A capture-recapture experiment was conducted to examine the effects of satiation state on the activity and foraging ef®ciency of the carabid beetle Pterostichus melanarius Ill., in four different habitats (a hedge, a transition zone between the hedge and the adjacent wheat crop, the margin of the wheat crop, and the centre of the wheat crop) in an agricultural ecosystem in France. 2. Prey biomass and density decreased from the centre of the hedge to the crop margin and to the centre of the crop, where prey density was low but prey biomass was moderate. Starved and satiated beetles were released in circular enclosures and recaptured using pitfall traps for 6 days. 3. Immediately after release, starved P. melanarius were signi®cantly more active than satiated beetles, but this difference disappeared after the ®rst day. At each recapture date, starved individuals had gained weight whereas satiated individuals had lost weight. The analysis of activity and weight variation at the end of the frst day indicated that the centre of the crop seemed to be the most favourable habitat for starved individuals to find food rapidly, whereas the behaviour of satiated individuals was identical in all habitats. 4. This experiment shows that habitat use of P. melanarius is flexible. The results are discussed in the context of the classical view of foraging behaviour in carabid beetles. The activity and foraging effciency in P. melanarius in different habitats can be explained as the result of the interaction between individual satiation state, prey availability in the habitat, and particular preferences of the species.