sex-related difference in spatial ability. Though they did not intend it to be an exhaustive review, we feel it important to stress that the Caplan et al. article does not constitute a thorough review either of the literature on sex differences in spatial ability or of the literature dealing with possible biological mechanisms that might contribute to the development of such a difference. Although we agree with the authors' denigration of a blind box-score approach to the sex difference question, we are critical of their omission of much of the pertinent literature, their failure to search for consistent patterns in the tallied studies, and especially of their failure to acknowledge patterns identified by others.There is at least one spatial test that produces a clear and consistent sex difference. This is the Shepard/Metzler Mental Rotation test devised by Vandenberg and Kuse (1978) after a task used originally by Shepard and Metzler. On this test, which requires recognition of fairly complex forms depicted after rotation in threedimensional space, males score approximately one standard deviation higher than females, and sex accounts for 12% to 16% of the variance. This has been found in college students (Bouchard & McGee, 1977;Sanders, Spares, & D'Aquila, 1982), high school seniors (Yen, 1975), parents of college students (Bouchard & McGee, 1977), and parents and offspring participating in the Hawaii Family Study (Wilson & Vandenberg, 1978). All of these results were cited in a 1982 article (Sanders et al., 1982). Findings obtained more recently are consonant with the 1982 review. The result has been replicated in college students (Sanders & Spares, in press;Went, 1985), and a difference of comparable magnitude has been found among ninth graders (Cohen, 1986).Research into the nature of the cognitive process (or processes) that might be responsible for the sex difference on mental rotation tests has been going on for a number of years (e.g., Kail, Carter, & Pellegrino, 1979;Petrusic, Varro, & Jamieson, 1978;Tapley & Bryden, 1977). Although it is certainly true that the sex difference is less pronounced on some mental rotation tasks than on others, the success of investigators in identifying processes responsible for the sex difference on tests of this sort attests to the reliability with which the difference is obtained.
