what colour of flowers do lepidoptera prefer for foraging?

Biologia 65/6: 1049—1056, 2010Section ZoologyDOI: 10.2478/s11756-010-0125-4

What colour of flowers do Lepidoptera prefer for foraging?

Selcuk Yurtsever, Zuhal Okyar & Necmettin Guler

Biology Department, Science Faculty, Trakya University, 22030 Edirne, Turkey; e-mail: [email protected]

Abstract: Food plant preferences of some Lepidoptera species associated with particular colour of the flowers were in-vestigated. Based on 1,329 field observations of 43 Lepidoptera and 66 plant species, Lepidoptera showed a high tendency(G-test, Gadj = 698.6, df = 6, P < 0.001) to use the yellow (29%) and pink (28%) coloured flowers for foraging. Comparedto the other colours it was evident that plants with red flowers (2%) were not preferred. Moreover, the plants with red(H = 0.435) and yellow-white (H = 0.543) flowers were not visited by diverse Lepidoptera species. Although yellow andpink flowers were most frequently visited, the highest degrees of the Lepidoptera diversity values were associated with theplants having blue (H = 0.647) and purple (H = 0.634) flowers. Species of Nymphalidae were most numerous (14 spp.) inthe study area and the members of this family were observed 430 times on 39 different plant species, but never on plantswith red flowers. Pieris rapae was the most abundant species that occurred 136 times on a total of 21 different plant speciesof which eight had yellow flowers. But, this species has never been seen while feeding on red flowers.

Key words: Lepidoptera; diversity; flower colour; foraging behaviour

Introduction

The co-evolution of flowering plants and their animalpollinators has been one of the most inspiring phenom-ena for scientists to document many detailed proper-ties of this subject in a diverse range of biology dis-ciplines. Pollinators involve many flying animals, butinsects, particularly bees and butterflies have receivedmost attention, since they have traditionally been con-sidered as the significant selective force to play role infloral evolution (Kudoh & Whigham 1998; Galen 1999;Goulson 1999). Plants have evolved a number of prop-erties to attract their potential pollinators (David &Gardiner 1961; Gardener & Gilman 2002; Arroyo et al.2007). But, the odour and colour of the flowers (Odellet al. 1999; Balkenius & Kelber 2004; Theis 2006; Os-orio & Vorobyev 2008) have accommodated pollinatorsto a much greater extent, even though this attractionhas sometimes had an array of costs (Galen 1999; Collinet al. 2002).Remarkable studies have been performed on the

colour vision of animals to reveal aspects of this be-haviour (Arikawa & Stavenga 1997; Briscoe & Chittka2001; Qiu & Arikawa 2003; Briscoe & Bernard 2005;Warrant & Nilsson 2006). Many vertebrates, deep-seacrustaceans and cephalopods have monochromatic vi-sion, i.e., they are colour blind (Warrant & Nilsson2006). It has been demonstrated that some nocturnalhawkmoths have colour sensitive eyes to detect flowerseven at dim starlight intensity when bees and humansare totally colour blind (Kelber et al. 2002). The pio-neer study on colour vision of insects dates back as faras to the late 1800s and was performed by Lubbock on

Daphnia and in 1914, Karl von Frisch proved that beeshave true colour vision (Warrant & Nilsson 2006). Ac-cording to a number of studies it is clear that spectralsensitivity and its functions in a particular species maydiffer for different visual tasks.Insect pollinators exhibit flower colour preference

(Lunau & Wacht 1994; Weiss 1997; Sutherland et al.1999; Briscoe & Chittka 2001) which varies among theindividuals and among the species studied. The colourpreference associations between insects and plants maybe due to several influences. For example, certaincolours are avoided (Vaidya 1969; Kelber 1999) andsome colours may be rewarding for nectar (Niesenbaumet al. 1999; Ida & Kudo 2003). Consequently, a partic-ular species may show some floral constancy associatedwith a variety of plant characters (Goulson & Wright1998; Goulson 1999), including flower colour (Kinoshita& Arikawa 2000; Kelber et al. 2002; Balkenius & Kelber2004).Butterflies are important pollinators of a wide

range of plant species since they frequently visit nu-merous flowers during their lifetime activities such asforaging, mating, and oviposition. Hence, they respondto several properties of plants such as scent, reflectance,size, outline and surface texture. But, the flower colourhas been one of the most profound impacts. A signifi-cant number of recent experimental studies confirmed(Kelber & Henique 1999; Kelber et al. 2002; Qiu &Arikawa 2003; Wakakuwa et al. 2004; Warrant & Nils-son 2006) that most butterflies have a unique eye struc-ture (Warrant & Nilsson 2006) and a highly developedcolour vision (Bernard & Remington 1991; Arikawa &Stavenga 1997; Awata et al. 2009), therefore it seems

c©2010 Institute of Zoology, Slovak Academy of Sciences

1050 S. Yurtsever et al.

that they are able to discriminate true colours. Thesestudies prove that the pigments which are present ina flower are key components of behaviour patterns ofcertain Lepidoptera species. Accordingly, a particularflower colour of a plant seems to be a signal for conduct-ing certain activities which will offer fitness for survivalfor the pollinator itself and for its next generation.In this paper, we present observations of a field

study which was performed on Lepidoptera visiting hostplants with particular flower colours.

Material and methods

The study was conducted in Trakya University (TU) in anisolated and protected natural woodland part of TU cam-pus. Thus, the study area was covered with a variety oftrees and herbaceous plants for providing suitable habitatsfor a diverse Lepidoptera fauna. The study area involved astream, a small pond and several meadow patches.

The rectangle shaped study area was about 2 km inlength and 500 m in width. The study was carried out fromearly March to late July in 2008. A total of 1,329 obser-vations were made during 20 sampling sessions. One sam-pling was performed every week periodically in the studyarea and about five hours were spent for each sampling.The observations were conducted between 10.00 AM and3.00 PM when adult Lepidoptera were most active duringthe day time. The observations included butterflies recordswhile they were clearly on a nectar feeding action on a par-ticular coloured flower of a plant species. The visiting scoreswere considered only when the butterfly/moth probed theflower. Perching butterflies or those not attempting to feedwere not considered for the foraging data. The sex of mostindividuals was not possible to record because we did notwant to prevent the foraging action.

The following tree species and herbaceous plants inthe study area where the observations were performedwere most abundant: Ulmus minor, Salix alba, Populus ni-gra, Robinia pseudoacacia, Prunus spinosa, Rubus sanc-tus, Crepis sancta, C. setosa, Trifolium repens, Capsellabursa-pastoris, Senecio vernalis, Anthemis cotula, Tyrim-nus leucographus, and Carduus pycnocephalus. A total of66 different plant species belonging to 17 families werecounted on which butterfly/moth nectar feeding exerciseswere recorded.

Additionally, a total of 428 adult Lepidoptera speci-mens belonging to 43 species and eight families were col-lected for morphometric analyses. Several measurements ofcertain parts of the plants including 10 specimens in eachspecies were taken. The results of these two aspects, namelythe correlations between the plant and Lepidoptera speciesconsidering certain morphometric characters will be ad-dressed elsewhere. During the course of the study, the but-terflies were captured using a standard insect net and thenkilled by squeezing the thorax. The materials were examinedin the laboratory using a binocular stereomicroscope.

Statistical analyses were carried out according to Zar(1996). The data from the species were pooled in each Lepi-doptera family to obtain appropriate quantity for statisticalanalyses. The index of Lepidopera diversity (H) on plantswith seven different colours of flowers was calculated usingthe Shannon-Wiener formula (Zar 1996):

H = −s∑

i=1

pi log pi

where pi is the proportion of the ith species and S is thenumber of species found in the case of a grouped studysite. This formula was modified to analyse if distinct groupsof plants with differently coloured flowers have comparablediversity values regarding the butterflies that visit them.Diversity indices between the seven types of flowers werecompared pair-wise using the t-test proposed by Hutcheson(1970), where the variance (S2H) of each H and the degrees offreedom (v) for the preceding t were approximated by usinga particular formula (Zar 1996). The Shannon’s equitability(E) was calculated by dividing H by Hmax (Hmax = lnS)which can be formulated as E = H/Hmax = H/lnS. Thedivergence (D) from equiprobability was calculated usingthe formula (D = Hmax − log4n −H). Thus, a low value ofD close to H indicates a high degree of diversity. Conversely,a high value of D indicates lower diversity and shows thatthe system has diverged substantially from equiprobability.

The G-test test (Fowler et al. 1998) was employed tosee if the observed Lepidoptera frequencies depend on par-ticular coloured flower groups used for foraging. William’scorrection factor was applied in the G-test irrespective ofthe number of degrees of freedom. Yates’ correction was ap-plied where there were only two categories of Lepidopteranumbers. The pair-wise comparisons included Lepidopteranumbers in the seven flower colour groups of the investi-gated plant species. The results of statistical analyses aresummarised in the tables with corresponding significancelevels.

Results

Forty-three different Lepidoptera species (Appendix 1)belonging to eight families which were recorded on66 different plant species belonging to 18 families(Appendix 2) were examined in the study area. Thehighest Lepidoptera species numbers were recorded inNymphalidae while three families had just a total of fivespecies since they included diurnal and nocturnal Lepi-doptera. Carduus nutans with pink coloured flowers wasthe most visited plant species during the study; it wasvisited 129 times by 19 different Lepidoptera species.On the other hand, the least visited plant species wasCapsella bursa pastoris with white coloured flower; itwas visited only four times by the Pieridae butterflyEuchloe eusonia.Furthermore, one of the most abundant Lepi-

doptera was Pieris rapae that occurred 136 times ona total of 21 different plant species. However, P. rapaehas never been seen while feeding on red or yellow-whiteflowers. The majority of this Pieridae butterfly occurredon plants with yellow flowers. Hence, P. rapae was ob-served 51 times on a total of 8 different plants with yel-low flowers. Besides, the Lycanidae butterfly Lycaenaalcipron was the most occasional species that was ob-served only two times on Carduus nutans having pinkflowers and being the most abundant plant species.With regard to Lepidoptera families (Table 1),

members of Hesperidae, Pieridae and Lycanidae wereobserved on all seven differently coloured flowers.Nymphalidae and Noctuidae have never been seen vis-iting red flowers. The same was true for Zyganidae.Sphingidae moths were observed on three differently

Colour preference in foraging Lepidoptera 1051

Table 1. Summary of Lepidoptera considered in analyses in thestudy.

Family A B C

HES 16 72 7PAP 2 15 2PIE 27 413 7LYC 31 315 7NYM 39 430 6NOC 7 19 6ZYG 6 29 5SPH 3 36 3

Explanations: For Lepidoptera family abbreviations see Ap-pendix 1; A – Number of plant species visited by particularLepidoptera family; B – Individual Lepidoptera numbers in eachfamily observed during the study; C – Number of different flowercolours that were visited by the Lepidoptera family.

coloured flowers – white, pink and purple. Papilionidaebutterflies occurred only on pink and yellow flowers.Nymphalidae butterflies with the highest species num-ber during the study were observed on 39 different plantspecies, but not on plants with red flowers. Moreover,Nymphalidae were observed 430 times on a wide vari-ety of plants, though the highest number of observa-tions (150) was associated with pink flowers. Althoughthe species number of Pieridae (9 species) was less thanin Nymphalidae (14 species), the flower visiting num-bers of Pieridae were very close to those in Nympha-lidae. Thus, Pieridae occurred 413 times on differentplants out of which 120 were on plants with pink flow-ers. Consequently, all of the eight Lepidoptera familiesincluding 30 different species visited pink coloured flow-ers for foraging. White, purple and yellow flowers werevisited by seven different Lepidoptera families, whilstyellow-white flowers were visited by six Lepidopterafamilies. However, only three Lepidoptera families withseven species, i.e., Aporia crataegi (PIE), Glaucopsy-che alexis (LYC), Pieris brassicae (PIE), Pyrgus malvae(HES), Spalia orbifer (HES) and Thymelicus sylvestris(HES), were observed while nectar feeding on plants

7

14

24

14

2022

30

35

97

380

53

141

373

250

0

5

10

15

20

25

30

White Pink Purple Blue Yellow Y-W Red

%

Fig. 1. Combined Lepidoptera number distributions for seven dif-ferent colour groups of flowers where particular species were ob-served. The y-axis represents the percentage. Blank and filledbars indicate the species numbers and individual Lepidopteranumbers, respectively. The numbers are given on the top of thebars. Y-W – yellow-white

with red flowers. Regarding red flowers, A. crataegi wasobserved twelve times on Dianthus corymbosus. The re-maining six species which were visitors of red flowerswere recorded almost in equal numbers. Crepis sanctawas the most preferred plant species with yellow flowerswhich had 42 visits by 11 different Lepidoptera species.The Lycanidae butterfly Polyommatus icarus had thehighest visiting record on yellow flowers of 6 differentplant species. Although the visiting numbers of Pierisrapae (51 visits) were less than those of P. icarus (58visits) on yellow flowers, the number of plant specieswith yellow flowers which were visited by P. rapae washigher (8 spp.) than for P. icarus.Our observations revealed that for nectar feeding,

Lepidoptera tend to prefer flowers of a particular colourwhich, based on the present study, are categorised intoseven groups (Fig. 1) (Gadj = 698.6, df = 6, P < 0.001).Thus, it appears that yellow (29%) and pink (28%)coloured flowers were preferred most by Lepidoptera.White (19%), purple (11%) and yellow-white (7%) flow-

Table 2. Pair-wise comparisons of Lepidoptera numbers (from Fig. 1) observed on flowers of seven different colours in the study area.

Colour Pink Purple Blue Yellow Yellow-white

Red

White 24.28(***)

30.39(***)

128.08(***)

26. 83(***)

67.46(***)

162.19(***)

Pink 104.72(***)

240.38(***)

0.065(N.S)

162.08(***)

280.01(***)

Purple 39.92(***.)

109.64(***)

8.13(**)

63.84(***)

Blue 246.95(***)

12.91(***)

3.68(N.S.)

Yellow 162.90(***)

286.81(***)

Yellow-white 29.12(***)

Explanations: ** P < 0.01, *** P < 0.001.


Table 3. Values of Lepidoptera diversity indices and other param-eters calculated for the seven different flower colours, using theShannon-Wiener diversity index (H values are at log2, the log10values of H are given in parentheses).

Flower colour H S2H D E

White 2.074 (0.624) 0.0005 0.925 69.134Pink 2.083 (0.627) 0.0004 0.916 69.448Purple 2.104 (0.634) 0.0009 0.905 70.166Blue 2.149 (0.647) 0.0007 0.850 71.651Yellow 2.013 (0.606) 0.0002 0.986 67.106Yellow-White 1.804 (0.543) 0.0013 1.195 60.148Red 1.446 (0.435) 0.0008 1.553 48.202

Table 4. Pair-wise comparisons of Lepidoptera diversity indicesbetween plants wit flowers of seven different colours in the studyarea.

Colour Pink Blue Yellow-white Red

White – – – (***)Pink – – (*) (***)Purple – – – (***)Blue – – (*) (***)Yellow – – – (***)Yellow-White (*) (*) – (**)

Explanations: Blank cells with a dash: P > 0.05 (not significant),* P < 0.05, ** P < 0.01, *** P < 0.001.

ers follow in the ranking order. However, in the 1,329observations during the present study, red (2%) andblue (4%) coloured flowers did not seem to be very pre-ferred by Lepidoptera. As evident from Table 2, therewere no significant statistical differences between theLepidoptera visiting frequencies observed on pink andyellow flowers. Thus, pink and yellow coloured flowersdid not differ in nectar feeding preference of butter-flies, although they were the most visited flowers amongall investigated. Similarly, the visiting frequency of theleast preferred flower colours, blue and red, was also notsignificantly different. In conclusion, in Lepidoptera theflower colour preference for nectar feeding can be di-vided into five groups: yellow and pink coloured flowerscome first, flowed by white, purple, and yellow-white.The blue and red coloured flowers create the last, fifthgroup.Diversity of Lepidoptera (Table 3) on the seven

different colour groups of flowers also showed some in-teresting results (Table 4). Thus, it appears that plantswith red and yellow-white flowers are not visited by di-verse Lepidoptera which visit them for nectar feeding.The remaining colour groups of flowers did not showany significant difference in the H values. Blue flowerswere visited by much more diverse Lepidoptera thanflowers of the other colours and the distribution of but-terflies in this colour group was more homogenous thanin the other flowers. Although pink and yellow flow-ers were visited by much more butterfly species dis-tributed in different families, the numbers in particularLepidoptera groups in the samples were highly dispro-

portional. Accordingly, based on Lepidoptera diversity,flowers of different colours can be ranked in a descend-ing order as follows: Blue > Purple > Pink >White >Yellow > Y-W > Red.

Discussion

According to our data, yellow flowers were the most vis-ited by Lepidoptera for foraging, though visiting pat-terns of the species examined showed a great variation.As it has been reported for other insect pollinators (Lu-nau & Wacht 1994; Sutherland et al. 1999; Briscoe &Chittka 2001), the colour preference by Lepidopteravaries among individuals and among the species stud-ied. For example, green colour is not attractive for feed-ing but it induces oviposition in females of some butter-flies, such as Papilio demoleus (L., 1758) (Vaidya 1969)and P. aegeus Donovan, 1805 (Kelber 1999). This isbecause the newly hatched caterpillars need food andbrightly green leaves give a sign of fresh and rich nutri-tion. The hawkmothsManduca sexta (L., 1763) (Goyretet al. 2008) and Macroglossum stellatarum (L., 1758)(Kelber 1997) have a strong innate preference for blueflowers as a food source but a weaker preference foryellow colour. Blue and green colours are stimuli bothfor feeding and egg-laying in Pieris brassicae (David& Gardiner 1961; Scherer & Kolb 1987; Weiss & Papa2003). It has been reported that the syrphid hoverflyEpisyrphus balteatus (De Geer, 1776) strongly prefersyellow flowers for foraging, moreover, small flowers havemore visitors than the large artificial flowers (Suther-land et al. 1999). Thus, several experimental studieshave clearly proved that preference of yellow is innatein many insect pollinators (Kay 1976; Lunau & Wacht1994; Odell et al. 1999).Nymphalidae butterflies comprised the highest

number of species, however, none of the 14 species ofthis family was seen on plants with red flowers, simi-larly as Noctuidae. Thus, all of their 430 visits were as-sociated with the 39 different plant species which haddifferent flower colours. It appears that some Nympha-lidae have no red-sensitive photoreceptors (Briscoe &Bernard 2005). Moreover, red flowers had the lowestnumber of visits by Lepidoptera, which included only 35specimens of seven species. According to the majorityof studies performed on the colour choice for foraging,the red colour does not seem attractive compared to theother colours. This may also be true for our study, be-cause red flowers had the least number of Lepidopteravisitors. Thus, our results do not rule out that someinsect species, including some butterflies, do not use orlack red visual receptors (Hasselmann 1962; Bandai etal. 1992; Briscoe & Bernard 2005). However, there isa remarkable diversity among the Lepidoptera speciesregarding wavelength for peak sensitivity (Briscoe &Chittka 2001; Briscoe & Bernard 2005). Accordingly,a number of reports based on certain species affirmedthat the butterflies are able to see red (Kevan & Baker1983; Scherer & Kolb 1987; Kelber & Pfaff 1999). Themajority of studies supporting the red vision of butter-


flies are experimental and involve learning behaviour.The experimental study of Goulson & Cory (1993) onPieris napi showed that the preference of red or blueflowers is not inherent. The data of Goulson & Wright(1998) based on a study in the field demonstrated thatthe hoverflies Episyrphus balteatus and Syrphus ribesii(L., 1758) showed no preference for a particular colourmorph within a plant species, but they exhibited a sig-nificant degree of constancy to plant species previouslyvisited. Nevertheless, a number of studies including thepresent one confirmed that butterflies exhibited thewidest visual range in the animals, thus colour mustbe an important signal for visiting a wide variety ofplants.As a matter of fact, butterflies respond to colour

of petals; the colour of the flowers change by time andis a cue for nectar availability. The quantity of nectarin a flower fluctuates through time as it has been de-pleted by several ways including animal foraging (Cor-bet 2003). Many flowers change in colour with age,the colour of young flowers, particularly of yellow oneswhich are usually nectar-rich turns to red (Niesenbaumet al. 1999; Ida & Kudo 2003) or from white to pur-ple (Sun et al. 2005) when nectar is no longer pro-duced (Casper & La Pine 1984; Weiss 1991; Niesen-baum et al. 1999). Evidence of preference for certainflower colours regarding foraging comes from a numberstudies. For example, white-flowered forms of the lark-spur, Delphinium nelsonii, produce fewer seeds than themore common blue-flowered form as a result of discrim-ination of white flowers by bumblebees and humming-birds (Waser & Price 1981). This was also true for thepresent study, because butterflies showed a tendencyto prefer flowers of particular colours. Although plantswith yellow flowers had more Lepidoptera visitors thanthe other flowers, the highest Lepidoptera diversity val-ues were associated with the plants having blue andpurple flowers. This is probably caused by the fact thatbutterflies visit these plants and find out that they havepoor food resource. Therefore they shortly leave theseflowers for yellow ones. Accordingly, the colour choiceby some butterflies may be innate (Kelber 1997; Weiss1997; Goyret et al. 2008) thus, they can learn and maychange colour preference as it has been shown in manystudies. Some Lepidoptera species can be trained tolearn rewarding colours. For example, for the hawkmothM. stellatarum and Deilephia elpenor L., 1758 a par-ticular colour may become more attractive than theirinnately preferred colour (Kelber et al. 2002; Balkenius& Kelber 2004). In the rewarding studies, nectar andsimilar food were usually used to attract the butterflies.Thus, it appears that Lepidoptera in general visit yel-low flowers significantly more often than those of othercolours. Presumably yellow flowers produce more nectarand pollen and therefore are rich in nutrition (Casper &La Pine 1984). Moreover, reproductive organs of mostflowers are yellow and the nutrition can be probed bythe galea of foraging Lepidoptera. Accordingly, yellowis a very simulative colour giving a sign of rich food.Hence, the colour choice behaviour may have in-

nate (inherited) or experienced (external influence)base for the rewarding colour, or both. Thus geneticand learned experiences enable Lepidoptera to avoidflowers which are poor sources of food and to adaptfor nectar availability in diverse environments. This isperhaps a perceptual mechanism to discover a partic-ular flower, i.e. a method similar to those in predatorssearching a prey (Goulson 2000). Hence, flower depen-dent species have evolved to learn rewarding colour andthis behaviour has an adaptive value for their fitness inthe field. There is also no doubt that many propertiesof particular Lepidoptera and plant species with regardto foraging have strong interaction (Corbet 2000; Gar-dener & Gillman 2002). These complex visual featuresare adaptive for natural plant populations and for theirpollinators (Kay 1976; Osorio & Vorobyev 2008) andhave played important role in colour choice and thusshaping the co-evolution.

Acknowledgements

We are very grateful to the editor M. Kazimirova, and theanonymous reviewers for kind suggestions and comments onthe manuscript.

References

Arikawa K. & Stavenga D.G. 1997. Random array of colour filtersin the eyes of butterflies. J. Exp. Biol. 200: 2501–2506.

Arroyo M.T.T., Till-Bottraud I., Torres C., Henriquez C.A. &Martinez J. 2007. Display size preferences and foraging habitsof high Andean butterflies pollinating Chaetanthera lycopo-dioides (Asteraceae) in the subnival of the central ChileanAndes. Arct. Antarct. Alp. Res. 39: 347–352.

Awata H., Wakakuwa M. & Arikawa K. 2009. Evolution of colourvision in pierid butterflies: blue opsin duplication, omma-tidial heterogeneity and eye regionalization in Colias erate. J.Comp. Physiol. A 195: 401–408. DOI: 10.1007/s00359–009–0418–7

Balkenius, A. & Kelber A. 2004. Colour constancy in diurnal andnocturnal hawkmoths. J. Exp. Biol. 207: 3307–3316. DOI:10.1242/jeb.01158

Bandai K., Arikawa K. & Eguchi E. 1992. Localization of spectralreceptors in the ommatidium of butterfly compound eye de-termined by polarization sensitivity. J. Comp. Physiol. 171:289–297.

Bernard G.D. & Remington C.L. 1991. Color vision in Lycaenabutterflies: Spectral tuning of receptor arrays in relation tobehavioural ecology. Proc. Nat. Acad. Sci. U.S.A. 88: 2783–2787.

Briscoe A.D. & Bernard G.D. 2005. Eyeshine and spectral tun-ing of long wavelength-sensitive rhodopsins: no evidence forred-sensitive photoreceptors among five Nymphalini butterflyspecies. J. Exp. Biol. 208: 687–696. DOI: 10.1242/jeb.01453

Briscoe A.D. & Chittka L. 2001. The evolution of colour visionin insects. Annu. Rev. Entomol. 46: 471–510.

Casper B.B. & La Pine T.R. 1984. Changes in corolla color andother floral characteristics in Cryptantha humilis (Boragi-naceae) cues to discourage pollinators? Evolution 38: 128–141.

Collin C.L., Pennings P.S., Rueffler C., Widmer A. & ShykoffJ.A. 2002. Natural enemies and sex: how seed predation andpathogens contribute to sex-differentiated reproductive suc-cess in a gynodioecious plant. Oecologia 131: 94–102. DOI:10.1007/s00442–001–0854–8

Corbet S.A. 2000. Butterfly nectaring flowers: butterfly andflower form. Entomol. Exp. Appl. 96: 289–298. DOI: 10.1023/A:1004096432758


Corbet S.A. 2003. Nectar sugar content: estimating standing cropand secretion rate in the field. Apidologie 34: 1–10. DOI:10.1051/apido:2002049

David W.A.L. & Gardiner O.C. 1961. Feeding behaviour of adultsof Pieris brassicae (L.) in a laboratory culture. Bull. Entomol.Res. 52: 741–762.

Fowler J., Kohen L. & Jarwis P. 1998. Practical Statistics forField Biology. Wiley, Chichester, 257 pp.

Galen C. 1999. Why do flowers vary: the functional ecology ofvariation in flower size and form within natural plant popu-lations. Bioscience 49: 631–640.

Gardener M.C. & Gillman M.P. 2002. The taste of nectar – aneglected area of pollination ecology. Oikos 98: 552–557. DOI:10.1034/j.1600-0706.2002.980322.x

Goulson D. 1999. Foraging strategies of insects for gathering nec-tar and pollen, and implications for plant ecology and evolu-tion. Perspect. Plant Ecol. Evol. Syst. 2: 185–209.

Goulson D. 2000. Are insects flower constant because they usesearch images to find flowers. Oikos 88: 547–552. DOI:10.1034/j.1600-0706.2000.880311.x

Goulson D. & Cory J.S. 1993. Flower constancy and learning inforaging preferences of the green-veined white butterfly Pierisnapi. Ecol. Entomol. 18: 315–320.

Goulson D. & Wright N.P. 1998. Flower constancy in the hover-flies Episyrphus balteatus (Degeer) and Syrphus ribesii (L.)(Syrphidae). Behav. Evol. 9: 213–219.

Goyret J., Pfaff M., Raguso R.A. & Kelber A. 2008. Why doManduca sexta feed from white flowers? Innate and learntcolour preferences in a hawkmoth. Naturwissenchaften 95:569–576.

Hasselmann E.-M. 1962. Uber de relative spektrale Empfindich-keit von Kafer- und Schemetterlingsaugen bei verschiedenenHelligkeiten. Zool. Jb. Physiol. 69: 537–576.

Hutcheson K. 1970. A test for comparing diversities based on theShannon formula. J. Theor. Biol. 29: 151–154.

Ida T.Y. & Kudo G. 2003. Floral colour change in Weigela mid-dendorffiana (Caprifoliaceae): Reduction of geitonogamouspollination by bumble bees. Am. J. Botany 90: 1751–1757.

Kay Q.O.N. 1976. Preferential pollination of yellow-floweredmorphs of Raphanus raphanistrum by Pieris and Eristalisspp. Nature 261: 230–232. DOI: 10.1038/261230a0

Kelber A. 1997. Innate preferences for flower features in the hawk-moth Macroglossum stellatarum. J. Exp. Biol. 200: 827–836.

Kelber A. 1999. Ovipositing butterflies use a red receptor to seegreen. J. Exp. Biol. 202: 2619–2630.

Kelber A., Balkenius A. & Warrant E.J. 2002. Scotopic colourvision in nocturnal hawk moths. Nature 419: 922–925. DOI:10.1038/nature01065

Kelber A. & Hénique U. 1999. Trichromatic colour vision inthe hummingbird hawkmoth, Macroglossum stellatarum. J.Comp. Physiol. A 184: 535–541. DOI: 10.1007/s003590050353

Kelber A. & Pfaff M 1999. True colour vision in the OrchardButterfly, Papilio aegeus. Naturwissenchaften 86: 221–224.

Kevan P.G. & Baker H.G. 1983. Insects as flower visitors andpollinators. Annu. Rev. Entomol. 28: 407–453

Kinoshita M. & Arikawa K. 2000. Colour constancy of the swal-lowtail butterfly Papilio xuthus. J. Exp. Biol. 203: 3521–3530.

Kudoh H. & Whigham D.F. 1998. The effect of petal size ma-nipulation on pollinator/seed-predator mediated female re-productive success of Hibiscus moscheutos. Oecologia 117:70–79.

Lunau K. & Wacht S. 1994. Optical releasers of the innate pro-boscis extension in the hoverfly Eristalis tenax L. (Syrphidae,Diptera). J. Comp. Physiol. A 174: 575–579.

Niesenbaum R.A., Patselas M.G. & Wein S.D. 1999. Does flowercolour change in Aster vimineus cue pollinators? Am. Midl.Nat. 141: 59–68.

Odell E., Raguso R.A. & Jones K.N. 1999. Bumblebee foragingresponses to variation in floral scent and color in snapdragons(Antirrhinum: Scrophullariaceae). Am. Midl. Nat. 142: 257–265.

Osorio D. & Vorobyev M. 2008. A review of the evolution ofanimal colour vision and visual communication signals. VisionRes. 48: 2042–2051.

Qiu X. & Arikawa K. 2003. The photoreceptor localization con-firms the spectral heterogeneity of ommatidia in the malesmall butterfly, Pieris rapae crucifora. J. Comp. Physiol A.189: 81–88.

Scherer C. & Kolb G. 1987. Behavioral experiments on the vi-sual processing of color stimuli in Pieris brassicae L. (Lepi-doptera). J. Comp. Physiol. 160: 645–656.

Sun S.-G., Liao K., Xia J. & Guo Y.H. 2005. Floral colourchange in Pedicularis monbeigiana (Orobanchaceae). PlantSyst. Evol. 255: 77–85.

Sutherland J.P., Sullivan M.S. & Poppy G.M. 1999. The influenceof floral character on the foraging of the hoverfly, Episyrphusbaltaeus. Entomol. Exp. Appl. 93: 157–164.

Theis N. 2006. Fragrance of Canada thistle (Circium arvanse)attracts both floral herbivores and pollinators. J. Chem. Ecol.32: 917–927. DOI: 10.1007/s10886-006-9051-x

Vaidya V.G. 1969. Investigations on the role of visual stimuli inthe egg-laying and resting behaviour of Papilio demoleus L.(Papilionidae, Lepidoptera). Anim. Behav. 17: 350–355.

Wakakuwa M., Stavenga D.G., Kurasawa M. & Arikawa K.2004. A unique visual pigment expressed in green, red anddeep-red receptors in the eye of the small white butterfly,Pieris rapae crucivora. J. Exp. Biol. 207: 2803–2810. DOI:10.1242/jeb.01078

Warrant E. & Nilsson D.-A. 2006. Invertebrate vision. CambridgeUniv. Press. 547 pp.

Waser N.M. & Price M.V. 1981. Polinator choice and stabilizingselection for flower color in Delphinium nelsonii Evolution35: 376–390.

Weiss M.R. 1991. Floral colour changes as cues for pollinators.Nature 354: 227–229. DOI: 10.1038/354227a0

Weiss M.R. 1997. Innate colour preferences and flexible colourlearning in the pipevine swallowtail. Anim. Behav. 53: 1043–1052.

Weiss M.R. & Papa D.R. 2003. Colour learning in two be-havioural contexts: how much can a butterfly keep in mind?Anim. Behav. 65: 425–434.

Zar H.J. 1996. Biostatistical Analysis. Prentice-Hall, Inc. NewYork, 662 pp.

Received May 12, 2010Accepted July 20, 2010


Appendix 1. List of Lepidoptera species belonging to eight families recorded on flowers of seven different colours, belongingto 66 different plant species in Edirne. The family names are abbreviated in three letters for convenience in the text, thusonly the first letters of the respective family are given. The species names are coded to enable comparison of the Lepidopteraspecies names by plant species names (as in Appendix 2).

Hesperiidae (HES)Thymelicus sylvestris (Poda, 1761) (A)Spialia orbifer (Hubner, 1823) (B)Pyrgus sidae (Esper, 1784) (C)Pyrgus malvae (L., 1758) (D)

Papilonidae (PAP)Iphiclides podalirius (L., 1758) (E)

Pieridae (PIE)Colias croceus (Fourcroy, 1785) (F)Gonopteryx rhamni (L., 1758) (G)Anthocharis cardamines (L., 1758) (H)Euchloe ausonia (Hubner, 1804) (I)Aporia crataegi (L., 1758) (J)Pieris brassica (L., 1758) (K)Pieris rapae (L., 1758) (L)Pieris napi (L., 1758) (M)Pontia daplidice (L., 1758) (N)

Lycaenidae (LYC)Lycaena phlaeas (L., 1761) (O)Lycaena tityrus (Poda,1761) (P)Lycaena alcipron (Rottemburg, 1775) (Q)Lycaena thersamon (Esper, 1784) (R)Satyrium acaciae (F., 1787) (S)Celastrina argiolus (L., 1758) (T)Glaucopsyche alexis (Poda, 1761) (U)Plebjus agestis (Denis et Schiffermuller, 1775) (V)

Plebejus artaxerxes (F., 1793) (W)Polyommatus icarus (Rottemburg, 1775) (X)

Nymphalidae (NYM)Kirinia roxelana (Cramer, 1777) (Y)Coenonympha pamphilus (L., 1758) (Z)Maniola jurtina (L., 1758) (ZA) aMelanargia galathea (L., 1758) (ZB)Limenitis reducta Staudinger, 1901 (ZC)Vanessa cardui (L., 1758) (ZD)Polygonia c-album (L., 1758) (ZE)Euphydryas aurina (Rottemburg, 1775) (ZF)Melitaea cinxia (L., 1758) (ZG)Melitaea phoebe (Goeze, 1779) (ZH)Melitaea didyma (Esper, 1778) (ZI)Argynnis pandora (Denis & Schiffermuller, 1775) (ZJ)Issoria lathonia (L., 1758) (ZK)Brenthis daphne (Bergstrasser,1780) (ZL)

Noctuidae (NOC)Autographa gamma (L., 1758) (ZM)Acontia lucida (Hufnagel, 1767) (ZN)Zygaenidae (ZYG)Adscita statices (L., 1758) (ZO)Zygaena purpuralis (Brunnich, 1763) (ZP)

Sphingidae (SPH)Macroglossum stellatarum (L., 1758) (ZQ)

Appendix 2. List of plant species (arranged alphabetically by family) and Lepidoptera species (abbreviations for Lepidopteraspecies are as in Appendix 1) of which foraging observations were conducted in Edirne.

Plant Species Lepidoptera species observed onthe plant

Apiaceae

Torilis japonica A, O, S, V, Z, ZA, ZNEryngium campestre FAsteraceaeAchillea coarctata A, SAnthemis austriaca A, N, O, Z, ZA, ZG, ZPAnthemis cotula S, ZAnthemis tinctoria XCarduus nutans A, C, F, J, K, L, N, Q, ZA, ZD,

ZF, ZG, ZI, ZJ, ZK, ZL, ZM, ZO,ZQ

Carduus pycno-cephalus

C, J, L, N, ZP

Centaurea cyanus ZA, ZGCentaurea solsti-tialis

E

Cirsium creticum ZJCrepis sancta B, L, M, N, U, V, X, ZF, ZG, ZK,

ZO

Crepis setosa C, L, Z, ZA, ZK, ZO, ZP

Echinops micro-cephalus

F, O, X, ZK

Matricaria chamo-milla

C, U, V, X, Z, ZG, ZN, ZO

Scolymus hispanicus F, ZKSenecio vernalis F, H, I, L, N, O, Z, ZA, ZNSilybum marianum K, V, ZA, ZG, ZH, ZJTaraxacum officinale O, L, ZKBoraginaceaeAnchusa officinalis A, B, J, L, P, ZEchium italicum L, ZQEchium plantagi-neum

A, J, P, ZG

Myosotis arvensis V, ZNMyosotis stricta V, ZNBrassicaceaeArabidopsis thaliana N, VBerteroa mutabilis L, N, P, V, W, ZCapsella bursa-pastoris

I

Cardaria draba H, L, N, V, Z, ZK, ZNDiplotaxis tenuifolia HErysimum diffusum LRaphanus raphanis-trum

L


Rorippa sylvestre C, H, N, Z, ZGRorippa thracica C, H, N, Z, ZGSisymbrium offici-nale

N, X

CaprifoliaceaeLonicera japonica H, ZQCaryophyllaceaeDianthus corym-bosus

A, J, K

DipsaceaeScabiosa triniifolia N, ZI, ZOEuphorbiaceaeEuphorbia seguie-riana ssp. niciciana

A, S, ZA, ZL, ZO

FabaceaeCoronilla varia ssp.varia

ZA, L

Lathyrus cicera U, VLotus corniculatusvar. tenuifolius

X

Medicago lupulina X, Z, ZDMedicago minima V, XMelilotus alba XTrifolium nigrescens L, XTrifolium repens A, X, Z, ZDTrifoliumresupinatum

X, Z

Vicia cracca A, Z, ZKVicia hirsuta B, VFumaria officinalis V, X

GeraniaceaeErodium cicutarium LGeranium disectum L, V, ZGeranium molle V, X, ZLamiaceaeClinopodium vulgare A, GLamium purpureum E, F, L, Z, ZD, ZK, ZN, ZQLiliaceaeOrnithogalum refrac-tum

I, L, N, V

RhamnaceaePaliurus spina-christii

O, ZA

RosaceaePotentilla argentea L, ZKPotentilla recta L, ZKPrunus spinosa I, T, ZERubus sanctus A, E, G, K, L, O, R, V, Y, Z, ZB,

ZC, ZD, ZK, ZL

Sanguisorba minorssp. muricata

A, B, D

RubiaceaeGalium aparine J, ZQGalium spurium J, ZQScrophulariaceaeParentucellia latifo-lia ssp. latifolia

Z, ZK

ViolaceaeViola kitaibeliana ZK

what colour of flowers do lepidoptera prefer for foraging?

Documents