This report discusses the Columbia Classroom Environments Project Autonomy in Learning An Exploration of Pupils' and Teachers Roles in Different Classroom Environments to develop Criteria and Procedures for Evaluation in Project Follow Through. The report begins with a discussion of a set of matrices which combine all of the behavior code categories which have been used in the research up to the time of this report. The construction of the matrices involved much working back and forth between the matrices and the pilot study data. This process and the evaluation of the matrices are on-going. The report first discusses the general structure of behavior in learning environments, suggesting some of the basic assumptions and hypotheses of the set of matrices and illustrating for each of the predicted locations of the four types of classrooms which it was planned to observe in spring 1971 in terms of the pilot observations made in fall 1970. Next, some of the argument up to this point as it bears on the definition and construct validity of autonomy is summarized. Third, the intra-individual consistency of children's behavior will be related to their inter-individual consistency, in terms of the pilot observations. Finally, the decisions which were made about the conduct of the spring observations are discussed. (Author/JM)