What is the interpretation of the facts reported? Does it not seem certain that _general intelligence_ is, as indicated by the high coefficients of correlation obtained between reading and intelligence, the chief consideration, in predicting whether or not a child will learn to read? Would it not appear that children of adequate general intelligence, and of normal sensory capacity, learn to read when given intensive training, whatever avenue of approach may be particularly stressed? It is not credible that all the non-readers found by Schmitt in Chicago, chanced to have a kind of disability approachable by phonics, and in no other way; that those discovered by Fernald and Keller in California were so constituted that they could be approached through motor exercises, and not otherwise; that Gates’ cases in the Scarborough School all happened to be susceptible to training through visual methods, and through no others. In fact, no investigator has established his or her method as the only method of successful approach to particular cases, by excluding other methods through experimental teaching. For non-readers such as have been described under the criteria laid down by the investigators quoted, it seems highly probable that the best method would be that wherein all the avenues of approach are fully utilized. Such a method would combine all the special exercises devised by the various investigators, in a proportion and sequence, which should be determined upon as optimum by experimental teaching. Such a method, when experimentally established, would be most suitable for all children—not for the extreme of the distribution exclusively. Here, as in so many questions of pedagogy, all children might profit from our study of the extreme cases, who differ from the typical in degree only. Children of normal sensory capacity, and of IQ average or superior, typically learn to read passably well, without approach through all the possible avenues, and without special attention on the part of the teacher to all the elements involved. A few such children require intensive teaching in order “to pass” in reading, because of specific idiosyncrasies. If the methods that succeed with the extreme cases were applied to the typical class, perhaps the children might learn to read, not “passably,” but very well. There might be a rise of ten points in norms for reading ability throughout the grades. Such perfection of method might or might not eliminate entirely the necessity for individual teaching of special cases. Probably it would not, in classes as large as those seen in most of our public schools to-day. IX. CASES OF SPECIAL ABILITY IN READING