The effect of different variables on the territory size and defence level of tawny owl (Strix aluco L.) populations was studied in two habitats: mixed farmland (1976–1978) and town (1986–1988). Territory size was determined by plotting the position of territorial challenges (hoots) of males which were individually recognizable in most cases. No differences were found in size, defence, number of competitors and biomass between territories in farmland and in town, but territories in town were more fragmented and showed a larger nearest neighbour distance. Size of rural territories was adequately predicted both by some structural habitat features and food supplies, while size of urban territories was affected only by habitat structure and by years of territory occupancy. No effect on territory size by competing neighbours was found in either habitat, thus confirming that behavioural interactions between the owners of territories and potential settlers prevent a decrease of territory size.

Defence level of urban territories was affected by 'social’variables (nearest neighbour distance and years of occupancy), structural variables (percentage of poplar grove and diversity), and food supply, while in rural territories only structural variables (fragmentation or natural vegetation) and food supplies affected defence level.

This was due to different settlement tactics in the two areas, depending on the extent and quality of suitable habitat.