We assessed the effects of combined metformin treatment and exercise training on body composition, on insulin concentration following glucose loading, on insulin-stimulated glucose transport in skeletal muscle, and on muscle glycogen content. Male Sprague-Dawley rats were treated for 35 days with or without metformin (320 mg/kg/day) and/or treadmill exercise training (20 min at 20 m/min, 5 days/wk). Because metformin reduces food intake, pair-fed controls were included. Metformin, training, and pair-feeding all decreased food intake, body weight, and insulin concentration following glucose loading. Metformin and training reduced intra-abdominal fat, but pair feeding did not. In isolated strips derived from soleus, epitrochlearis and extensor carpi ulnaris muscles, metformin increased insulin-stimulated transport of [3H]-2-deoxyglucose by 90%, 89% and 125%, respectively (P < 0.02) and training increased [3H]-2-deoxyglucose transport in the extensor carpi ulnaris muscle only (66%, P < 0.05). Pair-feeding did not alter [3H]-2-deoxyglucose transport. Training increased gastrocnemius muscle glycogen by 100% (P < 0.001). Metformin and pair-feeding did not alter muscle glycogen. We conclude that metformin reverses the maturation-induced impairment of insulin responsiveness in Sprague-Dawley rats by increasing insulin-stimulated glucose transport in skeletal muscle and that this effect is not secondary to reduced food intake. We also conclude that metformin and exercise training may increase insulin sensitivity by different mechanisms, with training causing increased glucose transport only in some muscles and also causing increased muscle glycogen storage.