In the early twentieth century behaviorist Edward L. Thorndike began the development and use of measurement scales to replace personal judgment to evaluate student compositions in U.S. public schools. In 1912, utilizing the Fullerton and Catell equal difference theorem, Milo B. Hillegas released the first scientifically designed scale to measure all of pupil composition. It was refined and modified over the years and numerous other scales created. Serious criticism began when scale use was at its height in 1925. Scale builder Stuart Courtis questioned scale validity, and Gestalt theorist Kurt Koffka challenged Thorndike's behavioristic psychology. In 1938 B. Othanel Smith, University of Illinois educator, thoroughly attacked the movement's logical base, and, by extension, its methods and outcomes. In that same year Courtis withdrew all his tests from the market. By 1940 there was almost no mention of composition measurement scales in the literature. Behaviorists failed to measure pupil English composition validly and reliably, which led to decline in use of Thorndikean initiated scales. However, the decades of experiment and dialogue provided increased understanding of and attention to major aspects of validity and reliability in judging composition. (ES)