Effects of coexistence on the performance of white clover-perennial ryegrass mixtures.

Yields from five white clover/perennial ryegrass mixtures, each based on coexisting components, were compared with those produced by ten mixtures whose constituents had no previous history of coexistence. The former group consisted of five paired clover/ryegrass populations, four collected from natural pastures and one from a sown pasture, whilst the latter group comprised the same five clover populations grown in turn with each of two ryegrass companions.In the first harvest year, when artificial fertilizer plus soil mineralization supplied most of the nitrogen to this experiment, the grasses were generally favoured at the expense of the clovers. Although those mixtures based on coexisting populations had significantly lower yields the clover within these mixtures made a significantly higher contribution to their productivity. This advantage was retained into the second harvest year when it was accompanied by a correlated response in the grasses, which by then may have become dependent upon the clover for their nitrogen supply. Consequently, by the end of the second harvest year those mixtures based on coexisting populations yielded over 20% more on average, due mainly to improved early season growth. Despite considerable differences in productivity, all five clover populations gave their highest yields when grown with their matching grass.