Research institutes seek solutions to reduce the use of pesticides in fruit production. They redesign new orchards, pesticides free (or nearly). It implies new forms of interactions with the local stakeholders as well as a change in the experimental paradigm and the content of work, with a key role of observations on the agroecosystem. This leads us to distinguish two territorial levels: the “ecological one” and the “sector one”. The first is related to the territory as a production environment (e.g. climate, type of soils, plants, animals, growers) whereas the second is related to the territory as a geographical space where different players of a production sector are located. We analyze how these two territorial levels are handled by the experimental stations. We discuss how the change in experimental paradigm brings questions to the experimental stations’ workers, regarding territories issues and suggest that giving more centrality to work activity might create new relations between the stations and their territories.