Two hundred and eight first graders were assigned to one of three treatment groups. Treatment A received initial instruction in letter names followed by sight words; treatment B received initial instruction in letter names and sounds followed by sight words; and treatment C received initial instruction in sight words followed by letter names and sounds. The time spent and the materials used in reading instruction outside the experimental situation were held constant for all three treatment groups. Pretesting measures included the Murphy-Durrell Reading Readiness Analysis, subtests of the Metropolitan Readiness Test, and the Lorge-Thorndike Intelligence Tests, Level I, Form B, Primary. Post-testing after 18 weeks included the word knowledge, word discrimination, and reading subtests of the Metropolitan Achievement Test, Form C, Primary I Battery. Statistically significant differences favored treatment group B over both treatments A and C in word knowledge and word discrimination. Differences were significant in favor of both B and C over A in comprehension, but no significant differences were found between B and C. Children of below-average readiness benefited in treatment A in word perception only, while children of average readiness benefited most from treatment B for all three criteria. Tables are included. (CM)