The peacock's tail should not exist. It is physiologically expensive to grow, metabolically expensive to carry, and makes the bird more conspicuous to predators. Every dimension of natural selection — Darwin's struggle for existence — predicts a smaller tail, or none. Darwin himself wrote in 1860: The sight of a feather in a peacock's tail, whenever I gaze at it, makes me sick! Twelve years later he resolved the puzzle in The Descent of Man, and Selection in Relation to Sex (1871) by proposing a second mechanism: sexual selection. Traits that reduce survival can spread if they increase mating success — and in many species, female mate choice and male-male competition produce traits whose costs Darwin's first mechanism would never tolerate.
Sexual selection works through two distinct mechanisms that often coexist. Intrasexual selection is same-sex competition — antlers, territorial fighting, the dominance hierarchies of elephant seals — and tends to produce weapons and bulk. Intersexual selection is mate choice, usually female choice on male display, and tends to produce the ornaments natural selection alone would never tolerate: peacock tails, bird-of-paradise plumage, the decorated bowers of bowerbirds, the choreographed dances of manakins. The intensity depends on the mating system. In strongly polygynous species, where a few males monopolize many females and most males never reproduce, sexual selection on males is fierce and dimorphism is extreme; in monogamous species the pressure is weaker; in sex-role reversal (pipefish, jacanas) the choosy and competitive roles flip with the parental-investment asymmetry.
The theoretical foundations split into two complementary stories. Fisher's runaway hypothesis (1930) showed that if females have any heritable preference for a male trait, the trait and the preference become genetically correlated, and both can amplify across generations — the runaway proceeds until offset by survival costs. Zahavi's handicap principle (1975) supplied the missing constraint by reading the cost as the message: extreme traits work as honest signals of underlying genetic quality precisely because only fit males can afford them. The immunocompetence handicap hypothesis refined the picture by noting that testosterone-dependent ornaments are immunosuppressive, so bright-plumage males who survive must have genuinely robust immune systems. The framework extends to humans, where moderate sexual dimorphism is consistent with mild ancestral polygyny and where Geoffrey Miller and others have argued that risk-taking, status-seeking, music, humor, and the cultural arts are partly products of sexual selection on cognition itself — empirically substantial claims, politically charged, methodologically hard to separate from purely cultural transmission.
Modern sexual-selection research uses genomic, neuroimaging, and computational tools the field's founders could not have anticipated. Mate-choice genomics has identified specific genes associated with mate preferences, including the MHC-disassortative mating pattern (preferences for partners with immunologically dissimilar major histocompatibility alleles, perhaps mediated by olfaction) found in many vertebrates including humans. Online dating platforms (Tinder, Hinge) have provided unprecedented quantitative data on revealed mate preferences, broadly consistent with sexual-selection theory but with substantial cultural variability layered on top. Same-sex sexual selection and the evolution of non-reproductive sexual behaviour, widespread across animals and long neglected after Bagemihl's Biological Exuberance, are now active research areas.