The enormous variation in values quoted in the literature concerning the occurrence of depressive diseases is caused by the lack of standardized diagnostic criteria. The difficulties of epidemiological research arise from nosological differentiation within the range of depressive diseases, from case identification and case definition. Further problems result from the limiting definitions of depressive disturbances, in respect to variations within the healthy condition on the one hand, and to signs and symptoms of other diseases, on the other hand. Moreover, it is not clear from most of the relevant epidemiological literature whether the investigation of depressive states refers to nosological diagnoses, to descriptive syndromes or to symptoms. We undertook the psychiatric investigation of the entire adult population (1337 in number) of an alpine valley. The comprehensive method of case finding in our seven-year epidemiological field study revealed an incidence of 3.9% of affective psychoses. Hence, the true prevalence rate of endogenous manic-depressive disease seems to lie somewhere between 3 and 4% of the total population. This rate is confirmed by projection of the results of other studies.