When a trait comes into existence largely due to female selection rather than utility. The basis for this theory is that females are adapted select mates who possess traits that enable their children to thrive. And when a trait appears that enables survival, females will begin to select for it. As females begin to select for it, the female preference for it becomes more pronounced via the Baldwin Effect.
Where this process can ‘run away’ is when the preference for the trait persists even once the male population achieves maximum practical utility through the trait. If a male bird for example, benefits from a longer tail, females will begin to select for the longer tail. Eventually, the tail will evolve to a point where it has achieved the physically, practically most effective length. But the female preference for longer tails does not disappear simply because the tail has reached the practically optimal length. So the female birds continue select for longer tails even though their “ideal” mates are now hindered by their tales being too long. At this point, one would say the sexual selection process has ‘run away’.
“Sir Ronald Fisher had suggested then that females need no better reason for preferring long tails than that other females also prefer long tails. At first such logic sounds suspiciously circular, but that is its beauty. Once most females are choosing to mate with some males rather than others and are using tail length as the criterion — a big once, granted, but we’ll return to that –then any female who bucks the trend and chooses a short-tailed male will have short-tailed sons. (This presumes that the sons inherit their father’s short tail.) All the other females are looking for long-tailed males, so those short-tailed sons will not have much success. At this point, choosing long-tailed males need be no more than an arbitrary fashion; it is still despotic. Each peahen is on a treadmill and dare not jump off lest she condemn her son to celibacy. The result is that the females’ arbitrary preferences have saddled the males of their species with ever more grotesque encumbrances. Even when those encumbrances themselves threaten the life of the male, the process can continue — as long as the threat to his life is smaller than the enhancement of his breeding success. In Fisher’s words: ‘The two characteristics affected by such a process, namely plumage development in the male and sexual preference in the female, must thus advance together, and so long as the process is unchecked by severe counterselection, will advance with ever-increasing speed.” (139, The Red Queen)
“Arbitrary ornaments can grow elaborate for no reason other reason than that females discriminate between males and end up following arbitrary fashions: and the more they discriminate, the more elaborate the ornaments become. What Fisher said in 1930 was right, but it left a lot of naturalists unconvinced for two reasons. First, Fisher assumed part of what he set out to prove: That females are already choosy is crucial to the theory. Fisher himself had an answer for this, which was that initially females chose long-tailed males for more utilitarian reasons — for example, that it indicated their superior size or vigor. This is not a foolish idea; after all, even the most monogamous species, in which every male wins a female (such as terns), are choosy. But it is an idea borrowed from the enemy camp. And the Good-geners can reply: ‘If you admit that our idea works initially, why rule it out later on?’
The second reason is more mundane. Proving the Fisher’s runaway selection could happen and the ornament get bigger with ever-increasing speed does not prove that it does not happen. Computers are not the real world. Nothing could satisfy the naturalists but an experiment, one demonstrating that sexiness of sons drove the evolution of an ornament
One other piece of evidence seems to weigh in the balance on the side of Fisher — the phenomenon of copying. If you watch a lek carefully, you see that females often do not make up their own minds individually; they follow one another. Sage grouse hens are more likely to mate with a cock who has just mated with another hen. In black grouse, which is also lek, the cocks tends to mate several times in a row if at all. A stuffed black female grouse (known in this species as a greyhen) placed in the male’s territory tends to draw other females to that territory — though not necessarily causing them to mate. In guppy fish, females that have been allowed to see two males, one of which is already courting a female, subsequently prefer that male to the other even if the female that was being courted is no longer present
Such copying is just what you would expect if Fisher was right because it is fashion-following for its own sake. It hardly matters whether the male chosen is the ‘best’ male; what counts is that he is the most fashionable, as his sons will be.” (145-6, The Red Queen)
The Red Queen