Juvenile queen conch (Strombus gigas L.) occur in discrete aggregations within seemingly uniform seagrass beds throughout the Exuma Cays, Bahamas, suggesting that the aggregations occupy ecologically unique sectors of the habitat or that conch gain fitness by aggregated distribution. To examine the structure of a juvenile aggregation and to determine the underlying mechanisms which affect juvenile conch distribution, we axamined density, size composition, growth, survivorship, and movement patterns within a typical tidal-flow field nursery over a 14 mo period (August 1989 to September 1990). At the beginning of the study in August 1989, the conch population occupied 16.7 ha, with densities>0.2 juvenile conch m-2. The aggregation formed an ellipse, with longitudinal axis parallel to the main axis of the tidal current. Surveys conducted every 2 mo showed that conch density in the aggregation center remained constant while all other zones had lower densities which varied with time. In areas of high population density within the aggregation, several mass migrations of juveniles (20 to 99 conch m-2) occurred in early 1990. Tagged juveniles transplanted to zones outside the aggregation had high growth rates but suffered higher losses than individuals transplanted to the aggregation center. A tethering experiment confirmed the hypothesis that predator-induced mortality is significantly higher outside than inside the conch aggregation. Our results suggest that the queen conch aggregation occupied only a portion of the habitat that is optimal for feeding and growth. Aggregations could be maintained by differential mortality over a site; however, predation rates are probably density-dependent. Gregariousness, observed in translocation experiments, may provide an active mechanism for maintaining aggregated distribution and reducing mortality in conch nurseries. The ecological significance of aggregations should be considered in fisheries management and stock enhancement programs with queen conch.