The purpose of this study was to determine to what extent hip joint separation occurs during normal gait on a treadmill and an abduction/adduction leg-lift maneuver in patients who have undergone total hip arthroplasty (THA). Eight patients who had a total of 10 successful unconstrained THAs (Harris Hip Scores >90) performed successive gait motions on an electronically powered treadmill and an abduction/adduction leg lift while under fluoroscopy. The fluoroscopic video images were analyzed using a 3-dimensional model-fitting technique that converts 2-dimensional fluoroscopic images into 3-dimensional real-time images. Hip joint separation was determined to be present if the amount of separation was >0.75 mm, the calculated linear error. During both activities, all 10 THAs experienced femoral head/acetabular component separation. For gait, the maximum amount of separation was 2.8 mm, while the minimum amount was 0.8 mm (average, 1.2 mm). For abduction/adduction leg lift, the maximum amount of separation was 3.0 mm, while the minimal amount was 1.7 mm (average, 2.4 mm). It appears that the femoral head separates from the acetabular component but remains in contact with the superior-most tip of the component. Potential detrimental effects resulting from hip joint separation include premature polyethylene wear and component loosening. Wear may be enhanced because of the creation of multidirectional wear vectors or excessive loads resulting from eccentric femoral head pivoting. These data may be valuable in hip simulation studies to better duplicate wear patterns observed in retrieval analysis.