The aim of this study was to assess whether it is possible to derive accurate vertical measurements of the mandibular ramus and condyle from panoramic radiographs. A human dry skull was positioned in a panoramic machine. The skull was displaced along the sagittal and transverse plane and rotated around the vertical and transverse axes. A set of 252 digital radiographs with defined positioning errors was compared with a set of 42 radiographs in the 'ideal' position. The distances between the metal markers that had been attached at the angle of the mandible at a distance of 60 mm in the condyle region to produce fixed reference points on the radiographs were measured. Statistical differences were investigated using Friedman repeated measures analysis of variance on ranks followed by the Dunnett's test for the comparison against the control group in the ideal position (α = 0.05). Vertical measurements were significantly affected when the skull was rotated around the vertical (P < 0.001) or shifted along the transverse axis (P < 0.001). Misalignment of the head affected the vertical measurement of the mandibular ramus and condyle. However, asymmetries of more than 6 per cent are probably not due to patient positioning in the panoramic machine.