We used three-dimensional finite-element models of the proximal end of the femur to examine the influence of stem material, stem geometry, and the use of a calcar collar on the stresses in and around implanted total hip replacement femoral components. Anatomical bone geometries and realistic prosthetic geometries were considered. A slender titanium-alloy stem with a collar allows creation of calcar stresses of approximately 80% of the anticipated normal levels. A similar stem of cobalt-chromium alloy creates calcar stresses of 67% of these normal values. Stem designs without a collar were shown to generate no more than 40% of normal values while larger, stiffer stems were seen to create less than 30% of normal values, with or without a collar. Proximal cement stresses were increased by the use of titanium-alloy stems, but were reduced to low levels by a functioning collar. The highest cement stresses in the system were found near the tip of the stem, where titanium-alloy stems create lower stresses than do corresponding cobalt-chromium-alloy stems. The achievability of calcar loading with a titanium prosthesis was demonstrated in in vitro strain-gauge tests. Clinical relevance: Loosening of the femoral stem and calcar resorption are problems that are seen in many long-term clinical series of total hip prostheses. In order to reduce the incidence of these problems, the goal of the designer of a prosthesis is to reduce cement and cement interface stresses around the femoral stem and to create stress distributions in the bone that will prevent resorption. We have demonstrated that, in a well bonded system, properly designed titanium-alloy stems with functioning collars create higher calcar bone stresses (nearer their normal values) and lower cement stresses than do equivalent cobalt-chromium-alloy stems. The ability to achieve calcar-collar contact and loading was demonstrated; however, the long-term effectiveness of the collar is unknown.