This article describes results of a reading comprehension intervention for students with adequate decoding but poor comprehension skills. Five teachers and 25 students in grades 3-5 from two rural public schools participated in this naturalistic experimental research study. Teachers met with identified students in a small group setting for 30 intervention sessions. The intervention involved explicit teaching and gradual release of instruction in three phases: metacognitive strategies, comprehension strategies, and peer-led discussions. To measure growth in reading comprehension, the Qualitative Reading Inventory-3 (Leslie & Caldwell, 2001) was administered as the pre- and posttest and analyzed through t-test comparisons. Interactive teaching is characterized by a dynamic flow of instruction with a powerful use of questioning used as a tool to assist students in understanding what they read. Recitative teaching is marked by static interactions that did not change across treatment intervention. Subsequently, the groups receiving the interactive instruction were compared to those receiving recitative instruction, and growth in reading comprehension for each group was compared. While all students gained in reading comprehension, students in the interactive teaching groups gained more in reading comprehension than those in the recitative teaching groups. Instructional implications of this research are presented and discussed, providing suggestions for teaching reading comprehension.