Oxidative stress (OS) is widely believed to be responsible for the generation of trade-offs in evolutionary ecology by means of constraining investment into a number of components of fitness. Yet, progress in understanding the true role of OS in ecology and evolution has remained elusive. Interpretation of current findings is particularly hampered by the scarcity of experiments demonstrating which of the many available parameters of oxidative status respond most sensitively to and are relevant for measuring OS. We addressed these questions in wild-caught captive greenfinches (Carduelis chloris) by experimental induction of OS by administration of the pro-oxidant compound paraquat with drinking water. Treatment induced 50% mortality, a significant drop in body mass and an increase in oxidative DNA damage and glutathione levels in erythrocytes among the survivors of the high paraquat (0.2 g l(-1) over 7 days) group. Samples taken 3 days after the end of paraquat treatment showed no effect on the peroxidation of lipids (plasma malondialdehyde), carbonylation of proteins (in erythrocytes), parameters of plasma antioxidant protection (total antioxidant capacity and oxygen radical absorbance), uric acid or carotenoids. Our findings of an increase in one marker of damage and one marker of protection from the multitude of measured variables indicate that detection of OS is difficult even under the most stringent experimental induction of oxidative insult. We hope that this study highlights the need for reconsideration of over-simplistic models of OS and draws attention to the limitations of detection of OS due to time-lagged and hormetic upregulation of protective mechanisms. This study also underpins the diagnostic value of measurement of oxidative damage to DNA bases and assessment of erythrocyte glutathione levels.