Robust redhorse, Moxostoma robustum, is a recently rediscovered, imperiled species of sucker (Catostomidae) that inhabits several rivers in the Atlantic Slope drainage and is subject to intense conservation efforts. Its spawning period frequently overlaps that of a sympatric congener, the notchlip redhorse (M. collapsum), making identifying the larvae of the species difficult. I measured various morphometrics, meristics, and developmental characteristics on lab-reared larvae of each species, fit a classification tree model to the data, and used the model to create a key discriminating between the species. The model had a leave-one-out, cross-validation expected error rate of 4.7%. The key formed from the model is highly accurate for fishes from 1020 mm total length: three independent verifiers used the key to identify fishes with a 95% accuracy rate. This key is one of a few that distinguishes between sympatric Moxostoma larvae and is the first to identify larval robust redhorse.