In this study, 31P nuclear magnetic resonance spectroscopy was used to monitor muscle metabolism in Type II diabetic subjects (n = 10) during an incremental exercise test. Also the exercise responses of diabetic subjects (n = 4) following submaximal endurance training were assessed and compared to healthy controls (n = 5). Responses to incremental exercise in the diabetic subjects were consistent over time despite minor fluctuations in metabolic control. In the diabetic and control groups, after 12 weeks of training the forearm flexor muscles, power output at the intracellular threshold of acidosis (IT) increased (p < .01) similarly: T0 versus T12: 0.90 +/- 0.09 versus 1.20 +/- 0.13 and 1.03 +/- 0.07 versus 1.22 +/- 0.10 W, respectively. Minimum intracellular pH reached at peak exercise was unchanged after training. The control group, however, became more acidic versus the diabetic group (p < .05) in response to progressive exercise. This difference was maintained over time. Endurance training elicited similar adaptations in forearm muscles of Type II diabetic and control subjects, although there were differences between the two groups in intracellular pH during exercise.