Tparla / inaturalist, CC BY-NC“Love makes fools of all of us,” wrote 19th-century novelist William Makepeace Thackeray. A moment spent watching the pigeons at your local park suggests he was right: males with puffed-up, shimmering necks hop, pirouette, coo, and bow to capture the attention of unimpressed females. But why do male pigeons express themselves through the passion of dance? The concept of sexual selection, first proposed by Charles Darwin, suggests that through extravagant displays, males give females information to identify their suitability as a partner. Females who choose the fittest and most well-coordinated dancers should produce fit and well-coordinated offspring.But this seems to raise a paradox. If females are only interested in the best dancers, then evolution should converge on a single optimal dance. Yet courtship displays, even among closely related species, are extraordinarily diverse.So, why hasn’t evolution danced itself into a corner? We set out to answer this question in our new study published in the journal Behavioural Ecology.Australia’s dancing dune fliesWe turned to an unlikely subject: flies that dance along beach dunes in eastern Australia (Apotropina ornatipennis). These tiny creatures, with patterned wings and reflective patches, perform courtship displays that rival some of the most complex dances in the animal kingdom. Our study was the first to characterise their courtship choreography of twists, turns and flicking wings. The courtship display of the dancing dune fly Apotropina ornatipennis. These dancing flies gave us an opportunity that many more conspicuous species don’t: distinctly isolated populations. Because they live on discrete stretches of coastline, separated by headlands and estuaries, populations have been evolving independently for generations. If evolution has room to wiggle through dance, we expect these isolated populations to develop different dance routines, the same way regional dialects emerge in human language and birdsong. We studied both the genetics and behaviour of these populations, mapping their 41 different dance moves and comparing their dances against their degree of genetic divergence.The results were surprising. Even when populations were clearly separated, their dance routines stayed consistent. Among all the moves in their repertoire, only a subtle change in the timing of one wing movement hinted at any divergence at all. The dancing dune fly (Apotropina ornatipennis). A male (right) following a female (left). Nathan Butterworth, CC BY-NC Honest signals and the cost of improvisingThis consistency suggests that males who try to invent new choreography pay a high cost: females might simply ignore them.A courtship display works best for females if it reflects the quality of the performer – what biologists call an “honest signal”. A physically demanding routine that requires precise execution should separate high-quality males from poor ones. So females don’t mind stale moves, as long as those moves provide proof of a male’s fitness. Rearranging the choreography can be risky if it departs from what females consider to be honest indicators of male quality. A male who deviates from the established routine might be performing in a language the female hasn’t learned or signalling that he hasn’t mastered the language everyone else speaks. This cost of innovation may explain why evolutionary changes to courtship dances are often minor improvisations, and why larger changes may only occur over long evolutionary timescales. The courtship display of the Western Parotia bird-of-paradise (Parotia sefilata). Ben Tsai, inaturalist/bentsai, CC BY-NC How do dances evolve?Courtship displays are not frozen in time. Behaviours can emerge or be abandoned under intense evolutionary pressure. A striking example comes from Hawaii, where a parasitic fly that hunts crickets by eavesdropping on their courtship songs invaded the islands. Within just 20 generations, some male crickets found a new strategy for reproductive success: abandoning their instruments and piggybacking on the efforts of other males that were foolish enough to keep singing. Often, genetic change is the origin of new behaviours. In many species, courtship behaviours are hardwired in the genome. In fruit flies, males of one species are born with the desire to vomit up nuptial “gifts” as part of their courtship ritual. Researchers identified the gene responsible for the vomiting behaviour, and when they triggered it in a different species that species also began vomiting up gifts. The courtship display of the banksia peacock spider (Maratus mungaich). Kerry Stuart, kerrystuart/inaturalist, CC BY-NC Social learning is another way displays might evolve, such as in lyrebirds and songbirds, where juveniles can learn by watching older individuals. In such scenarios, cultural drift can gradually reshape courtship over time. Small novelties creep in, other males copy them, females learn to prefer the new moves, and the dance slowly changes.Ultimately, when it comes to Dancing with the Flies, the judge’s panel is all female. No matter how fit a male is, a novel dance can only succeed if females find it appealing and if their daughters inherit or copy the preference. Fashionable dance in the animal kingdom is dictated largely by the selective momentum of females. A waltz over evolutionary timescalesThe remarkably diverse dance routines we see in pigeons, peacock spiders, and flies are all snapshots of an ancient and ongoing negotiation between the sexes. More than 150 years since Darwin introduced the idea of sexual selection, we are just beginning to unravel the complexities of courtship choreography. What our research adds is a sense of just how stable these routines might be, and how the rigorous aesthetic standards set by females may cause choreography to change far more slowly than previously thought.The authors do not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and have disclosed no relevant affiliations beyond their academic appointment.