This is an open access article d istributed under the terms of Creative Commons Attribution-NonCommer cial-ShareAl ike 4.0 International License, which permits unrestricted non-commercial use, d istribution, and reproduction in any med ium, provide d the original author and source are credited. INTRODUCTION China aster [Callistephus chinensis (L.) Nees] is a diploid (2n=2x=18) flowering annual belonging to the fa mily Aster a cea e a nd is a na tive of China (Navalinskien et al., 2005). The genus Callistephus derives its name from two Greek words ‘Kalistos’ and ‘Stephos’ meaning ‘most beautiful’ and ‘crown’, respectively. In India, China aster ranks third among the flowering annuals after chrysanthemum and marigold (Chakraborty et al., 2019). China aster is winter season annual crop. It is commercially grown for loose andcut flower, which are used in flower decoration, preparation of bouquets and garlands. It is also used in landscape gardening as a bedding plant to provide colour break and mass effect. It is gaining popularity in India, because of ease in cultivation, array of colours and varied uses (Bhargav et al., 2016). For any breeding programme, characterization and evaluation are crucial steps in developing a variety and further research. Cluster analysis and principal component analysis are two important parameters to determine the diversity of the crop. Considering the importance of the crop, the present investigation was carried out to assess the diversity among 42 genotypes of China aster based on the fourteen quantitative traits. MATERIALS AND METHODS The present study was conducted at the Division of Flower and Medicinal Crops, ICAR- Indian Institute of Horticultural Research, Bengaluru during 2015-16 a nd 2016-17. T he exper imenta l site wa s geographically located at 13o 58¹ N Latitude, 78oE Longitude and at an elevation of 890 m above mean sea level. The soil of experimental plot was red loamy with pH 7.35 and E.C. of 0.26 dsm-1. A total of 42 genotypes including 21 varieties and 21 stabilized lines were evaluated for vegetative growth, flowering, yield and postharvest life in randomized complete block design with two replications. Twenty plants per replication were planted at a spacing of 30 x 30 cm Original Research Paper J. Hortic. Sci. Vol. 18(1) : 84-89, 2023 https://doi.org/10.24154/jhs.v18i1.2138 Assessment of genetic diversity in China aster [Callistephus chinensis (L.) Nees] Bhargav V.1, Kumar R.2*, Bharathi T.U.2, Dhananjaya M.V.2 and Rao T.M.2 1Department of Horticulture College of Horticulture and Forestry, Central Agricultural University, Pasighat - 791102, Arunachal Pradesh, India 2Division of Flower and Medicinal Crops ICAR- Indian Institute of Horticultural Research, Bengaluru - 560089, Karnataka, India *Corresponding author Email : Rajiv.Kumar11@icar.gov.in ABSTRACT China aster [Callistephus chinensis (L.) Nees] is a flowering annual mainly cultivated for loose flower and cut flower, bedding and pot culture. To assess the genetic diversity, 42 genotypes were evaluated for fourteen quantitative traits. The genotypes were found to be highly variable for the traits such as plant height, plant spread, flower stalk length, 100 flower weight, number of flowers per plant, weight of flowers per plant and flower yield per hectare. However, low variability was recorded for vase life and shelf life. The genotypes were broadly grouped into two clusters, which were further divided into cluster 1a, 1b and cluster 2a, 2b, respectively. All the genotypes in cluster 1a were vigorous and medium flowering, whereas, genotypes in cluster 1b were tall, erect, vigorous and late flowering. The cluster 2a comprises of the genotypes with short stature, small flower and early flowering, however, cluster 2b contains only two genotypes. In principal component analysis (PCA) PC1 was highly correlated to flower yield, weight of flowers/plant, flower stalk length and plant height and PC2 was highly positively correlated to shelf life and vase life and negatively correlated to 100 flower weight. The results suggested that the existing variation in China aster genotypes could be used for the development of trait-specific novel genotypes. Keywords : China aster, cluster analysis, diversity, principal component analysis 85 J. Hortic. Sci. Vol. 18(1) : 84-89, 2023 under open field conditions. The recommended agronomical practices were adopted to raise the crop. Five random plants were selected for recording the various quantitative traits viz., plant height (cm), number of leaves per plant, plant spread (cm), number of branches per plant, days to first flowering, flower stalk length (cm), flower head diameter (cm), 100 flowers weight (g), number of flowers per plant, weight of flowers per plant (g), duration of flowering (days), vase life (days) and shelf life (days). Descriptive statistics (e.g., range, standard deviation, mean, standard error of mean), clustering based on average linkage, and euclidian distance and principal component analysis (PCA) were calculated using XLSTAT (Addinsoft, 2017). RESULTS AND DISCUSSION The diversity among the China aster accessions for quantitative traits was high (Table 1). Wide range of variation was observed for most of the characters such as plant height (8.20-61.80 cm), plant spread (8.75- 42.65 cm), flower stalk length (4.65-49.10 cm), 100 flower weight (105.00-548.25 g), number of flowers per plant (7.35-65.05), weight of flowers per plant (7.72-235.21 g), flower yield per hectare (6.48-197.57 q/ha). Highest variability was recorded for weight of flowers per plant, which directly represent the flower yield per hectare with a mean of 124.49 g and 104.57 q/ha, respectively having a C.V. of 48.65%. Minimum variability was observed for vase life (5.40-9.50 days) followed by shelf life (2.35-4.42 days) with a mean of 7.06 (C.V. 14.26%) and 3.42 days (C.V. 14.37%), respectively. Presence of such high genetic variability among the genotypes for these parameters will form the basis for effective selection of superior genotypes in China aster. Such wide variability for many quantitative traits was also reported by Gupta and Dutta (2005) a nd Ba ner ji et al. (2012) in chr ysa nthemum, a nd Kuma r et al. (2014) in bougainvillea. Cluster analysis was carried out to distinguish possible groups among the populations using Ward method (Fig. 1). The Agglomerative hierarchical clustering (AHC) allows sub-division of 42 genotypes into two major clusters based on the correlation that exists between the morphological traits among the genotypes. In the present study, cluster 1 comprised of 33 populations, which was further divided into two sub- groups cluster Ia contains 18 genotypes and Ib contains 15 genotypes. Nine genotypes were classified into cluster 2, which was again divided into two sub- clusters group IIa and group IIb containing 7 and 2 genotypes, respectively. In cluster Ia, all the genotypes were vigorous, medium flowering with big flowers, whereas, genotypes in cluster Ib were tall and erect, vigorous, late flowering with big flowers. Except genotype IIHRJ22, all the genotypes in cluster IIa belong to the Japanese Trait Range Mean ± SE(m) CV (%)Minimum Maximum Plant height (cm) 8.20 61.80 45.50 ± 0.66 24.97 Number of leaves/plant 9.20 32.35 19.50 ± 0.86 22.13 Plant spread (cm) 8.75 42.65 24.54 ± 0.72 32.59 Number of branches/plant 6.65 17.60 12.08 ± 0.30 19.83 Days to first flowering 46.85 100.15 66.71 ± 0.71 16.92 Flower stalk length (cm) 4.65 49.10 35.62 ± 0.72 29.34 Flower head diameter (cm) 3.54 6.74 5.39 ± 0.07 15.96 100 flowers weight (g) 105.00 548.25 292.02 ± 2.66 31.98 Number of flowers/plant 7.35 65.05 40.92 ± 0.46 34.82 Weight of flowers/plant (g) 7.72 235.21 124.49 ± 2.06 48.65 Duration of flowering (days) 16.58 34.40 24.92 ± 0.30 19.53 Flower yield/ hectare (q/ha) 6.48 197.57 104.57 ± 1.73 48.65 Vase life (days) 5.40 9.50 7.06 ± 0.12 14.26 Shelf life (days) 2.35 4.42 3.42 ± 0.08 14.37 Table 1 : Descriptive statistics of quantitative traits in China aster genotypes Assessment of genetic diversity in China aster [Callistephus chinensis (L.) Nees] 86 Fig. 1 : Dendrogram showing genetic relationship among 42 China aster genotypes based on morphological data originated short stature, early flowering and small flowered genotypes, however, IIb contains only two European originated genotypes namely Milady Scarlet and Milady White. Similar results were also obtained by Kumar et al. (2011) in snapdragon, and Bharathi and Jawaharlal (2014) in marigold. It was observed that genotypes representing particular geographic regions were grouped together. The heterogeneous origin of genotypes, place of release, different ploidy levels often grouped together in the same cluster, suggesting the ancestral relationship between the various genotypes (Bellundagi et al., 2013). For further improvement in the morphological and yield parameters, genotypes may be selected on the basis of genetic divergence. Crossing between highly genetic divergent genotypes could yield better results (Singh et al., 2016). Therefore, genotypes may be chosen for crossing on the basis of genetic divergence as depicted in the dendrogram. Based on the cluster distance, genotypes belonging to Matsumoto series, Milady Scarlet and Milady White, and genotypes belonging to Phule Ganesh, Arka, IIHRJ3-2 and IIHRG13 were most divergent. Therefore, crossing among the most divergent genotypes can achieve improvement in the morphological and yield attributes by getting desirable transgressive segregants. To determine the most significant characteristics of the data set and also to determine the distances between the genotypes ba sed on the da ta obta ined on morphological traits, set of 42 genotypes used for cluster a na lysis wer e subjected to Pr incipa l Component Analysis (PCA) (Table 2). The analysis a lso helped to under sta nd the contr ibution of morphological characters across the genotypes. The fir st two components of the seven consider ed accounted for most of the variation. The first principal component (PC1) explained 55.66% of the total variation and was positively correlated to all the traits; highly correlated to flower yield, weight of flowers/ plant, flower stalk length and plant height. The PC2 explained 10.55% of total variation and was highly positively correlated to shelf life and vase life, and negatively to 100 flower weight (Table 2). Because PC1 and PC2 accounting 66.21% of cumulative variance the compounds, a scatterplot was made for both (Fig. 2). J. Hortic. Sci. Vol. 18(1) : 84-89, 2023 Bhargav et al. 87 Fig. 2 : Principal Component Analysis (PCA) of China aster genotypes based on morphological characters 1. Arka Kamini 2, Arka Poornima, 3. Arka Shashank, 4. Arka Violet Cushion, 5. Arka Aadya, 6. Arka Archana, 7. Phule Ganesh Pink, 8. Phule Ganesh Purple, 9. Phule Ganesh White, 10. Phule Ganesh Violet, 11. Matsumoto Yellow, 12. Matsumoto White, 13. Matsumoto Rose, 14. Matsumoto Scarlet, 15. Matsumoto Red, 16. Matsumoto Pink, 17. Local Pink, 18. Local White, 19. Local Violet, 20. Milady Scarlet, 21. Milady White, 22. IIHRD5, 23. IIHRC5, 24. IIHRC42, 25. IIHRCC39, 26. IIHRCC5-1A, 27. IIHRCC31-2, 28. IIHRJ3, 29. IIHRJ22, 30. IIHRI1, 31. IIHRI66, 32. IIHRCC31A, 33. IIHRG13, 34. IIHRI69-2, 35. IIHRD2, 36. IIHRCC19, 37. IIHRJ3-2, 38. IIHRI69, 39. IIHRI16B, 40. IIHRH3, 41. IIHRE10, 42. IIHRC1. Table 2 : Principal component analysis in China aster genotypes Variable PC1 PC2 PC3 PC4 PC5 PC6 PC7 Eigen value 7.79 1.48 1.19 0.95 0.71 0.54 0.46 Cumulative variance (%) 55.66 66.21 74.68 81.49 86.57 90.41 93.70 Plant height (cm) 0.82 0.29 0.22 -0.25 -0.09 -0.02 -0.22 Number of leaves/plant 0.72 -0.25 -0.41 -0.19 0.05 0.39 0.02 Plant spread (cm) 0.82 -0.05 -0.27 0.38 -0.03 -0.04 -0.12 Number of branches/plant 0.75 -0.02 -0.17 0.43 -0.15 0.20 -0.17 Days to first flowering 0.71 0.04 -0.47 -0.36 0.24 0.12 0.04 Flower stalk length (cm) 0.86 0.13 0.30 -0.12 -0.05 0.08 -0.26 Flower head diameter (cm) 0.80 -0.08 -0.11 -0.05 -0.34 -0.31 -0.17 100 flowers weight (g) 0.81 -0.33 -0.17 -0.03 -0.22 -0.23 0.25 Number of flowers/plant 0.75 -0.02 0.56 0.04 0.24 0.17 -0.01 Weight of flowers/plant (g) 0.88 -0.30 0.28 -0.01 0.05 -0.03 0.22 Duration of flowering (days) 0.61 0.27 -0.18 0.25 0.59 -0.32 -0.04 Flower yield/hectare (q) 0.88 -0.30 0.28 -0.01 0.05 -0.03 0.22 Vase life (days) 0.46 0.67 -0.10 -0.41 -0.09 -0.09 0.13 Shelf life (days) 0.38 0.70 0.02 0.39 -0.20 0.17 0.30 J. Hortic. Sci. Vol. 18(1) : 84-89, 2023 Assessment of genetic diversity in China aster [Callistephus chinensis (L.) Nees] 88 Large variation was recorded in traits such as shelf life, vase life, plant height, 100 flower weight as mentioned by the relative length of the vectors in the biplot diagram. The biplot also signifies the positive correlation between the parameters viz., shelf life, vase life, plant height, duration of flowering, days to first flowering and 100 flower weight, flower yield per hectare, weight of flowers per plant and number of leaves per plant as indicated by the acute angle. The genotypes like Matsumoto, Milady Scarlet and Milady White which were short and early flowering for med a gr oup in one qua dr a nt a nd a ll a r e comparatively late flowering genotypes which are, tall with big flower s for med a nother gr oup a nd intermediate medium flowering forms the group in between. Most of the morphological traits contributed equally in grouping of genotypes except vase life and shelf life, which were distributed away from the genotypes. The result of PCA is consistent with that of the cluster analysis. A similar pattern was also observed for hips traits in Rosa sp. (Verma et al., 2013) and in pea (Esposito et al., 2007). CONCLUSION This study indicated that the quantitative traits are useful for preliminary evaluation of genetic diversity in China aster. PCA revealed that number of flowers per plant, flower yield per hectare, flower stalk length, pla nt height a nd pla nt spr ea d a r e key tr a its contributing to the maximum variation among the genotypes. The cluster analysis showed significant genetic variability among the evaluated China aster genotypes, which may provide an excellent opportunity for crop improvement through hybridization between the genotypes of different clusters, to assemble desirable traits with higher heterotic potential. ACKNOWLEDGEMENT We acknowledge ICAR-IIHR, Bengaluru for providing research facilities to carry out this study. The first author is thankful to ICAR-IARI, New Delhi for a wa r ding IARI Fellowship dur ing his Ph. D. programme. REFERENCES Addinsoft.2017. XLSTAT 2017: Data analysis and statistical solution for microsoft excel. Paris, France. https://www.xlstat.com/en/download Banerji, B.K., Batra , A., Dwivedi, A.K. 2012. Morphological and biochemical characterization of chrysanthemum. J. Hortic. Sci., 7(1): 51-55. Bellundagi, A., Singh, G.P., Singh, A.M., Arora, A., Jain, N., Prasad, S.S. Kumar, J., Ahlawat, A. and Ramya, P. 2013. Genetic diversity for moisture deficit stress adaptive traits in bread wheat (Triticum aestivum L.). Indian J. Plant Physiol., 18(2): 131-135. Bharathi, T.U. and Jawaharlal M. 2014. Genetic divergence of African marigold (Tagetes erecta L.). Trends Biosci., 7(16): 2233-2236. Bhargav, V., Sharma, B.P., Dilta, B.S., Gupta, Y.C. and Negi, N. 2016. Effect of different plant spacings and cultivars on growth, flowering and seed production of China aster [Callistephus chinensis (L.) Nees]. Res. Environ. Life Sci., 9(8): 970-972. Chakraborty, A., Bordolui, S.K., Mahato, M.K., Sadhukhan, R. and Sri Veda, D.J.M.S.N.K. 2019. Variation in seed production potential of China aster genotypes in the New Alluvial Zone of West Bengal. J. Crop Weed, 15(1): 201-204. Esposito, M.A., Martin, E.A., Cravero, V.P. and Cointr y, E. 2007. Cha r a cter iza tion of pea a ccessions by SRAP’s ma rkers. Sci. Hortic., 113(4): 329-335. Gupta, V.N. and Dutta, S.K. 2005. Morphological and chemica l cha ra cterization of thirty sma ll flowered chrysanthemum cultivars. J.Orn. Hortic., 8(2): 91-95. Kumar, P., Janakiram, T., Bhat, K.V., Ritu Jain, Prasad, K.V. and Prabhu, K.V. 2014. Molecular characterization and cultivar identification in Bougainvillea spp. using SSR markers. Indian J. Agric. Sci., 84(8): 1024-1030. Kumar, R., Kumar, S., Kumar, P. and Mer, R. 2011. Genetic variability and divergence analysis in snapdragon (Antirrhinum majus L.) under tarai conditions of Uttarakhand. Prog. Hortic., 43(2): 332-336. Navalinskien, Ė, M., SamuitienĖ, M., and Jomantien Ė,  R.  2005. Molecula r detection a nd J. Hortic. Sci. Vol. 18(1) : 84-89, 2023 Bhargav et al. 89 characterization of phytoplasma infecting Callistephus chinensis pla nts in Lithuania. Phytopathologia Polonica, 35: 109-112. Singh, D., Arya, R. K., Chandra, N., Niwas, R. and Salisbury, P. 2016. Genetic diversity studies in relation to seed yield and its component traits in Indian mustard (Brassica juncea L. Czern & Coss.). J. Oilseed Brassica, 1(1): 19-22. Verma, M. K., Lal, S., Ahmed, N. and Sagoo, P. A. 2013. Character association and path analysis in hip rose (Rosa sp.) genotypes collected from Nor th Wester n Hima la ya n r egion of Kashmir. Afr. J. Agric. Res., 8(39): 4949-4955. (Received : 03.07.2022; Revised : 23.06.2023; Accepted 25.06.2023) J. Hortic. Sci. Vol. 18(1) : 84-89, 2023 Assessment of genetic diversity in China aster [Callistephus chinensis (L.) Nees]