T HE initial stimulus for this paper' was provided by my inability to give simple answer to the simple question: Whom did you study in the field? The reasons for this inability concern the ways in which ethnologists demarcate ethnic and account for their survival. These are complex and important issues which require thoughtful and extensive research. It would be false to claim that this short paper raises all of the relevant questions and absurd to claim that it answers them. Nevertheless, I hope that this presentation of data about the Lue2 will both encourage others to recognize in their own work the need for identifying and delimiting ethnic entities and suggest some questions and field procedures which may help them to do this. Since comparison is basic to anthropology (Lewis 1956), it is important that our be comparable (K6bben 1952:132). Yet, it is apparent that the neat ethnic labels which we anthropologists use frequently deceive us. In reading about various areas of the world one frequently encounters ethnic names with unclear referents and groups of people with no constant label. Raoul Naroll, in recent article (1964), demonstrates that one source of confusion is our lack of agreement about the criteria which define the entities-variously called tribes, cultures, societies, peoples-which we describe. Such lack of agreement is obviously challenge to global comparisons. As this paper will demonstrate, it also has implications for ethnographic fieldwork. Naroll lists six criteria: trait distributions, territorial contiguity, political organization, language, ecological adjustment, and local community structurecommonly used to demarcate ethnic entities. In addition to Naroll's specific criticisms of them, I would add that these and similar criteria have three main shortcomings as delimiters of culture-bearing units (Naroll 1964:283). 1) Since language, culture, political organization, etc., do not correlate completely, the delimited by one criterion do not coincide with the delimited by another. 2) If by culture we wish to mean a pattern, set of plans, blueprint for living (Naroll 1964:288), then delimited by combinations of these criteria, including the combination which Naroll suggests, are only occasionally and accidentally culture-bearing units. 3) It is often difficult to discern discontinuities of language, culture, polity, society, or economy with sufficient clarity to draw boundaries. It is this which makes me suggest that the delimitation of ethnic entities is especially problematic in all parts of the world which are continuously inhabited but not divided into either sharp ecological zones or strong and durable states. Under such conditions, it becomes quite