Planktonic larvae are unlikely to be randomly distributed over time or in space. Fluctuations in larval density may result from a variety of physical and biological processes. To identify relatively high-frequency patterns of larval variation for an inner-shelf environment, zooplankton were sampled hourly for 3 d using an automated plankton pump. Moored ~80 cm above bottom, the pump was located in 8 m of water on the Outer Banks of North Carolina. Densities of all larval groups were highly variable: 5-746 m-3 for barnacle cyprids, 0-668 m-3 for polychaetes, 0-516 m-3 for bivalves, 6-414 m-3 for gastropods, and 0-86 m-3 for bryozoan cyphonautes. Moreover, maximal and submaximal peaks were nearly coincident for all taxa. One hypothesis to explain these results involves passive larval transport by wind-driven cross-shelf flows (upwelling and downwelling). Larval concentrations were persistently low in warm, downwelled water and highest during a period of cool, upwelled water. During upwelling, however, maximal and submaximal peaks for each taxon corresponded with brief relaxation events, dominated by downwelling-favorable winds. Thus, larvae tracking with cool water would have been moved upslope then offshore at the surface by upwelling currents or to the bottom by downwelling flows. A second hypothesis involves active diel vertical migration (DVM). During the cool-water period, peaks in near-bottom larval density occurred near noon, consistent with daytime descent. Daily peaks were not always separated by 24 h, however, probably due to modulation of DVM by physical processes (e.g., cross- or alongshelf advection or mixing). These relatively simple patterns of variation in larval abundance are surprising, given the complex hydrodynamic processes that typically operate on the inner shelf.