Because biological and physical processes alter the stable isotopic composition of atmospheric CO2, variations in isotopic content can be used to investigate those processes. Isotopic flux measurements of 13CO2 above terrestrial ecosystems can potentially be used to separate net ecosystem CO2 exchange (NEE) into its component fluxes, net photosynthetic assimilation (FA) and ecosystem respiration (FR). In this paper theory is developed to partition measured NEE into FA and FR, using measurements of fluxes of CO2 and 13CO2, and isotopic composition of respired CO2 and forest air. The theory is then applied to fluxes measured (or estimated, for 13CO2) in a temperate deciduous forest in eastern Tennessee (Walker Branch Watershed). It appears that there is indeed enough additional information in 13CO2 fluxes to partition NEE into its photosynthetic and respiratory components. Diurnal patterns in FA and FR were obtained, which are consistent in magnitude and shape with patterns obtained from NEE measurements and an exponential regression between night-time NEE and temperature (a standard technique which provides alternate estimates of FR and FA). The light response curve for photosynthesis (FA vs. PAR) was weakly nonlinear, indicating potential for saturation at high light intensities. Assimilation-weighted discrimination against 13CO2 for this forest during July 1999 was 16.8–17.1‰, depending on canopy conductance. The greatest uncertainties in this approach lie in the evaluation of canopy conductance and its effect on whole-canopy photosynthetic discrimination, and thus the indirect methods used to estimate isotopic fluxes. Direct eddy covariance measurements of 13CO2 flux are needed to assess the validity of the assumptions used and provide defensible isotope-based estimates of the component fluxes of net ecosystem exchange.