SEED PRODUCTION STUDIES ON PERENNIAL RYEGRASS, TIMOTHY AND PRAIRIE GRASS

A study over two years involving tiller identifica tion according to month of origin enabled the contribution of individual tillers of perennial rye grass Grasslands Ruanui, timothy Grasslands Kathu and a local New Zealand strain of prairie grass (Bromus unioloides) to be determined at seed harvest. In addition, the distribution of seed-heads in different arbitrary emergence groups was detemined so as to allow analysis of individual seedhead components within each group. In all species, tillers formed during the first 4 months following an autumn sowing in the first year, and also in the immediate post-harvest period and through the autumn in the second year, made a major contribution to seedhead numbers and seed yield at harvest. These tillers had a lower vegetative mortality rate than those formed in the spring. Primary tillers formed after sowing were highly persistent and became almost exclusively repro ductive. Many vegetative tillers died before har vest, behaving as annual tillers. Spring-formed tillers made a major contribution to the vegetative growth of the plant over the summer and early autumn. These tillers generally died before the following winter. Tillers formed in winter and early spring by ryegrass and prairie grass showed a high mortality, particularly in the second year. In timothy, mortality was highest amongst tillers formed in late spring. The reduced contribution of spring-formed tillers to the seed head population occurred as a result of fewer heads numerically compared with older tillers. In all species, most seedhead components varied according to time of ear emergence. Early emerged heads were generally longer, bore more spikelets and more florets per head and had a greater culm length than late-emerged heads. The major trend in all species was a general depression in head length, spikelet and floret number and culm length with grazing. These effects were in most cases partly or completely overcome by N application. The major effect of applied N was to increase floret number per head.