Management plans for the grazed parklands of the Auckland volcanic cones call for revegetation to restore a more natural and historic vegetation cover. The native grass Microlaena stipoides has been identified as a relevant species, and a field experiment was undertaken to determine the best approach for establishing this species into existing pasture. Seed obtained locally was broadcast or hydro-seeded in May 2007 onto kikuyu pasture, at three sowing rates (100, 250 and 750 seeds/m2) following control of the resident pasture with turf stripping or mowing and spraying, in a replicated factorial trial. After 12 months, the highest levels of Microlaena plant density (mean 56 plants/m2), cover (mean 50%) and accumulated herbage mass were observed on broadcast/sprayed plots at the highest sowing rate. Recovery of the resident pasture (dominantly Pennisetum clandestinum) occurred to a level of ~25% after 12 months, but was inhibited at the highest Microlaena sowing rate. Keywords: broadcast, hydro-seeding, kikuyu, Microlaena stipoides, re-vegetation