In vitro laboratory experiments were conducted to establish some empirical correlation between the metal ion concentration in the environment and the fretting wear volume of orthopedic implant metals, as well as to clarify the relative merits of metal ion concentration and traditional weight loss measurements for modular device testing. Included in this study were a number of characteristic examples for the fretting response of Co-Cr-Mo alloy, Ti-6Al-4V alloy, and 316L stainless steel for fretting against themselves, against each other, and against bone. For Ti-6Al-4V, nitrogen ion implanted surfaces were also studied. The test specimens were carefully designed to allow similar relative surface displacements to occur for all material combinations. Controlled fretting wear tests in 150 mL of Ringer's solution were performed for various periods of time with loads varying between 30 and 300 N (sine, 6 Hz). In some cases, different testing machines were employed for tests with identical material combinations to learn more about the test sensitivity to machine variation. Solution samples, which were taken at the end of each fretting wear test (after 0.5, 1, 1.5, and 2 × 106 cycles) and during the two-million-cycle test, were analyzed for base metal ions using inductively coupled plasma mass spectrometry. The fretted samples were removed and measured for weight loss at the end of the test. The correlations between the metal ion concentration and the fretted metal wear volume were determined for each fretting combination and showed good linear correlations. It was, however, also found that the correlation between the metal ion concentration and the fretting wear volume might be significantly influenced by the particle size of the fretting wear debris and by the occurrence of various chemical and electrochemical processes. Consequently, the correlation between the metal ion concentration and the fretting wear volume of a metal may depend considerably on the material it is fretted against. It was also suggested by these findings that the fretting wear volume calculated from the weight loss might often be considered only an apparent fretting wear volume and may underestimate the severity of fretting wear. Furthermore, the weight loss measurements do not seem to have adequate sensitivity in the size (weight) range of actual modular devices and cannot be carried out without interrupting the fretting wear test. In contrast, the metal ion concentration measurements offer an excellent method for assessment of implant fretting wear by simply monitoring the metal ion concentration throughout the test. Despite the complexity of events that might take place in these systems, the metal ion concentration is a direct indication of the overall interaction between the fretted metal and the testing environment. Being an extremely sensitive and semicontinuous method, the fretting wear performance of various material combinations and the effectiveness of surface modifications can also be successfully evaluated by the concentration measurements, as exemplified in this paper. With this technique, the fretting wear aspects of new modular devices, including particulate and metal ion release issues, can be better addressed without making any major changes in existing testing protocols. The use of metal ion concentration measurements for basic research is also valuable and strongly recommended.