The annual above-ground turnover of nutrients was compared among 14 wild plant species representing four life-forms (woody evergreen, woody deciduous, graminoid, herb) and growing in two contrasting habitat types, a meadow birch forest rich in nutrients and a nutrient-poor bog. Variations in plant nutrient-turnover characteristics were related to habitat, life-form and species. The aim was to determine whether nutrient use strategy is similar for plants coexisting in the same habitat, the alternatives being that it is determined by the plant hfe-form or varies individualistically among species. The nutrient turnover characteristics are discussed in relation to nutrient use efficiency and stress resistance syndrome. Total annual nitrogen (N) and phosphorus (P) losses due to reproduction, herbivory and litter production were lowest in evergreens. There were marked differences between habitats with respect to parameters related to the stress resistance syndrome: nutrient (N and P) pool size, leaf P concentration and net above-ground productivity were much higher in the meadow birch forest than in the bog. In contrast, variation in both parameters of the nutrient use strategy, i.e. above-ground mean nutrient residence time (MRTA) and above-ground nutrient productivity (A(A)), was related to life-form and species but not to habitat conditions. We found an inverse relationship between A(A) and MRTA. Woody evergreens had the longest MRTA and the lowest A(A) of all life-forms, while herbs and graminoids were characterised by a short MRTA and a high A(A). This inverse relationship is of great importance for the nutrient use strategy of different life-forms though it is underlain with an uncertain portion of correlation due to the definitions of parameters. We found no consistent correlations between above-ground nutrient use efficiency (NUEA), as currently defined (''nutrient use strategy''), and indices of NUEA based on nutrient concentrations in green or senescent leaves. Above-ground nutrient use efficiency was not affected by habitat, life-form or species. This is in agreement with the hypothesis that by increasing their NUEA per se plants cannot adapt to nutrient-poor conditions. However, our results support the idea that in low-nutrient habitats there is a selection for species showing a prolonged mean residence time of nutrients (evergreens), while in high-nutrient habitats selection is for species with high nutrient productivity (herbs, graminoids).