A glasshouse for the study of the effects of ambient levels of air pollution on plant growth is described. Two sealed sub-sections were force-ventilated, one received SO2-polluted ambient air while a second received air that had been passed through an activated carbon filter to remove SO2. Potential diffuse and direct incoming radiation, air temperature and humidity were examined in both chambers in order to determine whether differences in microclimatic factors other than air quality existed between the two ail treatments. Daily mean concentrations of SO2 and suspended particulate matter were measured in the two chambers and outside the glasshouse. During 2 years of continuous monitoring neither pollutant was detected in the filtered air while levels in the unfiltered section were usually below 100 μg m−3SO2 and 50 μg m−3 particulates. Discrepancies between the two chambers in the predicted light conditions and humidity and differences in environmental conditions between the glass enclosures and the open air are discussed critically. The possible occurrence of secondary gaseous pollutants in the unfiltered air in addition to SO2 is also considered.