BACKGROUND: Leaf-cutting ants (LCAs) are amongst the most important forestry pests in South America. Currently, their control is carried out almost exclusively through the application of toxic baits of restricted use. Here we evaluate a push-pull strategy (i.e., the simultaneous use of attractant and repellent stimuli in order to divert pests) to manage LCAs Acromyrmex spp. in young willow plantations in the area of Delta of the Parana River, Argentina, a wetland ecosystem. First, we surveyed ants' selection of farmland vegetation during one year. Then, we estimated ants' preferences between the willow Salix babylonica and a subsample of plant species from farmland vegetation under laboratory conditions. Finally, we designed and performed a fully crossed experimental field assay to evaluate a push-pull strategy by using farmland vegetation as pull stimulus. RESULTS: We surveyed 39 plant species in the area, 19 of which had been foraged by LCAs along the year. Plants were selected by species, not by abundance. In the lab, ants showed similar preference for the cultivated willow and the subsample of plant species. Push-pull was the only treatment that maintained willow remaining vegetation above 60-80% at the end of the growing season. CONCLUSIONS: For the first time the push-pull strategy was evaluated in social insects. We demonstrated that it can be successfully used to manage LCAs in young willow plantations. Our strategy generates biodiversity, which can improve the ecosystem functioning, and it can be easily implemented by producers since its design is based on regular willow plantations.

Leaf‐cutting ants are a serious pest of young forestry plantations. Currently, the main control method is the use of broad‐spectrum insecticides, which have a negative effect on non‐target organisms and the environment. In this work, plant‐based compounds were evaluated in laboratory assays with Acromyrmex ambiguus Emery (Hymenoptera: Formicidae) for their potential use as repellent and attractant stimuli to be used in a push‐pull strategy. Farnesol, a sesquiterpene present in many essential oils, was tested as a repellent at doses of 10, 50, and 100 mg. Its distance of action was studied by comparing the repellent effect of farnesol in a situation in which ants had to touch the farnesol in order to reach the food source in comparison to when ants could reach the food source without getting into direct contact with it. Different parts of the orange fruit (pulp and peel) were evaluated and compared as attractants, given that citrus‐based baits are among the most popular attractants used. Results from laboratory bioassays indicated that farnesol is repellent at doses of 50 mg and acts upon contact or at a very short distance. Furthermore, orange pulp was more attractive than the peel, and volatile compounds were highly responsible for the attraction. When both stimuli were tested simultaneously in a laboratory experiment, repellency of farnesol was enhanced in the presence of orange pulp odor. When tested in a field push‐pull experiment, the results also showed a good repellent effect of farnesol as well as an attractant effect of the orange pulp. These results encourage long‐term studies with these substances in a field setting and suggest that repellents can be enhanced by the use of attractants to manage leaf‐cutting ants behavior.

Leaf-cutting ants are considered major pests of agriculture and forestry in the Neotropics. Attractive toxic baits are the prevailing method for managing them. Thus, there is great interest in identifying attractants to incorporate into these baits. Moreover, leaf-cutting ants can avoid toxic baits by associating the attractant with the toxin. We evaluated attractiveness of heptyl butyrate, a volatile compound found in fresh apples and plums. We conducted field experiments with ten colonies of Acromyrmex lobicornis. First, we evaluated the behaviour of ants exposed to heptyl butyrate at 1% by surrounding resources. Then, we compared the attractiveness of heptyl butyrate and orange pulp, the most commonly used attractant. Finally, we evaluated whether heptyl butyrate increases the attractiveness of a carbohydrate resource at varying doses. Heptyl butyrate at 1% attracted 92% more ants than the control and that it was as attractive as orange pulp. Heptyl butyrate paired with sucrose at concentrations of 0.001 and 0.1% was more attractive than sucrose alone, but greater concentrations did not increase sucrose's attractiveness. Therefore, heptyl butyrate could be added to toxic baits to manage A. lobicornis as it is as attractive as the most commonly used attractant and can be applied directly to the pellets.