This study investigated whether children oversized at birth, like those undersized at birth, have an increased chance of mental subnormality. Subjects were 4-year-olds born of urban disadvantaged white mothers whose average education was 10.8 years. Stanford-Binet intelligence scores of the subjects were compared to their birth weights. Between mid-1959 and September 1964, 1,126 children, 564 boys and 562 girls, were followed and given intelligence tests. These children were divided into a low (below 2,501 gm.), a high (over 4,250 gm.), and a standard birth weight group (all others). The heaviest 5 percent of newborns was selected from the high group. The report indicates that of the heaviest 5 percent there was a significantly larger percentage with intelligence scores below 80 (subnormal) at 4 years of age. This percentage (23 percent) is more than double that of all other children tested who weighed over 2,500 gm, and is comparable to that found in children of low birth weight. A greater variability and lower mean IQ is also found in the oversized group than in those of usual size at birth. These results are sufficiently challenging to warrant investigation of fetal, maternal, and environmental factors which might have affected the results. (JF)