Cigarette smoking is associated acutely with elevated cortisol levels. However, the results of comparisons of cortisol levels in smokers and non-smokers have been inconsistent, and the significance of cortisol responses in smoking cessation is unclear. Here we describe one study comparing the cortisol profiles of smokers and nonsmokers over the day, and a second investigation in which cortisol was monitored during smoking cessation. In the first study, we collected saliva samples from 196 middle-aged men and women on working and weekend days repeatedly through the day. On both working and weekend days, cortisol levels were significantly higher in smokers after adjustment for age, gender and grade of employment. Cortisol responses to waking (the increase between waking and 30 min) were also greater in smokers. The elevation in cortisol among smokers is generally attributed to nicotine exposure. Nicotine replacement therapy substantially improves abstinence rates, and has become a standard component of smoking cessation treatments, but the effects of nicotine replacement on cortisol are not known. In the second study, cortisol was monitored over 6 weeks of abstinence in 112 smokers treated with behavioural support and 15 mg nicotine patches. Smoking cessation was accompanied by an abrupt decrease in salivary cortisol, and this was sustained over the abstinence period. There was a marginal association between the decrease in cortisol and smoking relapse rates. These results suggest that the nicotine supplied through patches was not sufficient to block the cortisol reduction following smoking cessation. The contribution of these findings to understanding the role of neuroendocrine function in smoking is described.