Only very few longitudinal studies on dental caries are available in adults and elderly. Therefore, the aim of the present study was to describe the patterns of lesion progression in a group of adult and elderly Chinese followed over a 10–year period of time. In 1984, 587 persons aged 20–80 years were examined and by 1994, 440 of these were still available and re–examined by the same two examiners. All restorations, periodontal conditions and dental caries were recorded for all teeth and surfaces. Only caries data are reported here. Dental caries on coronal surfaces was classified as sound, non–cavitated, cavitated without or with pulp involvement or filled. On root surfaces, lesions were classified as sound, active caries or restored. Prior to analysis of the data, a series of drop–out analyses was carried out showing that most of the drop–outs were above 60 years of age at baseline, and death of participants was the major drop–out reason. A total of 2,114 teeth were lost, 36% of which had a deep dentin lesion at baseline. The distribution of persons according to number of new lesions/restorations over 10 years was highly skewed in all age groups, the prevalence increasing with age. Molars had the highest risk of becoming carious/restored over the 10–year period, and a remarkably low proportion of teeth had the same caries diagnosis after 10 years, indicative of lesion progression. When comparing the 1984 DFT/DMFT with the figures for 1994, there were no statistically significant differences indicating that the period and cohort effects cancel out each other. There was a significant increase with age of the number of new DFT. It can be concluded that dental caries is a major health problem in contemporary adult and elderly Chinese populations, and that caries incidence increases with old age.