The bitewing radiographs of 1,731 English and Danish schoolchildren aged 13--15 years were examined to assess the prevalence of chronic periodontitis. Only one child (0.06%) was found to be affected in comparison with the high prevalence of 51.5% reported by Hull et al. (1975). Minute qualitative changes in the radiographic appearance of the alveolar crest, or changes in the visual representation of the width of the periodontal ligament space are not reliable or valid criteria for assessing chronic periodontitis. Horizontal bone loss observed on radiographs, which can be accurately measured using the cemento-enamel junction as a reference point, is a useful diagnostic criterion for the measurement of chronic periodontitis.