Analysis of Genotype‐Environment Interactions of Potato Yield1

Publisher Website
Google Scholar
Abstract
Three methods were used to study genotype‐environment (GE) interactions of potato (Solanum tuberosum L.) yield. They were chosen to represent single (stability variance σ1 of Shukla, 1972), two (genotypic stability parameters α, λ of Tai, 1971), and three (genotypic components v1, v2 and v3 of Tai, 1975) parameters of measurement of genotypic contributions to the GE interactions. Data of marketable yield (W) and three yield components: number of stems/plot (X), number of tubers/stem (Y), and mean tuber weight (Z) of 30 genotypes were collected from nine trials in New Brunswick. Significant differences were found for σ2, α, and λ among the genotypes. The estimates of σ2 were highly correlated to those of λ. Both parameters showed a curvilinear relationship with v1 and v2 suggesting low σ2 and λ values were due to moderate genotypic responses to the environmental components (r1 and r2) supporting X and Y. The linear response parameter α showed close association with v3 which measured the genotypic response to the environmental component (r3) supporting Z. The estimates of v3 showed a moderately positive correlation with the mean marketable yield (W) and were larger than those of v1 and v2 for most genotypes. The estimates of r3 were larger and more variable than those of r1 and r2 over the environments. It is thus concluded that difficulty would be encountered locating high yielding genotypes without sensitive responses to the variation of r3.