It was hypothesized that a tannin-rich legume such as sainfoin attenuates the negative postingestive effects of ergot alkaloids in tall fescue. Thirty-two 4-mo-old lambs were individually penned and randomly assigned to a 2 × 2 factorial arrangement with 2 legume species, sainfoin (SAN; 2.9% condensed tannins) or cicer milkvetch (CIC; without tannins) and a mixed ration containing tall fescue seed (50:30:20 seed:beet pulp:alfalfa) with 2 levels of endophyte infection (endophyte-infected tall fescue seed [E+; 3,150 ug/L ergovaline] or endophyte-free tall fescue seed [E-]). For a 10-d baseline period, half of the lambs were fed SAN and half were fed CIC and all lambs had ad libitum amounts of E-. In an ensuing 10-d experimental period, the protocol was the same except half of the lambs fed SAN or CIC received E+ instead of E-. Subsequently, all lambs could choose between their respective legume and seed-containing ration and between E+ and E-. Finally, an in vitro radial diffusion assay was conducted to determine whether tannins isolated from SAN would bind to alkaloids isolated from E+. All groups consumed similar amounts of E- during baseline period ( > 0.10), but lambs ate more E- than E+ during the experimental period ( < 0.05) and lambs offered SAN ate more E+ than lambs offered CIC ( < 0.05). Groups fed E- during the baseline and experimental periods had similar rectal temperatures ( > 0.10), but lambs fed E+ had lower rectal temperatures per gram of feed ingested when supplemented with SAN than with CIC ( < 0.05). Lambs fed E+ had greater concentrations of hemoglobin and more red blood cells than lambs fed E- ( < 0.05), but plasmatic concentrations of cortisol and prolactin did not differ among treatments ( > 0.10). All lambs preferred their treatment ration over their treatment legume, but lambs in the SAN and E+ treatment ate more legume + ration than lambs in the CIC and E+ (CIC-E+; < 0.05) treatment. All lambs preferred E- over E+, but lambs in the CIC-E+ treatment ate the least amount of E+ ( < 0.05). Binding of isolated SAN tannins to protein was reduced by the E+ isolate ( < 0.05), suggesting a tannin-alkaloid complexation but only from tannins extracted from SAN fed early in the experimental period. In summary, SAN supplementation increased intake of and preference for E+ and reduced rectal temperatures relative to CIC supplementation. Our results suggest that these effects were mediated by the condensed tannins in SAN through alkaloid inactivation, an interaction that declined with plant maturity.