Abstarct: Precise measuring the fracture toughness of rocks in tensile and shear modes is important as a key parameter in the investigation of rock fracture mechanics. Many different methods have been proposed to determine these parameters. Each method has it’s own advantages and disadvantages. This thesis attempts to provide a comprehensive comparision between commonly used methods for determining fracture toughness of rock. They include six methods for determining tensile mode and three ones for determining shear mode. The comparisions are baxsed on the sensivity of resualts to changes in specimens size and the length of crack in specimens. For this aim and due to the comparative nature of this study, like-rock specimens are used. The specimens are perfectly homogeneous and isotropic and they show linear elastic behavior characteristics. At first, 316 specimens are artificaly made of plaster materials and after the preparation, all of them are loaded until the moment of failure. Then the resualts are obtained from different methods are separetely analyzed. finally the sensivity of resualts to changes in specimen size and the length of crack in specimens are specified. In the last step by using the finite element software ABAQUS, the distribution of stress and displacemens in cracked circular disk and in semi-circular disk specimens are investigated. Also, failure behavior of laboratory specimens are compared with the resualts of numerical method. The experimental resuals verify the numerical results and show high performance of the applied method. Regarding to various factors such as the geometry of specimen, easy loading , precision of resualts and equipments that are required, it can be concluded the Cracked Chevron-Notch Brazilian Disc method (CCNBD) is the most appropriate method for both modes (tensile and shear). This method is one of the ISRM suggested methods to determine tensile mode of rock fracture toughness.