Community seedbanks promote conservation and the use of crop genetic diversity, as well as supporting farmer seed systems. This study analyses seed flow and access to crop genetic diversity over time in the Nakaseke, Rubaya, and Kibuga seedbanks of Uganda. The modes of operation of the banks were compared through scrutinizing records of crops and varieties being conserved, quantities of seed distributed, to whom, and quantities returned. The Nakaseke seed bank distributed the highest varietal diversity of common bean (Phaseolus vulgaris) and groundnuts (Arachis hypogaea L.), whereas the Rubaya seedbank distributed the highest quantity of common bean seed, followed by the Kibuga seedbank. There were no significant differences between the type of variety of seed, quantities of seed accessed, and seed returned to the seedbanks by women and men—except for the Nakaseke seedbank, where women returned significantly higher quantities of common bean seed. The Kibuga and Rubaya seedbanks dealt with individual farmers, whereas the Nakaseke seedbank dealt with individual farmers and groups. The extent to which core functions were achieved by a particular seedbank depended on the mode of operation, including actors, management, degree of development, socio-economic setting, among others. Further research is recommended to unpack these factors and come up with the most appropriate combinations for greater seedbank effectiveness.

Banana and plantain (Musa spp.) are very important crops in Ecuador. Agricultural production systems based on a single cultivar and high use of external inputs to increase yields may cause changes in the landscape structure and a loss in biodiversity. This loss may be responsible for a decrease in the complexity of arthropod food webs and, at the same time, related to a higher frequency and range of pest outbreaks. Very little is known either about the ecological mechanisms causing destabilization of these systems or the importance of the diversity of natural enemies to keep pests under control. Few studies have focused on this issue in tropical ecosystems. Here, we address this problem, comparing two Musa-based agroecosystems (monocultivar and mixed-species plantations) at two sites in Ecuador (La Maná and El Carmen) with different precipitation regimes. The diversity of soil macro fauna, represented by arthropods, was established, as indicators of the abovementioned disturbances. Our ultimate goal is the optimization of pest management by exploring more sustainable cropping systems with improved soil quality. Arthropod abundance was higher in the mixed system at both localities, which was clearly associated with the quality of the soils. In addition, we found Hymenoptera species with predatory or parasitic characteristics over the pests present in the agroecosystems under study. These highly beneficial species were more abundant at the locality of La Maná. The mixed type of production system provides plant diversity, which favors beneficial arthropod abundance and permits lower agrochemical application without yield penalties in comparison to the monoculture. These findings will help in the design of Musa-based agroecosystems to enhance pest control.

Mycotoxins are chemical contaminants that are invisible, tasteless, chemically stable and survive food processing. Contamination along the agri-food chain is difficult to control since their production and spreading are due to numerous factors including temperature, relative humidity, insect infestation, and susceptibility of the host plant. This is a pilot study which aims at assessing the contamination level of deoxynivalenol (DON), and its plant metabolites (3AcDON, 15 AcDON, DON 3G), nivalenol, T-2 and HT-2 toxins, and ochratoxin A in thirty-seven traditional varieties of Ethiopian durum wheat, and aflatoxins B1, B2, G1, and G2 in thirty-one varieties of Ugandan groundnuts grown in non-intensive rainfed production systems. Results indicate absence of mycotoxin contamination in all durum wheat samples and negligible levels of contamination (below the maximum levels tolerated by international standards) in groundnut samples. Further studies are required to assess if non-intensive production systems and varieties have a role in preventing and/or reducing mycotoxin contamination of the crops.