Among a large workplace population interviewed over a year, current sy
mptoms were assessed using the Hopkins Symptom Checklist (HSCL). Varia
tion in symptoms by date of assessment was observed among the 314 wome
n but not the 1556 men. Among women, symptoms were greatest during lat
e autumn and winter, and significant inverse correlations were found b
etween available daylight at the time of assessment and standard sympt
om dimensions of anxiety and somatisation, as well as an expanded mood
scale more inclusive of depressive symptoms within the check-list. Th
e amplitude of the seasonal effect was such that the prevalence of fem
ale 'cases', as defined by HSCL criteria, was twice as high during win
ter than during the rest of the year. The data are consistent with a r
ole for light-dark exposure in eliciting or synchronising annual mood
rhythms. The sex-by-season interactions may contribute to the sex diff
erences in overall prevalence of depression.