We wounded Pocillopora verrucosa to simulate injury caused by fish corallivores, and then evaluated impacts of this damage on coral performance under different temperatures (26.6 and 29.6 degrees C) and flow speeds (6 and 21 cm s(-1)) in microcosms. Colony growth (weight), photosynthetic efficiency (maximum dark-adapted quantum yield of PSII, F-v/F-m [where PSII is Photosystem II, F-v variable fluorescence and F-m maximum fluorescence yield in the dark]), and the healing of lesions were measured during 2 replicate 10 d trials. Injury caused growth to increase in 3 of 4 treatments: high flow, high temperature; low flow, high temperature; and low flow, low temperature. However, growth was greatest for uninjured corals in the high-flow, low temperature treatment, which appears to provide optimal conditions for P. verrucosa (when not injured). Temperature alone had little effect on growth but influenced F-v/F-m, as did flow; F-v/F-m was 5% greater at 26.6 degrees C than 29.6 degrees C, and 3% higher at 21 cm s(-1) than 6 cm s(-1). Injury had little effect on F-v/F-m, and neither temperature nor flow affected the rate of healing that occurred at 17 to 25% of the lesion area in 10 d. Results from a field experiment, in which growth of P. verrucosa was tested as a function of flow speed (similar to 14 cm s(-1) versus similar to 3 cm s(-1)) and fish predation (predators versus no predators), but not temperature, supported results of the microcosm experiment. Growth was greatest for corals in the high-flow, no predator treatment, and relatively high for injured corals in low flow. Together, these results suggest that P. verrucosa, a common branching coral, prioritizes overall growth over repair when injured by fish feeding, which differs from the outcome observed in a companion study in which juvenile colonies of massive Porites were subjected to similar injuries.