Significance of variances and mean performance
The analysis of variance for all the traits indicated highly significant variation among the genotypes (Table 2) which clearly shows that there is presence of sufficient variability in the germplasm. The presence of variability in the genotypes offers an opportunity in improvement of yield and its contributing traits through selection in Faba bean. Likewise, highly significant variations for all the yield related traits were also reported by
Alghamdi (2007).
The mean performances obtained for different traits are presented in Table 3. Two genotypes (FLIP15-183FB, FLIP15-196FB) have taken only 47.50 days for fifty percent flowering. While the genotype FLIP15-195FB took 66.50 days for 50% flowering. FLIP15-183FB took 63.50 days to 50% pod setting while first pod picking was done in (75.00 days) which was the earliest from all the other genotypes. The genotype Giza 4 was observed to be late which took 98.00 days for first pod picking. The tallest plants were observed in FLIP15-206FB (109.80 cm) and the shortest ones in FLIP15-195FB (72.10 cm). Regarding pod size, longest pod was recorded in FLIP15-183FB (8.65 cm) and widest pod in FLIP15-201FB (1.73 cm). The maximum pod weight was obtained in FLIP15-201FB (10.15 g) and the least for FLIP15-181FB (4.83 g). The least number of pods per plant was recorded for FLIP15-205FB (7.30) and the most for FLIP15-196FB (26.20). The maximum number of seeds per pod were obtained for FLIP15-199FB (4.05) followed by FLIP15-180FB (3.90) and FLIP15-183FB (3.80). The lowest hundred seed weight was observed for FLIP15-195FB (44.38 g) and the highest for FLIP15-198FB (90.32 g). The highest pod yield per plant was recorded in FLIP15-196FB (201.83 g) followed by FLIP15-201FB (151.77 g) and FLIP15-203FB (128.49 g). However, the lowest pod yield per plant was observed in genotype FLIP15-181FB (38.30 g). Faba bean plants vary widely in expression of yield and its contributing traits along with quality parameters that offers opportunities in Faba bean improvement, which were also reported by
Kalia and Pathania (2007);
Ammar et al., (2015).
Parameters of variability
The estimates of phenotypic and genotypic coefficients of variability, heritability (broad sense), genetic advance and genetic gain were statistically worked out to determine the response of selection for different traits, which are presented in Table 2. The perusal of the data showed that phenotypic coefficients of variability (PCV) were higher in magnitude than genotypic coefficients of variability (GCV) for all the characters. The results are in concurrence with Aziz and Osman (2015). Phenotypic coefficient of variation was the highest or pod yield per plant (PCV=55.87%) and lowest for pod width (PCV=8.46%). However, genotypic coefficient of variation was highest for pod yield per plant (GCV=53.90%) followed by number of pods per plant (GCV=40.32%) and pod weight (GCV=21.06%). Least genotypic variance was observed in days to first pod picking (GCV=6.28%). The highest genotypic and phenotypic coefficient of variation was observed for pod yield per plant followed by number of pods per plant and pod weight. It indicates selection can be applied for these traits to identify promising lines. The results are in accordance with the reports made by earlier workers (
Saghin, 2002;
Kalia, 2003;
Pekson, 2007). The differences between PCV and GCV were narrow that signified the importance of genetic variance in the inheritance of the characters
(Kumar et al., 2017).
Heritability (broad sense) determines to which degree of difference among phenotypes results from genetic causes. Heritability estimates ranged from 40.01% (Pod length) to 93.07% (Pod yield per plant). Pod length had lowest heritability estimates and genetic advance as percent of mean (40.01% and 8.50%). Maximum heritability was recorded for pod yield per plant (93.07%) followed by number of pods per plant (88.43%) and 100 dry seeds weight (86.07%).The broad sense heritability was highest for all the traits recorded. Pod yield per plant followed by 100 dry seeds weight and plant height registered high genetic advance. The characters that exhibit high heritability coupled with high genetic advance could be useful in the selection process as they are controlled by additive genes and less influenced by environment (
Panse and Sukhatme, 1985). Low to high heritability were also reported by
Kalia and Sood (2004),
Alghamdi (2007),
Mulualem et al., (2013). Some of the characters with high heritability and low genetic advance was observed in this study
viz., pod length (40.01%, 0.63), number of seeds per pod (41.18%, 0.29), pod width (56.59%, 0.14), number of primary branches (57.44%, 2.06) and pod weight (79.93%, 2.87). This may be accredited to non-additive gene effects and these characters could be improved through hybridisation. High heritability with low genetic advance for different characters was also reported by
Kalia and Pathania (2007).
Correlation studies
The estimates of phenotypic and genotypic correlation coefficients among characters were determined (Table 4). Genotypic correlation coefficients values were higher in magnitude than phenotypic correlation coefficients which may be due to reduction and modification in the inherent associations between the different characters under the influence of the environment.
Katiyar and Singh (1990) and
Singh et al., (2017) also reported that the genotypic correlation coefficient was higher than the phenotypic correlation values for all characters under studied and indicated that genotypic correlation is more dependable than phenotypic correlation and helps to identify the characters to be exploited in breeding activities.
The phenotypic and genotypic correlation coefficients among different characters pointed out that days to 50% flowering was significantly and positively associated with days to 50% pod setting, days to first pod picking and number of primary branches. However, days to 50% flowering was significantly and negatively correlated with plant height, pod length, pod weight, 100 dry seeds weight and pod yield per plant at genotypic level. This result showed that the early maturing genotypes have less yield potential than those of late maturing ones. These findings are in concurrence with
Alghamdi and Ali (2004);
Alghamdi (2007);
Tofiq et al., (2016).
Pod yield per plant was significantly and positively associated with plant height, number of primary branches, pod length, pod width, pod weight and number of pods per plant at genotypic level. At phenotypic level, pod yield per plant was significantly and positively correlated with pod width, pod weight and number of pods per plant only. Number of pods per plant was significant and positively correlated with number of primary branches, pod width and pod yield per plant at the genotypic level. Pod weight was significantly and positively correlated with plant height, pod length, pod width and pod yield per plant at the phenotypic and genotypic levels. These results indicate that selection for number of pods, pod length, pod width, pod weight and seed weight would ensures higher yield as these traits are significantly correlated with pod yield per plant. Significant positive correlations of yield per plant with other horticultural traits in Faba Bean were also reported by
Alghamdi (2007);
Tofiq et al., (2016). Pod yield per plant, pod weight, 100 dry seeds weight and number of primary branches significantly and negatively associated with days to 50% flowering, days to 50% pod setting and days to first pod picking. Some of the yield components exhibited negative associations with others that might have been due to competition between these components during their development (
Adams, 1967).
Path coefficient analysis
Path coefficient analysis at genotypic level used to study the effect of various traits on pod yield per plant. The perusals of phenotypic path coefficient analysis (direct and indirect effects) are presented in Table 5. The direct effect of days to 50% pod setting (-1.034), pod width (-0.118) and 100 dry seeds weight (-0.025) were negative on pod yield per plant which means these traits contributed for its negative correlation with pod yield per plant.
Tofiq et al., (2016) also reported negative direct effect of100 dry seeds weight on yield. Number of pods per plant showed the highest positive direct effect (0.845) followed by days to first pod picking (0.713), pod weight (0.489), days to 50% flowering (0.355), plant height (0.051), pod length (0.047)number of primary branches (0.035) and number of seeds per pod (0.015) on pod yield per plant.
Mridula et al., (1992), Ulukan et al., (2003) and
Tadesse et al., (2011) also observed the highest positive direct effect of number of pods per planttogether with plant height and number of seeds per pod on yield.
Kumar et al., (2017) reported that the number of branches per plant, number of pods per plant, number of seeds per pod and pod length exhibited positive direct effect on yield.
Number of pods per plant also showed high positive indirect effect via number of primary branches and pod width.
Kumar et al., (2017) also reported the positive indirect effects of number of pods per plant via branches per plant on yield.
Singh et al., (2017) observed that the positive indirect effects of days to 50% flowering via branches per plant and days to maturity on yield. These results suggest the importance of these traits in selection programme for the improvement of pod yield. The low magnitude of residual effect (0.006) indicated that the traits included in the study accounted for most of the variability present in the dependent variable.