In order to parametrize a leaf submodel of a canopy level gas-exchange model, a series of photosynthesis and stomatal conductance measurements were made on leaves of white oak (Quercus alba L.) and red maple (Acer rubrum L.) in a mature deciduous forest near Oak Ridge, TN. Gas-exchange characteristics of sun leaves growing at the top of a 30 m canopy and of shade leaves growing at a depth of 3–4 m from the top of the canopy were determined. Measured rates of net photosynthesis at a leaf temperature of 30°C and saturating photosynthetic photon flux density, expressed on a leaf area basis, were significantly lower (P = 0.01; n = 8) in shade leaves (7.9μmol m−2 s−1) than in sun leaves (11–5μmol m−2 s−1). Specific leaf area increased significantly with depth in the canopy, and when photosynthesis rates were expressed on a dry mass basis, they were not significantly different for shade and sun leaves. The percentage leaf nitrogen did not vary significantly with height in the canopy; thus, rates expressed on a per unit nitrogen basis were also not significantly different in shade and sun leaves. A widely used model integrating photosynthesis and stomatal conductance was parametrized independently for sun and shade leaves, enabling us to model successfully diurnal variations in photosynthesis and evapotranspiration of both classes of leaves. Key photosynthesis model parameters were found to scale with leaf nitrogen levels. The leaf model parametrizations were then incorporated into a canopy-scale gas-exchange model that is discussed and tested in a companion paper (Baldocchi & Harley 1995, Plant, Cell and Environment 18, 1157–1173).