We determined whether plant diversity and sequence of plant ingestion affected foraging when cattle chose from plants that varied in concentrations of alkaloids, tannins and saponins. We hypothesized cattle that ate high-alkaloid grasses (endophyteinfected tall fescue (TF) or reed canarygrass (RCG)) would prefer forages high in tannins (birdsfoot trefoil, BFT1) or saponins (alfalfa, ALF1), because tannins and saponins can bind to alkaloids, presumably reducing their absorption. We further hypothesized that forages with tannins or saponins consumed before, rather than after, foraging on high-alkaloid grasses would promote greater use of those grasses presumably by binding to alkaloids, thereby reducing their absorption. In Phase 1, cattle (n 5 32) grazed on either high (1) or low (2) alkaloid grass (TF or RCG) pastures for 30 min each morning at 0600 h and were then offered a choice of BFT1, BFT2, ALF1 and ALF2 for 60 min each day for 12 days. In Phase 2, cattle (n 5 32) were first offered a choice of BFT1 or ALF1 for 30 min at 0600 h and then placed on grass (TF1 or 2, or RCG1 or 2) pastures for 60 min for 12 days. In both phases, we had four spatial replications of four treatments with 2 per calves assigned to each of the 16 replications per treatment combinations. Scan samples of individuals at 2-min intervals were used to determine incidence of foraging on each plant species (%). Cattle grazed more on RCG than on TF in Phases 1 (62% v. 27%; P 5 0.0015) and 2 (71% v. 32%; P 5 0.0005). In Phase 1, cattle that first foraged on RCG1 or TF2 subsequently preferred ALF over BFT, whereas cattle offered RCG2 or TF1 foraged on ALF and BFT equally. Foraging by cattle on RCG was cyclic during Phase 1, whereas cattle foraging on TF markedly decreased incidence of use of TF from 41% to only 16% by the end of the 12-day trial (P 5 0.0029). Contrary to the cyclic (RCG) or steadily declining (TF) use of grasses in Phase 1, cattle steadily and dramatically increased foraging on both RCG and TF throughout Phase 2, when they first grazed BFT1 or ALF1 followed by high-alkaloid grasses (P 5 0.0159). Our findings suggest that in plant species the sequence of ingestion influenced foraging behavior of cattle and that secondary compounds influenced those responses.