This study examined the effect of the surgical approach used in total hip arthroplasty (THA) on gait mechanics six months following surgery. Quantitative gait analysis was performed on 29 subjects: 10 anterolateral (A-L) and 10 posterolateral (P-L) THA patients and nine able-bodied, velocity-matched subjects. Discriminant function analysis was used to determine the distinction of the groups with respect to sagittal plane hip range of motion, index of symmetry, trunk inclination, pelvic drop, hip abduction, and foot progression angles. The A-L group had the largest trunk inclination (3.0 ± 2.4°) and the smallest hip range of motion (34.0 ± 7.4°). Both THA groups demonstrated greater asymmetry as expressed by the smaller symmetry index (0.97 ± 0.04 for A-L and 0.98 ± 0.05 for the P-L) than the able-bodied group (0.99 ± 0.01). The classification procedure correctly classified 89% of the control group cases, 90% of the A-L cases, and 50% of the P-L cases. These results support the conclusion that six months following surgery, the gait of the majority (85%) of THA patients has not returned to normal. The A-L patients displayed distinct gait patterns, while a small percentage (30%) of the P-L patients demonstrated normal gait. While these differences are statistically significant, the clinical significance is unknown and linked to the duration that they persist. © 2003 Orthopaedic Research Society. Published by Elsevier Ltd. All rights reserved.