This study has arisen out of the work of the Newcastle group (Kiloh and Garside, 1963; Kay et al., 1969) and that of the London group (Kendell, 1968) on the forms of depressive illness. In the present investigation, unlike those mentioned, a subjective approach was employed; that is, what the patients actually reported about themselves rather than a psychiatric appraisal of the patient was used as the data. A similar approach was employed by Pilowsky, Levine and Boulton (1969) in their taxonomic study of depression. The object of this paper is firstly, to investigate the systematic and the taxonomy of depressive phenomena; secondly to look at this taxonomy in depressive illness and non-depressive illness patients; and thirdly to investigate the psychiatrist's notion of the endogenous-reactive dichotomy within this taxonomy. Essentially the question is whether there is a continuum of depression, at one pole being reactive depression and at the other pole endogenous depression or whether these two classical diagnoses represent separate empirical entities. The former view is taken by the London workers and the latter by the Newcastle workers. In a recent review article, Eysenck (1970) favoured the binary or separate entity model, although he suggests that both views of depressive illness are over simplistic in their formulations.