AbstractThe purpose of this paper is to propose a methodology for the naturalistic investigation of classroom environments in primary schools. At its centre, this proposal is derived from a critique of the position of the classroom ecologist Walter Doyle. The paper has three distinct phases. The first strand is concerned with analysis of aspects of Doyle's work, especially in connection with his position on classroom ecology. This section will include an overview of his approach to naturalistic research, together with an outline of his view on the definitions of tasks and activities. Reference will also be made to the source of his work, ecological psychology. The second phase of the paper relates to Doyle's methodological position, including the use of classroom narratives as a research method together with an outline and critique of the role of dimensions of task ambiguity within naturalistic research. Finally the concept of dimensions of classroom activity ambiguity is proposed. This is used as a foundation for suggestions for empirical research methods for natural investigation.