Previous studies have reported significant region-dependent differences in the fiber-type composition of human skeletal muscle. It is therefore hypothesized that there is a difference between the deep and superficial parts of muscle energy metabolism during exercise. We hypothesized that the inorganic phosphate (Pi)/phosphocreatine (PCr) ratio of the superficial parts would be higher, compared with the deep parts, as the work rate increases, because the muscle fiber-type composition of the fast-type may be greater in the superficial parts compared with the deep parts. This study used two-dimensional 31Phosphorus Chemical Shift Imaging (31P-CSI) to detect differences between the deep and superficial parts of the human leg muscles during dynamic knee extension exercise. Six healthy men participated in this study (age 27±1 year, height 169.4±4.1 cm, weight 65.9±8.4 kg). The experiments were carried out with a 1.5-T superconducting magnet with a 5-in. diameter circular surface coil. The subjects performed dynamic one-legged knee extension exercise in the prone position, with the transmit-receive coil placed under the right quadriceps muscles in the magnet. The subjects pulled down an elastic rubber band attached to the ankle at a frequency of 0.25, 0.5 and 1 Hz for 320 s each. The intracellular pH (pHi) was calculated from the median chemical shift of the Pi peak relative to PCr. No significant difference in Pi/PCr was observed between the deep and the superficial parts of the quadriceps muscles at rest. The Pi/PCr of the superficial parts was not significantly increased with increasing work rate. Compared with the superficial areas, the Pi/PCr of the deep parts was significantly higher (p<0.05) at 1 Hz. The pHi showed no significant difference between the two parts. These results suggest that muscle oxidative metabolism is different between deep and superficial parts of quadriceps muscles during dynamic exercise.