Smooth brome (Bromus inermis Leyss.), an introduced perennial grass is an aggressive invader of prairie dominated by plains rough fescue (Festuca altaica Trin. subsp. hallii [Vasey] Harms). We (1) compared richness and density of plant species in brome and fescue stands that were unburned or burned in spring or fall; (2) determined the effects of wick application of a 33% glyphosate solution, applied when smooth brome was in the boot stage, on the desity of brome and native flora in unburned plots and plots burned in spring or fall, and (3) determined the composition of the seedbank for its potential contribution to natural revegetation following the control of smooth brome. Stem densities of native species and plains rough fescue were about two-and fivefold greater, respectively, in fescue plots than in brome plots. Species richness was generally slightly greater in fescue than in brome plots. Burning had no significant effect on stem density of smooth brome. At one site, changes in the density of smooth brome were affected by the interacting effects of burn treatments and glyphosate application. Glyphosate eliminated brome in the spring burn plots and reduced densities 76 and 50% (SE+-6.4) in the fall burn and unburned plots respectively. At a second site, smooth brome densities were reduced by glyphosate but not by burn treatments. There was, however, a trend for greater reduction of smooth brome with glyphosate application in the spring burn (98%) than in the fall burn (40%) and unburned (56%) plots (SE +-15.5). Glyphosate reduced the density of native graminoids 91% (SE +-10.0), but plains rough fescue and native forbs were not affected. Twenty-three species emerged from the seedbank in fescue plots, whereas twenty emerged from soils collected in brome plots. The total number of seedlings emerging from the seedbank was similar in brome and fescue plots, averaging 1794/m2 and 2078/m2 (SE+-252), respectively. The proportion of seedlings emerging was lower for native graminoids (33 vs 41%, SE +-2) and greater for native forbs (56 vs 48%, KSE +-3) in brome as compared to fescue plots. Smooth brome seedlings emerged only from soils collected in brome plots, averaging 3/m2. These studies indicate that excellent control of smooth brome can be achieved with spring burning followed by wick application of glyphosate, however, native species were also reudced by glyphosate. Additonal glyphosate applications will be required for complete control of smooth. The number of seeds of native species in the soils in smooth brome stands approximates that in stands of plains rough fescue, suggesting that there is an adequate seedbank for natural recovery of vegetation following control of smooth brome. © Natural Areas Association. Abstract reproduced by permission. Further information Email: naa@natareas.org