Cross‐sectional evidence suggests that maximal O2 uptake (VO2max) is the consequence of well‐matched O2 supply and demand in untrained skeletal muscle (symmorphosis) and in trained skeletal muscle is limited, not by mitochondrial O2 capacity, but, by convective and diffusive O2 transport (dysmorphosis). However, it is unknown if, longitudinally, aerobic exercise training transitions skeletal muscle VO2max from a sym‐ to a dys‐morphosis of O2 transport and utilization capacities. The purpose of this study was to test the hypothesis that, compared to normoxia, hyperoxia would increase skeletal muscle mean capillary PO2 (capPO2) and intracellular PO2 (iPO2) both pre‐ and post‐training, but would only increase VO2max post‐training. We tested this in 5 healthy sedentary males (mean±SD; age: 27±5 yrs, height: 175±6 cm, mass: 73±7 kg, cycle VO2max: 33±4 ml/kg/min) using the unique combination of human quadriceps single‐leg knee‐extensor exercise (KE), directly measured arterial and femoral PO2, thermodilution measured leg blood flow (LBF), and 1H nuclear magnetic resonance spectroscopy measurement of myoglobin (Mb) desaturation at 100% maximal work rate in randomized conditions of hypoxia, normoxia, and hyperoxia (12, 21, and 100 % O2, respectively). Mb desaturation was converted to iPO2 using an O2 half‐saturation pressure of 3.2 mmHg. All measurements were made pre‐ and post‐training, which consisted of eight weeks of KE (1 hour/visit, 3 visits/week). Pre‐training, increasing inspired O2 increased capPO2 (33.0±4.8, 41.4±4.1, and 49.6±8.7 mmHg, all p<0.019), Mb desaturation decreased from hypoxia to normoxia (46±4 and 31±4 %, p=0.004), but not from normoxia to hyperoxia (29±14 %, p=0.79), iPO2 increased from hypoxia to normoxia (3.8±0.6 and 7.2±1.4 mmHg, p=0.009), but not from normoxia to hyperoxia (10.0±6.6 mmHg, p=0.375), and VO2max was not altered by inspired O2 (0.47±0.10, 0.52±0.13, and 0.54±.01 l/min, all p>0.285). Post‐training, increasing inspired O2 increased capPO2 (32.8±2.6, 39.6±2.8, and 47.5±6.1 mmHg, all p<0.010), did not decrease Mb desaturation (79±20, 53±11, and 53±8 %, all p>0.050), iPO2 increased from hypoxia to normoxia (1.1±1.2 and 3.1±1.1 mmHg, p=0.019), but not from normoxia to hyperoxia (3.0±1.1 mmHg, p=0.954), and VO2max increased with greater inspired O2(0.59±0.11, 0.68±0.11, and 0.76±0.09 l/min, all p<0.044). There was a pre‐ to post‐training difference within all three inspired O2 conditions for VO2max (all p<0.010) and in hypoxia and normoxia for both Mb desaturation (both p<0.020) and iPO2 (both p<0.011). When untrained, the concomitant increase in capPO2 and iPO2, with no change in VO2max, from normoxia to hyperoxia infers a mitochondrial O2 capacity limited skeletal muscle VO2max. In trained skeletal muscle, the concomitant increase in capPO2 and VO2max, with minimal changes in iPO2, from normoxia to hyperoxia infers an O2 transport limited VO2max.