The effect of elevated [CO2] on the productivity of spring wheat, winter wheat and faba bean was studied in experiments in climatized crop enclosures in the Wageningen Rhizolab in 1991–93. Simulation models for crop growth were used to explore possible causes for the observed differences in the CO2 response. Measurements of the canopy gas exchange (CO2 and water vapour) were made continuously from emergence until harvest. At an external [CO2] of 700 μmol mol−1 Maximum Canopy CO2 Exchange Rate (CCERmax) at canopy closure was stimulated by 51% for spring wheat and by 71% for faba bean. At the end of the growing season, above ground biomass increase at 700 μmol mol−1 was 58% (faba bean), 35% (spring wheat) and 19% (winter wheat) and the harvest index did not change. For model exploration, weather data sets for the period 1975-88 and 1991–93 were used, assuming adequate water supply and [CO2] at 350 and 700 μmol mol−1. For spring wheat the simulated responses (35–50%) were at the upper end of the experimental results. In agreement with experiments, simulations showed smaller responses for winter wheat and larger responses for faba bean. Further model explorations showed that this differential effect in the CO2 response may not be primarily due to fundamental physiological differences between the crops, but may be at least partly due to differences in the daily air temperatures during comparable stages of growth of these crops. Simulations also showed that variations between years in CO2 response can be largely explained by differences in weather conditions (especially temperature) between growing seasons.