Developing teeth are used to assess maturity and estimate age in a number of disciplines, however the accuracy of different methods has not been systematically investigated. The aim of this study was to determine the accuracy of several methods. Tooth formation was assessed from radiographs of healthy children attending a dental teaching hospital. The sample was 946 children (491 boys, 455 girls, aged 3-16.99 years) with similar number of children from Bangladeshi and British Caucasian ethnic origin. Panoramic radiographs were examined and seven mandibular teeth staged according to Demirjian's dental maturity scale [A. Demirjian, Dental development, CD-ROM, Silver Platter Education, University of Montreal, Montreal, 1993-1994; A. Demirjian, H. Goldstein, J.M. Tanner, A new system of dental age assessment, Hum. Biol. 45 (1973) 211-227; A. Demirjian, H. Goldstein, New systems for dental maturity based on seven and four teeth, Ann. Hum. Biol. 3 (1976) 411-421], Nolla [C.M. Nolla, The development of the permanent teeth, J. Dent. Child. 27 (1960) 254-266] and Haavikko [K. Haavikko, The formation and the alveolar and clinical eruption of the permanent teeth. An orthopantomographic study. Proc. Finn. Dent. Soc. 66 (1970) 103-170]. Dental age was calculated for each method, including an adaptation of Demirjian's method with updated scoring [G. Willems, A. Van Olmen, B. Spiessens, C. Carels, Dental age estimation in Belgian children: Demirjian's technique revisited, J. Forensic Sci. 46 (2001) 893-895]. The mean difference (+/-S.D. in years) between dental and real age was calculated for each method and in the case of Haavikko, each tooth type; and tested using t-test. Mean difference was also calculated for the age group 3-13.99 years for Haavikko (mean and individual teeth). Results show that the most accurate method was by Willems [G. Willems, A. Van Olmen, B. Spiessens, C. Carels, Dental age estimation in Belgian children: Demirjian's technique revisited, J. Forensic Sci. 46 (2001) 893-895] (boys -0.05+/-0.81, girls -0.20+/-0.89, both -0.12 y+/-0.85), Demirjian [A. Demirjian, Dental development, CD-ROM, Silver Platter Education, University of Montreal, Montreal, 1993-1994] overestimated age (boys 0.25+/-0.84, girls 0.23+/-0.84, both 0.24 y+/-0.86), while Nolla [C.M. Nolla, The development of the permanent teeth, J. Dent. Child. 27 (1960) 254-266] and Haavikko's [K. Haavikko, The formation and the alveolar and clinical eruption of the permanent teeth. An orthopantomographic study, Proc. Finn. Dent. Soc. 66 (1970) 103-170] methods under-estimated age (boys -0.87+/-0.87, girls -1.18+/-0.96, both -1.02 y+/-0.93; boys -0.56+/-0.91, girls -0.79+/-1.11, both -0.67 y+/-1.01, respectively). For individual teeth using Haavikko's method, first premolar and second molar were most accurate; and more accurate than the mean value of all developing teeth. The 95% confidence interval of the mean was least for mean of all developing teeth using Haavikko (age 3-13.99 years), followed by identical values for Demirjian and Willems (sexes combined).