books
By faith and honour,
Our madams mock at us, and plainly say Our mettle is bred out; and they will give Their bodies to the lust of English youth To new-store France with bastard warriors.
Shakespeare, King Henry V, III, v, 27-31
The idea that females have anything at all to gain by copulating with more than one male has had a rocky history. With hind- sight, the way sperm competition as a discipline has developed, particularly with respect to the female, seems to have been perverse. Almost from the beginning, feminists interested in sexual selection ranted and gnashed their teeth in frustration at
1
its androcentric bias, but until recently to no avail. However,
few areas of science flow logically from step to step. We were wrong-footed initially by Darwin, who, as we saw in chapter 1, assumed females to be monogamous. But even a hundred years later when Bob Trivers gave sexual selection an intellectual facelift - which included the recognition that sperm competition
- 2-
might be important females still got short shrift. There is a
certain amount of both irony and predictability in this. Trivers’s brilliant insights revolutionized our ideas about sexual selection and helped to spawn the new field of behavioural ecology. But following so hard in Darwin’s footprints, it is perhaps not too surprising that he should have retained a male-orientated view - after all, almost everyone else had done so. None the less, Trivers’s ideas were so influential that for over a decade they perpetuated the male bias in sexual selection and in sperm competition in particular.
As we have seen, a basic assumption throughout much of the brief history of sperm competition has been that males have lots
Number of copulation partners Number of copulation partners
The full results from Bateman’s experiments on Drosophila. The upper half of the figure is the same as figure 1. The lower half of the figure shows the results of other experiments by Bateman (1948). The upper figures show that male repro- ductive success increases with the number of copulation partners, but female reproductive success does not. The lower figures, derived from experiments in which food was limiting, show that female reproductive success also increased with the number of copulation partners. This result probably arose because sperm production and ejaculate size were smaller when food was limiting and females copulated repeatedly in order to obtain sufficient sperm to fertilize their eggs (from Arnold and Duvall (1994)).
to gain from copulating with multiple partners, but there is little or no benefit for females behaving similarly. This assumption arose from Trivers’s interpretation of Bateman’s earlier experi- ments on fruitflies: rampant males and acquiescent females. However, until recently it was not widely appreciated that some of Bateman’s experiments did show that females benefited from
copulating with several males (figure 15). Moreover, Bateman identified what this benefit was: a renewed supply of sperm.3 Accidentally, the conditions in which Bateman maintained his flies in these particular experiments
were poorer (but also probably more natural) than in the others, as reflected by their overall lower reproductive success. But an additional consequence of the poorer conditions was that males probably inseminated fewer sperm in each ejaculate. As a result, females needed to recopulate to replenish their sperm supplies sooner than they would otherwise have done. It is unfortunate that only half of Bateman’s experiments showed this effect, and even more unfortunate that Trivers chose to focus on the other half. 4 The fact that Bateman got these two different sets of results was a consequence of the species of fruitfly he used and the design of
his experiments.
Bateman used the species almost everyone else used at that
time: the black-bellied fruitfly. Male fruitflies are ready to copulate with any female fly they meet. Virgin female fruitflies are also keen to copulate because they need a supply of sperm, but thereafter they are reluctant suitors. In part this is because males have made them reluctant by adding an anti-aphrodisiac to the semen they inseminate into them, but this effect is relatively short-lived. In general, female fruitflies copulate again only when their supply of sperm starts to dwindle. This is the key point. A single copulation will provide a female black- bellied fruitfly with enough sperm to maintain a high level of egg fertility for about four days. After this, because females don’t utilize sperm especially efficiently, some of the eggs a female lays will not be fertilized. Consequently, about four days after their initial copulation females become interested in copulating again. Because Bateman’s experiments ran for only three or four days, relatively few of his females needed to recopulate. Over this period, only those flies reared under the poor conditions needed to do so. Bateman therefore missed the females’ renewed interest in copulating that would have occurred had his experiments gone on for longer.
There are several fruitfly'species where females store so few sperm that they have to recopulate much more frequently than every three or four days to maintain the fertility of their eggs. The most extreme cases involve females recopulating after only a few hours. Had Bateman chosen a species
that typically recopulates more often than every three or four days, Trivers would not have been able to disregard those results which did not fit his preconception about sexually motivated males and reluctant females. However, Trivers was not alone in this view. Coming at sexual selection from a slightly different direction, Geoff Parker arrived at a similar conclusion.5 For males more copulations with more partners meant more offspring, but the best females could hope for from copulating with additional partners was better-quality offspring. In evolutionary terms,
Parker concluded, quantity was more important than quality. An important turning point, at least for biologists studying sperm competition in vertebrates, was a paper by Susan Smith
of Mount Holyoke College, Massachusetts, published in 1988.
In fourteen years of watching black-capped chickadees Smith
6
witnessed just thirteen extra-pair copulations. In nine of these
the incident took place inside the extra-pair male’s territory, strongly suggesting that the female had gone looking for illicit
sex. Not only that, but every single
involved females copulating with males that had been dominant
over their partner during the previous winter. Females therefore
appeared to go up-market for their
Smith’s study was pivotal in alerting us to the possibility that
females might gain something
activities. Further confirmation was provided by a group of Belgian behavioural ecologists studying another socially mono- gamous species, the closely related blue tit. In a paper published in Nature they showed that extra-pair copulations and extra- pair paternity were all down to females. The senior author of this paper, Bart Kempenaers, then at the University of Antwerp, was able to watch female blue tits during the time they were
fertile/ His observations revealed that, far from being reluctant 198
or passive participants in
actually initiated them. These two studies were soon followed by others which also found females to be in control of the number of sexual partners. The myth of the reluctant female had started to crumble.
Once it was recognized that females actively seek copulations from more than one male other issues immediately became important. First, it strongly suggested that females did have something to gain from promiscuity. Second, if promiscuous females left more descendants than those which were not, one could infer that sexual selection on females might be much more important than had previously been thought.
We shall consider these two issues in turn and start by looking at what possible benefits females might obtain from copulating with more than one male. As will become apparent, there has been no shortage of ideas. Although inappropriate in many circumstances, thinking about why human females might have several sexual partners can give us some insights here. There are two broad possibilities. Females may trade sex for resources: money, food, a house, parental care or fertility. Alternatively, or in addition, females may engage in extra-pair
copulations in order to improve the genetic quality of their offspring.
Before looking at these two classes of female benefit, we shall briefly consider (and rapidly dismiss) another possibility: that females are promiscuous simply because males are promiscu- ous. Just as peahens carry traces of the green iridescent plumage that renders peacocks so spectacular, females might carry some of the male’s genetic baggage that controls their copulation behaviour. Put another way, polyandry may be merely a by- product of selection on males to copulate at the drop of a hat. As soon as the suggestion had been made that female copulation frequency might be merely a non-adaptive consequence of selection on males, several biologists responded by identifying
studies that showed that this was not the case. A striking example is provided by chickens subject to artificial selection
extra-pair copulations, females
for either high- or low-copulation frequency. After twenty gen- erations there was a striking change in mating frequency among cockerels, but no change in the propensity for hens to copulate, showing clearly that there was no correlation between male and
8 female copulation frequency.
The types of benefit that females might accrue from copulat- ing with more than one male, which we have just considered for humans, can be divided into those you can see and those you can’t. In behavioural ecology terminology, these are, respec- tively, direct benefits - those a female gets for herself - and indirect or genetic benefits - those she gets for her offspring.
The Direct Benefit of Fertility
A relative of mine is a surgeon. A couple came to see him to discuss and arrange the husband’s vasectomy. They had three children, and they now needed a form of birth control. The man was subsequently booked in, but on opening up his scrotum on one side, he was found to be suffering from a congenital deformity in which the vas deferens is unattached to the testicle. It was the same on the other side. There was obviously no way in which this man could have fathered his children and clearly the woman had copulated with one or more other men in order to become pregnant. Strictly speaking, this example does not involve sperm competition because the woman’s partner was ‘azoospermic’ and produced no sperm. This was not an isolated incident. Midwives tell of pregnant women on the maternity ward who confide in them, saying, ‘I hope it doesn’t look like its real dad.’ More quantitatively, one study followed the fate of seventeen women married to azoospermic men and seeking fertility treatment. No fewer than four (23 per cent) of the women became pregnant before receiving any treatment and subsequently admitted to engaging in extra-marital sex. 9 As the author of this report points out, this apparently high frequency may have been partly a consequence of the National Health
Service’s long waiting lists for women seeking donor insemina- 200
establish whether extra-pair
red-wing blackbirds, researchers
territory-holding males. Despite the inability of these males to produce an ejaculate, their partners had no problem producing fertile eggs - with the help of the intact males next door. 10
If you were to ask anyone on the street what the main benefit from polyandry might be for a female, most would say ‘ferti- lity’. The idea that females copulate with several different males as an insurance in case their partner is infertile is a widely held belief. As a general explanation for polyandry, the evidence for the fertility hypothesis is mixed, however.
Prior to the 1960s and 1970s many entomologists assumed that female insects could get all the sperm they required from a single copulation. Since some female insects, such as ants, were known to copulate just once and yet produce fertile eggs for years, there was abundant circumstantial evidence that made this idea credible. In addition, stuck with Darwin’s female monogamy myth, many entomologists felt that females ought to copulate just once. However, as the study of sperm competi- tion expanded from its beginnings in the 1970s, it became clear that the females of many insects were not monogamous and regularly copulated with several males during their lifetime. Mark Ridley found that the more polyandrous females were, the greater their reproductive success, strongly suggesting that their reproductive output was often limited by a shortage of sperm." If females use repeated copulation to replenish their
sperm supplies this raises other questions. Why don’t males transfer enough sperm to fill a female’s sperm stores? And why don’t females store enough of a male’s sperm to ensure a lifetime’s supply?
A possible explanation for why males fail to inseminate enough sperm to keep a female going for life is that in species in which females copulate with several males it might not be
tion. On the other hand, the result is exactly what one would expect if females were behaving in an adaptive manner.
There is a nice parallel to this story. In an early attempt to
copulations were adaptive in vasectomized a number of
worth it for a male - either because at some time in the future his sperm might be displaced by another male, or because he won’t have enough sperm if he encounters another female. Putting in more than enough may simply not make economic sense for males. The cost of producing sperm is probably not trivial (chapter 3) and selection would rapidly penalize those males who were profligate and favour those who used their sperm prudently. A male’s best option might be to inseminate the number of sperm that maximizes his chances of fertilizing the female’s next batch of eggs, but no more. The solution to the puzzle of how much sperm a male should transfer depends on the likelihood that the female will copulate with another male and the likelihood that a subsequent insemination will displace
or devalue the sperm of the original male.
parable of the prudent male seems plausible, I think it is an unlikely explanation because in many species males inseminate many more sperm than a female can utilize. If this is generally true we are left with the second question: why don’t females store enough sperm?
One answer could be that keeping sperm alive for weeks is costly for females. We know nothing about the energetics of sperm storage and we do not know how the responsibility is divided up between the sperm and the female. Nevertheless, it seems likely that the female incurs some energetic cost from keeping sperm alive inside her. If so, then it may pay females to store only as many sperm as they need in the short term, copulate when they require sperm, use it up on a current batch of eggs and then copulate again for the next batch of eggs. Another possibility is that storing a large amount of sperm may take up space within the female’s body. In some insects the sperm-storage structures are fairly bulky and occupy a substan- tial part of the female’s abdomen. The size of the storage organ will depend on the number of sperm the female needs to store, and this in turn may be determined in part by the size of individual sperm. As we have seen, fruitfly species show more variation in sperm length than any other group of animals. In
the fruitflies that produce
organ, the sperm receptacle, is also much enlarged, but the larger the sperm, the fewer she is able to store and the more frequently she has to copulate. This is well illustrated by two species: the sperm of D. acanthoptera are just 5mm in length and a female stores about 1,400 of them which last her for about three weeks. In contrast, D. hydei has z3mm-long sperm, and females store just zoo; to maintain their fertility females of this species must copulate five or six times every day. 13 The more frequently a female recopulates, the more likely it is that sperm competition will occur, depending, of course, on when remating takes place relative to how many sperm the female has in her sperm store. If a female waits to remate until her sperm stores are empty, then sperm competition will be non-existent. But if, as occurs in nature, she remates as soon as she’s less than half full, sperm competition will be intense.
Although many female insects are polyandrous to ensure that they have enough sperm to fertilize their eggs, fertility insurance does not provide a general explanation for polyandry in other animals. The idea of a fertility benefit has been explored in most detail in birds. When I was collecting information for a book on sperm competition in birds I asked dozens of ornithologists whether they had ever come across any evidence for infertile males. There was just one instance: a kestrel, who when paired to a particular female produced a clutch in two successive seasons, both of which failed to hatch. On its own this doesn’t provide evidence for male infertility. The hatching failure may just as well have been due to the female. However, in the following year the male switched partner and produced two further unhatched clutches, while his original partner success- fully reared chicks with a new male. 14
Several studies of birds, however, have proposed that fertility insurance may be one reason why female birds seek extra-pair copulations. Jon Wetton and David Parkin of Nottingham University looked at parentage in house sparrows and found that broods that included extra-pair offspring also tended to
long sperm the female’s storage
contain at least one unhatched egg. They suggested that this association between illegitimate offspring and unhatched eggs arose because the female’s social partner suffered a shortage of sperm and the female sought another male to fertilize her eggs. This argument has a certain appeal but it does not fully explain the results. If the female really was copulating with another male to ensure the fertility of her eggs, her best strategy would have been to copulate early enough so that none of her eggs failed to be fertilized. In other words, there should have been no link between extra-pair paternity and unhatched eggs. The other problem is that it was assumed that unhatched eggs were infertile. This might not be true. If the embryo dies at a very early stage of development, say in the first one or two days after fertilization, the lack of development makes the egg appear infertile. This can be checked relatively easily, and I did this by collaborating with a Spanish biologist, Jose Veiga at the
Zoology Museum in Madrid, who was also studying house sparrows. You can tell whether an egg is fertile or not by ‘candling’ it - shining a bright light through the shell: a fertile egg reveals the pink mass of the developing embryo. The eggs of most small birds, including sparrows, show signs of embryo development after two days, so we collected eggs that had been incubated for this long but showed no sign of development. We removed the layer of tissue that surrounds the yolk chapter 6) and, using a fluorescent dye to stain the sperm nuclei, we checked for the presence of sperm. To our surprise, about 80 per cent of all eggs had sperm present, suggesting that most unhatched eggs had been fertilized, but that the embryo had died. Obviously, this does not support the idea that female sparrows engage in extra-pair copulations for fertility
pur- the embryos could have died for a number of reasons, including the possibility that the male and female were genetically in- compatible. Genetic incompatibility is well known in humans, as evidenced by the spontaneous abortion of very early embryos when both parents are of the same HLA haplotype; that is,
poses.'^ It does, however, suggest another possibility:
have the same human leukocyte antigen system — part of the human version of the MHC (major histocompatibility complex). Wetton and Parkin’s sparrows might therefore have copulated with other males in an effort to find a genetically
Evidence for male infertility is more convincing in the red- winged blackbird - of vasectomy fame. There is a greater likelihood of males of this species suffering from a temporary shortage of sperm for two reasons. First, males are polygynous and may have as many as twenty females in their harem, several of which may require sperm on the same day. Second, we know that male red-winged blackbirds typically transfer about 12 million sperm during each copulation and that their sperm stores are limited to about iiz million sperm, allowing them about ten copulations per day. Elizabeth Gray of the University of Washington found that the incidence of unhatched eggs in red-winged blackbirds increased with the size a male’s harem,
which certainly suggests a fertility problem.
contrast to the Nottingham sparrow study, that broods with extra-pair offspring had fewer unhatched eggs. Together these two observations provide some of the most convincing circum- stantial evidence that female red-winged blackbirds copulate with other males to minimize the risk of their eggs remaining unfertilized.
A female could potentially gain a fertility benefit from copulating with additional males in one of two ways. In situations where females have no indication of whether a particular male is sterile they could copulate with others to reduce the chances of being without sperm. In a population with rather frequent male sterility, selection would rapidly favour this random form of polyandry. Alternatively, in cir- cumstances where females can judge whether particular males are capable of fertilizing their eggs, they will be selected to avoid the no-hopers. Among socially monogamous birds, females often engage in extra-pair copulations with males that are more attractive than their regular partner and it has
been suggested that they might do this to ensure an adequate supply of sperm. It is possible - indeed, it seems entirely logical
that a male with well-developed secondary sexual traits should also have well-developed testes and abundant, good- quality sperm. However, we have only to think of the occa- sional young male heart-throb declared sterile to realize that good sperm do not automatically follow good looks. Indeed, there is no evidence, in birds at least, that they do. Female zebra finch females are attracted to males with a red beak and a high song rate as both social and extra-pair partners, but in two separate studies there was no hint that attractive males had larger testes or better or more sperm than other males.18
One species where a secondary sexual trait does appear to correlate with sperm attributes is the guppy. The females of these fish choose males on the basis of their colour (preferring males with orange markings), size and the vigour with which they display. Males who displayed more (but interestingly not those with more orange markings) tended to have larger sperm reserves. Guppies have no pair bonds; the only reason for male courtship is copulation (guppies are one of the few fish with internal fertilization). The association between
sperm numbers in guppies may arise simply because males are
prepared to display only when they have an adequate supply of
sperm.
Another explanation for females having several copulation
partners is not so much to do with ensuring sufficient sperm, but with ensuring sufficient good-quality sperm. Although females of many species can store sperm for long periods (see chapter 3), sperm don’t last for ever and it is possible that even though a female may have adequate numbers to start with, the ability of the sperm to produce a viable embryo may deteriorate with time. There is good evidence for this kind zo
of effect in birds.
poultry biologists started to look in detail at just how long a hen could continue to produce fertile eggs, they found that females could occasionally produce fertile eggs three or four weeks after
a single insemination. However,
The relationship between the proportion of eggs with dead embryos and the age of stored sperm in the female domestic fowl reproductive tract. The longer sperm are stored before being used to fertilize the more likely the embryo is to die before hatching (from Lodge et ol. (1971)).
stored the sperm, the greater the likelihood of the embryo dying before hatching (figure 16). The message here for hens is not to rely on stored sperm for too long. Since free-living hens copulate at least once a day, this is unlikely to be a problem.
For many years John Hoogland of the University of Mary-
land has studied prairie dogs, those highly social rodents of the
21
American Midwest. In several prairie dog species, including
Gunnison’s, females usually copulate with several males during their single afternoon of oestrus and their litters are often multiply sired. Hoogland has looked at the advantages female Gunnison’s prairie dogs might gain from having several copula- tion partners. His results are among the most striking of any
Promiscuity in female Gunnison’s prairie dogs. The more partners a female has, the more young she produces (from Hoogtand (1998)).
study so far. Females who copulated with only a single male had a 92 per cent chance of pregnancy and birth, whereas 100 per cent of those with several copulation partners became pregnant and produced a litter. This result strongly suggests that multiple partners serve as fertility insurance. The other, much more impressive result, was that females who copulated with several males produced larger litters (figure 17). Hoogland argues that it is unlikely that these females ovulated more eggs, so this effect must have some other cause. Again, genetic compatibility provides an explanation. Just as with the adders and sand lizards discussed in the last chapter, female prairie dogs with a greater number of partners may have been more likely to acquire sperm compatible with their ova, with more embryos
A detailed study of a North American songbird, the dark- eyed junco, led by Ellen Ketterson of Indiana University, revealed an association between female reproductive success
and the number of copulation partners.
observation and analysis, they were unable to find any reason why females should do better with more copulation partners. They were forced to conclude that their result might simply be an artefact of more productive females attracting more males. If this is true, the number of copulation partners a female has may be irrelevant. For a male, on the other hand, focusing on fecund females might be a very smart move. In his study of prairie dogs Hoogland was able to control for the ‘better female’ effect. It is well known among prairie dog biologists that heavier and older female prairie dogs produce larger litters; but even when the effects of female mass and age were controlled, females with multiple copulation partners still produced more offspring. Hoogland’s study therefore suggests that females benefit from copulating with several males through a direct fertility benefit or a genetic compatibility benefit, or possibly both.
The idea of exchanging gifts for sex is a familiar one. In humans it takes a variety of forms, some subtle, some not so subtle. Among animals, trading sex for material benefits also appears in a number of guises and, in some cases at least, can explain why females copulate more often than they need to in order to fertilize their eggs.
Most females require additional food to enable them to produce eggs or nurture their young. Most males want pater- nity, so they are often happy to trade food for sex. The ultimate strategy from a male’s point of view would be to combine sperm with a gift in such a way that a female has to accept his sperm in order to secure the gift. Precisely this system has evolved in a number of insects in which nutrients are either dissolved in the
PROMISCUITY
seminal fluid or are contained in the packaging in which the
For many crickets and grasshoppers copulation comprises the male attaching a large spermatophore externally to the female’s genitalia. While the sperm leak out into her sperm store the female bends round and eats the wrapper - a proteinaceous coating. In some crickets the nutritious part of the spermato- phore is relatively enormous, representing one-third of the male’s body weight. Radio-tracers have been used to demon- strate that the nutrients in spermatophores are incorporated into a female’s eggs, and it has also been found that the more a female copulates, or the more spermatophores she eats, the greater her output of offspring. The situation in some grass- hopper species is extreme since the amount of nutrient a male puts into his spermatophore means that it can take him a week to produce another one. As you might expect, males are extremely fussy about who they give it to and focus their attentions on the largest, most fecund females. The corollary of this, in contrast to most other situations, is that females often have to compete for males rather than the other way round.
In some instances females may benefit from copulating with several males because they are able to digest excess sperm, exploiting the fact that males usually inseminate many more sperm than females need for fertilization. Rather than simply ejecting the leftovers, it makes sense for females to utilize the nutrients in excess sperm. Flatworms excel at few things, but they are extremely good at digesting sperm. Like many other hermaphrodites, individual flatworms exchange sperm recipro- cally during copulation. However, as we have seen, this is a risky business because it is vulnerable to exploitation. One individual may accept sperm from another, but not then reciprocate. Once an individual receives a sperm donation it can use it in any way it wishes, and many flatworms treat sex as nothing more than a free meal. Since food is often in short supply for flatworms, using another flatworm’s sperm is a good ploy. However, as soon as flatworms start doing this, selection
will favour any individual with a counter-ploy. Some flatworms have responded by either delivering their sperm wrapped in a digestion-proof spermatophore or including chemicals in their seminal fluid that neutralize the recipient’s digestive enzymes.24
Unconventional nuptial gifts are widespread in insects. Sev-
eral species have specialized glands
saliva or other valuable substances
Others have taken this even further and
for example, the males possess special, fleshy hind-wings which the female consumes during copulation.
The ultimate nutri- tional gift a male can provide to a female is himself. Sexual cannibalism occurs in a range of species but is best known in
spiders and praying mantis. Discussing the situation
European praying mantis in the 1890s, J. H. Fabre wrote, ‘If
the poor fellow is loved by his lady as the vivifier of her ovaries,
he is also loved as a piece of highly flavoured game.’ Ever since,
the idea of sexual cannibalism has excited speculation because it
was difficult to imagine how anything so apparently detrimental
could have evolved. It was therefore assumed that what Fabre
had so colourfully described was an artefact of observing mantis
in captivity. However, a selfish-gene view of sexual cannibalism
provides a much more compelling explanation. First, whether
something is good for the species is irrelevant; what matters
is the individual. Second, although sexual cannibalism may
be beneficial to females, males might not actually sacrifice them-
With these ideas in her head, Susan Fawrence of Sheffield University set off to Coimbra in Portugal to be one of the first
people to look at what mantis get up to in the wild.
that in about one-third of all copulations the male was killed and eaten by the female. She also discovered that, far from sacrificing themselves, males did everything in their power to avoid being eaten. This was blatantly obvious from watching copulatory sequences in nature. A couple of weeks after her final moult the female mantis is ready to copulate. She signals her readiness by releasing a pheromone and on catching a whiff
that produce nutritious on which females feed. in certain crickets,
of this a male will fly upwind until he locates her. However, he does not come winging in to alight beside her: this is far too dangerous. Instead, he usually alights about thirty or forty centimetres away. Both sexes have big eyes and excellent vision so the male waits until he can see the female. He then stalks her — extremely carefully. His behaviour is remarkable: if she turns to look in his direction he freezes; when she turns away he moves forward. If there’s a breeze, he edges forward using the movement of the fluttering vegetation to disguise his progress. Once he is within ten to twenty centimetres of the female he waits. He composes himself and when the female is looking elsewhere, he leaps on to her back in a single adroit movement. If he does it right, he’s safe. But if he misjudges his timing he ends up in her prickly embrace. Let’s assume our male got it right. Grasping the female and lying longitudinally along the female’s back, the male’s prehensile abdomen tip starts to seek out hers, and copulation takes place. After copulating for
about two hours the male either jumps or flies off the female’s back as quickly as you can imagine. Again, if he messes this up, he’s dead, but usually what happens is that he springs away from the female and hits the ground running. He does not want to be dinner.
Now let’s consider scenario two, in which the approaching male mistimes his leap and is caught by the female. She starts to consume him straight away and, depending on how the male was caught, she may or may not start with his head. If she eats his head she also consumes the nervous tissue that controls his copulation behaviour. Having lost this particular ganglion the male also loses all inhibitions about sex and his abdominal tip
goes into overdrive seeking out the female’s genital opening. Once connected he is perfectly capable of transferring his sperm - despite being headless. So, males don’t want to be eaten, but sometimes they are. If they are, then their reproductive end can still function reasonably well and they have a good chance of inseminating the female. On the other hand, females almost always want to eat their partner, since he represents a good
meal which enables her to produce more eggs. Even if he is eaten, providing the female does not copulate with another male, he has a good chance of fertilizing her eggs. However, females can and do copulate with several males, so the male has no guarantee of paternity.
A fundamentally different cannibalistic situation occurs in
the redback spider. The consumption of the male by the female
during copulation is routine in this species. But, in contrast to
the mantis, there is no sexual conflict here, since the male wants
to die — and in a most unusual way. Within seconds of starting
to inseminate the female the male redback does a somersault,
landing directly on the female’s mouth. Alice herself would
have been hard pressed to find a more obvious signal to ‘Eat
Me’ and in the majority of cases this is exactly what happens.
So, what’s in it for the male? There are two possibilities. His
body may, as in the praying mantis, contribute towards the
quality or quantity of offspring produced. Or his complicity in
cannibalism may increase his certainty of paternity. This could
be important because female redbacks often copulate with
more than one male. Working in Australia, Maydianne
Andrade of the University of Toronto showed that male
sacrifice is unlikely to result in increased offspring quality or
2
quantity. As in many spiders the male redback is much
smaller than the female, just i per cent of the female’s body mass, and only 2.5 per cent of the egg mass she later produces, so eating him would provide little extra nourishment. But in terms of paternity there were two advantages to being eaten. Copulations involving cannibalism last almost twice as long (25 minutes) as those that don’t (11 minutes), and the longer a copulation lasts the more likely the female is to use that male’s sperm to fertilize her eggs. In addition, following a cannibalistic copulation, females were much less eager to copulate with other males, thereby reducing the likelihood of sperm competition. In other words, in this instance it really seems to be adaptive for males to be eaten!
The males of many bird species, such as terns and eagles, 213
bring food to their female partner during the days before she lays. Depending on the species, the male may provide nearly all the female’s food requirements, and sometimes the delivery of a food item is followed by copulation. It seems logical that a female might be able to acquire even more food by accepting gifts from other males, in exchange for copulation. Surprisingly, female birds rarely seem to do this. An antipodean seabird, the red-billed gull, does it occasionally, as do some birds of prey, but these are the only known instances. Moreover, transactions have a hidden meaning: females that accept food from strangers are usually those whose own partner is less than adequate. A poor provider spells doom for a female: the quality
and quantity of eggs she produces will be reduced and any chicks that hatch will not be adequately provided for. Under these circumstances the female’s best bet is to find someone better, and in these species extra-pair copulation may be the
8 prelude to switching partners/
The bonobo is remarkable among primates
reasons and not least, as we have seen, because it substitutes sex for aggression. Females are receptive virtually throughout their cycle, and sex between group members, regardless of age or gender, is frequent. In bonobo society various forms of sex serve to dissipate tension, and sexual behaviour is especially frequent when there is competition for food. Part of the increase in sexual activity in response to food
females occasionally solicit copulations from males in exchange
for food/9
Finally, in this section, there is one instance in birds where
females seem to trade sex for an important resource other than food. Like most long-lived birds, Adelie penguins are socially monogamous. Using chunky pebbles, they build a raised nest to protect their eggs from melt water and the other filth that accumulates in seabird colonies. Stones are in short supply, because they are usually already part of a bird’s nest. When owners are away both male and female penguins steal stones from other nests. But if a male is alone at a nest, a female from
another pair may take a stone in exchange for copulation with the male nest-owner. This remarkable behaviour was discov- ered by Fiona Hunter, then at Cambridge University, working in the Antarctic, and when the media got hold of it, the headline was ‘Penguin Prostitutes’. 30 A
moment’s thought, however, would have revealed that the female penguin’s behaviour differs from human prostitution in several important ways. First, in humans it is usually males that seek out females. Second, males use prostitutes for sexual gratification, not to increase their chances of paternity. The fact that female Adelies can be so casual about copulating with other males is remarkable and has several implications. It could mean that female Adelies don’t care who the father of their offspring is. It could mean that females have exquisite control of sperm and can eject it after insemination. Or it could be that females choose their extra-pair partners rather carefully, and get the best of both worlds: stones
The Direct Benefit of Parental Care
A third, popular explanation for why females copulate with different males is to trade sex for parental care. If males are prepared to provide care for the offspring of females with whom they have copulated, females can benefit from multiple copu- lation partners. The benefit to such females is increased reproductive success because offspring cared for by more than one male may have a greater chance of survival. Parental care is most prevalent in birds and mammals, and can take a number of different forms. The most obvious is the provision of food, but males can also protect offspring from predators.
Polyandrous marriages in humans are extremely rare and occur only in cultures where, for ecological or social reasons, resources are extremely limited. In most instances polyandrous marriages involve two or more brothers sharing a single wife. From a male perspective a polyandrous marriage appears to be a poor deal since sharing a wife obviously limits male
reproductive success. It was probably this chauvinistic stand- point that led Joseph Hooker disparagingly to liken polyandrous people in Bhutan to primitive animals. Early anthropologists perpetuated this view by referring to them as survivors from a more primitive epoch. It is rather disappointing that Hooker, Darwin's closest ally and confidant, should have failed to see that polyandrous marriage might have had an adaptive expla- nation - even from a male perspective. Certainly from a female standpoint the parental care, love and attention of several husbands seems very advantageous. This is verified by the fact that in Ladakh the reproductive success of women in polyan- drous marriages, with an average of 5.2 children, was higher than those in monogamous marriages, who produced only 3.1
children. The social and evolutionary advantage of polyandrous marriage for males is that when resources are limited, as they are among feudal societies in the depauperate valleys of the Himalayas, two or more males are capable of supporting a wife whereas a single man cannot.31 Thus, under difficult ecological circumstances, polyandrous marriage is a case of men making the best of a bad job. Moreover, if a male has to share his paternity with another man, in evolutionary terms he is better off doing so with a close relative rather than with a non-relative. The genetic costs of sharing paternity are offset to a consider- able extent if co-husbands are relatives. Because brothers share half their genes, by helping to rear a brother’s offspring, a male
perpetuates some of his own genes. None the less, since any male would do even better if he were the sole husband, conflict between males in polyandrous marriages can be high. This is clearly illustrated by a natural experiment in Sri Lanka where co-husbands in polyandrous marriages may be either brothers or non-relatives. Among the former, marriages are more stable
and more enduring than those involving unrelated males.
In most polyandrous marriages the oldest brother is typically a few years younger than his wife. This in turn means that the wife is slightly, or in some cases considerably, older than her husband’s younger brothers, her co-husbands. The consequence
of this disparity in ages is very different reproductive success for the different males. Among the Nyinba, a Tibetan group living in north-western Nepal, where co-husbands are brothers and where wives place great importance on knowing paternity, the reproductive success of the eldest brothers was greater than second and third brothers. Not surprisingly, as soon as their financial situation permits it, younger brothers leave to strike out on their own.
Competition for sexual access to the wife and sperm compe- tition would appear to be almost inevitable in polyandrous marriages. However, in some societies, such as the Ladakhi Buddhists studied by John Crook, conflict is avoided by a clever cultural ploy. The brothers most likely to compete for paternity are the senior husband and his closest brother. To minimize potential disputes, second sons are honoured by being chosen as monks. Being selected as a monk is a sufficiently great honour for second sons to comply willingly - even though their reproductive success will be zero. For this reason, despite the apparent similarity between polyandrous marriage in humans and polyandrous mating systems in birds, the potential for sperm competition among humans is probably rather limited.
A more specialized form of polyandrous marriage is spouse exchange or co-marriage. In the early days of Arctic exploration the Inuit were infamous for allowing southern fur-trappers, whalers and traders to copulate with their wives. Inuit men would make their wife available to other men as a form of hospitality, especially if visitors were wealthy, as southerners often appeared to be. To those early explorers Inuit morals were merely lax, but a godsend for the likes of Peter Freuchen and Knud Ramussen so far from their own wives and girlfriends. Rather than being simply immoral, however, spouse exchange was an important part of Inuit culture, with clear adaptive value. Most cases of spouse exchange were actually co- marriages: the enduring co-operative and sometimes sexual relationship between two Inuit couples - a form of long- distance reciprocity with immense survival value. Co-marriage
instantly increased each participant’s circle of relatives, and as a result his or her likelihood of survival. Arctic conditions are often harsh, hunting success unpredictable and starvation all too familiar. Because family members were morally obliged to care for each other, having others that one could depend on must have more than offset any loss of paternity. But, if having additional family was so crucial, why didn’t the Inuit exchange their spouses more widely? What was so special about recipro- cating pairs of couples? The answer to this question has come not from the study of people but from research on the co-
operative polyandrous mating systems in birds.
During his brief but seminal visit to the Galapagos Islands
Darwin was impressed by the tameness of the birds and in a letter home he described how he had pushed a hawk off a branch with the muzzle of his gun. Darwin did not know it, but the Galapagos hawk has a mating arrangement with females blatantly trading sex for paternal care. The mating system of these birds is extremely unusual and is described as co-operative polyandry. It comprises two or three males paired to a single female. All the males copulate with the female, often in rapid succession, and subsequently all work
together to rear the conducted in the early 1990s revealed that each male in a group had a similar chance of being a father, hence explaining why it is worthwhile for each of
them to help rear the chicks.33
A similar system occurs in the dunnock, and Nicholas Davies
elegantly demonstrated the benefits females accrue by having two partners. By making sure both males copulate with her, the female trades sex for paternal care, and in so doing produces more offspring and increases her own reproductive success. This looks like a case of the more the merrier for the female. But when Davies and Ian Hartley cleverly arranged things so that female dunnocks could take a third husband, they declined. The reason was that males adjust how much effort they put into
feeding the chicks in relation to their share of paternity. In a
group comprising three husbands 218
reduced chance of being a father and so would reduce his chick- feeding effort accordingly, resulting in an overall reduction in the amount of male care. For females two husbands were better than one and much better than three. When Hartley and Davies looked at the other bird species whose mating system was co- operative polyandry they found a similar pattern. In almost all cases the number of males associated with one female was rarely more than two. With more than two males in a group a male’s chance of paternity is so reduced that it is not worth helping to rear offspring. Exactly the same explanation applies to co- marriage among the Inuit. Because Inuit co-families usually live in different communities, fathers provide parental care primarily for the offspring produced by their main wife. From the primary husband’s perspective, his chances of being the genetic father of the offspring he helps to rear are relatively high. However, if the system of spouse exchange involved more than two couples, each male would be much less certain of his paternity and the value of his parental care would be
We can now ask whether the provision of paternal care by extra-pair males explains the high incidence of extra-pair paternity we find in socially monogamous birds. The answer is that, under some circumstances, it might. In her study of red- winged blackbirds Elizabeth Gray found that females who had engaged in extra-pair copulations with neighbouring males were
.
less likely to lose their chicks to predators 35 She also showed
that when she placed a stuffed magpie - which in real life is a potential predator of blackbird chicks - near a female’s nest containing chicks, only her partner and those males from adjacent territories with whom she had copulated came to attack the predator. This looks like a clear case of females trading paternity for protection from predators. Unfortunately, this is unlikely to provide a general explanation for extra-pair copulation behaviour in socially monogamous birds. In another study of parentage in red-winged blackbirds, females with illegitimate young in their nest had reduced rather than
enhanced breeding success and in other bird species there is no evidence that extra-pair males provide any kind of parental care.
A somewhat perverse form of paternal care is not being
aggressive to your offspring.36
renowned for being unpleasant to babies, but only if they are not their own. One way females can avoid this abuse of their offspring is to copulate with several males and persuade each of them that they might be the father. In that way it would never be in a male’s interest to injure her offspring. Polyandry may be
beneficial because it confuses paternity, which may in turn increase the chances of copulating males caring for offspring, and simultaneously minimize the likelihood of infant abuse or infanticide by males. Although there has been no formal test of the avoiding-infant-abuse hypothesis, several observations are consistent with it. The most likely perpetrators of infant abuse or infanticide are males that have recently joined a group or that
might take over a group - extra-group males. Females seem to be particularly keen to solicit copulations from such males. Male troop take-overs and subsequent infanticide are regular events among langurs and
Some of the most dramatic evidence that certainty of pater- nity influences the way males behave towards juveniles comes from our own species. When a woman rears a family consisting of offspring from both a previous partner and a current partner, the conflict between stepfather and stepchildren is considerably
greater than that between biological parent and offspring. In the 1980s Martin Daly and Margo Wilson showed that stepchildren were much more likely to suffer from abuse than were children
extra-group males while
striking still, after a take-over, females will solicit copulations from the new males, even though they are already pregnant!
with both biological
abuse is a side-effect of the psychological
suggested that child mechanisms that to offspring.37 You might be forgiven for thinking that human females are unlikely
females will even copulate a take-over is in progress.
to trade sex for paternal care in contemporary Western society (although I suspect we can all think of isolated examples), hut in the Ache, a group of preliterate South American Indians, there is clear evidence for such an effect.
The first contact between the Ache people and white anthro- pologists occurred as recently as 1972 — with the inevitable hut disastrous result that 40 per cent of them died from Western diseases. In reassembling a picture of Ache life before 1972, an- thropologists Kim Hill and and Magdalena Hurtado of the University of New Mexico discovered an extremely brutal society in which male aggression ruled; club fights between men were common and as a result children were often left fatherless.38 Fathers sometimes also abandoned their partner and offspring of their own accord - presumably seeking better reproductive opportunities elsewhere. Children without fathers had little better than a 50 per cent chance of survival, compared with an 86 per cent chance with their father present. In this tough hunter-gatherer society pragmatism rode roughshod over compassion, and abandoned children were usually buried alive - unless they had a secondary father. There is no marriage as such in Ache society, and women are typically polyandrous, with at least two partners: a primary partner - the one most likely to be the father of a particular child - and a secondary one, who still has some likelihood, albeit rather less, of being that child’s father. Through this system of multiple sexual
partners, females generated uncertainty about paternity, which ensured that if the primary father died, there would be someone else to care for and feed their children: a clear case of human females trading sex for paternal investment.
The Direct Benefit of Avoiding Rejection Costs
On the outskirts of Sheffield where I live there are numerous ponds or dams which in past times provided the power for the early steel industry. The ponds are occupied by numerous mallard ducks and each spring we are forced to witness the
sexual subjugation of females by males. Forced copulation is a ubiquitous feature of mallard reproduction: fleeing females are
followed by their partner and one
Eventually the female is forced,
sometimes into the middle of the school playground, sometimes back on to the pond. She is then forcibly copulated while her partner fights valiantly to remove the extra-pair male. brutal, brutish and extremely dangerous for females, several of which die each year as a result. But, to judge from the molecular evidence, forced extra-pair copulation is a successful strategy for male mallards 40
One option the female has under these circumstances is to give in to the male and allow him to copulate so that she will be left in peace. This explanation assumes that copulating with a harassing male carries no costs. This seems unlikely because often the males that resort to coercive tactics are defective in some way: low status, poor quality, or diseased and with little chance of securing a partner using more conventional tactics. If this is true then simply giving in to a male to save time and harassment does not sound like a good idea. To put it another way, this kind of female strategy would not be evolutionarily
stable. If all females simply rolled over and acquiesced every time a male made a nuisance of himself, there would be chaos since all males would be at it. Whenever copulations are costly for females, for whatever reason, it will usually pay them to resist. Of course, the corollary of this is that there will be selection on males to make an ever greater nuisance of them- selves. By continuing to resist, females may be trying to make
persistence disproportionately
energy for males. 4 ' If you watch mallard ducks in early spring
as they are chased and harassed by male after male, this cat and
the most to accept males’ advances, but then to dump their sperm. If they could do this, some of the costs at least would be eliminated, and because males would stand to gain nothing in terms of paternity, there
would be no evolutionary incentive for the behaviour to persist. The fact that coercive male behaviour is widespread, and in mallards at least is quite often successful, suggests that females do not have total control over the fate of inseminated sperm.
When females do not appear to obtain direct benefits for themselves from copulating with more than one male, the possibility exists that they might acquire indirect benefits: that is, genetic benefits for their offspring. At first sight the idea of genetic benefits seems entirely plausible, but in terms of both theory and reality it is a quagmire.4*
There are four possible types of genetic benefit, (i) By having several sexual partners females might increase the genetic diversity of their offspring. In a rapidly changing world this might ensure that at least some of a female’s offspring are adapted to prevailing environmental conditions and so will survive to reproduce. This seems sensible, but population gen- eticists consider that the additional genetic variation a female would gain from being inseminated by more than one male would be trivial. The point can be most clearly seen by thinking of ourselves: we each have 23 pairs of chromosomes and, as a result of meiosis and recombination, a man can potentially produce 2*3 (i.e. an astronomical number of) different sperm types. It has therefore been assumed that by copulating with several partners females will do little to increase the genetic diversity of their offspring.43 The result is that this particular genetic benefit has received little empirical attention.
(2) As was suggested by Trivers in 1972, females might base their choice of partner on genetic complementarity. At first no one bothered to follow up Trivers’s idea because no one could imagine how, on the basis of a male’s appearance, a female could assess whether his genotype would match her own. Only later, when it was clear that females routinely copulated with several males and that some females at least had the ability to
discriminate between the sperm of different males, as we saw in chapter 6, did the idea of post-insemination choice based on compatibility seem feasible. Jeanne Zeh of Rice University, Texas, and her husband David Zeh, now at the University of Reno, have championed the view that females copulate with different males to minimize their chances of being fertilized by an incompatible male. They point out that there is abundant evidence for genetic incompatibility - inbreeding depression being the most obvious example. They also draw attention to a recently discovered cause of incompatibility - the existence of
genetic conflicts within and between the nucleus and cytoplasm. Such intragenomic conflict results in genomes being dynamic and evolving entities. This state of genomic flux will of course increase the likelihood that your genes will not be compatible with those of your partner.44
We now come to two genetic benefits which have been the major focus of both theory and practical research for the last twenty years. These two ideas have also been the focus of considerable controversy, partly because biologists have been divided over whether they might work, and partly because,
despite this, several studies claim to have shown that they exist.
They comprise two
proposed by the geneticist and statistician R. A. Fisher in the
1930s and is concerned with genes for attractiveness. Fisher
proposed that females choose attractive
these males are better in any way, but simply because they are attractive. He proposed that if genes for a female preference for particular male characteristics were linked to the attractive trait in males, a runaway process would occur. In this
females always choose the most attractive males and produce attractive sons so that over successive generations there is very
strong directional selection which — in the most extreme cases — results in the dramatic sexual dimorphism we see in species such as birds of paradise. By choosing to copulate with an attractive male a female will produce attractive sons, who in turn will be successful in reproduction.45
T he idea here is that male secondary sexual traits, such as shiny plumage, a brightly coloured bill, or a complicated song, are not arbitrary features, but have evolved to reflect a male’s inherent quality. In this scenario attractive males are also extremely viable males. By having her eggs fertilized by an attractive male a female acquires viability genes for her offspring — genes that will help her
offspring survive to reproduce.
These latter two classes of benefit have become the focus of a
phenomenon known as the ‘paradox of the lek’.4" The lek mating system of many animals, including some fish, frogs, birds and mammals, in which males congregate and females turn up simply to copulate with one of the males, is a paradox because biologists cannot see what females get from it. In most lekking species the majority of females copulate with only a handful of males, so the distribution of copulations among males is very uneven. After a female has been inseminated she leaves the lek to rear her offspring alone. It is generally assumed that she gets no direct benefits from her choice of partner at the lek, and it is therefore inferred that she must obtain some indirect benefit from choosing a particular male, and the benefit is generally assumed to be genes - either for attractiveness or for viability.
But there is a major problem with the idea that female choice on leks is driven by either of these genetic benefits. If females all agree about what sort of male they want to copulate with and females always copulate with this type of male, it creates incredibly strong selection on males. Any male that doesn’t conform to the female’s ideal is genetically dead. He’s not chosen, never copulates and the genes he carries go nowhere. The problem with directional selection is that very soon all the genetic variation in male attractiveness is used up. Breeders of domestic animals soon found this out when they put their breeding programmes into practice - they imposed strong directional artificial selection on cattle to improve milk yield, or on chickens to improve egg production. But within a few
The final type of genetic benefit (4) is viability.
generations all individuals were much the same and no further improvement could be made, simply because all the genetic variation for these traits had been used up. If this were true of sexual selection among wild animals, then there would be no point in females exerting any choice, but they do!
One solution to this conundrum, suggested by Bill Hamilton
and Marlene Zuk, is a system in which selection is not
consistently in the same direction. Hamilton and Zuk proposed
that parasites, by continuously evolving new varieties (and at
faster rates than their hosts) might provide just this kind of
regime 48 They suggested that males infested with parasites are .
incapable of producing secondary sexual traits to the same extent as healthy males. By choosing males with the most elaborate traits, females acquire viability genes, specifically genes for disease resistance, for their offspring. The ingenious aspect of Hamilton’s and Zuk’s idea hinges on the fact that disease organisms are constantly evolving, and the consequence of this is that selection for disease resistance via female choice never continues long enough in the same direction to use up all the genetic variation.
Testing these ideas has been extremely problematic. Demonstrating indirect benefits at all has been difficult, and distinguishing between genetic benefits for attractiveness and viability has been all but impossible. There are two problems to overcome. First, researchers cannot start to consider genes for attractiveness until they have eliminated the possibility that females are gaining genes for viability and vice versa. Second, among wild animals, especially those on leks, it is extremely difficult to conduct experiments that would allow researchers to tackle the idea of indirect benefits. Ideally, what one would like to do is to see what happens when females cannot choose, but since females on leks always choose and it is difficult to make them copulate with another male, this is something of a dead end. We know from a number of laboratory studies of more tractable species, such as fruitflies, that if you allow a female to choose her partner her reproductive success is greater than if her
partner is allocated to her at random by the experimenter. Lekking sage grouse or Uganda kob hardly lend themselves to this type of manipulation and so it looks as though the paradox of the lek may remain unresolved.
There is a feasible alternative, however. It does not involve lekking species, but comprises another situation where females may obtain indirect benefits from copulating with particular males. Imagine a female swallow migrating back from Africa to breed in Europe. She is heading back to the farmyard where she bred in the previous year, but this time she is late, delayed by a sandstorm over the Sahara. When she arrives almost all the other females are paired, and paired to the most attractive males. She has to make do with Mr Average. She pairs up with him and they build a nest. However, in the middle of her fertile period she gives her partner the slip, finds one of the really attractive males, and copulates with him. She then returns to her nest as though nothing has happened. Pairing up with Mr Average has provided the female with the direct benefit of an opportunity to breed at all, but by surreptitiously copulating at the optimal time for fertilization with Mr Much-better-than- average she can be fairly sure at least some of her chicks are fathered by the better male. So here we have a situation where a female is producing two types of baby: legitimate offspring from her social partner whom she doesn’t really rate, and illegitimate offspring from her extra-pair partner whom she specifically chose to copulate with. Suddenly we have a sperm competition
situation in which it might be possible to see whether females gain genetic benefits from copulating with particular males.
On the basis of this scenario we can make several predictions. If a female gets viability genes from her extra-pair partner, then her illegitimate offspring should survive better than her legit- imate offspring. If there is no difference in the viability of the offspring from her different partners we can forget this parti- cular genetic benefit and look at genes for attractiveness. The prediction in this situation is that, when they grow up, the
illegitimate offspring will be more attractive to females than the
This brings us to the question of what genes could confer viability benefits. The answer seems to be, for some species at least, those that provide protection from pathogens. Some of the most convincing results come from a study of the MHC - the major histocompatibility complex - of house mice. The MHC comprises a closely linked set of genes in vertebrates associated with disease resistance and the immune system. Wayne Potts from the University of Florida and his colleagues found that female mice living in the territory of a male whose MHC was similar to their own were more likely to engage in copulations with other males, especially those with a different MHC.50 A
plausible explanation for this female mate choice is that off- spring fathered by a male whose MHC is different from their mother’s acquire a more varied set of MHC genes and hence a more effective pathogen resistance
distinguish different
odours, it seems likely that females ‘know’
extra-pair copulations simply on the basis of their partner’s smell. It remains to be shown in this system that the offspring of mice with compatible MHCs are more viable than those whose MHCs are incompatible. However, this seems to be a very clear example of a case where females can decide on the genetic compatibility of their partner simply on the basis of his scent. There is some tantalizing, but controversial, evidence from a study by Claus Wedekind that MHC may also affect choice of
partners in humans, this time mediated by body odours. What is not known is whether female choice for males with particular
MHC attributes is simply to avoid
inbreeding, or whether this is part of the process of sexual selection in which females acquire specific genetic benefits for
their offspring.
Birds are renowned for their lack of any sense of smell. None
the less there have been several studies that suggest that females might gain viability benefits in the form of disease-resistance
genes via their choice of extra-pair partner. In Anders Moller’s now famous study of swallows, females preferentially pair with and specifically seek extra-pair copulations with long-tailed individuals. As a consequence, the reproductive success of long-tailed males is much greater than that of other males since they father offspring both in their own nest and in the nests of other males. Females also appear to benefit, this time indirectly, because long-tailed males appear to have a better- developed immune system, indicating that the reason offspring fathered by long-tailed males survive better is because they resist
pathogens more effectively than those fathered by other males. In another study, the indirect benefits females accrue from infidelity appear to be genes for attractiveness rather than
viability. In the scrubby regions of Australia lives a handful of tiny jewel-like wren species whose superlative names - splendid, lovely and superb - describe the appearance of the males. The superb fairy wrens breeding in the Canberra Botanical Garden have been the subject of a detailed study over the past decade by a team led by Andrew Cockburn of Australia’s National
These birds live in co-operative groups which comprise a breeding pair aided by several grown sons. Astonishingly, over two-thirds of all offspring are fathered by extra-group males. During the breeding season males are like persistent salesmen relentlessly touting their wares in front of other males’ females. They do this by plucking a yellow flower petal, which contrasts perfectly with their iridescent cobalt-blue plumage, and presenting this to a female. These displays eventually lead to extra-pair copulations when females surrep- titiously slip away from home in the dim light of dawn to visit one of these males in his own territory. Extra-pair males provide no direct benefits to females but the fact that most extra-pair fertilizations are achieved by a dynasty of males, stemming from one particularly successful individual, strongly suggests that the benefits females accrue from extra-pair copulations are genetic. Cockburn and his colleagues were able to eliminate the idea that these genetic benefits were for viability, since there was no
evidence whatsoever that £xtra-pair offspring had any greater reproductive success or were more likely to survive than other offspring, leaving researchers with the possibility of attractive- ness genes.
What can we conclude from all this? In
instances where females choose to copulate with more than one male it seems very likely that they benefit from doing so. The evidence would seem particularly compelling if there exists a positive association between the number of partners a female has and her reproductive success. However, we should bear in mind that this association could arise even when females gain no benefit from multiple copulation partners - simply because high-quality females attract more males. Fortunately, in some instances it is clear that it is advantageous for females to have multiple partners - obtaining direct benefits, such as a fresh
supply of sperm as in fruitflies, or parental care as in dunnocks.
choice of copulation partner, but this idea is even more difficult to test.
Finally, there is one further type of genetic benefit that avoids
the mechanistic complexities
attractiveness genes, and that is Jeanne and David Zeh’s idea that females choose their copulation partners, or their sperm, on the basis of their genetic complementarity 54 Several studies of
Although genetic benefits continue
because the mechanisms by which they might operate are still poorly understood - several studies have produced convincing evidence that females produce better-quality offspring by copu- lating with several different males. There is less evidence for the idea that females acquire genes for attractiveness through their
pre-copulatory choice and cryptic female choice in particular have provided tantalizing evidence which is consistent with the idea of compatibility. In some instances females may be able to discriminate between males if they advertise which particular ‘type’ they are - as in mice, and possibly humans, which signal their MHC type via their odour. However, in the absence of such signals females may do best by copulating with several
catching HIV from chimpanzees is defenceless 56
different males and allowing their reproductive tract to sort out compatible from incompatible sperm.
There is now some compelling evidence for the genetic compatibility idea, and it comes from an unlikely study organ- ism — a tiny pseudoscorpion which inhabits rotting wood in
Panamanian forests 55 This animal is unusual in a number of
ways: the first is that it gets to new rotting stumps by hitching a lift under the wings of a giant harlequin beetle. The second is its revealing mode of reproduction. Uniquely, the pseudoscorpion lays its eggs into a translucent external brood sac - effectively an external womb - enabling researchers to monitor the hatching and survival of the babies as they grow, something that is impossible in other animals which produce live young. The researchers found that female pseudoscorpions that had been inseminated by only a single male aborted a much higher prop- ortion of their developing embryos than females that had been allowed to copulate with several partners. Cleverly designed experiments in which females obtained the same numbers of sperm from different partners revealed that abortions occurred because of genetic incompatibility between certain partners. Polyandry, therefore, was beneficial to females as a pre-emptive strategy to avoid this kind of reproductive failure.
Finally, a note of caution: I do not want to give the impression that females benefit unreservedly from polyandry. In sexually monogamous species, promiscuity either confers no benefits or imposes a cost on females. Even in species where polyandry is routine, female behaviour may be a trade-off between the costs and benefits. The most apparent cost of polyandry, at least for humans, is the risk of contracting a sexually transmitted disease, such as HIV, which may compro- mise the lifetime reproductive success of both females and males. The selection pressures created by HIV on humans are far too recent for us to have adapted, but intriguingly our more promiscuous cousin, the chimpanzee, seem to have done just that. Chimpanzees are relatively immune to HIV, but a human
InthepreviouschaptersI havetriedtoaddressthedouble-edged question of why sex is a battle. We have explored this issue principally by considering historical and biological evidence and, while not ignoring it, I have avoided becoming embroiled in the politics of gender bias among researchers. There has undoubtedly been a battle between female and male researchers to force biologists to consider the female perspec- tive, and many with an interest in feminism may feel that, while they may not have won, there has been considerable progress. From what was a distinctly androcentric starting point, recent research has made it abundantly clear that females are not passive participants in sexual reproduction. None the less, I think we also have to accept the truth of Geoff Parker’s assertion, made over twenty years ago, that sexual selection acts more intensively on males than on females. That sexual
selection pushes males harder than females is largely down to the size of their sex cells. When one set of contestants comprises a swarm of tiny, highly mobile sperm and the other a few large, immotile eggs, there is a certain inevitability about the type of battle that will ensue. The most important consequence of the difference between male and female sex
cells is that the reproductive potential of males far outstrips that of females. The most successful males produce many more offspring than
the most successful females. However,
everything their own way and for species with internal fertiliz- ation, at least, the evolutionary battleground on which sexual conflict occurs - the female reproductive tract - is one designed by evolution to counter the ability of sperm to run circles round eggs. This is, perhaps, the most significant discovery of the past two decades — that male and female reproductive attributes co-
.
evolve 57 Adaptation and counter-adaptation - both sexes in a
state of dynamic flux — each evolving, now, in response to
adaptations in the other. Where males
copulating, females have a range of subtle behavioural and
physiological counter-strategies. Where females evolve long and convoluted reproductive tracts better to regulate sperm uptake, males respond by evolving longer sperm. The idea that there exists a battle between the sexes implies that there are winners and losers, but if we think about sexual interactions as part of a co-evolutionary process, it is not obvious that either sex can ever be a clear winner. At any moment in time one sex may have slightly more control than the other, but the battle between the sexes is an evolutionary see-saw - subtle, sophisticated and inevitable.