Understanding the inconsistent rock fracture toughness ( K Ic ) measurement results from different test specimen geometries helps provide suitable fracture parameters for engineering applications, predict rock fracture load, and assess the safety of flawed rock engineering structures. In this study, fracture experiments using full- or half-disc specimens with chevron notches or straight-through notches were conducted. Experimental results show that the notch types and loading methods (Brazilian-type diametric compression and three-point bending) significantly affect the K Ic measurements. It is indicated that only considering the T-stress or fracture process zone (FPZ) alone cannot explain the test results well, while a good agreement is found between the experimental results and the interpretation based on the combined effects of T-stress and FPZ. Moreover, Brazilian-type compression and three-point bending under a short support span can make the full- or half-disc specimens have higher negative T-stress and larger FPZ, thus producing lower K Ic values than three-point bending under a relatively long support span. Compared with K Ic measurements using the straight-through notch specimens, those utilizing the chevron-notched ones are less affected by FPZ and yield higher K Ic results, providing that the loading method is the same. The notch types have little impact on the T-stresses of the specimens. This study sheds light on the combined influence of T-stress and FPZ on rock fracturing.