Annual grass weed control and switchgrass cultivar response to PRE-applied pendimethalin and POST-applied mesotrione and quinclorac was evaluated in 2005 and 2006 near Paterson, WA, in both newly seeded and 1-yr-old established switchgrass. Pendimethalin applied to newly planted switchgrass at 1.1 kg ai ha−1 at the one-leaf stage in 2005 or at 0.67 kg ha−1 PRE in 2006 severely injured and greatly reduced switchgrass stands. Mesotrione applied POST at 0.07 kg ai ha−1 injured newly planted switchgrass, reduced switchgrass height for several weeks after treatment, and reduced final switchgrass biomass by 54% both years. ‘Kanlow’ and ‘Cave-in-Rock’ cultivars were injured less by mesotrione than ‘Shawnee’ in 2005, whereas in 2006, Kanlow was injured less than Shawnee and Cave-in-Rock. Quinclorac applied POST at 0.56 kg ai ha−1 injured newly planted switchgrass less than mesotrione and pendimethalin but reduced final switchgrass biomass by 33% both years compared with treatment with atrazine alone. All three herbicide treatments controlled large crabgrass in the year of establishment. Green foxtail counts were reduced 93% or more by pendimethalin and quinclorac compared with nontreated controls, but mesotrione failed to control green foxtail. Pendimethalin applied PRE at 1.1 kg ha−1 did not injure 1-yr-old established switchgrass or reduce switchgrass biomass. Quinclorac applied POST at 0.56 kg ha−1 to established switchgrass reduced switchgrass biomass of the first harvest by 16% in 1 of 2 yr. Mesotrione applied POST at 0.07 kg ha−1 injured established switchgrass and reduced biomass of the first harvest by 33 and 17% in 2005 and 2006, respectively. Kanlow was injured the least by mesotrione in both years. Established switchgrass suppressed late-emerging annual grass weeds sufficiently to avoid the need for a grass-specific herbicide application.