The results of forty-eight resection arthroplasties (Girdlestone procedures) in forty-three patients were analyzed. The indications were sepsis after a total hip replacement (thirty-three procedures), aseptic loosening of a total hip replacement (ten procedures), and primary septic arthritis (five procedures). The resection arthroplasty effectively eradicated the sepsis in all but three patients, who had a recurrence. Pain was alleviated in nearly all of the patients, but the ability to walk and the level of activity improved only slightly. Shortening of the limb ranged from three to eleven centimeters, and all patients used a support for walking. The six patients who had a bilateral procedure were able to walk using double supports. With regard to relief of pain, walking, and function, the results were significantly poorer in the women, particularly the older ones. The results were also significantly inferior in the patients who had had sepsis after a total hip replacement. Generally, walking, function, and the level of activity were better when much of the proximal end of the femur had been retained. Resection arthroplasty should spare as much of the proximal end of the femur as possible if a future replacement of the hip is contemplated.