A study was conducted to determine how well gifted students in grades 4-6 should be reading and whether traditional reading expectancy formulas are useful in predicting reading achievement of gifted students in the upper elementary grades. The total reading scores on the Comprehensive Test of Basic Skills (Form Q2) for 624 fourth and sixth graders were divided into categories according to the student's IQ range and grade level. Average reading achievement ranged from two to nearly four years above grade level. Reading achievement increased in direct relationship to IQ. No significant differences were found between the achievement of males and females within each IQ range at each grade level studied. Three reading expectancy formulas, the Harris, Los Angeles, and Bond and Tinker, were used to compare the actual reading achievement of the gifted fourth and sixth graders with predictions of their performance at this time. The Harris formula predicted actual reading achievement most accurately. It was concluded that the Harris formula is a more accurate predictor, that it should prove useful in predicting individual achievement in reading, and that it could be useful in identifying remedial readers among the gifted population. (KM)