We evaluated the reliability of using cortisol concentrations in urine and feces to indicate stress responses in captive Rocky Mountain bighorn sheep (Ovis canadensis canadensis). Tests for parallelism and quantitative recovery of cortisol in bighorn plasma, urine, and fecal supernatant revealed that radioimmunoassay reliably measures cortisol in these media. High performance liquid chromatography confirmed the presence of cortisol in fecal extracts. In a randomized complete block experiment, we observed cortisol responses of eight bighorn sheep injected with either 0.0 or 0.50 U/kg of adrenocorticotrophic hormone (ACTH) gel delivered at 48 h intervals over a 29-day period. Treatment elevated (P < 0.01) cortisol in urine and fecal supernatant during 0–24 h following ACTH injections but levels dropped below those of controls during 24–48 h posttreatment. Magnitude of cortisol responses did not change with time during the experiment (time × treatment, P > 0.05). In a separate experiment, we observed short-term (0, 2, 4, 8, 12, 24 h) cortisol responses in plasma, urine, and feces from eight bighorns following a single challenge with 0.0, 0.25, 0.5, or 1.0 U/kg of repository corticotropin. Plasma cortisol concentrations rose within 2 h of treatment (P < 0.025) and remained elevated for at least 8 h. Urine cortisol excretion showed a similar trend with peak response delayed 2–6 h. Fecal cortisol trends resembled those of plasma and urine but were more variable. We failed to observe a dose response above the 0.5 U/kg level in any media. Concentrations of cortisol (ng/mL) in plasma correlated (r2 = 0.76) with cortisol:creatinine ratios (ng/mg) in urine. We conclude that measuring cortisol in excreta of bighorn sheep offers promise as a noninvasive means of sampling their responses to environmental stressors.