Effect of genotypes and GE interaction on starch content of cassava

Objective of this study is to explore the effect of genotype and genotype x environment on starch content percentage of 5 cassava genotypes grown under several environments in early and late rainy seasons. The results indicated that there was less seasonal effect on genotypic performance. There was more GE interaction in late rainy season. The sum of squares of GE interaction was partitioned using 3 techniques, viz. linear regression (REG), using Eberhart and Russell (1966) stability, and Additive Main effects and the Multiplicative Interaction (AMMI) model in 2 principal component axes. The AMMI model can explain a large portion of sum of squares, and thus more effective in partitioning the GE interaction SS than the REG technique. Grouping of homogeneous environments using cluster analysis resulted in almost the same genotype ranking in each group. Environmental grouping can then be used to reduce the reversal GE interaction.