Data collected during the standardization of the 1976 Metropolitan Readiness Tests (MRT) were examined to provide information on sex differences in prereading skills. The MRT are designed for administration either at the beginning of kindergarten or between kindergarten and first grade. Performance on the six subtests and the total test on both forms of the MRT favors girls. Boys' scores are consistently more variable than girls' scores. Performance differences favoring girls occur not just at the mean, but throughout the distribution. Boy/girl differences are also apparent when absolute performance standards are set. Item data show that 111 of the 152 items from the two forms of the MRT favor girls, 16 significantly. A stepwise multiple regression analysis shows that subject sex is not a significant predictor of achievement, but that the two best predictors of end-of-year reading achievement (letter recognition and visual matching) are two of the MRT subtests where sex differences are largest. Although the data support the conclusion that girls have better prereading skills at the time of initial school entrance, they do not support the use of differential instructional treatment for boys and girls. (RL)