Signals of humans and other animals have evolved through natural selection. For example, consider two individual animals of the same species, differing slightly in size and strength, facing each other over some resource that would benefit either individual. It would be advantageous to both individuals to exchange signals that accurately indicate their relative strength, and hence the likely outcome of a fight. By avoiding a fight, the weaker individ-ual is spared the likelihood of injury or death, while the stronger individual saves energy and risk.

How do animal signals evolve? What do they actually convey? That is, are they wholly arbitrary, or do they pos-sess any deeper meaning? What serves to ensure reliability and to minimize cheating? We shall now explore these questions about the body signals of humans, especially our signals related to sex. However, it is useful to begin with an overview of signals in other animal species, for which we can gain clearer insights through doing controlled experiments impossible to do on humans. As we shall see, zoologists have been able to gain insights into animal signals by means of standardized surgical modifications of animals' bodies. Some humans do ask plastic surgeons to modify their bodies, but the result does not constitute a well-controlled experiment.

Animals signal each other through many channels of communication. Among the most familiar to us are auditory signals, such as the territorial songs by which birds attract mates and announce possession to rivals, or the alarm calls by which birds warn each other of dangerous predators in the vicinity. Equally familiar to us are behavioral signals: dog lovers know that a dog with its ears, tail, and hair on the neck raised is aggressive, but a dog with its ears and tail lowered and neck hair flat is submissive or conciliatory. Olfactory signals are used by many mammals to mark a territory (as when a dog marks a fire hydrant with the odors in its urine) and by ants to mark a trail to a food source. Still other modalities, such as the electrical signals exchanged by electric fishes, are unfamiliar and imperceptible to us.

While these signals that I have just mentioned can be rapidly turned on and off, other signals are wired either permanently or for extended times into an animal's anatomy to convey various types of messages. An animal's sex is indicated by the male/female differences in plumage of many bird species or by the differences in head shape between male and female gorillas or orangutans. As discussed in chapter 4, females of many primate species advertise their time of ovulation by swollen, brightly colored skin on the buttocks or around the vagina. Sexually immature juveniles of most bird species differ in plumage from adults; sexually mature male gorillas acquire a saddle of silvery hairs on the back. Age is signaled more finely in Herring Gulls, which have distinct plumages as juveniles and at one, two, three, and four or more years of age.

Animal signals can be studied experimentally by creating a modified animal or dummy with altered signals. For instance, among individuals of the same sex, appeal to the opposite sex may depend on specific parts of the body, as is well known for humans. In an experiment demonstrating this point, the tails of male Long-Tailed Widowbirds, an African species in which the male's sixteen-inch tail was suspected of playing a role in attracting females, were lengthened or shortened. It turns out that a male whose tail is experimentally cut down to six inches attracts few mates, while a male with a tail extended to twenty-six inches by attaching an extra piece with glue attracts extra mates. A newly hatched Herring Gull chick pecks at the red spot on its parent's lower bill, thereby inducing the parent to vomit up half-digested stomach contents to feed the chick. Being pecked on the bill stimulates the parent to vomit, but seeing a red spot against a pale background on an elongated object stimulates the chick to peck. An artificial bill with a red dot receives four times as many pecks as a bill lacking the dot, while an artificial bill of any other color receives only half as many pecks as a red bill. As a final example, a European bird species called the Great Tit has a black stripe on the breast that serves as a signal of social status. Experiments with radio-controlled, motor-operated tit models placed at bird feeders show that live tits flying into the feeder retreat if and only if the model's stripe is wider than the intruder's stripe.

One has to wonder how on Earth animals evolved so that something seemingly so arbitrary as the length of a tail, the color of a spot on a bill, or the width of a black stripe produces such big behavioral responses. Why should a perfectly good Great Tit retreat from food just because it sees another bird with a slightly wider black stripe? What is it about a wide black stripe that implies intimidating strength? One would think that an otherwise inferior Great Tit with a gene for a wide stripe could thereby gain undeserved social status. Why doesn't such cheating become rampant and destroy the meaning of the signal?

These questions are still unresolved and much debated by zoologists, in part because the answers vary for different signals and different animal species. Let's consider these questions for body sexual signals-that is, structures on the body of one sex but not the opposite sex of the same species, and that are used as a signal to attract potential mates of the opposite sex or to impress rivals of the same sex. Three competing theories attempt to account for such sexual signals.

The first theory, put forward by the British geneticist Sir Ronald Fisher, is termed Fisher's runaway selection model. Human females, and females of all other animal species, face the dilemma of selecting a male with which to mate, preferably one bearing good genes that will be passed on to the female's offspring. That's a difficult task because, as every woman knows all too well, females have no direct way to assess the quality of a male's genes. Suppose that a female somehow became genetically programmed to bo sexually attracted to males bearing a certain structure that gives the males some slight advantage at surviving compared to other males. Those males with the preferred structure would thereby gain an additional advantage: they would attract more females as mates and hence transmit their genes to more offspring. Females who preferred males with the structure would also gain an advantage: they would transmit the gene for the structure to their sons, who would in turn be preferred by other females.

A runaway process of selection would then ensue, favoring those males with genes for the structure in an exaggerated size and favoring those females with genes for an exaggerated preference for the structure. From generation to generation the structure would grow in size or conspicu-ousness until it lost its original slight beneficial effect on survival. For instance, a slightly longer tail might be useful for flying, but a peacock's gigantic tail is surely no use in flying. The evolutionary runaway process would halt only when further exaggeration of the trait would become detrimental for survival.

A second theory, proposed by the Israeli zoologist Amotz Zahavi, notes that many structures functioning as body sexual signals are so big or conspicuous that they must indeed be detrimental to their owner's survival. For instance, a peacock's or widowbird's tail not only doesn't help the bird survive but actually makes life more difficult. Having a heavy, long, broad tail makes it hard to slip through dense vegetation, take flight, keep flying, and thereby escape predators. Many sexual signals, like a bowerbird's golden crest, are big, bright, conspicuous structures that tend to attract a predator's attention. In addition, growing a big tail or crest is costly in that it uses up a lot of an animal's biosynthetic energy. As a result, argues Zahavi, any male that manages to survive despite such a costly handicap is in effect advertising to females that he must have terrific genes in other respects. When a female sees a male with that handicap, she is guaranteed that he is not cheating by carrying the gene for a big tail and being otherwise inferior. He would not have been able to afford to make the structure, and would not still be alive, unless he were truly superior.