Genetic Diversity and Association Studies in Greengram [Vigna Radiata (L.) Wilczek]

Green gram [Vigna radiata (L.) Wilczek] is the third important pulse crop grown in India after chickpea and pigeonpea. It belongs to the subgenus Ceratotropis in the genus Vigna. Vigna radiata is a self-pollinating diploid grain legume (2n = 22) and it is a major source of dietary protein for the predominantly vegetarian population of India. Greengram also called mungbean was originated in India. Greengram is one of the important pulse crops because of its adaptation to short growth duration, low water requirement, soil fertility, easy digestibility and low production of flatulence (Shil and Bandopadhya, 2007). Average protein content in the seed is around 24.0 per cent. The protein is comparatively rich in lysine, which is predominantly deficient in cereal grains (Baskaran et al., 2009). India is the largest producer of greengram with more than 50% of the world’s production in an area of 43.0 lakh hectares with an annual production of 20.7 lakh tons with an average productivity of 481 kg per hectare (Annual Report, DPD 2016-17).
       
To meet out the ever increasing demand for greengram, there is a need to break the yield barriers by developing high yield and biotic/abiotic resistant varieties. Yield is a complex character which is influenced by many dependent characters. The dependent characters which are associated with yield component traits are themselves interrelated. Selection based on yield may not be rewarding, therefore improvement in yield is possible through selecting yield component traits which show close association with yield. Correlation and path analysis provide the information on the extent of association between yield and yield components and bring out the direct and indirect cause of association and thus give a clear understanding of their association with yield. Cluster analysis and principal component analysis (PCA) help in identification of genotypes from distant clusters and selection of important traits which contribute towards the total variation. Identified genetically divergent parents when used in crossing program are expected to throw superior and desirable segregants in the segregating generations. Hence, the present study was carried out to assess genetic diversity and the factors determining the seed yield in greengram through correlation coefficient and path analysis.

The present investigation was carried out with 44 genotypes of greengram collected from Indian Institute of Pulses Research (IIPR), Kanpur, National Pulses Research Centre (NPRC), Vamban and Rice Research Station (RRS), Tirur were raised at Rice Research Station, Tirur during Rabi, 2012 which is located at about 13°N latitude and 79°E longitude at an altitude of 40 meters above MSL. The genotypes were evaluated in a randomized block design with two replications. Each genotype was accommodated in a single row of three meter length with a spacing of 30 cm between rows and 10 cm between plants. The recommended agronomical and plant protection practices were followed to maintain healthy stand of the plant.
       
Observation were recorded on five randomly selected plants per replication for ten agro-morphological traits viz., days to 50% flowering (DFF), days to maturity (DTM), plant height (PH), number of branches per plant (NBP), number of pod clusters per plant (NPC), pod length (PL), number of pods per plant (NPP), number of seeds per pod (NSP), 100 seed weight (HSW) and seed yield per plant (SYP).
       
The mean data were subjected to the following statistical analysis. Descriptive statistics like mean, maximum minimum, SD, CV were obtained using MS Excel. Biometrical methods were followed to estimate correlation and path coefficient analysis (Singh and Chaudhary, 1979). Principal component analysis was worked out through correlation matrix using PAST 3. Dissimilarity between the genotypes was calculated based on Euclidean distance and the genetic divergence was computed using Wards D minimum method of clustering through R 3.5.1 statistical package.

The results of basic descriptive statistics viz., mean maximum, minimum, standard deviation (SD) and coefficient of variation (CV) for the ten quantitative traits studied in greengram genotypes are presented in Table 1.
 

       
The genotype CO-5 had taken the maximum number of days (44 days) to attain 50% flowering, while the genotype MH-318 reached 50% flowering within a short span of 27 days. In case of maturity the genotype MH-521 matured early (59 days), while the genotype CO-5 took 74 days to complete the maturity. Plant height ranged from 32.25 to 95.00 cm. The number of branches per plant ranged from 1.52 to 6.00, TM-11-34 recording the maximum number of branches and Pusa 0871 recording less number of branches. The genotype EC-591338 produced less number of pod clusters, while CO-7 produced more number of pod clusters per plant. On an average the pod length among the genotypes was 8.04 cm. The number of pods per plant ranged from 15.30 (EC-591338) to 54.23 (EC-496839). The mean 100 SW among the genotypes is 3.85 g. With respect to seed yield per plant the genotype CO-7 produced higher seed yield of 13.21g, while the genotype EC-591338 produced 4.71g with an average of 9.30 g among the genotypes studied.
       
Among the ten traits studied, the largest variation was observed for number of pods per plant with CV amounting to 24.04 per cent, followed by seed yield per plant (23.23), plant height (18.37) and number of pod clusters per plant (13.72). Least coefficient of variation of 3.11 was observed for pod length. 
 
Correlation coefficient analysis
 
Seed yield is a complex trait and it is very difficult to improve by selecting the genotypes for yield. Therefore, identifying the characters, which are closely related and contributed to yield, becomes highly essential. Correlation coefficient is a statistical measure, which is used to find out the degree of relationship between two or more variables. In plant breeding, correlation coefficient measures the mutual relationship between various plant characters and determines the component characters on which selection can be relied upon for genetic improvement of yield. The results revealed that the genotypic correlation coefficients were higher than the phenotypic correlation indicating the preponderance of genetic variance in expression of traits. In the present study, number of pods per plant, number of pod clusters per plant, number of seeds per pod and number of branches per plant showed as significant and positive association with seed yield per plant (Table 2). The positive association of pods per plant, pod clusters per plant, seeds per pod and branches per plant with seed yield per plant has been reported by Raje and Rao (2000) and Singh et al., (2018). Reddy et al., (2011) and Hemavathy et al., (2015) reported positive association of pods per plant and seeds per pod with seed yield per plant. On contrary negative association of number of pod clusters per with seed yield was reported by Tabasum et al., (2010). Selection of parents based on these traits can help in yield improvement in greengram.
 

       
Correlation among the component traits revealed that days to 50% flowering and days to maturity had positive significant correlation between each other. Similar finding were reported by Hemavathy et al., (2015). Number of branches per plant had positive significant association with number of pods clusters plant and number of pods per plant. With the Increase in number of branches per plant the number of pod clusters per plant also increases, this resulted in the production of more number of flowers and pods per plant. Number of pod clusters per plant had significant positive association with number of pods per plant and number of seeds per pod and it also had non-significant association with 100 seed weight. This indicated that number of pod clusters increases the number of pods per plant and seeds per pod resulting in less seed weight. Similar observations have been reported by Raje and Rao (2000).


 
Path analysis
 
Information obtained from correlation study does not give a complete idea about the contributions of each component character. Path coefficient analysis is useful for partially direct and indirect causes of correlation and also enables us to compare the causal factors on the basis of their relative contributions. As the correlation coefficient is not sufficient to explain true relationship for an effective manipulation of the character, path coefficient was worked out. In the present study, Path analysis showed that the maximum positive direct effect contributing to single plant yield was exhibited by days to 50% flowering followed by number of pods per plant (0.480), number of seeds per pod (0.409), number of pod clusters per plant (0.212) which implies that selection for these traits would improve seed yield per plant (Table 3). These results were in correspondence with the findings of Srivastava and Singh (2012) and Prasad and Prasad (2013). Genotypes with short duration, more number of pod clusters and pods per with more number of seeds and determinate growth habit are the ideal plant types which resulted in high seed yield per plant.
 

       
Among the component traits studied, high indirect effect on seed yield was attributed by days to maturity via days to 50% flowering (0.454) and number of pod clusters via number of pods per plant (0.218). Such an observation was reported by Biradar et al., (2007). The residual effect of path analysis was moderate (0.3995), which shows that some more traits may be included in the study to see the pattern of interaction on yield. From the path analysis traits viz., days to 50% flowering, number of pods per plant and number of seeds per pod showed maximum direct effects on seed yield per plant. Among these, number of pods per plant and number of seeds per pod exhibited highly significant and positive association with seed yield. Therefore, to increase yield in greengram the emphasis should be given to the selection of these traits.
 
Principal component analysis (PCA)
 
Principal component analysis (PCA) was applied as a dimensionality-reduction tool for the multivariate data to analyze the structure of the genetic diversity in the experimental material. In the present study it revealed that first five principal components in the PCA contributed to a maximum of 79.12% of the total phenotypic diversity among the 44 genotypes (Table 4 and Fig 2). The first principal component (PC1) with an eigen value of 2.68 explained 26.77% of the total variation. PC1 was associated mainly with seed yield per plant, number of pod clusters per plant, number of pods per plant and number of branches per plant as well as number of seeds per pod. The second principal component (PC2) accounted for 21.47% of the total variation and was mainly related to days to 50% flowering and maturity. PC3 accounted for 11.63% of the total variation and was characterized by 100 seed weight. The PC4 and PC5 explained about 10.62 and 8.63 per cent of total variation and were contributed by pod length and number of seeds per pod. The first two PCs which contributed to 48.24% of the total variance were plotted graphically to demonstrate the relationships between accessions (Fig 3).  It can be inferred that there exists wide genetic variability among the genotypes based on the distribution pattern of the genotypes on the biplot. Mehandi et al., (2015) studied twenty one greengram genotypes and reported that PC1 was positively contributed by short plant height and number of seeds per pod. They also reported that the major contribution of PC2 through number of pods per plant, 100 seed weight, pod length, seeds per pod and number of pod clusters per plant whereas PC3 was attributed by number of pod clusters per plant.
 

 

 

 
Cluster analysis
 
Cluster analysis was used to determine the genetic relationship among the genotypes and find out the suitable genotypes for future breeding programme. Cluster analysis based on ten quantitative traits for 44 genotypes grouped them into six discrete and well defined clusters. Cluster III was the largest consisting of 12 genotypes (Table 5 and Fig 4) whereas, the smallest was cluster VI with four genotypes. It is noted that genotypes from of different geographical origin were grouped in the same cluster indicating the absence of relationship between genetic diversity and geographical diversity.
 

 

 
Presence of variability in the 44 genotypes was also reflected in the cluster means for the ten traits evaluated (Table 6). Cluster I comprised of seven genotypes characterized by genotypes with more number of pods per plant. Cluster II comprised of five genotypes with more branches and more pod clusters resulted in high yielding genotypes (11.92g). Cluster III comprised of 12 genotypes predominantly early maturing genotypes with small pods. Cluster IV comprised of five genotypes which were characterized by small seeds having low seed weight (2.75g) and less number of seeds per pod (7.49). Cluster V comprised of 11 genotypes which were characterized by late flowering (69 days) and dwarf genotypes (49.90cm). Cluster VI comprised of four genotypes which were predominantly characterized by tall genotypes (70.90cm) having long pods (8.99). The selection of diverse parents should be based on the component characters of yield that leads to better adaptation of the crop. Several researchers viz., Katiyar et al., (2009) and Singh et al., (2014) also gave emphasis on need of high genetic diversity to create the high genetic variation and genetic gain under selection. In the present study the genotypes grouped under cluster II, having more number of branches, pod clusters, seeds per pod and single plant yield can be utilized as potential donors in crossing program for improving yield.