In freshwater rainbow trout (Oncorhynchus mykiss), two common responses to acute waterborne copper (Cu) exposure are reductions in ammonia excretion and Na(+) uptake at the gills, with the latter representing the likely lethal mechanism of action for Cu in adult fish. Larval fish, however, lack a functional gill following hatch and rely predominantly on cutaneous exchange, yet represent the most Cu-sensitive life stage. It is not known if Cu toxicity in larval fish occurs via the skin or gills. The present study utilized divided chambers to assess cutaneous and branchial Cu toxicity over larval development, using disruptions in ammonia excretion (Jamm) and Na(+) uptake (Jin(Na)) as toxicological endpoints. Early in development (early; 3 days post-hatch; dph), approximately 95% of Jamm and 78% of Jin(Na) occurred cutaneously, while in the late developmental stage (late; 25 dph), the gills were the dominant site of exchange (83 and 87% of Jamm and Jin(Na), respectively). Exposure to 50 μg/l Cu led to a 49% inhibition of Jamm in the late developmental stage only, while in the early and middle developmental (mid; 17 dph) stages, Cu had no effect on Jamm. Jin(Na), however, was significantly inhibited by Cu exposure at the early (53% reduction) and late (47% reduction) stages. Inhibition at the early stage of development was mediated by a reduction in cutaneous uptake, representing the first evidence of cutaneous metal toxicity in an intact aquatic organism. The inhibitions of both Jamm and Jin(Na) in the late developmental stage occurred via a reduction in branchial exchange only. The differential responses of the skin and gills to Cu exposure suggest that the mechanisms of Jamm and Jin(Na) and/or Cu toxicity differ between these tissues. Exposure to 20μg/l Cu revealed that Jamm is the more Cu-sensitive process. The results presented here have important implications in predicting metal toxicity in larval fish. The Biotic Ligand Model (BLM) is currently used to predict metal toxicity in aquatic organisms. However, for rainbow trout this is based on gill binding constants from juvenile fish. This may not be appropriate for post-hatch larval fish where the skin is the site of toxic action of Cu. Determining Cu binding constants and lethal accumulation concentrations for both skin and gills in larval fish may aid in developing a larval fish-specific BLM. Overall, the changing site of toxic action and physiology of developing larval fish present an interesting and exciting avenue for future research.