An experimental study was undertaken in an attempt 1) to describe the possibilities to detect lingual root resorptions of upper incisors in intraoral radiographs and 2) to determine the amount of reduction that is necessary in order to detect resorptions in the radiograph. The upper right central incisor was extracted from a dry skull. In a standardized way, the root was repeatedly reduced by grinding on the lingual side at an angle of 16 degrees measured from apex to the long axis of the tooth. After each reduction the tooth was replaced in the dry skull and radiographed in a standardized manner. The radiographs were placed in two different charts and at random within each chart. On two different occasions three groups of observers, 5 orthodontists, 5 pedodontists and 5 radiologists were asked to register any detected "root resorptions" on the right first and second incisors. The first chart included radiographs with a reduction of 0-3 mm and the second chart 0-6 mm. The orthodontists and pedodontists were not very consistent in their observations and false positive registrations were very frequent. The radiologists were more accurate in their observations and detected resorptions when the reduction incorporated the apex of the root at a 3 mm level. Thus, intraoral radiographs seems to be a very blunt instrument for detecting lingual root resorptions.