The present study tests the hypothesis that reoxygenation after cerebral tissue hypoxia with 100% oxygen increases generation of oxygen free radicals compared to reoxygenation with room air. Studies were performed in anesthetized, ventilated newborn piglets subjected to 1 hr of hypoxia(FiO2 0.05-0.07, PaO2 < 20 mmHg). Following hypoxia, animals were randomized to Group 1 (reoxygenation with 21% O2, n=8) or Group 2(reoxygenation with 100% O2, n=8). Near infrared spectroscopy (NIRS) was used with wavelengths of 716 nm and 850 nm to monitor tissue oxygenation using a phase modulation spectrometer with an interprobe distance of 22 mm. A phase shift in the NIRS signal was observed at 50±10 sec following initiation of reoxygenation. Cortical biopsies were obtained via cranial windows for free radical measurements using electron spin resonance spectroscopy (ESR) immediately following appearance of the phase shift. ESR spectra of electron spin adducts were obtained in duplicate using a Varian-E-109 spectrometer. Signal heights were measured from the resultant graph and divided by tissue weight. Reference ESR signals were measured from tissues incubated at room temperature for 10 min to confirm the absence of free radicals. Tissue was homogenized in 100 mMα-phenyl-tert-butyl-nitrone (PBN) and extracted with 2:1 chloroform:methanol. Intensity of the ESR signal from the PBN extract was expressed as signal height (mm)/g tissue after subtracting the reference value determined in control animals. Free radical signal height was 130±30 mm/g brain in Group 1 and 230±90 mm/g brain in Group 2 (p <0.05). Reoxygenation with 100% O2 was associated with significantly more free radical generation than reoxygenation with 21% O2. We speculate that mitochondria which were anaerobic during the hypoxic phase produce a surge of free radicals during reoxygenation and that the amount of free radicals produced increases with increased oxygen concentration. (NIH#HD-20337 and AAP/AHA/Neonatal Resuscitation Program)