Abstract Cades Cove, Great Smoky Mountains National Park, U.S.A. was historically cleared largely for pastoral purposes; it is now comprised of recently abandoned pastures dominated by non-native pasture species. To investigate the potential for reducing non-native species relative to native species, park managers initiated an experiment in 1995 that included mowing, herbicide application, planting of seed, and burning of replicate 20 × 50–m plots at each of two sites within Cades Cove. Between 1995 and 2001 we evaluated the response of the plant community (i.e., species-specific cover and frequency, biomass, diversity) to this suite of treatments and compared it with unmanipulated control plots at each site. Four years after treatment initiation abundance measures of Plantago lanceolata, Setaria geniculata, and Trifolium spp. averaged one-third lower in treated than control plots. Frequency of Festuca pratensis was lower in treated than in control plots for 2 years, but after 4 years its frequency, cover, and biomass did not differ between treated and control plots. By 2000 the cover of Sorghastrum nutans in treated plots increased to 23–47%, depending on the site. Total biomass and diversity increased in treated plots. The dominance of Lespedeza cuneata at one site apparently reduced planting success, biomass production, and diversity and evenness. Post-treatment lags in response for several species, coupled with interannual variation in response to environmental conditions, suggest that evaluations of treatment success would differ greatly depending on when the evaluation was conducted.