Ten inclusive middle school social studies classes, including 133 general education students, and 24 students with mild disabilities (21 with learning disabilities and 3 with emotional disabilities), were assigned at random to a traditional instruction condition, or an experimental condition involving classwide peer tutoring with specialized materials and parent training. After 18 weeks of instruction, posttest data revealed that students in the experimental condition gained significantly more than students in the traditional instruction condition. These effects were observed on content included in the tutoring intervention, as well as on related content that was taught but not included in the tutoring intervention. Results are discussed within the context of recent research on inclusive secondary content area instruction.