No abstract

Herbage accumulation and botanical composition of pastures containing 4 morphologically-contrasting white clovers (Kent wild white, Southland uncertified, 'Grasslands Huia', and 'Grasslands Pitau') are described. The clovers were grown in monoculture and in association with either 'Grasslands Ruanui' or 'Nui' perennial ryegrass under grazing frequencies of 4 and 6 weeks. Clover production in monoculture and mixtures was least from the small-leaved Kent, intermediate for medium-leaved clovers, and greatest from the largeleaved Pitau. Kent was inferior in all seasons, whereas the superiority of Pitau reflected good autumn-winter growth. In all mixtures, the ryegrass and clover components showed improved production under infrequent grazing for all seasons. Within a grazing frequency, pastures containing Kent produced the lowest clover yields but had the largest associated grass yields, whereas pastures containing the larger-leaved Pitau showed the converse. This compensatory effect resulted in the pasture mixtures producing similar total herbage. Differences in grass/clover compositions were related to the morphology of clover types. This has implications for competitive effects in grasslclover mixtures and for nutrient cycling in pastures. A desirable clover type would incorporate the potential to produce large quantities of nutritious feed and the ability to support an efficient transfer of nutrients to grasses, thus ensuring a reliable distribution of herbage yield.

(Festuca arundinacea Schreb.), established successfully at most sites. Findlay-Wilkinson analysis of annual total herbage accumulation (THA) of four sites, found that Maru production was above average and resident production was below average. Variation in THA between sites and years was related to variation in annual rainfall (621-1484 mmI year). Wana was the most persistent grass at all sites (60% of THA after 4-6 years), followed by Nui (51 % of THA after 4-6 years). Maru was only persistent at Southland, Wairarapa, and Taupo (averaging 39% ofTHA after 4-6 years). Matua was only persistent at Wairarapa (32% of THA over 5 years), and Roa was only persistent at Taupo (58% of THA over 5 years). At most sites, at least one grass species (in many instances more) outperformed the resident sward.

Evaluations of Kahu timothy (Phleum prutense L.) included agronomy under mowing (1976-1979) and grazing (1980-1983), and animal performance from grazed pasture (1988-1993) and silage feeding (1989-1991). Agronomic testing showed that Kahu timothy was slow to establish, producing only 28% of the Grasslands Ruanui ryegrass (L.&urn perenne L.) swards in the 6 months after sowing. Frequent, close defoliation severely reduced the yields of timothy and caused losses of plants, reducing longevity. A less frequent defoliation regime with a residual sward height of 30-50 mm gave the greatest sustainable yields. Under mowing Kahu produced significantly less than S51 timothy in summer and winter but under grazing Kahu was superior in spring and summer, resulting in significantly greater annual production. Animal performance testing showed hoggets grazing timothy in spring grew at 218 g/day, significantly greater than the 164 g/day achieved on Grasslands Nui ryegrass. The spring stocking rate on timothy was also significantly greater than on ryegrass, though utilisation of the forage was similar. When fed timothy/red clover silage lambs fed in winter grew at 98 g/day while lambs on standard ryegrass/ cocksfoot/white clover silage averaged a growth rate of only 44 g/day. From the assessment of the physiological state of individual tillers it was observed that timothy had a significantly higher proportion of its tillers with growing points present above ground through late spring and summer than ryegrass. This is suggested as a major reason for the depletion of timothy populations -under close grazing. When grazed correctly in late spring and summer timothy can provide a perennial forage of high feeding value for livestock systems. Keywords: animal performance, Grasslands Kahu, pasture quality, Phleum pratense

Nine subterranean clover (Trifolium subterraneum L.) cultivarsand 10 whiteclover(T. repensL.) lines were evaluated i.n small plo~f?r production and persistence at a hIll country SIte In Southland. The late-flowering subterranean clovers 'Nangeela' and 'Tallarook' and the mid-season flowering 'Woogenellup' showed the highest seedlingregeneration and herbageyieldsin thefirst year. However, regeneration declined to 20% of levelsrecordedin the first year and yieldsfell from 1700to 50kg DM/haby the secondyear.Continued lowproduction wasexpected as noneof thecultivars developed hardseed. TheNewZealand whiteclover lines ('Grasslands Kopu', 'Grasslands Huia', 'GrasslandsPitau', 'Italy X NewZealand' and'Hill Country') showed the best herbage yields and persistence. The overseas lines ('Haifa', 'Tamar', 'Clarence Valley', and 'Louisiana S1') were consistently poor.Thesetypes wereunable tosustain active growth because of low stolon numbers per unit area of pasture. Re-seeding was unimportant over the evaluation period at this site.