A STUDY WAS CONDUCTED OF THE DIFFERENCES AMONG PUPILS OF DIFFERING ABILITY LEVELS BOTH IN PROGRESS AND ACHIEVEMENT. A COMPARISON OF THESE DIFFERENCES WAS THEN MADE WITH THOSE USING CONVENTIONAL INSTRUCTION PROGRAMS. TWENTY-THREE HYPOTHESES WERE TESTED TO DETERMINE IF SIGNIFICANT DIFFERENCES EXISTED AMONG LOW-, AVERAGE-, AND HIGH-ABILITY STUDENTS IN DIFFERENT GRADE LEVELS IN SCORES ON ARITHMETIC ACHIEVEMENT TESTS. THE EXPERIMENTAL GROUP CONSISTED OF 66 FOURTH-, FIFTH-, AND SIXTH-GRADE STUDENTS OF DIFFERING ABILITY LEVELS. ALL OF THESE STUDENTS STUDIED ARITHMETIC UNDER THE INDIVIDUALLY PRESCRIBED INSTRUCTION (IPI) PROGRAM. THE CONTROL GROUPS CONSISTED OF 399 STUDENTS MATCHED IN ABILITY AND GRADE LEVELS AS THE EXPERIMENTAL GROUP. THESE STUDENTS RECEIVED ARITHMETIC IN A TRADITIONAL CLASSROOM PROGRAM. VARIOUS ACHIEVEMENT MEASURES WERE IMPOSED FOR DATA COLLECTION. ANALYSIS OF COVARIANCE WAS USED TO TEST THE SIGNIFICANT DIFFERENCES. THE FINDINGS INDICATED, IN GENERAL, HIGHER ABILITY STUDENTS PERFORMED AND MASTERED MORE THAN DID THE LOWER ABILITY STUDENTS UNDER THE IPI PROGRAM. THE CONCLUSIONS INDICATED, HOWEVER, NO SIGNIFICANT DIFFERENCE WAS FOUND AMONG HIGH-, AVERAGE, AND LOW-ABILITY STUDENTS IN ARITHMETIC PROBLEM-SOLVING SCORES WHEN ACCOUNTING FOR PRETEST PERFORMANCE. A NUMBER OF RECOMMENDATIONS FOR FURTHER RESEARCH WERE DISCUSSED. (RS)