A continuous-culture fermentor study was conducted to assess nutrient digestibilities, volatile fatty acid (VFA) concentrations, microbial protein synthesis, bacterial nitrogen (N) efficiency, and enteric methane (CH 4 ) production of four 50:50 grass-legume diets, randomly assigned in a 4 × 4 Latin square design. Four legumes with different concentrations of condensed tannins (CT) were tested: alfalfa [ALF; Medicago sativa L., non-CT legume]; birdsfoot trefoil [BFT; Lotus corniculatus L., low-CT legume]; crown vetch [CV; Securigera varia (L.) Lassen, moderate-CT legume]; and sericea lespedeza [SL; Lespedeza cuneata (Dum. Cours.) G. Don, high-CT legume]. Orchardgrass (Dactylis glomerata L.) was the common forage used in all diets. Four fermentors were evaluated over four 10-d periods by feeding 82 g of dry matter (DM)/d in 4 equal feedings. Methane output was recorded every 10 min. Effluent samples were collected during the last 3 d of the experiment,composited by fermentor and period, and analyzed for pH and VFA, as well as DM, organic matter, crude protein, neutral detergent fiber, and acid detergent fiber for determination of apparent and true nutrient digestibilities. Microbial protein synthesis and bacterial efficiency were estimated by analysis of N flows and purines. The CT concentrations were 3, 21, 38 and 76 g/kg of DM for ALF, BFT, CV, and SL diets, respectively. The SL diet had decreased fiber digestibilities and total VFA concentrations compared with the other diets. This resulted in the least total CH 4 production in the SL diet. Bacterial N efficiency per kilogram of organic matter truly digested was lower in the SL diet than in the BFT and CV diets. The lowest CH 4 production per unit of digestible nutrients was also found in the SL diet. Further work should be conducted to find optimal diets (by testing other legumes, rations, and sources of CT) for reducing CH 4 emissions without negatively affecting ruminal digestion to maintain or improve productivity.