We selected nine normal subjects (8M, 1F; aged 25-43 yr) with brisk hypoxic ventilatory responses, and studied their ventilatory response to sustained isocapnic hypoxia (SaO2 82 (SEM 0.1) % for 25 min) in the presence and absence of 0.1% inspired halothane. Halothane had no significant effect on baseline ventilation or gas exchange. In the absence of halothane, ventilation increased initially from mean 7.57 (0.35) litre min-1 to 14.54 (0.91) litre min-1, and decreased subsequently to 10.74 (0.32) litre min-1 during hypoxia (both P < 0.05). In the presence of 0.1% inspired halothane, ventilation increased initially from 7.19 (0.47) litre min-1 to 12.08 (0.99) litre min-1 (P < 0.05), then decreased to 10.12 (0.28) litre min-1 during sustained hypoxia (ns compared with baseline normoxic ventilation). Halothane reduced significantly the initial increase in ventilation (P < 0.05), but did not enhance the subsequent decrease. These results confirm that a sub-anaesthetic concentration of halothane depresses the initial hypoxic ventilatory response; the response during prolonged periods of hypoxia is, however, less than the initial response and is reduced in the presence or absence of a sub-anaesthetic concentration of halothane.