Sniffing out patterns of sexual dimorphism in floral scent

Summary 1.  A major transition in flowering plants has been the evolution of separate sexes (dioecy) from combined sexes (hermaphroditism). This transition is often, but not always, accompanied by the evolution of sexual dimorphism in attractive traits, and floral scent is no exception. 2.  In this review I aim to improve our understanding of variation in sexual dimorphism in floral scent characteristics by first explicating the relevant hypotheses, and then deriving explicit predictions for the pattern of floral scent from each. 3.  Next, I synthesize and qualitatively review published data on floral volatile emission rate and composition in 33 gender dimorphic species to identify emerging patterns, and evaluate these in light of predictions derived from the hypotheses. 4.  Although conclusions must be viewed as preliminary, a handful of strong patterns were revealed: (1) in the majority of the species studied males emit more volatiles per flower than females, (2) in over half of the species studied the sexes differed in at least one aspect of scent composition, (3) sexual dimorphism in scent composition was less common in species with rewardless females and/or brood site pollination than those that offered nectar and/or pollen, (4) a one‐to‐one correspondence between sex differences in overall scent composition and male organ‐specific scent production is largely not found. 5.  This review has highlighted gaps in our understanding of the genesis of patterns of sexual dimorphism in floral scent, and makes clear that to move the field forward we need to shift our focus from pattern to process, and this will be best achieved by simultaneously testing alternative hypotheses at the same level of analysis.