We examined bone-to-implant relative displacement of acetabular prostheses acutely and after ingrowth in a canine model. Uncemented hemispherical acetabular cups with titanium mesh pads comprising approximately 26% of the surface of the cup were inserted in eight adult canine hemipelves ex vivo. The acetabular prostheses were fixed with 13 mm titanium screws. Zero, one, and two-screw configurations were tested, with the order of testing randomly assigned. A load simulating 1,000 cycles of canine gait as applied to the acetabular component, and relative displacements were measured at three locations between implant and bone to determine acute fixation. A repeated measures analysis of variance showed that two screws produced only 42% of the average relative displacement of one screw and 14% that of zero screws. Eight adult mixed-breed dogs then underwent unilateral total hip arthroplasty. All acetabula were biologically fixed with two cancellous screws. The results at 4 months showed significantly less relative displacement between the implant and bone than was measured in ex vivo implantations (p = 0.014). Bone ingrowth filled 20 +/- 6% (mean +/- SD) of the available space. The relative displacements of these implants were small in all cases (12 +/- 13 microns) and did not correlate with the amount of bone ingrowth. These data suggest that acetabular fixation with two screws can lead to bone ingrowth and reduced relative motion of the prosthesis under functional loading.