The ventilatory effects of a propofol infusion were studied in 10 females premedicated with atropine and nine with papaveretum and atropine. The infusion, at a rate of 20 mg kg-1 h-1 for 5 min, reducing to 12 mg kg-1 h-1 for 10 min and then 6 mg kg-1 h-1 thereafter, was known to produce a steady-state plasma propofol concentration for 20-25 min after 25 min from commencement. Minute ventilation, tidal volume, frequency and response to breathing carbon dioxide were measured before the infusion and during the steady-state period. Propofol decreased minute ventilation to 56% and 46% (P less than 0.01) of their mean control values in the atropine and papaveretum groups, respectively. Mean tidal volume was decreased to 41-44% (P less than 0.02) by propofol, but a tachypnoea observed in the atropine group during the infusion was absent in the papaveretum group. Propofol alone had no effect on the slopes of the carbon dioxide response curves but did produce a shift to the right (P less than 0.05). Following papaveretum premedication the minute ventilation-carbon dioxide response curve slope, was decreased to 55% of its mean control volume value by the infusion, but this failed to reach statistical significance.