USE OF AMMI MODEL TO ANALYZE GENOTYPE- ENVIRONMENT INTERACTION FOR EGYPTIAN COTTON

The additive main effects and multiplicative interaction (AMMI) was applied to analyze the genotype-environment interaction for thirty (Gossypium barbadense L.) genotypes over five environments. The mean squares for environments (E), genotype (G), and GE interaction were highly significant for all traits. The mean square for first interaction prin­cipal component axis (IPCA 1) was highly singnificant for all traits. The second interaction principal component axis (IPCA 2) was highly signifi­cant for boll weight and lint yield. The third interaction principal compo­nent axis (IPCA 3) was highly significant for lint yield. Two strains (F5 733/91 and F8 899/91) showed high boll weight and minimum IPCA 1. Three strains (F5 691/91, F6 744/91 and G 75R6022) showed minimum IPCA 1 and IPCA 2. Those strains were high­ly stable for boll weight. For lint yield, two strains (F5 678/91 and F6 744/91) were highly productive with minimum IPCA 1. The most stable strains for lint yield were M5 491/91 and F5 795/91. Four strains (M5 507/91, F6 757/91, F7 891/91 and F7 893/91) showed the minimum IPCA 1 and high lint percentage. The most stable strains for lint percent­age were F5 689/91, F5 720/91, F7 893/91 and F8 901/91.