INTRODUCTION. Ceramic-on-metal hip replacements (COM, where the head is a Biolox Delta ceramic and liner is Co Cr alloy), have demonstrated reduced wear under standard conditions in vitro compared to metal-on-metal (MOM) [1]. Early clinical results are also encouraging [2]. Recently concerns have been raised regarding the poor clinical performance of MOM hip resurfacings [3], particularly when cups are steeply inclined. Laboratory hip simulator testing has been used to replicate edge loading, also demonstrating elevated wear [4]. Therefore, a range of conditions to replicate sub-optimal use clinically to better predict in vivo performance should be used. The aim of this study was to compare the wear rates of MOM and COM under adverse edge loading conditions in an in vitro hip simulator test. METHODS. Ceramic-on-metal (n=3) and metal-on-metal (n=3) 36mm hip prostheses (supplied by DePuy International Ltd, UK) were tested in the Leeds Physiological Anatomical Hip Joint Simulator. Liners were mounted to provide a clinical angle of 45o, and stems positioned anatomically. A simplified gait cycle and microseparation was applied as previously described [5] for two million cycles in 25% new born calf serum. Gravimetric analysis was completed every million cycles and wear volumes calculated. RESULTS. The overall mean volumetric wear rate of COM bearings was 0.36 ± 0.55mm3 per million cycles, this was significantly less than the MOM bearing wear (1.32 ± 0.91mm3 per million cycles). For both COM and MOM bearings wear under these edge loading conditions was significantly greater if compared to previously reported wear under standard conditions [1]. DISCUSSION. The reduced wear of COM has been attributed to the differential hardness decreasing adhesive wear and reduced corrosive wear [6]. Wear under the harsh edge-loading conditions in this study is also reported to be significantly less in COM bearings compared to MOM. In MOM bearings in edge contact conditions, the wear zone becomes starved of lubrication, this elevates wear and increases damage at the edge of the cup. In COM bearings the harder head does not become damaged when there is lubricant starvation and hence wear does not accelerate in the same way. In conclusion, COM bearings show reduced wear compared to MOM bearings under standard and adverse conditions and there is some early evidence to support this finding clinically. ACKNOWLEDGEMENTS. Supported by DePuy International Ltd. SW is supported by a Royal Academy of Engineering/EPSRC (UK) fellowship