This study investigated individual differences in the diurnal cycle of cortisol and explored their relation to several psychosocial variables and to upper-respiratory symptoms. Cortisol and daily experience were assessed for 2 days in 109 healthy employed and unemployed community residents (mean age = 36.4 +/- 12.1, 69% female); self-report upper respiratory illness (URI) symptoms were assessed for an additional 10 days. Fifty-six (51%) participants showed typical declines in cortisol during both days, 19 (17%) showed no significant diurnal pattern on both days, and 34 (31%) showed different diurnal patterns on the 2 days. Individuals with no cycles did not differ from those with normal or inconsistent cycles on demographic factors, baseline psychological measures, health behaviors, or daily experiences over the two assessment days. Individuals without cortisol cycles, however, reported fewer URI symptoms than the remaining subjects. That 17% of our sample did not exhibit diurnal cycles of cortisol was surprising, given established views of normal endocrine function. Although average daily level of cortisol is related to a number of psychosocial and psychiatric factors (e.g. stress and depression), pattern of diurnal cycle was not related to any demographic or psychosocial measures in this study. The finding that flat cycles were related to fewer reports of URI symptoms suggests that perturbations in cycle may be related to processes associated with symptom susceptibility or symptom expression. (C) 1997 Elsevier Science Ltd.