Cortical bone is a composite material, with a complex hierarchical, heterogeneous and anisotropic microstructure. Moreover, it is a living material, which undergoes considerable changes in composition and structure, according to the mechanical and physiological environment. Despite this self-adaptive capacity, which helps to prevent fracture, bone fractures are still quite common. Predicting the fracture risk of cortical bone is an important topic due to its high socio-economic impact. In particular, predicting the age-related changes of cortical bone strength become increasingly important in light of the rising average population age. The bone mineral density had been used as the main predictor of age-related bone fracture risk, and as the main indicator of the benefits of several drug therapies. However, there is mounting evidence that other factors contribute to age-related bone fragility. In this context, the Fracture Mechanics emerged as an essential tool to assess “bone tissue quality”. To date, most assessments of the toughness of cortical bone have involved linear-elastic fracture mechanics (LEFM). Most works have focused on the mode I crack initiation, both in the longitudinal and transverse axis of long bones, and very few works have addressed the mode II and III fracture initiation, even if cracks in bone seldom experience pure tensile loading. However, LEFM is an improper theoretical model as a result of R-curve behaviour of cortical bone, allied to the size-scales involved in its facture testing. In light of this, several authors had recently applied cohesive damage models for analyzing initiation and propagation of cracks in cortical bone. However, a number of issues associated to this approach are yet unresolved, including the test methods and the data reduction methods more appropriate for accurate identification of the cohesive law. This approach has a great potential inasmuch as it allows analysing the influence of several factors that control the fracture behaviour of cortical bone, thus contributing to our capability of prediction of bone fracture risks. The overall objective of this project is the development of appropriate tests and data reduction methods to determine the fracture properties of cortical bone under mode I, II and III. In order to achieve this goal, we intend to apply a cohesive damage model previously developed. This model combines strength of materials and Fracture Mechanics concepts, and allows the simulation of damage initiation and growth. Several tests will be simulated aimed at selecting the most appropriate ones for each crack face orientation and crack propagation direction, allowed by the anisotropic structure of cortical bone (longitudinal, radial and circumferential directions). The following mode I fracture tests are envisaged: (1) compact tension and compact sandwich tension tests; (2) single-edge notched bending and single-edge notched sandwich bending tests; (3) double cantilever beam test. On the other hand, the following mode II fracture tests are anticipated: (1) compact shear specimen; (2) end-notched flexure and end-notched sandwich flexure tests; (3) tapered sandwich end-notched flexure test. The definition of a mode III test will be tried for at least one natural plane of cortical bone symmetry. A special attention will be given to the definition of appropriate data reduction methods to obtain the R-curves, which take into account the variability of mechanical properties of cortical bone and overcome the need for a direct measurement of crack length. Moreover, an inverse analysis technique will be developed to extract the cohesive law parameters from the force-displacement curve, using a genetic algorithm. This part of the project will be based on the experience of the team on the measurement of fracture properties of wood. The selected fracture tests will be performed experimentally, using cortical bone tissue from the mid-diaphyses of bovine femur. In order to evaluate the capabilities of test methods in discriminating the effects of compositional and morphological changes in the fracture behaviour of cortical bone, both fresh and dried samples of young and adult bovine bone tissue will be tested. Indeed, in bovine long bones most plexiform bone is replaced by osteonal bone as cow mature; furthermore, the fracture properties are affected by the hydration state of bone tissue. The digital image correlation will be used to evaluate the displacement field around the crack tip so as to observe the extension of fracture process zone. The fracture tests will be complemented with the measurement of Vickers microhardness, bone mineral density, and weight fractions of the mineral, organic and water phases of bone samples. Moreover, the scanning electron microscopy will be used to analyse the fracture surfaces of tested specimens, and the optical microscopy will be employed for histological evaluations.