Prediction of a difficult airway is a continuing problem. Several tests have been developed in order to overcome this problem. It has been proposed that using more than one test could increase the degree of predictability. One hundred patients scheduled for general surgery were successfully evaluated using three methods: 1. The size of the tongue in relation to the oral cavity (Mallampati gradation). 2. The thyromental and the mandibular length. 3. The atlanto-occipital angle. Seven patients had a difficult laryngoscopy and were difficult to intubate. It was possible to predict differenties in all seven patients, but it was found that only the Mallampati gradation and the atlanto-occipital angle contributed significantly to predictability. The cost of predicting further difficulties in all patients with a difficult laryngoscopy was a false positive prediction in a further 13 patients who had no problems with laryngoscopy or intubation. By increasing the specificity of the tests the number of false positive classifications could be reduced to 10 patients, but this would have led to the missing of one patient with a difficult laryngoscopy.