SUMMARYOzone concentrations in many parts of Europe exceed the proposed UN-ECE Critical Levels for the protection of ecosystems but there is so little known about the effects on native, herbaceous species, that it is presently impossible to judge what effect, if any, these concentrations are having on this group of plants. This paper reports the results of an initial screening of 32 taxa that was used to select a small number of species for further work. All were exposed to 70 nl 1"^ Og given for 7 h d"' for two weeks and the response measured in terms of nnean relative growth rate {R), the allometric root/shoot coefficient {K), and changes in induced chlorophyll fluorescence (/^).Only six species showed visible symptoms but 14 showed a significant reduction in R. There was no relationship between the extent of visible symptoms and growth reduction, confirming that visible symptoms are often of little value in determining growth sensitivity to O3 stress. There were marked differences within genera and within species; a population of Plantago major collected in Athens showed no effects, while a Derbyshire population, with a 24 % reduction in R, was the most sensitive of taxa tested, including Bel-W3 tobacco which has been specifically hred for O3 sensitivity.There was a significant regression of R\ change on the R of controls {R% = 12-4-104 x ^, P = 0-0007, r = -0-56), implying that > 30°(, of the variation in response between taxa was related to the inherent R; faster growing taxa tending to show a greater reduction in R on exposure to O3. It is tentatively suggested that this may be related to differences in stomatal conductance.Ozone caused a significant shift in K" in 14 taxa, with many of the biggest changes being in grasses. Although O3 reduced K in some taxa, contrary to several published accounts, it increased it in others, and in some there was an effect on K with no effect on R. Possible reasons for this are discussed.Chlorophyll fluorescence was examined as a tool for detecting Oj stress. The maximum rate of rise of induced fluorescence, Fj^, was very sensitive to O3 but showed a complex pattern of response dependent on species and time. Most taxa showed a significant change in Fj^, even when there was no effect of the O3 on R or K. Whilst the sensitivity of F^ makes it a useful method for in vivo detection of O3 stress, the complexity of the response makes it very difficult to quantify and use for comparisons between species.