And these might be no more than suggestive coincidences; correlation does not imply cause. Three kinds of evidence are needed to turn their conjecture into a fact: first, that there are regular genetic cycles in hosts and parasites; second, that ornaments are especially good at demonstrating freedom from parasites; third, that females choose the most resistant males for that reason rather than the males just happening to be the most resistant.
The evidence has been pouring in since Hamilton and Zuk first published their theory: Some of it supports them, some does not. None quite meets all the criteria set forth above. Just as the theory predicts that the more flamboyant species should be the ones most troubled by parasites, so it predicts that within a species the more flamboyant a male 's ornament, the lower his parasite burden.
This proves to be true in diverse cases; it is also true that females generally favor males with fewer parasites. This holds for sage grouse, bowerbirds, frogs, guppies, even crickets:" In swallows, females prefer males with longer tails; those males have fewer lice, and their offspring inherit louse resistance even when reared by foster swallow parents." Something similar is suspected in pheasants and jungle fowl (the wild species to which domestic chickens belong)." Yet these are deeply unshocking results. It would have been far more surprising to find females being seduced by sick, scrawny males than to find them succumbing to the charms of the healthiest. After all, they might be avoiding a sick male for no better reason than that they do not wish to catch his bug."°
Experiments done on sage grouse have begun to satisfy some of the skeptics. Mark Boyce and his colleagues at the University of Wyoming found that male grouse sick with malaria do poorly, and so do males covered with lice. They noticed, too, that the
::: 152 :::
The Red Queen
lice were easy to notice because they left spots on the males ' inflat-ed air sacs. By painting such spots on a healthy male 's sac, Boyce and his colleagues were able to reduce his mating success." If they could go on to show cycles from one resistance gene to another mediated by female choice, they would have given the Good-gene theory a significant boost.
THE SYMMETRY OF BEAUTY
In 1991, Anders Moller and Andrew Pomiankowski stumbled on a possible way of settling the civil war between Fisher and Good-genes: symmetry. It is a well-known developmental accident that animals ' bodies are more symmetrical if they were in good condition when growing up, and they are less symmetrical if they were stressed while growing: For example, scorpionflies develop more symmetrically when fathered by well-fed fathers that could afford to feed their wives: The reason for this is simply the old wrench-in-the-works argument: Making something symmetrical is not easy. If things go wrong, the chances are it will come out asymmetricaclass="underline" "
Most body parts, such as wings and beaks, should therefore be most symmetrical when they are just the right size and be the least symmetrical when stress has left them too small or too large: If Good-geners are right, ornaments should be the most symmetrical when they are the largest because large ornaments indicate the best genes and the least stress: If Fisherians are right, you would expect no relationship between ornament size and symmetry; if anything, the largest ornaments should be the least symmetrical because they reflect nothing about the owner other than that he can grow the largest ornament.
Moller noticed that, among the swallows he studied, the longest tails of the males were also the most symmetrical. This was quite unlike the pattern of other feathers, such as wings, which obeyed the usual rule: The most symmetrical were the ones closest to the average length: In other words, whereas most feathers show a U-shaped curve of asymmetry against length, tail streamers show a THE PEACOCK ' S TALE
::: 153 :::
steady upward progression: Since the swallows with the longest tails are the most successful in securing mates, it follows that the most symmetrical tails are also doing better. So Moller cut or elongated the tail feathers of certain males and at the same time enhanced or reduced the symmetry of the tails. Those with longer tails got mates sooner and reared more offspring, but within each class of length, those with enhanced symmetry did better than those with reduced symmetry."
Moller interprets this as unambiguous evidence in favor of Good-genes, for it shows that a condition-dependent trait—symmetry—is sexually selected: He joined forces with Pomiankowski to begin to separate those ornaments that show a correlation between symmetry and size from those that do not—in effect, to separate Good-genes from Fisher: Their initial conclusion was that animals with single ornaments—such as a swallow with a long tail—are Good-geners and show increasing symmetry with increasing size, whereas animals with multiple ornaments—such as a pheasant with its long tail, red facial roses, and colorful feather patterns—are mostly Fisherian, showing no relationship between size and symmetry. Since then, Pomiankowski has returned to the subject from a different angle, arguing that Fisher and many ornaments are likely to predominate when the cost to females of choosing is cheap; Good-genes will predominate when the cost of choosing is high: Again we reach the same conclusion: Peacocks are Fisherian; swallows are Good-geners.'°
HONEST JUNGLE FOWL
So far I have considered the evolution of male ornaments mainly from the female 's point of view because it is her preferences that drive that evolution. But in a species such as a peafowl, where female choice of mate rules, the male is not entirely a passive spectator of his evolutionary fate. He is both an ardent suitor and an eager salesman: He has a product to sell—his genes, perhaps—and information to impart about that product, but he does not simply
::: 154 :::
The Red Queen
hand the information over and await the peahen's decision. He is out to persuade her, to seduce her. And just as she is descended from females who made a careful choice, so he is descended from males who made a hard sell.
The analogy of the sales pitch is revealing, for advertisers do not promote their product merely by providing information about it. They fib, exaggerate, and try to associate it with pleasur-able images: They sell ice cream using sexy pictures, airplane tickets using couples walking hand in hand on beaches, instant coffee using romance, and cigarettes using cowboys.
When a man wants to seduce a woman, he does not send her a copy of his bank statement but a pearl necklace. He does not send her his doctor 's report but lets slip that he runs twenty miles a week and never gets colds. He does not tell her what degree he got but instead dazzles her with wit: He does not display testa-ments to how thoughtful he is but sends her roses on her birthday.
Each gesture has a message: I 'm rich, I'm fit, I'm clever, I 'm nice.
But the information is packaged to be more seductive and more effective, just as the message "Buy my ice cream " catches the eye when it is accompanied by a picture of two good-looking people seducing each other.
In courtship, as in the world of advertising, there is a discrepancy of interests between the buyer and the seller: The female needs to know the truth about the male: his health, wealth, and genes. The male wants to exaggerate the information: The female wants, the truth; the male wants to lie: The very word seduction implies trickery and manipulation."
Seduction therefore becomes a classic Red Queen contest, although this time the two protagonists are male and female, not host and disease: Zahavi 's handicap theory, as explored by Hamilton and Zuk, predicted that honesty would eventually prevail and males who cheat would be revealed: This is because the handicap is the female's criterion of choice for the very reason that it reveals the male 's state of health.