Dr N. S. Farrar asked about an illustration in which a general mechanism of fracture was proposed in which a fracture plane emanated from under the loading platens during a compressional failure. Had the authors experienced such fractures in experimental work, as it was his own view that such an occurrence seemed highly improbable? However, many published reports of testing do indicate fracture planes originating from the edges of the loading system and running diagonally across the specimen. These suggest a shearing failure with the material possessing parameters of cohesion and internal friction similar to those in the compressional testing of soft sediments. It has been the experience of the speaker that this type of diagonal failure in rock only occurs in the fracture of specimens which have been loaded in an unsymmetrical manner. Fractures of such material when loaded axially should fail in a manner which is symmetrical about the axis of loading.
Also, under normal practical conditions existing in a compression test the inevitable presence of friction plays a dominant role in conditioning the actual mechanism of fracture. The speaker suggested that the majority of well prepared brittle materials will thus fracture with the formation of end cones or pyramids. The section of the material between these ends then fails in the load-parallel manner described by the authors. A few materials, such as glass, appear to fail in a totally load-parallel direction, which may be occasioned by different frictional properties of these specimens. However it is considered that