THE MAJOR PROBLEM OF THIS INVESTIGATION WAS TO DETERMINE TO WHAT EXTENT FIRST-GRADE PUPILS ARE CAPABLE OF LEARNING ALGEBRAIC STRUCTURES THROUGH PROGRAMED INSTRUCTION. IN THE EXPERIMENT APPROXIMATELY 130 FIRST-GRADERS WERE INSTRUCTED THROUGH AUDIOVISUAL TEACHING MACHINES FOR APPROXIMATELY 15 WEEKS. AT THE END OF THE PROGRAM, THE CHILDREN WERE TESTED THROUGH ORAL INTERVIEWS AND GROUP TESTS ON THEIR UNDERSTANDING OF BASIC MATHEMATICAL PRINCIPLES. THEIR PERFORMANCE WAS COMPARED WITH THAT OF TWO CONTROL GROUPS, WHO RECEIVED ONLY CONVENTIONAL INSTRUCTION. ONE CONTROL GROUP WAS DRAWN FROM THE FIRST GRADE, THE OTHER FROM THE FOURTH. IT WAS DISCOVERED THAT THE AUTOINSTRUCTIONAL PROGRAM WAS SUCCESSFUL IN DEVELOPING CERTAIN COMPETENCIES IN ARITHMETIC, SUCH AS ORDERING OF NUMBERS, WITH NO APPARENT LOSS IN ABILITY TO PERFORM SIMPLE ADDITION AND SUBTRACTION. HOWEVER, THE PRINCIPAL INVESTIGATION POINTS OUT THAT A CONCLUSION REGARDING THE MAJOR HYPOTHESIS, THE EXTENT TO WHICH FIRST-GRADERS ARE CAPABLE OF LEARNING ALGEBRAIC STRUCTURES, HINGES ON ONE'S JUDGMENT OF THE VALIDITY OF THE INDIVIDUAL TEST USED AS THE MAJOR CRITERION. FURTHER STUDY AND LONG-RANGE EXPERIMENTATION WERE RECOMMENDED. (JC)