Recent studies using reptiles and other ectothermic species have shown that hypoxia lowers the set-point for the control of body temperature. This is characterized by a preference for cooler ambient (T(sub a)) and deep body temperatures (T(sub b)) when placed in a temperature gradient. To elucidate the presence of this effect in mammals, the selected T(sub a) and T(sub b) of three rodent species, mouse, hamster, and rat, were measured while subjected to graded hypoxia in a temperature gradient. Individual animals were placed in the gradient for 30 min. Percentage O2 of air entering the gradient was then reduced to a constant level for a period of 60 min by dilution with nitrogen. T(sub b) was significantly reduced in all species at O2 levels of 5.5 to 10%. Selected T(sub a) was significantly reduced in the mouse at O2 levels of 5.5 and 7.3%. Selected T(sub a) of the hamster and rat were reduced slightly at percentage O2 levels of 5.8 and 7.4%, respectively; however, the effect was not statistically significant. Both species exhibited a significant reduction in selected T(sub a) during hypoxia concomitant with hypothermia. These data support the hypothesis that hypoxia lowers the set-point for the control of body temperature in rodents.