Solidago altissima populations consistently contain a minority of 'ducking', or 'candy-cane', stems. The goals of this study were to investigate whether these candy-cane stems may be an adaptation to resist herbivory, and to look for costs (namely, resistance tradeoffs or reduced reproduction) that might constrain the spread of the ducking trait. In this study, herbivory and seed set were recorded for 272 erect and 272 candy-cane stems in a field population of S. altissima. Candy-cane plants were twice as resistant as erect plants to two common apex-attacking gall midges, but were 26% more susceptible than erect plants to a more abundant apex-boring caterpillar. The two stem morphs were equally resistant to all other herbivores surveyed. Candy-cane plants were 11% less likely to experience apex damage and 3% more likely to set seed than erect-stemmed plants. Damaged candy-cane stems were 9% more likely to produce seeds than damaged erect stems. Although ducking is an effective means of resisting apex-galling herbivores, its spread may be constrained by susceptibility to an apex-boring caterpillar, which may enjoy enemy-reduced space on candy-cane stems. The evolution of ducking does not seem to be constrained by a reduced likelihood of sexual reproduction, or reduced tolerance of apex damage.