The CO2 assimilation of primary foliage of red maple (Acer rubrum L.) and red oak (Quercus rubra L.), and of regrowth foliage produced in response to simulated insect defoliation, was measured throughout the season by infrared gas analysis: parallel measurements of leaf conductance were obtained by ventilated diffusion porometry. The rate of net photosynthesis, measured at a quantum flux density of 1,150 μmol m-2s-1, of primary foliage of both species increased from slightly negative values to about 5 μmol m-2s-1 by early June. Thereafter the rate of photosynthesis of maple slowly declined to about 4 μmol m-2s-1 before onset of a senescent decline in early September, while that of oak slowly increased to about 8 μmol m-2s-1 before onset of senescence. Manual defoliation to simulate insect attack in mid-June elicited refoliation proportional to the severity of defoliation in early July. After 100% defoliation, fully expanded regrowth foliage of maple, but not of oak, had a rate of net photosynthesis from mid-July through September that was about 50% higher than in the primary foliage of undefoliated trees. A 30 to 60% enhancement of photosynthesis of residual primary foliage remaining on 50 and 75% defoliated trees during July was also observed. The seasonal patterns of CO2 exchange of primary and regrowth foliage, and the enhancement of CO2 assimilation in residual foliage, was paralleled by similar changes in leaf conductance to water vapour. Carbon budgets of leaf canopies of each species showed that the net assimilation of the leaf canopy of both species ranged from 19 to 67% more than what would have been expected solely from replacement of leaf area. This response was greater in maple than in oak, presumably a reflection of the high rate of CO2 assimilation of regrowth maple foliage compared with that of the undefoliated control in maple. The increased CO2 assimilation of regrowth maple foliage and the increases in CO2 assimilation of residual primary foliage after defoliation offer evidence that heretofore unanticipated physiological mechanisms may be important to perennial species coping with herbivory.