The evolution of fruit is a reward for animal dispersal of seeds, but the nutrients in fruit that attract dispersers also attract microbial pathogens.   Selectively deterring these microbes while retaining legitimate dispersers is thus critical for the success of fruiting plants, and it is widely believed that fruit secondary metabolites mediate this interaction. This argument is attractive because it would explain an evolutionary paradox: many wild fruits contain compounds that are so distasteful that they reduce dispersal by deterring legitimate dispersers. Yet the assertion that fruit secondary metabolites increase plant fitness by deterring microbial attack remains largely untested, as it requires explicit evaluation of both fitness benefits and costs.

Here we focus on benefits and costs of capsaicinoids, responsible for pungency or “heat” in chilies.  We use a naturally occurring polymorphism for capsaicinoid production in a South American wild chili, Capsicum chacoense to examine the selection on pungency imposed by microbial, invertebrate, and vertebrate consumers before, during, and after dispersal.  We show that the vast majority of pre-dispersal seed loss takes place due to Fusarium fungi. Fusarium exploits the foraging of Lygaeid seed bugs to gain access to seeds and fruit, and Fusarium attack of seeds causes large decreases in seed viability. Further, we show that the two primary capsaicinoids in wild chilies directly reduce Fusarium growth. Capsaicinoids thus provide chemical protection to fruit and seeds before dispersal.  Yet this chemical protection appears to trade-off with physical protection of seeds: Pungent fruits contain seeds with thinner, more fragile seed coats.  Thus seeds from pungent plants are more vulnerable to damage during dispersal, resulting in lost viability when retention times are long, and they are more vulnerable to seed predation after dispersal, as the primary predators prefer seeds with thin seed coats.