Analysis of variance (ANOVA)
The analysis of variance (ANOVA) revealed a wide range of variation and significant (P=0.001) differences among the 30 yard long bean accessions for all the characters studied (Table 1). It indicated that elucidating the presence of genetic variation for the respective traits exists in the yard long bean accessions under study. Hence, there is an opportunity for plant breeder to evaluate further breeding pursuit like yard long bean hybridization programme. Substantial variability in yard long bean accessions were previously reported by
Zakia Sultana et al., (2020) and
Mithun Kumar et al., (2021).
Estimates of mean and range
The descriptive statistics, including maximum and minimum quantitative traits mean value, standard error of the means and critical differences (1% and 5%) for all nine pod yield and yield contributing traits are summarized in (Table 2). The widest range was recorded for pod yield per plant (259.06-1361.05) followed by plant height (164.71-377.70), days to maturity (84.56-125.33), number of pods per plant (22.87-76.55), days to first flowering (37.74-52.16), pod length (24.38-59.31), number of seeds per pod (9.26-22.55), test weight (8.19-20.44) and pod girth (1.43-3.72) indicating the presence of considerable variation among the genotypes used in the present study. This would help in selecting the best genotypes from existing selection. The existence of high variability for different traits among yard long bean accessions had been earlier reported by
Rambabu et al., (2016).
Phenotypic and genotypic coefficient of variation
Genotypic and phenotypic coefficient of variation for pod yield and yield components are presented in (Table 2). The components used to measure the variability present in a population are GCV and PCV. The GCV provides a valid basis for appraising and comparing the range of genetic variability for quantitative traits. PCV estimates the augment of total variation. Expectedly phenotypic coefficient of variation (PCV) was higher than the genotypic coefficient of variation (GCV) in all the characters studied. Result from the present study in this context, records highest PCV and GCV ((45.07, 44.15) for pod yield per plant followed by number of pods per plant (34.85, 34.77), test weight (29.37, 29.28), pod girth (21.11, 20.96), number of seeds per pod (20.33, 20.20) and plant height (20.84, 20.26). PCV generally ranged between 7.44 for days to maturity to 45.07 for pod yield per plant, indicating that the presence of inherent variability among the accessions. Similarly, GCV range varied from 7.12 for days to maturity to 44.15 for pod yield per plant. The slight edge of PCV over GCV indicated good scope for making selection for the improvement of these characters. Similar findings of higher PCV than GCV were also reported by
Sivakumar et al., (2014).
Moderate estimates of PCV and GCV were recorded in present study for pod length. Similar findings were also obtained by
Haque et al., (2021).
Relatively least difference between the magnitude of PCV and GCV were observed for days to fifty per cent flowering and days to maturity indicating a lack of variability and limited scope for improvement through selection for these traits among the genotypes. Similar findings were earlier studied by
Yadea et al., (2011) for days to fifty per cent flowering and days to maturity.
Heritability and genetic advance
The heritability estimates assist in identifying genotypes from a broad genetic pool. Heritability values helps in predicting the expected progeny to be achieved through the selection process. In the present study, the estimates of broad sense heritability ranged from 88.65% for days to fifty percent flowering to 99.49% for number of pods per plant (Table 2). In present study tested genotypes showed high heritability for all the characters. Highest heritability estimates were observed for number of pods per plant (99.49%), test weight (99.39%), number of seeds per pod (98.76%), pod yield per plant (95.96%), pod girth (98.59%), pod length (98.47%), plant height (94.52%), days to maturity (91.55%) and days to fifty percent flowering ( 88.65%), which indicated that characters are least influenced by environmental factors and potency of these characters for transmission to next generation. In addition phenotypic selection would also be effective.
The heritability alone does not given the idea about expected gain in the subsequent generation. But still, it should be considered in connection with the genetic advance. The characters exhibiting highest value of genetic advance over per cent as mean (GAM) was obtained for pod yield per plant (89.11%) followed by number of pods per plant (71.44%), test weight (60.14%), pod girth (42.89%), number of seeds per pod (41.36%), plant height (40.58%) and pod length (37.60%) while other traits such as days to maturity (14.03%) and days to fifty percent flowering (12.25%), showed less than 20% of genetic advance mean (Table 3).
In the present investigation, high heritability coupled with high genetic advance as per cent of mean was recorded for the traits, pod yield per plant, number of pods per plant, test weight, pod girth, number of seeds per pod, plant height and pod length which indicated lesser environmental influence on these traits and predominance of additive gene action. Hence, simple selection may be practiced to improve these traits. This is in conformity with findings of
Manju Devi and Jayamani (2018) for plant height and test weight and single plant yield.
High heritability accompanied with moderate genetic advance per cent as mean recorded for days to fifty per cent flowering and days to maturity suggests that these traits are controlled by both additive and non-additive gene action. Similar kind of high heritability and moderate genetic advance was exerted for days to fifty per cent flowering and days to maturity by
Rambabu et al., (2016).
Clustering pattern
On the basis of relative magnitude of D
2 values, the 30 yard ling bean genotypes were grouped into seven clusters following Tocher’s method (Table 3). Cluster size varied from 1 to 12 genotypes. Cluster I was having maximum number of genotypes (12), followed by cluster II had 11 genotypes and cluster III consisted 3 genotypes. The remaining four clusters namely cluster IV, V, VI and VII comprised only one genotype each, as these could not be clubbed with any other genotypes, respectively. The group of clustering constellation in the present study culled out that eco- geographical diversity was not related to the genetic diversity. The clustering pattern reflected the presence of substantial amount of genetic divergence in the genetic accessions. The existence of genetic divergence in the yard long bean accessions were earlier reported by
Widyawan et al., (2020).
Cluster-distances
Based on the procedure given by
Singh and Choudhary (1979), the average intra and inter cluster distance values were computed and presented in (Table 4). The values of average intra cluster D
2 values ranged from 0.0 (cluster IV to VII) to 98.10 (cluster III) respectively. The highest value of intra cluster distance III (D
2 = 98.10) possessing 3 genotypes, followed by I ((D
2 = 94.59) consisted 12 genotypes and minimum distance was observed by cluster II (D
2 = 83.58). Data further revealed that there is good scope for selection for many traits of economic and agricultural importance within a cluster as indicated by the high magnitude of intra cluster distance among clusters. Hence, selection of divergent genotypes from the clusters I and III indicated that these cluster consisted genotypes which are highly diverse.
The minimum distance was noted between cluster IV and VII (D
2 = 113.23), indicating close relatedness among the genotypes included. The maximum inter cluster distance was found between cluster I and cluster VII (D
2 = 550.41) followed by the cluster I and IV (D
2 = 458.96) indicating wider genetic diversity among the genotypes between these clusters. To exploit genetic diversity through hybridization program inter-cluster distance must be taken into consideration. Hence, hybridization between the genotypes from these clusters would result in maximum hybrid vigour and high number of useful heterotic hybrids. Similar findings were reported by
Valarmathi et al., (2007).
Cluster means performance
Considerable differences in cluster mean values were evident for all the characters studied (Table 5). In the present study, dwarf plant (183.02) earliest flowering (41.44) and earlier maturity (101.55), number of seeds per pod (21.89) were observed in cluster IV and VI respectively. Cluster VI ranked first and recorded highest pod yield per plant (962.15), test weight (21.11), pod length (51.03) while this also produced minimum pod girth (3.01). Cluster V exhibited maximum number of pods per plant (75.05).
The genotypes suggested for use as parents for hybridization based on their mean value and genetic divergence for traits are listed in (Table 6). The genotypes with maximum mean value are used as parental material in progressive breeding. Based on the genetic distance values, cluster constellation, the most divergent genotypes were from clusters I, IV, V and VII which could be involved in a crop improvement programme. Genotype IC- 630381 present in clusterVII has highest pod yield per plant and can be used in future breeding programmes for developing variety with high pod yield content. Genotype IC-626143 Cluster (I), IC-630379 (cluster V) and IC-630383 (Cluster IV) might be used in breeding for improved early varieties of yard long beans. These findings were supported by the results of
Ullah et al., (2011).
Character’s contribution to diversity
The ranking technique was adopted to rank the characters in the order of the contribution to total genetic divergence. The contribution towards the total genetic divergence was represented in (Table 7). The relative contribution of individual character towards the expression of genetic diversity studied, revealed maximum contributions of 68.27 per cent by pod length followed by test weight (13.10 per cent), days to maturity (7.81 per cent) and pod yield per plant (3.94 per cent). Remaining characters had very less contribution to genetic divergence. Similar studies were conducted by
Sulanthi et al., (2007); Hossain et al., (2013).