The main objective of any plant breeding programme is to produce high yielding varieties suitable for the cultivation area and breeding strategy like selection can only be imposed when there is ample availability of desirable genetic variability
(Chandra et al., 2019b) and variability can be revealed through analysis of variance (ANOVA). Therefore, total variability was divided into different components presented in Table 2 for yield and its contributing characters of twelve mung bean genotypes along with four checks evaluated at Transitional Plain of Inland Drainage zone of Rajasthan. Results from individual ANOVA analysis for
kharif 2018 and 2019 revealed that variability in mung bean germplasm found to be significant for all the characters under study. However, pooled analysis of both season exhibited variation only in pod length, test weight and seed yield. Whereas interaction effect of year × treatment showed that days to 50% flowering, days to maturity, plant height, number of seeds/pod and test weight were only significant. Similar reports have been obtained from
Belay et al., (2019) stating that analysis of variance for six mung bean varieties exhibited differences in varietal characters
viz. days to 50% flowering, plant height, number of seeds per pod, seed yield and thousand seed weight (p≤0.05) except number of pods per plant.
The significant differences revealed that there is ample genetic variability among mung bean germplasm. After exploitation of genetic variability in available germplasm, it is better to separate heritable portion from the non-heritable part and to plan the accurate breeding programme and this is the opportunity to a breeder that now he/she can improve characters with high heritability
via selection
(Parida et al., 2018).
Piepho et al., (2008) explains that BLUP has good predictive accuracy in plant breeding experiments. Table 3 presents the summary statistics, heritability (broad sense) and BLUPs values of mung bean germplasm for yield and its ancillary characters. Where it clearly depicts that the seed yield drastically decreased
i.e. 41.23% in 2019 compared to 2018
kharif season and days to maturity also revealed that the genotypes responded early in maturity in 2019 since there is 17.98% reduction in maturity days. Similarly other traits under study also revealed percent reduction in 2019
viz. test weight (8.85%), number of seed/pod (8.15%) and plant height (3.63%) whereas delayed days to 50% flowering (1.46%) and increased pod length (1.94%) have been observed. The reduction in yield in
kharif, 2019 might be due to high rainfall coinciding with maturity stage of mung bean during 39
th meteorological week (24
th September- 30
th September) and early crop maturity whereas no rainfall coincides with maturity in 2018 as depicted from Fig 7 and Fig 8.
Johnson et al., (1955) has classified heritability as low (below 30%); medium (30-60%) and high (above 60%). In both the seasons only days to 50% flowering and test weight shown high heritability (>60%). This implies that expected gain from selection will be high if these traits are to be used as selection criteria in mung bean crop improvement
(Degefa et al., 2014).
The data collected from present study were also used for estimating the variability, normality and character association for yield and ancillary characters of mung bean germplasm and presented in Fig 3, where distribution of each variable has been depicted diagonally. The bivariate scatter plots with a fitted line are given below the diagonal and correlation values above the diagonal. All the traits under study followed normal distribution. The x-axis represents column variable and y-axis represent row variable in each scatter plot. The correlation values under scatter plots were also supported using correlograms among yield and ancillary characters of mung bean genotypes (Fig 4). In 2018, seed yield was positively correlated with days to 50% flowering, days to maturity and number of seeds/pod, but negatively correlated with plant height. Similar findings reported from
Makeen et al., (2007). However, in 2019 the similar pattern was not observed and seed yield was found to be negatively correlated with days to 50% flowering and number of seeds/pod, while positive correlation with plant height and pod length. Observing a negative correlation of seed yield with number of seeds/pod is not in agreement with other reports and this negative correlation might be due to shriveled and poor pod filling due to high humidity by rainfall at maturity stage of mung been (Fig 8).
Further to understand the significant difference between genotypes under study, Duncan’s new multiple range test of mung bean germplasm for yield and its ancillary characters were studied and the results are presented in Table 4. Genotypes RMG 1098 and RMG 1132 were jointly ranked first for seed yield with the value of 1129.67 kg/ha during
kharif 2018. However, genotypes RMG 1087, RMG 1094, RMG 1134, RMG 1137, RMG 1138, RMG 1139, RMG 1147 and RMG 492 were also statistically on par. In
kharif 2019, genotype RMG 1147 ranked first for seed yield. Lowest yield was obtained from MSJ 118 in 2018 and IPM 02-3 in 2019. Therefore, during both the year, genotype RMG 1147 was found to be suitable and outperformed even in condition of rainfall coinciding with maturity (
kharif 2019). The seed yield of genotypes RMG 1098, RMG 1132, RMG 1134, RMG 1139 and RMG 1147 were superior over the zonal and state checks.
Fig 5 revealed box plot description of mung bean yield and its ancillary characteristics
. Williamson et al., (1989) states that box plot convey the level, spread and symmetry of distribution of data. Fig 5(a) for days to 50% flowering depicts that a wide range has been observed in 2019 compare to 2018 and median value revealed delayed flowering in 2019 than 2018. This might be due to more rainfall in 2019
i.e. 414 mm total rainfall and 33 rainy days compare to 281.5 mm rainfall and 27 rainy days in 2018 which tends to extend of vegetative growth and lead to delay flowering (Fig 7 and Fig 8).
All the studied genotypes showed delayed maturity during 2018 as compared to 2019 (Fig 5b). Thus, during 2019, early maturity might be due to high humidity at critical stage tends to invite the infestation of disease and pest and plant forced to complete their life cycle under stress condition
(Oplinger et al., 1990; Nair et al., 2019). Genotypes showed more number of seeds/pod during 2018 (Fig 5c). Hence, observed seed yield was less during 2019 than 2018 (Fig 5g). A higher mean pod length was shown by horizontal line in box plot for 2019 compare to 2018 which can be revealed from Fig 5(e).
It is inevitable that pod length, number of seeds/pod and seed yield are interrelated characters. High humidity (Fig 7 and Fig 8) at critical plant stages may attract disease and pest. In 2019, genotypic seed yield was less compare than 2018 due to infestation of fungal even though there was higher pod length. It tends to incomplete pod filling and less number of seeds per pod and resulted in yield penalty. Apart from this, lack of sunlight facilitated vegetation to grow between the sown seeds that used most of the nutrients. Medium value shows that there was no much difference with reference to plant height as depicted from Fig 5(d). However, a wide range has been observed in 2018 compared to 2019. A higher test weight has been observed in 2018 compared to 2019 as depicted in Fig 5(f). Therefore, it clearly depicts that the grain formed in 2019 were shriveled and small.
Kumar et al., (2013) also reported that heavy rain leads to yield loss in green gram.
Clustered heat map of yield and ancillary characters of mung bean genotypes during 2018 and 2019 has been presented in Fig 6. Heat map is a two way display of data matrix in which the individual cells are displayed as coloured rectangles. The colour of a cell is proportional to its position along a colour gradient. Column variable of the matrix represents as the column of heat map and row of matrix as rows of the heat map. The order of rows is determined by performing hierarchical cluster analysis of the row tends to position similar rows together on the plot. The order of the column is determined similarly. Heat map clustered seed yield in one cluster and remaining all the characters in other clusters. Looking at the dendrogram we can see the two clusters for genotypes during both the seasons which explain that all the experimental material can be classified into two major groups. However, again if we look into the sub clustering then complexity of classification will be increased.
