## Abstract

Research into the effects of long-term SO<latex>$_2$</latex> exposures on crop growth in the absence of visible foliar injury has progressed to a point where the data are being examined for (a) threshold concentrations that avoid damage and (b) dose-response relations that assist cost-benefit analysis of emission control. The data from 125 exposures of 21 crops to constant concentrations of SO<latex>$_2$</latex> alone in chambers have been analysed with a view to identifying threshold values and dose-response relations. The most appropriate functional form for the dose-response relation was a linear regression between SO<latex>$_2$</latex> concentration and percentage yield loss. The correlation was not improved by a number of transformations of either variable, nor by normalizing for duration of exposure. Exclusion of the studies with chambers using low airflows did not significantly alter the regression equations, but did result in a substantial improvement in the correlation between SO<latex>$_2$</latex> and yield loss. This analysis confirms that there is considerable variability in response between species and both between and within different exposure chambers. Reasons for the differences between chambers are discussed with reference to recent information on the effect of restricted SO<latex>$_2$</latex> flux, growth rate, plant age and other stresses on responses to long-term SO<latex>$_2$</latex> exposures. Recent research has focused on the interactions between SO<latex>$_2$</latex> exposure and other stresses particularly pests and diseases, frost injury, wind, frequent cutting, nutrient deficiency and possibly drought. The effects of long-term exposures to SO<latex>$_2$</latex> in outdoor chambers were greatest over winter when the metabolic capacity to de-toxify SO<latex>$_2$</latex> absorbed during periods of slow growth was limited by low temperatures and low irradiance. The lower sensitivity of crops to SO<latex>$_2$</latex> under optimal conditions of indoor chambers may also be because of the reduced intensity of other stresses. Dose-response relations used in recent cost-benefit analysis of sulphur oxide emission control in Europe predict much greater economic losses than the regression equations developed in this paper. However, the relevance of dose-response relations developed from chamber studies with constant levels of SO<latex>$_2$</latex> alone will also depend upon (a) the effects of fluctuating concentrations in the field and (b) the effects of pollutant mixtures rather than SO<latex>$_2$</latex> alone. It should soon be possible to develop improved dose-response relations solely from studies carried out in outdoor chambers over the complete annual cropping cycle. Many of the problems inherent in chamber studies on cereals in particular, will be avoided by the recent developments in field fumigation systems.