The thickness of the cement mantle around the femoral component of total hip replacements is a contributing factor to aseptic loosening and revision. Nevertheless, various designs of stems and surgical tooling lead to cement mantles of differing thicknesses. This thesis is concerned with variability in cement thickness around the Stanmore Hip, due to surgical approach, broach size and stem orientation, and its effects on stress and cracking in the cement. The extent to which cement mantle thickness varies in clinical practice as a result of surgical approach was investigated, through retrospective radiographic analysis. The posterior approach was associated with a thicker and more uniform cement mantle. Stress distributions in the cement mantle around Stanmore hips and surrounding cortical bone were investigated using finite element analysis, under a simulated gait load. Thicker cement was found to produce lower maximum principal stress in the cement and to reduce the likelihood of bone resorption due to stress shielding. Bone density was found to strongly affect stress levels in the bone and cement, but not to alter this result. Stanmore Hips were implanted into synthetic femurs with different cement thicknesses and stem alignments. Crack measurement and analysis was conducted, following fatigue testing under a simulated stair-climbing load. Crack length was found to be independent of overall cement mantle thickness, suggesting that thinner mantles would fail sooner. However, crack length was highly sensitive to local cement thickness, with regions of 1 mm or less containing longer and more concentrated cracks. Stem alignment and cement mantle uniformity are thus more critical to cement damage than broach size. Finite element simulations incorporating creep and nonlinear damage accumulation were performed to investigate cracking in the cement mantles around Stanmore Hips with varied cement thickness, interfacial bonding and collar design. Simulations represented the simple stair-climbing joint contact load used in fatigue tests and a more realistic stair-climbing load incorporating muscle forces and physiological bone properties. In all cases, damage levels were much higher when the stem-cement interface was bonded. Cement mantle thickness was of limited importance to cement damage in debonded cases, in agreement with fatigue test results, but was critical in bonded and collarless cases. Damage around a smooth, debonded stem with a collar is thus much less sensitive to cement thickness than around rough/bonded or collarless stems.