Spearman's ''Law of Diminishing Returns'' with regard to ability is tested in a dataset from the National Longitudinal Study of Youth. The dataset consisted of a sample of 6980 children aged 12-16 from the 1997 cohort. The subjects were tested with a computer administrated adaptive format of the Armed Services Vocational Aptitude Battery consisting of 12 subtests. Spearman's ''Law of Diminishing Returns'' was tested by two methods both dividing the sample into Low/High ability based either on the total score on the test or on the score one of the 12 subtests. Subsequently the ability groups were factor analysed separately. The eigenvalue of the first principal component and the first principal axis factor, and the average inter-correlation of the subtests were used as estimates of the g saturation and compared across groups. The study could not confirm Spearman's ''Law of Diminishing Returns'' for any of the methods applied and did not find any relevant differences across methods applied.