This study reports correlations among scores from two commonly used individual measures of intelligence, the Revised Stanford Binet and the Peabody Picture Vocabulary Test, and reading subtests of a standardized achievement test with a group of elementary-aged disadvantaged children. The present study employs longitudinal design to assess correlations among beginning-first grade IQ scores and measures of reading performance obtained toward the end of the first, second, third, and fourth grades. The content of both reading and IQ tests thus remained essentially constant, thereby avoiding the confounding of reading ability with the assessment of intelligence. A complementary objective of the study was to determine the test-retest reliabilities of the two IQ measures for disadvantaged children during the same time interval. Determining reliability and validity of these measures is of particular interest since both tests were standardized on white children of essentially middle socioeconomic backgrounds. The sample included 36 subjects, 18 boys and 18 girls, from the public schools of metropolitan Nashville, Tennessee. The sample comprised children who had participated in an experimental reading project in which they were taught to read by the initial teaching alphabet method during the first and second grades. The findings suggest that verbal ability, particularly verbal proficiency, becomes more highly associated with reading performance as the child proceeds through school. (Author/JM)