For humans, the results of those marriages with advertised ovulations would be awful. Fathers would rarely be at home, mothers would be unable to rear kids unassisted, and babies would die in droves. That would be bad for both mothers and fathers, because neither would succeed in propagating their genes.

Now let's picture the reverse scenario, in which a husband has no clue to his wife's fertile days. He then has to stay at home and make love with her on as many days of the month as possible if he wants to have much chance of fertilizing her. Another motive for him to stay at home is to guard her constantly against other men, since she might prove to be fertile on any day that he is away. If the philandering husband has the bad luck to be in bed with another woman on the night when his wife happens to be ovulating, some other man might be in the philanderer's bed fertilizing his wife, while the philanderer himself is wasting his adulterous sperm on another woman unlikely to be ovulating then anyway. Under this reverse scenario, a man has less reason to wander, since he can't identify which of his neighbor's wives are fertile. The heartwarming outcome: fathers hang around and share baby care, with the result that babies survive. That's good for mothers as well as fathers, both of whom now succeed in transmitting their genes.

In effect, Alexander and Noonan argue that the peculiar physiology of the human female forces husbands to stay at home (at least, more than they would otherwise). The woman gains by recruiting an active coparent. But the man also gains, provided that he cooperates and plays by the rules of his wife's body. By staying home, he acquires confidence that the child whom he is helping to rear really does carry his genes. He needn't be fearful that, while he is off hunting, his wife (like a female baboon) may start flashing a bright red derriere as an advertisement for her imminent ovulation, thereby attracting swarms of suitors and publicly mating with every man around. Men accept these ground rules to such a degree that they continue to have sex with their wives during pregnancy and after menopause, when even men know that fertilization is impossible. Thus, in Alexander and Noonan's view, women's concealed ovulations and constant receptivity evolved in order to promote monogamy, paternal care, and fathers' confidence in their paternity.

Competing with this view is the many-fathers theory developed by the anthropologist Sarah Hrdy of the University of California at Davis. Anthropologists have long recognized that infanticide used to be common in many traditional human societies, although modern states now have laws against it. Until recent field studies by Hrdy and others, though, zoologists had no appreciation for how often infanticide occurs among animals as well. The species in which it has been documented now include our closest animal relatives, chimpanzees and gorillas, in addition to a wide range of other species from lions to African hunting dogs. Infanticide is especially likely to be committed by adult males against infants of females with whom they have never copulated-for example, when intruding males try to supplant resident males and acquire their harem of females. The usurper thus “knows” that the infants killed are not his own.

Naturally, infanticide horrifies us and makes us ask why animals (and formerly humans) do it so often. On reflection, one can see that the murderer gains a grisly genetic advantage. A female is unlikely to ovulate as long as she is nursing an infant. But a murderous intruder is genetically unrelated to the infants of a troop that he has just taken over. By killing such an infant, he terminates its mother's lactation and stimulates her to resume estrus cycles. In many or most cases of animal infanticide and takeovers, the murderer proceeds to fertilize the bereaved mother, who bears an infant carrying the murderer's own genes.

As a major cause of infant death, infanticide is a serious evolutionary problem for animal mothers, who thereby lose their genetic investment in murdered offspring. For instance, a typical female gorilla over the course of her lifetime loses at least one of her offspring to infanticidal intruding male gorillas attempting to take over the harem to which she belongs. Indeed, over one-third of all infant gorilla deaths are due to infanticide. If a female has only a brief, conspicuously advertised estrus, a dominant male can easily monopolize her during that time. All other males consequently “know” that the resulting infant was sired by their rival, and they have no compunctions about killing the infant.

Suppose, though, that the female has concealed ovulations and constant sexual receptivity. She can exploit those advantages to copulate with many males-even if she has to do it sneakily, when her consort isn't looking. While no male can then be confident of his paternity, many males recognize that they might have sired the mother's eventual infant. If such a male later succeeds in driving out the mother's consort and taking her over, he avoids killing her infant because it could be his own. He might even help the infant with protection and other forms of paternal care. The mother's concealed ovulation will also serve to decrease fighting between adult males within her troop because any single copulation is unlikely to result in conception and hence is no longer worth fighting over.

As an example of how widely females may thus use concealed ovulation to confuse paternity, consider the African monkeys called vervets, familiar to anyone who has visited an East African game park. Vervets live in troops consisting of up to seven adult males and ten adult females. Since female vervets give no anatomical or behavioral signs of ovulation, the biologist Sandy Andelman sought out an acacia tree with a troop of vervets, stood under the tree, held up a funnel and bottle, collected urine when a female relieved herself, and analyzed the urine for hormonal signs of ovulation. Andelman also kept track of copulations. It turned out that females started to copulate long before they ovulated, continued long after they ovu-lated, and did not reach their peak sexual receptivity until the first half of pregnancy.

At that time the female's belly was not yet visibly bulging, and the deceived males had no idea that they were utterly wasting their efforts. Females finally ceased copulating during the latter half of pregnancy, when the males could no longer be deceived. That still left most males in the troop ample time to have sex with most of the troop's females. One-third of the males were able to copulate with every single female. Thus, through concealed ovulation female vervets ensured the benevolent neutrality of almost all of the potentially murderous males in their immediate neighborhood.

In short, Hrdy considers concealed ovulation an evolutionary adaptation by females to minimize the big threat to their offsprings' survival posed by adult males. Whereas Alexander and Noonan view concealed ovulation as clarifying paternity and reinforcing monogamy, Hrdy sees it as confusing paternity and effectively undoing monogamy.

At this point, you may be starting to wonder about a potential complication in both the daddy-at-home theory and the many-fathers theory. Why is human ovulation concealed from women as well, when all that's required by either theory is for women to conceal ovulation from men? For example, why couldn't women keep their derrieres the same shade of red every day of the month to deceive men, while still remaining aware of sensations of ovulation and just faking an interest in sex with lusty men on non-ovulatory days?

The answer to that objection should be obvious: it would be hard for a woman convincingly to fake sexual receptivity if she felt turned off and knew that she was currently infertile. That point applies with particular force to the daddy-at-home theory. When a woman is involved in a long-lasting monogamous relationship in which the partners come to know each other intimately, it would be hard for her to deceive her husband unless she herself were deceived as well.