Interactions between perennial ryegrass (Lolium perenne L.) and white clover (Trifolium repens L.) under contrasting nitrogen availability: productivity, seasonal patterns of species composition, N-2 fixation, N transfer and N recovery

Nitrogen (N) fertiliser and clover cultivar choice affect competition and productivity in grass-clover mixtures. Pure stands and mixtures of perennial ryegrass and white clover cultivars with contrasting growth habits were examined. The aim of this work was to study the effect of repetitive nitrogen (N) application and cultivar combination on competition and productivity, N yield in the harvested herbage, N-2 fixation in mixtures and pure stands, and transfer of N from clover to the companion grass. Large-leaved white clover cultivar Alice and small-leaved cv. Gwenda and perennial ryegrass cvs. Barlet (erect) and Heraut (prostrate) were sown in pure stands and as four binary grass-clover mixtures on a sandy soil in 1995. In the mixtures, two levels of N fertiliser were applied: 0 (-N) and 150 and 180 kg ha(-1) y(-1) N (+N) in 1996 and 1997, respectively, while the grass monocultures received three N levels (0, 140/180 and 280/360 kg ha(-1)) in 1996 and 1997, respectively. No N was applied to pure clover. The plots were cut five times during 1996 and six times during 1997. Fertiliser N was applied in early spring and after every harvest. The treatments were continued until the summer of 1999. In pure grass, the applied N was effectively recovered. In mixtures, N application affected competition by enhancing grass growth and the overall effect of N application was 17 kg DM per kg N applied in 1996. However, there was no yield response to N fertilizer in 1997, because this was compensated for by a higher clover production in unfertilised mixtures. In 1997, -N mixtures yielded more N than +N mixtures, owing to the higher clover content and N-2 fixation. Large-leaved clover cv. Alice was better able to withstand the negative effect of repetitive N application on clover production in mixtures and increased its proportion during the growing season of the second harvest year. In 1997, mixtures with Alice yielded more N than mixtures with Gwenda, but in pure clover swards, there was no cultivar effect on N yield. Also, during the autumn of 1998 and the spring of 1999, the clover content was highest in mixtures with Alice. Harvested N and apparent N-2 fixation were almost twice as high in 1997 as in 1996. N yield and apparent N-2 fixation were higher in pure clover than in mixtures. In mixtures, the apparent N-2 fixation in 1996 was 142 kg N ha(-1), irrespective of cultivar or N treatment. In 1997, it was on average 337 kg N ha(-1), and higher in -N mixtures and in mixtures with Alice. For each tonne of clover DM in the harvested herbage, 65 and 57 kg N was harvested in 1996 and 1997 in -N mixtures, respectively. The apparent transfer of clover-derived N to grass was on average 29 and 70 kg N ha(-1) yr(-1) in 1996 and 1997, respectively. It was highest in +N mixtures and highest in mixtures with Gwenda in 1997. In contrast to clover, the grass cultivars were very similar in their productivity and seasonal patterns, despite their contrasting growth habits. Seasonal trends in N yield, N transfer and N recovery are discussed in relation to fertilizer application regimes and variation in production patterns in mixtures and pure stands.