Additive main effects and multiplicative interaction analysis in lentil genotypes

Estimation of genotype by environment interaction (GE) is one of the main interests for a plant breeder. The statistical analysis، AMMI، is combination of ANOVA and PCA methods that calculate additive and multiplicative interactions. Nine advanced lines obtained from yield trials accompanied with Gachsaran cultivar (check) planted using a randomized complete block design replicated four times and 2 years (2002-2003) in 5 locations of Iran. Combined ANOVA showed that GE interaction was significant (at 1% level test). In this research AMMI methods used for analyzing of GE interaction. Three first IPCA were significant at 1% level test، next to model correction with Cornelious proposed method only IPCA1 was significant at 1% level test and IPCA2 was significant at 5% level test. ASV، SIPC1 and AMGE calculated، these are most important of AMMI parameters. Result showed genotype 9 (ILL 6199) had smallest ASV & SIPC1 (3. 98 and 1. 64 respectively) and highest mean yield (1265 kg. ha-1) was stable genotype in this research. Genotype no. 8 (ILL 6037) ranked as second in this research. Correlation among other stability statistics and AMMI parameter showed that ASV parameter had positive and high correlation with Wricke Ecovalence and Shukla stability variance (r = 0. 911**) and negative and high correlation with coefficient of determination (r = -0. 851**). SIPC1 had high correlation with regression coefficient (r = 0. 818**) and correlated with R2 (r = 0. 628*). AMGE parameter had no correlation with AMMI parameters and other stability statistics also we propose that in use of this parameter for yield stability has more attention.