The effect of eseroline on the normoxic hypercapnic ventilatory response was assessed in nine alpha-chloralose-urethane-anaesthetized cats. The ventilatory responses to step changes in end-tidal PCO2 were determined before (control), during i.v. infusion of eseroline (bolus of 1.2 mg.kg-1 followed by 0.65 mg.kg-1 x h-1) and 1 h after the end of the infusion. Each response was separated into central and peripheral chemoreflexes, characterized by CO2 sensitivity, time constant, time delay and apnoeic threshold. We found that eseroline depressed ventilation by affecting both tidal volume and breathing frequency. The ventilatory response to CO2 was depressed due to a decrease in the CO2 sensitivity of peripheral chemoreceptors from 0.20 to 0.12 l.min-1 x kPa-1 and in the CO2 sensitivity of central chemoreceptors from 1.04 to 0.50 l.min-1 x kPa-1 (P < 0.01). However, the ratio of these sensitivities was not changed, like the apnoeic threshold. The depressant effect was reversed by naloxone. We conclude that the depressant effect of eseroline on ventilatory response to CO2 is mainly due to an action on the respiratory integrating centres in the brainstem rather than on the CO2 sensitivity of peripheral and central chemoreceptors.