The Attributional Style Questionnaire was given to three groups of 15 adult patients: a group of paranoid patients who were not depressed, a group of depressed patients who were not paranoid, and a group of patients who were both paranoid and depressed. As predicted, the paranoid patients manifested an attributional style opposite to that of the depressed patients: that is, they tended to attribute good events to themselves and bad events to others or to chance, whereas the depressed patients tended to attribute bad events to themselves and good events to others or to chance. These findings confirm those of Kaney & Bentall (1989). The patients who were both paranoid and depressed fell in between the two other groups with respect to their attributions of good events but did not differ from the paranoid group in their attributions of bad events. In addition, differences among groups were manifest with respect to self-esteem, delusional content and suicidal ideation. The implications of these findings for therapy are discussed.