Evaluation stability of seed yield of faba bean (Vicia faba L.) genotypes by parametric and non-parametric methods

Faba bean (Vicia faba L.) is one of the most important legumes for its seed high protein content and nutritional value, and is still considered to have great potential as rainfed crop. Faba bean is a very valuable legume crop that contributes to the sustainability of cropping systems through its ability of biological N2 fixation, diversification of cropping systems leading to decreased disease, pest and weed. The decrease in area under cultivation in Iran, mainly because of high yield instability. This can be attributed to biotic and abiotic stresses Faba bean yield is strongly influenced by environments and breeders often determine the stability of high yield genotypes across environments before recommending a stable cultivar for release. The purpose of this study is the distinguishing the superior genotypes in terms of yield and yield stability in 15 selected faba bean lines from preliminary yield test.

15 selected lines obtained from crosses between ICARDA lines and Iranian improved and landrace varieties and resulted from preliminary yield test of 2013-2014 cropping season, along with control cultivar Barekat, Saraziri, Zereshki and Baluchi were evaluated in a Gorgan, Dezful, Borojerd and Iranshar at two cropping seasons (2015-2017) in a randomized complete block design with three replications. Stability analysis were performed with Eberhart-Russell analysis of variance, Shukla's variance (2i), Wrick equivalence (Wi), environmental variance (S2i), coefficient of environment variation (CVi), coefficient of determination (R2), Finlay and Wilkinson regression coefficient (bi), Lin and Binns superiority index (Pi), yield stability index (YSi) and nonparametric methods such as , , Z1, Z2 and TOP and mean and standard deviation of rank.

Combined analysis of variance was performed after Bartlett test and not significant of it and assurance of uniformity of experimental errors. Combined analysis of variance indicated a significant effect of genotype, environment and genotype × environment interaction on seed yield. Based on Wrick equivalence, Shukla variance and RMSE indices, genotypes 5, 13, 14 and 15 were more stable. Genotype 9 had lowest environmental variance and coefficient of environmental variation and was stable than the other genotypes. genotypes 9 and 11 had the highest R2and the lowest deviation from regression. The coefficient of regression in genotypes 6, 9 and 16 were lower than one and their seed yield were higher than total average seed yield and therefore were the stable genotypes. Genotypes were stable genotypes. According to the priority index in all of environments and favourable environments, genotypes 11, 12, 14 and 15 were adapted to the unpredictable high yielding environments(1, 3, 5 and 7), while 6, 9, 14, 16 and 12 were adapted to low yielding environments(2, 4, 6 and 8).Simultaneous analysis based on seed yield and stability (YSi) indicated genotypes 6, 7, 9, 11, 12, 13, 14, 15 and 16 were as stable genotypes. Based on Top statistic, genotypes 12 and 14 and based on and , genotype 11 were stable than the other genotypes. Results of Selection index of ideal genotype (SIIG) method, Based on all non-parametric methods, showed that 14 and 15 genotypes whit 3478 and 3439 kgha-1 grain yield were the most stable genotypes. ConclusionBased on all indices, genotypes 9 and 11 were the most stable genotype in terms of seed yield and could be considered a good breeding material stock in any future breeding program.