This paper reviews evidence indicating that adrenal corticosteroids modulate the responsiveness of mice and rats to nicotine. Adrenalectomy increases, and both acute and chronic corticosteroid administration decrease, some of the physiological and behavioral effects of nicotine. One function of adrenal steroids may be to regulate stress-induced changes in nicotine sensitivity. Another is to mediate the development of chronic tolerance when nicotine is given intermittently, and when the resulting tolerance has a learned component. A role of glucocorticoids in the development of tolerance to nicotine is suggested by the findings that a conditioned elevation of plasma corticosterone, which anticipates nicotine delivery, accompanies the development of chronic tolerance and that environmental cues evoke a conditioned corticosterone response, but only after they have become associated with nicotine delivery. The mechanisms by which adrenal steroids modulate nicotine sensitivity are not known, although recent in vitro evidence suggests that steroids can rapidly and reversibly reduce nicotinic receptor function. While most of the data are consistent with the hypothesis that corticosteroids reduce nicotine responsiveness, and thus promote a learned form of tolerance, there are new findings that corticosteroids increase the development of sensitization to the locomotor-activating effects of nicotine. These data suggest that formulations postulating a unidirectional effect of corticosteroids on nicotine's actions (e.g. decreased sensitivity) must be revised to take into account interacting variables such as the specific nicotine effect being studied and whether that effect normally exhibits tolerance or sensitization. Finally, research is presented which indicates that the corticosterone-elevating effects of nicotine, previously reported for experimenter-administered drug, are also produced when nicotine administration is contingent on an operant response, and at a dose which sustains the development of nicotine self-administration in rats. These findings highlight the feasibility of using self-administration models in future explorations of the relationship between adrenal steroids and nicotine function.