Pastures based on the white clover cultivars Grasslands Tahora and Grasslands Huia ",:ere established by oversowing into unimproved hIll country at Ballantrae in May-June 1986. Each pasture type occupied 2.2 ha, and a further 2.2 ha of unamended pasture containing the resident clover ecotype was also included. From August 1988, four ferti1i~er treatments (control, receiving no fertiliser; reactive phosphate rock; partially acidulated phosphate rock; and a superphosphate/phosphate rock blend, the latter three applied at 35 kg Plha per year) were superimposed upon pasture types. Emphasis was placed on identifying the population and plant characteristics which responded to treatment, and defining the relative benefits to pasture performance resulting from the two approaches to pasture improvement (cultivars versus fertiliser). Based on measurements of the frequency of occurrence of cyanogenic plants in populations nearly 4 years after sowing, 58% of white clover plants in the Tahora-sown pastures and 43% of plants in the Huia-sown pastures were estimated to have derived from the original seedlines. The rest A92063 Received 1 December 1992; accepted 15 February 1993 of these populations were volunteer plants of the resident ecotype; these were much smaller than plants ~f Tahora and Huia (mean total plant weights, Includmg root material, were 49.5, 112.2, and 139.2 mg, respectively). When both genotype frequency and plant weight differences were accounted for, the introduced cultivars were estimated to have contributed 70% of the total white clover biomass present in sown pastures, averaged across fertiliser treatments. White clover herbage accumulation was greater in the sown pastures than in the resident pasture (558 and 586 kg DMlha per year for Huia-sown and Tahora-sown pastures respectively, versus 305 kg DMlha per year for resident pasture, means for Years 3 and 4 after sowing), and in the fertilised treatments compared to the Control (518-597 kg DMlha per year versus 231 kg DMlha per year respectively). Applying 35 kg Plha per year had a greater influence on clover herbage accumulation than cultivar introduction, though returns on cultivar introduction can be seen as accruing each year (if cultivars are persistent) for a one-off cost whereas fertiliser must be financed annually. Fertiliser application increased mainly the population density of clover plants (number/m2) whereas cultivar introduction affected mainly the mean size of plants in the population. Limited evidence was obtained to suggest that Tahora and Huia responded differently to fertiliser application.