It is well documented that an important cause of osteolysis and subsequent loosening of replacement hip joints is polyethylene wear debris. To avoid this, interest has been renewed in metal-on-metal and ceramic-on-ceramic prostheses. Various workers have assessed the lubrication modes of different joints by measuring the friction at the bearing surfaces, using different lubricants. Measurements of friction factors of a series of hip prostheses were undertaken using carboxymethyl cellulose (CMC) fluids, silicone fluids, synovial fluid and different concentrations of bovine serum as the lubricant. The experimental results were compared with theoretical predictions of film thicknesses and lubrication modes. A strong correlation was observed between experiment and theory when employing CMC fluids or silicone fluids as the lubricant. Mixed lubrication was found to occur in the metal-on-metal (CoCrMo/CoCrMo) joints with all lubricants at a viscosity within the physiological range. This was also the case for the metal-on-plastic (CoCrMo/ultra-high molecular weight polyethylene) joints. The ceramic-on-ceramic (Al2O3/Al2O3) joints, however, exhibited full fluid film lubrication with the synthetic lubricants but mixed lubrication with the biological lubricants. Employing a biological fluid as the lubricant affected the friction to varying degrees when compared with the synthetic lubricants. In the case of the ceramic-on-ceramic joints it acted to increase the friction factor tenfold; however, for the metal-on-metal joints, biological fluids gave slightly lower friction than the synthetic lubricants did. This suggests that, when measuring friction and wear of artificial joints, a standard lubricant should be used.