The measurement of salivary cortisol provides a simple, non-invasive, and stress-free measure frequently used in studies of the hypothalamic-pituitary-adrenal axis activity. In research projects, samples are often required to be stored for longer periods of time either because of the protocol of the project or because of lack of funding for analysis. The aim of the present study was to explore the effects of long-term storage of samples on the amounts of measurable cortisol. Ten pools of saliva were collected on polyester Salivette tampons from five subjects. After centrifugation the samples were either stored in small vials or spiked to polyester Salivette tampons before analysis for cortisol using Spectria RIA kits. The effects of storage were evaluated by a linear regression model (mixed procedure) on a logarithmic scale. No effects on cortisol concentrations were found after storage of saliva at 5 degrees C for up to 3 months or at -20 degrees C and -80 degrees C for up to one year. In contrast, concentrations of cortisol were found to decrease by 9.2% (95% confidence interval (CI): 3.8%; 14.3%) per month in samples stored at room temperature. Repeated freezing and thawing of samples up to four times before analysis did not affect the measured concentrations of cortisol. The coefficient of residual variation (CVresid) for samples stored on Salivette tampons were twice the CVresid for samples stored in separate vials after centrifugation. In conclusion, centrifuged saliva samples for analysis of cortisol may be stored at 5 degrees C for up to 3 months or at -20 degrees C or -80 degrees C for at least one year. However, long-term storage at room temperature cannot be recommended. Repeated cycles of freezing and thawing did not appear to affect the concentrations of cortisol.