Since the Wechsler Intelligence Scale for Children-Third Edition (WISC-III) was published in 1991, it has been reported that fewer students are qualifying for gifted programs that use the WISC-III as a criterion measure. WISC-III differs from the WISC-Revised (WISC-R) in having a greater emphasis on speed of response, which could "penalize" reflective gifted children. The WISC-III was administered to 141 rural West Virginia children aged 6-12.5 who had full-scale IQ scores above 114. The children were categorized according to level of IQ as bright (115-123), superior (124-131), or gifted (132-148). Multivariate analysis of covariance (MANCOVA) was used to compare the groups on subtest scores, verbal and performance IQ scores, and two of the four WISC-III factorial indices--verbal comprehension index (VCI) and perceptual organization index (POI). When adjusted for full-scale IQ as the covariate, analyses showed significant differences between the IQ groups for four subtests, for VCI and POI, and for untimed and speed-bonus groups of subtests. The bright group showed comparatively lower scores on subtests yielding bonus points for quick performance; this deficit was not observed for superior and gifted groups. Bright group scores were similar to those of the superior group for VCI, but well below the superior group on POI. Although perceptual organization skills are important in advanced learning, it would appear that WISC-III does not measure these skills in gifted children, but instead measures the "speed" with which children organize perceptual materials. Implications for identification and placement in gifted programs are discussed. Contains 23 references. (SV)