WHITE CLOVER CONTRIBUTION TO CONTINUOUSLY STOCKED SHEEP PASTURES IN ASSOCIATION WITH CONTRASTING PERENNIAL RYEGRASSES

White clover content and herbage production in grass/clover swards wer e studied at contrasting altitudes in Wales. The lowland experiment (1 ) compared the performance of three clover varieties grown with four d iploid perennial ryegrass (Lolium perenne) varieties with a range of h eading dates. In the upland experiment (2) the productivity and persis tence of Aberystwyth S184 (small-leaved) clover was studied in associa tion with three perennial ryegrasses. In experiment 1, clover content and dry matter production with Aurora (very early) ryegrass was superi or to that with Frances (early), Talbot (intermediate) and Melle (late ). Furthermore, the performance of Kent wild white clover (small-leave d) was better than that of medium-leaved Menna and Grasslands Huia. Av eraged over two years (1985 and 1986) mean clover content of Kent was 22% compared with 9% and 10% of Menna and Huia. Performance of S184 cl over during the five years (1986-90) of experiment 2 was better with M eltra (late tetraploid) than with Aberystwyth S23 (late diploid) and i ntermediate with Aurora. Clover stolon length and growing point number declined to a low level in 1988 but increased dramatically in 1989 an d 1990 after cessation of applying fertilizer N. Results from both sit es were related to animal performance data previously published. It is concluded that small-leaved clover varieties are needed to ensure a h igh proportion of clover in sheep pastures and also that choice of com panion ryegrass can have a significant effect on the performance of th e legume component.