Unilateral triple pelvic osteotomy was performed with three variations in technique on canine cadaver pelves. The following variables were studied: the angle of the ilial osteotomy (perpendicular to the long axis of the ilium and 10d̀, 20d̀, and 30d̀ off perpendicular); the degree of axial rotation (20d̀, 30d̀, and 45d̀); and the type of fixation (canine pelvic osteotomy plate [CPOP] or a 2.7‐mm dynamic compression plate [DCP]). Structural changes measured were the pelvic inlet and acetabular area, interischiatic tuberosity distance, and degree of acetabular version. Means and standard error of the means were determined for all pelvic measurements and were analyzed by three‐way analysis of variance (P<.05). As the axial rotation increased from 20d̀ to 45d̀ an ilial osteotomy angled 10d̀, 20d̀, or 30d̀ resulted in a significantly greater decrease in the pelvic inlet area and significantly less deviation of the interischiatic tuberosity distance and degree of acetabular version from normal than an osteotomy directed perpendicular to the long axis of the ilium. Compared with the CPOP, as the axial rotation increased from 20d̀ to 45d̀, the DCP resulted in a significantly greater decrease in the pelvic inlet area, an increase in the interischiatic tuberosity distance and degree of acetabular version, and less of an increase in the acetabular area. The results of this study suggest that to maximize dorsal acetabular coverage, while minimizing disruption of normal pelvic architecture, a CPOP and an ilial osteotomy angled 10d̀ to 30d̀ are preferred for all degrees