It has been hypothesized that mitral valve prolapse may account for a substantial number of patients who have symptoms of chronic anxiety neurosis. In a previous investigation, this hypothesis was confirmed in eight of 21 patients who had anxiety neurosis. In the present investigation, we reevaluated the hypothesis that persons with anxiety neurosis have impaired exercise ability by exercising 20 of the anxiety neurotics according to a standard treadmill exercise protocol. Compared with the control group, the anxiety neurotics required less exercise to achieve an equivalent heart rate and therefore their estimated maximum oxygen consumption was less, thus confirming the hypothesis. However, this difference was due entirely to the anxiety neurotics with mitral valve prolapse, and those without prolapse did not differ significantly from the controls. This suggests that impaired exercise tolerance in anxiety neurotics may be attributable to a subgroup of these patients with mitral valve prolapse.