The importance of research into the possible role of dietary intake in depressive symptoms is emphasized by the fact that diet is modifiable. We systematically reviewed observational studies investigating the association between dietary intake and depressive symptoms published in English as of December 2008. Using the PubMed database, 34 publications (23 cross-sectional, 10 prospective cohort, and 1 case-control studies) were identified. The number of subjects (n=80-27,111), age of subjects (15-97 years), dietary assessment method (dietary record, diet history interview, and validated and non-validated dietary questionnaire), depressive symptom assessment (discharge diagnosis, established scale, and self-reported information) varied among studies. Dietary variables most frequently investigated included long chain n-3 PUFA, fish, folate, and other B vitamins. Most studies found no association between dietary variables and depressive symptoms. However, most studies included at least one important methodological limitation, such as no inference for causality, unreliable or rough assessment of diet or depressive symptoms, inadequate treatment of potential confounding factors, and ignorance of the possible mediating or confounding influence of other dietary variables. Further evidence from well-designed observational studies is required to confirm or refute the association between dietary intake and depressive symptoms in free-living settings.