C HILDREN do learn to spell. Can teachers claim, however, that this learning is a result of their teaching? The schools have in the past generally made that claim. Is it well founded? As a means of investigating this situation, a simple but thorough experiment was tried out in the schools of Staunton, Illinois. The results were rather surprising and should cause educators to re-evaluate all the work in spelling, especially that which follows day by day the present standard textbooks in spelling. The pupils of the Staunton schools for three and a half years used a standard series of workbook spellers which were made by specialists in the field and which are widely known and used throughout the country. The day-by-day directions of these spellers were carefully followed. From all appearances, the teachers were doing the correct thing in the correct way and were securing typically good results. Certain indications, however, caused the superintendent to wonder exactly what function the textbook was serving in the children's learning of spelling. He therefore devised a testing program to secure evidence on this point. For testing purposes a spelling list was made up of seventy words from each book from Grade II to Grade VII and eighty words from Book VIII, a total of five hundred words. These were chosen at random and represented one word out of each 7.63 words in the entire spelling program. From these five hundred words, ten tests of fifty words were made, each test consisting of seven words from each of Grades II-VII, inclusive, and eight words from Grade VIII. The tests provided for pronunciation of each word,.use in a sentence, and 584