Hitherto we have simply talked of 'depression' and ignored characteristics of the condition itself. The fact that provoking agents play the same role for onset cases in the general population and for the more severe in-patient cases suggests that they do not directly influence severity of depression. But we have also shown that vulnerability factors playa role in the presence of a provoking agent in determining whether a woman develops a case or borderline case condition and here vulnerability factors could be said in some sense to playa role in determining severity. But these points have been thrown up in the course of other analyses. We need to consider the question directly. Are there symptom-formation factors that influence the form and severity of depressive conditions? Are these distinct from the aetiological factors so far established?