ArticlePDF Available

Abstract and Figures

Background and Aims Flower colour is a key feature in plant-pollinator interactions that make the flowers visible amid the surrounding green vegetation. Green flowers are expected to be scarcely conspicuous to pollinators; however, many of them are visited by pollinators even in the absence of other traits that might attract pollinators (e.g., floral scents). In this study, we investigate how entomophilous species with green flowers are perceived by pollinators. Methods We obtained reflectance spectra data of 30 European species that display green or green-yellow flowers to the human eye. These data were used to perform spectral analyses, calculate both chromatic (colour contrast against the background) and achromatic (colour contrast that relies on the signals from the green-sensitive photoreceptors) cues, and model colour perception by hymenopterans (bees) and dipterans (flies). Key Results The visibility of green flowers to bees and flies (i.e., their chromatic contrast values) was lower compared to other floral colours commonly pollinated by these insects, whereas green-yellow flowers were as conspicuous as the other flower colours. Green flowers with low chromatic contrast values exhibited higher achromatic contrasts, which is used to detect distant flowers at narrow visual angles, than green-yellow flowers. Additionally, the marker points (i.e., sharp transition in floral reflectance that aid pollinators in locating them) of green and green-yellow flowers aligned to some degree with the colour discrimination abilities of bees and flies. Conclusions We found that many entomophilous green and green-yellow flowers are conspicuous to bees and flies through their chromatic or achromatic contrasts. While acquiring pigments like carotenoids, which impart a yellowish hue to flowers and enhances their visibility to pollinators, could increase their conspicuousness, the metabolic costs of pigment production, along with the use of alternative strategies to attract pollinators, may have constrained carotenoid emergence in certain lineages of green-flowered species.
Content may be subject to copyright.
Annals of Botany XX: 1–12, 2025
https://doi.org/10.1093/aob/mcae213, available online at www.academic.oup.com/aob
ORIGINAL ARTICLE
Green owers need yellow to get noticed in a green world
José C. del Valle1,*, Melissa León-Osper2, Carlos Domínguez-González2, Mª Luisa Buide2, Montserrat Arista1,
Pedro L. Ortiz1, Justen B. Whittall3, and Eduardo Narbona2
1Department of Plant Biology and Ecology, Facultad de Biología, Universidad de Sevilla, Seville, Spain, 2Department of
Molecular Biology and Biochemical Engineering, Facultad de Ciencias Experimentales, Universidad Pablo de Olavide, Seville,
Spain and 3Department of Biology, Santa Clara University, Santa Clara, CA, USA
*For correspondence. E-mail jvgarcia@us.es
Received: 17 August 2024 Returned for revision: 17 November 2024 Editorial decision: 3 December 2024 Accepted: 11 December 2024
Background and Aims Flower colour is a key feature in plant–pollinator interactions that makes the owers
visible amid the surrounding green vegetation. Green owers are expected to be scarcely conspicuous to pollin-
ators; however, many of them are visited by pollinators even in the absence of other traits that might attract pollin-
ators (e.g. oral scents). In this study, we investigate how entomophilous species with green owers are perceived
by pollinators.
• Methods We obtained reectance spectra data of 30 European species that display green or green–yellow
owers to the human eye. These data were used to perform spectral analyses, to calculate both chromatic
(colour contrast against the background) and achromatic (colour contrast that relies on the signals from the
green-sensitive photoreceptors) cues and to model colour perception by hymenopterans (bees) and dipterans
(ies).
Key Results The visibility of green owers to bees and ies (i.e. their chromatic contrast values) was lower
compared with other oral colours commonly pollinated by these insects, whereas green–yellow owers were
as conspicuous as the other ower colours. Green owers with low chromatic contrast values exhibited higher
achromatic contrast, which is used to detect distant owers at narrow visual angles, than green–yellow owers.
Additionally, the marker points (i.e. sharp transition in oral reectance that aids pollinators in locating them)
of green and green–yellow owers aligned to some degree with the colour discrimination abilities of bees and
ies.
• Conclusions We found that many entomophilous green and green–yellow owers are conspicuous to bees and
ies through their chromatic or achromatic contrasts. Although acquiring pigments such as carotenoids, which
impart a yellowish hue to owers and enhance their visibility to pollinators, could increase their conspicuousness,
the metabolic costs of pigment production, along with the use of alternative strategies to attract pollinators, might
have constrained carotenoid emergence in certain lineages of green-owered species.
Key words: Chromatic contrast, ower conspicuousness, green owers, plant–pollinator interaction, reectance
spectra, visual modelling, yellowish hue.
INTRODUCTION
Colour plays a key role in ecological interactions and evolu-
tionary processes. Conspicuous colours and ornamentations
are a widespread phenomenon in the animal kingdom that
can function in different ways during animal communication
(Endler, 1978, 1992; Stevens, 2013). Extravagant and colourful
ornaments in males are hypothesized to signal the overall in-
dividual quality to females, obtaining greater tness benets
through multiple mating (Darwin, 1871; but see Nolazco et
al., 2022). Aposematic animals are brightly coloured to adver-
tise that they are unprotable to potential predators, whereas
cryptic species use coloration to blend in with their environ-
ments and avoid detection (Strauss and Cacho, 2013; Cuthill
et al., 2017). In plants, colour serves as a vital cue in mutualist
interactions, such as pollination or seed dispersal (Willson and
Whelan, 1990; Stournaras et al., 2013). Flower colour is one of
the main components of plant–pollinator signalling, crucial for
enhancing ower visibility against the surrounding background
and for drawing the attention of pollinators (Van Der Kooi et
al., 2019). Flowers that maximize contrast with the natural
background, most often composed of green foliage, are more
easily detected by pollinators and can inuence their foraging
choices (De Ibarra et al., 2015; Bukovac et al., 2017; Finnell
and Koski, 2021).
Green owers, as perceived by humans, are relatively rare
in nature, representing approximately <10 % of species in dif-
ferent regions around the world (Weevers, 1952; Utech and
Kawano, 1975; Dyer et al., 2021; Wang et al., 2022). These
owers are commonly associated with wind pollination owing
to the absence of striking oral pigments that attract pol-
linators (Faegri and van der Pijl, 1979; Fenster et al., 2004;
© The Author(s) 2024. Published by Oxford University Press on behalf of the Annals of Botany Company.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/
by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited.
Downloaded from https://academic.oup.com/aob/advance-article/doi/10.1093/aob/mcae213/7920006 by guest on 07 January 2025
del Valle et al. Pollinators’ perception of green owers requires yellow
2
Friedman and Barret, 2009; Rosas-Guerrero et al., 2014).
Dyer et al. (2021) reported that ~45 % of the total European
ora that are pollinated abiotically (e.g. wind or water pollin-
ation) displayed green owers. Likewise, historical records of
the ora from eastern North America showed that 82 % of na-
tive green-owered species were wind pollinated, while small
and sometimes imperfect, entomophilous green owers were
sparingly visited by non-specialized insects and were also
self-compatible (Lovell, 1912). A likely explanation for the
rarity of pollinator-dependent species exhibiting green owers
is the expected low contrast they create against the predomin-
antly green background vegetation (Chittka et al., 1994; Moré
et al., 2020). In fact, many greenish and brownish ower col-
ours are inconspicuous to ower herbivores, thus protecting
plants from orivory (Kemp and Ellis, 2019). However, some
species that display green or greenish owers are successful
in attracting pollinators (Herrera et al., 2001; Alonso, 2005).
Some of these owers produce oral scents to attract pollin-
ators, such as in the case of two African Eucomis species,
whose green owers emit aromatic and monoterpene scents
that attract pompilid wasps (Shuttleworth and Johnson, 2009).
Hence, a specialized pollination syndrome has been dened
that combines cryptic green owers with scents as oral attract-
ants (Johnson et al., 2007; Brodmann et al., 2008). However,
other species that display green or greenish owers appear to
be odourless, at least to humans, such as Daphne laureola and
several species of Euphorbia, yet they are visited by generalist
pollinators (Herrera et al., 2001; Narbona et al., 2002; Alonso,
2005; Asenbaum et al., 2021). The owers of these species are
visited mainly by non-specialized pollinators from the orders
Hymenoptera and Diptera (Fenster et al., 2004; Lukas et al.,
2020; Asenbaum et al., 2021). Although rare, bird pollination
has also been documented in some other green-owered spe-
cies (Faegri and van der Pijl, 1979; Rebelo and Siegfried, 1985;
Rathcke, 2000; Mizuno et al., 2024).
Although abiotic factors might inuence ower colour evolu-
tion, this trait is shaped primarily by biotic selection pressures,
with pollinators acting as key agents and driving oral reect-
ance to align with their visual sensitivities (Fenster et al., 2004;
Shrestha et al., 2013b; Dyer et al., 2021; Dorin et al., 2023).
In this context, it is crucial to determine how green owers are
perceived visually by their most common pollinators, namely
hymenopterans and dipterans (Supplementary Data Table S1).
Regarding hymenopterans, the honeybee Apis mellifera is used
as a model organism for studies on colour vision (Van Der Kooi
et al., 2021). Honeybees and other hymenopterans are trichro-
matic, having three photoreceptors, with peak sensitivities in the
ultraviolet (UV), blue and green regions of the light spectrum
(Peitsch et al., 1992; Briscoe and Chittka, 2001; Chittka and
Raine, 2006; Dyer et al., 2008). Dipterans are another important
group of pollinators, although their visual system has received
less attention (Garcia et al., 2022). Blowies and hoveries
typically have four sensitivity peaks covering the UV, blue and
green–yellow wavelengths (Troje, 1993; An et al., 2018; Hannah
et al., 2019), although spectral sensitivity varies widely between
species (reviewed by Van Der Kooi et al., 2021). Therefore, the
visual systems of pollinating bees and ies, both sensitive to
the green region of the visible spectrum, suggest that they can
perceive green owers. In addition, the visual system of pol-
linators demonstrates optimal colour discrimination in specic
regions of the light spectrum, typically aligning with the wave-
length range where the sensitivity of two photoreceptors over-
laps (Chittka and Menzel, 1992; Shrestha et al., 2013b). Yet, the
reectance spectra of owers can display areas characterized by
abrupt transitions (referred to as marker points), indicating sharp
changes in the spectrum (Shrestha et al., 2013). If marker points
align with the wavelengths associated with optimal discrimin-
ation ability of a pollinator group, it might indicate an adaptation
of the oral colour to the visual system of the animal (Dyer et
al., 2012; Shrestha et al., 2013a, 2016; Camargo et al., 2019).
Thus, bees show their maximum discrimination wavelengths
at 400 and 500 nm (Chittka and Menzel, 1992; Shrestha et al.,
2013a), whereas the wavelengths of maximum discrimination
for ies have not yet been studied.
In addition to the ability of pollinators to perceive ower col-
ours, foraging relies on the contrast between the oral stimulus
and its background (i.e. chromatic contrast), typically green fo-
liage, and this contrast aids in ower detection, attracting pol-
linators to the food source (Menzel and Backhaus, 1991). For
bees, this parameter is crucial, because they use chromatic sig-
nals to detect a target when foraging in short distances (Giurfa et
al., 1996), and it is likely to play a key role in stimulus detection
in ies (An et al., 2018). Additionally, achromatic contrast (i.e.
contrast based on signals from green-sensitive photoreceptors)
has been suggested to be important in detection and discrimin-
ation tasks in bees, because they rely on this parameter when
foraging from long distances (Spaethe et al., 2001). However,
achromatic contrast is less well characterized in ies (van der
Kooi and Kelber, 2022). For green owers, the degree of con-
spicuousness (i.e. the chromatic and achromatic contrasts) of
these owers against a background of green foliage has not yet
been quantied, nor has their attractiveness to pollinators been
evaluated relative to other oral colours.
The goal of this study was to assess how owers that appear
green or greenish to human eyes are perceived by pollinators.
We determined whether these owers are inconspicuous, as
traditionally thought based on human perception, or whether,
after accounting for pollinator visual systems, they are more
conspicuous than expected. Specically, we analysed the re-
ectance spectra of 30 entomophilous species that display
green owers found in Europe, where hymenopterans (bees)
and dipterans (ies) constitute the primary pollinator groups
(Ashworth et al., 2015). Although green owers are rare,
greenish-yellow owers are more common, hence we chose
to include both groups and compare their conspicuousness to
pollinators in contrast to their backgrounds. We examined the
chromatic and achromatic contrasts of these owers according
to the vision models of their main pollinators and compared
their conspicuousness with other ower colours. Furthermore,
the position of the marker points in the reectance spectra of
the green and green–yellow owers were compared with the
spectral discrimination sensitivities of bees to assess the cor-
respondence between reectance spectra of these owers and
the colour discrimination ability of this pollinator (Chittka
and Menzel, 1992; Shrestha et al., 2013a). We specically ad-
dressed the following questions. What is the perception by the
main pollinators (i.e. bees and ies) based on their visual sys-
tems, of colour stimuli generated by green and green–yellow
owers? Do green and green–yellow owers have higher or
lower chromatic and achromatic contrast than owers of other
Downloaded from https://academic.oup.com/aob/advance-article/doi/10.1093/aob/mcae213/7920006 by guest on 07 January 2025
del Valle et al. Pollinators’ perception of green owers requires yellow 3
colours in the visual systems of bees and ies? Are green and
green–yellow owers equally visible to these pollinators?
Do the spectral characteristics of these owers align with the
colour discrimination abilities of pollinators? All this informa-
tion would illuminate whether green and green–yellow owers
are similarly conspicuous to their pollinators.
MATERIALS AND METHODS
Study species
We studied 30 entomophilous European species that display
green owers as perceived by the human eye, although some
of them have a yellowish hue in human vision (we refer to
them as ‘green–yellow owers’ hereafter). The spectrum of all
owers generally showed very low reectance values in the UV
(300–400 nm) and blue (400–500 nm) regions of the visible
(VIS) spectrum, with a primary reectance peak occurring in
the green (500–550 nm) region of the VIS light, which then
decreased until reaching a valley at ~670 nm (Supplementary
Data Fig. S1). A principal component analysis using reect-
ance data from 300 to 700 nm and colour as a grouping variable
(i.e. green and green–yellow) was conducted to conrm the val-
idity of the separation between green and green–yellow owers
based on human colour perception (Supplementary Data Fig.
S2). Examples of species displaying green owers and those
with a yellowish coloration include Rhamnus lycioides and
Daphne laureola, respectively (for additional examples, see
Fig. 1).
Floral reectance data
For 19 of the 30 species, we conducted measurements of nat-
urally growing owers in the eld from populations located in
A
D
G H I
E F
B C
F. 1. Pictures of some of the species displaying green (A–G) or green–yellow (H, I) owers included in this study: (A) Aristolochia paucinervis (Aristolochiaceae);
(B) Arum italicum (Araceae); (C) Helleborus foetidus (Ranunculaceae); (D) Rhamnus lycioides (Rhamnaceae); (E) Paris quadrifolia (Melanthiaceae); (F) Rubia
peregrina (Rubiaceae); (G) Euphorbia nicaeensis (Euphorbiaceae); (H) Daphne laureola (Thymelaeaceae); and (I) Viscum cruciatum (Santalaceae).
Downloaded from https://academic.oup.com/aob/advance-article/doi/10.1093/aob/mcae213/7920006 by guest on 07 January 2025
del Valle et al. Pollinators’ perception of green owers requires yellow
4
southern Spain (Supplementary Data Table S1). We acquired
the oral reectance data in the laboratory from owers col-
lected in the eld using a Jaz portable spectrometer equipped
with a deuterium–tungsten halogen light source (Ocean Optics,
Dunedin, FL, USA) and a black metal probe holder (6 mm
diameter opening at 45°). We measured reectance curves from
300–700 nm wavelengths at 0.4 nm intervals, setting an inte-
gration time of 2 s and smoothing boxcar width of 12 (Del Valle
et al., 2018). Reectance measurements were calibrated with a
white standard (WS-1-SL, Ocean Optics) and analysed using
both Spectra Suite v.10.7.1 and OceanView v.2.0 software. For
each species, we selected between one and ten owers (always
using a single ower per plant) to measure the reectance of
petals. Alternatively, we measured the reectance of bracts,
spathes or the larger oral structures of an inorescence that
could attract pollinators (for further details, see Supplementary
Data Table S1). We used the ‘aggplot’ function of the pavo R
package (Maia et al., 2019) to aggregate all the spectra from
the same species, then to obtain the average and standard de-
viation (s.d.) values. For the remaining 11 species, we down-
loaded oral reectance data (one spectrum per species;
Supplementary Data Table S1) from the Floral Reectance
Database (http://www.reectance.co.uk/; hereafter ‘FReD’)
(Armold et al., 2010). We conducted a search in FReD, l-
tering for the plant attribute ‘Human Colour: green’ (accessed
25 November 2021). We considered only those entries located
in Europe and the Eastern Mediterranean. The results obtained
were validated with a subsequent bibliographical search to
conrm that the oral coloration of these species was predom-
inantly or entirely human green. Data from both direct meas-
urements and FReD were processed equally, as described in the
next section, to be used in spectral analyses.
Spectral analysis
For reectance spectra analyses, we considered wavelengths
between 300 and 700 nm (Briscoe and Chittka, 2001; Endler
and Mielke, 2005). We used the ‘procspec’ function of the pavo
R package (Maia et al., 2019) to process reectance curves.
We smoothed reectance curves with a smoothness parameter
of 0.20. Subsequently, we removed negative values from four
samples to correct minor deviations during calibrations by set-
ting the minimum value to zero and scaling other values ac-
cordingly (White et al., 2015).
Perception of green and green–yellow owers by pollinators
We used animal vision models to represent how a colour
stimulus is perceived by the two main functional groups of pol-
linators considered in this study: hymenopterans (bees) and dip-
terans (ies). We plotted the processed reectance curves as loci
in the vision models of both pollinator groups using the function
colspace’ in the pavo R package (Maia et al., 2019), as described
by Narbona et al., (2021a). For bees, the most widely used vi-
sion model is the so-called ‘colour hexagon’ of the honeybee A.
mellifera, which represents a continuous spectrum of colour per-
ception, encompassing UV, UV–blue, blue, blue–green, green
and UV–green regions (Chittka, 1992; Chittka and Wells, 2004;
Chittka and Kevan, 2005). For ies, we used the space model
for the tetravariant visual system of Lucilia sp. developed by
Troje (1993). This colour space has four quadrants that repre-
sent a continuous spectrum of colours, including UV, purple,
blue and yellow (Troje, 1993; Lunau, 2014). Although bee and
y colour vision models are categorical, recent studies demon-
strate that colour discrimination of both groups of insects could
be continuous (Garcia et al., 2017, 2022; Hannah et al., 2019).
We implemented von Kries transformation for both models,
which assumes that eye photoreceptors are adapted to the spec-
tral properties of the illumination and background (White et al.,
2015). We used the standard daylight function (D65 irradiance
function) as illuminant and the average spectrum of green fo-
liage of 230 species proposed by Chittka (1992) as background
in the vision models, which is the most commonly used set-up
in similar studies (Chen et al., 2020; Dyer et al., 2021; Narbona
et al., 2021a; León-Osper and Narbona, 2022). For the specic
case of bees, we considered a hyperbolic transformed quantum
catch. We obtained the chromatic contrast against the back-
ground, i.e. the distance between the colour loci of the ower
and the achromatic centre measured in Euclidean units (here-
after, EU), according to the photoreceptor spectral sensitivities
of the two colour space models (Rohde et al., 2013; Hannah
et al., 2019; Garcia et al., 2022). Additionally, we considered
the empirically determined discriminability thresholds of 0.11
EU for A. mellifera and 0.096 EU for Eristalis tenax (putatively
extendable to other hymenopterans and dipterans, respectively)
as an estimate of whether a colour stimulus is easily discrim-
inated from the background (Dyer et al., 2012; Garcia et al.,
2022). It should be noted that these thresholds must be handled
with caution, because they were established in laboratory con-
ditions and do not consider additional factors that might signi-
cantly impact the chromatic contrast of the ower, such as the
surrounding background or the presence of other nearby stimuli
(Giurfa et al., 1997; Bukovac et al., 2017). For bees, we also cal-
culated the achromatic contrast (i.e. green contrast), which is the
modulation of the green receptor against the background and is
calculated as the excitation value of the green photoreceptor dif-
ferent from 0.5, which represents adaptation to the background
(Spaethe et al., 2001).
Flower colour conspicuousness in relationship to other ower
colours
To compare the conspicuousness (i.e. chromatic and achro-
matic contrasts) of green (n = 19) and green–yellow (n = 11)
owers relative to more common ower colours in nature, we
used a dataset comprising 100 species categorized into four
colour groups based on human perception of their owers:
yellow (n = 38), blue–violet (n = 28), pink (n = 25) and white
(n = 9) (Supplementary Data Fig. S3). The reectance data
for non-green owers were sourced from a previous study on
ower colour perception considering the visual systems of the
main groups of pollinators present in the Mediterranean Basin
(Narbona et al., 2021a).
Spectral signals: marker points
To identify marker points of green and green–yellow owers
in the UV–VIS range (300–700 nm), we used the software
Downloaded from https://academic.oup.com/aob/advance-article/doi/10.1093/aob/mcae213/7920006 by guest on 07 January 2025
del Valle et al. Pollinators’ perception of green owers requires yellow 5
Spectral-MP developed by Dorin et al. (2020) with the recom-
mended settings: changes in reectance of ≥10 % occurring
over a wavelength range of 50 nm with a smoothing window of
±10 data points and considering ve data points to look ahead
when performing slope change detection. We also calculated
the frequency of marker points in wavelength bins of 10 nm
across the UV–VIS reectance spectra for all species. Then,
we measured the matching between detected marker points
and the wavelengths of maximum discrimination of bees. To
do so, we calculated two metrics: the mean absolute deviation
(MAD) and the minimal absolute deviation (minAD) (Shrestha
et al., 2013a). MAD measures the average proximity of ower
marker points to each wavelength of maximum discrimin-
ation, and minAD measures the minimum distance between
marker points and each specic wavelength of optimal visual
discrimination (for bees, we thus obtain one MAD value and
two minAD values: minAD400 and minAD500) (Shrestha et al.,
2013a). For both metrics, the smaller values imply a closer t
between oral reectance spectra to a particular visual system.
For comparison, we also identied the marker points and cal-
culated the MAD and minAD spectral metrics for the 28 blue–
violet owers considered in this study. This comparison was
performed considering that bees have an innate preference for
blue and violet owers (Giurfa et al., 1995).
Statistical analyses
Differences in conspicuousness (i.e. achromatic and chro-
matic contrasts) between green owers and other ower colour
categories, and between green and green–yellow owers, were
analysed using phylANOVAs (Garland et al., 1993). This ap-
proach controls for the potential inuence of phylogeny when
analysing differences among ower colour groups. Initially,
we built a phylogenetic tree using the ‘phylo.maker’ function
implemented in the V.PhyloMaker R package (Jin and Qian,
2019), which relies on the angiosperm megatree as a phylo-
genetic backbone (Zanne et al., 2014; Smith and Brown,
2018). Then, we performed phylANOVAs using the phytools
R package (Revell, 2012), with 10 000 simulations for each
test and Holm-adjusted P-values for post hoc comparisons.
PhylANOVAs were also used to test for differences in the
MAD and minAD parameters between green, green–yellow
and blue–violet owers. The principal component analysis used
to assess the suitability of grouping the species into green or
green–yellow categories was performed using the R package
factoextra (Kassambara and Mundt, 2020). We used RStudio
v.2022.12.0 (RStudio Team, 2022) to perform all statistical
analyses.
RESULTS
Pollinators’ perception
For the bee vision, most green (13 of 19) and green–
yellow (9 of 11) owers occupied the green category in the
hexagon colour space (Fig. 2A; Supplementary Data Table
S2). The chromatic contrast of the green owers was vari-
able, ranging from 0.035 to 0.338 EU, with a mean ± s.d. of
0.107 ± 0.068 EU (Fig. 2C), and the mean of green–yellow
owers was signicantly higher (0.293 ± 0.082 EU;
phylANOVA: F1,29 = 44.8, P < 0.001), with values ran-
ging between 0.184 and 0.424 EU. Eleven of the 19 green
owers showed a chromatic contrast of <0.11 EU, the the-
oretical discriminability threshold for bees, whereas none
of the green–yellow owers did (Fig. 2C; Supplementary
Data Table S2). These data suggest that the green owers
are, presumably, more difcult for bees to perceive. When
examining the achromatic contrast of green owers, the mean
was 0.176 ± 0.089 EU (range = 0.036–0.322 EU), which was
statistically higher than green–yellow owers(mean = 0.102
± 0.073 EU; range = 0.001–0.200 EU; phylANOVA for spe-
cic comparison green vs. green–yellow owers: F1,29 = 5.51,
P = 0.045; Supplementary Data Fig. S4; Table S2).
For the y vision, green owers concentrated the majority
of colour loci in the yellow and the purple sectors (10 and 6
of 19, respectively), whereas all green–yellow owers except
one occupied the yellow section (Fig. 2B; Supplementary
Data Table S2). The chromatic contrast was variable, ranging
from 0.041 to 0.350 EU for green owers, with a mean ± s.d.
of 0.124 ± 0.070 EU. The chromatic contrast values obtained
from green–yellow owers was signicantly higher than for
green owers (0.312 ± 0.087 EU; phylANOVA: F1,29 = 42.3,
P < 0.001), ranging between 0.197 and 0.453 EU. Seven of
the 19 green owers and none of the green–yellow owers
were below the theoretical discriminability threshold for ies
(0.096 EU; Fig. 2D; Supplementary Data Table S2), suggesting
that they might be more difcult for ies to perceive. As in-
dicated above, achromatic contrast is not available for the y
visual system because it is not yet well dened.
Conspicuousness of green and green–yellow owers compared
with other ower colours
The phylogenetic ANOVA revealed signicant differences
among ower colours regarding their visibility (i.e. chromatic
contrast) to bees (phylANOVA: F5,129 = 18.02, P < 0.001) and
to ies (phylANOVA: F5,129 = 16.52, P < 0.001). For both bees
and ies, green owers showed the lowest chromatic contrast
values. For bees, the chromatic contrast of green owers did not
differ from those of blue–violet, pink and white owers (Fig.
3A), although this was not the case for ies (Fig. 3B). Green–
yellow owers exhibited signicant higher conspicuousness
than green owers for both bees and ies (Fig. 3). For bees,
we also found signicant differences among ower colours
when analysing the contrast to the green receptor of bees (i.e.
achromatic contrast; phylANOVA: F5,129 = 23.57, P < 0.001).
Achromatic contrast values were statistically greater in green
relative to green–yellow owers, and white owers had the
highest achromatic contrast values (Supplementary Data
Fig.S5).
Analysis of marker points
Four of the 30 studied species (13 %) presented reect-
ance spectra that did not generate marker points with the re-
commended settings (Supplementary Data Fig. S6). Green
owers exhibited most marker points clustered near 500 nm
(45.2 % of inection points were in the 480–520 nm range;
Downloaded from https://academic.oup.com/aob/advance-article/doi/10.1093/aob/mcae213/7920006 by guest on 07 January 2025
del Valle et al. Pollinators’ perception of green owers requires yellow
6
Fig. 4A) and 600 nm (54.8 %). Green–yellow owers, in
contrast, concentrated a higher proportion of marker points
at 500 nm (61.1 %), with the remaining found at 600 nm
(38.9 %) (Fig. 4B). When analysing marker points in the re-
ectance spectra of blue–violet owers, we found an aggre-
gation around 500 nm (22.9 % in the 480–520 nm range) and
at wavelengths of >600 nm (37.1 %). However, in this case,
marker points were also concentrated near 400 nm (27.1 % in
the 380–420 nm range) or even within the UV range (12.9 %
in the 300–380 nm; Fig. 4C).
When comparing MAD and minAD values among green,
green–yellow and blue–violet owers, we found no statistical
differences for the MAD and minAD500 parameters (Table 1).
However, minAD400 differed signicantly among oral colours
(phylANOVA: F2,57 = 91.1, P = 0.001; Table 1). Blue–violet
owers demonstrated a signicantly better match for optimal
bee discrimination at 400 nm, with values nearly 6-fold lower
(closer to the bee optimal) than those obtained for the green and
green–yellow owers.
DISCUSSION
Our results suggest that a considerable number of species with
green owers in Europe, particularly those with a green–yellow
hue, appear to be detectable visually by hymenopterans and
dipterans, the two most important pollinator groups in this re-
gion. Indeed, green–yellow owers were as conspicuous as
other ower colours commonly visited by both groups of pol-
linators. Although the canonical view is that green owers lack
both chromatic and achromatic cues, we demonstrated that the
spectral characteristics of green and green–yellow owers en-
able many of them, particularly the green–yellow ones, to be
Blue-Green
–1.0 –0.5 0 0.5 1.0
–1.0 –0.5 0 0.5 1.0
Blue
Blue-UV
DC
BA
UV Green
UV-Green
Euclidean Units Euclidean Units
0 0.1 0.2 0.3 0.4 0.5 0 0.1 0.2 0.3 0.4 0.5
R7y – R8y
R7p – R8p
Yellow
Blue UV
Purple
Euonymus europaeus
Alchemilla alpina
Aristolochia paucinervis
Arum italicum
Osyris lanceolata
Matricaria discoidea
Euphorbia peplus
Euphorbia nicaeensis
Euphorbia segetalis
Euphorbia boetica
Euphorbia terracina
Euphorbia helioscopia
Rhamnus lycioides
Euphorbia serrata
Rubia peregrina
Asparagus horridus
Narcissus viridiflorus
Paris quadrofolia
Alchemilla fissa
Tofieldia calyculata
Withania frutescens
Chrysosplenium alternifolium
Euphorbia hierosolymitana
Alchemilla vulgaris
Alchemilla glabra
Rhamnus oleoides
Helleborus foetidus
Viscum cruciatum
Matricaria aurea
Daphne laureola
Euonymus europaeus
Alchemilla alpina
Aristolochia paucinervis
Arum italicum
Osyris lanceolata
Matricaria discoidea
Euphorbia peplus
Euphorbia nicaeensis
Euphorbia segetalis
Euphorbia boetica
Euphorbia terracina
Euphorbia helioscopia
Rhamnus lycioides
Euphorbia serrata
Rubia peregrina
Asparagus horridus
Narcissus viridiflorus
Paris quadrofolia
Alchemilla fissa
Tofieldia calyculata
Withania frutescens
Chrysosplenium alternifolium
Euphorbia hierosolymitana
Alchemilla vulgaris
Alchemilla glabra
Rhamnus oleoides
Helleborus foetidus
Viscum cruciatum
Matricaria aurea
Daphne laureola
F. 2. (A, B) The colour loci of the 30 species that display green (dark green dots) or green–yellow (lime green dots) owers are represented in the hexagon
colour space of bees (A) and the tetravariant visual system of ies (B). (C, D) The chromatic contrast values of the 30 species that display green (dark green bars)
or green–yellow (lime green bars) owers, obtained in the pollinator visual systems of bees (C) and ies (D), are shown. Species have been sorted by increasing
values obtained from each vision model. Chromatic contrast values are expressed in Euclidean units (EU). The dotted lines represent the theoretical discrimin-
ability threshold for hymenopteran and dipteran vision models (0.11 and 0.096 EU, respectively). Animal silhouettes were taken from Divulgare (http://divulgare.
net/) under a Creative Commons licence.
Downloaded from https://academic.oup.com/aob/advance-article/doi/10.1093/aob/mcae213/7920006 by guest on 07 January 2025
del Valle et al. Pollinators’ perception of green owers requires yellow 7
distinguished visually from the surrounding green background.
This distinctiveness renders them conspicuous to both bees and
ies. Below, we discuss the potential adaptive signicance of
green and green–yellow owers with respect to the visual sys-
tems of their pollinators.
Not all green owers are equally conspicuous to bees and ies
Green owers were less visible to pollinators, on average,
in comparison to the green–yellow ones, which were nearly
three times more conspicuous. Approximately 58 % of the
pure green-owered species studied here showed chromatic
contrast values lower than the theoretical discriminability
threshold for bees, and 37 % showed values lower than the
threshold for ies. This suggests that these owers might be
more challenging for these insects to distinguish from the
background. Green owers showed lower chromatic con-
trast than some of the other ower colours, although adding a
yellowish hue enhanced their visibility to both bees and ies.
For bees, green owers showed lower chromatic contrast than
green–yellow and yellow owers but not blue, white or pink
owers. For ies, green owers showed lower chromatic con-
trast relative to all other ower colours tested. As far as we
know, green and green–yellow owers have not been included
explicitly in previous studies analysing chromatic contrast in
the visual systems of honeybees and ies. The mean chromatic
contrast values for other typical bee-pollinated ower colours
were comparable to those observed in green and green–yellow
0.8
A
B
0.6
0.4
c
b
bc bc
bc
a
0.2
0
Euclidean distanceEuclidean distance
0.8
0.6
0.4
b
a
aa
a
a
0.2
0
Green Green-yellow Blue-violet WhitePink Yellow
F. 3. Violin plots with boxplots represent the distribution of chromatic contrast values, expressed in Euclidean units (EU), obtained from the vision model of
bees (A) and ies (B). The colour of each violin plot corresponds to the six ower colour categories considered: green (n = 19), green–yellow (n = 11), blue–violet
(n = 28), pink (n = 25), white (n = 9) and yellow (n = 38). Boxplots represent the interquartile range, the thin black line depicts 1.5 × interquartile range, the cen-
tral line displays the median, and the red and small black points represent the mean and all sample values, respectively. Violin plots show the full distribution of
sample values for each ower colour category. Different letters indicate signicant differences at the 0.05 level. Animal silhouettes were taken from Divulgare
(http://divulgare.net/) under a Creative Commons licence.
Downloaded from https://academic.oup.com/aob/advance-article/doi/10.1093/aob/mcae213/7920006 by guest on 07 January 2025
del Valle et al. Pollinators’ perception of green owers requires yellow
8
owers. For instance, Coimbra et al. (2020) found that the
average chromatic contrast for bee-pollinated owers across
>200 species worldwide ranged from 0.13 EU for pink owers
to 0.22 EU for white owers. This range of values encom-
passes those we found for approximately one-third of green
owers and for all green–yellow owers (0.131–0.424 EU).
Likewise, in another study on bee-pollinated species from the
Mediterranean Basin displaying different ower colours, the
mean chromatic contrast value was 0.23 EU, slightly lower
than that of the green–yellow owers presented here (0.293 EU
on average) (Ortiz et al., 2021). Our results suggest that cer-
tain species with green owers, particularly when combined
with yellow hues, might be as conspicuous to bees as other
ower colours, at least based on their chromatic contrast. For
ies, however, green owers would be likely to benet from
the addition of yellow pigmentation to enhance their visual at-
traction. Although the perceived chromatic contrast between a
ower and its surroundings plays an important role in ower
detection and recognition by pollinators (Giurfa et al., 1996;
Kelber et al., 2003; Dyer and Chittka, 2004; van der Kooi et
al., 2019), it is important to note that chromatic contrast is only
one of several factors that inuence pollinator preferences.
Attributes such as oral colour, ower size, scent, rewards and
pollinator learning also play signicant roles, because colour
represents only one component of the multisensory cues that
guide pollinator foraging.
The comparison of marker points among green, green–
yellow and blue–violet owers revealed both differences and
similarities. All three ower colours showed a signicant
number of marker points in the red region (600–700 nm) of the
light spectrum. This region lacks relevance for bees owing to
their long-wavelength photoreceptors having low sensitivity at
these wavelengths (Chittka and Wells, 2004), but probably en-
hances detection by other insects that might have sensitivities
close to these longer wavelengths. Typically, bee-pollinated
owers show a high number of marker points in two regions
of the light spectrum, around 400 nm (violet–blue) and 500 nm
(blue–green) (Giurfa et al., 1995; Dyer et al., 2012; Bischoff et
al., 2013), where the visual system of bees is maximally sen-
sitive (Chittka and Menzel, 1992). In our study, blue–violet
owers had marker points in these two regions of the light
spectrum, suggesting integration of this ower colour with
the visual system of bees (Shrestha et al., 2013a). Conversely,
green and green–yellow owers clustered their spectral marker
points primarily around 500 nm, consistent with the statis-
tical differences found in the comparison of minAD400 but not
in the case of the minAD500 spectral metric. Thus, the spectral
signatures of green owers would activate only one region of
maximum discrimination at 500 nm, as opposed to the two re-
gions in blue owers (400 and 500 nm). Moreover, bees have
an innate, precise ability to discriminate colours around 400 nm
(Menzel, 1967; Giurfa et al., 1995; Chittka and Wells, 2004).
The implications of this are not entirely clear. Although it is true
that ower-naïve bees frequently prefer blue–violet and green
colours, their innate preference for the former is linked to the
higher nectar rewards associated with these owers (Giurfa et
al., 1995). This preference is thought to be inherited, reecting
a phylogenetically ancient tendency (Chittka and Wells, 2004).
0.4
A
B
C
0.3
0.2
Relative frequency
0.1
0
0.4
0.3
0.2
Relative frequency
0.1
0
0.15
0.10
0.05
Relative frequency
0300 350 400 450 500
Wavelength (nm)
550600 650700
F. 4. Relative frequency distribution of marker points found in species
displaying green (A), green–yellow (B) and blue–violet (C) owers. Dotted
lines at 400 and 500 nm indicate the hue-discrimination optima of the hymen-
opteran visual system. Animal silhouettes were taken from Divulgare (http://
divulgare.net/) under a Creative Commons licence.
T . Comparisons of mean absolute deviation (MAD) and
minimal absolute deviation (minAD) values for hymenopterans be-
tween green, green–yellow and blue–violet owers. Mean ± s.d.
values for MAD and minAD parameters are shown; ‘ns’ indicates
a non-signicant P-value.
Parameter Green
owers
Green–yellow
owers
Blue–violet
owers
Signicance
MAD
Hymenoptera 96.5 ± 57.4 70.8 ± 48.5 77.0 ± 15.8 ns
minAD400 125.3 ± 45.5 116.5 ± 4.9 19.9 ± 19.1 0.001
minAD500 25.3 ± 45.5 16.5 ± 4.9 42.0 ± 41.4 ns
Downloaded from https://academic.oup.com/aob/advance-article/doi/10.1093/aob/mcae213/7920006 by guest on 07 January 2025
del Valle et al. Pollinators’ perception of green owers requires yellow 9
Unlike bees, ies typically prefer owers with higher re-
ectance at longer wavelengths, especially yellow owers
(560–580 nm) (Briscoe and Chittka, 2001; Lunau, 2014;
Hannah et al., 2019). Shrestha et al. (2019) found that, con-
trary to bee-pollinated orchids, y-pollinated ones present
their marker points at longer wavelengths, but never at wave-
lengths of <500 nm. Likewise, the oral spectra of native ora
from the Macquarie Islands, consisting of pale cream–green
owers exclusively pollinated by ies, were characterized by
spectral marker points predominantly occurring around 510
and 690 nm, occupying the yellow quadrant in the colour vi-
sion system of ies (Shrestha et al., 2016). Thus, the concen-
tration of marker points in long wavelengths (500–700 nm) of
European species with green and green–yellow owers might,
potentially, facilitate colour discrimination by ies, at least for
syrphid ies (An et al., 2018; Hannah et al., 2019). In addition,
a substantial proportion of the species studied here clustered in
the yellow region of the colour space model for the y vision,
especially among green–yellow owers, which is consistent
with the preference of ies for long-wavelength-rich, ‘yellow’
coloration (Lunau, 1990; Shrestha et al., 2016, 2019; An et al.,
2018). Finally, the low UV reectance of green–yellow owers
is also in line with the innate preferences of generalist dipterans
for yellow colours without UV reection, probably owing to
the effects of UV wavelengths on ower brightness and colour
saturation, which inuences colour choice in ies (An et al.,
2018). Our ndings suggest that the spectral signatures of most
green and green–yellow owers, including their marker points,
appear to be adjusted, to some extent, to the visual capabilities
of dipterans.
One of the most remarkable results of this study is the en-
hanced conspicuousness of green–yellow relative to green
owers for both bees and ies. Given that no distinctive fea-
tures of surface cells responsible for structural colours have
been reported previously in green or green–yellow owers
(Kay et al., 1981; Yuan et al., 2023), we can infer that ower
conspicuousness is inuenced primarily by pigments accumu-
lated in cells of oral advertising structures (Van Der Kooi et
al., 2019; van der Kooi, 2021). Except for rare green–bluish
colours produced by highly decorated anthocyanins (Mizuno
et al., 2021), the green colour of owers is produced primarily
by the accumulation of chlorophylls (Vignolini et al., 2012;
Narbona et al., 2021b; Yuan et al., 2023). Indeed, the decrease
from 550 to 650 nm and the dip at 670 nm observed in the re-
ectance spectra of green or green–yellow owers align with
the light absorption characteristics of chlorophylls, particularly
chlorophyll a (Lee, 2007; Narbona et al., 2021a). Likewise,
the absence or very low reectance between 400 and 500 nm
in some green owers, particularly in the green–yellow ones,
is likely to be attributable to light absorption by carotenoids
(Lee, 2007; Narbona et al., 2021b), which typically co-occur
with chlorophylls (Vignolini et al., 2012; Narbona et al.,
2021b; Yuan et al., 2023). We propose that the selective pres-
sures favouring increased visibility to pollinators might op-
erate by reducing chlorophyll concentrations and increasing
carotenoid levels in green owers, thereby distinguishing the
green–yellow owers from the leaf background. However, not
all green owers need to adopt this strategy to enhance their
visibility to pollinators. The development of green–yellow pig-
mentation to enhance conspicuousness might be constrained
by a combination of phylogenetic, ecological and evolutionary
factors, as plants adopt diverse strategies to attract pollinators
based on their specic ecological niches. Furthermore, pig-
ment production involves a metabolic cost (ChalkerScott,
1999; Gould, 2004), which suggests that most angiosperms
are likely to produce only the minimum necessary pigments in
their owers, presumably compensating with alternative strat-
egies to attract pollinators.
Less noticeable green owers might display additional traits to
attract pollinators
We identied several green-owered species with low chro-
matic contrasts that appear less conspicuous to bees and ies.
However, this does not necessarily mean that these subtler
owers go unnoticed by pollinators. Colour detection in the
visual system of hymenopterans involves both chromatic and
achromatic visual processing, with the use of each or their com-
bination depending on the size and distance to a ower (Giurfa
et al., 1996; Dyer et al., 2008; Ortiz et al., 2021; van der Kooi
and Kelber, 2022). Additionally, these factors can vary consid-
erably depending on surrounding stimuli, lighting conditions
or even environmental context, among others. As a result, the
same physical stimulus might be interpreted as an entirely dif-
ferent colour, potentially inuencing how bees respond to iden-
tical spectral cues (Garcia et al., 2021; Lunau and Dyer, 2024;
Shrestha et al., 2024). In this study, species such as Alchemilla
alpina, Arum italicum and most species of Euphorbia showed
low chromatic contrasts while displaying high achromatic con-
trasts (Supplementary Data Fig. S4). High achromatic contrasts
might benet these less visually striking species by allowing
honeybees and bumblebees to detect their owers from long
distances (Giurfa et al., 1996; Spaethe et al., 2001; Dyer et al.,
2008; but see Ng et al., 2018). Green owers might have other
colour-related characteristics that enhance their attractiveness
to pollinators, such as colour patterns, that improve ower de-
tection by pollinators (De Ibarra et al., 2015), as seen in Nigella
orientalis (Yuan et al., 2023). Likewise, green owers can use
alternative or complementary cues to lure bees or other insects.
For instance, green-owered species not included in this study,
such as Veratum album or Hedera helix, produce a foul scent
that attracts ies and ants (Faegri and van der Pijl, 1979; Kato
et al., 2009; Lukas et al., 2020). Pollinator foraging decisions
are inuenced by a variety of factors, with colour being one
of the most signicant, although alternative or complementary
cues are clearly important in ower identication. For instance,
the signicance of scent and colour can vary, shaped by the par-
ticular combination of these traits and the unique preferences
of each bee (Chittka and Wells, 2004). Hence, other oral traits
might enhance the detectability of owers that are less prom-
inent, such as the green ones in this study, making them more
noticeable to pollinators.
CONCLUSION
In conclusion, entomophilous green owers have traditionally
been considered invisible to pollinators, relying instead on scent
cues for their attractiveness (Faegri and van der Pijl, 1979). Our
results question this assertion, at least in part, because many
Downloaded from https://academic.oup.com/aob/advance-article/doi/10.1093/aob/mcae213/7920006 by guest on 07 January 2025
del Valle et al. Pollinators’ perception of green owers requires yellow
10
species with green or green–yellow owers can be perceived
by both bees and ies. The combination of chlorophylls and
carotenoids appears to play an important role in making green
owers visually more noticeable to pollinators (e.g. Martins et
al., 2021; Ortiz et al., 2021). An intriguing question stemming
from our study is why entomophilous green owers persist in-
stead of transitioning to green–yellow owers, which we predict
would be advantageous for pollinator detection. Although many
green owers can obtain a fraction of energy and carbon directly
through photosynthesis within their tissues (Bazzaz et al., 1979),
the metabolic cost associated with acquiring new pigments
(ChalkerScott, 1999; Gould, 2004), especially when alternative
strategies to attract pollinators might be available, might have
constrained the emergence of carotenoids in certain lineages.
SUPPLEMENTARY DATA
Supplementary data are available at Annals of Botany online
and consist of the following.
Table S1: information about the 30 species with green or
green–yellow owers analysed in this study. Table S2: list of
species used in this study along with their ower colour as per-
ceived by the human eye. Figure S1: UV–VIS spectral reect-
ance of the 30 species that display green (A) or green–yellow
(B) owers included in this study. Figure S2: principal compo-
nent analysis of the reectance data from 30 species, grouped
according to human visual perception into green (dark green
circles) and green–yellow (lime green triangles) owers. Figure
S3: average UV–VIS spectral reectance of typical blue–violet
(n = 28), pink (n = 25), white (n = 9) and yellow (n = 38)
owers. Figure S4: achromatic contrast values of the 30 spe-
cies that display green (dark green bars) or green–yellow (lime
green bars) owers, obtained in the pollinator visual systems
of bees, are shown. Figure S5: violin plots with boxplots rep-
resent the distribution of achromatic contrast values, expressed
in Euclidean distance units, obtained from the vision model of
bees. Figure S6: spectra of the four species with green owers
that do not generate marker points.
FUNDING
This work was supported by the project PID2020-116222GB-I00,
funded by MICIU/AEI/10.13039/501100011033 to M.A. and
E.N.
ACKNOWLEDGEMENTS
We thank Piluca Álvarez and Julia Fernández for obtaining
spectral data from the collected samples. We thank Dr. Modesto
Luceño for providing several photographs of green-owered
species.
AUTHOR CONTRIBUTIONS
E.N., P.L.O., M.A. and M.L.B. conceived the study and car-
ried out the sampling. E.N., J.C.V. and M.L.-O. established the
methodology. J.C.V. and C.D.-G. analysed the spectral data
and performed the statistical analyses. E.N., J.C.V., M.A. and
J.B.W. drafted the manuscript. All authors contributed to the
nal manuscript and approved the submitted version.
CONFLICT OF INTEREST
The authors declare that they have no known competing nan-
cial interests or personal relationships that could have appeared
to inuence the work reported in this paper.
LITERATURE CITED
Alonso C. 2005. Pollination success across an elevation and sex ratio gra-
dient in gynodioecious Daphne laureola. American Journal of Botany 92:
1264–1269.
An L, Neimann A, Eberling E, Algora H, Brings S, Lunau K. 2018. The
yellow specialist: droney Eristalis tenax prefers different yellow colours
for landing and proboscis extension. Journal of Experimental Biology
221: jeb184788.
Armold SEJ, Faruq S, Savolainen V, McOwan PW, Chittka L. 2010.
FReD: the oral reectance database — a web portal for analyses of
ower colour. PLoS One 5: e14287.
Asenbaum J, Schäfer I, Etl F, Dötterl S, Schönenberger J, Chartier M.
2021. Comparative pollination ecology of ve European Euphorbia spe-
cies. International Journal of Plant Sciences 182: 763–777.
Ashworth L, Aguilar R, Martén-Rodríguez S, et al. 2015. Pollination
syndromes: a global pattern of convergent evolution driven by the
most effective pollinator. In: Pontarotti P, ed. Evolutionary biology:
biodiversication from genotype to phenotype. Cham, Switzerland:
Springer, 203–224.
Bazzaz FA, Carlson RW, Harper JL. 1979. Contribution to reproductive ef-
fort by photosynthesis of owers and fruits. Nature 279: 554–555.
Bischoff M, Lord JM, Robertson AW, Dyer AG. 2013. Hymenopteran
pollinators as agents of selection on ower colour in the New Zealand
mountains: salient chromatic signals enhance ower discrimination. New
Zealand Journal of Botany 51: 181–193.
Briscoe AD, Chittka L. 2001. The evolution of color vision in insects. Annual
Review of Entomology 46: 471–510.
Brodmann J, Twele R, Francke W, Hölzler G, Zhang QH, Ayasse M. 2008.
Orchids mimic green-leaf volatiles to attract prey-hunting wasps for pol-
lination. Current Biology: CB 18: 740–744.
Bukovac Z, Shrestha M, Garcia JE, Burd M, Dorin A, Dyer AG. 2017. Why
background colour matters to bees and owers. Journal of Comparative
Physiology. A, Neuroethology, Sensory, Neural, and Behavioral Physiology
203: 369–380.
Camargo MGG, Lunau K, Batalha MA, Brings S, Brito VLG, Morellato
LPC. 2019. How ower colour signals allure bees and hummingbirds: a
community-level test of the bee avoidance hypothesis. New Phytologist
222: 1112–1122.
ChalkerScott L. 1999. Environmental signicance of anthocyanins in plant
stress responses. Photochemistry and Photobiology 70: 1–9.
Chen Z, Niu Y, Liu CQ, Sun H. 2020. Red owers differ in shades be-
tween pollination systems and across continents. Annals of Botany 126:
837–848.
Chittka L. 1992. The colour hexagon: a chromaticity diagram based on photo-
receptor excitations as a generalized representation of colour opponency.
Journal of Comparative Physiology A 170: 533–543.
Chittka L, Kevan PG. 2005. Flower colour as advertisement. In: Dafni A,
Kevan P, Husband BC, eds. Practical pollination biology. Canada:
Enviroquest Ltd, 157–196.
Chittka L, Menzel R. 1992. The evolutionary adaptation of ower colours and
the insect pollinators’ colour vision. Journal of Comparative Physiology
A 171: 171–181.
Chittka L, Raine NE. 2006. Recognition of owers by pollinators. Current
Opinion in Plant Biology 9: 428–435.
Chittka L, Wells H. 2004. Color vision in bees: mechanisms, ecology and
evolution. In: Prete F, ed. How simple nervous systems create complex
perceptual worlds. Boston, IL: MIT Press, 165–191.
Chittka L, Troje N, Menzel R. 1994. Ultraviolet as a component of ower
reections, and the color perception of Hymenoptera. Vision Research 34:
1489–1508.
Downloaded from https://academic.oup.com/aob/advance-article/doi/10.1093/aob/mcae213/7920006 by guest on 07 January 2025
del Valle et al. Pollinators’ perception of green owers requires yellow 11
Coimbra G, Araujo C, Bergamo PJ, Freitas L, Rodríguez-Gironés MA.
2020. Flower conspicuousness to bees across pollination systems: a gen-
eralized test of the bee-avoidance hypothesis. Frontiers in Plant Science
11: 558684.
Cuthill IC, Allen WL, Arbuckle K, et al. 2017. The biology of color. Science
357: eaan0221.
Darwin C. 1871. The descent of man, and selection in relation to sex.
London, UK: John Murray.
De Ibarra NH, Langridge KV, Vorobyev M. 2015. More than colour attrac-
tion: behavioural functions of ower patterns. Current Opinion in Insect
Science 12: 64–70.
Del Valle JC, Gallardo-López A, Buide ML, Whittall JB, Narbona E. 2018.
Digital photography provides a fast, reliable, and noninvasive method to
estimate anthocyanin pigment concentration in reproductive and vegeta-
tive plant tissues. Ecology and Evolution 8: 3064–3076.
Dorin A, Shrestha M, Herrmann M, Burd M, Dyer AG. 2020. Automated
calculation of spectral-reectance marker-points to enable analysis of
plant colour-signalling to pollinators. MethodsX 7: 100827.
Dorin A, Shrestha M, Garcia JE, Burd M, Dyer AG. 2023. Ancient insect
vision tuned for ight among rocks and plants underpins natural ower
colour diversity. Proceedings of the Royal Society B: Biological Sciences
290: 20232018.
Dyer AG, Chittka L. 2004. Biological signicance of distinguishing be-
tween similar colours in spectrally variable illumination: bumblebees
(Bombus terrestris) as a case study. Journal of Comparative Physiology.
A, Neuroethology, Sensory, Neural, and Behavioral Physiology 190:
105–114.
Dyer AG, Spaethe J, Prack S. 2008. Comparative psychophysics of
bumblebee and honeybee colour discrimination and object detection.
Journal of Comparative Physiology. A, Neuroethology, Sensory, Neural,
and Behavioral Physiology 194: 617–627.
Dyer AG, Boyd-Gerny S, Mcloughlin S, Rosa MGP, Simonov V, Wong
BBM. 2012. Parallel evolution of angiosperm colour signals: common
evolutionary pressures linked to hymenopteran vision. Proceedings of the
Royal Society B: Biological Sciences 279: 3606–3615.
Dyer AG, Jentsch A, Burd M, et al. 2021. Fragmentary blue: resolving the
rarity paradox in ower colors. Frontiers in Plant Science 11: 618203.
Endler JA. 1978. A predator’s view of animal color patterns. In: Hecht MK,
Steere WC, Wallace B, eds. Evolutionary biology. Boston, MA: Springer,
319–364.
Endler JA. 1992. Signals, signal conditions, and the direction of evolution. The
American Naturalist 139: S125–S153.
Endler JA, Mielke PW. 2005. Comparing entire colour patterns as birds see
them. Biological Journal of the Linnean Society 86: 405–431.
Faegri K, van der Pijl P. 1979. The principles of pollination ecology. Oxford:
Pergamon Press.
Fenster CB, Armbruster WS, Wilson P, Dudash MR, Thomson JD.
2004. Pollination syndromes and oral specialization. Annual Review of
Ecology, Evolution, and Systematics 35: 375–403.
Finnell LM, Koski MH. 2021. A test of sensory drive in plant–pollinator
interactions: heterogeneity in the signalling environment shapes pol-
linator preference for a oral visual signal. New Phytologist 232:
1436–1448.
Friedman J, Barret SCH. 2009. Wind of change: new insights on the ecology
and evolution of pollination and mating in wind-pollinated plants. Annals
of Botany 103: 1515–1527.
Garcia JE, Spaethe J, Dyer AG. 2017. The path to colour discrimination
is S-shaped: behaviour determines the interpretation of colour models.
Journal of Comparative Physiology. A, Neuroethology, Sensory, Neural,
and Behavioral Physiology 203: 983–997.
Garcia JE, Dyer AG, Burd M, Shrestha M. 2021. Flower colour and size sig-
nals differ depending on geographical location and altitude region. Plant
Biology (Stuttgart, Germany) 23: 905–914.
Garcia JE, Hannah L, Shrestha M, Burd M, Dyer AG. 2022. Fly pollin-
ation drives convergence of ower coloration. The New Phytologist 233:
52–61.
Garland T, Dickerman AW, Janis CM, Jason A. 1993. Phylogenetic analysis
of covariance by computer simulation. Systematic Biology 42: 265–292.
Giurfa M, Núñez J, Chittka L, Menzel R. 1995. Colour preferences of ower-
naive honeybees. Journal of Comparative Physiology A 177: 247–259.
Giurfa M, Vorobyev M, Kevan P, Menzel R. 1996. Detection of coloured
stimuli by honeybees: minimum visual angles and receptor specic con-
trasts. Journal of Comparative Physiology A 178: 699–709.
Giurfa M, Vorobyev M, Brandt R, Posner B, Menzel R. 1997. Discrimination
of coloured stimuli by honeybees: alternative use of achromatic and chro-
matic signals. Journal of Comparative Physiology A Sensory Neural and
Behavioral Physiology 180: 235–243.
Gould KS. 2004. Nature’s Swiss army knife: the diverse protective roles of
anthocyanins in leaves. Journal of Biomedicine and Biotechnology 2004:
314–320.
Hannah L, Dyer AG, Garcia JE, Dorin A, Burd M. 2019. Psychophysics
of the hovery: categorical or continuous color discrimination? Current
Zoology 65: 483–492.
Herrera CM, Sánchez-Lafuente AM, Medrano M, Guitián J, Cerdá X,
Rey P. 2001. Geographical variation in autonomous self-pollination levels
unrelated to pollinator service in Helleborus foetidus (Ranunculaceae).
American Journal of Botany 88: 1025–1032.
Jin Y, Qian H. 2019. V.PhyloMaker: an R package that can generate very large
phylogenies for vascular plants. Ecography 42: 1353–1359.
Johnson SD, Ellis A, Dötterl S. 2007. Specialization for pollination by
beetles and wasps: the role of lollipop hairs and fragrance in Satyrium
microrrhynchum (Orchidaceae). American Journal of Botany 94: 47–55.
Kassambara A, Mundt F. 2020. factoextra: Extract and Visualize the Results
of Multivariate Data Analyses. R package version 1.0.7. (accessed on 17
August 2024) https://CRAN.R-project.org/package=factoextra
Kato Y, Araki K, Ohara M. 2009. Breeding system and oral visitors of
Veratrum album subsp. oxysepalum (Melanthiaceae). Plant Species
Biology 24: 42–46.
Kay QON, Daoud HS, Stirton CH. 1981. Pigment distribution, light reec-
tion and cell structure in petals. Botanical Journal of the Linnean Society
83: 57–83.
Kelber A, Vorobyev M, Osorio D. 2003. Animal colour vision — behavioural
tests and physiological concepts. Biological Reviews of the Cambridge
Philosophical Society 78: 81–118.
Kemp JE, Ellis AG. 2019. Cryptic petal coloration decreases oral apparency
and herbivory in nocturnally closing daisies. Functional Ecology 33:
2130–2141.
Lee DW. 2007. Nature’s palette. Chicago, IL: University of Chicago Press.
León-Osper M, Narbona E. 2022. Unravelling the mystery of red owers in
the Mediterranean Basin: how to be conspicuous in a place dominated by
hymenopteran pollinators. Functional Ecology 36: 2774–2790.
Lovell JH. 1912. The color sense of the honey-bee: the pollination of green
owers. The American Naturalist 46: 83–107.
Lukas K, Dötterl S, Ayasse M, Burger H. 2020. Olfactory and visual
oral signals of Hedera helix and Heracleum sphondylium involved in
host nding by nectar-foraging social wasps. Frontiers in Ecology and
Evolution 8: 571454.
Lunau K. 1990. Colour saturation triggers innate reactions to ower signals:
ower dummy experiments with bumblebees. Journal of Comparative
Physiology A 166: 827–834.
Lunau K. 2014. Visual ecology of ies with particular reference to colour
vision and colour preferences. Journal of Comparative Physiology.
A, Neuroethology, Sensory, Neural, and Behavioral Physiology 200:
497–512.
Lunau K, Dyer AG. 2024. The modelling of ower colour: spectral purity or
colour contrast as biologically relevant descriptors of ower colour sig-
nals for bees depending upon the perceptual task. Plant Biology (Stuttgart,
Germany) 26: 896–910.
Maia R, Gruson H, Endler JA, White TE. 2019.  2: new tools for the
spectral and spatial analysis of colour in . Methods in Ecology and
Evolution 10: 1097–1107.
Martins AE, Arista M, Morellato LPC, Camargo MGG. 2021. Color sig-
nals of bee-pollinated owers: the signicance of natural leaf background.
American Journal of Botany 108: 788–797.
Menzel R. 1967. Untersuchungen zum Erlernen von Spektralfarben durch die
Honigbiene (Apis mellica). Zeitschrift für Vergleichende Physiologie 56:
22–62.
Menzel R, Backhaus W. 1991. Colour vision in insects. Vision and Visual
Dysfunction 6: 262–293.
Mizuno T, Sugahara K, Tsutsumi C, et al. 2021. Identication of antho-
cyanin and other avonoids from the green–blue petals of Puya alpestris
(Bromeliaceae) and a clarication of their coloration mechanism.
Phytochemistry 181: 112581.
Mizuno T, Mori S, Sugahara K, Yukawa T, Koi S, Iwashina T. 2024. Floral
pigments and their perception by avian pollinators in three Chilean Puya
species. Journal of Plant Research 137: 395–409.
Downloaded from https://academic.oup.com/aob/advance-article/doi/10.1093/aob/mcae213/7920006 by guest on 07 January 2025
del Valle et al. Pollinators’ perception of green owers requires yellow
12
Moré M, Ibañez AC, Drewniak ME, Cocucci AA, Raguso RA. 2020. Flower
diversication across ‘pollinator climates’: sensory aspects of corolla
color evolution in the orally diverse South American genus Jaborosa
(Solanaceae). Frontiers in Plant Science 11: 601975.
Narbona E, Ortiz PL, Arista M. 2002. Functional andromonoecy in
Euphorbia (Euphorbiaceae). Annals of Botany 89: 571–577.
Narbona E, del Valle JC, Arista M, Buide ML, Ortiz PL. 2021a. Major
ower pigments originate different colour signals to pollinators. Frontiers
in Ecology and Evolution 9: 743850.
Narbona E, Del Valle JC, Whittall JB. 2021b. Painting the green canvas: how
pigments produce ower colours. The Biochemist 43: 6–12.
Ng L, Garcia JE, Dyer AG. 2018. Why colour is complex: evidence that bees
perceive neither brightness nor green contrast in colour signal processing.
FACETS 3: 800–817.
Nolazco S, Delhey K, Nakagawa S, Peters A. 2022. Ornaments are equally
informative in male and female birds. Nature Communications 13: 5917.
Ortiz PL, Fernández-Díaz P, Pareja D, Escudero M, Arista M. 2021. Do
visual traits honestly signal oral rewards at community level? Functional
Ecology 35: 369–383.
Peitsch D, Fietz A, Hertel H, de Souza J, Ventura DF, Menzel R. 1992. The
spectral input systems of hymenopteran insects and their receptor-based
colour vision. Journal of Comparative Physiology. A, Sensory, Neural,
and Behavioral Physiology 170: 23–40.
Rathcke BJ. 2000. Birds, pollination reliability, and green owers in an endemic
island shrub, Pavonia Bahamensis (Malvaceae). Rhodora 102: 392–414.
Rebelo AG, Siegfried WR. 1985. Colour and size of owers in relation to pol-
lination of Erica species. Oecologia 65: 584–590.
Revell LJ. 2012. phytools: an R package for phylogenetic comparative biology
(and other things). Methods in Ecology and Evolution 3: 217–223.
Rohde K, Papiorek S, Lunau K. 2013. Bumblebees (Bombus terrestris) and
honeybees (Apis mellifera) prefer similar colours of higher spectral purity
over trained colours. Journal of Comparative Physiology. A, Neuroethology,
Sensory, Neural, and Behavioral Physiology 199: 197–210.
Rosas-Guerrero V, Aguilar R, Martén-Rodríguez S, et al. 2014. A quan-
titative review of pollination syndromes: do oral traits predict effective
pollinators? Ecology Letters 17: 388–400.
RStudio Team. 2022. RStudio: integrated development environment for R.
RStudio, PBC. Boston, MA, USA.
Shrestha M, Dyer AG, Boyd-Gerny S, Wong BBM, Burd M. 2013a. Shades
of red: bird-pollinated owers target the specic colour discrimination
abilities of avian vision. The New Phytologist 198: 301–310.
Shrestha M, Dyer AG, Burd M. 2013b. Evaluating the spectral discrimin-
ation capabilities of different pollinators and their effect on the evolution
of ower colors. Communicative & Integrative Biology 6: e24000.
Shrestha M, Lunau K, Dorin A, et al. 2016. Floral colours in a world without
birds and bees: the plants of Macquarie Island. Plant Biology (Stuttgart,
Germany) 18: 842–850.
Shrestha M, Burd M, Garcia JE, Dorin A, Dyer AG. 2019. Colour evolution
within orchids depends on whether the pollinator is a bee or a y. Plant
Biology (Stuttgart, Germany) 21: 745–752.
Shrestha M, Tai KC, Dyer AG, et al. 2024. Flower colour and size-signals
vary with altitude and resulting climate on the tropical-subtropical islands
of Taiwan. Frontiers in Plant Science 15: 1304849.
Shuttleworth A, Johnson SD. 2009. A key role for oral scent in a wasp-
pollination system in Eucomis (Hyacinthaceae). Annals of Botany 103:
715–725.
Smith SA, Brown JW. 2018. Constructing a broadly inclusive seed plant phyl-
ogeny. American Journal of Botany 105: 302–314.
Spaethe J, Tautz J, Chittka L. 2001. Visual constraints in foraging bumble-
bees: ower size and color affect search time and ight behavior.
Proceedings of the National Academy of Sciences of the United States of
America 98: 3898–3903.
Stevens M. 2013. Sensory ecology, behaviour, and evolution. Oxford, UK:
Oxford University Press.
Stournaras KE, Lo E, Böhning-Gaese K, et al. 2013. How colorful are fruits?
Limited color diversity in eshy fruits on local and global scales. The New
Phytologist 198: 617–629.
Strauss SY, Cacho NI. 2013. Nowhere to run, nowhere to hide: the importance
of enemies and apparency in adaptation to harsh soil environments. The
American Naturalist 182: E1–E14.
Troje N. 1993. Spectral categories in the learning behaviour of blowies.
Zeitschrift für Naturforschung C 48: 96–104.
Utech FH, Kawano S. 1975. Spectral polymorphisms in angiosperm owers
determined by differential ultraviolet reectance. The Botanical Magazine
Tokyo 88: 9–30.
van der Kooi CJ. 2021. How much pigment should owers have? Flowers with
moderate pigmentation have highest color contrast. Frontiers in Ecology
and Evolution 9: 731626.
van der Kooi CJ, Kelber A. 2022. Achromatic cues are important for ower
visibility to hawkmoths and other insects. Frontiers in Ecology and
Evolution 10: 1–14. doi:10.3389/fevo.2022.819436.
Van Der Kooi CJ, Dyer AG, Kevan PG, Lunau K. 2019. Functional signi-
cance of the optical properties of owers for visual signalling. Annals of
Botany 123: 263–276.
Van Der Kooi CJ, Stavenga DG, Arikawa K, Belušič G, Kelber A. 2021.
Evolution of insect color vision: from spectral sensitivity to visual ecology.
Annual Review of Entomology 66: 435–461.
Vignolini S, Davey MP, Bateman RM, et al. 2012. The mirror crack’d:
both pigment and structure contribute to the glossy blue appearance
of the mirror orchid, Ophrys speculum. The New Phytologist 196:
1038–1047.
Wang Q, Guo Q, Chi X, Zhu S, Tang Z. 2022. Evolutionary history and cli-
mate conditions constrain the ower colours of woody plants in China.
Journal of Plant Ecology 15: 196–207.
Weevers TH. 1952. Flower colours and their frequency. Acta Botanica
Neerlandica 1: 81–92.
White TE, Dalrymple RL, Noble DWA, O’Hanlon JC, Zurek DB, Umbers
KDL. 2015. Reproducible research in the study of biological coloration.
Animal Behaviour 106: 51–57.
Willson MF, Whelan CJ. 1990. The evolution of fruit color in eshy-fruited
plants. The American Naturalist 136: 790–809.
Yuan Y, Li X, Yao X, et al. 2023. Mechanisms underlying the formation of
complex color patterns on Nigella orientalis (Ranunculaceae) petals. The
New Phytologist 237: 2450–2466.
Zanne AE, Tank DC, Cornwell WK, et al. 2014. Three keys to the radiation
of angiosperms into freezing environments. Nature 506: 89–92.
Downloaded from https://academic.oup.com/aob/advance-article/doi/10.1093/aob/mcae213/7920006 by guest on 07 January 2025
Preprint
Full-text available
Flower coloration arises primarily from pigments that serve dual functions: attracting pollinators and mitigating environmental stresses. Among major flower pigment groups, anthocyanins and UV-absorbing phenylpropanoids are particularly notable for fulfilling both roles. Their importance suggests that both pigment types should be widespread in flowers. Here, we analyze the UV-Vis absorption profiles of major floral pigments to assess their potential protective role against UV radiation and demonstrate that this protection is largely confined to UV-absorbing phenylpropanoids. We also analyzed the floral pigment composition of 926 animal-pollinated species from California, southern Spain, and southeastern Brazil. UV-absorbing phenylpropanoids were ubiquitous, while anthocyanins occurred in ~56% of species, carotenoids in ~37%, and chlorophylls in ~17%. Pigment frequencies varied with abiotic and biotic factors, such as light environment in Spain, and pollinator type, such as insect versus hummingbird, in California. Despite these local differences, our findings reveal a consistent regional distribution of floral pigments.
Article
Full-text available
Flower colour is an important mediator of plant-pollinator interactions. While the reflectance of light from the flower surface and background are governed by physical properties, the perceptual interpretation of such information is generated by complex multilayered visual processing. Should quantitative modelling of flower signals strive for repeatable consistency enabled by parameter simplification, or should modelling reflect the dynamic way in which bees are known to process signals? We discuss why colour is an interpretation of spectral information by the brain of an animal. Different species, or individuals within a species, may respond differently to colour signals depending on sensory apparatus and/or individual experience. Humans and bees have different spectral ranges, but colour theory is strongly rooted in human colour perception and many principles of colour vision appear to be common. We discuss bee colour perception based on physiological, neuroanatomical and behavioural evidence to provide a pathway for modelling flower colours. We examine whether flower petals and floral guides as viewed against spectrally different backgrounds should be considered as a simple colour contrast problem or require a more dynamic consideration of how bees make perceptual decisions. We discuss that plants such as deceptive orchids may present signals to exploit bee perception, whilst many plants do provide honest signalling where perceived saturation indicates the probability of collecting nutritional rewards towards the centre of a flower that then facilitates effective pollination.
Article
Full-text available
The diversity of flower colours in nature provides quantifiable evidence for how visitations by colour sensing insect pollinators can drive the evolution of angiosperm visual signalling. Recent research shows that both biotic and abiotic factors may influence flower signalling, and that harsher climate conditions may also promote salient signalling to entice scarcer pollinators to visit. In parallel, a more sophisticated appreciation of the visual task foragers face reveals that bees have a complex visual system that uses achromatic vision when moving fast, whilst colour vision requires slower, more careful inspection of targets. Spectra of 714 native flowering species across Taiwan from sea level to mountainous regions 3,300 m above sea level (a.s.l.) were measured. We modelled how the visual system of key bee pollinators process signals, including flower size. By using phylogenetically informed analyses, we observed that at lower altitudes including foothills and submontane landscapes, there is a significant relationship between colour contrast and achromatic signals. Overall, the frequency of flowers with high colour contrast increases with altitude, whilst flower size decreases. The evidence that flower colour signaling becomes increasingly salient in higher altitude conditions supports that abiotic factors influence pollinator foraging in a way that directly influences how flowering plants need to advertise.
Article
Full-text available
Understanding the origins of flower colour signalling to pollinators is fundamental to evolutionary biology and ecology. Flower colour evolves under pressure from visual systems of pollinators, like birds and insects, to establish global signatures among flowers with similar pollinators. However, an understanding of the ancient origins of this relationship remains elusive. Here, we employ computer simulations to generate artificial flower backgrounds assembled from real material sample spectra of rocks, leaves and dead plant materials, against which to test flowers' visibility to birds and bees. Our results indicate how flower colours differ from their backgrounds in strength, and the distributions of salient reflectance features when perceived by these key pollinators, to reveal the possible origins of their colours. Since Hymenopteran visual perception evolved before flowers, the terrestrial chromatic context for its evolution to facilitate flight and orientation consisted of rocks, leaves, sticks and bark. Flowers exploited these pre-evolved visual capacities of their visitors, in response evolving chromatic features to signal to bees, and differently to birds, against a backdrop of other natural materials. Consequently, it appears that today's flower colours may be an evolutionary response to the vision of diurnal pollinators navigating their world millennia prior to the first flowers.
Chapter
Full-text available
The authors of Complex Worlds from Simpler Nervous Systems explain how animals with small, often minuscule, nervous systems—jumping spiders, bees, praying mantids, toads, and others—are not the simple "reflex machines" they were once thought to be. Because these animals live in the same world as do much larger species, they must meet the same environmental challenges. They do so by constructing complex perceptual worlds within which they can weigh options, make decisions, integrate unique experiences, apply complex algorithms, and execute plans—and they must do this with thousands rather than the billions of neurons necessary for their larger counterparts. The authors of each chapter, leading neuroscientists and animal behaviorists, present their research in ways that allow the reader to understand this process from the animal's perspective. The first of the book's three parts, "Creating Visual Worlds: Using Abstract Representations and Algorithms," examines the visual worlds of jumping spiders, honeybees, praying mantids, and toads. Part II, "Enhancing the Visual Basics: Using Color and Polarization," explores color vision and light polarization perception in honeybees, butterflies, crayfish, mantis shrimps, and octopuses. The final part, "Out of Sight: Creating Extravisual Worlds," examines the complex integration of visual and mechanosensory information in the cockroach and the unique auditory world of the unusual bladder grasshopper. All of these fascinating stories can be read both for what they teach us about the perceptual worlds of little animals, and for what they suggest about the general organizing principles of all central nervous systems, both "simple" and complex. Bradford Books imprint
Article
Full-text available
Complex color patterns on petals are widespread in flowering plants, yet the mechanisms underlying their formation remain largely unclear. Here, by conducting detailed morphological, anatomical, biochemical, optical, transcriptomic, and functional studies, we investigated the cellular bases, chromogenic substances, reflectance spectra, developmental processes, and underlying mechanisms of complex color pattern formation on Nigella orientalis petals. We found that the complexity of the N. orientalis petals in color pattern is reflected at multiple levels, with the amount and arrangement of different pigmented cells being the key. We also found that biosynthesis of the chromogenic substances of different colors is sequential, so that one color/pattern is superimposed on another. Expression and functional studies further revealed that a pair of R2R3‐MYB genes function cooperatively to specify the formation of the eyebrow‐like horizontal stripe and the Mohawk haircut‐like splatters. Specifically, while NiorMYB113‐1 functions to draw a large splatter region, NiorMYB113‐2 functions to suppress the production of anthocyanins from the region where a gap will form, thereby forming the highly specialized pattern. Our results provide a detailed portrait for the spatiotemporal dynamics of the coloration of N. orientalis petals and help better understand the mechanisms underlying complex color pattern formation in plants.
Article
Full-text available
Female ornaments are often reduced, male-like traits. Although these were long perceived as non-functional, it is now broadly accepted that female ornaments can be adaptive. However, it is unclear whether this is as common in females as it is in males, and whether ornaments fulfil similar signalling roles. Here, we apply a bivariate meta-analysis to a large dataset of ornaments in mutually ornamented birds. As expected, female ornament expression tends to be reduced compared to males. However, ornaments are equally strongly associated with indicators of condition and aspects of reproductive success in both sexes, regardless of the degree of sexual dimorphism. Thus, we show here in a paired comparison within-and-across species, that ornaments in birds provide similar information in both sexes: more ornamented individuals are in better condition and achieve higher reproductive success. Although limited by their correlational nature, these outcomes imply that female ornaments could widely function in a similar manner as male ornaments.
Article
Full-text available
Red‐flowered species have traditionally been related to ornithophily, and the ‘bee avoidance’ hypothesis, that is, red flower colouration is a way to reduce visits from hymenopterans, has been proposed to explain this association. In the Mediterranean Basin, ornithophily is almost absent, and hymenopterans are the most common group of pollinators. The fact that hymenopterans are virtually red‐blind raises the question of how red‐flowered species are pollinated in this region. Are these flowers pollinated by other groups of red‐sensitive insects such as lepidopterans and coleopterans, or do they have visual cues that make them attractive to hymenopterans? We examined the reflectance spectra of 51 red‐flowered species from the Mediterranean Basin and modelled these spectra in the visual system of hymenopterans, dipterans, coleopterans and lepidopterans to obtain colour and conspicuousness. According to their reflectance curves, species were classified as pure red and UV‐red flowers, and the presence of more than one flower colour (patterned flowers) was studied. We evaluated the match between flower reflectance spectra and the maximum discrimination abilities of hymenopteran and lepidopteran visual system. All these metrics were analysed in a phylogenetically explicit framework, and a literature review of potential pollinators was performed. The vast majority of red‐flowered species in the Mediterranean Basin are potentially pollinated by hymenopterans, and only three species are exclusively visited by coleopterans. We found that 90% of these species showed at least one colour signal strategy that helps to enhance conspicuousness to hymenopterans: to produce UV‐red flower type spectra and/or patterned flowers. The UV‐red colour showed a significant phylogenetic signal, but the presence of patterned flowers did not. Even though the red‐flowered species of the Mediterranean Basin did not optimally match the colour vision of hymenopterans or lepidopterans, the presence of patterned and UV‐red colours suggests an improvement in detection and discrimination by hymenopterans. The bee‐avoidance hypothesis seems to be ruled out for the red‐flowered species of the Mediterranean Basin. Our results suggest that red‐flowered species are mostly pollinated by hymenopterans and show different flower colour signal strategies that can be interpreted as signs of adaptation to these pollinators. Read the free Plain Language Summary for this article on the Journal blog.
Article
Full-text available
Studies on animal colour vision typically focus on the chromatic aspect of colour, which is related to the spectral distribution, and disregard the achromatic aspect, which is related to the intensity (“brightness”) of a stimulus. Although the chromatic component of vision is often most reliable for object recognition because it is fairly context independent, the achromatic component may provide a reliable signal under specific conditions, for example at night when light intensity is low. Here we make a case for the importance of achromatic cues in plant-pollinator signalling, based on experimental data on naïve Deilephila elpenor and Macroglossum stellatarum hawkmoths, optical modelling and synthesising published experiments on bees, flies, butterflies and moths. Our experiments show that in ecologically relevant light levels hawkmoths express a strong preference for brighter stimuli. Published experiments suggest that for flower-visiting bees, butterflies, moths and flies, achromatic cues may be more important for object detection than often considered. Our optical modelling enabled disentangling the contribution of pigments and scattering structures to the flower’s achromatic contrast, and illustrates how flower anatomy and background are important mediating factors. We discuss our findings in the context of the often-assumed dichotomy between detection and discrimination, chromatic versus achromatic vision, and the evolution of floral visual signals.
Article
Full-text available
Flower colour is mainly due to the presence and type of pigments. Pollinator preferences impose selection on flower colour that ultimately acts on flower pigments. Knowing how pollinators perceive flowers with different pigments becomes crucial for a comprehensive understanding of plant-pollinator communication and flower colour evolution. Based on colour space models, we studied whether main groups of pollinators, specifically hymenopterans, dipterans, lepidopterans and birds, differentially perceive flower colours generated by major pigment groups. We obtain reflectance data and conspicuousness to pollinators of flowers containing one of the pigment groups more frequent in flowers: chlorophylls, carotenoids and flavonoids. Flavonoids were subsequently classified in UV-absorbing flavonoids, aurones-chalcones and the anthocyanins cyanidin, pelargonidin, delphinidin, and malvidin derivatives. We found that flower colour loci of chlorophylls, carotenoids, UV-absorbing flavonoids, aurones-chalcones, and anthocyanins occupied different regions of the colour space models of these pollinators. The four groups of anthocyanins produced a unique cluster of colour loci. Interestingly, differences in colour conspicuousness among the pigment groups were almost similar in the bee, fly, butterfly, and bird visual space models. Aurones-chalcones showed the highest chromatic contrast values, carotenoids displayed intermediate values, and chlorophylls, UV-absorbing flavonoids and anthocyanins presented the lowest values. In the visual model of bees, flowers with UV-absorbing flavonoids (i.e., white flowers) generated the highest achromatic contrasts. Ours findings suggest that in spite of the almost omnipresence of floral anthocyanins in angiosperms, carotenoids and aurones-chalcones generates higher colour conspicuousness for main functional groups of pollinators.
Article
The Chilean Puya species, Puya coerulea var. violacea and P. chilensis bear blue and pale-yellow flowers, respectively, while P. alpestris considered to be their hybrid-derived species has unique turquoise flowers. In this study, the chemical basis underlying the different coloration of the three Puya species was explored. We first isolated and identified three anthocyanins: delphinidin 3,3ʹ,5ʹ-tri-O-glucoside, delphinidin 3,3′-di-O-glucoside and delphinidin 3-O-glucoside; seven flavonols: quercetin 3-O-rutinoside-3′-O-glucoside, quercetin 3,3ʹ-di-O-glucoside, quercetin 3-O-rutinoside, isorhamnetin 3-O-rutinoside, myricetin 3,3ʹ,5ʹ-tri-O-glucoside, myricetin 3,3ʹ-di-O-glucoside and laricitrin 3,5ʹ-di-O-glucoside; and six flavones: luteolin 4ʹ-O-glucoside, apigenin 4′-O-glucoside, tricetin 4ʹ-O-glucoside, tricetin 3ʹ,5ʹ-di-O-glucoside, tricetin 3ʹ-O-glucoside and selagin 5ʹ-O-glucoside, which is a previously undescribed flavone, from their petals. We also compared compositions of floral flavonoid and their aglycone among these species, which suggested that the turquoise species P. alpestris has an essentially intermediate composition between the blue and pale-yellow species. The vacuolar pH was relatively higher in the turquoise (pH 6.2) and pale-yellow (pH 6.2) flower species, while that of blue flower species was usual (pH 5.2). The flower color was reconstructed in vitro using isolated anthocyanin, flavonol and flavone at neutral and acidic pH, and its color was analyzed by reflectance spectra and the visual modeling of their avian pollinators. The modeling demonstrated that the higher pH of the turquoise and pale-yellow species enhances the chromatic contrast and spectral purity. The precise regulation of flower color by flavonoid composition and vacuolar pH may be adapted to the visual perception of their avian pollinator vision.