To assess adverse effects of ambient ozone on injury and growth, EDU (ethylenediurea) and non-EDU-treated plants of subterranean clover (Trifolium subterraneum cv. Geraldton) were exposed to ambient air at four rural sites in the Netherlands. In each of two successive experiments of eight weeks during the 1994 growing season, an intermediate harvest (cutting) was performed after four weeks of exposure. Concentrations of ambient ozone were high in Experiment 1 (June 21–August 16) and relatively low in Experiment 2 (August 23–October 18). Subterranean clover displayed characteristic ozone injury at all sites in Experiment 1 and in Experiment 2 until the intermediate harvest (September 20). The proportion of leaves injured in non-EDU-treated plants was higher than that in plants treated with the antioxidant. Both the degree of injury and the protective effect by EDU differed between sites and were not related to the measured ozone levels. The proposed short-term critical levels for injury development by ozone would not have protected clover against ozone injury in Experiment 2. The site-dependent pattern of injury in non-EDU-treated plants was similar for the various harvests and experiments and a maximal degree of 23% injury was observed after four weeks of exposure. The protective effect by EDU was related positively to the injury intensity in non-EDU-treated plants, irrespective of site, harvest and experiment. EDU caused variable effects on leaf biomass production under conditions of low ozone in Experiment 2 whereas EDU generally increased leaf biomass when the plants were exposed to the relatively high concentrations in Experiment 1. The EDU-induced increase in leaf biomass production in this experiment did not vary between sites and was on average 5–7%, depending on the harvest. The involvement of elevated concentrations of ambient ozone measured at the four sites in the Netherlands in 1994 in the observed increase of leaf biomass by EDU is discussed.