The Wechsler Intelligence Scale for Children - Revised (WISC-R) and the Wechsler Intelligence Scale for Children (WISC) were administered in counter-balanced order to 81 children aged 7-12 in New South Wales, Australia. Canonical correlation analysis showed the total redundancy values of WISC-R given WISC and WISC given WISC-R to be .58 and .61 respectively. The correlation between the first pair of canonical variates of the two tests was .97. The first canonical variates were found to be g-type factors and accounted for the major portion of the overlap between the two tests. Analyses of variance indicated significance for test form, order of administration and for the test form X order of administration interaction effect. WISC-R produced consistently lower intelligence quotient estimates than WISC, and all students to whom WISC-R had been administered first performed better on both tests than students to whom WISC had been administered first. The degree of similarity of the internal structures of the tests was investigated by factor analysis. A two factor solution representing verbal and performance factors was selected as the best solution for both WISC-R and WISC. (Author/RL)