THE YIELD AND COMPOSITION OF LUCERNE, GRASS AND CLOVER UNDER DIEEERENT SYSTEMS OE MANAGEMENT

The yields of two varieties of lucerne sown alone and in associalion with each of three grasses were compared with similar red clover/grass mixtures over the period 1953–56 under two levels of nitrogenous and potassic fertilizers and two cutting treatments. The main object of the experiment was to gain a wider knowledge of the productivity and chemical composition of lucerne sown alone, and in mixtures, under relatively high rainfall conditions. The highest production was obtained from Du Puits and its mixtures, which exceeded Grimm and S 123 red clover mixtures. The yields of the lucerne/grass mixtures were not significantly better than those of the pure sown lucerne. Under the most favourable conditions for lucerne, Du Puits eliminated practically all the sown grasses, viz., S 23 ryegrass, S 51 timothy and S 215 meadow fescue. The application of nitrogen encouraged the growth of the grasses, often at the expense of the legume, especially the slower growing Grimm and S 123. The effect on Du Puits was less marked, but in no case was there a stibstantial increase in actual legume yield. Increasing the level of potassium, although not immediately affecting the yield, proved beneficial to lucerne survival in the later years. The protein yield from the best lucerne plots was found to be double that from pure grass swards heavily fertilized with nitrogen. Three cuts of lucerne in the first two harvest years had a disastrous effecl on the carry‐over of lucerne from the second to the third year. Both varieties of lucerne and their mixtures behaved in a similar fashion, although S 123 red clover appeared to be indifferent to these conditions. It is suggested that the cause of this was the unfavourable weather experienced in 1954.