Tumour oxygenation and bioenergetic status were measured in the same tumour and these results related to radiobiological hypoxia. A C3H mouse mammary carcinoma grown in the feet of CDF1 mice was used. Bioenergetic status was assessed by 31P MRS using a SISCO 7 Tesla magnet, oxygen measurements were done by a polarographic electrode and the hypoxic fraction was determined from direct analysis of the radiation dose-response data. During all examinations restrained, non-anaesthetized mice were allowed to breathe either 100% oxygen, carbogen, normal air, carbon monoxide (CO) at 75, 220, or 660 ppm or had blood flow occluded by clamping. Results showed a significant correlation between the radiobiological hypoxic fraction and % pO2 < or = 5 mmHg under the different treatment conditions, whereas no correlation was found between beta nucleosidetriphosphate/inorganic phosphate (beta-NTP/Pi) ratio and either the hypoxic fraction or the % of pO2 values < or = 5 mmHg under the different treatment conditions. In conclusion, oxygen electrode measurements were sensitive to changes in tumour hypoxia whereas the bioenergetic status alone seemed to be a less precise measure of hypoxia in this tumour model. Furthermore, the present study demonstrated that tumour cells in vivo can actually maintain the bioenergetic status during a period of severe hypoxia.