Six self-rating mood scales were administered to a group of psychiatric inpatients. Two main factors emerged from this study, factors similar to the two broad dimensions of psychiatric disturbance derived by Welsh from the Minnesota Multiphasic Personality inventory. These factors did not correspond to the clinical dimensions of anxiety and depression; most of these scales related closely to each other as part of the first factor. Further, the intercorrelation between the scores obtained on the individual scales were, in the main, all high. This inability to demonstrate that these scales could separate patients into two groups (anxious and depressed) raises important questions about their use and the relationship between depression and anxiety.