FOUR NULL HYPOTHESES WERE TESTED TO EVALUATE SELECTED READING READINESS TESTS AS PREDICTORS OF FIRST-GRADE READING ACHIEVEMENT. FIVE SCHOOLS IN THE BANGOR, MAINE, CITY SCHOOL SYSTEM WERE RANDOMLY CHOSEN. IN EACH SCHOOL, ONE CLASS USED THE EXPERIMENTAL PROGRAM WHICH EMPHASIZED SUPPLEMENTARY VOCABULARY INSTRUCTION WITH READINESS TRAINING. ANOTHER CLASS USED THE CONVENTIONAL PROGRAM WITHOUT VOCABULARY INSTRUCTION. BOTH PROGRAMS WERE BASED ON THE HARPER-ROW BASAL READERS. READING READINESS WAS DETERMINED BY THE SELECTED READING READINESS TESTS DURING THE SECOND WEEK OF THE SCHOOL YEAR. READING ACHIEVEMENT WAS TESTED BY INDIVIDUAL WORD RECOGNITION TESTS BASED ON THE VOCABULARY OF THE FOUR PREPRIMERS USED AND BY SPACHE'S DIAGNOSTIC READING SCALES. RESULTS STATISTICALLY ANALYZED AT THE UNIVERSITY OF MAINE COMPUTING CENTER INDICATED THAT PREDICTION OF SUCCESS IN FIRST-GRADE READING WAS SIGNIFICANTLY IMPROVED BY USING COMPOSITE SUBTESTS, BY CLEARLY DEFINING THE ACHIEVEMENT CRITERIA, AND BY ORGANIZING AND CONTROLLING INSTRUCTIONAL VARIABLES. NINE SUBTESTS WERE IDENTIFIED, AND IT WAS FOUND THAT INTACT TESTS DID NOT REALLY MEASURE FEATURES OF READINESS. AN EXTENSIVE BIBLIOGRAPHY AND TABULATED TEST RESULTS ARE INCLUDED IN THIS FINAL REPORT OF A STUDY DONE UNDER CONTRACT WITH THE DEPARTMENT OF HEALTH, EDUCATION, AND WELFARE. (NS)