To evaluate whether changes in theory and research on the basic processes in reading comprehension have led to changes in classroom practice, a study conducted a modified replication of a 1967 study by F. J. Guszak. In the 1967 study, Guszak recorded and analyzed reading group story discussions for students and teachers in grades 2, 4, and 6, revealing an overall emphasis at all grades on literal level questions, with a few more evaluation level questions than either conjecture or explanation questions. Fifteen teachers and their classrooms participated in the current study. All class discussions were videotaped and transcribed for analysis. Using similar analysis schemes, this study reported a sharp reduction in literal level questions, with a large increase in the explanation category, but with the percentage of evaluation and conjecture questions comparable to the original study. While significant changes were found in the type of questions governing discussions, almost no change was found in the patterns of interactions, such as the frequent repetition of IREs (teacher initiates an interaction, student responds, and teacher evaluates). Observations indicate that teachers still dominate discussions and control interactions, but rely more on questions that invoke students' knowledge and less on text-based questions. (Five tables of data are included and 41 references are attached. Appendixes describe and report interaction categories and provide inter-rater agreements.) (MM)