SKIN-DEEP
We now understand quite a bit about the genetic changes that led to light skin in northern Eurasians. At least we know what happened in Europe and Asia (China and Japan), the non- African populations studied in the HapMap.
In each of these populations, a number of genes have been replaced by—or are in the process of being replaced by—new variants that produce lighter skin color than the dark skin seen in sub-Saharan Africans, who mostly have the ancestral human alleles. Interestingly, the sets of changes driving light skin color in China are almost entirely different from those performing a similar function in Europe. In most cases the mutations involve changes in different genes, and even when the same gene is involved, usually the common mutations at the opposite ends of Eurasia are not the same. So in this example, as in lactase persistence and a number of other cases, it turns out that similar traits in different populations are the product of convergent evolution and are quite different at the level of biochemistry and genetics. Sometimes racial similarities are only skin-deep.
Many of these changes seem to be quite recent. The mutation that appears to have the greatest effect on skin color among Europeans and neighboring peoples, a variant of SLC24A5, has spread with astonishing speed. Linkage disequilibrium— that is, the degree to which the genome is surprisingly uniformaround this gene—suggests that it came into existence about 5,800 years ago, but it has a frequency of about 99 percent throughout Europe and is found at significant levels in North Africa, East Africa, and as far east as India and Ceylon. If it is indeed that recent, it must have had a huge selective advantage, perhaps as high as 20 percent. It would have spread so rapidly that, over a long lifetime, a farmer could have noticed the change in appearance in his village. Again, if it is that recent, it must have had a more limited distribution in early historical times, particularly in peripheral areas: In fact, this may explain the Roman impression that the Picts of Scotland were dark-skinned.
As noted in Chapter 3, the recent sweeps of genes causing light skin might have been driven by an increased need for vitamin D among farmers living in high-latitude regions with low levels of ultraviolet radiation. But there are other possibilities. In the Old World tropics, such as sub-Saharan Africa, Melanesia, and New Guinea, the ancestral condition—dark skin—was favored by selection. Palefaces didn't prosper. But in higher- latitude regions, such as Europe and northern Asia, skin could be lighter. Many genes have more than one function: It may be that genes that produce dark skin pigments were now free to change in ways that enhanced some of their other functions, giving some kind of benefit other than increased vitamin D production.
We know of an example in fish that illustrates the same principle. In humans, OCA2 (for oculocutaneous albinism II) is a gene involved in the melanin pathway—if you have two broken copies, you're an albino. It also affects eye color: A particular variant that has increased rapidly in Europe is the main cause of blue eyes. Species of fish trapped in caves—this all relates,trust us—lose their eyesight and become albinos over many generations. But researchers have found that OCA2 is mutated in a number of different species of blind cave fish in Mexico, all descended from tetras. The mutations causing albinism in these fish are different from each other and originated independently. Since we see changes in OCA2 in each case, however, there must have been some advantage in knocking out OCA2, at least in that underground environment. The advantage cannot lie in increased UV absorption, since there's no sunlight in those caves.4
There are hints that knocking out OCA2, or at least reducing its activity, may be advantageous (probably in some way unconnected with vitamin D) in humans who can get away with it. We see a pattern that suggests that having one inactive copy of OCA2 is somehow favored even in some quite sunny regions. In southern Africa, a knocked-out version of OCA2 is fairly common: The gene frequency is over 1 percent.5 Individuals with two copies are albinos and have problems such as skin cancer and blindness as well as social rejection and persecution. Yet it's the most common genetic disease in southern Africa, with the great majority of cases caused by the same mutation. There's a similar story among Amerindians in the American Southwest: A form of OCA2 albinism is common among the Navajo and other neighboring tribes, with gene frequencies as high as 4.5 percent.6 The same pattern appears in southern Mexico, eastern Panama, and southern Brazil. All of which suggests that heterozygotes—that is, those carrying one copy of the broken version of OCA2—may have some advantage.
Something else that makes us wonder whether vitamin D was the key factor behind recent changes in skin color is the fact, mentioned before, that the genetic changes underlyinglight skin in Europe and East Asia are almost entirely different. If a reduced-function version of a gene involved in melanin synthesis was strongly favored in Europe, why wouldn't a similar reduced-function version of that same gene arise and spread in China? Mutations that reduce function are quite common. In addition, selection on genes affecting skin color, eye color, and hair color somehow created lots of variety in Europeans: redheads and blondes, blue eyes and green eyes. Nowhere else in the world is that sort of variety common. In most parts of the world, even in temperate regions, everyone has dark eyes and dark hair. To us these facts suggest that there was something fundamentally different in the selective forces affecting skin color in Europe and East Asia. If those forces were different, at least one of them was probably selecting for something other than vitamin D.
DEM BONES
The skeletal record clearly supports the idea that there has been rapid evolutionary change in humans over the past 10,000 years. The human skeleton has become more gracile—more lightly built—though more so in some populations than others. Our jaws have shrunk, our long bones have become lighter, and brow ridges have disappeared in most populations (with the notable exception of Australian aborigines, who have also changed, but not as much; they still have brow ridges, and their skulls are about twice as thick as those of other peoples.)7 Skull volume has decreased, apparently in all populations: In Europeans, volume is down about 10 percent from the high point about 20,000 years ago. These changes were spread out over time, of course.
For example, if you look at Bronze Age skeletons from Europe (around 3,000 years ago), you find that some people still had brow ridges like those of Australian Aborigines. Hardly any Europeans have brow ridges today.
Genome-selection surveys may have found some of the alleles affecting these processes. One group of researchers discussed two genes involved in bone growth that showed signs of selection in Europeans, another in the same gene family that showed selection among East Asians, and a fourth that showed signs of selection in both populations.8 Even though we see similar skeletal changes in many populations, the genetic underpinnings are generally different, much like the pattern underlying skin color.
Some changes can be seen even over the past 1,000 years. English researchers recently compared skulls from people who died in the Black Death («650 years ago), from the crew of the Mary Rose, a ship that sank in Tudor times («450 years ago), and from our contemporaries. The shape of the skull changed noticeably over that brief period—which is particularly interesting because we know there has been no massive population replacement in England over the past 700 years. The height of the cranial vault of our contemporaries was about 15 percent larger than that of the earlier populations, and the part of the skull containing the frontal lobes was thus larger.9
CHEATERS PROSPER
Usually a new version of a gene increases in frequency because it aids the bearer in some way—although it may not aid the species as a whole. Some alleles take this a step farther andsucceed by helping themselves, not the bearer. They're called "driving genes."
Everyone has two copies of all chromosomes other than the sex chromosomes, and so everyone has two copies of each gene on those autosomal chromosomes. In the process of meiosis (the process that forms germ cells), a diploid germ cell replicates its DNA and then divides twice, forming four haploid cells e
ach having a complete set of autosomal chromosomes and a single sex chromosome. In sperm production, all four haploid cells become gametes, while in females only one of the haploid cells becomes an egg.
Generally, each of the two copies of each gene has an equal chance of showing up in a gamete. The system is designed (in an evolutionary sense) to give alleles a fair shake. But sometimes, mutation creates a new allele that has a better chance of getting into a gamete—greater than 50 percent. Think of it as a line-cutter. Along the same line, a mutation might increase a gamete's chance of success—say, by making sperm swim faster—and this might be the case for SPAG6 (for sperm associated antigen 6), a gene involved in sperm motility that has apparently undergone a sweep in Europeans over the past few thousand years.10
Driving versions of genes must have come into existence more often as human population increased, just as lightning strikes more Texans than Kansans. In a small population, a driving allele would occasionally come into existence and rapidly go to fixation, but the population might spend most of the time in between such sweeps. The much larger populations associated with behavioral modernity, in particular with agriculture, should have generated driving alleles at a rate perhaps two orders ofmagnitude higher than the populations of the Old Stone Age. Those new driving alleles would not have taken all that long to spread, since they would have grown exponentially in a well- mixed population. Therefore, modern humans should have an unusually large number of driving genes, either recently fixed or on their way to fixation. By the same argument, any species whose numbers have recently soared—for example, after domestication—is likely to also have an unusually large number of driving alleles.
Recent work may have identified such driving alleles. One study11 found evidence of selective sweeps in a number of cen- tromeric regions—the centromere being a central region that holds together the two halves of the chromosome and that plays a key role in meiosis and in ordinary cell division (mitosis). Since these regions have relatively few active genes, the centromeres themselves may be the target of selection. There is reason to believe that centromere mutations can affect the way in which alleles end up in the egg or in polar bodies, which are dead-end by-products of the egg's division. Any allele with an increased chance of ending up in the egg, instead of a polar body, would have an advantage, possibly a large one. The researchers found evidence of sweeps in eight out of seventeen chromosomes for which data were available. Those sweeps were regional, mainly in the European and Asian samples, which suggests that they came after the expansion out of Africa.
In the long run, a large population would develop more defenses against driving genes as well as more driving genes. But in the short run, just after a dramatic population expansion, driving genes might be both unusually numerous and unusually troublesome, since selective pressures favoring defenses andmodifiers only come into existence after the driving alleles become common. This isn't just a theoretical concern—it might have something to do with the mysteriously high miscarriage rate in humans. The fraction of conceptions that lead to a healthy baby may be as low as 25 percent, far lower than in most other mammals, and it seems that most of the miscarriages are caused by chromosomal anomalies. An unusually large number of driving genes may play some causal role in this. Too many alleles trying to shoulder their way into the egg at the same time could lead to trouble, just like Stooges all trying to get through the door at the same time.
CHANGING MINDS
The most interesting kind of genetic changes are those that affect human personality and cognition, and the evidence is good that such changes have indeed occurred.
A number of the new, rapidly spreading alleles found in the recent selection surveys have to do with the central nervous system. There are new versions of neurotransmitter receptors and transporters—neurotransmitters being molecules that relay and influence signals between nerve cells. Several of the new alleles have effects on serotonin, a neurotransmitter involved in the regulation of mood and emotion. Many recreational and therapeutic drugs (particularly antidepressants) modulate serotonin metabolism. And there are new versions of genes that play a role in brain development: genes that affect axon growth, synapse formation, formation of the layers of the cerebral cortex, and overall brain growth. Again, most of these new variants are regional: Human evolution is madly galloping off in all directions.
We see new versions of several genes in factors having to do with muscle fibers and brain function. Dystrophin is a protein (coded by the longest of all known human genes) that has an important structural role in muscle fibers and the brain; the dystrophin complex is a set of proteins that are physically associated with dystrophin. Major defects in the dystrophin gene itself cause Duchenne muscular dystrophy, which has very severe effects, while lesser defects cause Becker's muscular dystrophy, which is milder. These are among the most common genetic diseases, apparently because the extremely large and structurally complex dystrophin gene has so many ways of going wrong. Dystrophin's dual role has medical consequences, in that boys with Duchenne muscular dystrophy suffer reduced IQ as well as muscular weakness.
The dystrophin-associated sweeping alleles that we see in the selection surveys (which do not cause disease) raise the interesting possibility of direct trade-offs between muscle and brain function in the recent past. We have reason to think that humans circa 100,000 BC had stronger muscles than today— and so changes in the dystrophin complex may have sacrificed muscle strength for higher intelligence.
Another very intriguing pattern involves new versions of genes that affect the inner ear.12 We wonder if this is a consequence of recent increases in language complexity sufficiently recent that our ears (and presumably our brains, throats, and tongues) are still adapting to those changes. Or, since some of the sweeping genes involving the inner ear are regional and recent, could some populations be adapting to characteristics of particular languages or language families? It seems that all humans can learn any human language, but we don't know whether everyone is inherently just as good as everyone else atlearning every language, communicating in every language, or eavesdropping in every language.
More generally, these sweeping neurological genes could be responses to the new challenges posed by agriculture itself and the dense hierarchical societies it made possible. In the following sections, we discuss those challenges and likely adaptive responses to them.
THE MALTHUSIAN TRAP
In An Essay on the Principle of Population, Thomas Malthus in 1798 observed that population tends to outrun food supply, since population increases geometrically while food supply increases arithmetically. He wrote:
The power of population is so superior to the power of the earth to produce subsistence for man, that premature death
must in some shape or other visit the human race. The vices of mankind are active and able ministers of depopulation. They are the precursors in the great army of destruction, and often finish the dreadful work themselves. But should they fail in this war of extermination, sickly seasons, epidemics, pestilence, and plague advance in terrific array, and sweep off their thousands and tens of thousands. Should success be still incomplete, gigantic inevitable famine stalks in the rear, and with one mighty blow levels the population with the food of the world.
Imagine that a population of farmers is doing well: They have plenty to eat. It's easy for them to raise more than twochildren per family—they do so, and the population increases. It continues to increase as long as conditions remain the same. More people need more food, but then there are more workers producing food. As long as per capita production stays the same, the standard of living does not change, even as population increases. However, eventually this expanding population runs out of land, and farmers in the next generation have to farm smaller plots. They may be able to keep per capita production the same by working harder, but in the next generation plots become even smaller. If the methods of food production remain the same, eventually per capita production must decrease as po
pulation increases and per capita resources decrease. That decrease will continue until the average farmer produces just enough food to raise two children, at which point population growth stops.
Suppose that farming methods improve, so that productivity per acre goes up by a factor of ten. The population begins to grow—let's say fairly slowly, with each family managing to raise 2.5 children (on average) to adulthood. The population is growing 25 percent per generation. In ten generations—about 250 years—the population has caught up with those improved methods. Living standards are low again, and population growth stops. But 2.5 children per family is by no means an especially high rate of population growth: In colonial America, the average family raised more than 7 children to adulthood. At that rate, population growth could catch up with a tenfold increase in productivity in just two generations.
The point is that even moderate rates of population growth can rapidly catch up with all plausible improvements in food production. Thus, populations should spend most of the time near a Malthusian limit, and there should be no lasting improvementin the standard of living. Malthus himself pointed out that factors other than food shortages can also limit population. Any negative factor that intensifies as population density increases can be the limiting factor—starvation and malnutrition are not the only possibilities. The key is which negative factor shows up at the lowest population density. We believe that the nature of the key limiting factor—which is not necessarily the same in all human populations—can have important effects on human evolution, including the recent changes we have been discussing.
The 10,000 Year Explosion Page 9