O'Boyle and Hoff (Neuropsychologia, 25, 977-982, 1987) reported that females were faster and more accurate than males at mirror-tracing the outline of random shapes. To account for this differential performance, the authors advanced two alternative explanations. The first was a 'Spatial Stroop' effect which suggested that, because of their enhanced spatial ability, males are especially disadvantaged in escaping the misleading visual feedback provided by the mirror, resulting in slower and less accurate tracings. The second was a 'manipulospatial' hypothesis which suggested that the tracing advantage was related to female superiority in fine-detailed motor control, and perhaps, differential practice at performing skilled motor tasks in mirror-reversed contexts. In the present study, two experiments were conducted to assess the relative viability of these explanations. In Experiment 1, the WAIS-R Block Design task was performed within and outside the context of a mirror. This was done to determine if the observed female advantage was restricted to mirror-tracing per se, or generalizable to other manipulospatial tasks. In Experiment 2, a mental rotation task was performed within and outside the context of a mirror. The latter was designed to reveal if the removal of the motor component would affect the obtained sex difference. The present findings suggest that as long as some form of precision motor manipulation is required, females are superior to males at mirror-reversed spatial tasks. However, when the motor component is eliminated, a male performance advantage emerges in both normal and mirror-reversed contexts, suggesting that the manipulospatial hypothesis is the more viable explanation.