The robust redhorse Moxostoma robustum occurs in an 85-km stretch of the Oconee River, Georgia, downstream of a hydropower dam. The population consists primarily of older individuals and recruitment in recent years has been minimal. Operation of the hydropower dam may have affected recruitment negatively by displacing newly hatched larvae downstream and away from nursery habitats. Our null hypothesis was that larval robust redhorse can tolerate water velocities that occur in the Oconee River during peak river discharge related to hydropower generation. We measured swimming speeds for three size-classes of larvae (means: 13.1, 16.2, and 20.4 mm total length) and modeled low-velocity habitat (i.e., as defined by larval swimming speeds) in the Oconee River. We used logistic regression to calculate prolonged swimming speeds (i.e., water velocity at which 50% of fish failed to swim for 1 h) for each size-class and to predict the proportion of larvae in the water column that could maintain their position in the river. Prolonged swimming speeds were 6.9, 10.6, and 11.7 cm/s for 13.1-, 16.2-, and 20.4-mm fish, respectively. Habitat modeling suggested that low-velocity areas were present in the river and that there was not a strong relationship between low-velocity habitat and discharge. However, low-velocity habitats were dynamic during fluctuating discharge, and the ability of larval robust redhorse to access these dynamic areas is unknown. ?? Copyright by the American Fisheries Society 2000.