Using five transplantable murine tumors (SCC-VII, B16F1, KHT-C, KHT-LP1, RIF-1), measurements of tumor hypoxia have been made with two techniques which have the potential to be used for assessing oxygenation in human tumors (the Eppendorf pO2 Histograph and binding of [3H]misonidazole) and have been compared with an established radiobiological technique, the paired survival assay. There were significant differences in the pO2 measurements made in individual tumors both within and between the five different tumor types. Significant differences between the tumor types were also found for the [3H]misonidazole binding. A correlation was observed between the mean values of the hypoxic proportion as measured by the paired survival assay and the mean binding of [3H]misonidazole as measured by both tumor activity in dpm/100 mg tissue (r = 0.94, P = 0.02) and the tumor-to-muscle activity ratio (r = 0.87, P = 0.06). No biologically significant correlation was seen between the mean values of the hypoxic proportion from anesthetized mice as measured by the paired survival assay (range 20-58%) and the pooled Eppendorf pO2 Histograph measurements made on groups of tumors. These results with the Eppendorf pO2 Histograph are similar to those reported by others. When both Eppendorf pO2 Histograph measurements and paired survival measurements were made on the same individual KHT-C tumors, it was again found that there was no correlation between the two measurements of hypoxia.