It has been argued that the male performance advantage on the Mental Rotations Test, a measure of three-dimensional rotational ability, reflects a slower problem-solving strategy in females. A recent study of a high-ability sample suggested that sex differences do not reach significance when the test is administered without time limits (Goldstein, Haldane, & Mitchell, 1990). To further explore the effects of time limits, unspeeded versions of the Mental Rotations Test and a vocabulary test were administered to 88 males and 94 females, who represented a less restricted range of ability levels. Males obtained significantly higher scores than females on the Mental Rotations Test (p = .0001), despite comparable performance for the two groups on the vocabulary test. In addition, distributions of Mental Rotations scores were significantly different, with greater proportions of men at the high and women at the low ends. Thus, a difference in speed of problem solving strategy is not a sufficient explanation of the sex difference on the Mental Rotations Test.