Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Natural Variation in Petal Color in Lycoris longituba Revealed by Anthocyanin Components

  • Qiuling He,

    Affiliations College of Forest Resources and Environment, Nanjing Forestry University, Nanjing, People's Republic of China, Department of Agronomy, College of Agriculture and Biotechnology, Zhejiang University, Hangzhou, People's Republic of China

  • Ye Shen,

    Affiliation College of Forest Resources and Environment, Nanjing Forestry University, Nanjing, People's Republic of China

  • Mingxiu Wang,

    Affiliation College of Forest Resources and Environment, Nanjing Forestry University, Nanjing, People's Republic of China

  • Minren Huang , (MH); (LW); (RW)

    Affiliation College of Forest Resources and Environment, Nanjing Forestry University, Nanjing, People's Republic of China

  • Ruizhen Yang,

    Affiliation Beijing Botanical Garden, Institute of Botany, Chinese Academy of Sciences, Beijing, People's Republic of China

  • Shuijin Zhu,

    Affiliation Department of Agronomy, College of Agriculture and Biotechnology, Zhejiang University, Hangzhou, People's Republic of China

  • Liangsheng Wang , (MH); (LW); (RW)

    Affiliation Beijing Botanical Garden, Institute of Botany, Chinese Academy of Sciences, Beijing, People's Republic of China

  • Yanjun Xu,

    Affiliation Department of Applied Chemistry, China Agricultural University, Beijing, People's Republic of China

  • Rongling Wu (MH); (LW); (RW)

    Affiliations Center for Computational Biology, Beijing Forestry University, Beijing, People's Republic of China, National Engineering Laboratory for Tree Breeding, Beijing Forestry University, Beijing, People's Republic of China, Key Laboratory of Genetics and Breeding in Forest Trees and Ornamental Plants, Beijing Forestry University, Beijing, People's Republic of China

Natural Variation in Petal Color in Lycoris longituba Revealed by Anthocyanin Components

  • Qiuling He, 
  • Ye Shen, 
  • Mingxiu Wang, 
  • Minren Huang, 
  • Ruizhen Yang, 
  • Shuijin Zhu, 
  • Liangsheng Wang, 
  • Yanjun Xu, 
  • Rongling Wu


Lycoris longituba is one of the species belonging to the Amaryllidaceae family. Despite its limited distribution, endemic to central eastern China, this species displays an exceptionally wide diversity of flower colors from purple, red, orange, to yellow, in nature. We study the natural variation of floral color in L. longituba by testing the components of water-soluble vacuolar pigments – anthocyanins – in its petals using high-performance liquid chromatography coupled with photodiode array detection and electrospray ionization mass spectrometry. Four anthocyanins were identified, cyanidin-3-sophoroside (Cy3So), cyanidin-3-xylosylglucoside (Cy3XyGlc), cyanidin-3-sambubioside (Cy3Sa), and pelargonidin-3-xylosylglucoside (Pg3XyGlc), which occur at various amounts in L. longituba petals of different colors. A multivariate analysis was used to explore the relationship between pigments and flower color. Anthocyanins have been thought to play a major role in acting as a UV screen that protects the plant's DNA from sunlight damage and attracting insects for the purpose of pollination. Thus, knowledge about the content and type of anthocyanins determining the petal coloration of Lycoris longituba will help to study the adaptive evolution of flowers and provide useful information for the ornamental breeding of this species.


Endemic in China, Lycoris longituba (Y. Hsu et Q.J. Fan), a species from the Amaryllidaceae family, is characterized by a few biological features that are interesting to evolutionary studies. First, this species is naturally distributed in a limited range, including the Langya Mountain of Anhui Province, the Baohua Mountain and Xuyi City of Jiangsu Province, all about 50 km from one another [1], but it displays an incredible diversity of floral colors. As shown in Fig. 1, four distinct representatives of floral colors, purple, red, orange, and yellow, commonly occur in L. longituba. Such great variability provides an excellent model system for identifying the biochemical determinants of flower colors and explaining these determinants in terms of plant ecological and evolutionary processes [2]. Second, leaves of L. longituba sprout in spring and wither in early summer, followed by flowering. These discrete phases of vegetative growth and reproduction, a rare phenomenon in angiosperms, may offer an unprecedented opportunity to study the evolution of fitness traits in plants. In addition, L. longituba has many ornamental and medical values and can be potentially interesting to plant breeding.

Figure 1. Flowers of four representative colors, purple (A), red (B), orange (C), and yellow (D), for Lycoris longituba variants sampled from a natural population.

As part of the ongoing Lycoris longituba evolutionary genetics project, we attempted to study the phytochemical components that form floral colors in this species and further use these components to interpret the mechanism of why there is a diversity of flower colors within such a limited range of distribution. Floral colors are due to biological pigments including a variety of different kinds of molecules, such as porphyrins, carotenoids, anthocyanins and betalains [3]. These biological pigments absorb some wavelengths of light and reflect others. The absorbed light may be used by the plant to perform chemical reactions, while the reflected wavelengths of light determine the color of the flower.

Anthocyanin-type pigments, found only in terrestrial plants, are thought to be evolutionarily adaptive for several different plants processes [4][6]. In photosynthetic tissues, such as leaves and sometimes stems, anthocyanins serve as a sunscreen blocking blue-green and UV light, thus protecting cells from excessive light damage and the tissues from photoinhibition or excessive light stress [7], [8]. In flowers, anthocyanins may appear red, purple, or blue, according to pH values [9], [10], which helps in attracting pollinators [11]. In fruits, the colorful skins derived from anthocyanins also attract animals to eat the fruits and, therefore, disperse the seeds [12]. Because of their significant roles in plant function, anthocyanins have been used to study the molecular genetic basis of variation and evolution in morphological traits [11] and potentially used as an indirect trait for plant breeding [6].

There has been a wealth of literature on determining anthocyanin components in the Amaryllidaceae family. For example, four anthocyanin components, pelargonidin-3-glucosides (Pg3G), pelargonidin-3-xylosylglucosides (Pg3XyGlc), cyanidin-3-glucosides (Cy3G), and cyanidin-3-xylosylglucosides (Cy3XyGlc) were identified in L. radiata (L'Her.) Herb [13]. In Nerine, a genus of the Amaryllidaceae family, Arisumi and Shioya [14] detected Cy3G, Pg3G, cyanidin-3,5-diglucoside (Cy3G5G), peonidin-3,5-diglucoside (Pn3G5G), pelargonidin- 3,5-diglucoside (Pg3G5G), and cyanidin-3-sophoroside (Cy3So). In the flowers of Hippeastrum, cyaniding-3-O-(6″-O-α-rhamnopyranosyl-β-glucopyranosides) and pelargonidin-3-O- (6″-O-α-rhamnopyranosyl-β-glucopyranosides) were detected, with Pg3G existing as a minor component [15], [16].

In this study, we hypothesized that the floral flowers of L. longituba are assocated with the components of anthocyanins, a group of water-soluble vacuolar pigments, which can be measured readily by using high-performance liquid chromatography, coupled with a photodiode array detection and electrospray ionization mass spectrometry (HPLC–ESI–MSn) analysis.

We measured anthocyanin components in different colors of flowers derived from 44 variants of L. longituba. These variants were sampled randomly from a population in the natural distribution of this species. Four major components in the petal extracts, Cy3So, Cy3XyGlc, cyanidin-3-sophoroside (Cy3Sa), and Pg3XyGlc, were identified. Through a multivariate analysis for the correlations between color and anthocyanin contents, we suggested that different amounts of these anthocyanin components are important determinants for the natural variation in flower colors for L. longituba.


Petal Colors

Of the 44 variants sampled, one has white flowers, which is viewed as an ancestral type of L. longituba [17]. Floral colors were quantified by the Royal Horticultural Society Colour Chart (RHSCC) [18], with RHSCC values ranging from 4D to 159C except for the white. A software package called CIE 1976 L*a*b* (CIELAB) was used to measure different aspects of a flower color. CIELAB is a one-color system that takes into account all aspects of the color described by L*, a*, and b* parameters. The L* describes the lightness of the color, ranging from black (L* = 0) to white (L* = 100). The a* negative and positive are for green and red, and the b* negative and positive are for blue and yellow, respectively [19]. The color coordinates measured were shown as follows: a* values ranged from −6.99 to 40.87, b* values from −14.36 to 56.71, and the L* values from 45.95 to 93.03 (Fig. 2).

Figure 2. Distribution of Lycoris longituba variants ants by flower colors based on a bivariate (a* and b*) (A) and trivariate (a*, b*, and L*) CIELAB data (B), respectively.

The flower colors were identified by the RHSCC value.

Using Tian's approach [20], we derived two new parameters from a* and b*, which are Chroma [C* = (a*2+b*2)1/2] and hue angle [h = arctan b*/a* ]. The Chroma parameter describes the saturation of the color, a measure of how far the color is from the grey tone. The higher the C* value, the more saturated is the color. The hue angle parameter describes the hue of the color, i.e. color tonalities (red, yellow, blue, etc.) [21], [22]. A red color has h around 0° while yellow is described by an h around 90°. Thus, h = 45° corresponds to orange [15], [19]. Based on the hue angles measured, 43 variants were classified into four groups in terms of floral colors (Fig. 1), purple (14), red (10), orange (9), and yellow (10) (Table 1). All the groups have a high L* value (except for LL105). However, L* values of the orange group were lower than those of the other groups, with the maximum value of 64.60 in the orange group, compared with 92.69 in the yellow group.

Table 1. Petal colors and color parameters of 44 Lycoris longituba variants.

Anthocyanin Identification

Figure 3 shows an HPLC–ESI–MSn analysis of anthocyanins in L. longituba variants. UV–vis detection (by light wavelength 516 nm) showed the presence of two major and three minor anthocyanins. The change of the mobile phase helped the separation of peak 1 from peaks 2, 3, 4 and 5. Chemical structures of anthocyanins were confirmed by HPLC–ESI–MSn. The HPLC retention time (Rt), elution order, and their UV–vis spectral properties together with mass spectrometric data were shown in Table 2. Peaks 1, 2, and 3 all produced a major fragment at m/z 287, indicating that they were cyanidin (Cy) derivatives (Fig. 3). They exhibited molecular ions at m/z 611 and 581. The cyanidin anthocyanin with a molecular ion of m/z 611 detected by the HPLC–ESI–MSn analysis was identified as cyanidin 3-sophoroside (Fig. 4A), because this compound only produced one major fragment at m/z 287 [23]. Similarly, peak 3 was identified as Cy3Sa (Fig. 4C). The molecular ion of m/z 581 was fragmented into two major fragments at m/z 287 and 449 in peak 2 identified as Cy3XyGlc (Fig. 4B). The minor anthocyanin (m/z 565, peak 4 identified as Pg3XyGlc in Fig. 2) was identified as a pelargonidin anthocyanin, which was a precursor of m/z 271 (Fig. 4D). The molecular ion at m/z 565 was fragmented at m/z 271 and 433. In addition, the identities of Cy3So and Cy3Sa were further confirmed using their respective standards. Peak 5 produced a major fragment at m/z 303, but its identity was not confirmed because of sample limitation. The amounts of various components of anthocyanins measured for individual L. longituba varieties are given in Table 3.

Figure 3. The HPLC chromatogram of Lycoris longituba petal extracts detected by light wavelength = 516 nm.

Peak labels: (1) cyanidin 3-sophoroside (Cy3So); (2) cyanidin 3-xylosylglucoside) (Cy3XyGlc); (3) cyanidin 3-sambubioside (Cy3Sa); (4) pelargonidin 3- xylosylglucoside (Pg3XyGlc); (5) Unidentified. Two colors, red and blue, denote the HPLC chromatograms of two different L. longituba varieties, LL52 and LL27, respectively.

Figure 4. The HPLC–ESI/MS/MS of anthocyanins in Lycoris longituba.

Mass spectra of anthocyanin components. (A) cyanidin 3-sophoroside, (B) cyanidin 3-xylosylglucoside, (C) cyanidin 3-sambubioside, and (D) pelargonidin 3-xylosylglucoside.

Table 2. Chromatographic and spectral data of anthocyanins detected in 44 Lycoris longituba varieties.

Table 3. Petal anthocyanin mean values (in peak areas %) of 44 Lycoris longituba varieties.

Relationships between Color Parameters and Anthocyanin Compositions

Relationship between C* and L* values.

In all the groups, except for yellow, the L* value was negatively correlated with the C* value for all groups (Fig. 5), with a steeper slope for purple, followed by red, orange and yellow. This suggests that the less bright the flower, the more saturated the flower color.

Figure 5. The multivariate relationships between C* and L* values, showing different patterns depending on petal colors, purple, red, orange, and yellow (see Fig. 1), in Lycoris longituba flowers.

Relationship between L* value and the amount of total anthocyanins (TA).

There was the same tendency of a negative correlation between these two variables in red, purple, and orange groups. The L* value increased as the TA amount decreased. It means that the brighter the flower petals, the less the TA amount. The coefficient of determination (R) of the red group (0.92) was much larger than those of purple (0.64) and orange groups (0.58). In white and yellow groups, the TA amounts were nearly zero.

Relationship between a* value and TA.

The red and orange groups showed a positive correlation (R = 0.93 and 0.49, respectively), whereas the correlation between a* value and TA was not found in the purple group. In general, the a* value increased as TA increased.

Relationship between a* value and CI.

The co-pigmentation effect occurs when the TF/TA ratio (co-pigmentation index, CI) exceeds 5 [24]. This phenomenon was observed in all the groups. The red and orange groups showed a negative correlation between a* value and CI. Since the TA values of the white and yellow groups were zero, they were excluded from the CI analysis.

Multiple linear regression (MLR) analysis of color parameters and anthocyanin compositions.

The MLR analysis chooses the following equations that best fit the relationships between color parameters and anthocyanin components (p<0.01):

From this MLR analysis, it can be seen that the Cy derivatives were the major factors that affect color parameters. Cy3XyGlc and Cy3So had positive effects on the value of a*, whereas Cy3So and Pg3XyGlc had negative effects on the value of b*. TA was the key factor that reduces the L* value. In addition, Cy3Sa had a positive effect on the value of L*.


Lycoris longituba, an annual species, is characteristic of a few unique biological features [1]. For example, a distinct diversity of floral colors pervades its narrow natural distribution. In general, the flower colors of this species are grouped into four types, purple, red, orange, and yellow (Fig. 1). In this study, we studied the natural variation of floral color in L. longituba by quantifying the components of anthocyanins in flower petals of each type. As water-soluble vacuolar pigments, anthocyanins are an important phenotype for studying the evolution of floral colors and may be used as an indirect trait to select elite genotypes that are optimally adaptive to changing environment [4][6], [9][12]. Through tests and comparsions, we suggest that differences in anthocyanin type and quantity were responsible for diverse floral colors of L. longituba. This result may not only provide a biochemical basis for floral colors, but also helps to understand the adaptation and evolution of floral colors to changing environment.

By using high-performance liquid chromatography, coupled with photodiode array detection and electrospray ionization mass spectrometry, we identified four major types of anthocyanin, cyanidin-3-sophoroside (Cy3So), cyanidin-3-xylosylglucoside (Cy3XyGlc), cyanidin-3-sambubioside (Cy3Sa), and pelargonidin-3-xylosylglucoside (Pg3XyGlc), in the petal extracts of L. longituba. To our best knowledge, Cy3So and Cy3Sa have not been reported in any species of the Lycoris genus, but they were found to be an important determinant of floral colors in L. longituba. Cy3XyGlc had a wide distribution in purple, red, and orange varieties, occurring in more than 70% variants. Cy3So occurs in all but one sample, accompanied by Pg3XyGlc in the purple group. Furthermore, Pg3XyGlc was only detected in the purple group. But in the red group, no Cy3So was detected. We found that the major flower flavonols of L. longituba were 3-xylosylglucoside of kaempferol and quercetin (Km3XyGlc and Qu3XyGlc) (Q. L. He, unpublished data). This suggests that the same glycosylation process occurred for flavonols and anthocyanins.

By estimating and using quantitative color parameters with color measuring systems, we are able to interpret the tremendous variation of floral color in L. longituba naturally distributed within a small range. It is suggested from our study that a wide array of floral colors is mainly due to the chemical structures of different anthocyanins accumulated in the flower. The correlations between color parameters, L* and C* (Fig. 5), L* and TA value, a* and CI value, showed negative tendencies in all the groups except for yellow. An inverse tendency was detected between a* and TA in red and orange groups. A correlation analysis indicates that the L* value increased with increasing proportions of Cy3XyGlc and Cy3Sa. These results can be integrated into an evolutionary study to study the ecological and evolutionary relationship of floral colors with environmental factors [2]. In practice, these also help to develop appropriate criteria for breeding favorable colors through indirect selection.

The type and content of anthocyanins are important determinants for flower coloration. All four anthocyanins occur in most variants from the purple group, whereas only three (Cy3So, Cy3XyGlc, and Cy3Sa) exist in some orange variants and only two (Cy3XyGlc and Cy3Sa) exist in all red variants. Thus, the in vivo colors of L. longituba seem to be correlated with the anthocyanin types in the petals, although there were some exceptions. In some variants, Cy3XyGlc is the sole anthocyanin, but their petals display orange and purple-red colors. In a study of tulips, Nieuwhof et al. [25] found that yellow and orange colors of petals may be due to carotenoids. The petals of some plants have a capacity to modify carotenoid biosynthesis, producing an orange-red color [26]. Astaxanthin is a ketocarotenoid that is produced in a few plants such as Adonis aestivalis; it furnishes an attractive orange-red color [27]. The orange petals of calendula (Calendula officinalis) contain reddish carotenoids that are absent in yellow petals [28]. These species contain very few types of anthocyanins, and their yellow, purple and orange colors indeed reveal that other factors than anthocyanins (carotenoids, co-pigmentation, pH, etc.) may influence the flower color [3]. However, carotenoids are highly unstable and difficult to extract and measure in practice [29]. Taken together, we suggest that anthocyanins can be practically more useful to explain the floral colors of L. longituba and their tremendous diversity in nature and can further be used as an indirect trait to select elite genotypes in flower color.

L. longituba has a limited range of distribution, but is characterized by a high diversity of floral colors. A next step for studying this species should be to understand the pattern of genetic diversity using molecular markers and the evolution of phytochrome function operated by anthocyanins [2]. More recently, we have exploited an array of informative microsatellite markers [30] and also developed a series of statistical models for estimating and testing multiallelic Hardy-Weinberg disequilibria and multiallelic linkage disequilibria with multilocus data [31]. By associating with marker polymorphisms with floral colors, we can identify significant genetic variants that contribute to flower variation, shedding light on the genetic architecture of color formation in L. longituba.

Materials and Methods

Plant Material and Petal Color Measurement

Flowers of 44 L. longituba variants were collected from the Lycoris plantation of Nanjing Forestry University in Nanjing, China. Three petal samples of each variant were collected at a full-bloom stage from late July to early August in 2008. The 44 variants were divided into four groups according to flower color – purple, red, orange, and yellow.

The colors of fresh petals of the 44 variants were first compared with the Royal Horticultural Society Colour Chart (RHSCC) [17] (Royal Horticultural Society 2001), and the surface color of fresh flowers were also analyzed with a spectrophotometer NF333 (Nippon Denshoku Industries Co. Ltd., Japan) using the CIELAB system at CIE C/2° illumination/viewer condition.

Quantification of Anthocyanin Components

Petals of each sample (approximate 100–150 mg dry weight) were extracted with 5 mL 70% methanol aqueous solution containing 0.1% HCl (pH 2.08) in a test tube at 4°C in the dark for 24 h and shaken in a Gene-2 vortex (Scientific Industries INC, USA) every 6 h [32]. The liquid was separated from the solid mixture by filtration through a sheet of qualitative filter paper (Hangzhou Xinghua Paper Industry, Zhejiang, China). The filtrate passed through 0.2 µm membrane filters (Membrana, Wuppertal, Germany). This experiment was performed three times for each cultivar and the extracts from each experiment were separately used for identification and quantification of anthocyanins.

Chromatographic analysis was carried out on a HPLC (Agilent Corporation, Waldbronn, Germany) equipped with a G1311A HPLC Pump, Thermostatted Column Compartment G1316A and a Diode-Array G1315B Detector. The HPLC column was ZORBAX Eclipse XDB-C18 (250 mm×4.6 mm i.d.; 5 µm) (Agilent Corporation, Sunnyvale, CA, USA) and was protected with a Micro-Guard Cation-H Cartridge (Bio-RAD Laboratories, Hercules, CA, USA). A volume of 10 µL sample solution was injected for HPLC analysis. Flow rate was 0.8 mL/min and the mobile phase A consisted of 10% (v/v) formic acid in water (10∶90; v/v, HCOOH∶H2O), while CH3CN∶CH3OH (85∶15, v/v) was used as the mobile phase B [33]. The column was eluted at 25°C using linear gradients which was 5% B at 0 min, 15% B at 15 min and 90% B for another 6 min. Data were recorded on a computer with the ChemStation software (Agilent Corporation, Waldbronn, Germany). Detection was carried out at 516 nm for anthocyanins and 350 nm for flavones and flavonols, the photodiode array spectra were recorded between 200 and 800 nm. All determinations were performed with three replicates.

Anthocyanins were identified by HPLC–ESI–MSn (Agilent Technologies, Palo Alto, CA, USA). The interface was an ESI–electrospray (Agilent 1100 LC/MSD Trap VL) operating in positive ionization mode monitoring of the protonated and deprotonated molecular ions at the following operating parameters: the operating parameters in positive mode were as follows: gas (N2) temperature, 350°C; flow rate, 6.0 L/min; nebulizer pressure, 241.3 kPa; HV voltage, 3.5 kV; octopole RF amplitude, 150 Vpp; skim 1 voltage, 55.6 V; skim 2 voltage, 6.0 V; capillary exit, 140 V; cap exit offset, 84.4 V, and scan range, m/z 100–1000 units.

The amount of total anthocyanins (TA) and total flavones and flavonols (TF) were calculated in milligrams per 100 mg dry weight (as a quantity of Mv3G5G mg/100 mg, and as a quantity of Rutin mg/100 mg, respectively). Co-pigmentation index (CI) was calculated by TF/TA [34].

Statistical Analysis

Correlations between petal color parameters and anthocyanin compositions of individual varieties were analyzed using a multiple linear regression (MLR) (SAS 8.1 software). Color parameters including L*, a*, and b* values were considered as dependent variables, and the contents of individual anthocyanins and the total anthocyanin content were considered as independent variables.


We are grateful to Huiyu Zhou (Professor, Department of Botany, Nanjing Forestry University, China) for identification of all the varieties. We also would like to thank Shaojie Cui, Lili Shen, Sheng Zhu and Linnan Ma for their help in collecting plant materials and chemical analysis and Arthur Berg for redrawing Figs. 2 and 5.

Author Contributions

Conceived and designed the experiments: MH LW MW. Performed the experiments: QH YS RY SZ YX. Analyzed the data: QH RW. Contributed reagents/materials/analysis tools: QH YX. Wrote the paper: QH RW.


  1. 1. Deng CL, Zhou J, Gao WJ, Sun FC, Qin RY, et al. (2006) Assessment of genetic diversity of Lycoris longituba (Amaryllidaceae) detected by RAPDs. J Genet 85: 205–207.
  2. 2. Smith H (2000) Phytochromes and light signal perception by plants—an emerging synthesis. Nature 407: 585–591.
  3. 3. Bohren CF, Clothiaux E (2006) Fundamentals of Atmospheric Radiation: An Introduction with 400 Problems Wiley-VCH, Weinheim, Germany.
  4. 4. Hoch WA, Zeldin EL, McCown BH (2001) Physiological significance of anthocyanins during autumnal leaf senescence. Tree Physiol 21: 1–8.
  5. 5. Whittal JB, Voelckel C, Kliebenstein DJ, Hodges SA (2006) Convergence, constraint and there role of gene expression during adaptive radiation: floral anthocyanins in Aquilegia. Mol Ecol 15: 4645–4657.
  6. 6. Strauss S, Whittall JB (2006) Non-pollinator agents of selection on floral traits. In: Harder L, Barrett S, editors. Ecology and Evolution of Flowers. Oxford University Press. pp. 120–138.
  7. 7. Barritt BH, Torre LC (1973) Cellulose thin-layer chromatographic separation of Rubus fruit anthocyanins. J Chromatogr 75: 151–155.
  8. 8. Ancos B, Gonzalez E, Cano MP (1999) Differentiation of raspberry varieties according to anthocyanin composition. Z Lebensm Unters Forsch A 208: 33–38.
  9. 9. Hada H, Hidema J, Maekawa M, Kumagai T (2003) Higher amounts of anthocyanins and UV-absorbing compounds effectively lowered CPD photorepair in purple rice (Oryza sativa L.). Plant Cell Environ 26: 1691–1701.
  10. 10. Steyn W, Wand S, Holcroft D, Jacobs G (2002) Anthocyanins in vegetative tissues: a proposed unified function in photoprotection. New Phytol 155: 349–361.
  11. 11. Thomson JD, Wilson P (2008) Explaining evolutionary shifts between bee and hummingbird pollination: Convergence, divergence, and directionality. Intl J Plant Sci 169: 23–38.
  12. 12. Steyn WJ (2009) Prevalence and functions of anthocyanins in fruits. In: Anthocyanins K, Could , et al., editors. Springer Science and Business Media. New York, USA: 85 p. In.
  13. 13. Arisumi K (1971) Flower colors in Amaryllidaceae. II. Anthocyanin constitu- tion of Lycoris. Yamaguchi Daigaku Nogakubu Gakjustsu Hokoku 22: 171–180.
  14. 14. Arisumi K, Shioya H (1970) Flower colors in Amaryllidaceae. II. Anthocyanin constitution of Nerine. Yamaguchi Daigaku Nogakubu Gakjustsu Hokoku 21: 65–72.14.
  15. 15. Byamukama R, Jordheim M, Kiremire B, Namukobe J, Andersen ØM (2006) Anthocyanins from flowers of Hippeastrum cultivars. Scientia Hort 109: 262–266.
  16. 16. Hrazdina G (1988) Purification and properties of a UDPglucose:flavonoid 3-Oglucosyltransferase from Hippeastrum petals. Biochim. Biophys Acta 955: 301–309.
  17. 17. Hu SY (1976) The tour of a botanist in China. Arnoldia 35: 264–295.
  18. 18. Royal Horticultural SocietyBritish Colour Council (ed.) (2001) Royal Horticultural Society colour chart. Royal Horticultural Society, London, UK.
  19. 19. Gonnet JF (1998) Colour effects of co-pigmentation of anthocyanins revisted-Acolorimetric denition using the CIELAB scale. Food Chem 63: 409–415.
  20. 20. Tian QG, Giusti MM, Stoner GD, Schwartz SJ (2005) Screening for anthocyanins using high-performance liquid chromatography coupled to electrospray ionization tandem mass spectrometry with precursor-ion analysis, product-ion analysis, common-neutral-loss analysis, and selected reaction monitoring. J Chromatogr A 1091: 72–82.
  21. 21. Stintzing FC, Schieber A, Carle R (2003) Evaluation of color properties and chemical quality parameters of cactus juices. Euro Food Res Technol 216: 303–311.
  22. 22. White SA, Scoggins HL (2005) Fertilizer concentration affects growth response and leaf color of Tradescantia virginiana L. J Plant Nutr 28: 1767–1783.
  23. 23. Gonnet JF, Hieu H (1992) Insitu microspectrophotometric and microspectrophotocolorimetric investigation of vacuolar pigments in flowers of cultivars of carnation (Dianthus caryophyllus). J Hort Sci 67: 663–676.22.
  24. 24. Asen S, Stewart RN, Norris KH (1971) Co-pigmentation effect of quercetin glycosides on absorption characteristics of cyanidin glycosides and color of Red Wing azalea. Phytochemistry 10: 171–175.
  25. 25. Nieuwhof M, Eijk JP, Eikelboom W (1989) Relation between flower color and pigment composition of tulip (Tulipa L.). Neth J Agric Sci 37: 365–370.
  26. 26. Tanaka Y, Sasaki N, Ohmiya A (2008) Biosynthesis of plant pigments: anthocyanins, betalains and carotenoids. Plant J 54: 733–749.
  27. 27. Cunningham FXJ, Gantt E (2005) A study in scarlet: enzymes of ketocarotenoid biosynthesis in the flowers of Adonis aestivalis. Plant J 41: 478–492.
  28. 28. Kishimoto S, Maoka T, Sumitomo K, Ohmiya A (2005) Analysis of carotenoid composition in petals of calendula (Calendula officinalis L.). Biosci Biotechnol Biochem 69: 2122–2128.
  29. 29. Kamffer Z, Bindon KA, Oberholster A (2010) Optimization of a method for the extraction and quantification of carotenoids and chlorophylls during ripening in grape berries (Vitis vinifera cv. Merlot). J Agric Food Chem 58: 6578–6586.
  30. 30. He QL, Shen Y, Chen Y, Zhou Y, Berg A, et al. (2009) Development of 16 polymorphic simple sequence repeat markers for Lycoris longituba from expressed sequence tags. Mole Ecol Resour 9: 278–280.
  31. 31. Berg A, He QL, Shen Y, Chen Y, Huang MR, et al. (2010) Trilocus disequilibrium analysis of multiallelic markers in outcrossing populations. Stat Appl Mol Genet Biol 9(1): Article 16.
  32. 32. Yang RZ, Wei XL, Gao FF, Wang LS, Zhang HJ, et al. (2009) Simultaneous analysis of anthocyanins and flavonols in petals of lotus (Nelumbo) cultivars by high-performance liquid chromatography-photodiode array detection-electrospray ionization mass spectrometry. J Chromatogr A 1216: 106–112.
  33. 33. Buchert J, Koponen JM, Suutarinen M, Mustranta A, Lille M, et al. (2005) Effect of enzyme-aided pressing on anthocyanin yield and profiles in bilberry and blackcurrant juices. J Sci Food Agric 85: 2548–2556.
  34. 34. Wang LS, Hashimoto F, Shiraishi A, Aoki N, Li JJ, et al. (2001) Phenetics in tree peony species from China by flower pigment cluster analysis. J Plant Res 114: 213–221.