Seedlings of three poplar clones (I-488, Rimini, and D-64) grown in plastic pots were submitted to three water regimes: (1) irrigated and well-drained control (T); (2) flooded with running water (Hr = 6 mg center dot L-1 of O-2); and (3) flooded with stagnant water (Hs = 2 mg center dot L-1 of O-2). A permanent water table was maintained at 5 cm above the soil surface for 2 months. The Hr and Hs treatments simulate the conditions of hydromorphic soils chosen for planting poplar in Tunisia where the attempts failed because of the lack of knowledge relating to the degree of tolerance of this species to waterlogging. Waterlogging significantly reduced growth (leaf initiation was inhibited, root decay and early leaf abscission occurred) and modified photosynthetic activity (stomatal closure and reduction of CO2 net assimilation rates). Intracellular CO2 values did not significantly differ between treatments (T, Hr, and Hs), indicating that both stomatal and nonstomatal limitations could be responsible for reducing CO2 net assimilation. However, the degree of disruption varied according to the treatment. Stagnant regime Hs had negative and irreversible effects on CO2 net assimilation rates and internal CO2 conductance for the full length of the treatments with stomatal closure after 40 days in Rimini and D-64 clones. The Hr treatment produced limited reactions; the development of hypertrophied lenticels and adventitious roots alleviated the negative effects of waterlogging with a distinct recuperation of net photosynthesis and CO2 internal conductance particularly for clone I-488. In the same way, the development of root adaptations was greater for I-488 than Rimini and D-64 clones, probably indicating a higher tolerance to flooding in the former clone. This approach could be used to select the clones intended for afforestation of the nonsaline hydromorphic areas.