Histological methods for estimating age at death using osteon population densities for the rib, clavicle, and rib and clavicle combined are presented. Predicting formulas were generated from a sample of 40 individuals of known age, sex, and race. Independent samples of 12 ribs and 7 clavicles were used to test the formulas. Mean differences between known and predicted ages were 1.1 years, 2.6 years, and 3.4 years for the clavicle, rib and clavicle combined, and rib formulas respectively. An analysis of variance found no significant differences among the means for predicted and known ages. Since the formula based upon rib and clavicle combined has the higher standard error and r2, and includes data from different bones, it should provide better overall accuracy and reliability, and is recommended whenever both bones are available.