This randomized controlled trial examined the impact of adaptive, web-based, scaffolded silent reading instruction on 426 fourth- and fifth-grade students in an urban US school district. Reading proficiency was evaluated in the fall and spring using the Group Reading Assessment Diagnostic Evaluation (GRADE) and an eye movement recording system (Visagraph). Fall GRADE scores and demographic factors were used to pair students. One member of each pair was then randomly assigned to the treatment or control condition and their pair to the alternate condition. During scheduled 25-minute supplemental literacy blocks, students in the control group received "business-as-usual" reading instruction, whereas the treatment group engaged in scaffolded silent reading instruction. Structural equation modeling indicated that scaffolded silent reading instruction produced significantly larger gains on measures of reading efficiency in grade 4 and significantly larger gains on the GRADE reading achievement measures in grade 5. Implications for practice and future research are discussed.