Elaborate morphological ornaments can evolve if they increase the reproductive success of the bearer during competition for mates. However, ornament evolution is incredibly rare in females, and the type and intensity of selection required to develop female-specific ornamentation is poorly understood. The main goals of my thesis are to clarify the relationship between the type and intensity of sexual selection that drives the evolution of female ornamentation, and investigate alternative hypotheses that might be limiting or contributing to the development of female ornaments. I investigated the ecology and evolution of female-specific ornaments within and between species of dance flies from the subfamily Empidinae (Diptera: Empididae). The dance flies display incredible mating system diversity including those with elaborate female-specific ornaments, lek-like mating swarms, aerial copulation and nuptial gift giving. To elucidate the form of sexual selection involved in female-ornament evolution, I experimentally investigated the role of sexual conflict in the evolution of multiple female- specific ornaments in the species Rhamphomyia longicauda. Through manipulative field experiments, I found that variation in the attractiveness of two ornaments displayed by females indicates that sexual conflict, causing a coevolutionary arms race, is an important force in the evolution of multiple extravagant female ornaments. Using R. longicauda again, I tested for a role of functional load-lifting constraints on the aerial mating ability of males who paired with females displaying multiple large ornaments. I found no evidence of functional constraints influencing the mating opportunities of elaborately ornate females, but instead discovered a relationship consistent with positive assortative mating for mass. Biased sex ratios are predicted to increase the intensity of sexual selection in a population, which in turn, is predicted to influence the evolution of ornamentation. I measured the incidence and prevalence of vertically transmitted symbiotic bacteria that has been observed to distort the sex ratio in other Dipteran hosts. While my survey revealed that symbionts occur at high incidence and variable prevalence across dance fly hosts, I found no effect of symbiont infection levels on population sex ratios, or female- specific ornament evolution. Further investigation into the relationship between sex ratios and female-ornament evolution using the comparative method revealed that the operational sex ratio (OSR) of a population did not predict continuous measures of female ornamentation across species. However, female-ornament evolution did predict male relative testis investment across species indicating that female ornaments likely indicate increased levels of polyandry. My thesis reveals that sexual selection theory developed to describe male-specific ornament evolution cannot readily be translated to apply to females. I show that male mate choice, rather than functional constraints or ecological associations with bacteria, is likely driving the evolution of female-specific ornaments. I also identify sexual conflict as an important selective force in the evolution of female-specific ornaments.