Sulphur (S) depletion of grassland soils has occurred in Europe for many decades. This is known to promote a decrease in ecosystem productivity and is suspected to alter plant community structure. Considering the strong links between nitrogen (N) and S metabolism in plants, these effects should depend on N availability. We tested this hypothesis in a pot experiment, considering a four grassland species plant mixture (three Poace ae s: Lolium perenne, Agrostis capillaris and Poa pratensis and one Fabace : Trifolium repens), and submitted it to a double N and S gradient. We used labelled N-15-fertilizer and S-34-fertilizer in order to determine both nutrient use efficiencies by each species and to analyze the influence of competition for these nutrients on plant mixture dynamics. We compared species relative physiological performance (RPP) in the monoculture and their relative ecological performance (REP) in the mixture of the four species. We analysed gradient effects at establishment and at regrowth after cutting. At establishment, grass production and S use efficiency increased along the N gradient. The S gradient slightly favoured the dominance of L. perenne, increased A. capillaris production and enhanced N use efficiency of both species. At regrowth, increased S promoted more significant effects, enhancing T. repens performance in increasing its N-2 fixation ability and maintaining this at high N. It also induced a change in grass species relative performance (dry matter production and N use efficiency) at high N, enhancing that of L. perenne and decreasing that of A. capillaris. At both establishment and regrowth, RPP did not reflect REP, meaning that species behave differently along the gradient when grown in mixture. Finally, the S gradient and the N gradient modulated relative plant species abundance. It appears that modulation of S availability could be used as a tool to drive grassland community structure.