Objective To investigate the age-related changes of biomechanical properties for humerus, femur and tibia in male rats and their application values in age estimation. Methods According to different weeks of age, 90 healthy male SD rats were divided into 2, 4, 6, 8, 17, 26, 52, 78 and 104-week groups with 10 rats in eachgroup. After the rats were executed by excessive anesthesia, humerus, femur, and tibia were separated and the attached soft tissues were removed. The length of the above-mentioned bones and the diameter of the middle section (compression site) were measured with vernier caliper, and the three-point bending test was conducted with electronic universal material testing machine to detect the ultimate load and displacement under ultimate load. Results There were significant differences in the ultimate load of humerus, femur and tibia among male rats in different age groups (P<0. 05). With the increase of week age, the ultimate loads of the humerus, femur and tibia increased first and then decreased, and reached the peak value in 52-week age group, showing a strong positive correlation with week age before 52 weeks ( r = 0. 884,0. 933,0. 929, P<0. 05). There was no significant difference in humerus and tibia. The displacement of femur under ultimate load was weakly positively correlated with week age (R= 0. 406,P<0. 05). The age prediction accuracy for automatic linear modeling of ultimate load for humerus, femur, tibia and three above-mentioned bones in rats before 52-week age was 78. 2% , 86. 8% , 84. 1% and 88. 3% , respectively. There was a strong positive correlation between the length of humerus, femur and tibia and the ultimate load (R= 0. 904, 0. 897, 0. 814, P<0. 05). The diameters of humerus, femur and tibia were strongly positively correlated with the ultimate load (R = 0. 759, 0. 814 and 0. 745, P<0. 05). Conclusions The ultimate loads of humerus, femur and tibia in male rats increased first and then decreased with age, and were positively correlated with age before 52 weeks, which could be used for age inference. The highest accuracy of age estimation was ultimate loads of three bones, followed by femur. The length/ middle diameter of humerus, femur and tibia were strongly positively correlated with the ultimate load.