Chronological age cannot be inferred from somatic maturation in children. Alternate criteria, such as skeletal age, have been used by several methods, although sample heterogenicity adds questions about their reliability as chronological age predictors.To determine gender differences in skeletal age calculated by the Fishman Skeletal Maturity Assessment (SMA) and the Hägg-Taranger method, and to define in which of their indicators the difference with chronological ages was minimal.Skeletal age was assessed in 2751 hand-wrist films of 9- to 17-year- old Bogotanian children with the Fishman SMA, and a randomly selected subsample of 364 films with the Hägg-Taranger method. Matched-pair signed-rank tests were used for statistical analysis of data.Differences from zero (P < 0.001) between means of skeletal and chronological ages were found with the Hägg-Taranger (0.549) and with the Fishman SMA methods (0.38). As chronological age predictors, both methods showed high coefficients; however, the former method was higher (r2 = 0.732) than the latter (r2 = 0.680).Specific hand-wrist skeletal indicators are suitable only to girls with ages between 9 and 16 years and to boys between 10 and 17 years. Other methods should be considered for prediction of chronological age in younger children.Differences in skeletal age means with white individuals suggest the need of obtaining population standards separated by race and gender. Both methods are reliable as predictors of chronological age with greater accuracy of the Hägg-Taranger method.