Activation of the hypothalamic-pituitary-adrenal (HPA) axis is one of the hallmarks of the physiological responses to psychosocial stressors. The most common method of assessing HPA function is via the measurement of plasma cortisol levels. However, venipuncture involves capture and restraint, which can modify HPA function. We validated a noninvasive procedure for monitoring HPA responses to stressors by measuring excretion of free urinary cortisol. Samples collected throughout the day displayed marked circadian variation, with low cortisol values in first-void samples, followed by a mid-morning peak in cortisol excretion. Concentrations of excreted cortisol declined throughout the day. Exposing marmosets to mild and moderate stressors (11 h isolation in a small cage and manual restraint) increased excreted cortisol concentrations in a dose-dependent fashion: isolation in a small cage led to elevated cortisol in afternoon samples, while manual restraint and isolation produced elevated cortisol in both morning and afternoon samples. The marmoset HPA is differentially sensitive to rather subtle variations in stressors, and these results show that urinary cortisol excretion is a valid and sensitive index of the HPA response to these stressors.