Timely assessment of forage productivity and quality during the growing season is important in livestock and pasture management. Traditional laboratory determination of forage quality is usually time consuming and costly. Remote sensing of canopy reflectance may be used for the real-time estimation of forage biomass and quality. Canopy hyperspectral reflectance was measured using a portable spectroradiometer in five warm season grass pastures during the 2002 and 2003 growing seasons to determine relationships between forage biomass and quality variables and canopy reflectance. Pasture biomass, crude protein (CP) concentration, and CP availability correlated closely and linearly with the two-band reflectance ratios of R1145/R1205, R1695/R605, and R875/R735 (r² = 0.36-0.61****, n = 207), respectively. Although correlations of neutral detergent fiber (NDF) and acid detergent fiber (ADF) with reflectance ratios were significant (P < 0.0001), the best reflectance ratios could only explain 13 to 35% of NDF and ADF variations (r² = 0.13-0.35). Compared to a simple linear regression of the quality variables with a two-band reflectance ratio, multiple regression (MAXR) with a total of 10-waveband entrances improved the relationships between forage quality and reflectance values (r² = 0.27-0.74****). Validation of developed equations indicated that the forage biomass, CP concentration, and CP availability could be well predicted using either 10-waveband reflectances or the two-band reflectance ratios. Pasture NDF could also be predicted using the MAXR equation. Our results suggest that biomass and major quality parameters of warm season grass pastures can be nondestructively predicted using canopy reflectance measurements.