Two readability formulas were applied to two math and two reading series textbooks at the fourth, fifth, and sixth grade levels to see whether the publishers' suggested reading levels were accurate. Five samples were taken from each text and compared to the publishers' suggested usage according to grade level. Results showed that although formulas agreed with the publishers' general readability averages, vast ranges of readability existed within specific units of texts. The study concludes that when considering a textbook for any course of study it would be wiser to look at specific unit readability levels rather than the general statistical average. (RL)