In this month's Archives of General Psychiatry, a study by Sanchez-Villegas et al. is published showing a strong association between lower rates of depression, and consuming a Mediterranean diet (lots of vegetables, fruits, nuts, whole grains, and fish, with low intake of meat, moderate intake of alcohol & dairy, and lots of monounsaturated fatty acids compared to saturated fatty acids). Data was gathered prospectively during a period averaging over 4 years, and was based on following about 10 000 initially healthy students in Spain who reported food intake on questionnaires.
I'll have to look closely at the full text of the article. I'm interested to consider the question of whether the results strongly suggest causation, or whether the results could be due to non-causal association. That is, perhaps people in Spain with a higher tendency to become depressed tend to choose non-Mediterranean diets. Another issue is cultural: the study was done in Spain, where a Mediterranean diet may be associated with certain--perhaps more traditional--cultural or subcultural features, and this cultural factor may then mediate the association with depressive risk.
In any case, in the meantime, given the preponderance of other data showing health benefits from a Mediterranean-style diet, I wholeheartedly (!) recommend consuming more nuts, vegetables, olive oil, fish, whole grains, and fruit; and less red meat.