THE OBJECTIVE WAS TO TEST TWO HYPOTHESES--(1) DAY CLASS TRAINING WOULD PRODUCE DESIRABLE CHANGES IN THE BEHAVIOR OF THE TRAINABLE MENTALLY RETARDED CHILDREN OVER AND ABOVE THOSE WHICH RESULT WHEN THE CHILD REMAINS AT HOME AND HAS NO TRAINING IN A SCHOOL SETTING, AND (2) SCHOOL PLACEMENT OF SUCH CHILDREN WOULD PRODUCE DESIRABLE CHANGES IN THE HOME ADJUSTMENT OF THE FAMILIES INVOLVED. DURING PHASE 1 A PICTURE VOCABULARY SCALE WAS DEVELOPED TO MEASURE CHANGES IN THIS ASPECT OF THE DEVELOPMENT OF TRAINABLE CHILDREN. THEN THE EXPERIMENTAL DESIGN OF THE STUDY WAS PERFECTED, SUBJECTS WERE IDENTIFIED, AND A PILOT STUDY WAS MADE CENTERING ON THE TWO HYPOTHESES. PHASE 2 OF THE STUDY CARRIED OUT THE EXPERIMENTAL DESIGN. IN SPITE OF SOME POSITIVE TRENDS, THERE WAS AN INDICATION THAT THE CHILDREN IN THE SPECIAL CLASSES DID NOT MAKE MORE PROGRESS IN SOCIALIZATION, INTELLECTUAL DEVELOPMENT, SELF-CARE, OR PARENT-CHILD RELATIONS THAN THE CHILDREN WHO REMAINED AT HOME. HOWEVER, THE SPECIAL CLASSES AS OPERATED APPEARED TO PRODUCE GREATER CHANGES IN THE HIGH IQ GROUP. LITTLE EVIDENCE ACCRUED TO SUPPORT THE CONTENTION THAT DAY CLASS TRAINING FOR TRAINABLE CHILDREN, AS PRESENTLY CONSTITUTED, IS EFFECTIVE. (GC)