Models provide a quantitative means to evaluate yield response of forage grasses to applied N and water availability. The objective of this analysis was to estimate model parameters (A, b, c) for bermu· dagrasses and bunchgrasses grown at the same location and during the same time period. The logistic equation was used to relate annual dry matter production to applied N for three bermudagrasses [C)· nodon dactylon L., 'Coastal', 'Alicia', and 'Coastcross-1'] and four bunchgrasses [Eragrostis curvuw (Schrad.) Nees, 'Morpa' and 'Ren· ner' lovegrass; Panicum coloratum L., 'Selection-75' kleingrass, and Cenchrus ciliDris L., 'Strain 18-35' buffelgrass]. Data were from 4 yr of field experiments at Stephenville, TX. Analysis of variance was used to show that the N coefficients b and c were independent of year and climate for each grass, in agreement with recently published results. Applied N required to provide one-half of maximum yield was approximately 150 kg ha -I for bermudagrass and 50 kg ha -I for lovegrass. Variation among years was accounted for in the linear coefficient A, which was then related to rainfall. Bunchgrasses typically had lower A and b, but similar c values compared with bermudagrasses.The logistic model allows estimates of yields in terms of applied N and rainfall, and quantifies the interaction between these two factors. Bermudagrass yielded approximately 45% more dry matter than lovegrass for the same rainfall. Correlation coefficients exceeded 0.95 throughout the analysis.