An estimate of the change in intelligence in ten English education authorities during the past 10 years has been made by comparing the scores of complete year groups of 11-year-old children in 1947 with scores made by similar groups in the same test on an earlier occasion. The first test was taken by 31,728 children and the second by 28,505 children, boys and girls in approximately equal numbers. For boys and girls together there was a fall of 0.0034 point of intelligence quotient per year, which was far from significant. In seven districts where we had separate scores for boys and girls on both occasions there was a significant fall in intelligence quotient of 0.0921 point per year amongst boys and a significant rise of 0.0921 point per year amongst girls. The adventitious influences which might invalidate a comparison of the two sets of scores have been considered; they probably tend to depress the score at the second test. A study of sex differences in attainment tests at 11 years and in intelligence tests at 13 years and later suggests that the scores of the boys in 1947 are underestimates of their ability, and that a truer measure of the trend of intelligence amongst the population is represented by the change in the girls' scores. By assuming a certain parallelism between the present survey and that of the Scottish children, the rise in intelligence quotient of the latter is estimated at 0.0543 point per year. A significant increase in the dispersion of intelligence test score of both boys and girls during the past 10 years has been recorded.