The ability to sense changes in oxygen tension is shared by a variety of tissues, though there is evidence to suggest that the same mechanisms and transduction pathways are not employed by all. The carotid body glomus cell, neuroepithelial cells of the lung and pulmonary vascular smooth muscle all respond to an acute fall in oxygen with a rapid rise in intracellular calcium, with a consequent increase in efferent nerve activity or hypoxic pulmonary vasoconstriction (HPV) respectively. These mechanisms are primarily focused on matching and optimising gas exchange in the lung to metabolism. More prolonged hypoxia affects gene transcription in many tissues, often mediated via HIF, leading to adaptive alterations in protein expression and often modulation of cell proliferation and tissue remodelling. Changes in oxygen tension from the norm, either up or down, may therefore have profound effects on development of the fetus. For example, the relative hyperoxia suffered by ventilated premature neonates is believed to contribute to the development of bronchopulmonary dysplasia. In the adult, chronic hypoxic lung disease can lead to pulmonary vascular remodelling and pulmonary hypertension. Despite numerous studies, the cellular mechanisms responsible for oxygen sensing remain elusive, although a consensus is emerging that electron transport chain (ETC) of the mitochondrion plays a central role, at least in the carotid body and pulmonary vasculature. However both the location and mechanism of the ETC sensor, and the signal transduction pathways linking the mitochondria to the rise in intracellular calcium, remain highly controversial. There are three main competing hypotheses, all essentially based on an hypoxia-induced reduction in electron flux through the ETC: The first hypothesis is that during hypoxia there is a fall in production of reactive oxygen species by the ETC and a reduced cytosolic redox state; potassium channels are therefore reduced and inhibited, leading to depolarisation (e.g. Michelakis et al., 2002). In direct contrast, the second hypothesis proposes that hypoxia causes an increase in generation of ROS from complex III of the ETC, which then acts as the signalling moiety (e.g. Waypa et al., 2002). Although the precise cellular targets of the latter are unclear, several potential candidates have been suggested (Ward et al., 2004). The third hypothesis essentially proposes that signalling is related, either directly or indirectly, to a fall in activity of the ATP F1F0 synthase and ATP production (e.g. Wyatt & Buckler, 2004). This last hypothesis is largely based on studies of the glomus cell, which may well differ from pulmonary vascular smooth muscle. Different predictions can be made for each hypothesis, in particular concerning the effects of mitochondrial inhibitors acting at different points in the ETC. Our own data on HPV in isolated small pulmonary arteries would largely tend to support hypothesis 2, with signalling derived from complex III of the ETC (Ward et al., 2004). ROS have also been implicated in the effects of changes in oxygen tension on proliferation of human fetal airway smooth muscle, but here the relationship may be even more complicated. Both oxygen tension and exogenous peroxide caused a bell shaped response in proliferation, with a peak corresponding to normoxic conditions for adults, but which is significantly hyperoxic for the fetus (Pandya et al., 2002). This and other studies raise questions over the therapeutic use of antioxidants in premature babies, as they could potentially interfere with normal development.