We have studied the effect of bolus doses of alfentanil on the arousal response to intubation, using the auditory evoked response (AER) of the electroencephalogram (EEG) in 45 anaesthetized patients. After induction of anaesthesia with propofol and 20 min of positive pressure ventilation via a laryngeal mask airway with 0.6 MAC of isoflurane and 50% nitrous-oxide in oxygen, patients received saline (control), or alfentanil 15 or 50 micrograms kg-1 before intubation. The early cortical AER before and after intubation in each group was measured and compared. After intubation, Pa amplitude increased in the saline group by 60% (95% CI +10 to +130), decreased in the low-dose alfentanil group by 11% (-38 to +29) and decreased further in the high-dose alfentanil group by 26% (-49 to +7). There were significant (P < 0.005) linear trends in the three group means for Pa amplitude. Similarly, Nb latency increased in the saline group by 30% (-2 to +73), decreased in the low-dose alfentanil group by 10% (-32 to +19) and decreased further in the high-dose alfentanil group by 19% (-39 to +7). There were significant (P = 0.02) linear trends in the three group means for Nb amplitude. Systolic and diastolic arterial pressures increased after intubation in the control group, and to a lesser extent in the low-dose alfentanil group, but decreased after high-dose alfentanil. Heart rate increased after intubation in the control group but decreased in both alfentanil groups, decreasing to a greater degree in the high-dose group. There were significant (P < 0.001) linear trends for all three haemodynamic variables in the three groups. We conclude that the increase in Pa amplitude after intubation was blocked by bolus administration of alfentanil and that this effect was dose dependent.