Three years ago, at the VIIth International Congress of Radiology in Otorhinolaryngology, held in Copenhagen, the early impressive results of computer-assisted tomography (CAT) in otorhinolaryngology were presented. Since considerable technologic and diagnostic progress of CAT has taken place in the meantime, questions about the increase in CAT's value and the expansion of its limits are appropriate now. Computed and, in some cases, conventional tomograms of facial lesions are used in the discussion of these questions. There is no doubt of CAT's advance within the last few years; there are, however, limits of CAT to be emphasized: (a) coronal CAT might be uncomfortable (especially if the gantry cannot be tilted), unsuccessful (e.g., in cases of neck stiffness), and even dangerous (when head injury is combined with trauma to the cervical spine); (b) enhancement is often not useful (except when intracranial tumor invasion is suspected); (c) specific diagnosis, based on CAT findings only, is impossible in most cases; (d) the radiation dose, which increases with improvement in the quality of the image, must be considered in each case; and (e) CAT does not replace conventional x-ray techniques, least of all polytomography.