We studied the success rates for tracheal intubation in 64 healthy patients during simulated grade III laryngoscopy after induction of anaesthesia, using either the single-use bougie or oral flexible intubating fibrescope, both in conjunction with conventional Macintosh laryngoscopy. Patients were randomly allocated to either simulated grade IIIa or grade IIIb laryngoscopy, and also to one of the two study devices. Success rates for tracheal intubation (primary outcome measure) and times taken to achieve intubation (secondary outcome measure) were recorded. For the simulated grade IIIa laryngoscopy group, the fibreoptic scope was more successful than the bougie (16/16 successful intubations vs. 8/16; p = 0.02). For the simulated grade IIIb laryngoscopy group, the fibreoptic scope was also more successful than the bougie (8/16 successful intubations vs. 1/16; p = 0.02), but clearly use of the fibreoptic scope was not as successful as it had been in simulated grade IIIa laryngoscopy (p = 0.04). With either device, median (range) total tracheal intubation times for successful attempts with either grade of laryngoscopy were less than 60 s (19-109) and there were no clinically important differences. We conclude that the fibrescope used in conjunction with Macintosh laryngoscopy is a more reliable method of tracheal intubation than the single-use bougie in both types of grade III laryngoscopy. This finding has implications for the management of patients in whom grade III laryngoscopy is encountered unexpectedly after induction of anaesthesia, and also for the management of patients previously known to have grade III view at laryngoscopy.