A 3-YEAR EXPERIMENTAL PRIMARY-GRADE READING PROGRAM CONDUCTED WITH A 50 PERCENT REDUCTION OF READING CLASS SIZE WAS EVALUATED IN THE RIVERSIDE, CALIFORNIA, UNIFIED SCHOOL DISTRICT FOR THE YEARS 1962-1965. TEST SCORES WERE ANALYZED FOR 656 CHILDREN IN THE EXPERIMENTAL GROUP WHO HAD 2 OR MORE YEARS OF EXPERIENCE IN THE EXPERIMENTAL PROGRAM AND FOR 602 CHILDREN IN THE CONTROL GROUP WHO HAD 1 YEAR OR LESS IN THE EXPERIMENTAL PROGRAM. THE SMALL-CLASS PROGRAM (AN AVERAGE OF 15 STUDENTS PER CLASS) WAS INITIATED IN A STRATIFIED RANDOM SAMPLE OF SEVEN ELEMENTARY SCHOOLS THE 1ST YEAR, WITH SIX SCHOOLS ADDED THE 2D YEAR AND EIGHT MORE ADDED THE 3D YEAR. DATA GATHERED FROM THE METROPOLITAN READINESS TESTS IN GRADE 1, METROPOLITAN ACHIEVEMENT TEST IN GRADES 2 AND 3, THE CALIFORNIA SHORT FORM TEST OF MENTAL MATURITY IN GRADE 2, AND THE SCHOOL AND COLLEGE APTITUDE TEST IN GRADE 4 WERE ANALYZED ACCORDING TO EXPERIMENTAL AND CONTROL GROUP READINESS, INTELLIGENCE, READING ACHIEVEMENT, AND SEX DIFFERENCES. THE EXPERIMENTAL GROUP ACHIEVED SIGNIFICANTLY HIGHER THAN THE CONTROL GROUP, ACCORDING TO THE AMOUNT OF TIME SPENT ON THE EXPERIMENTAL PROGRAM. FINDINGS INDICATED THAT FIRST GRADE INSTRUCTION WAS MOST CRUCIAL AND THAT BOYS BENEFITED MORE FROM REDUCED CLASS SIZE THAN GIRLS. TABLES ARE INCLUDED. (LS)