A self-rating scale was developed to measure the severity of fatigue. Two-hundred and seventy-four new registrations on a general practice list completed a 14-item fatigue scale. In addition, 100 consecutive attenders to a general practice completed the fatigue scale and the fatigue item of the revised Clinical Interview Schedule (CIS-R). These were compared by the application of Relative Operating Characteristic (ROC) analysis. Tests of internal consistency and principal components analyses were performed on both sets of data. The scale was found to be both reliable and valid. There was a high degree of internal consistency, and the principal components analysis supported the notion of a two-factor solution (physical and mental fatigue). The validation coefficients for the fatigue scale, using an arbitrary cut off score of 3/4 and the item on the CIS-R were: sensitivity 75.5 and specificity 74.5.