Abstract

---

32 faba bean genotypes were tested at Hashenge and Aiba in 2015 to evaluate the variability of large seeded faba bean genotypes for their yield and yield components in water logging areas. Alpha lattice design with two replications was used for the experiment. ANOVA has showed high statistical significant difference (P<0.01) for thousand seed weight over locations. Based on this, genotype EH 06007-2 scored highest 1000 seed weight (1111gm) followed by EH 06088-6 (971.5) and EH 06007-4 (938.5 gm) over locations. But least seed weight was obtained from the local genotype (505.5 gm) followed by the standard check Walki (587.3) and ET 07017-bulk (648.0 gm) genotypes. Moreover, grain yield has also showed significant difference (p<0.05) at Aiba location but no significant difference in the other location (Hashenge). At Aiba, genotype ET 07013-1 gave the highest grain yield (59.31 qt/ha) followed by genotype ET 07005-1 (57.85) and EH 06088-1 (54.77 qt/ha). Significant positive correlations were also recorded TSW with DM (0.624), number of seeds per pod (0.567) and NTPP (0.427) but it was negatively correlated with NPPP (-0.487). Grain yield was significantly and positively correlated with plant height (0.49) and NPPP (0.369). The highest distance (0.692) was between cluster V and IV, which suggested that the members of these clusters diverge on most of the studied traits and could be used in breeding programs. Principal component (PC) analysis revealed that the first four PCs having eigenvalues >1 explained 83.7% of the total variation. The variance explained by PC1 was mostly due to traits related to DM, TSW, NSPP and NTPP, whereas PC II was mostly related to grain yield, plant height, NPPP and thousand seed weight traits. In this experiment, the PC analysis ultimately showed the amount of variability for the traits that could be used for the improvement of large seed sized faba bean genotypes

---