Both depressive symptoms and vitamin D insufficiency are common during winter. This study examined the association between serum 25-hydroxyvitamin D and depressive symptoms by survey season. Subjects were 527 municipal employees aged 21-67 years of two municipal offices in Japan. Overall, there was no measurable association. However, in the workplace surveyed in November, multivariate-adjusted odds ratios (95% confidence interval) of having depressive symptoms (Center for Epidemiologic Studies Depression score of >or=16) for the lowest through highest quartiles of serum 25-hydroxyvitamin D were 1.00 (reference), 0.84 (0.45-1.58), 0.83 (0.44-1.58) and 0.59 (0.30-1.15), respectively (trend P=0.14). The association with the severe depressive state was more pronounced. By contrast, there was no such association in the workplace surveyed in July. Overall, this study did not provide evidence linking higher blood vitamin D levels with decreased depressive symptoms. The suggestive inverse association in sun-deprived season warrants further investigation.