*Department of Agricultural Botany, J. V. College, Baraut
**Department of Forage Research, Haryana Agricultural University, Hissar
1Part of a Ph.D. thesis submitted by the senior author to the Meerut University, Meerut.
Genotype-environment interactions for grain yield in bread wheat (Triticum aestivum L.), studied by growing 21 genotypes in eight environments, revealed that the genotypes interacted significantly with the environments and a large portion of the interactions was accounted for by the linear function of the environmental means. High yielding genotypes ‘Kalyansona’, ‘C 306 × Hira’ F1 and ‘Agra Local × C 306’ F1 were found to be highly responsive but unstable, whereas genotype ‘Hira’ was least responsive as well as highly stable. It was also possible to predict the mean performance of most of the genotypes within the limits of sampling error. Use of only one stable genotype for the purpose of prediction was found to be quite useful. It was observed that for genotype-environment interaction studies, a minimum of three, preferably contrasting, environments would be sufficient both in order to select the most stable genotypes as well as to predict the mean performance.
