There’s something about sex that seems to inspire whimsy. The scientific literature on how it all began is spiced with clever turns of phrase, witty asides, and the occasional risque double entendre. perhaps the strain of accounting for such an absurd way of making babies goes to the heads of even the sober men and women of science.
You’d think there would hardly be cause for such exercise about the original reason for sex. After all, the problem was considered pretty well solved 30 years ago. Sex was said to be good for the species, says Richard Michod, a 41-year-old professor of ecology and evolutionary biology at the University of Arizona in Tucson and one of the key figures in the current debate. By sex, of course, he means the mating of genetic material from two individuals to produce one with a new combination of genes. By ensuring that offspring were slightly different from their parents, sex increased the chances that a species would produce a new, improved model capable of surviving environmental changes or of getting the jump on a rival or predator. It provided genetic variability, so sexual populations evolved faster, and that was that. It was in all the textbooks, says Michod.
In the mid-seventies, however, evolutionary biologists began to question that conventional wisdom. Chief among the worriers was John Maynard Smith, a professor of biology at the University of Sussex in England and, dare one say it, a seminal figure in the field. He was troubled that the standard explanation for sex invoked a slightly dubious mechanism (dubious in his view, anyway) known as group selection.
The classic example of group selection in action is the animal that gives a warning cry to alert the group to a dangerous predator and thereby puts itself at risk. Why does self-sacrificing behavior enter into a discussion of sex? Because no organism in its right mind would opt for sex with another organism as a way to create offspring. It’s too darned expensive, genetically speaking.
Most higher organisms that go in for sexual reproduction package their genes into pairs of chromosomes (we humans have 23 such pairs). But any sexually reproducing organism throws half its genes overboard when it makes sex cells--that is, eggs or sperm--because its sex cells contain only one chromosome from each chromosome pair. (This is called the haploid, or halved, condition; the union of egg and sperm in sexual reproduction restores the diploid, or paired chromosome, condition.) Thrifty asexual organisms, on the other hand, transmit all their genes to the next generation.
This thorny fact of life presents a major problem for evolutionary theory. If natural selection acts on individuals, rewarding those who get the most copies of their genes into the next generation, then sex would seem to make no sense. All organisms ought to opt for efficient cloning, not mad, untidy mating. Cloning--which exists in many single- celled organisms, some plants, and a few insects, fish, and reptiles-- involves no apparently pointless halving of paired chromosomes, no compulsory union with other organisms in order to form a complete whole. Clones simply split in two or bud to generate identical copies of themselves, or produce self-sufficient diploid eggs that don’t require fertilization by sperm.
Furthermore, many organisms invest enormous amounts of time in pursuing mates, time that they might better spend on eating or on avoiding being eaten. Females end up paying a particularly high price for sexual reproduction; most females gestate and nurture their young, while males contribute nothing but their genes.
The group selection theory that worried Maynard Smith was a way to explain--or explain away--the toll that sex apparently exacted on the individual. According to the rules of natural selection, sex must be good because there’s so much of it around. (Current estimates are that 99.9 percent of higher organisms are sexual.) Unfortunately, it looks very bad for the individual, so it must be the group that it’s good for. Sex, the argument went, produces genetic variation by combining the genes of two individuals; very well, then, variation must be advantageous to the species.
Thus sex evolved to produce variation as a sort of group benefit plan, allowing groups that reproduced sexually to keep up with changes in their environment. It wasn’t a pretty explanation, but it got the job done.
There things rested until the publication in 1978 of Maynard Smith’s The Evolution of Sex, in which he wrote of his distaste for the Panglossian belief that if some characteristic can be seen as benefiting the species, then all is explained. The Evolution of Sex is an unusual scientific book, elegantly written and steeped in perplexity. Phrases like It is hard to see how this could be so, There is an obvious difficulty in arguing, and the koan-like What is it that goes extinct? appear on almost every page. Yet nearly to a person, young researchers in the field recall being inspired by its puzzled lucidity. An important problem--maybe even the problem--in evolutionary biology had been put up for grabs again. Why bother with sex?
What’s more, the near-universality of sex--it occurs in microorganisms, plants, and animals high and low--makes it a scientist’s delight. It can be studied nearly everywhere, from petri dishes to field stations to supercomputer labs, by molecular biologists, botanists, ecologists, and mathematicians. As a result, there has been a virtual gold rush of young scientists eager to tackle what has been called (lyrically, of course) the queen of problems in evolutionary biology: Why did sex evolve, and what keeps it going?
There’s a comedy routine beloved of Mel Brooks fans in which Brooks assumes his funniest persona, the 2,000-Year-Old Man. This ancient has been alive since the birth of Christ, and he’s more than willing to editorialize on what he’s seen in his long life. Asked to name the greatest scientific advance in his two millennia, Brooks promptly replies, Saran Wrap. Saran Wrap? says startled straight man Carl Reiner. What about the conquest of space? Oh, says Brooks, generously, that was a good thing too.
Nowadays you could almost summarize the most influential arguments to explain the origin of sex as, Oh, variability, that was a good thing, too. These arguments hark back to ancestral forms of proto- sex--the mixing of genetic material from two different organisms that paved the way for sexual reproduction proper. Researchers working at the molecular level regard these transactions as anything from a strategy for DNA repair to an accident, the consequence of a parasitic bit of DNA becoming especially persistent about getting itself copied. On one point, however, the newer views agree. All the genetic variability that sex affords is good, so good that it may well be what kept sex going once it got started, but it wasn’t why sex evolved in the first place.
One of the arguments currently dominating the competition is championed by Michael Rose, a mathematical geneticist at the University of California at Irvine, and his colleague Donal Hickey of the University of Ottawa. Along with many others in their field, they believe that a bacterial phenomenon known as conjugation constitutes an ancient form of proto-sex. Conjugation is a property of some but not all bacteria in a given colony. It involves the extension of a projection called a pilus from one bacterium to another, and the journey along that bridge of a self- contained, parasitic loop of genetic material called a plasmid. (There is a certain morphological similarity between conjugation and higher sex, notes Michod delicately.)
The bacteria seem to gain nothing from this transaction. In fact, if this is proto-sex, it’s proto-bad-sex, because neither bacterium can be described as consenting. The plasmid contains the quintessential selfish gene, a bit of DNA whose only mission is to reproduce itself, thus driving the plasmid to distribute as many copies of itself to as many hosts as possible. In the process, bits of the original bacterium’s genome occasionally cling to the plasmid like foxtails on a dog’s coat and find themselves in the new host. Eventually, explains Rose, some hosts begin to use and benefit from the inadvertent gift of another individual’s DNA.
Rose and Hickey have gone on to propose that selfish DNA could account for a primitive form of sex that’s closer to sex as we now know it. In some early single-celled organisms, they theorize, selfish DNA didn’t merely cause a bridge to form so that it could travel from one individual to another--it impelled the two organisms to actually fuse, in a primitive anticipation of what sperm and egg do during fertilization. This parasitic DNA could then spread contagiously until the whole population was committed to sex.
How widely accepted is this scenario? There are three stages in the life cycle of any scientific idea, says the 36-year-old Rose. First, it’s treated as a joke. Next, it’s taken seriously but considered to be impossible. Finally, people admit that it’s possible, but they insist that it’s trivial. Rose says he came on the scene in 1983, during the second stage, performing the mathematics to demonstrate that selfish DNA is a powerful evolutionary force. Now, he deadpans, we’re in the third.
Michod is thoughtful when asked to comment on the Hickey-Rose theory of gene transfer. It is certainly a reasonable explanation for the origin of sex, he says. In fact, I think it’s the main competition to the DNA repair view.
Michod’s idea that the reshuffling of genes from two organisms originated as a mechanism to mend damaged chromosomes is another of the theories in current contention. Influenced by Maynard Smith, Michod refused to buy the argument that genetic variation was enough justification for sex.
Look at us, he says, adult organisms who have already passed muster, evolutionarily speaking. We’ve survived, so our genomes must be in reasonable shape. But what is the most striking effect of sexual reproduction? It scrambles up that perfectly good genome. What are the odds that that will be an improvement? And even if it is, then what? You can produce a superkid, but she’ll just reproduce and scramble the genome even more. Everything sex does, it partly undoes in the next generation.
Michod reasoned that since DNA is a way of conveying information, perhaps sex was initially a way of getting the message straight: it might be about error correction, not variation. In 1988 he and his team demonstrated sex-for-DNA-repair in a bacterium called Bacillus subtilis. These microbes engage in an activity called transformation, which involves incorporating bits of DNA floating in their environment. (Not to be too lurid about it, but this DNA originates from the disintegrating corpses of neighboring B. subtilis.) Michod believes that they use this spare DNA to repair breaks in their own chromosomes caused by exposure to environmental insults, such as excessive oxygen or ultraviolet light. The evidence? Damaged bacteria use more DNA than undamaged bacteria, and repaired bacteria replicate more successfully than unrepaired bacteria. (Comments Rose, Sex with dead bacteria is apparently better than no sex at all.)
None of this means that either Rose or Michod underestimates the significance of variation. Look, says Michod, diversity is the fuel of evolution, and gene recombination produces diversity. We’re just saying that recombination--proto-sex if you will--didn’t come into existence to produce variation. Variation, in other words, is an effect of sex, one that’s turned out to be extraordinarily useful, but it’s not the original reason for sex. There must be some short-term, individual benefit to recombination, says Michod, and in his view, it’s DNA repair.
Rosemary Redfield agrees about the short-term benefit, but she has her own ideas on what it might be. Redfield, 43, a biochemist at the University of British Columbia in Vancouver, dubs her take on bacterial proto-sex Having your cake and eating it, too. (I’m rather pleased with that description, she says.) Accepting her idea, which she plans to lay out for her colleagues next month at a conference on the evolution of sex, doesn’t mean rejecting Michod’s, she adds; the two can coexist quite nicely.
Redfield agrees with Michod’s observation that bacteria go to a great deal of trouble to incorporate external DNA, but she notes that the patching they achieve is hit-and-miss, as likely to be bad as good. What else, then, might motivate transformation, this DNA absorption that seems to be the harbinger of sex? In Redfield’s opinion it’s that other great physiological drive: hunger.
The spine of the DNA molecule is made up of alternating sugars and phosphates, she explains, with a chemical base hanging off each sugar. When DNA is broken down, it’s really sugars and a base, she says. I think of it as molecular candy, rather like the candy on a string we used to eat as children. When a bacterium feels hungry--runs out of its usual sugar supplies--it becomes capable of taking up external DNA. Through a poorly understood mechanism--Though it must be something like slurping spaghetti, Redfield says--it sucks a string of DNA though a pore in its wall and sets about digesting it.
This explains only half of Redfield’s catchy aphorism. The bacterium can eat its DNA confection; what about having it? DNA, recall, consists of a twist of two complementary strands. When a bacterium goes to work on a DNA fragment, it degrades one strand for the sugars, leaving the other one floating free. The second strand may subsequently be digested also. But if it matches a stretch of the bacterium’s own DNA (especially a damaged bit), it knocks out that bit and replaces it. The discarded DNA can then be digested too. Redfield notes that while individual steps in this scenario have been observed, she has yet to prove the whole story. But she feels confident that the Haemophilus influenzae colonies in her petri dishes are going to confirm the benefits of proto-sex: a fill-up and, with luck, a tune-up.
And that, one might think, is that. Having constructed a couple of reasonable scenarios for the origins of sex, having set the recombination process humming in a modest way, having tackled the question of how sex might have arisen to the extent that questions about distant origins can ever be answered, why don’t evolutionary biologists put sex behind them and get on with life?
Because the sex riddle is still only partially solved. Evolutionary biologists may have some idea of what made sex possible in the first place. But just because something is possible doesn’t guarantee it will catch on. Why did so many organisms stick with sex after trying it? Why didn’t they revert to cloning? This is where some researchers think variability really comes into play: it’s what made sex such an enduring success.
On paper, clones look unbeatable. A clone wastes no time looking for a mate, runs no risk of scrambling a perfectly serviceable genotype, and can put more copies of itself into the next generation than a sexual organism. Yet while clones abound in the lowest ranks of life, they constitute a fairly exclusive club in the higher animal kingdom. True, some earthworms, spiders, and the water flea Daphnia can switch from sexual to clonal reproduction depending on environmental conditions. Certain all- female species of whiptail lizard have opted for total parthenogenesis-- they have eggs that develop without fertilization--but they are nearly unique among terrestrial vertebrates. And quite a number of fish and amphibians come in sexual and asexual versions. But in all these cases the asexuals almost certainly had sexual ancestors, making them evolutionary backsliders. As a rule, the higher you go on the evolutionary ladder, the less likely a group is to have a clonal variant.
That’s a puzzle: Why don’t more organisms backslide and revert to cloning? Are we sexuals simply in a rut, so deeply invested in the visible and invisible machinery of sex that we can no longer throw it away? Actually, recent research suggests this may be at least part of the explanation for the persistence of sex. Mouse embryos in which both copies of a chromosome were engineered to come from one parent--rather than the standard, sexually obtained complement of one maternal and one paternal copy--died early in development. Somehow the sex of the parent does leave an indelible and absolutely necessary imprint on offspring. Thus males have become, once and for all, indispensable.
Of course, blinkered as we humans are by our own sexuality, we don’t usually spare much thought for clones, let alone wonder why there aren’t more of them. (Perhaps we’re half afraid to. In the popular imagination--or is it the collective unconscious?--clones are the stuff of sci-fi nightmare. Who can forget the evil pod people overrunning San Francisco in The Invasion of the Body Snatchers?) But to a student of sex evolution, clones are a godsend. Commented one researcher, To learn about health, you study disease. To learn about sex, you study clones.
In the first round of studying clones, researchers tried to discover if there were any consistencies in the habitat preferences of these organisms that had gotten off the sexual merry-go-round. The results of their experiments were, to put it mildly, surprising.
Earlier theorists had assumed that sex was advantageous in the long run because it produced variability in gross features like size and shape, thus equipping species to adapt and roll with the inevitable environmental punches. If that were the case, then sexual organisms ought to turn up in harsh areas on the frontiers of an organism’s habitat, and clones ought to live only in cushy environments. In fact, nearly the opposite is true: clones tend to predominate in frontier settings, while sexual organisms fill the niches in environmentally stable zones.
It appears that in difficult habitats, or on the fringes of a range where populations are low and finding a mate may not be a trivial effort, clones do well. For one thing, they can reproduce quickly, which makes them excellent colonizers: they’re homesteaders who don’t need to wait for mail-order brides. But clones evidently falter in the dense jostling of stable ecosystems, which teem with life and where competition between and within species is fierce. Studying the juxtaposition of sexual and asexual variants of the same species thus offers scientists a rare opportunity to study what made sex, once it got going, such a hit; and some of the most exciting new work does just this.
Atop a brutally modern building on a nondescript college campus on the bleak high plains of southeastern Washington State, Steven Kelley has his greenhouse. Outside, the gray winter afternoon is as cold as iron; inside the greenhouse, it’s May in a meadow. Under the orange light of tungsten lamps, Anthoxanthum odoratum is releasing its perfume. Kelley’s favorite plant doesn’t look like much, just your standard untidy grass. But it smells like heaven, and it’s helping him test whether there is a quantifiable advantage to sex. Not bad for a weed.
Kelley is a 35-year-old botanist at Washington State University, and like others of his generation, he too was attracted to the question of sex by Maynard Smith’s deft exposure of its paradoxes. His initial study, published four years ago, had a seductively tidy result. Kelley raised hundreds of pots of sweet vernal grass in a greenhouse. These young plants were of two types: clones propagated from slips, and sexuals grown from seed. He then divided and transplanted nearly 4,000 of these young plants into a field, arranging them around their parents in a pattern like the one that occurs when the wind scatters A. odoratum’s seeds. The clones and the sexual offspring were competing directly, so any differences between them would be highlighted.
After more than two years Kelley measured the number of flower stems on each plant. (What passes for a flower on A. odoratum is predictably modest: two tiny anthers and a small, feathery stigma.) He found that the sexually propagated plants showed nearly 1.5 times the inflorescences of the clones. Since the number of flower stems is a measure of future reproductive volume--flowers produce seeds--those numbers translate directly into what evolutionary biologists refer to as fitness. A 1.5 fitness advantage is a whopping edge in evolutionary terms. Perhaps it’s not enough to offset the theoretical cost of sex--of throwing half your genes overboard--but it’s an extremely healthy start. Moreover, other work suggests that the longer the plants live, the further the sexuals would outdistance the clones, because clones tend to die off at younger ages. It was kind of a textbook result, recalls Kelley. There was a substantial short-term benefit for the sexual plants. It stood to reason there must be an advantage of something like this magnitude, but it was really exciting to see it.
Provocative as it is, this experiment only documents that sex offers an edge. It doesn’t show us why it offers this edge, though Kelley has his suspicions. Pathogens--viruses, bacteria, and parasites--and the constant need for organisms to defend themselves against these infectious invaders may be the unseen force that makes sex such a success. In all likelihood, thinks Kelley, his sexual grasses flourish because they are less likely to be attacked by microorganisms that cause disease. Pathogens are ubiquitous and sex is ubiquitous, observes Kelley. It seems reasonable to think they might be connected.
Kelley’s hunch puts him into what is rapidly becoming the mainstream. Many researchers are coming around to the idea that all living creatures are trying to outrun pathogens just to stay in the evolutionary race. Pathogens, though tiny, have the advantage of speed and numbers: they can reproduce (usually asexually) in seconds and mutate many times while their hosts are held to slower reproductive timetables. The genetic variability afforded by sex gives us hosts at least a fighting chance against our various nemeses, and the little breathing room it provides is what makes sex worth the trouble. That idea stems from what has been dubbed the Red Queen hypothesis, after the irascible royal in Lewis Carroll’s Through the Looking Glass: Now, here, you see, says the Queen to Alice, it takes all the running you can do, to keep in the same place. If you want to get somewhere else, you must run at least twice as fast as that.
It’s nearly impossible to prove or disprove an evolutionary hypothesis like the Red Queen; the best that scientists can do is ask how organisms would behave if the Red Queen were running the show, and then see if organisms actually behave that way.
Let’s suppose that organisms opted for sex to confuse pathogens by creating groups of genetically distinct individuals, each with its own uniquely challenging immune defenses. As the flip side of that idea, all sorts of enemies--viruses, bacteria, parasites--should be pretty adroit at infecting populations with whom they’ve coevolved and whose defenses they’ve had a chance to study, though they should be stymied by organisms they’ve never met. In other words, the bad guys should exhibit a sort of home-field advantage.
One researcher who is enthusiastically putting such ideas to the test is Curtis Lively, a 38-year-old population biologist at Indiana University in Bloomington. One of his first crucial experiments involved an old tussle between two species living in the lakes of New Zealand. The protagonists were an aquatic snail, Potamopyrgus antipodarum, and its parasitic nemesis, the rather unkindly named worm Microphallus. If the coevolutionary arms race between pathogen and host predicated by the Red Queen was occurring, Lively reasoned, you’d expect to see a genetic basis for parasite susceptibility in these snails. (This is something the legendary evolutionary thinker J.B.S. Haldane had mused about as long ago as 1949. Recent work on mice strongly suggests that it is true, and that mice have an intriguing way to counteract and confer optimal immune defenses to their offspring. See box on page 38.)
Accordingly, Lively drew snails and parasites from each of two lakes separated by some 10,000 feet of New Zealand alp--a barrier high enough to discourage even the most determined Microphallus. One lake had smooth-shelled snails, while the other had spiky ones, making it possible to put both snail types into the same containers and readily tell them apart. Lively then infected some containers with worms from one lake and some with worms from the other.
Just as the theory predicted, the parasitic worms showed a home- field advantage. They handily infected the snails with which they had coevolved and whose defenses they knew, but they were stumped by snails from the alien lake with an unfamiliar genotype. That suggests there could indeed be a pathogen-induced advantage to producing variable offspring, says Lively. His observation, moreover, meshed with a previous, more general survey of New Zealand snails. In lakes heavily infested with parasites, sexual snails, with their more variable genotypes, greatly outnumbered the clones; whereas in lakes with fewer parasites, clones formed a larger part of the population. Apparently, the more Microphallus there were nipping at the snails’ heels, the greater the incentive for sex.
Lively has since participated in a study that clearly demonstrates the advantage of sex in a small Mexican minnow, Poeciliopsis monacha, which has both sexual and asexual variants. In rock pools where both coexist, the study found, parasitic worms made a beeline for the most common genotype--all those of the familiar-looking clones. The sexual fish outnumbered the clones by four to one.
If there were no parasites in the world, says Lively, a clone could take over in many species. The world being the pest-ridden place it is, sexuals predominate. By mixing up genes, sex produces variable offspring with rare genotypes that can stay one step ahead of their enemies. In contrast, clones are easy pickings for a parasite. (Seen one clone, seen ’em all.) Clones, observes Lively philosophically, pay a high price for being so uniform.
Every once in a while, though, clones have their day in the sun. Robert Vrijenhoek, the leader of the minnow study, can attest to it. Vrijenhoek, 46, is an evolutionary biologist at the Center for Theoretical and Applied Genetics at Rutgers University in New Jersey; he has been working with fish for 20 years, both in a Rutgers lab awash with aquariums and in the rugged highlands of Sonora, Mexico. In that time he has amassed plenty of data to affirm that sex is here to stay. However, one ecological drama that unfolded before his eyes in the rock pools of the upper Sonora cast clones as real, if momentary, contenders for evolutionary success (and cast Vrijenhoek, temporarily, as a gonzo biologist). The players in this drama were, once again, Poeciliopsis and an asexual variant, and a parasitic trematode worm.
In 1978, during one of the periodically recurring droughts that characterize the Sonora region, one of the pools dried up. When the drought eased and streams flowed, the pool was recolonized by a few clones and sexuals that had swum upstream from another pool. But because the population was so tiny, variation among the few surviving sexuals was quite low. In this isolated pool, with its mix of clones and sexuals, the clones could play their strong suit of rapid population growth, and the genetically depauperate sexuals couldn’t give them any real competition. The clones invaded with a vengeance: within a year they constituted 95 percent of the total minnow population. And in a further reversal of fortune, it was the inbred sexual fish that were getting clobbered by parasitic worms.
Then, in 1983, the rancher through whose land the stream ran decided to dam the stream to provide water for his stock. Worse yet, he planned to farm fish and introduce Tilapia, the omnivorous Nile perch that invariably spells doom for native species. I felt sorry for those minnows, recalls Vrijenhoek in a tone both defiant and conspiratorial. But since the ecosystem was going to be destroyed anyway, I felt I could put a bunch of downstream sexual female minnows in the pool and see what happened.
What happened was that Vrijenhoek went ahead with his plan to put the variation back into sex by introducing new, alluring females. But someone else, perhaps the goddess of native species, was looking out for Poeciliopsis. While building the dam, the property owner had a serious encounter with a machete. During his lengthy recuperation he had a change of heart, deciding he was too old for dam building and fish farming, and the plan was scrapped.
But the genetic cavalry had already arrived, courtesy of Vrijenhoek. The result was dramatic. Within two years, he says with satisfaction, the sexual fish were up to eighty percent of the population again. And once again it was the clones that were getting decimated by parasites. The one-shot infusion of new genes was all it took for sexual reproduction to reassert its superiority over cloning.
The importance of variability was confirmed, with a remarkably swift payoff for the sexual fish. But one can’t help feeling a little sorry for the clones. Sure, worms would probably have gotten them in the end. The weight of the evidence suggests strongly that parasites have a field day with clones, and that sex helps organisms evade their micro-pursuers. It does seem a shame, though, that the clones had so little time to demonstrate their opposition to what 99.9 percent of higher species and 100 percent of biologists accept as a given: sex may be time consuming and inefficient, but it’s become the way to go.
Led by the Nose
Female Mus musculus domesticus seeks male w/right stuff for intense, short-term relationship. Object--superior babies. Your place okay?
If rodents had personal columns, they might run ads like that. Evidently, female mice seek mates whose genetically coded defenses against disease--called MHC, or major histocompatibility complex--will best complement their own. The result of their behavior? Avoidance of inbreeding and production of more genetically diverse, disease-resistant offspring. Females try to slip away from closely related males on their home territory for an assignation with Mr. Right--a more alluring stranger whom they appear to recognize by a signature scent in his urine that broadcasts his MHC complex.
A research team led by geneticist Wayne K. Potts of the University of Florida at Gainesville discovered the phenomenon while seeking to explain the variability among genes coding for MHC. (Most genes have only one allele, or alternative form, but MHC genes may have over 100.)
For the study, which was published in 1991, Potts and co-workers chose semiwild house mice, earmarked them, sampled their DNA, and put them in enclosures large enough for the males to establish their normal mating territories. Then they let the mice get on with it. When the researchers checked the DNA of the resultant offspring, they found that half of the litters whose parentage was well established involved fathers who were not the mothers’ turf mates. Moreover, they established the payoff for the roving females. By following their noses and mating with males whose MHC profiles were very unlike their own, the females produced litters of mice with a much wider range of MHC alleles than if they had mated with their territorial males. According to conventional wisdom, having a wider range of MHC alleles increases a litter’s chance of resisting parasites and disease. Perhaps, to paraphrase Pascal, The genes have reasons that reason does not know.