Schizoaffective disorder is relatively common in clinical settings, with studies suggesting that the disorder accounts for 10-30% of admissions to inpatients mental health facilities for functional psychosis. Lifetime prevalence is estimated to be between 0.5 and 0.8%. Since its initial description by Kasanin in 1933, the concept has undergone many changes, moving on from that of acute psychosis toward a position located between affective and schizophrenic disorder. In current classification systems this position seems to be closer to the latter one. However DSM IV and ICD-10 differ according to the nature of symptoms, especially the psychotic ones, and the temporal relationship of the psychotic symptoms to the mood symptoms. Psychotic features may include paranoid, cognitive and negative symptoms but mood-congruent and mood-incongruent psychotic features can also be found during mood episodes. The disorder has an episodic course and a prognosis usually intermediate between that of schizophrenia and affective disorder. It has been commonly subtyped into a bipolar and a depressive type, but more recently a division into a concurrent and a sequential type has been proposed too, based on longitudinal findings. Family studies using current diagnostic criteria show that the pattern of illness in relatives suggests that schizoaffective probands may have as high a familial liability to schizophrenia and affective illness as do those probands with only one of the disorders. In addition to biological and neuroendocrine investigation, schizoaffective disorder has been addressed in some neuroimaging, neuropathological, cognitive and neurodevelopmental studies which tend however to confirm its location between the two disorders, but closer to the schizophrenic pole. Several hypotheses have been proposed to account for the nature of schizoaffective disorder. Current data tend to favor the ranging of the disorder within a unitary spectrum of functional psychosis the diathesis of which could be activated by an episode of mood disorder. In contrast to the extensive research devoted to the development of pharmacologic agents for the treatment of schizophrenia and mood disorders, the optimal pharmacological treatment of schizoaffective disorder has been relatively unstudied which may, for instance, explain the lack of treatment guidelines. Nonetheless, despite the paucity of data in this area, some authors recommend the use of atypical anti-psychotics alone or in combination treatment as first line medi-cations both for the short and long term treatment. Research on schizoaffective disorder has been marred by the variability of its definition. Even if doubts have been expressed about the reliability of current diagnostic criteria, they seem to define a syndrome that differs meaningfully from both schizophrenia and affective disorder. Further studies are however still necessary, especially with regard to the subtyping of the disorder and its pharmacological treatment.