Genotype-environment interaction of yield in yellow corn hybrids, using AMMI and SREG

Abstract

It is indispensable for corn (Zea mays L.) plant breeding programs to select homogeneous materials, with high yield and with stable agronomic attributes; also, that they have a good adaptability in contrasting environments. The objective of the work was to evaluate the stability and genotype-environment interaction of the yield of 36 hard yellow corn hybrids, evaluated in seven environments of Peru, during 2016-2018, these materials were analyzed using the AMMI (additive main effects and multiplicative interaction) and SREG (site regression) models. The design used in each experiment was a 6×6 lattice with three repetitions, and the response variable was grain yield. A combined analysis of variance was performed, in which statistical differences between them (p≤ 0.05) were detected, then the Tukey mean test (p≤ 0.05) was applied, finally the AMMI and SREG models were run and the biplot graphs of each statistical model were obtained. Of the interaction between PC1 and PC2, AMMI explained 45.5% and 15.3%, respectively, and SREG with 59.8% and 12.2%, for the same components. The trilinear hybrids Dk-5005 and AG-01 outperformed the double-cross hybrids. The AMMI model detected the existing GE interaction in grain yield, and the SREG accurately grouped the assessment sites into six mega-environments. The three environments of La Molina and that of Huánuco identified the two hybrids (Dk-5005 and AG-01) with the highest grain yield (11.524 and 11.359 t ha-1, respectively).