Experimental Fracture Mechanics is based on the study of specimens containing “machinednatural” cracks, a term which is self-contradicting. In practice, an artificial (machined) notch of considerable thickness is opened in the specimen by using various techniques and, then, a “natural” crack extension of the notch is created through fatigue. This procedure is, in details, described by ASTM standards [
], where an artificial notch is considered as acceptable when, among other restrictions, it must have a radius of curvature of the tip equal to ∼0.25×10
m and the through-fatigue crack extension has a quiet uncontrolled length of (1–2)×10
m. Geometrically, this “natural” crack satisfies two of the requirements of a slit to be a natural crack, i.e. no loss of mass and small (but not near- zero) tip-radius. In many cases, it is not straight or forms a small angle with the notch axis. In addition, this method cannot be applied in case of notches inclined to the loading axis and special loading apparatuses are required. To facilitate and improve ASTM standard many attempts have been presented (e.g. [
]). However, an important side-effect of this method has not been discussed. It is that fatigue causes changes of unpredictable severity in the mechanical properties of the material, exactly in the tip area, where the final crack is expected to initiate. In this area material fails and around the tip of the “natural” crack plasticized zones and hardening processes are observed. Consequently, fracture criteria are based on a slippery ground.