A decrease in maximal O2 uptake has been demonstrated with increasing altitude. However, direct measurements of individual links in the O2 transport chain at extreme altitude have not been obtained previously. In this study we examined eight healthy males, aged 21-31 yr, at rest and during steady-state exercise at sea level and the following inspired O2 pressures (PIO2): 80, 63, 49, and 43 Torr, during a 40-day simulated ascent of Mt. Everest. The subjects exercised on a cycle ergometer, and heart rate was recorded by an electrocardiograph; ventilation, O2 uptake, and CO2 output were measured by open circuit. Arterial and mixed venous blood samples were collected from indwelling radial or brachial and pulmonary arterial catheters for analysis of blood gases, O2 saturation and content, and lactate. As PIO2 decreased, maximal O2 uptake decreased from 3.98 ±0.20 l/min at sea level to 1.17 ± 0.08 l/min at PIO2 43 Torr. This was associated with profound hypoxemia and hypocapnia; at 60 W of exercise at PIO2 43 Torr, arterial PO2 = 28 ± 1 Torr and PCO2 = 11 ± 1 Torr, with a marked reduction in mixed venous PO2 [14.8 ± 1 (SE) Torr]. Considering the major factors responsible for transfer of O2 from the atmosphere to the tissues, the most important adaptations occurred in ventilation where a fourfold increase in alveolar ventilation was observed. Diffusion from alveolus to end-capillary blood was unchanged with altitude. The mass circulatory transport of O2 to the tissue capillaries was also unaffected by altitude except at PIO2 43 Torr where cardiac output was increased for a given O2 uptake. Diffusion from the capillary to the tissue mitochondria, reflected by mixed venous PO2, was also increased with altitude. With increasing altitude, blood lactate was progressively reduced at maximal exercise, whereas at any absolute and relative submaximal work load, blood lactate was higher. These findings suggest that although glycogenolysis may be accentuated at low work loads, it may not be maximally activated at exhaustion.