The objective of the research described here was to compare the milk production from paspalum (Paspalum dilatatum Poir.)-dominant swards with milk production from perennial ryegrass (Lolium perenne L.)–white clover (Trifolium repens L.)-dominant swards. Eight experiments were conducted over a 2-year period, during which half the cows grazed pastures that had been established for more than 15 years (old pasture) and were dominated by paspalum in summer, while the remaining cows grazed pastures sown within the previous 2–3 years (new pasture), which were dominated by temperate species. Groups of four to five lactating Friesian cows grazed either old or new pasture and were either unsupplemented or were offered ~5 kg DM/cow.day of a high energy supplement (75% barley and 25% wheat). The hypothesis was that milk production would be greater from cows grazing new pastures than old pastures (at common pasture allowances), but that marginal milk responses when cereal grain-based supplements were fed would be greater when cows grazed old pastures; this was only partially supported. Milk yields were generally greater when cows grazed new pastures, but in six out of eight experiments, the advantage was only ~1 kg milk (100 g fat plus protein)/cow or less. In these six experiments, the difference in metabolisable energy density between the pastures was 0.7 MJ/kg DM or less. When concentrates were fed to supplement grazed pasture, there were seasonal variations, but similar marginal responses in milk production were recorded on both pastures throughout the study.