The presence of floral nectaries is a particularly interesting example of convergent traits in flowering plants and is considered to be associated with the adaptation and diversification of numerous angiosperm lineages. For instance, the petal spurs of Aquilegia flowers are regarded as a key innovation, but the adaptive radiation of the New World Aquilegia species would have been unlikely to have occurred if there were no nectaries at the distal tips of the petal spurs to reward pollinators. To date, we have very little knowledge of what genes contribute to nectary development and evolution, particularly in non‐core eudicot taxa. In this study, we investigated the functions of genes from the STYLISH (STY) family in A. coerulea. Single and triple gene knockdown revealed that these three genes function redundantly in style and nectary development, and triple silencing led to the absence of nectary. No previous study among the core eudicot has reported STY genes to function in nectary development, suggesting the genes have been co‐opted to this role in Aquilegia. Moreover, strong expression of STY homologs in nectary‐bearing petals has also been detected in Delphinium and Epimedium, suggesting that this co‐option event is likely to have occurred before the diversification of the family Ranunculaceae and Berberidaceae. Since the identification of the first gene controlling nectary development in 1999, the STY homologs of the Ranunculaceae are the only alternative loci for the control of nectary development in flowering plants, providing a critical data point in understanding the evolutionary origin and developmental basis of nectaries.