Prevailing concepts of pulmonary oxygen toxicity pathogenesis and hyperoxic tolerance have evolved over the past two decades. The “free radical” theory stating that lung cells poison themselves by producing an endogenous excess of reactive oxygen species (ROS) within various organelles had garnered considerable experimental support in the 1980s (1). The possible role of inflammation in compounding or propagating injury during sublethal hyperoxic exposure was also acknowledged. Subsequently, considerable evidence was obtained to support an association between increasing lung antioxidant enzyme activities and acquired tolerance to hyperoxia. Studies using pharmacologic supplementation and genetically altered mice have provided additional support for the importance of antioxidant enzymes, particularly the mitochondrial MnSOD. Nonetheless, certain inevitable intramitochondrial events, such as inactivation of aconitase with concomitant liberation of ferrous iron from the active site in the matrix and an associated decline in respiration (2), cannot be prevented by MnSOD induction via cytokines or genetic manipulation (3).