Papadopoulos, Y. A., McElroy, M. S., Fillmore, S. A. E., McRae, K. B., Duyinsveld, J. L. and Fredeen, A. H. 2012. Sward complexity and grass species composition affects the performance of grass-white clover pasture mixtures. Can. J. Plant Sci. 92: 1199–1205. The productivity of managed permanent pastures is closely associated with the species composition of seeded mixtures. Ecological theory suggests that increasing plant species' diversity will result in higher productivity, resilience, and resistance to invasive species. To better understand the relationship between sward species composition and pasture productivity, mixtures of four common pasture grass species, timothy (Phleum pratense L.), Kentucky bluegrass (Poa pratensis L.), reed canarygrass (Phalaris arundinacea L.), and meadow fescue (Festuca pratensis L.), were seeded in 2004 in binary (two-grass), tertiary (three-grass), and quaternary (four-grass) combinations with white clover (Trifolium repens L.). Plots were rotationally grazed for 5 yr, with yield determined in the first post-establishment year (2005) and in 3 subsequent production years (2007, 2008, and 2009). Mean dry matter yield (DMY) increased appreciably from post-establishment (3801 kg ha−1) to the production years (6613 kg ha−1). Contrasts revealed significantly higher DMY production in quaternary mixtures versus less complex plots in production years. Repeated measure analysis found significant quadratic trends in DMY for plots containing bluegrass and timothy, showing different patterns of growth between the respective mixtures. Principal component analysis (PCA) of averaged yields over the production years revealed a strong association between seeded grass growth and DMY. This relationship between DMY and seeded grass growth was strongest in swards containing a combination of timothy and bluegrass. In general, the growth of unseeded grasses and forbs increased in mixtures with large proportions of timothy and reed canarygrass. In spite of the fact that mixtures containing timothy and bluegrass were shown to produce high DMY, it appears the aggressiveness of bluegrass suppressed the yield potential of timothy. White clover yields and unseeded grasses/forbs were both significantly lower in mixtures containing bluegrass. Also, the PCA revealed a favourable compatibility between meadow fescue and white clover growth, while bluegrass appears to suppress meadow fescue growth. Results show that species complexity increases sward productivity over the long term, and that the presence of two grass species, in particular (timothy and bluegrass), has considerable influence on DMY.