Increasing productivity of mixtures as compared to monocultures has been reported from numerous experimental studies, but so far the variable contribution of individual species to higher mixture productivity in biodiversity experiments is not well understood. To address this issue, we quantified the productivity of 60 species in monocultures and mixtures of varying species richness (2, 4, 8, 16, 60) and functional group number and composition (1, 2, 3, 4; legumes, grasses, small herbs, tall herbs) and tested how species properties are related to species performance in mixtures in the third year after sowing. We analysed monoculture biomass, plant biomass from separately grown plant individuals (=estimate of plant growth rates), and the monoculture resource-use characteristics canopy height and structure (leaf area index) as indicators for light acquisition, and soil nitrate concentration (=estimate of depletion of plant available nitrogen) and biomass:N ratios (=estimate of biomass produced per unit plant N) as indicators for nitrogen acquisition and use. High monoculture productivity was related to different combinations of resource-use characteristics. The biomass of a species and its proportional contribution to mixture biomass correlated positively with species relative yields, suggesting that highly productive mixture species were most important for an overyielding of mixtures. Although monoculture biomass was a significant predictor for species performance in mixtures except for grasses, a combination of monoculture biomass, plant growth rates and resource-use traits associated with nutrient and light acquisition explained non-legume species performance best. Legume performance was best associated with their monoculture biomass and traits associated with light acquisition. In spite of the fact that high species performance in mixtures was associated with a species' competitive ability as represented by monoculture productivity, growth rates and resource-use traits, our results suggest that species uniqueness in resource acquisition strategies increases the chance for niche differentiation among overyielding species.