Three hundred and fifty consecutive patients (322 non-obstetric, 28 obstetric; 185 female) were assessed before operation using the modified Mallampati test and by measuring thyromental and sternomental distances, forward protrusion of the mandible and interincisor gap with the mouth fully open. Tracheal intubation was difficult in 17 (4.9%) patients, of whom four (1.14%) had a grade III or IV view on laryngoscopy. A sternomental distance of 12.5 cm or less with the head fully extended on the neck and the mouth closed predicted 14 of the 17 patients in whom tracheal intubation was difficult. As a screening test, sternomental distance appeared to be more sensitive (82.4%) and more specific (88.6%) than thyromental distance (64.7% and 81.4%, respectively), the modified Mallampati test (64.7% and 66.1%, respectively) and forward protrusion of the mandible (29.4% and 85.0%, respectively). The interincisor gap ranged from 2.0 to 5.0 cm in all patients except one. There was no correlation between the interincisor gap and the view on laryngoscopy (P > 0.05, one-way ANOVA). There was also no difference in mean interincisor gap between those patients who presented no difficulty with tracheal intubation and those who did (P = 0.7-0.8, two sample t test). Sternomental distance may be a useful bedside screening test for preoperative prediction of difficult tracheal intubation.