We assessed the central-chemoreflex ventilatory responses to carbon dioxide in six male volunteers using a hyperoxic rebreathing technique. Hyperventilation prior to rebreathing allowed both the threshold and the sensitivity of the response to be measured. We used immersion in water to control the body temperature (tympanic). The water temperature was adjusted to be either thermo-neutral or hot so that body temperature either remained normal [+0.2 (0.04) degrees C, mean (SEM)] or was elevated by 1.5 (0.08) degrees C. The sensitivities of the central-chemoreflex ventilatory responses to carbon dioxide were increased at elevated body temperatures, changing from a mean of 1.8 (0.2) 1.min-1. Torr-1 to 2.7 (0.1) 1.min-1. Torr-1. However, the thresholds did not change with temperature, and the mean threshold was 48(1) Torr at both normal and elevated temperatures. For all of the volunteers, ventilation was increased at elevated body temperatures for all levels of carbon dioxide, mainly by changes in respiratory frequency due to reductions in expiratory times. At subthreshold levels of carbon dioxide, mean ventilation changed significantly from 6.3(1.1) 1.min-1 at normal temperatures to 10.8 (1.9) 1.min-1 at elevated temperatures. Heart rates also increased significantly with temperature, changing from a mean of 66 (4) beats.min-1 to 102 (3) beats.min-1 at threshold levels of carbon dioxide. The mean rates of rise of carbon dioxide partial pressure during rebreathing were significantly increased with temperature as well, changing from 0.075 (0.008) Torr.min-1 to 0.089 (0.004) Torr.min-1. We concluded that elevating the body temperatures of our subjects not only increased their ventilation, heart rates and metabolic rates at all levels of carbon dioxide, but it also increased the sensitivity of their central chemoreflex ventilatory responses to carbon dioxide. Despite these increases, the thresholds of the central-chemoreflex ventilatory responses to carbon dioxide remained unchanged.