Selenium-accumulator plants are reputed to be unpalatable to large ungulates. Elk (Cervus canadensis) populations in southeastern Idaho overlap with populations of Se-rich plants, but there is no information on the influence of plant Se concentration on elk dietary preferences. The objective of this study was to determine, under controlled conditions, the preference of elk for feeds varying in Se concentrations. Seven yearling female elk (128 ± 5 kg) were purchased from a commercial elk farm in southeastern Idaho and adapted to low-Se alfalfa pellets. Three feeding trials using pellets with predetermined Se concentrations were conducted. Alfalfa pellets were commercially made with the addition of Symphyotrichum ascendens (western aster) so that the pellets contained 0.4, 5, 20, 50, or 100 mg/kg (DM basis) Se. In trial 1, 5 Se-containing alfalfa pellets (0.4, 5, 20, 50, and 100 mg/kg Se) were offered for 10 d; trial 2 used 4 Se-containing alfalfa pellet choices (0.4, 20, 50, and 100 mg/kg), and trial 3 used 3 pellet choices (0.4, 50, and 100 mg/kg) for 6 d. In trial 1, consumption of the control pellets by elk was greater than each of the other pellet choices (P 0.50) in consumption of the 20-, 50-, or 100-mg/kg Se pellets. In trial 2, elk consumed more (P 0.99) in elk consumption of the 50- and 100-mg/kg Se pellets. In trial 3, elk consumption of the control and 50- and 100-mg/kg Se pellets differed (P ≤ 0.03) from one another each day except that on d 1 and 2, where elk consumption of the 50- and 100-mg/kg Se pellets did not differ (P ≥ 0.32). Elk clearly discriminated against pellets with the highest Se concentrations when they were given pellets with differing Se concentrations. These results suggest that elk are not likely to select forages with high Se concentrations, particularly when high-Se plants are present in a rangeland situation with numerous other forage choices.