The epistemological roles of sociology and geography are discussed in an attempt to identify the scope of social geography. The examination begins by rejecting the methodological exceptiona]ism of either sociology or geography, partly on the basis of an assertion that both require the scientific investigation of multiple cases and the development of understanding by means of subjectively guided integrations, sometimes of specific cases. It is concluded that the two subjects are defined most satisfactorily by their focal subject matters. The wide scope ascribed to sociology by Comte is supported even though it implies that social geography is virtually synonymous with human geography. Emphasis is placed on the role of geography as the study of places. Both the nature of places and the methods required for their study demonstrate again the interdependence of nomothetic spatial investigations and of place-related syntheses. The final sections of the paper argue pragmatically that although the programme for social geography can be defined as no less than that of human geography, in practice social geographical studies might reasonably concentrate on those aspects which are neglected by economic, political and historical geographers. These include the geographical aspects of those microscopic and mesoscopic groups and institutions which express affective, affirmative, cultural, recreational and spiritual aspects of social life. It is also accepted that social geography has recently neglected a task which no other subject approaches, the writing of explanatory descriptions of areally defined social complexes with emphasis upon their geographical context and structure.