On farm assessment of agronomic performance for 12 yam cultivars (Dioscorea rotundata) in Benin using Additive Main effects and Multiplicative Interaction models
Yam (Dioscorea cayenensis - D. rotundata) occupies an important place in terms of food, economic and socio-cultural domain in Benin. Despite the existence of a large number of successful cultivars, the choice of these remains difficult for producers because of the variability of the additive effect of the environment and soil and climate conditions of the different production areas. The present study aims to orient seed producers and farmers in the choice of yam cultivars through high production areas. For that, eight popular local cultivars and four yam hybrids were subjected to agronomic evaluation using RCBD with three replications in six farmers’ field selected from the major producing areas. The yield data collected on the different sites were analyzed using Additive Main effects and Multiplicative Interaction (AMMI) models. The results show a high variability within genotypes and locations (P˂0.001). The yield varies not only from one genotype to another, but also across the locations. In addition, significant interactions (Pr (> F) <2.2e-16) were observed between cultivars and the targeted regions. Based on yield performance, the hybrid TDr-A21-2003 responds better in Tchakalakou agro-ecology, Wokourou in Tcheti zone, Amoula for Adihinlidji region, Hêapala in Sakarou region and Soussouka in the regions of Tchaourou and Ouaké. However the hybrid TDr-A39-2003 is the most stable cultivar across regions. Data sheets can be developed for these cultivars to facilitate their registration in the national catalogue and release for the benefit of farmers.
Copyright (c) 2025 Bulletin de la Recherche Agronomique du Bénin

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Les articles publiés par le Bulletin de la Recherche Agronomique du Bénin sont en libre accès. Ils sont gratuits pour tout le monde, immédiatement téléchargeables dès la publication et distribués sous la licence CC BY-NC-ND (https://creativecommons.org/licenses/by-nc-nd/4.0/).