Hypoxic pulmonary vasoconstriction (HPV) becomes activated in response to alveolar hypoxia and, although the characteristics of HPV have been well described, the underlying mechanism of O(2) sensing which initiates the HPV response has not been fully established. Mitochondria have long been considered as a putative site of oxygen sensing because they consume O(2) and therefore represent the intracellular site with the lowest oxygen tension. However, two opposing theories have emerged regarding mitochondria-dependent O(2) sensing during hypoxia. One model suggests that there is a decrease in mitochondrial reactive oxygen species (ROS) levels during the transition from normoxia to hypoxia, resulting in the shift in cytosolic redox to a more reduced state. An alternative model proposes that hypoxia paradoxically increases mitochondrial ROS signalling in pulmonary arterial smooth muscle. Experimental resolution of the question of whether the mitochondrial ROS levels increase or decrease during hypoxia has been problematic owing to the technical limitations of the tools used to assess oxidant stress as well as the pharmacological agents used to inhibit the mitochondrial electron transport chain. However, recent developments in genetic techniques and redox-sensitive probes may allow us eventually to reach a consensus concerning the O(2) sensing mechanism underlying HPV.