||Slow germination and low seedling vigor limit establishment of perennial warm-season forage grasses, and temperature is a major environmental factor influencing both processes. Therefore, three experiments were conducted to determine temperature effects on germination of big bluestem (BB) (Andropogon gerardi Vitman), caucasian bluestem (CB) [Bothriochloa caucasica (Trin.) C.E. Hubb.], indiangrass (IG) [Sorghastrum nutans (L.) Nash.], and switchgrass (SG) (Panicum virgatum L.), as well as an annual species, crabgrass (CG) [Digitaria sanguinalis (L.) Scop.]. In Exp. I and II seeds were chilled at 4°C for 2 weeks before the germination test. In Exp. I, germination temperatures were 15, 20, 25, 30, 35, and 30/20°C with one entry of each species, except for two of SG. In Exp. II, two to four entries of BB, IG, and SG were tested at 20, 25, and 30° C. In Exp. III, unchilled seeds of two strains of BB, IG and SG, and one of CB were germinated at 9, 12, 15, 20, 25, and 30°C. A logistic function was used to describe cumulative germination. Germination was maximum between 20 and 30°C for chilled seeds compared with 12 to 20°C for unchilled seeds because lower temperatures probably met partially the chilling requirement. Chilling also enhanced speed and percentage of germination of perennial species over unchilled seeds. Calculated minimum temperatures for germination of IG, BB, SG, and CB were 8.6, 8.9, 10.3, and 10.9°C, respectively. Time to reach 50% of final germination (Gt50) decreased with increasing temperature. Further, Arrhenius plots of Gt50 indicated a change in temperature response between 20 and 25°C for perennial species. True germination rate (% day-1) was maximum at Gt50 for each treatment. Corrected germination rate index and GR30 (reciprocal of Gt50) were highly correlated and might be good indices to estimate speed of germination. The lower minimum temperatures for germination of IG and BB may allow these species to be planted earlier in spring than SG and CB.