How far can conflict be avoided by the separation of actual or potential adversary groups into politically discrete territories? The experience of Greeks and Turks suggests that such stability cannot be guaranteed in the long term because the strategic, demographic and other elements taken into account in a 'settlement' are themselves likely to change, perhaps radically and rapidly. The Greek-Turkish interface shows a variety of types of boundary/frontier at different stages over the course of time. It shows that where population (and other 'movables' such as political institutions) are relocated to conform to a new boundary, the line in question is neither wholly antecedent nor subsequent to the pattern of distribution of such population but is a special, composite case. Today, the stability of the boundary between Greece and Turkey established as a result of the 1923 Treaty of Lausanne is questioned because the significance of the Aegean itself has changed--especially for Turkey which looks to new questions about resources and access. Moreover, both countries share a new concern for another variable in their relationship, the question of Cyprus. Meanwhile, at the level of non-spatial, inter-communal relationships, the experience of Greece and Turkey suggests that the concepts of boundary and frontier may be as relevant as they are to patterns of territorial space. Such intergroup relationships may be demarcated (i.e. institutionally defined) with the precision which attaches to political boundaries or they may be broad 'cultural frontier' zones of contact and intermixing.