Through a field experiment, we show that a predator has negative nonconsumptive effects (NCEs) on different life-history stages of the same prey species. Shortly before the recruitment season of the barnacle Semibalanus balanoides (spring), we established experimental cages in rocky intertidal habitats in Nova Scotia, Canada. The cages were used to manipulate the presence and absence of dogwhelks, Nucella lapillus, the main predators of barnacles. At the centre of each cage, we installed a tile where barnacle pelagic larvae could settle and the resulting recruits grow. Mesh prevented caged dogwhelks from accessing the tiles, but allowed waterborne dogwhelk cues to reach the tiles. During the recruitment season, barnacle larvae settled preferentially on tiles from cages without dogwhelks. In the fall, at the end of the dogwhelk activity period and once the barnacle recruits had grown to adult size, barnacle body mass was lower in the presence of dogwhelks. This limitation may have resulted from a lower barnacle feeding activity with nearby dogwhelks, as found in a previous study. The observed larval and adult responses in barnacles are consistent with attempts to decrease predation risk. In the fall, dogwhelk cues also limited barnacle reproductive output, a possible consequence of the limited growth of barnacles. Overall, the results of this study suggest that a predator might influence trait evolution in a prey mediated by NCEs on different life-history stages.