Based on variation in agro morphological, yield and nutritional traits variations, it was attempted to group the chickpea genotypes and identify each one of them through descriptors (Table 1; Fig 1). Anthocyanin pigmentation was classified into two classes as pigmentation absent in 3 and present in 96 genotypes. Variations were observed among genotypes for time of 50% flowering with at least one open flower in 50% plant population was noted and categorized into two viz
., early (40-60 days) and medium (61-80 days). 67 genotypes flowered early as compared to 32 genotypes of medium duration and none was recorded in the category of extra early and late duration. Plant growth habit is a distinguishing feature in characterization. 18 genotypes were observed as erect type, 81 genotypes showed semi-erect type of growth habit. Four groups were made based on intensity of green colour of foliage in which 3 genotypes showed light green colour, 94 genotypes showed medium, 1 genotype showed dark green and greenish purple colour. With regard to size of leaflets, the variation was observed and categorized into two groups; 44 genotypes recorded small size (10 mm), 55 genotypes recorded medium (10-15 mm). For leaf pattern, 98 genotypes had pinnate leaves and one genotype had simple leaf. No variation was found among the genotypes for number of flowers per peduncle as all genotypes had single flower per peduncle. Three groups were made based on variation in flower colour; one genotype had white, 97 genotypes showed pink, only one genotype showed blue colour flower. Genotypes were examined for the stripes on standard petal of the flower, the stripes were absent in 1 and present in 97 genotypes. Morphological features of genotypes have been a major component of varietal identification. However, it is not possible to identify varieties using any single parameter. A detailed morphological description of plants and seeds should therefore be prepared. Utilization of these features in sequential fashion is useful and convenient to distinguish different genotypes. Similarly, genotypes identification based on distinguishable morphological characters were reported by Lalitha (2007)
Table 1: Agromorphological variation in chickpea germplasm.
Fig 1: Histogram of eleven yield and yield attributing traits in chickpea germplasm.
Significant variability differences were observed in the yield traits. Maximum variability was recorded in plant yield followed by iron concentration (mg/kg), hundred seed weight and number of pods/plant. Mean of DTF was 58.13 days which ranged from 47 days to 63 days. Maturity days ranged from 97 to 108 days with the mean of 103.63 days. Plant height was classified into two classes, 75 having short (<45 cm), 24 with medium (45-65 cm) height. The trait possessed a mean of 42.47 cm ranging from 22.40 cm to 60.00 cm. Primary branches and secondary branches also emphasized on yield, thus, had a mean of 2.30 and 7.75, respectively.
Considerable variation did exist in traits namely, pods/plant, hundred seed weight, plant yield, protein (%), zinc and iron concentration. Genotype (ICC12539) showed high number of pods (19.40). Seed index is the important character, it influences seed yield to a great extent. Seventy five accessions had seed weight less than 20 g (very small); nineteen fell into small (20-25 g) and remaining five in medium class (26-35 g). Plant yield was recorded highest in RG 2016-117 (83 g) ranging from 7 g to 83 g with the mean of 24.94 g.
Highest protein % was observed in RGH- 4 with variability of 14.48%. RGH-4, RGH-56, RG2016-84, ICC251762, RGH53, IPC98-12, RG2016-03, ICC1053 and RGH58 recorded high protein content (>20%). These entries possessing high protein had pink flower. ICC251811 beard blue flower with 18% protein. Research says, high protein content in blue flowered, could be because of their reduced seed size compared to pink flowered and white flowered. RGH-4 and RG-2016-84 showed high protein with pink flower and recorded small seed size followed by RGH-56, ICC251762, RGH 53, IPC 98-12, RG 2016-03, ICC1053 and RGH 58 with very small seed size (<20 g). It is observed that blue flowered plants with smaller seeds had higher protein content compared to pink flowered plants with larger seeds. They suggested linkages between genes for flower colour, protein and seed weight. Existence of linkage between flower colour and seed size was also reported by Atta et al., (2003).
Zinc and Iron (mg/kg)
High variability in iron concentration was found as compared to zinc. Iron content ranged from 10 to 73.65 mg/kg whereas, zinc ranged from 21.70 to 136.00 mg/kg with the mean of 35.65 and 50.07 mg/kg, respectively. Vijay recorded very high zinc (136 mg/kg) followed by RGH-7 (81.90) and ICC269712 (79.90). RGH52 and RGH11 recorded high iron concentrations of 73.65 and 72.89 mg/kg, respectively. Iron concentration tended to be less affected by market class, although small-seeded chickpea contained more Fe than large seeded. RGH53 hadhigh values for both protein and zinc, with an average of 21.86% and 73.00 ppm, respectively, but it showed low iron concentration (20.77 ppm).
Results revealed that days to flowering showed very strong relation with days to maturity. Plant height recorded significant association with primary branches, secondary branches; pods/plant, hundred seed weight and plot yield (Table 2). However, negative association was recorded between plant height and protein. Primary branches possessed strong association with secondary branches and pods per plant; secondary branches recorded significant relation with pods per plant and hundred seed weight. Plant yield represented strong association with pods per plant and hundred seed weight. A positive correlation was seen between hundred seed weight and pods / plant. Except plant height, no other trait gave any relationship with protein. As far as iron and zinc is concerned, none of the trait showed any kind of significant relationship with zinc, whereas, for iron concentration, negative significant trend was seen with days to flowering, maturity, plot yield and protein. Primary branch indicated that hike in this trait will increase the iron. Negative but non-significant trend was seen between seed weight and protein content. A negative relationship between seed size and protein content implies that as seed increases in size there is an increased amount of starch deposition altering the starch/protein ratio (Bahl et al., 1979).
Protein content has been found to be negatively correlated in chickpea (Frimpong et al., 2009; Gangola et al., 2012).
Similar to this study, protein content was found to be negatively correlated with seed size in pigeonpea. However, breeding lines combining high protein content with medium seed size were successfully developed. The existence of a negative correlation between protein content and seed size and significant associations of protein content with flower color, seed coat color and seed shape suggests that development of chickpea cultivars with high protein content and desired seed traits (size, shape and color) would require large segregating populations and the selection of desired recombinants. Blue flower color, grey seed coat color and pea seed shape of high protein line show pleiotropic effects of gene (s) and these traits were also associated with seed size and protein content. Thus, it would be important to search for other high protein lines in the germplasm and use diverse sources of high protein content in breeding programs for development of high protein chickpea cultivars with desired seed traits (size, shape and color).
Principal component and cluster analysis
Table 2: Correlation coefficients of eight yield and yield attributing traits in chickpea genotypes.
Relationships among the different traits were evaluated by principal component analysis (PCA). The first six principal components provided a reasonable summary of the data and explained 80.19% of the total variation. In the chickpea genotypes, PC1 explained variation in plant height, primary branches, secondary branches, number of pods/plant and grain yield per plant, PC2 in DTF and maturity and iron concentration, PC4 in protein %, PC5 and PC6 for zinc and 100-seed weight, respectively. PC3, however, did not report variation for any of the trait (Table 3 and Fig 2).
Table 3: Eigen vectors and eigen values for the six principal components in chickpea genotypes.
Fig 2: Principal coordinates between PC1 and PC2 based on active observations and variables.
In cluster analysis, 99 genotypes were subjected to cluster analysis using Agglomeration method: Un-weighted pair group average based on dissimilarity estimated by Euclidean distance (Table 4 and Fig 3). Results indicated that the entire material was grouped into ten clusters. 33 genotypes were categorized into cluster I followed by cluster III (28), cluster V (23) and cluster II and IV with 6 and 4 genotypes, respectively. Rest of the five clusters had one entry each. The inter and intra cluster distances were also worked out to understand the relationship of the genotypes within and between the clusters based on genetic distances (Table 5). Inference can be made that maximum intra cluster distance was observed in cluster V, I and III. All these three clusters also beard more number of genotypes. The inter cluster distance was found high in clusters VI and IX followed by VI and VII and IV and VI. Thus, for chickpea improvement program the genotypes from these clusters based on inter cluster distances can be taken as donors. For DTF, ICC12365 flowered early (cluster I), two genotypes, one each from cluster II (RG 2016-50) and VII (RG 2016-117) recorded early and late maturity. High means of plant height were shown by RG 2003-28 from cluster X. Three genotypes (RG 2016-50; RG 2016-29 and RGH7) from clusters (II, V and VII) showed high cluster means for primary branches. One entry each from cluster IV (RGH 35) for secondary branches, VII (RG 2016-117) for number of pod/plant, IV (RGH 85) for hundred seed weight, VII (RG 2016-117) for plant yield, IX (RGH 58) for protein per cent, VI (Vijay) for zinc and IX (RGH 58) for high iron. Genetic diversity among the parents is a prerequisitefor ensuring the chances of selecting improved segregants for various characters (Dwevedi et al., 2009).
Success of hybridization in highly self-pollinated crops like chickpea mainly depends on the genetic diversity of the parents. Cluster analysis was found to be a potential tool in quantifying the degree of divergence in germplasm collections of crop plants.
Table 4: Cluster analysis presenting ten clusters along with genotypes based on inter and intra cluster distances.
Fig 3: Dendrogram of ninety nine genotypes based on UPGMA-Euclidean distances.