Intended to help dispel several unsupported "myths" about reading instruction, this book analyzes a selected group of teaching practices that have been supported by reading experts but not by research findings. The "myths" discussed in the first 12 chapters of the book are as follows: (1) phonics hinders comprehension; (2) unpredictable spelling invalidates phonics; (3) children learning to read should first be taught whole, or sight, words; (4) reading is best taught in sentences; (5) context clues are all-important to beginning readers; (6) word length makes no difference in beginning reading instruction; (7) learning modalities and instruction should be matched for children to learn best; (8) letter names are unimportant; (9) children need to be taught dictionary syllabication to help them acquire reading skills; (10) reading tests should be replaced with oral reading miscue analysis; (11) subvocalization is bad; and (12) oral reading is dangerous. The remaining three chapters of the book discuss why the myths prevail and how they can be dispelled, and present an annotated bibliography of reviews of research on phonics. (FL)