ABSTRACT Furniture commonly fails at joints due to long-term repeated loads. This study aimed to investigate the effects of the selected factors (loading type, tenon fit, and load amplitude) on cyclic load resistance (CLR) and count of the cycle (COC) of mortise-and-tenon (M-T) joints made of beech (Fagus orientalis). The tests were conducted using a self-designed air-operated multi-function fatigue testing machine. The results showed that using a U-shaped circle to apply load on the stretcher was the optimal loading type for testing the cyclic performance of M-T joints. Tenon fit had a greater effect on CLR than loading type, and the effect of load amplitude on COC was more significant than those of loading type and tenon fit at 80% CLR. Although tenon fit had a significant effect on COC, the differences in COC among M-T joints with different tenon fits subjected to the same load amplitude were much greater than those at the same load amplitude. A strong exponential relationship was observed between COC and load amplitude. The fatigue limit load of the M-T joint evaluated was approximately 150 N or 47% CLR. This study provided an efficient method and extensive data for evaluating fatigue life of M-T joints.