It was then found that many of the thirteen thus selected had particular difficulty with phonetics—could not readily connect the sounds of letters with the visual symbol. They could all match words. All could point out the difference in two words differing in one letter. Those who were old enough to have learned to write could transpose a page of print into script. These facts are taken as evidence that the difficulty must be central and not sensory. Four of the children were followed up with teaching in a special class. Phonics were taught. The easiest letters—those that can be prolonged—were taught first (r, f, l, m, s). In the effort to make the work interesting, the phonics were presented in a story, associating each sound with parts of the story. As many associations as possible were established to fix the _sound_ of _the letter_. All who received this special training are reported to have improved greatly, in a short time. Schmitt concludes that there are a few children who are so constituted that they cannot learn readily by the word and sentence method. “Every teacher uses this (the phonic method) to some extent, but to a very slight extent. The average child quickly learns to associate the printed letters and words with their vocal prototypes, without special emphasis on phonics, or special attention to associations. “Whatever may be said for the word and sentence methods, it is really by the phonic method that the child becomes independent of the teacher.” In 1920 and 1921 Freeman and Gray respectively presented well-studied cases of individual pupils. Freeman’s case was that of a girl 9 years and 6 months old, in the fourth grade of the University Elementary School, at Chicago. General intelligence was “better than average,” the exact IQ not being stated. The child’s father and paternal aunt had also had marked difficulty in reading. Both parents were above average in social-economic status, and hence probably also in intelligence. An oculist had made a diagnosis of word blindness, with a very discouraging prognosis for learning to read. As a result of careful psychological analysis, it was decided that there was no deficiency of general intelligence, and no disorder of vision or of visual perception. There was no motor deficiency or general language disturbance. “The defect, therefore, must be a highly specialized one,” apparently consisting in lack of aptitude for associating visual symbols with prescribed sounds.