Genotype x environment (GE) interactions are of interest to plant breeders because of their influence on progress from selection. This study was conducted to examine the effects of GE interactions for grain yield in three maturity groups of maize (
Zea mays
L.) cultivars adapted to the lowland tropics. Nine early (90-95 days), eight medium (105-110 days) and nine late (115-120 days) maturing maize cultivars were evaluated for grain yield potential in 32 to 36 environments across coastal savanna, forest, forest-savanna transition, and Guinea savanna zones of Ghana (Lat. 4 ° 44' - 11 ° 11' N, Long. 1° 11' E - 3 ° 11' W) from 1995 to 1998. Analyses of variance combined over locations and years within each maturity group indicated highly significant (P < 0.01) genotype x location x year interactions for yield in the three maturity groups. The genotype x year and genotype x location interactions were also significant in the intermediate and late maturity groups whereas only genotype x location interaction was significant in the early group. Spearman rank correlations, used to test the consistency of ranking of genotypes across locations, showed crossover interactions were important in all maturity groups and suggest the need to stratify the environment to minimize GE interactions. Sub-division of sites into variety testing zones corresponded to the major agro-ecologies in Ghana for the early and intermediate maturity groups but not the late group. The data also showed that testing using at least two replications per site at 10 locations in two years would be effective for identifying high-yielding genotypes across environments within each maturity group.