The effect of four levels of phosphorus (0, 33, 66 and 98 kgjha), and four levels of calcium carbonate (0, 2, 4 and 8 ton/ha) on a fescue and subterranean c10ver mixture, was studied at Pumalallocation, on an Ultisol Metrenco, during two years. The trial was sown in March 1986, in a complete randomized blocks design, with four replications. Each plot consisted of 5 rows, 6 m long, 0.2 m aparto Fescue cv. Clarine was sown with 12 kgjha of seed and subterranean clover cv. Mount Barker at 8 kgjha of seed, previously inoculated with the specific Rhizobium and pelletized. Legume plant population, nodules clover plant, dry matter production, phosphorus response efficiency, botanical composition ofthe mixture through time, soil pH and residual phosphorus at the end of each growing season, were measured. No significant differences (P > 0.05) were found in plant population and nodulesjplant. In both seasons, dry matter production increased significantly (P < 0.01) as a result of phosphorus applications. Calcium carbonate application was effective (P < 0.01) only the first season and the seed pelleting showed no effects. Phosphorus applications positivelyinfluenced clover populations (only the first season) and residual phosphorus, at the end of each growing season. Calcium carbonate application increased Iinearly the soil pH and their effects remained at the end of the second year.