VERBAL INTERACTION BETWEEN TEACHERS AND PUPILS WHEN THEY ARE READING CRITICALLY IS REPORTED. SIX HUNDRED FIFTY-ONE CHILDREN AND 24 TEACHERS FROM SEVEN ELEMENTARY SCHOOLS IN COLUMBUS, OHIO, SERVED AS SUBJECTS DURING THE 9-MONTH INVESTIGATION. TWELVE CLASSES, TWO AT EACH OF THE SIX ELEMENTARY GRADE LEVELS, WERE GIVEN TRAINING IN CRITICAL READING WHILE 12 CLASSES WERE INSTRUCTED IN LITERATURE. TEACHER QUESTIONS AND STUDENT RESPONSES WERE THE MAIN FOCUS OF THE STUDY. AN INSTRUMENT WAS DEVISED FOR OBSERVING VERBAL BEHAVIOR. EIGHT CATEGORIES OF TEACHER QUESTIONS WERE INFLUENCED BY BLOOM'S APPROACH, AND FIVE PUPIL CATEGORIES, REPRESENTING LEVELS OF THOUGHT, WERE INFLUENCED BY GUILFORD'S STRUCTURE. TEACHERS WERE INFORMED OF FORTHCOMING CLASSROOM OBSERVATIONS WHICH TOTALED SIX IN NUMBER AND LASTED FOR 25 MINUTES. CHI-SQUARE WAS USED TO ANALYZE THE DATA. THE INCLUDED RESULTS INDICATED THAT--(1) THERE IS A DEFINITE RELATIONSHIP BETWEEN TEACHER QUESTIONS AND QUALITY OF PUPIL RESPONSES, (2) TEACHERS IMPROVED IN THEIR ABILITY TO ASK QUESTIONS, (3) TRAINING OF TEACHERS AND SPECIAL INSTRUCTIONAL MATERIALS INFLUENCED VERBAL BEHAVIOR, (4) LIMITED GRADE LEVEL TRENDS WERE DISCERNABLE IN TEACHERS QUESTIONS, AND (5) DEVELOPMENTAL TRENDS IN PUPIL RESPONSES WERE IDENTIFIABLE IN THE EXPERIMENTAL GROUP. TABLES AND THE OBSERVATION DIRECTIONS ARE INCLUDED. (BK)