Vegetable gardens are important for maintenance of food security and biodiversity. This is particularly true for insects, as the presence of species in vegetable gardens is defined by traditional control methods that have been applied for generations and are linked to the knowledge of traditional communities. In this context, we aimed to identify the knowledge of rural communities regarding family vegetable gardens, including the vegetables cultivated, the main vegetable pests, their control methods, and the role of ants in these environments. We interviewed 46 families from five rural communities in Santa Rita de Caldas, Minas Gerais, Brazil. Among them, 44 families reported that they currently have or have had a vegetable garden at their residence. A limited number of vegetables ethnospecies are commonly cultivated, primarily lettuce and kale. The families possess a broad knowledge of alternative pest control methods. Regarding ants, the communities generally view them only as pests to vegetables. In this context, strategies to preserve traditional knowledge and disseminate it among traditional communities are essential for the preservation of the regional culture and the conservation of agrobiodiversity.
Diaspore removal by ants is a crucial stage for successful myrmecochory and can be directly or indirectly affected by natural or anthropic changes to environments. Among the consequences of such changes is variation in habitat attributes, such as changes in conditions and resources and, consequently, decreased diaspore removal or even the loss of this ecological function. The aim of this study was to assess whether canopy and litter cover affect diaspore removal by ants in the Cerrado. We considered canopy and litter cover as proxies of humidity and temperature and evaluated whether changes in these environmental conditions could affect diaspore removal by ants. We hypothesized that the greater the canopy and litter cover (higher humidity and lower temperature), the greater the number of diaspores removed by ants. We tested this hypothesis by establishing three classes of cover for each proxy: low, intermediate, and high. We placed artificial diaspores under each cover class and quantified the number of diaspores removed. We found that variation in canopy and litter cover did not affect the number of diaspores removed by ants in areas of cerrado sensu stricto. Variation in habitat attributes in natural environments were less important for diaspore removal than in modified areas. Our results indicate that understanding the processes and habitat atributes involved in diaspore removal by ants is important for conserving the Cerrado.