THE PURPOSE WAS THE DISCOVERY AND EVALUATION OF GENERAL PROGRAMING PRINCIPLES FOR TEACHING MACHINES. THE DEVELOPMENT OF A SELF-INSTRUCTIONAL PROGRAM IN INFORMAL GEOMETRY FOR SECOND-GRADERS WAS THE FIRST STEP OF THE PROCEDURE. THE 170-ITEM PROGRAM CONSISTED OF FIVE UNITS DESIGNED (1) TO FAMILIARIZE THE CHILD WITH MACHINE OPERATION AND (2) TO TEACH SUCH CONCEPTS AS THE STRAIGHT LINE, THE ANGLE, AND THE TRIANGLE. ACHIEVEMENT TESTS WERE DEVELOPED TO ASSESS THE EFFECTS OF THE SELF-INSTRUCTIONAL PROGRAM, AND TO FACILITATE THE SELECTION AND DESCRIPTION OF RESEARCH SUBJECTS. FIFTEEN PUPILS WERE TAKEN THROUGH THE PROGRAM AND COMPARED WITH MATCHED CONTROLS WHO DID NOT GO THROUGH THE PROGRAM. A STUDY WAS CONDUCTED TO CHECK THE ASSUMPTION THAT THE KNOWLEDGE OF GEOMETRY COVERED IN THE CHILDREN'S PROGRAM VARIED LITTLE DURING THE PRIMARY GRADES. A STUDY WAS ALSO CONDUCTED TO DETERMINE THE IMPORTANCE OF PRESENTING PROGRAM ITEMS IN THE ORIGINAL AND LOGICAL ORDER IN WHICH THEY HAD BEEN DEVELOPED. MATCHED GROUPS OF 18 SUBJECTS EACH WERE TAKEN THROUGH TWO VERSIONS OF THE PROGRAM. THE CONTROL GROUP WENT THROUGH THE STUDENT PROGRAM, WHILE THE EXPERIMENTAL GROUP WENT THROUGH A VERSION OF THE PROGRAM IN WHICH THE ITEM OF ONE UNIT HAD BEEN RANDOMLY SCRAMBLED WITHIN BLOCKS OF 20 ITEMS. THE LEARNING ACHIEVEMENT OF THE EXPERIMENTAL GROUP WAS NOT SIGNIFICANTLY DIFFERENT FROM THAT OF THE CONTROL GROUP. (HB)