There has been a resurgence of interest in the use of metal on metal bearings in total hip arthroplasty. Although the use of metal on metal bearing couples would eliminate or substantially reduce particulate polyethylene generation (depending on the presence or absence of polyethylene in the implant system), there is concern about the potential for increased particulate and ionic metal generation in comparison with polyethylene on metal bearings. These metallic degradation products may be transported away from the implant site and distributed systemically. Chromium concentrations in the serum and urine and cobalt concentrations in the serum were measured in subjects with cobalt chromium alloy metal on metal total hip replacements and in controls without implants. Eight subjects with long term (> 20 years) McKee-Farrar total hip replacements had 9-fold elevations in serum chromium, 35-fold elevations in urine chromium, and at least 3-fold elevations in serum cobalt concentrations in comparison with controls. Six subjects with short term (< 2 years) metal on metal surface replacement arthroplasties had 3-fold elevations in serum chromium, 4-fold elevations in urine chromium, and 4-fold elevations in serum cobalt concentrations in comparison with subjects with McKee-Farrar implants. Although the toxicologic importance of these trace metal elevations has not been established, serum and urine metal concentrations may be useful markers for the tribologic performance of metal on metal bearings.