The purpose of this study was to examine the development of fatigue in isolated, single skeletal muscle fibers when O2 availability was reduced but not to levels considered rate limiting to mitochondrial respiration. Tetanic force was measured in single living muscle fibers (n = 6) from Xenopus laevis while being stimulated at increasing contraction rates (0.25, 0.33, 0.5, and 1 Hz) in a sequential manner, with each stimulation frequency lasting 2 min. Muscle fatigue (determined as 75% of initial maximum force) was measured during three separate work bouts (with 45 min of rest between) as the perfusate PO2 was switched between values of 30 +/- 1.9, 76 +/- 3.0, or 159 Torr in a blocked-order design. No significant differences were found in the initial peak tensions between the high-, intermediate-, and low-PO2 treatments (323 +/- 22, 298 +/- 27, and 331 +/- 24 kPa, respectively). The time to fatigue was reached significantly sooner (P < 0.05) during the 30-Torr treatment (233 +/- 39 s) compared with the 76- (385 +/- 62 s) or 159-Torr (416 +/- 65 s) treatments. The calculated critical extracellular PO2 necessary to develop an anoxic core within these fibers was 13 +/- 1 Torr, indicating that the extracellular PO2 of 30 Torr should not have been rate limiting to mitochondrial respiration. The magnitude of an unstirred layer (243 +/- 64 micron) or an intracellular O2 diffusion coefficient (0.45 +/- 0.04 x 10(-5) cm2/s) necessary to develop an anoxic core under the conditions of the study was unlikely. The earlier initiation of fatigue during the lowest extracellular PO2 condition, at physiologically high intracellular PO2 levels, suggests that muscle performance may be O2 dependent even when mitochondrial respiration is not necessarily compromised.