Objective: To better understand the association between skeletal muscle and exercise intolerance in peripheral arterial disease (PAD), we assessed treadmill-walking performance and gastrocnemius muscle phenotype in healthy control subjects and in patients with PAD. We hypothesized that gastrocnemius muscle characteristics would be altered in PAD compared with control subjects and that exercise tolerance in patients PAD would be related to muscle phenotype. Methods: Sixteen patients with PAD and intermittent claudication and 13 healthy controls of the same age participated. Each subject completed a graded treadmill-walking test and underwent a resting muscle biopsy. Muscle biopsy samples were obtained from the medial gastrocnemius muscle of the most ischemic limb in PAD and a limb chosen at random in controls. Samples were analyzed for fiber type and cross-sectional area, capillary-to-fiber ratio, the number of capillaries in contact with each fiber type, and the optical density of glycogen within each fiber by using histochemical procedures. Total muscle glycogen content was determined biochemically. Results: Exercise capacity measured on the incremental walking test in the PAD group was only 30% to 40% of that observed in controls. The PAD group had a lower proportion of type I muscle fibers (P < .05), fewer capillaries per muscle fiber (P < .05), and tended to have smaller fiber areas (P = .08). The relative area of type I fibers, the capillary-to-fiber ratio, capillary contacts with type I and IIa fibers, and the optical density of glycogen in type I fibers were all positively correlated with exercise tolerance in the PAD group(P < .05) but not controls. Conclusions: These data suggest that muscle phenotype is altered in PAD and that such alterations are associated with the exercise intolerance in these patients. In light of these findings, therapies such as resistance training or electrical stimulation that target skeletal muscle in PAD may prove beneficial, and further investigation of such therapies is warranted.