To justify the use of new materials or designs of total hip arthroplasties, in vitro test methods always should replicate wear rates, patterns, and mechanisms seen clinically. Retrieved first-generation alumina-alumina hip joints typically have shown a stripe of wear on the femoral heads, an associated worn area in the acetabular cup, and a low incidence of osteolysis. The wear rates of well-positioned retrieved alumina-alumina prostheses have been reported to be on average 1 to 5 mm3 per annum [1], which is at least an order of magnitude lower than average wear rates of metal–ultra-high-molecular-weight polyethylene (UHMWPE) prostheses. Publications of in vitro test data to date have shown even lower wear rates, however [2,3]. The aim of this study was to reproduce clinically relevant wear in a physiologic hip simulator to test the new generation of devices more thoroughly. This is especially important as novel prostheses, such as those using alternative materials and geometries, are being proposed for younger patients. Two modern alumina-alumina total hip prostheses were studied (manufactured since 1995). Both had been well positioned and stable before either trauma or a low-grade infection, which had necessitated revision after approximately 1 year. Both of the femoral heads showed small wear stripes similar to those seen in previous retrieval studies of firstgeneration alumina-alumina prostheses. The only visible wear on the cups was a small band of wear around the rim. This band was the first evidence that modern alumina-alumina hip joint components also exhibited the stripe wear morphology frequently found in retrieved first-generation alumina-alumina prostheses. Fluoroscopy studies have shown separation of the ball and socket during the swing phase of walking for metal-UHMWPE hip prostheses [4]. It was hypothesized that this microseparation could occur with any hip prosthesis and could be a factor in stripe wear initiation in ceramic-ceramic hip prostheses. The small clearances of the head and socket (typical radial clearances are 30 μm) meant that it was possible that the femoral head translated inferiorly and laterally if microseparation occurred. These displacements typically could have been ,1 mm for a well-positioned prosthesis. It was postulated that after microseparation during the swing phase, on heel-strike, the head would translate superiorly and contact the rim before relocating in the cup. This rim contact would occur under high stresses and could initiate surface damage in the form of a wear stripe on the head. The postulated mechanism of the generation of stripe wear is shown in Fig. 1. From the *Department of Mechanical Engineering and †Department of Microbiology, University of Leeds, Leeds; ‡Stryker Howmedica Osteonics, Newbury, Berkshire, UK; §Stryker Howmedica Osteonics, Kilchberg, Switzerland; \Bradford Royal Infirmary, Bradford, UK; and ¶Mater Hospital, Sydney, Australia. Submitted November 11, 1999; accepted February 29, 2000. Benefits or funds were received in partial or total support of the research material described in this article from Stryker Howmedica Osteonics, Kilchberg, Switzerland. Reprint requests: J. Fisher, Department of Mechanical Engineering, University of Leeds, Leeds, LS2 9JT UK. Copyright r 2000 by Churchill Livingstonet 0883-5403/00/1506-0016$10.00/0 doi:10.1054/arth.2000.8100 The Journal of Arthroplasty Vol. 15 No. 6 2000