A process-based leaf gas exchange model for C3 plants was developed which specifically describes the effects observed along light gradients of shifting nitrogen investment in carboxylation and bioenergetics and modified leaf thickness due to altered stacking of photosynthetic units. The model was parametrized for the late-successional, shade-tolerant deciduous species Acer saccharum Marsh. The specific activity of ribulose-1,5-bisphosphate carboxylase (Rubisco) and the maximum photosynthetic electron transport rate per unit cytochrome f (cyt f) were used as indices that vary proportionally with nitrogen investment in the capacities for carboxylation and electron transport. Rubisco and cyt f per unit leaf area are related in the model to leaf dry mass per area (MA), leaf nitrogen content per unit leaf dry mass (Nm), and partitioning coefficients for leaf nitrogen in Rubisco (PR) and in bioenergetics (PB). These partitioning coefficients are estimated from characteristic response curves of photosynthesis along with information on lear structure and composition. While PR and PB determine the light-saturated value of photosynthesis, the fraction of leaf nitrogen in thylakoid light-harvesting components (PL) and the ratio of leaf chlorophyll to leaf nitrogen invested in light harvesting (CB), which is dependent on thylakoid stoichiometry, determine the initial photosynthetic light utilization efficiency in the model. Carbon loss due to mitochondrial respiration, which also changes along light gradients, was considered to vary in proportion with carboxylation capacity. Key model parameters - Nm, PR, PB, PLCB and stomatal sensitivity with respect to changes in net photosynthesis (Gr) – were examined as a function of MA, which is linearly related to irradiance during growth of the leaves. The results of the analysis applied to A. saccharum indicate that PB and PR increase, and Gf, PL and CB decrease with increasing MA. As a result of these effects of irradiaiice on nitrogen partitioning, the slope of the light-saturated net photosynthesis rate per unit leaf dry mass (Ammax) versus Nm relationship increased with increasing growth irradiance in mid-season. Furthermore, the nitrogen partitioning coefficients as well as the slopes of Ammax versus Nm were independent of season, except during development of the leaf photosynthetic apparatus. Simulations revealed that the acclimation to high light increased Ammax by 40% with respect to the low light regime. However, light-saturated photosynthesis per leaf area (Aamax) varied 3-fold between these habitats, suggesting that the acclimation to high light was dominated by adjustments in leaf anatomy (Aamax=AmmaxMA) rather than in foliar biochemistry. This differed from adaptation to low light, where the alterations in foliar biochemistry were predicted to be at least as important as anatomical modifications. Due to the light-related accumulation of photosynthetic mass per unit area, Aamax depended on MA and leaf nitrogen per unit area (Na). However, Na conceals the variation in both MA and Nm (Na=NmMA), and prevents clear separation of anatomical adjustments in foliage structure and biochemical modifications in foliar composition. Given the large seasonal and site nutrient availability-related variation in Nm, and the influences of growth irradiance on nitrogen partitioning, the relationship between Aamax and Na is universal neither in time nor in space and in natural canopies at mid-season is mostly driven by variability in MA. Thus, we conclude that analyses of the effects of nitrogen investments on potential carbon acquisition should use mass-based rather than area-based expressions.