The stability of grain yield of 11 late maize hybrids with a check cultivar were studied in a randomized complete block design with 4 replications for 2 years at 11 sites in multi-environment trials in Iran. The obtained data were analyzed with a GGE [genotype + (genotype × environment interaction)] biplot method. The first 2 principal components (PC1 and PC2) were used to create a 2-dimensional GGEbiplot and explained 44% and 27% of GGE sum of squares (SS), respectively. Ranking of the genotypes based on the ideal genotype revealed that grain yields of G2, G3, and G12 were higher together with being more stable. Ranking of environments for the ideal environment showed that site Rasht (RAS) was the most discriminating, whereas site Hamedan (HAM) was the most representative. There were 5 winning genotypes and 3 mega-environments in our study. The GGE biplot suggests 3 late maize mega-environments in Iran: a minor mega-environment (northern Iran consisting of Rasht and Ghaemshahr), an average one (southern and central Iran consisting of Shiraz (SHI) and Esfahan (ESF)), and a major one (western and central Iran consisting of Arak (ARA), HAM, Karaj (KAR), Kermanshah (KER), Khoramabad (KHO), Varamin (VAR), and Sanandaj (SAN)). As a result, the findings from our study are as follows: (1) genotype G2 was the most stable and is thus recommended for commercial release in Iran; (2) the GGEbiplot method can be used to identify superior genotypes for target sites in Iran and sites in other parts of the world.

The stability of grain yield of 11 late maize hybrids with a check cultivar were studied in a randomized complete block design with 4 replications for 2 years at 11 sites in multi-environment trials in Iran. The obtained data were analyzed with a GGE [genotype + (genotype × environment interaction)] biplot method. The first 2 principal components (PC1 and PC2) were used to create a 2-dimensional GGEbiplot and explained 44% and 27% of GGE sum of squares (SS), respectively. Ranking of the genotypes based on the ideal genotype revealed that grain yields of G2, G3, and G12 were higher together with being more stable. Ranking of environments for the ideal environment showed that site Rasht (RAS) was the most discriminating, whereas site Hamedan (HAM) was the most representative. There were 5 winning genotypes and 3 mega-environments in our study. The GGE biplot suggests 3 late maize mega-environments in Iran: a minor mega-environment (northern Iran consisting of Rasht and Ghaemshahr), an average one (southern and central Iran consisting of Shiraz (SHI) and Esfahan (ESF)), and a major one (western and central Iran consisting of Arak (ARA), HAM, Karaj (KAR), Kermanshah (KER), Khoramabad (KHO), Varamin (VAR), and Sanandaj (SAN)). As a result, the findings from our study are as follows: (1) genotype G2 was the most stable and is thus recommended for commercial release in Iran; (2) the GGEbiplot method can be used to identify superior genotypes for target sites in Iran and sites in other parts of the world.