The lasting integrity of the bond between cement and bone defines the long-term stability of cemented acetabular replacements. So far experimental evidences seem to suggest that debonding of the bone-cement interface is responsible for the fixation failure in the cemented acetabular replacements. Both mechanical and biological effects may jointly contribute to implant loosening (Huiskes 1993, Sundfeldt et aI., 2006). However, mechanical analysis of pelvic bone with cemented acetabular replacement and quantitative assessment of fatigue behaviour under physiological condition on acetabular replacement have been difficult, due to the complexity of the pelvic bone geometry and the associated loading conditions. In this work, the mechanical analysis of the pelvic bone with cemented acetabular replacements was carried out using the finite element method, and the fatigue behaviour of the cement fixation in acetabular replacements under physiological condition was studied on a specially designed hip simulator. A series of finite element models of pelvic bone with cemented acetabular replacements were developed and validated, including a 3D simplified model, 3D half pelvic bone model and 3D whole pelvic bone model. The basic mechanical characteristics of the pelvic bone and pelvic bone with cemented acetabular replacement were analysed to examine the influence of cemented acetabular reconstruction. Stress and strain distributions in the cement mantle, especially at the bone-cement interface, were obtained and analysed. Parametric studies were performed, including parameters such as cup size, cement thickness, cup orientations, and the influences of load variations and insertion of muscle forces on the cement mantle. Numerical results seem to suggest that cement thickness and cup orientation have a significant effect on the stress distributions in the cement mantle and at the bone-cement interface. The stress distribution in the cement mantle seems to be most favourable in the standard cup position, as opposed to open and retroverted cup positions. The hip simulator for fixation endurance testing was utilized to study the long-term mechanical response of cemented acetabular reconstructs, i.e. damage initiation and development, under a combined loading blocks representative of patient typical routine activities (Bergmann et aI., 2001), as well as stair climbing and normal walking loading conditions. A micro-CT scanner was used to detect and monitor damage development at regular intervals of the experiments. Microscopic studies post testing were carried out to verify the damage detected by the CT scanning. The results show that debonding at the bone-cement interface defined the failure of the cement fixation in all cases, and debondings initiated near the dome of the acetabulum in the superior-anterior quadrant, consistent with the high stress regions identified from the finite element analysis of implanted acetabular models. An environmental chamber that mimics in vivo condition was also developed. Preliminary results show that bone-cement interfacial debonding was identified in all Delee zones in implanted acetabula tested with an environmental chamber. The introduction of simulated body fluid into the hip simulator seems to be significant in that a much reduced survival life of the bone-cement interface was reported, due to a combination of mechani~al loading and environmental effects. This combination might also be responsible for early bone-cement interfacial debonding, more consistent with the clinical observations. Results from the present study may be useful for orthopaedic surgeons towards improving surgical procedures to achieve long-term stability of acetabular cement fixation.