Background/Context: The achievement effects of Comprehensive School Reform (CSR) programs have been studied through the use of input-output models, in which type of CSR program is the input and student achievement is the output. Although specific programs have been found to be more effective and evaluated more than others, teaching practices in CSR schools have received less attention. This study focuses on observations of math and reading/language arts lessons in classrooms implementing an array of CSR programs to better understand what occurs in CSR classrooms. Research Question: This article describes observed instructional practices and teacher-student dynamics that occurred in CSR classrooms in two different subject areas, reading/language arts and math, in Grades 3-5. Reading/language arts included several literacy-related areas such as spelling, vocabulary, phonetics, and writing. Math primarily consisted of computations and math applications. Our primary research questions were: (1) Does subject matter matter in CSR classrooms? (2) How does instruction in CSR classrooms differ among Grades 3, 4, and 5? (3) Are there major differences in classroom practices between fall and spring? Population: Teachers (N = 104) in Grades 3-5 in 16 CSR schools, totaling 248 observation periods in math and reading/language arts lessons. Research Design: Observational study using a systematic coding system to observe student/teacher classroom behaviors and dispositions. Data Analysis: Data were analyzed using basic descriptive statistics and analysis of variance procedures. Conclusion: We found that students were productively involved in assigned tasks and that classrooms were pleasant and task oriented in both mathematics and reading/language arts. Some subject matter differences were notable, particularly that math lessons were more structured and rigid than were reading/language arts lessons. Also of interest, third- and fifth-grade classrooms experienced more positive teacher-student relationships than did fourth-grade classrooms. Furthermore, instruction in the fall was more structured and more focused on basic skills than in the spring. Overall, students appeared to be engaged in learning basic facts/skills in an uninterrupted teacher-directed classroom. Students did what was asked of them, were given little choice about their social and academic tasks, and were in mostly comfortable classroom environments.