The hypothesis that oxidative stress can be induced by hypoxia was tested by measuring the concentration of hydrogen peroxide by a luminometric technique in the breath samples of rats exposed to hypoxia and paraquat. The group of animals (n=15) exposed to normobaric hypoxia (10% O2) for three days had an increased amount of H2O2 (200%, P<0.001) in their breath in comparison to control animals. After 7 days of recovery in air, the exposed animals still produced significantly increased levels of H2O2 (152%, P<0.001). Paraquat administration was used as a positive control, since it is a redox cycling compound producing free radicals. In the animals treated with a toxic dose of paraquat, the peak H2O2 production was observed 5 h after i.p. injection (156%, P<0.02). Within the next 2 h it decreased to the control level and stayed constant for 48 h, when the animals began to die. It is suggested that H2O2, observed in the breath samples, is a product of a metabolic pathway that could itself be sensitive to oxidative damage.