The use of manipulative experiments to study directly the effects of climate change on natural and managed systems is expensive if performed on a large scale; it also relies upon accurate predictions of future climatic conditions. Predictive modelling is less expensive and permits a range of scenarios to be considered but this is only as reliable as the underpinning model. Predicting the suitability of England and Wales for the cultivation of forage maize under climate change was attempted using an established crop growth model by Muchow et al. (1990) integrated with climate data. The biological and climatic inputs to the model were the thermal requirements for the stages of crop growth, leaf number, leaf area, harvest index, the daily maximum and minimum air temperature, precipitation and solar radiation. These values were obtained from a daily meteorological database for ninety-three sites in England and Wales for the period 1951-80. Model outputs as point values of potential crop yield predicting current production-'baseline'-were statistically validated using actual crop yield data collated from bibliographic analysis. The baseline results indicated that parts of the south-east appeared to be too dry and the north too cold. The model was run again using an artificial sensitivity test (temperature +2°C, precipitation ±10%). Increased precipitation led to a predicted increase in geographical suitability of the U K for forage maize production to the north and west. Under reduced precipitation there was a decrease in suitability in the south-east, possibly owing to moisture stress.