Near-infrared spectroscopy is currently used to assess changes in the oxygen saturation of the muscle during exercise. The primary purpose of this study was to assess the reliability of near-infrared spectroscopy in determining muscle oxygen saturation (StO[subscript 2]) in the vastus lateralis during cycling and the gastrocnemius during running for exercise intensities at lactate threshold and maximal effort. Test-retest reliability was determined from an intraclass correlation coefficient obtained from a one-way analysis of variance. Reliability of muscle StO[subscript 2] for the gastrocnemius at lactate threshold was R = 0.87, and R = 0.88 at maximal effort. Reliability of muscle StO[subscript 2] for the vastus lateralis at lactate threshold was R = 0.94 and R = 0.99 at maximal effort. (Contains 4 tables and 3 figures.)