Our objective was to determine the influence of phosphorus (P) and potassium (K) nutrition on yield and composition of switchgrass (Panicum virgatum L.).   In 2007 switchgrass was over seeded into an array of plots  previously used to study the impact of P and K nutrition on  alfalfa (Medicago sativa L., 1997 to  2005) and maize (Zea mays L., 2006).  These plots had received historical applications of P at 0, 25, 50, 75 kg/ha and K at 0, 100, 200, 300, 400 kg/ha last applied in April of 2004.  Nitrogen (N) was applied at 75 kg/ha each May from 2007 to 2010.  Biomass was measured, and herbage and soil were analyzed for P and K concentrations.  Biomass was analyzed for neutral detergent fiber (NDF), acid detergent fiber (ADF), acid detergent lignin (ADL) and ash, and cellulose and hemicellulose concentrations were calculated.  Sugar and starch concentrations were also measured.  Cluster analysis was used to compare performance of alfalfa to maize and switchgrass. Low soil P and/or K that impaired alfalfa yield also reduced maize biomass and grain yield, but these effects did not carry over to switchgrass.  Large differences in soil test P (2 to 60 mg/kg) and K (50 to 250 mg/kg) were not associated with variation in switchgrass yield, and did not markedly alter biomass composition.  Similarly, variation in herbage P (0.5 to 2.0 mg/g) and K (2 to 16 mg/g) concentrations were not closely associated with switchgrass yield.  However, using linear-plateau regressions soil test P and K levels (0 to 20 cm) and tissue P and K concentrations were associated.  Annual P and K removal were not associated with biomass yield, but were closely associated with variation in tissue P and  K concentrations.  Regrouping plots by tissue K (high versus low) revealed that high yield and low tissue K concentrations were achievable in switchgrass.