FOURTEEN STUDENTS AT THE UNIVERSITY OF SOUTH CAROLINA, MOST OF THEM EXPERIENCED TEACHERS, ENROLLED IN AN EXPERIMENTAL TRAINING PROGRAM IN EARLY CHILDHOOD EDUCATION ATTEMPTING TO DETERMINE WHETHER OR NOT SHORT PERIODS OF INDIVIDUALIZED TRAINING OVER A PERIOD OF 5 WEEKS WOULD BE REFLECTED IN PUPILS' READINESS AS MEASURED BY THE METROPOLITAN READINESS TEST. FOURTEEN HEAD START CLASSES WERE ASSIGNED RANDOMLY TO THE EXPERIMENTAL TEACHER-STUDENTS AND 14 HEAD START CLASSES TO THE CONTROL TEACHER-STUDENTS. ONE MALE AND ONE FEMALE WERE SELECTED RANDOMLY FROM EACH OF THE EXPERIMENTAL AND CONTROL CLASSES. ON FORM A, THE PRETEST, THE T-RATIO DID NOT APPROACH SIGNIFICANCE. ON FORM B, THE POST-TEST, THE T-RATIO APPROACHED SIGNIFICANCE AT THE 10 PERCENT LEVEL OF CONFIDENCE. DATA SHOWED THAT PUPILS WHO RECEIVED SPECIAL INSTRUCTION GAINED DURING THE PRE- TO POST-TESTING PERIODS AND THAT MOST PUPILS IN THE CONTROL GROUP EITHER LOST OR REMAINED AT ABOUT THE SAME LEVEL. THE STUDY RECOMMENDED THAT A SIMILAR EXPERIMENT BE CONDUCTED FOR VARIOUS TIME PERIODS. IT IS RECOMMENDED THAT TEACHERS LEARN MORE ABOUT DIAGNOSING PUPILS' INSTRUCTIONAL LEVELS AND ABOUT STRATEGIES FOR APPROPRIATE TREATMENTS. THIS PAPER WAS PRESENTED AT THE AMERICAN EDUCATIONAL RESEARCH ASSOCIATION CONFERENCE (CHICAGO, FEBRUARY 6-10, 1968). (JM)