The relationship between the plasma concentration of morphine and morphine-induced changes in ventilation and the ventilatory response to carbon dioxide was studied in 17 healthy adults undergoing elective surgery under general anaesthesia. Each subject was given morphine sulphate 0.15 mg kg-1 i.m.; ventilation (Ve), end-tidal PCO2(PE'CO2), mixed venous PCO2(PVCO2) and ventilatory response to carbon dioxide (delta Ve/deltaPCO2) were measured before and within 90 min after injection. Mixed venous PCO2 and deltaVe/deltaPCO2 were measured by standard rebreathing methods; plasma morphine concentration was measured by radioimmunoassay. Maximum plasma morphine ranged from 30 to 120 ng ml-1, between 4 and 60 min after injection. There was a significant increase in mixed venous PCO2 (P less than 0.001), and PE'CO2 (P less than 0.01) after morphine while Ve decreased insignificantly. Morphine displaced the carbon dioxide response curve to the right (P less than 0.01) and delta Ve/delta PCO2 decreased from 12.3 to 10.0 litre min-1 kPa-1 (P less than 0.05). The magnitude of changes in Ve and deltaVe/deltaPCO2 were not related to the peak plasma concentration of morphine or to the mean concentration immediately before and after the carbon dioxide response measurement. Plasma concentrations of morphine, under the conditions of the present study, are not an objective indicator of pharmacological activity between one patient and another.