The aim of this paper is to consider the nosological issues of depressive syndrome predominantly from the viewpoint of the modern neurobiology of psychiatric disorders. The author particularly addresses the following two questions for general psychiatrists to facilitate their understanding：“Can the response to medication treatment distinguish each subcategory of depressive syndrome?”and“Can we clearly define a boundary between the grief reaction due to bereavement and major depressive disorder?”For the former, the categorical classification based on the response to different psychotropic drugs is losing the significance it had at the time DSM‒Ⅲ was published in 1980. The discrimination of melancholia and atypical depression based on a differential response to the specific type of antidepressant drugs（e. g., tricyclic antidepressants or monoamine oxidase inhibitors）is unlikely to be proven. Moreover, because melancholia and atypical depression overlap each other longitudinally in clinical features, that may continue for both mood disorder subtypes in pathophysiology. For the latter question, the DSM‒5 draft contains a proposal deleting the exclusion criteria of bereavement from major depressive disorder. Although this proposal has become controversial, the functional brain imaging studies suggest a potential continuity between the bereavement reaction and major depression. In conclusion, a growing body of evidence from biological research does not seem to support the validity of the categorical diagnosis of the depressive syndrome subdivision. At present, such biological studies may require the current DSM designers to incorporate a dimensional approach in DSM‒5.