BACKGROUND, AIMS: In 1973, a random sample of 574 dentate individuals aged 15, 20, 30, 40, 50, and 60 years in the city of Jönköping, Sweden, were examined clinically and radiographically to assess oral health and overall treatment needs. Periodontal examination included registration of plaque, gingivitis, probing depths at four aspects of each tooth, and interproximal bone height measurements on full-mouth intraoral radiographs. In 1990, 17 years later, the same individuals were invited to participate in a new investigation. Of these, 433 (75%) agreed to participate in the investigation and were re-examined (Hugoson & Laurell 2000). The proximal alveolar bone height at all interproximal sites was measured and expressed as per cent of tooth length. Only teeth that were present in both 1973 and 1990 were included in the assessment of changes in bone score. From the age of 30 years, about 80% of the population had one or more sites with a bone loss of 2-3 mm or more. Seventeen per cent of the individuals had more than six such sites, indicating destructive periodontal disease. Bone loss occurred at sites both with and without previous bone loss. The present study was undertaken to test the hypothesis that sites with a bone loss of 10% or more of the tooth length (2-3 mm) during the 17 years were randomly distributed in the dentition. MATERIAL AND METHODS: Of the 13,197 sites examined in individuals 20-60 years at baseline, 1201 sites (9.0%) in 998 teeth with a bone loss corresponding to 10% or more of the tooth length were found and included in the analysis. A probability test for binomial distribution was used to test the null hypothesis that all teeth had the same risk of losing bone regardless of its position in the dentition. The valid risk for each tooth was 3.571% and the null hypothesis was rejected at the 95% confidence interval. RESULTS: Although all tooth types were affected by tooth loss, some teeth, namely 17, 16, 42, 41, and 31, showed a higher incidence of sites losing bone, whereas 46, 45, 44, and 36 had a lower incidence. Loser sites in smokers appeared more at random. CONCLUSION: Sites that will develop periodontal break-down over time may appear at random, although with higher risk at maxillary molars and lower incisors. For the early detection of destructive periodontitis, periodontal examination that includes all teeth should be made routine in every dental check-up.