The impact of dispersing juveniles on population and community dynamics is a topic of considerable debate, particularly for marine species with broadly dispersing larvae. We argue that sampling problems continue to pose a serious impediment to progress. The root of the problem is the large variation in dispersal dynamics over space and time. For short-distance dispersal the variance in dispersal has been largely ignored. By contrast, for long-distance dispersal the variance in dispersal may have been overemphasized due to the inappropriate time scales of sampling. We present data for an intertidal barnacle, Semibalanus balanoides, to contrast conclusions reached from standard sampling of long-distance juvenile dispersal with those reached by techniques that more appropriately integrate the dynamics of dispersers over time. The standard sampling approach misrepresents the true juvenile dynamics.