An examination of the methodological perspectives of Mead and Blumer establishes considerable divergence in emphasis and consequence. Their diverse epistemologies rest upon equally diverse ontological assumptions. Blumer's naturalistic inquiry neither compliments Mead's methodological perspective nor facilitates the examination of Mead's theoretical ideas. Several experimental studies of "taking the attitude of other" are discussed, one by Mead's student Leonard S. Cottrell, which are consistent with Mead's methodological perspective. That perspective is equally applicable when exploratory-descriptive research is required. Many problems of naturalistic methodology are avoided by recognizing the different domains of meaning for investigator and investigated, by reducing the scope of behavior to be examined, and by increasing control of the investigator's behavior regarding that which is examined. An illustrative study is discussed.