Analysis of variance revealed that cluster bean genotypes under study exhibited significant differences for all the characters studied
viz., plant height, number of branches, days to 50% flowering, days to first fruit set, days to first harvest, number of pod clusters/plant, number of pods/cluster, number of pods/plant, pod length, pod girth, pod weight, number of seeds/pod, pod yield/plant. Results indicated that consider variability exist for all the characters studied and sufficient improvement could be achieved in most of these traits.
Estimate of genotypic and phenotypic variance are of greater use in determining extend of variability present within the breeding material. The estimate of range, mean components of variance, heritability and genetic advance for growth, earliness and yield parameters in cluster bean is given in Table 1. In the present study, high estimates of Phenotypic coefficient of variation (PCV) was recorded for number of branches (40.35), number of pod clusters/plant (26.90), number of pods/plant (36.37), number of pods/cluster (42.37), pod weight (23.23) and pod yield/plant (28.41) which indicated presence high variability for these traits.
As the estimate of phenotypic variability cannot differentiate between environmental and genetic effects, study of genetic variability is required to partition the real genetic difference from environmental effects. Higher the genotypic coefficient of variation (GCV) more the chances of improvement in that character. In the present study, high estimates of GCV (>20%) was recorded for number of branches (39.90), number of pod clusters/plant (26.21), number of pods/plant (36.17), number of pods/cluster (41.64), pod weight (23.21) and pod yield/plant (28.11). This indicated the presence of high amount of variability in the genetic stock studied. Hence, selection could be employed for the improvement of these traits. This is in confirmation with the results obtained by
Anandhi and Ommen (2007); Vir and Singh (2015); Kumar et al., (2015); Patil et al., (2016); Rishitha et al., (2019); Kgasudi et al., (2019).
Moderate estimate of PCV and GCV (11-20%) were observed for plant height (16.74), pod length (15.96), pod girth (14.04). This indicated presence of moderate amount of variability.
Malaghan et al., (2012) and
Girish et al., (2013) also observed moderate GCV for plant height, pod length and girth in cluster bean. Low GCV(<11%) were observed for days to 50% flowering (8.41), days to first fruit set (3.23), days to first harvest (2.90) and number of seeds/pod (7.76). This indicated the narrow genetic base for these traits in the collected accessions and there is limited scope for improvement of these traits through selection. These results are in accordance with
Hanchinamani (2004); Girish et al., (2013); Gowd et al., (2019).
Heritability estimates is the percentage of phenotypic variance that is attributed to genetic variance. High heritability of a trait implies that improvement is possible through selection. Whereas, characters with low heritability shows less response to selection. In the present study, very high broad sense heritability (>80%) was observed for plant height (97.59%), number of branches (97.81%), number of pod clusters/plant (94.90%), number of pods/cluster (96.55%), number of pods/plant (98.91%), pod length (99.28%), pod girth (87.92%), pod weight (99.87%) and pod yield/plant (97.87%). Traits like days to 50% flowering (52.47%), day to first harvest (39.63%) had moderate heritability whereas, days to first fruit set had very low heritability (10.56%). Similar results were reported by
Anandhi and Ommen (2007); Muthuselvi and Shanthi (2013); Patil et al., (2016); Santhosha et al., (2017); Rishitha et al., (2019). Heritability estimates alone is not enough to provide information about genetic progress that could be made through selection. Hence, heritability estimates along with genetic advance is more useful in predicting the response to selection.
Considering heritability in broad sense along with genetic advance revealed the prevalence of addictive or non- addictive genetic variance and thus helps in judging the effectiveness of selection more accurately. Genetic advance as percent mean (GAM) for the traits varied from 2.17% (days to first fruit set) to 84.27% (number of pods/cluster). In the present study, high heritability coupled with high genetic advance was predicted for plant height, number of branches, number of pod clusters/plant, number of pods/cluster, number of pods/plant, pod length, pod weight and pod yield/plant. Indicating that these characters may be controlled by addictive gene actions and are likely to respond better to selection. The findings of
Muthuselvi and Shanthi (2013); Patil et al., (2016); Kumar et al., (2015); Boghara et al., (2016); Santhosha et al., (2017); Kgasudi et al., (2019) in cluster bean were in same line.
High heritability coupled with moderate to low GAM were observed for number of seeds/pod and days to 50% flowering. This indicated the prevalence of non-addictive gene actions. Hence, there will be less response to selection.
Gowd et al., (2019) also reported similar results for number of seeds/pod in cluster bean. Moderate to low heritability and GAM was observed for days to first fruit set and days to first harvest. These traits also had high PCV compared to GCV. This indicated the influence of environment on these traits and hence, selection would be ineffective. They could be exploited through manifestation of dominance and epistatic components through hybridization followed by selection. These results are in line with the findings of
Rai et al., (2012) in cluster bean,
Islam et al., (2011) in dolichos bean and
Kamaluddin and Ahmed (2011) and
Jhanavi et al., (2018) in french bean.