Introduction

Safflower (Carthamus tinctorious L.; 2n = 24) is a member of the family compositae or Asteraceae. It has tremendous potential for growing under varied conditions (Golkar, 2014). Safflower is a long-season crop with a deep taproot that can draw moisture from deep in the subsoil. It is drought and heat tolerant and can grow in arid and semi-arid areas (Li and Mundel, 1996) where other oilseed crops fail to do so (Weiss, 2000).

Safflower is cultivated mainly for its seed as vegetable oil for human consumption and feed for the birds (Li and Mundel, 1996; Yadava et al., 2012). It has been used for various traditional purposes such as a source of carthamin for coloring foods, fabric painting, vegetable, and medicine (McGuire et al., 2012). It is also used as hay or silage (Smith, 1996) and as a snack food in Ethiopia and Sudan (Belayneh and Wolde-Mariam, 1989). In Ethiopia, the crop is cultivated since antiquity (Weiss, 2000) though its production is negligible, as compared to other oil crops commonly grow in the country (Wijnands et al., 2007). Nevertheless, Ethiopia is the fourth world leading producer of safflower after India, USA, and Mexico (Yadava et al., 2012). In the country, according to agricultural sample survey data, an area of 7,853 ha was under cultivation and 6581.4 tonnes of safflower produced with a productivity of 0.9 tonne/ha (CSA, 2008), but since then its production has declined. This decline in production might have resulted from lack of awareness about the crop, its relatively low productivity, and scanty research attention. However, safflower is an important underutilized crop, which can be used as an alternative oil crop to sustain crop production under changing environmental conditions such as rainfall and temperature.

Demand for vegetable oil is increasing in Ethiopia, and more than 90% of vegetable oil is imported, and the country spends about 432 million USD annually on vegetable oil (EIAR, 2016). Safflower seed oil content varies from 28% to 36% (Golkar, 2014), and the oil is rich in polyunsaturated fatty acids that helps to reduce blood cholesterol (Weiss, 2000). In order to enhance the production and productivity of safflower, researchers have to put effort into its improvement. Breeding materials with genetic variation in characters of interest is a prerequisite for genetic improvement in any crop. The value of breeding materials to the crop improvement is influenced by the genotype and environment. The relative extent of genetic variation in characters could be understood through their heritability estimates (Marwede et al., 2004; Toker, 2004; Camas and Esendal, 2006; Golkar et al., 2011; Reddy and Dhole, 2015). Pahlavani et al. (2007) reported considerable genetic variability among the parents for all evaluated traits in safflower, and they found that all the traits had high broad sense heritability. Camas and Esendal (2006) reported high heritability for plant height, number of seeds per head, and 1000-seed weight and low heritability for seed yield and number of branches. Though various studies of genetic variation in characters of safflower have been carried out using morphological, biochemical, and molecular markers (Ashri, 1975; Ramachandram and Goud, 1981; Pascual-Villalobos and Alburquerque, 1996; Jaradat and Shahid, 2006; Johnson et al., 2007; Yang et al., 2007; Amni et al., 2008; Khan et al., 2009; Mohamadi and Pourdad, 2009; Elfadl et al., 2010), limited effort has been put into investigation of Ethiopian safflower germplasm. Therefore, the objective of this study was to assess the genetic variation in some agro-morphological characters among 36 Ethiopian safflower germplasm accessions.

1 Results and Analysis

1.1 Genetic variation

The germplasm accessions were evaluated at two different locations (Holetta and Ginchi). The combined analysis showed that accessions were significantly (p < 0.05) different in all agro-morphological characters. Genotype x environment interaction was significant for days to maturity, plant height, and primary branches, which showed the presence of environmental influence in the accessions for these characters. On the other hand, for days to flowering and seed yield, genotype x environment interaction was non-significant (Table 1). Duncan Multiple Range Test (Table 2) indicated that accessions varied with days to flowering, which ranged between 144.5 and 151.0 with a mean of 148.0, and accessions varied with days to maturity, which ranged between 210.3 and 214.0 with a mean of 211.8. The maximum and minimum days of flowering were observed in accessions such as ACC.231421 (151 days) and ACC.231419 (144.5 days) respectively. Accessions showed variation in plant height, which ranged from 43.0 cm to 70.0 cm with a mean of 56.9 cm. Accessions such as ACC.205069 (70.0 cm) and ACC.241793 (43.0 cm) had the maximum and minimum height respectively. Accession ACC.241793 could be a good source of gene for short height cultivar breeding. There was wide variation in seed yield, which ranged between 39.5 kg/ha and 880.3 kg/ha with a mean of 395.4 kg/ha. Accession ACC.200487 had the highest seed yield (880.3 kg/ha).

1.2 Heritability and genetic advance

Heritability and genetic advance analysis is presented in Table 3. Relatively high proportion of genotypic variance in days to flowering and seed yield were observed in accessions. Days to maturity, plant height, and primary branches per plant had high phenotypic variance. Heritability in the broad sense were grouped into high (> 50%), moderate (20-50%), and low (< 20%) according to Stansfield (1988). High broad sense heritability was revealed to days to flowering (58.7%) and seed yield (64.6%). Plant height (50.7%) and primary branches per plant (45.0%) had moderately high broad sense heritability. Low heritability in broad sense was revealed to days to maturity (8.3%). Seed yield (72.7%) had the highest genetic advance as percent of mean followed by primary branches per plant (7.4%), plant height (6.5%), and days to flowering (1.2%). Days to maturity (0.07%) had the least genetic advance as percent of mean. High heritability along with high genetic advance as percent of mean was observed in seed yield of the accessions.

1.3 Cluster analysis

Accessions were fell into three major clusters (Table 4 and Figure 1). Cluster II (52.8%) had the highest proportion of the accessions followed by cluster III (25%) and cluster I (22.2%). Days to flowering ranged from 146.8 days to 149.5 days with a mean of 148.1 days in cluster I while it ranged from 144.5 days to 150.8 days with a mean of 148 days in cluster II. In cluster III, days to flowering ranged from 146 days to 151 days with a mean of 147.9 days. Cluster III comprised the high yielding accessions with seed yield varied from 579 kg/ha to 880.3 kg/ha with a mean of 683.0 kg/ha, where as cluster II and cluster I comprised the intermediate and low yielding accessions respectively.

1.4 Principal component analysis

The first three principal components together explained 78.6% of the total variation (Table 5). The first principal component explained 33.4% of the total variation while the second and third principal components explained 23.8 % and 21.5% of the total variation respectively. Days to maturity was the highest positive contributor to the variation under the first principal component followed by primary branches per plant and days to flowering. Under the second principal component, seed yield had the highest but negative contribution to the variation. Plant height was a major positive contributor to the variation under the third principal component.

2 Discussions

This study shows significant genetic variation between the germplasm accessions that will help towards genetic improvement in safflower. High broad sense heritability for days to flowering and seed yield showed the genetic control over the phenotype of these characters. Similarly, high heritability of seed yield was reported in faba bean (Toker, 2004) and safflower (Camas and Esendal, 2006; Mohammadi and Pourdad, 2009). These authors also indicated that plant height had high heritability, whereas present investigation showed moderate heritability for this character. Number of primary branches also had moderate heritability that conforms to the result reported elsewhere (Camas and Esendal, 2006), which indicated that the distinctiveness of these characters were under both genetic and environmental influences. On the other hand, days to maturity had low heritability, which is contrary to the result of research on faba bean (Toker, 2004). The discrepancy in the results above might occur due to different plant materials and/or different environmental conditions used (Falconer, 1981). High heritability in broad sense along with high genetic advance as percent of mean was observed in seed yield, and this agrees with the result reported on rapeseed (Sadat et al., 2010), which indicated that phenotype of this character is controlled by additive gene action, and  early generation selection of this character would be effective. Predominantly dominant genes were more important in controlling seed yield of safflower (Singh et al., 2008).

Accessions were grouped into three major classes in order that genotypes can be selected for hybridization. Increasing seed yield of safflower is one of the major breeding strategies; germplasm accessions in cluster III can be used for increasing seed yield. Plant height is also one of the most important agronomic characters of safflower; hence, desirable genotypes could be selected for breeding from among the accessions in cluster II. Days to maturity, primary branches per plant, and days to flowering were positive contributors to variation under the first principal component. Under the second principal component, days to flowering and plant height were major positive contributors, whereas seed yield was a major negative contributor, which indicates that there are a possibility of simultaneous improvement in seed yield, early flowering, and short height. Under climate change and global warming, understanding of flowering time is crucial for crop improvement, and it is one of the major important agronomic characters of crop plants (Long et al., 2007). Hence, genotypes could be selected for desirable flowering time from among the present accessions investigated. Khan et al. (2009) reported considerable variation among 193 safflower germplasm. These authors also showed that the studied germplam were belonged to eight categories, and the first three principal components explained 59 % of the total variation.

Since this investigation was comprised of limited accessions, further study of variation between large number of germplasm using both conventional and biotechnological tools would be helpful for improving safflower. However, this study shows wide genetic variation among 36 germpasm accessions; variations within major groups could be used for selection programme. Seed yield can be improved through early generation selection, but other characters should be improved through advanced generation selection. Accessions with desirable agronomic characters such as ACC.231419 and ACC.238275 could be used as a source of gene for early flowering and short height respectively.

3 Materials and Methods

3.1 Plant materials and experimental design

Thirty six safflower germplasm accessions, including standard check (Turkana) and local check were grown at Holetta and Ginchi (38^oE and 9^oN) during the cropping season of 2012. Accessions were obtained from Ethiopian Institute of Biodiversity Conservation. The experiment was laid out using 6 x 6 simple lattice design with two replications. Each accession was sown in a plot of six rows, 5 m long, and 30 cm between rows. Other cultural practices were followed as recommended. Ten plants from each accession were randomly selected for measuring agro-morphological characters such as plant height (PH) and number of primary branches per plant (PB). Other agro-morphological characters such as days to flowering (DF), days to maturity (DM), and seed yield (SY) were measured on a plot basis.

3.2 Data analysis

Combined variance, mean estimate, cluster, and principal component analyses were done using SAS software version 9.00 (SAS Institute, 2002). Estimates of phenotypic and genotypic coefficients of variation (Burton and De Vane, 1953), broad sense heritability (h²) (Camas and Esendal, 2006), and the expected genetic advance under selection (GA) (Johnson et al., 1955) assuming selection intensity of 5% (2.063) were obtained.

Acknowledgments

Author is grateful to the staff of Highland Oil Crops Commodity Research for a concerted effort to collect data for this study.

References

Amni F., Saeidi G., and Arzani A., 2008, Study of geneic diversity in safflower genotypes using agro-morphological traits and RAPD markers, Euphytica, 163: 21-30

https://doi.org/10.1007/s10681-007-9556-6

Ashri A., 1975, Evaluation of the germplasm collection of safflower (Carthamus tinctorius L) V. Distribution and regional divergence for morphological characters, Euphytica, 24: 651-659

https://doi.org/10.1007/BF00132903

Belayneh H., and Wolde-Mariam Y., 1991, Safflower production, utilization and research in Ethiopia, in: V. Ranga Rao, M. Ramachandran (Eds.), Proceedings Second International Safflower Conference, Hyderabad, India, 9-13 Jan. 1989, Indian Society of Oilseeds Research, Directorate of Oilseeds Research, Hyderabad, India, pp. 43-55

Burton G.W., and De Vane E.H., 1953, Estimating heritability in all fescue (Festuca arundinacea) from replicated clonal material, Agronomy Journal, 45: 478-481

https://doi.org/10.2134/agronj1953.00021962004500100005x

Camas N., and Esendal E., 2006, Estimation of broad-sense heritability for seed yield and yield components of safflower (Carthamus tinctorius L.), Hereditas, 143: 55-57

https://doi.org/10.1111/j.2006.0018-0661.01914.x
PMid:17362334

CSA, 2008/09, Agricultural sample survey: Report on area and production of major crops, Volume I , Central Statistical Agency (CSA) of the Federal Democratic Republic of Ethiopia, Addis Ababa, Ethiopia

Dhole V.J., and Reddy K.S., 2015, Genetic variation for phytic acid content in mungbean (Vigna radiata L. Wilczek), The Crop Journal, 3: 157-162

https://doi.org/10.1016/j.cj.2014.12.002

EIAR, 2016, National highland oilseeds commodity research strategy fifteen years (2016-2030), Addis Ababa, Ethiopia

Elfadl E., Reinbreeht C., and Claupein W., 2010, Evaluation of phenotypic variation in a worldwide germplasm collection of safflower (Carthamus tinctorius L) grown under organic farming conditions in Germany, Genetic Resource and Crop Evolution, 57: 155-170

https://doi.org/10.1007/s10722-009-9458-7

Falconer D.S., 1981, Introduction to quantitative genetics, second ed. John Wiley and Sons, Inc., New York, UK,  pp. 148-165

Golkar P., 2014, Breeding improvements in safflower (Carthamus tinctorius L.): A review, Australian Journal of Crop Science, 8:1079-1085

Golkar P., Arzani A., and Rezaei A.M., 2011, Genetic analysis of oil content and fatty acid composition in safflower (Carthamus tinctorius L.), Journal of American Oil Chemist Society, 88: 975-982

https://doi.org/10.1007/s11746-011-1758-3

Jaradat A.A., and Shahid M., 2006, Patterns of phenotypic variation in a germplasm collection of Carthamus tinctorius L. from the Middle East, Genetic Resource and Crop Evolution, 53: 225-244

https://doi.org/10.1007/s10722-004-6150-9

Johnson H.W., Robinson H.F., and Comstock R.E., 1955, Estimates of genetic and environmental variability in soybeans, Agronomy Journal, 47: 314–318

https://doi.org/10.2134/agronj1955.00021962004700070009x

Johnson R.C., Kisha T.J., and Evans M.A., Characterizing safflower germplasm with AFLP molecular markers, Crop Science, 47:1728-1736

https://doi.org/10.2135/cropsci2006.12.0757

Khan M.A., Von Witzke-Ehbrecht S., Maas B.L., and Becker H.C., 2009, Relationships among different geographical groups, agro-morphology, fatty acid composition and RAPD marker diversity in safflower (Carthamus tinctorius L), Genetic Resource and Crop Evolution, 56: 19-30

https://doi.org/10.1007/s10722-008-9338-6

Li D., and Mündel H.H., 1996, Safflower, Carthamus tinctorius L. Promoting the conservation and use of underutilized and neglected crops, 7. Institute of Plant Genetics and Crop Plant Research, Gatersleben/International Plant Genetic Resources Institute, Rome, Italy

Long Y., Shi J., Qiu D., Li R., Zhang C., Wang J., Hou J., Zhao J., Shi L., Park B.S., Choi S.R., Lim Y.P., and Meng J., 2007, Flowering Time Quantitative Trait Loci Analysis of Oilseed Brassica in Multiple Environments and Genome wide Alignment with Arabidopsis, Genetics, 177: 2433-2444

PMid:18073439 PMCid:PMC2219480

Marwede V., Schierholt A., Moillers C., and Becker H.C., 2004, Genotype x environment interactions and heritability of tocopherol contents in canola, Crop Science, 44: 728-731

https://doi.org/10.2135/cropsci2004.7280
https://doi.org/10.2135/cropsci2004.0728

McGuire P.E., Damania A.B., and Qualset C.O., 2012, Safflower in California, The Paulden F. Knowles personal history of plant exploration and research on evolution, genetics, and breeding, Agronomy Progress Report No. 313, Dept. of Plant Sciences, University of California, Davis CA, USA

Mohammadi R., and Pourdad S.S., 2009, Estimation, interrelationships and repeatability ofgenetic variability parameters in spring safflower using multi-environment trial data, Euphytica, 165: 313-324

https://doi.org/10.1007/s10681-008-9789-z

Pascual-Villalobos M.J., and Alburquerque N., 1996, Genetic variation of safflower germplasm collection grown as a winter crop in Southern Spain, Euphytica, 92: 327-332

https://doi.org/10.1007/BF00037116

Pahlavani M.H., Saeidi G., and Mirlohi A.F., 2007, Genetic analysis of seed yield and oil content in safflower using F1 and F2 progenies of diallel crosses, International Journal of Plant Production, 1(2): 129-140

Ramachandram M., and Goud J.V., 1981, Genetic analysis of seed yield, oil content and their components in safflower (Carthamus tinctorius L), Theoretical and. Applied Genetics, 60: 191-195

https://doi.org/10.1007/BF00264529
PMid:24276693

Sadat A.H., Nematzadeh  A.G., Jelodar N.B., and Chapi G.O., 2010, Genetic evaluation of yield and yield components at advanced generations in rapeseed (Brassica napus L.), African Journal of Agricultural Research, 5: 1958-1964

SAS Institute, 2002, SAS/STAT guide for personal computers, version 9.00 edition. Cary, NC: SAS Institute Inc.

Singh V.N., and Kolekar M., 2008, Nimbkar, Breeding strategy for improvement of flower and seed yield in safflower, 7^th international Safflower conference, Wagga, New South Wales, Australia

Smith J.R., 1996, Safflower, AOCS Press, Champaign, IL, USA

https://doi.org/10.1201/9781439832080

Stansfield W.D., 1988, Theory and problems of genetics, M. C. Grow Hill Book Co. New York USA, pp. 220-221

Toker C., 2004, Estimates of broad-sense heritability for seed yield and yield criteria in faba bean (Vicia faba L.), Hereditas 140: 222-225

https://doi.org/10.1111/j.1601-5223.2004.01780.x
PMid:15198712

Weiss E.A., 2000, Oil Seed Crops, Blackwell Science Ltd, Oxford, UK

Wijnands J.H.M., Biersteker J., and Hiel R., 2007, Oilseeds business opportunities in Ethiopia. Ministry of Agriculture, Nature and Food Quality, the Netherlands http://library.wur.nl/WebQuery/wurpubs/361069

Yadava D.K., Vasudev S., Singh N., Mohapatra T., and Prabhu K.V., 2012, Breeding major oil crops: Present status and future research needs. In: Technological Innovations in Major World Oil Crops, S.K. Gupta, (Eds.), Springer Science+Business Media, LLC, pp. 17-51

Yang Y.X., Wu W., Zheng Y.L., Chen L., Liu R.J., and Huang C.Y., 2007, Genetic diversity and relationships among Safflower (Carthamus tinctorius L) analyzed by inter simple sequence repeats (ISSRs), Genetic Resource and Crop Evolution, 54: 1043-1051

https://doi.org/10.1007/s10722-006-9192-3