Leafcutter ants in the genus Atta (Hymenoptera: Formicidae: Attini) are considered major pests of agriculture and forestry in the Neotropics. Phorid flies (Diptera: Phoridae) have been proposed as viable candidates for biological control of ants because of the importance of their trait-mediated effects on their hosts. However, the impact of different densities of phorid flies has never been assessed in the field. Experiments were conducted by isolating 3-m sections of Atta vollenweideri Forel foraging trails with tunnels, and sampling ants in trails with 0, 1, or 4 Eibesfeldtphora trilobata Disney female parasitoid flies. Samples were collected every 30 min from these trails. We also collected a sample before introducing the parasitoids and another one 30 min after removing them from the trail. We measured traffic of ants on the trails, weight and type of plant material transported, and the proportion and size of the workers collected. The presence of phorids on the trails reduced the ant traffic and amount of plant material transported into the nests and decreased the proportion of workers on the trails in the size range preferred as hosts by the flies. The effect on worker size, as well as the lag effect recorded after phorids were removed from the tunnels, was more pronounced with four phorids. The presence of phorids also affected the weight of monocotyledon and dicotyledon material transported. Even at the minimum density possible, phorids significantly influenced a key aspect of the colony life, the food intake through foraging. From an applied point of view, our results show that releases of these phorids into the field should not necessarily involve many individuals to reduce foraging by A. vollenweideri, making them potentially useful candidates for biological control of these ants.