Studies on genetic variabilityand heritability in Asiatic hybrid lily (Lilium x elegans L)  

M. R. Dhiman , Chander Parkash , Raj Kumar , M S Guleria , Munish Dhiman
Indian Agricultural Research Institute, Regional Station, Katrain, Kullu Valley, Himachal Pradesh 175129
Author    Correspondence author
Molecular Plant Breeding, 2015, Vol. 6, No. 2   doi: 10.5376/mpb.2015.06.0002
Received: 10 Dec., 2014    Accepted: 29 Jan., 2015    Published: 10 Feb., 2015
© 2015 BioPublisher Publishing Platform
This is an open access article published under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Preferred citation for this article:

Dhiman et al., Topic: Studies on genetic variabilityand heritability in Asiatic hybrid lily (Lilium x elegans L), Molecular Plant Breeding, 2015, Vol.6, No. 2 1-8 (doi: 10.5376/mpb.2015.06.0002)


Seventeen genotypes of Asiatic hybrid lily were evaluated for various vegetative and floral traits to ascertain genetic parameters such as variability, heritability, genetic (GCV) and phenotypic (PCV) coefficient of variation, genetic advance and correlation during 2012-2013. PKLH-2 was the earliest to flower and PKLH-12 took maximum days to flower. The maximum plant height and flower size was recorded in Stargater and Prato, respectively. PKLH-2 produced maximum number of shoots per plant. PKLH-8 exhibited the maximum number of flowers per plant. The investigation revealed wide variation for all the characters indicating sufficient genetic variability to be exploited in breeding programme. The maximum value of PCV and GCV was recorded for number of spots/cm2. High heritability coupled with high genetic advance was observed for plant height and number of leaves per plant. Selection on the basis of these characters would be more effective for improvement of Lilium. Plant height, the economically important trait revealed a highly significant and positive correlation with number of leaves per plant, leaf area, flower diameter, number of flowers per plant and inflorescence length. Hence, direct selection from germpalsm lines may be effective for improvement in closely related traits.

Genetic variability; Heritability; Correlation; Lilium

Lilies are very important ornamental plants and are used worldwide as cut flowers and as potted and garden-grown plants. The genus Lilium comprises of > 90 species (Asano, 1989) and is classified into sections (Comber 1949, Smyth et al. 1989). Most species of section Sinomartagon, such as Lilium dauricum, L. maculatum, L. concolor, L. leichtlinii, L. davidii and L. cernuum, are distributed in East Asia. The Asiatic hybrid lily, one of the most popular ornamental plants world-wide, which has been derived from interspecific crosses of species of section Sinomartagon (Leslie, 1982).Cultivars of hybrid lily have a wide range of flower colour, shape, size and morphological characteristics. As a cut flower, lily is now ranked as the fourth most important crop in the Netherlands. In India, lilium was introduced during 80’s, and since then, a large number of varieties have been introduced from exotic sources at different Central and State Research Centers and AgriculturalUniversities.There is a tremendous scope for its commercial cultivation in Himachal Pradesh, Jammu & Kashmir, Uttarakhand and similar other hilly terrains of India. The main strength of lilium cultivation in these areas lies in the suitability of climate for quality cut flower production and bulb multiplication, availability of manpower, lesser cost of cultivation compared to major temperate bulb growing countries. However, the non-availability of any Indian cultivars has been the major constraint for expanding its area of cultivation and therefore farmers import planting material from abroad every time. Hence, to boost the cultivation of lilium and cut down the production cost, there is a need to breed indigenous cultivars so that, the farmers can get planting materials at reasonable rates.

The performance of any crop or variety largely depends on genotypic and environmental interactions. As a result, cultivars which perform well at one region may not show same performance in other region having different climatic conditions. Therefore, it is necessary to evaluate the germpalsm to find out the most promising genotypes suitable for various quality traits. The Asiatic hybrids of lilium have originated from more than 12 very different lilium species (De Graaff and HornBack, 1967; Wadekamper, 1977). For that reason, there is a great genetic variation among its cultivars. Thus, determining genetic diversity through variation between genotypes, genotype groups or populations is the most important breeding tool to select better genotypes for improvement in desired traits. Since, lilium is a vegetatively propagated crop and selection is an easy method for its improvement, an estimate of genetic advance along with heritability is helpful in assessing the reliability of character for selection. Knowledge of the mechanisms underlying the correlations between different traits is fundamental for understanding the degree of integration of the phenotype and to resolve the constraints imposed on evolutionary processes. All these measures are important for the identification of genetically distant parental combinations, aiming to use distinct gene sets in crossings for getting superior hybrids and segregants, to evaluate the degree of genetic erosion, or even to determine the extent of the genetic base of cultivated forms to develop heterotic groups. With the background in view, the present study was undertaken to assess and estimate the magnitude and nature of variation among 17 Asiatic lilium genotypes with respect to various vegetative and floral characteristics that would be helpful for further crop improvement programme.
1 Materials and Methods
The present investigation was carried out during 2012-13, at the Indian Agricultural Research Institute, Regional Station, Katrain, Kullu, Himachal Pradesh-175129. The experimental material consisted of 7 Asiatic lily cultivars and 10 intervarietal Asiatic lily hybrids developed at the station. Ten bulbs of commercial size (12-14 cm) from each cultivar and hybrid were planted in three replications. Bulbs were planted in 1 m wide beds, 15 cm a part rows, 10 cm depth and 20 cm between rows. The experiment was planned under low cost polyhouse fabricated with UV stabilized polyfilm as a cladding material at the top. All the selected genotypes were given uniform management practices for healthy growth and development. The biometrical observations were recorded on 5 random competitive plants from each replication after discarding side plants. The observations were recorded on various vegetative and floral traits, viz. days to sprouting, days to bud formation, days to flowering, plant height, tepal diameter (cm), number of spots/cm2, number of shoots per plant, number of leaves per plant, stem diameter (cm), flower diameter (cm), number of flowers per stem and inflorescence length (cm). Analysis of variance was performed following the standard procedures. The phenotypic and genotypic coefficient of variation were calculated according to Burton and De Vane (1953), heritability, genetic advance, genetic gain and correlation coefficients were calculated according to the formulae of Johnson et al. (1955). Leaf area was estimated by the formula as suggested by Erwin et al. (1991)
2 Results and Discussion
Analysis of variances revealed that highly significant differences existed among lilium genotypes for all the growth and flowering traits (Table 1) indicating that this variability among different traits would be useful for considerable improvement in this valuable crop. Results have shown that a single genotype was not superior for all traits (Table 2). Hence, different superior genotypes were identified, which were having ideal vegetative and floral traits. Earliest sprouting was recorded in genotype PKLH-8 (15.9 days) followed by PKLH-3 (17.1 days). Number of days taken to bud formation and flowering is an important trait in lilium because early and late flowering genotypes may be useful for regular availability of flowers. The desirable genotypes for early flowering were PKLH-2 (142.8 days) followed by PKLH-6 (150.0 days) and PKLH-1 (151.3 days). Based on days taken to flowering- the lilium genotypes were classified as early (=142.0 days), mid (>142.0 to 155.0 days) and late (> 155.0 days) flowering types. PKLH-6, PKLH-3, PKLH-8, PKLH-11, Pollyanna, Prato, Navona and Stargater were categorised as mid-season flowering type. Eight genotypes were late flowering type and only one genotype i.e. PKLH-2 was classified as early flowering type. Accordingly the genotypes may be utilized for prolonging the blooming period. Flowering time in lilies is genetically controlled (McRae, 1980). Mynett (1985) documented that homozygotic in early flowering species, conditioned by multigenes of earliness, gave a cumulative effect of early and very-early flowering plants. He also reported that the genes responsible for late flowering are dominant and gave in the high majority of late flowering plants.

Table 1 Analysis of variance (ANOVA) for different traits in lily


Table 2 Mean performance of 17 Asiatic hybrid lily genotypes for various vegetative and floral traits

Plant height is the most important trait which determines the quality of cut lilium. The maximum plant height was recorded in Stargater (94.0 cm) and minimum in PKLH-3 (14.8 cm) (Table 2). Tepal diameter was maximum in Prato (8.7 cm) and minimum in PKLH-10 (5.1cm). Most of lilium hybrids have spots on their tepals. These spots are anthocyanin (Banba, 1967). As cultivars with no or fewer spots are in demand. Genetic basis of flower colour and spot formations are little understood because of the heterozygous genome structure of Asiatic hybrid lily. Number of spots varied from 0 to 9.3. Number of spots/cm2 was recorded maximum in PKLH-2 (9.3). Multiple sprouting of flower stalks is an attractive trait and considered useful for developing novel lily cultivars. In lilium this phenomena is known as “feathering”. The desirable genotypes with multiple sprouting are PKLH-2 (2.9) followed by PKLH-8 (1.9). Oomiya et al. (2005) also observed multiple shooting in ‘Li-9’, which is a hybrid between ‘Mona’ x L. concolor Salisb. var. pulchellum (Fisch) Baker; nevertheless, the factors controlling the trait have not been revealed. Number of leaves per plant varied form 23.6 (PKLH-3) to 79.3 (Sumplon). Leaf area was noticed maximum in PKLH-1 (36.04 cm2) followed by Prato (35.1 cm2). However, minimum leaf area was recorded in Shiraj (9.8cm2). The stem diameter is an important trait because it relates to the stem sturdiness. The desirable genotypes for more stem diameters are PKLH-3 (4.2 cm) followed by PKLH-2 (4.1 cm). Flower diameter ranged from 13.8 cm to 20.2 cm. Significantly large sized flowers were recorded in Stargater (20.6 cm) followed by Prato (20.2 cm) and PKLH-1 (19.3 cm). These results confirm that the flower diameter could serve as a varietal trait (De Hertogh, 1996).
The productivity of a cultivar is determined by the number of flowers per stem. The more flowers on a stem, the more attractive are the plant with a longer flowering duration. The desirable genotypes for more number of flowers were PKLH-8 (7.3), PKLH-13 (7.1) and PKLH-2 (7.0). Besides this, inflorescence length is also an important trait which determines the number of flowers produced per stem. The inflorescence length was recorded maximum in PKLH-2 (22.3 cm) and minimum in PKLH-3 (11.2 cm).
Variability for various vegetative and floral traits is attributed to genetic make-up of the genotypes. Hence, these diverse genotypes with superior traits could be involved in the hybridization programme for assembling of desirable traits. Phenotypic co-efficient of variation was higher than genotypic co-efficient of variation for all the observed traits indicating that genotypic expression was superimposed by the environmental influence (Table 2&3). However, high genotypic (86.75) and phenotypic (86.85) co-efficient of variation were found for number of spots/cm2 followed by plant height (GCV=36.85, PCV=37.09) and leaf area (GCV=34.9, PCV=35.76). Grassotti et al. (1990) and Balode (2010) also reported higher phenotypic variability for plant height in lilium. Higher genotypic co-efficient of variation for above mentioned characters can be effectively utilized in future breeding programme. Singh and Sen (2000) suggested that if the value of phenotypic co-efficient of variation is greater than genotypic co-efficient of variation, the apparent variation is not only due to genotypes, but also due to influence of environment and hence selection may be misleading.
The estimates of phenotypic and genotypic co-efficient of variance showed narrow difference for days to bud formation, days to flowering and number of leaves per plant indicating that these traits were least influenced by environment and also indicating that phenotypic variability could be a reliable measure of genotypic variability (Table 3). Bhatia et al. (2013) have also reported low PCV and GCV for days to flowering in tulip. This facilitates for direct selection for improving the performance of specific traits. Genotypic co-efficient of variation helps in measurement of the range of genetic diversity in a character and provide means to compare the genetic variability in the quantitative characters. The GCV along with heritability estimates provides a better picture of the amount of genetic advance to be expected by phenotypic selection (Burton, 1952). Heritability estimates give a measure of transmission of characters from one generation to another, thus giving an idea of heritable portion of variability and enabling the plant breeder in isolating the elite selection in the crop. Heritability and genetic advance
increase the efficiency of the selection in a breeding programme by assessing the influence of environmental factors and additive gene action. The heritable portion of variability was thus determined with the help of broad sense heritability. The broad sense heritability estimates were recorded high for almost all the characters and it ranged from 75.3% (stem diameter) to 99.7% (number of spots /cm2) (Table 4). Heritability estimates were very high (>90%) for days to sprouting, days to bud formation, days to flowering, plant height, tepal diameter, number of spots/cm2, number of leaves per plant and leaf area. Such high estimates of heritability have been found useful in making selection of superior genotypes on the phenotypic performance with respect to quantitative traits. These results on broad sense heritability corroborate the findings of Sestra et al. (2007) and Bhatia et al. (2013) in tulip. Stem diameter, inflorescence length, number of flowers per stem, flower diameter, and number of shoots per plant exhibited moderate to low heritability. The low heritability estimates for these characters might be due to the predominance of genotype x environment interaction.

Table 3 Estimates of mean, range and co-efficient of variation for different vegetative and floral traits in lily

GCV and heritability in broader sense are not sufficient to determine the amount of variations which are heritable. The heritable variations could be determined with a greater degree of accuracy when heritability is studied in conjunction with genetic advance. Johnson et al. (1955) reported that heritability estimates together with expected genetic gain are more reliable than either of these parameters alone in predicting the resultant effects of selecting the best individuals and therefore, the genetic advance should be considered along with heritability in streamlining the coherent selection in breeding programme. Genetic advance was estimated at 2.06 % selection intensity and converted into expected genetic gain as percent of mean. Most of the characters exhibited moderate to high genetic advance as percent of mean (Table 4). The genetic advance varied from 9.64 for days to flowering to 95.53% for plant height. Low genetic advance for days to flowering was also reported by Bhatia et al. (2013) in tulip. The high (>50%) estimates of GA were observed for plant height, number of spots/cm2, number of leaves per plant, leaf area, number of flowers/plant and inflorescence length. Most of these traits also exhibited high heritability. Bhatia et al. (2013) also observed similar results for wrapper leaf area, spike length and plant height in tulip.

Table 4 Estimates of genetic parameters for vegetative and floral traits in lily genotypes

High heritability with high genetic advance indicate that the trait is governed by the additive gene action. Selection on the basis of these characters would be more effective for the improvement of lilium. Days to sprouting, tepal diameter, number of shoots per plant and stem diameter exhibited high heritability with moderate genetic advance indicating presence of dominant and epistatic genes and these traits can be improved through hybridization (Kumar et al. 2012). Similar results for heritability and genetic advance were obtained earlier in tulip by Jhon et al. (2006) and Balamurugan et al. (2002) in gladiolus. High heritability with low genetic advance gain was observed only for days to flowering indicating that the trait is governed by the non-additive gene action and for this selection with adequate progeny testing is practiced for improvement. High heritability with low genetic gain for day to flowering has also reported by Bhatia et al. (2013) in tulip.
The estimates of genotypic correlation in general were higher than the phenotypic correlation, indicating the presence of inherent association between various characters (Table 5). The study showed a positive and significant correlation of days to sprouting with days to bud initiation, tepal diameter, leaf area and inflorescence length and negative correlation with number of shoots per plant, stem diameter and number of flowers per plant. Significant and positive correlation was also observed between days to bud initiation and days to flowering. Similarly, days to flowering had a positive and significant correlation with flower diameter. These results are in agreement with the findings of Kumar and Tewari (2003) in lilium. Ohkawa (1977) and Roh (1985) also demonstrated that a delay in flowering correlates with an increase in leaf number in a variety of L. speciosum rubrum and in ‘Kiyotsubeni’.

Table 5 Correlation matrix for biometrical characters in lilium

Plant height had a positive and significant correlation with tepal diameter, number of spots/cm2, number of leaves, leaf area, flower diameter, number of flowers per plant and inflorescence length. Therefore, it was evident that plant height, an important character for cut flower production; could be increased with the increase in any of these characters. Bhatia et al. (2013) also observed significant positive correlation for scape length with number of leaves/plant wrapper leaf area and flower size in tulip. Positive and significant correlation was observed between tepal diameter and number of spots per flower, leaf area, flower diameter and inflorescence length. Number of shoots per plant was also positively and significantly correlated with stem diameter and number of flowers per stem. Significantly positive correlation was found between leaf area with flower diameter and inflorescence length. Flower diameter showed highly significant and positive correlation with inflorescence length. It is evident from this association that cut flower quality parameters of lilium such as inflorescence length can be improved simultaneously with the improvement in flower diameter.
The present investigation concluded that the existence of a wider variation for various vegetative and floral traits in 17 genotypes of lilium will be helpful for future to be breeding programmes. The commercially important traits like plant height, number of leaves, leaf area, exhibited high heritability coupled with high genetic advance signifying additive gene action. Hence, selection would be more effective for the improvement of these traits. Plant height exhibited a high degree of significant positive correlation with tepal diameter, number of leaves/plant, leaf area, flower diameter, number of flowers/plant and inflorescence length, so a direct selection from germpalsm lines may be effective for improvement for improving the cut flower quality.
Asano Y., 1989, Lilium L. In: The Grand Dictionary of Horticulture, Vol. 5. Edited by Tsukamoto, Y. pp. 198-199. Syogakukan, Tokyo (in Japanese)
Banba H., 1967, Pigments of lily flowers 1. Survey of anthocyanin (in Japanese), Journal of Japanese Society for Horticultural Science, 36:61-65
Balamurugan., Rengasamy P., and Arumugam., 2002, Variability studies in gladiolus, Journal of Ornamental Horticulture, 5 (1): 38-39
Balode A., 2010, Phenotypic analysis of hybrids and their parents in lilium spp. breeding, Annual 16th International Scientific Conference Proceedings. Research for Rural Development, 1: 65-70
Bhatia R., Dhiman M. R., Chander Parkash., and Dey S. S., 2013, Genetic variability and character association in tulip (Tulipa gesneriana) for various quantitative traits, Indian Journal of Agricultural Sciences, 83 (7): 773-780
Burton G. W., and Devane E. H,. 1953, Estimation of heritability in tall fescue from replicated clonal material, Agronomy Journal, 45:478-487
Burton G. W., 1952, Quantitive inheritance in grasses, Proceedings of 6th International Grassland Congress, pp. 277-283
Comber H. F., 1949, A new classification of the genus Lilium. In Lily Yearbook, Vol. 13. pp. 85 – 105, Royal Horticultural Society, London
DeFraaff J., and Hornback E., 1967, Origin of the mid-century hybrids, The Royal Horticultural Society, The lily year book,30:26-29
De Hertogh A. A., 1996, Marketing and research requirements for Lilium in North America, Acta Horticulture, 414: 17-24
Erwin J. E., Heins R. D., and Moe R., 1991, Temperature and photoperiod effects on Fuchsia x hybrida morphology, Journal of American Society for Horticultural Science, 116(6): 955-960
Grassotti A., Torrini F., Mercuri A., and Schiva T., 1990, Genetic improvement of lilium in Italy, Acta Horticulture, 266: 339-347
Jhon A. Q., Khan F. U., and Rather Z. A., 2006, Genetic variability studies in tulip, Applied Biological Research, 8 (1-2): 37-39
Johnson H. W., Robinson H. F., and Comstock R. E., 1955, Estimate of genetic and environmental variability in soybean, Agronomy Journal, 46: 314-318
Kumar D., and Tewari G. N., 2003, Correlation studies in Asiatic hybrid lily, Journal of Ornamental Horticulture, 6(3): 268-270
Leslie A.C., 1982, The International Lily Register, 3rd edn., The Royal Horticultural Society, London.
McRae Freedman J., 1980, Genetics for lily breeder, The Lily Yearbook of the North American Lily Society, Queen Lily Printers Inc., Burlington, Vermont, 33: 98-110
Mynett K., 1985, Breeding for early flowering of Asiatic lilies, Acta Horticulture, 177: 393-397
Ohkawa K., 1977, Studies on the physiology and control of flowering in Lilium speciosumrubrum, Special Bulletin of the Kanagewa Horticultural Experimental Station, pp.1-73
Oomiya T., Tamagake H., Tsutsui S., and Ubukata M., 2005, A new cultivar of the Asiatic hybrid lily, ‘Li-9’, with multiple scapes and small flowers, Acta Horticulture, 673: 559-622
Roh M. S., 1985, Flowering response of Mid-Century hybrid lilies to bulb vernalization and shoot photoperiod treatment, Horticultural Science, 20:710-713
Sestra R., Mihalte L., Sestra A., Bondrea I., and Baciu A., 2007, The variability and heritability of several traits at different cultivars of tulips, Buletinul Usamv-CN, 64:1-2
Singh D., and Sen N. L., 2000, Genetic variability, heritability and genetic advance in marigold, Journal of Ornamental Horticulture, 3(2): 75-78
Smyth D. R., Kongsuwan K., and Wisudharomn S., 1989, A survey of C-band patterns in chromosomes of Lilium (Liliaceae), Plant Systematic and Evolution, 163: 53-69

Wadekamper J., 1977, “Connecticut King” as a parent in lily hybridizing, The lily yearbook N.A.L.S, 30: 115-118

Molecular Plant Breeding
• Volume 6
View Options
. PDF(355KB)
. Online fPDF
Associated material
. Readers' comments
Other articles by authors
pornliz suckporn porndick pornstereo . M. R. Dhiman
. Chander Parkash
. Raj Kumar
. M S Guleria
. Munish Dhiman
Related articles
. Genetic variability
. Heritability
. Correlation
. Lilium
. Email to a friend
. Post a comment