes get reshuffled in complicated ways when our bodies make the sperm and egg cells needed to have children, the drawing of marbles from a bowl captures exactly how genes and their alleles are passed from generation to generation.17 And that we humans have two copies of each gene makes no difference either: a pool of four human genes corresp
onds to a “population” of two people, each with two copies of each gene. What I just described is how this gene pool would evolve if these two people had two children, a girl and a boy, who grow up and have another girl and a boy, who have another girl and a boy, and so on. Eventually, this highly inbred family would have all blue or brown eyes, and nobody could tell in advance which it was going to be.18
Genetic drift affects larger populations as well. As a rule of thumb, in a pool of ten genes—five individuals with two alleles each—allele numbers fluctuate by about 10 percent each generation. In a pool of one hundred genes, they fluctuate by only 1 percent, and in a pool of a thousand, by 0.1 percent. Moreover, the time a blue or brown allele takes to spread through a gene pool and become fixed is longer in large populations. For a gene pool that is ten times as large, it would take ten times as long for all individuals to carry the same allele. That’s why inbreeding is not as much of a worry in very large populations. Yes, drift does affect a gene pool of a billion, but it causes only tiny jitters in allele numbers, causing them to fluctuate by 0.0000001 percent every generation.19
A DNA mutation creates only a single new copy of any one allele, and only in a single individual of a population. For the many genetic diseases where two bad copies need to come together in the same individual, this rare allele can spread through a population via genetic drift, generation by generation, unmolested by natural selection, until the allele has become frequent enough that two bad copies begin to come together in some individuals. When these individuals die as a result, natural selection is kicking in. The smaller the population, the faster this will happen. And because we have so many genes, chances are high that bad alleles will come together for at least one or a few genes.20
That’s where the diseases caused by inbreeding come from, and that’s why different diseases occur in different inbred populations. If you could reset evolution’s clock at will and start to inbreed the same small population, you would find that the bad alleles that genetic drift helps spread are different each time.
All this means that the negative effects of inbreeding are a consequence of genetic drift in small populations. It does not matter if an island forces a population to be small, if royal politics—or conceit—restricts an aristocrat’s potential mates, if a mountain range isolates a population of villagers, or if a breeder repeatedly mates individuals from the same lineage. The result is the same: genes that would otherwise be wiped out can persist and spread through a population.
Once again, this doesn’t mean that drift and inbreeding necessarily have bad consequences. Both are simply indifferent to the concept of good or bad—they will spread bad alleles just as they spread good ones. What’s more, genetic drift is not necessarily going to lead to inbreeding. Some single-celled fungi and algae have only one copy of each gene, which means that the coming together of two bad alleles is impossible. Bacteria also have only one copy of each gene, and they reproduce without sex as we know it, so they, too, are not subject to inbreeding.21 Yet genetic drift is at work in their populations, just as it is at work in those of algae, fungi, or any other organism, because the blind drawing of genes to fill a gene pool is fundamental to all life-forms, regardless of how they reproduce. Inbreeding is not universal to life, but genetic drift is.
To visualize what genetic drift does to a population, recall the fitness landscapes of the previous chapter, and the relentless uphill pull that natural selection exerts on populations evolving on these landscapes. Because drift causes random and directionless changes in a population’s gene pool, you can think of it as an unceasing tremor, like an earthquake that causes this landscape to tremble and shake without pause.22 Any mountaineer’s ascent would be slowed by an earthquake, and a hill-climbing population is no exception. Because genetic drift is directionless, these tremors can take the population in any direction—uphill, downhill, or sideways (Figure 3.2). If the tremors are weak—in a large population—they will not delay the climbing population by much, but the stronger they are—the smaller the population—the more violently they will jerk the population around, and the more they will slow it down. Downhill lurches are especially momentous because they can bring a population to a valley where it does as poorly as those inbred royals and worm-infested sheep.
Figure 3.2.
The genetic math can also tell us how strong the tremors—how small the population—must be to override the uphill pull of selection. Take a population whose organisms differ by 5 percent in their fitness—one type of bacterium dividing 5 percent faster than another, one apple tree producing 5 percent more seeds, or one squirrel with a 5 percent greater chance of surviving a harsh winter. In such populations, drift can overpower selection if it jiggles allele numbers by more than 5 percent. That happens in populations of twenty or fewer individuals.23 If, however, a population’s individuals differed in their fitness by only 1 percent, drift could overpower selection even in somewhat larger populations because it has to jiggle allele numbers by much less—only more than about 1 percent—to overcome selection. The math shows that in this case, populations with fewer than one hundred individuals are small enough to create jitters strong enough to overpower selection and prevent the population’s ascent. The general pattern is simple: if individuals differ in their fitness by a tenth of a percent, drift could overpower selection in a population smaller than a thousand individuals, and if they differ in fitness by a hundredth of a percent, a population of fewer than ten thousand individuals would suffice.
Fitness differences as small as these are important in evolution because they are natural selection’s most abundant raw material. We know this because we can measure how strongly mutations alter fitness. The vast majority turn out to change fitness just a wee bit.24 Such small effects, however, can accumulate since evolution unfolds over millions of generations or years. In the long run, even mutations that cause fitness differences smaller than a millionth of a percent can mean the difference between survival and eventual extinction25—as long as they occur in a population larger than 100 million individuals. In a population that is any smaller, selection would be powerless to overcome genetic drift.
I have used alleles that affect eye color to illustrate the action of random genetic drift because eye color seems an especially innocuous trait, but it turns out that people with lighter, bluer irises are slightly more prone to getting some cancers of the eye.26 Thousands of other genes in our genomes affect traits much less innocuous than eye color—the strength of our bones, the vigor of our immune system, or plain old fertility. Because all genes are inherited according to the same rules, “bad” alleles of these genes can also spread through a population via random genetic drift as long as a population is small enough and drift is strong enough.
In the fourteenth century, the poet Dante Alighieri reached a summit of world literature with the epic known as the Divine Comedy. The creative process guiding him is lost to history, but for some creators, this process may resemble the journey that the Divine Comedy’s protagonist has to endure: he must descend through the nine circles of hell, whose inmates suffer increasingly gruesome and imaginative forms of torture, before he can eventually ascend through purgatory into the nine celestial spheres of heaven.
Those philistines who are indifferent to this epic’s flowering allegories might just see it as an elaboration of an age-old principle: for things to get better, sometimes they first need to get worse. This principle, it turns out, is much more profound than they realize because it applies far beyond the human realm. Life’s four-billion-year-long journey of evolution would not have gotten very far without it. That’s because a population stuck at a dead-end, low peak of an adaptive landscape cannot be led away from it by natural selection, which is powerless to go anywhere but up. Genetic drift, however, can help the population descend into the hellish cauldron where new and successful combinations of genes are forged.
Unfortunately, descending into this genetic hell is more than just dangerous: many populati
ons never even reach purgatory. They simply go extinct. And only their remains can tell us about their tragic fate.
These remains are especially eloquent on islands, where colonizers are few and genetic drift is strong. On Hawaii alone, more than 30 percent of all flowering plant species that colonized the islands never made it. And insects fared even worse. A combination of ill-adapted genes and a hostile environment extinguished more than 150 of their species, some 80 percent of the original colonists.27
But to those populations that were lucky enough to persist and turn the vale of tears into their base camp, many new peaks became reachable, together with new forms of making a living. Once again, islands tell this story best. They are not only graveyards of extinct species, but also wellsprings of evolution’s creativity.
On the tiny Galápagos Islands, a single founding colony of finches diversified into fourteen different species, some of which Charles Darwin discovered when he visited on the HMS Beagle in 1835.28 On Hawaii, at least thirty species of nectar-feeding honeycreepers evolved, and on the Canary Islands off the African west coast, twenty-three new plant species appeared in the genus Echium—relatives of the blueweed, a modest flowering plant with an eye-catching blue inflorescence.29 More than 90 percent of one thousand species of flowering plants and more than 98 percent of five thousand species of insects found today on Hawaii have emerged there.
Even more remarkable than these numbers is the explosive speed at which evolution created them. The oldest islands in both the Galápagos and Hawaii have been around for barely five million years, about the same time that separates humans from chimpanzees—a brief moment in evolutionary time that sufficed to create thousands of new island species.30 But nature’s creativity is not just about speed and the number of species. Many new island species also have new lifestyles.31 The first finches on Galápagos fed on soft insects, but some of today’s species have evolved oversized nutcracker-like beaks to crush the hardest seeds to be found. On the Canary Islands, some relatives of the modest blueweed have evolved into eighteen-foot-high wooden giants supported by a drought-resistant root system and crowned by a gaudy cylindrical inflorescence beloved by gardeners.