We analysed and compared the functioning of UV-B screening pigments in plants from marine, fresh water and terrestrial ecosystems, along the evolutionary line of cyanobacteria, unicellular algae, primitive multicellular algae, charophycean algae, lichens, mosses and higher plants, including amphibious macrophytes. Lichens were also included in the study. We were interested in the following key aspects: (a) does the water column function effectively as an 'external UV-B filter'?; (b) do aquatic plants need less 'internal UV-B screening' than terrestrial plants?; (c) what role does UV screening play in protecting the various plant groups from UV-B damage, such as the formation of thymine dimers?; and (d) since early land 'plants' (such as the predecessors of present-day cyanobacteria, lichens and mosses) experienced higher UV-B fluxes than higher plants, which evolved later, are primitive aquatic and land organisms (cyanobacteria, algae, lichens, mosses) better adapted to present-day levels of UV-B than higher plants? Furthermore, polychromatic action spectra for the induction of UV screening pigments of aquatic organisms have been determined. This is relevant for translating 'physical' radiation measurements of solar UV-B into 'biological' and 'ecological' effects. From the action spectra, radiation amplification factors (RAFs) have been calculated. These action spectra allow us to determine any mitigating or antagonistic effects in the ecosystems and therefore qualify the damage prediction for the ecosystems under study. We summarize and discuss the main results based on three years of research of four European research groups. The central theme of the work was the investigation of the effectiveness of the various screening compounds from the different species studied in order to gain some perspective of the evolutionary adaptations from lower to higher plant forms. The induction of mycosporine-like amino acids (MAAs) was studied in the marine dinoflagellate Gyrodinium dorsum, the green algal species Prasiola stipitata and in the cyanobacterium Anabaena sp. While visible (400-700 nm) and long wavelength UV-A (315-400 nm) showed only a slight effect, MAAs were effectively induced by UV-B (280-315 nm). The growth of the lower land organisms studied, i.e. the lichens Cladina portentosa, Cladina foliacaea and Cladonia arbuscula, and the club moss Lycopodiumannotinum, was not significantly reduced when grown under elevated UV-B radiation (simulating 15% ozone depletion). The growth in length of the moss Tortula ruralis was reduced under elevated UV-B. Of the aquatic plants investigated the charophytes Chara aspera showed decreased longitudinal growth under elevated UV-B. In the 'aquatic higher plants' studied, Ceratophyllum demersum, Batrachium trichophyllum and Potamogeton alpinus, there was no such depressed growth with enhanced UV-B. In Chara aspera, neither MAAs nor flavonoids could be detected. Of the terrestrial higher plants studied. Fagopyrum esculentum, Deschampsia antarctica, Vicia faba, Calamagrostis epigejos and Carex arenaria, the growth of the first species was depressed with enhanced UV-B, in the second species length growth was decreased, but the shoot number was increased, and in the latter two species of a dune grassland there was no reduced growth with enhanced UV-B. In the dune grassland species studied outdoors, at least five different flavonoids appeared in shoot tissue. Some of the flavonoids in the monocot species, which were identified and quantified with HPLC, included orientin, luteolin, tricin and apigenin. A greenhouse study with Vicia faba showed that two flavonoids (aglycones) respond particularly to enhanced UV-B. Of these, quercetin is UV-B inducible and mainly located in epidermal cells, while kaempferol occurs constitutively. In addition to its UV-screening function, quercetin may also act as an antioxidant. Polychromatic action spectra were determined for induction of the UV-absorbing pigments in three photosynthetic organisms, representing very different taxonomic groups and different habitats. In ultraviolet photobiology, action spectra mainly serve two purposes: (1) identification of the molecular species involved in light absorption; and (2) calculation of radiation amplification factors for assessing the effect of ozone depletion. Radiation amplification factors (RAFs) were calculated from the action spectra. In a somewhat simplified way, RAF can be defined as the percent increase of radiation damage for a 1% depletion of the ozone layer. Central European summer conditions were used in the calculations, but it has been shown that RAF values are not critically dependent on latitude or season. If only the ultraviolet spectral region is considered, the RAF values obtained are 0.7 for the green alga Prasiola stipitata, 0.4 for the dinoflagellate Gyrodinium dorsum, and 1.0 for the cyanobacterium Anabaena sp. In the case of P. stipitata, however, the effect of visible light (PAR, photosynthetically active radiation, 400-700 nm) is sufficient to lower the RAF to about 0.4, while the PAR effect for G. dorsum is negligible. RAFs for some damage processes, such as for DNA damage (RAF=2.1 if protective effects or photorepair are not considered ), are higher than those above. Our interpretation of this is that if the ozone layer is depleted, increased damaging radiation could overrule increased synthesis of protective pigments. In addition to investigating the functional effectiveness of the different screening compounds, direct UV effects on a number of key processes were also studied in order to gain further insight into the ability of the organisms to withstand enhanced UV-B radiation. To this end, the temperature-dependent repair of cyclobutane dimers (CPD) and (6-4) photoproducts induced by enhanced UV-B was studied in Nicotiana tabacum, and the UV-B induction of CPD was studied in the lichen Cladonia arbuscula . Also, photosynthesis and motility were monitored and the response related to the potential function of the screening compounds of the specific organism. © 2002 Published by Elsevier Science B.V.