Artificial selection for food colour

rspb.royalsocietypublishing.org preferences
Gemma L. Cole and John A. Endler
Centre for Integrative Ecology, School of Life and Environmental Sciences, Deakin University, Waurn Ponds,
Victoria 3216, Australia
Research
Colour is an important factor in food detection and acquisition by animals
Cite this article: Cole GL, Endler JA. 2015
using visually based foraging. Colour can be used to identify the suitability
Artificial selection for food colour preferences. of a food source or improve the efficiency of food detection, and can even be
Proc. R. Soc. B 282: 20143108. linked to mate choice. Food colour preferences are known to exist, but whether
http://dx.doi.org/10.1098/rspb.2014.3108 these preferences are heritable and how these preferences evolve is unknown.
Using the freshwater fish Poecilia reticulata, we artificially selected for chase
behaviour towards two different-coloured moving stimuli: red and blue
spots. A response to selection was only seen for chase behaviours towards
Received: 22 December 2014 the red, with realized heritabilities ranging from 0.25 to 0.30. Despite intense
Accepted: 27 January 2015 selection, no significant chase response was recorded for the blue-selected
lines. This lack of response may be due to the motion-detection mechanism
in the guppy visual system and may have novel implications for the evolvabil-
ity of responses to colour-related signals. The behavioural response to several
colours after five generations of selection suggests that the colour opponency
Subject Areas: system of the fish may regulate the response to selection.
ecology, evolution, behaviour

Keywords:
sensory bias, artificial selection, evolution,
foraging, vision, motion detection
1. Introduction
Evaluating the visual appearance of potential food is important for accurate
detection and acquisition in visually based foraging. Consequently, many ani-
mals have developed highly stereotyped food preferences. For example, bees
Author for correspondence: have consistent preferences for objects that contain symmetrically radial pat-
Gemma L. Cole terns (an efficient way to recognize flowers) [1], flower colour facilitates
e-mail: [email protected] learning of foraging behaviour in bees [2], butterflies [3] and birds [4,5], and
innate colour preferences allow an efficient harvest of local nectar in the bum-
blebee Bombus terrestris [6]. Preferences for coloured food items may also have
shaped visual sensitivities in primates, where the maintenance of trichromatic
vision is thought to be linked to food detection [7– 10].
Despite the importance of food colour preferences to many aspects of behav-
ioural and evolutionary ecology [11–13], the heritability of these preferences is
unknown. Many foraging preferences are thought to be flexible towards unpre-
dictable environments and variability in food reward [13]; however, this may
be a risky option when foraging attracts costs. Innate preferences may represent
a safer and more efficient method of food detection, provided that the environ-
ment is predictable enough for heritable rules to work. Colour preferences are
known in a wide variety of species, and can affect food choice [11,14], but how
these preferences evolve is largely unknown and untested.
One such species that has distinct food colour preferences is the guppy, Poecilia
reticulata. Food colour preferences in this species are thought to be of particular
importance as they may have been co-opted into mate choice [15]. A study by
Rodd et al. [15] showed that guppies have a food colour preference for orange.
This matches a female mate preference for the same colour [16,17], indicating
that mate preferences may have co-evolved with food colour preferences; males
may have exploited a pre-existing sensory bias in the visual system that evolved
Electronic supplementary material is available as a response to food detection or recognition. However, the heritability of this
food colour preference, an essential assumption of this hypothesis, has never
at http://dx.doi.org/10.1098/rspb.2014.3108 or
been investigated. In fact, to the best of our knowledge, the heritability of food
via http://rspb.royalsocietypublishing.org. colour preferences, in general, has never been tested.

& 2015 The Author(s) Published by the Royal Society. All rights reserved.
 Using guppies as a model system, we conducted artificial was standardized with a neutral density wheel for each laser 2
selection to investigate whether these food colour preferences colour to equal total irradiance (2.8 mmol photons m21 s21) for
both spots, so that only colour and not luminance was different

rspb.royalsocietypublishing.org
are heritable and to test how quickly any evolutionary response
is likely to occur. Guppies live in continuously flowing fresh- between the treatments. The resulting radiance off the sand was
0.4 and 0.6 mmol photons m22 s21 str21 for the blue and red
water streams, and are omnivorous, opportunistic foragers
spots, respectively, owing to the tan sand colour. We calculated
that feed on insect larvae, small invertebrates, fruit and algae
the perceived luminance of the two laser spots, based upon the
[18–20]. Because benthic vegetation is often absent in such
guppy double cone guppy visual parameters (optical media pro-
streams, and food items tend to move with the current of the vided by Dr Ron Douglas; cone peak sensitivities from [22–24])
water, we used moving-simulated prey items to conduct this and the radiance spectra of the laser spots (see [17,27–29]). Lumi-
study. We used two different colours, red and blue, to select nance (sometimes known as brightness) represents the total light
on the opposite ends of the guppy spectral sensitivity range intensity as measured by the cone types responsible for detecting
(300–700 nm [21]). Because food colour may be important in luminance contrast (the double cones). Based on these calculations,

Proc. R. Soc. B 282: 20143108

mate choice in this species [15], we predicted that both food the perceived luminance for the blue spot was 1.06 times that of the
colour preferences will be heritable. We also wanted to provide red. The reduced difference in luminance compared with total
further evidence as to whether co-option is likely to have led to radiance is due to the incandescent adapting light; it is more
intense in long wavelengths, so chromatic adaptation makes the
known female preferences in this species.
cones less sensitive to red than blue light. Chromatic adaptation
was allowed for in the von Kries correction (see [29]). We calcu-
lated the number of just noticeable differences (JNDs) between
2. Material and methods the two stimuli to determine whether the observed difference in
perceived luminance would be apparent to the fish (see [29,30]).
(a) Husbandry This resulted in a JND of 0.01, well below the threshold of 1, indi-
Guppies used for the experimental populations were first to cating that no difference in luminance could be detected by the
second generation wild-caught fish from Alligator Creek, Queens- fish. The spectra for the double cones and laser spots used in the
land, Australia (19o26.790 S 146o58.65 E0 ). Fish were maintained at a calculations are given in the electronic supplementary material,
24 + 18C and 12 h light/dark regime with brown and green flake figure S3.
food provided daily. Individuals were housed in large 194 l mixed-
sex glass tanks containing around 150 juveniles and adults of both
sexes (sex ratio approx. 1 : 1) prior to use in the experiments.
(c) Recording foraging behaviour
In order to record foraging behaviour, individual fish were
(b) Visual stimuli placed into a 6 l tank that had a sand substrate and black mesh
In order to create a selectable proxy for foraging behaviour, visual around the sides. The sand substrate enabled the observer to
stimuli from red and blue laser pointers (wavelengths of 650 identify fish behaviour easily owing to high visual contrast,
and 450 nm, respectively) were used to create slowly moving and the black mesh was used to minimize spot reflection from
spots inside a black-walled wooden box measuring 48 cm the tank walls. The tank was filled to a depth of 6 cm above
(height)  40 cm (depth)  32 cm (width) containing a removable the substrate with water that had been taken from their mixed-
6 l tank with a sand substrate (see electronic supplementary sex maturation tank to assist acclimation. Fish were allowed to
material, figure S1). We named this apparatus the ‘maculator’ (lit- acclimate for 6 min in these tanks prior to the fish and tank
erally, spot generator, from the Latin ‘macula’, meaning spot). being placed into the maculator; this allowed fish to return to
These wavelengths were chosen so that selection acted on very normal behaviour after being moved. Fish then had a further
different guppy cones; guppies have peak visual sensitivities at 2 min to both behaviourally and chromatically acclimate inside
359, 408, 464, 533, 543 and 572 nm (see electronic supplementary the apparatus. During this time, the spot motor was running,
material, figure S2) [22–24]. Both the red and blue spots were but the mirror was not illuminated, ensuring fish were accli-
approximately 2 mm in diameter when measured on the floor of mated to vibrations and noise created by the motor. After
the tank containing sand and water—about 10 larger than the 2 min, the mirror was illuminated by either a red or blue light
minimum spatial resolution of guppies when seen from 2 cm [25]. beam from the laser pointer, creating a single moving coloured
On the internal ceiling of the maculator, at a height of 37 cm, spot on the floor of the tank. The octagonal mirror ensured
a motor (later referred to as the ‘spot motor’) turns a mirror- that once the first spot had travelled the floor of the tank another
covered octagonal cylinder. When a mirror is illuminated by a spot would follow; this continued until the end of the trial.
laser, a single spot is projected onto and moves in a straight Thirty-six consecutive spots were seen in each trial. Those fish
line across the tank floor below. The motor was set at a constant that chased the spot continued to do so despite no food
speed of 12 revolutions a minute, so that the spot moved across reward being given. The observation session commenced when
the sand floor at about 16 cm s21. The moving spots simulated the first spot appeared. A video monitor connected to a camera
food items being carried at a speed similar to that of a inside the maculator allowed the observer to watch behaviour,
medium-flowing stream (the type of habitat in which a guppy so that there were no disturbances to the fish during each session.
would be found in the wild). Guppies will chase a number of Fish were observed for 3 min and their behaviour recorded.
moving prey items, including brine shrimp, daphnia and their The number of chases of the moving light spot (c), the number of
own offspring. orientations towards the spot (o) and the latency until the start of
The irradiance spectra of the spots and the ambient light in the the first behaviour (either chase or orientation; l ) were combined
chamber were measured using a calibrated Li-Cor photometer (LI- into a total score T ¼ (3c þ o)/l . T is higher when a fish shows
189) placed on the bottom of the tank. The radiance of the spots any or all of the chase-related behaviours. For example, a fish
reflected off the sand was measured with an Ocean Optics that chased after 2 s of the trial commencing and that carried out
USB2000 þ spectrometer connected to a calibrated sensor aimed four chases and two orientations during the trial would receive a
at the spot from 2 cm away. The ambient light inside the maculator score of ((4  3) þ 2)/2 ¼ 7. A fish that orientated after 50 s of
(6.6 mmol photons m22 s21) was created by a 15 watt incandescent the trial commencing with no further behaviours would receive
bulb placed at 25 cm above the floor of the apparatus—within the a score of ((0  3) þ 1)/50 ¼ 0.02. Fish that did not chase or orien-
natural range in the wild [26]. The spot light intensity on the sand tate towards the spot received a latency score of 180, which
resulted in an overall score of 0. Chases were weighted three times and the order in which they encountered the two colours was 3
more than orientations because they were the most obvious indi- alternated to control for any order effects. These are referred to
cator of the simulated foraging behaviour. The number of nips as CB1 and CR1 for the first C replicate, and CB2 and CR2 for

rspb.royalsocietypublishing.org
was also measured, but this was not used in the selection criteria the second; CB1 and CR1 represent the fish in C1, CB2 and
because it was initially relatively rare, particularly for blue spot CR2 represent the fish in C2.
stimuli (in generation 0, the mean number of nips towards the The chase score T was determined for 200 fish (100 of each
blue and red spots was 0.013 and 0.11, respectively). sex) within each replicate of the three selection lines (R, B, C)
in the first selection (generation 0) and 160 (80 of each sex) in
the following four generations (1 – 4). This represents a compro-
(d) Test of foraging behaviour mise between time to measure and effective population size. A
We verified that the blue and red light spots simulated foraging total of 5040 fish were measured, and 6720 scores were recorded,
behaviour by testing two groups of fish with a high and low because control fish were measured twice in order to obtain both
motivation for foraging and observing their behaviour inside their red and blue scores.

Proc. R. Soc. B 282: 20143108

the maculator. To create the treatment groups, males and females Once all fish had been scored within a given treatment, the 40
were selected at random from the stock tanks and within-sex size males and 40 females with the best T in their replicate were
matched within 2 mm. These fish were then divided into two placed into a 194 l glass aquarium and allowed to mate for two
treatments: fed and relatively food-deprived. Fish in the fed treat- weeks. Fish in the control lines were selected at random and
ment received floating flaked fish food once daily as normal, 40 of each sex collected from the population allowed to mate in
whereas fish in the food-deprived treatment received food on the same manner. After two weeks of mating opportunity,
alternate days. Food-deprived fish were still able to forage on females were removed and housed in individual 6 l tanks
microflora and fauna on the gravel of their tanks (most of their containing 5 mm plastic mesh (to protect offspring from canni-
natural food); this ensured that the fish were never in danger balism) until the females gave birth. Males were moved to a
of actually being starved. Within these food treatments, the fish new 54 l home tank according to their treatment.
were separated into two further treatments corresponding to Once a female gave birth, she was removed from her tank and
spot colour. This resulted in four treatments—red hungry (RH), her fry were left to mature. Sexes were separated prior to matu-
red fed (RF), blue hungry (BH) and blue fed (BF)—with 16 fish ration. Up to four female and four male juveniles (depending
of each sex in each treatment (128 fish in total). upon family size) were taken from each family and placed, accord-
Fish were housed in 6 l tanks in single sex groups of four ing to treatment, into 54 l glass aquaria to mature. Sexes were again
according to food and colour treatment. Sexes were separated housed separately. Males were housed with both virgin and non-
to control for energy expended from mating effort and female virgin females from stock tanks in order to maintain normal
conditioning to colours seen in males. To control for differences mating behaviour. After approximately 20 weeks, these new gen-
in behaviour over time and to ensure all fish had the same treat- eration fish were scored using the maculator and the process
ment time (not all fish could be measured in one day), four tanks repeated. T scores from five generations were obtained (gener-
for each treatment were set up on alternate days for 4 days. For ations 0–4). Offspring mortality rates, indicative of the level of
example, on the first day, two BH and two RH tanks were set inbreeding, were recorded for each generation. Offspring mortality
up with four males and four females each, the second day BF rose from 5.2% and 6.9% in generations 1 and 2, respectively, and
and RF, and so on. After 4 days of treatment, fish were tested to 28.8% and 27.2% in generations 3 and 4, respectively.
individually in the maculator with the spot colour determined
by the designated treatment (R or B). Differences in the chase be-
haviour of fish in the four treatments were analysed using (g) Calculation of heritablilies
Bonferroni-corrected Mann – Whitney U-tests in the program R Realized heritabilities (h 2) were calculated for each selection line
[31], because the data were not normally distributed. using the slope coefficient of the regression of the response to
selection for increased T, against cumulative selection differential
[33]. Realized heritabilities were also calculated for control-
(e) Experimental population corrected selection lines; control correction was performed by
We conducted artificial selection experiments to investigate subtracting the mean of both control replicates from the individ-
whether chase behaviour for red or blue spots was heritable and ual values of the selected line, using red controls for R lines and
to estimate its magnitude. Up to eight juveniles (four males and blue controls for B lines.
four females when family size allowed) from 180 different females
were used to initiate the experimental population, representing a
compromise between increased effective population size and (h) Mechanistic test of visual stimuli
brood size. Siblings were split, by sex, into one of six spot colour To test possible visual mechanisms by which different chase beha-
treatments: blue (B), red (R) or control (C), each with two replicates, viours evolved, we tested the final generation (5; the offspring of
blue (B1, B2), red (R1, R2) and control (C1, C2). A total of 1200 indi- generation 4) with a green spot (532 nm) in addition to the blue
viduals made up the initial experimental population, with 200 and red spots we used in the previous generations. Between 10
individuals (100 males and 100 females) assigned to each of the and 20 fish (owing to time and fish inbreeding constraints) of
six treatments. In order to maximize artificial selection efficiency, each sex from each line (B1, B2, R1, R2, C1 and C2) were selected
males and females were housed separately to ensure all females randomly and assigned to one of the three spot colours. In order
were virgins after artificial selection in each generation; guppies to avoid possible habituation to the moving spot stimulus, test
have sperm storage [32]. fish were only exposed to one spot colour treatment. The chase
score T was recorded in the same way as before. Mann–Whitney
U-tests were used to identify significant differences between the
(f ) Artificial selection treatments. Control correction was again performed by subtracting
After maturation (approx. 20 weeks), fish were placed in the the mean of both control replicates from the individual values of
maculator, and their tendency to chase the moving light spot the selected line, using red controls for R lines and blue controls
was observed and recorded. A chase score (T ) was obtained for B lines.
for each individual. R replicates were scored with moving red The perceived luminance of the green laser spot on reflection
spots and B replicates were scored with moving blue spots. C from the sand was 5.5 times that of the red spot and 5.2 times
replicate fish were scored for both red and blue spot chasing, that of the blue spot, owing to the sand colour. This resulted in
 chase score no. chases 4
6

rspb.royalsocietypublishing.org
8 5
c
b b
6 4

3
4
2 c
2 c a c a c
1 a c a c
a
a a b b a a
0 0

Proc. R. Soc. B 282: 20143108

latency no. orientations
200 6
a
5
150 ac
a 4
a
100 3 a
a a a ac bc
b a 2 b
50 b bc
1
b
0 0
BH male
BH female
BF male
BF female
RH male
RH female
RF male
RF female

BH male
BH female
BF male
BF female
RH male
RH female
RF male
RF female
Figure 1. Box plots showing the median (horizontal line), upper and lower quartiles (box length), and range of the effect of feeding treatment on the selection
criteria: chase score (T ), number of chases (c), number of orientations (o) and latency (l ). Whiskers show the lowest and highest values within 1.5 times of the
interquartile range. White boxes represent hungry fish and shaded boxes represent fed fish. Thinner outlines represent the blue treatment and thicker outlines
represent the red treatment. n ¼ 32 for each of the treatment combinations (128 fish were tested in total). Fish that did not chase or orientate towards the
spot received a latency score (l ) of 180 and a T of 0. Different letters denote significant differences (at p , 0.05) between treatments with the Bonferroni correction.

JNDs of 0.74 and 0.73, respectively. This is again below the (b) Artificial selection
threshold of 1, indicating that the fish would not perceive
Artificial selection will reveal additive genetic variation if
this difference in luminance. The radiance of the green laser on
present. A response to artificial selection for increased T
reflection from the sand was 0.08 mmol photons m22 s21 str21.
was seen immediately in the two red lines R1 and R2
(figure 2). Although an initial increase in T was seen between
generation 0 and generation 1 in the blue lines (B1,B2), no
increase was observed after this. No significant response
3. Results was observed in the control lines C1 and C2, although the
(a) Test of feeding behaviour mean T fluctuated between 0.58 and 0.9. The highest mean
Within treatments, both red and blue food-deprived fish T was seen at generation 1 for both B1 and B2, and generation
(RH, BH) exhibited more spot pursuit behaviour than fed 3 for both R1 and R2. The subsequent decrease in the R chase
fish (RF, BF), indicating that our selection criteria were related scores after generation 3 may have been due to the highly
to foraging behaviour (figure 1). Chase scores (T ) were sig- elevated mortalities. Table 1 shows the realized heritabilities
nificantly higher in both RH and BH trials than in the BF both with and without standardization by the controls. The
and RF trials (Mann –Whitney U-tests: RH versus RF, W ¼ variation in scores in the R lines was notably higher than
733.5 p ¼ 0.018; BH versus BF, W ¼ 723.5, p ¼ 0.023). RH that of the B lines. Consistent with the food treatment exper-
fish exhibited more chases than their RF counterparts iment, no significant difference was observed between the
(Mann –Whitney U-tests: chases, W ¼ 752, p ¼ 0.0041), but sexes, again indicating that this is not a sex-linked behaviour
latency to chase and the number of orientations were not sig- and is therefore unlikely to be due to sexual preferences.
nificantly different (orientations, W ¼ 680.5, p ¼ 0.12; latency,
W ¼ 653, p ¼ 0.35). BH fish exhibited a higher number
of orientations and a lower latency to chase (orienta- (c) Test of possible visual mechanisms
tions, W ¼ 773, p ¼ 0.0018; latency, W ¼ 707.5, p ¼ 0.045) The chase score (T ) was recorded for three different spot col-
than BF fish. No significant difference was observed ours (blue, red and green) for each of the treatments at the
between sexes (W , 145, p . 0.33). All comparisons were final generation (5; the offspring of generation 4) to test
Holm –Bonferroni-corrected [34]. whether there was a correlated response between colours.
Across treatments, fish in the red treatments scored a sig- Figure 3 shows the results for the treatment replicates com-
nificantly higher T (food-deprived, W ¼ 717, p ¼ 0.0023; fed, bined (note that CR1/CB1 and CR2/CB2 are actually C1
W ¼ 711.5, p ¼ 0.0065) than the blue treatment. and C2, respectively; the mean of these is represented by
 (a) 5

rspb.royalsocietypublishing.org
6

mean chase score

B1
B2
4 CB1
CB2
CR1
CR2
R1
R2

Proc. R. Soc. B 282: 20143108

2

(b)
control-corrected mean chase score

B1
B2
R1
2 R2

0 1 2 3 4
generation
Figure 2. Mean chase scores (T) per generation with standard errors. (a) Mean scores and (b) control-corrected mean scores (experimental values – control values). R1
and R2 are the red-selected lines, B1 and B2 are the blue-selected lines, and CR1, CR2, CB1 and CB2 are red and blue control lines, respectively. Although the same fish
were measured for both the red and blue control lines, these are shown separately on the graph as CB1 and CR1, and CB2 and CR2. (Online version in colour.)

Table 1. Tables showing realized heritabilities (h 2) and their standard errors in parentheses, for generations zero to four, for each selection line, before control
correction and after control correction. Asterisks denotes significance at *p , 0.05 and **p , 0.01.

B1 B2 R1 R2

before control correction 0.096 (0.137) 0.081 (0.075) 0.298 (0.084)* 0.23 (0.068)
after control correction 0.030 (0.098) 0.070 (0.092) 0.304 (0.049)** 0.249 (0.116)

the control line in figure 3). Control-corrected scores were (532 nm) spot, and intermediate chase scores towards the
calculated by subtracting the mean of the C lines from the blue (450 nm) and red (650 nm) spots. Pair-wise Mann –
mean of the R and B lines. There was no difference in chase Whitney U comparisons on the uncorrected data indicate
score between the sexes ( p . 0.09) nor between the treatment that the R lines had significantly lower T towards the blue
replicates ( p . 0.06) within colours. (450 nm) spot and significantly higher T towards the red
The general trend was for both R and B selection lines to (650 nm) spot than the B and C lines (table 2). The C lines
have a significantly reduced response to the colours for also have a significantly higher T towards the green
which they were not selected, relative to the C lines. When (532 nm) spot than both the B and R lines.
control-corrected, the B-selected lines had the highest T
towards the blue (450 nm) spot, but lower T towards the
green (532 nm) and red (650 nm) spots. The R-selected lines
had the highest T towards the red (650 nm) spot, but lower 4. Discussion
scores towards the blue (450 nm) and green (532 nm) spots. Our study has shown that hunger-related colour preferences
The C lines had the highest chase score for the green are heritable and can respond significantly to artificial
 (a) species [35], but this is despite very high levels of inbreeding 6
5 and mortality in the final two generations, which would pre-

rspb.royalsocietypublishing.org
sumably have increased the effects of genetic drift. Previous
studies have shown that the colour sensitivity in the guppy
4 does respond to selection [21].
It can be difficult to determine exactly what artificial selec-
tion affects. In order to understand whether it was behaviour,
motivation or vision that we were selecting, we tested the
mean chase score

3 chase tendency of our final generation with three wavelengths:

blue (450 nm), green (532 nm) and red (650 m). If we were
blue selecting on a general motivation to chase, we would be likely
control
red to see an increase in the red lines’ chase scores towards the

Proc. R. Soc. B 282: 20143108

2 blue (450 nm) spot, and both red and blue lines should chase
the green (532 nm) spot as frequently as the control lines. How-
ever, our results show a decreased tendency to chase the spot
colours for which the lines were not selected. Disentangling a
1
motivation to chase red items and an increase in the visual sen-
sitivities for red is more difficult, although our results suggest
that changes in the visual system may have occurred.
0 Because our stimuli were of a very narrow bandwidth
(lasers), any correlated response to selection is most likely to
be owing to the broad spectral sensitivity of each guppy
(b)
cone. Guppies have cone types with maximum sensitivities
3 at 359 (ultraviolet, abundance  1%), 408 (short wavelength
violet, abundance  25%), 465 (short wavelength blue,
abundance  35%) and 533/548/572 nm (medium-long wave-
length, abundance  35%) [22,24]. Spectral sensitivity is
control-corrected mean chase score

2 known to respond to selection [21]. If one of our stimuli

strongly stimulated a cone, then the response should evolve
rapidly, but if it stimulated the cone weakly, the response
should be weaker, or even non-existent if stimulation is
blue below the cone’s sensitivity threshold.
1
red Selection on one wavelength should yield an apparent cor-
related response in responses to other wavelengths provided
they are sensed by the same cone. Colour is coded in vertebrate
(and invertebrate) visual systems by the differences between
0
cone outputs; this is called opponency [36,37]. If the response
to selection is in cone opponency, then we would expect a posi-
tive response to colours close to the favoured cone’s peak
sensitivity and a negative response to colours matching cones
–1 involved in opponency to the favoured cone. In the red lines,
there was a strong increase in the response to red, but a
400 450 500 550 600 650 decrease in responses to green and blue, compared with the
wavelength
controls. Similarly, the blue lines showed a significant decrease
Figure 3. Mean chase scores (T ) measured across three wavelengths (450 nm, in response to green and red compared with the control, and a
532 nm and 650 nm) for each treatment (replicates combined) with standard non-significant increase in their response to blue (figure 3).
errors. (a) Mean score and (b) control-corrected mean score. Control-corrected Although the opponency mechanism in guppies is not
scores were calculated by subtracting the mean of the C (control) lines from known, the pattern of positive and negative responses to selec-
the mean of the R (red) and B (blue) lines. C is the dashed line, B is the tion suggests that selection may have affected the colour
solid thick line and R is the solid thin line. Fish in the C lines were selected opponency mechanism rather than simply the cones most sen-
at random and each measured twice, once for chasing the red spot (CR) and sitive to the selected colour. The mechanism may lie in either
once for chasing the blue (CB) spot; the control line in the figure represents the retina or brain, but opponencies are well known in ver-
these fish. (Online version in colour.) tebrate retinas, involving the horizontal and/or amacrine
cells. Given this, it is simpler to propose that retinal opponency
may have evolved in the selected lines.
selection within only four generations. Surprisingly, we As it is chase tendency that we have selected, it is not sur-
found a response for only one of the two selected colours, prising that the red treatment lines had a strong chase tendency
which suggests limited evolutionary potential in our blue and response to selection. Guppies have multiple opsin genes
lines based solely on differences in food colour. We found that encode their long wavelength colour vision, which may
heritability (h 2) values of 0.25 (s.e. 0.12) and 0.30 (s.e. 0.05) provide broader visual sensitivity in the long wavelength
for the red treatment replicates, and 0.03 (s.e. 0.10) and 0.07 range of the colour spectrum [22–24]. This visual mode may
(s.e. 0.09) for the blue (h 2 ffi 0 for blue). These values are have evolved in response to high competition for orange- and
lower than others reported for foraging behaviours in this red-coloured fruit on which the guppies forage; these fruits
Table 2. Results of Mann–Whitney U-tests of the pair-wise comparison of chase score (T ) for selection lines towards the three light spots, red (650 nm, n ¼ 140), 7
green (532 nm, n ¼ 120) and blue (450 nm, n ¼ 140). Comparisons are Bonferroni-corrected.

rspb.royalsocietypublishing.org
blue spot (450 nm) green spot (532 nm) red spot (650 nm)

R versus B W ¼ 1493.5, p ¼ 0.03 W ¼ 804, p ¼ 1 W ¼ 587.5, p , 0.0001

B versus C W ¼ 787.5, p ¼ 1 W ¼ 498, p ¼ 0.01 W ¼ 1062, p ¼ 0.94
R versus C W ¼ 1545, p ¼ 0.0093 W ¼ 1096, p ¼ 0.012 W ¼ 485.5, p ¼ 0.0067

Proc. R. Soc. B 282: 20143108

are relatively rare and provide a favoured source of nutrients, acquisition; animals that are predated by species bearing a
creating high levels of competition [38]. Additionally, selection given colour should not actively forage on food of the same
on increased chase tendency for long wavelengths has resulted colour. This interaction could lead to interesting dynamics
in decreased chase tendency towards short wavelengths. regarding two opposing naturally selected traits—foraging
Having a relatively increased sensitivity for long wavelengths and predation—that have not been considered before. Unfortu-
may enable individuals to forage on food items of this colour nately, the only natural predator having blue reflectance is the
more efficiently and would result in a higher chase tendency very weak predator Aequidens pulchur; the other natural guppy
for the red spot over the blue; blue food items (fruit) are rarer predators are much more dangerous [25] and lack blue entirely.
in the wild than red or orange fruit in the rainforests above The best explanation for the lack of blue response is the
the native guppy streams. Changes in the visual system are colour sensitivity of vertebrate motion detectors. Motion detec-
likely to be altered through the up- and downregulation of tion in fishes is thought to be mediated via cones containing
the opsin genes [39–41]. Given the extensive complement of long-wavelength-sensitive opsins [47,48]. We selected guppies
opsins, particularly long-wavelength-sensitive opsins in gup- to chase moving blue spots, but if the motion-detection system
pies [22], spectral sensitivity can easily be modified by is insensitive to the short-wavelength blue spots, then this
evolved changes in opsin regulation patterns. means that there is no genetic variation for favouring blue
This sensory bias, potentially from food detection, has also moving objects, and hence no evolutionary response is poss-
been linked to female mate choice in this species [15]. Female ible. The fish were able to orientate towards the spot, and
guppies have a preference for orange coloration in males, and this may have been due to seeing the spot, but a lack of
this is prevalent in many populations [16,42,43], but not in all motion detection for blue would make following the spot’s
[17,44]. It has been suggested that this has evolved through path difficult. The colour-sensitive nature of motion detection
the visual sensitivity towards long wavelengths that probably suggests that blues, violets and UV-based colours should not
originated from carotenoid-rich food sources such as orange be used in sexual displays involving a lot of movement, and
fruits [15,45]. Our study helps support these findings by con- this prediction holds in any species where shorter wavelengths
firming not only a foraging preference for long wavelengths, are outside the spectral sensitivity of motion-detection systems.
but also the evolutionary potential for it to be selected; we We have demonstrated artificial selection for foraging behav-
have shown that there is a high heritability for foraging behav- iour based solely on food colour preferences. Our results indicate
iour towards red-coloured objects. This is essential for this trait that a response to selection can be elicited in a very short period
to be co-opted as a mate choice criterion. of time, which may serve to allow populations to respond readily
The lack of response in the blue line is more difficult to to a change in foraging conditions, such as increased water flow
explain. Although we might have expected the initial chase or high competition for food resources. We have also demon-
tendency of the blue lines to be lower than those of the red, strated that foraging behaviour towards some food colours
because blue fruit is rare in the wild, we would still expect may not be able to evolve. This may be due to a number of
some behavioural and evolutionary response. Our study reasons, the most probable being the visual physiology of the
has shown that hungry fish had a higher chase tendency fish. The relationship between colour and evolvability may
than fed fish, even for blue spots, and so we are confident affect the form of many different kinds of visual signals, and
than it is foraging behaviour that we are measuring. Unlike may well extend to other aspects of visual signals.
the red treatment, there is no historic association with blue
food, which would make the response to selection slower. Ethics statement. The experiment was approved by the Deakin Univer-
sity Animal Ethics Committee, approval numbers A21-2010 and
Carotenoid-based foods may be nutritionally more attractive
G01-2012.
than blue fruits and, if so, a mechanism to exploit them as a
Data accessibility. The raw data for this study can be found in the Dryad
food source may have never evolved. It may also be that the data repository: doi:10.5061/dryad.k6j0v.
reward of foraging on blue food items is highly variable, and Acknowledgements. The authors thank Priti Singh and Xandy Kranz for
therefore a higher degree of learning has evolved. A lack of their assistance with fish husbandry, and Ron Douglas for providing
genetic variation is a barrier to selection [46], and it is poss- the guppy cornea and lens transmission spectrum. We are also grateful
ible that this is the case in this study system, where the fish to the Australian Research Council for support (grant no.
DP110101421) and to three anonymous reviewers whose comments
have limited behavioural variation towards the blue light.
helped to improve the manuscript.
Another explanation is that this colour has implications out-
Author’s contributions. The experiment was designed primarily by J.A.E.
side of foraging, such as for predator avoidance. If blue is linked and the data were collected, and analysed, by G.L.C. The manuscript
to predators, then this could explain avoidance of blue items, was prepared by G.L.C and was revised by J.A.E.
particularly moving ones. This leads to interesting predictions Competing interests statement. The authors do not have any competing
regarding a trade-off between predator avoidance and food interests.
References 8

rspb.royalsocietypublishing.org
1. Lehrer M, Horridge GA, Zhang SW, Gadagkar R. coturnix japonica). Science 207, 549–551. (doi:10. 86, 405 –431. (doi:10.1111/j.1095-8312.2005.
1995 Shape vision in bees: innate preference 1126/science.7352267) 00540.x)
for flower-like patterns. Phil. Trans. R. Soc. 15. Rodd FH, Hughes KA, Grether GF, Baril CT. 30. Endler JA. 2012 A framework for analysing
Lond. B 347, 123–137. (doi:10.1098/rstb. 2002 A possible non-sexual origin of mate colour pattern geometry: adjacent colours.
1995.0017) preference: are male guppies mimicking fruit? Biol. J. Linn. Soc. 107, 233 –253. (doi:10.1111/j.
2. Hill PSM, Wells PH, Wells H. 1997 Spontaneous Proc. R. Soc. Lond. B 269, 475– 481. (doi:10.1098/ 1095-8312.2012.01937.x)
flower constancy and learning in honey bees as a rspb.2001.1891) 31. R Core Team. 2013 R: a language and
function of colour. Anim. Behav. 54, 615 –627. 16. Kodric-Brown A. 1985 Female preference and sexual environment for statistical computing. Vienna,
(doi:10.1006/anbe.1996.0467) selection for male coloration in the guppy (Poecilia Austria: R Foundation for Statistical Computing.

Proc. R. Soc. B 282: 20143108

3. Goulson D, Cory JS. 1993 Flower constancy reticulata). Behav. Ecol. Sociobiol. 17, 199 –205. 32. Winge O. 1937 Succession of broods in Lebistes.
and learning in foraging preferences of the green- (doi:10.1007/BF00300137) Nature 140, 467. (doi:10.1038/140467b0)
veined white butterfly Pleris napi. Ecol. Entomol. 17. Endler JA, Houde AE. 1995 Geographic-variation in 33. Falconer DS, Mackay TF, Frankham R. 1996
18, 315–320. (doi:10.1111/j.1365-2311.1993. female preferences for male traits in Poecilia Introduction to quantitative genetics (4th edn).
tb01107.x). reticulata. Evolution 49, 456–468. (doi:10.2307/ Trends Genet. 12, 280.
4. Meléndez-Ackerman E, Campbell DR, Waser NM. 2410270). 34. Holm S. 1979 A simple sequentially rejective
1997 Hummingbird behavior and mechanisms of 18. Dussault GV, Kramer DL. 1981 Food and feeding multiple test procedure. Scand. J. Stat. 6, 65 –70.
selection on flower color in ipomoPSIS. Ecology 78, behavior of the guppy, Poecilia reticulata (Pisces: (doi:10.2307/4615733)
2532–2541. (doi:10.1890/0012-9658(1997)078 Poeciliidae). Can. J. Zool. 59, 684–701. (doi:10. 35. Karino K, Utagawa T, Shinjo S. 2005 Heritability of
[2532:HBAMOS]2.0.CO;2) 1139/z81-098) the algal-foraging ability: an indirect benefit of
5. Whitfield M, Köhler A, Nicolson SW. 2014 Sunbirds 19. Houde AE. 1997 Sex, color, and mate choice female mate preference for males’ carotenoid-based
increase foraging success by using color as a cue for in guppies. Princeton, NJ: Princeton University coloration in the guppy, Poecilia reticulata. Behav.
nectar quality. Behav. Ecol. 25, 328–334. (doi:10. Press. Ecol. Sociobiol. 59, 1–5. (doi:10.1007/s00265-005-
1093/beheco/art115) 20. Magurran AE. 2005 Evolutionary ecology. The 0001-3)
6. Raine NE, Chittka L. 2007 The adaptive significance trinidadian guppy. Oxford, UK: Oxford University 36. Hering E. 1878 Zur Lehre vom Lichtsinne: Sechs
of sensory bias in a foraging context: floral colour Press. Mitteilungen an die Kaiserl. Vienna, Austria:
preferences in the bumblebee Bombus terrestris. 21. Endler JA, Basolo A, Glowacki S, Zerr J. Akademie der Wissenschaften; C. Gerold’s Sohn.
PLoS ONE 2, e556. (doi:10.1371/journal.pone. 2001 Variation in response to artificial 37. Hurvich LM, Jameson D. 1957 An opponent-process
0000556) selection for light sensitivity in guppies theory of color vision. Psychol. Rev. 64, 384 –404.
7. Smith AC, Buchanan-Smith HM, Surridge AK, (Poecilia reticulata). Am. Nat. 158, 36 –48. (doi:10. (doi:10.1037/h0041403)
Osorio D, Mundy NI. 2003 The effect of colour vision 1086/320862) 38. Reznick D, Butler IV MJ, Rodd H. 2001 Life-
status on the detection and selection of fruits by 22. Laver CR, Taylor JS. 2011 RT-qPCR reveals history evolution in guppies. VII. The
tamarins (Saguinus spp.). J. Exp. Biol. 206(Pt 18), opsin gene upregulation associated with age and comparative ecology of high-and low-predation
3159–3165. (doi:10.1242/jeb.00536) sex in guppies (Poecilia reticulata)—a species with environments. Am. Nat. 157, 126–140. (doi:10.
8. Dominy NJ, Lucas PW. 2001 Ecological importance color-based sexual selection and 11 visual-opsin 1086/318627)
of trichromatic vision to primates. Nature 410, genes. BMC Evol. Biol. 11, 81. (doi:10.1186/1471- 39. Parry JWL, Carleton KL, Spady T, Carboo A,
363–366. (doi:10.1038/35066567) 2148-11-81) Hunt DM, Bowmaker JK. 2005 Mix and match
9. Regan BC, Julliot C, Simmen B, Vienot F, Charles- 23. Archer SN, Endler JA, Partridge JC. 1987 Visual color vision: tuning spectral sensitivity by
Dominique P, Mollon JD. 2001 Fruits, foliage and pigment polymorphism in the guppy Poecilia differential opsin gene expression in lake
the evolution of primate colour vision. Phil. reticulata. Vision Res. 27, 1243–1252. (doi:10.1016/ Malawi cichlids. Curr. Biol. 15, 1734–1739.
Trans. R. Soc. Lond. B 356, 229– 283. (doi:10.1098/ 0042-6989(87)90200-8) (doi:10.1016/j.cub.2005.08.010)
rstb.2000.0773) 24. Archer SN, Lythgoe JN. 1990 The visual pigment 40. Carleton KL, Kocher TD. 2001 Cone opsin genes of
10. Osorio D, Vorobyev M. 1996 Colour vision as basis for cone polymorphism in the guppy, Poecilia African cichlid fishes: tuning spectral sensitivity by
an adaptation to frugivory in primates. Proc. R. reticulata. Vision Res. 30, 225 –233. (doi:10.1016/ differential gene expression. Mol. Biol. Evol. 18,
Soc. Lond. B 263, 593–599. (doi:10.1098/rspb. 0042-6989(90)90038-M) 1540– 1550. (doi:10.1093/oxfordjournals.molbev.
1996.0089) 25. Endler JA. 1978 A predator’s view of animal color a003940)
11. Schaefer HM, McGraw K, Catoni C. 2008 Birds use patterns. Evol. Biol. 11, 319– 364. 41. Fuller RC, Carleton KL, Fadool JM, Spady TC,
fruit colour as honest signal of dietary antioxidant 26. Endler JA. 1993 Some general comments on the Travis J. 2004 Population variation in opsin
rewards. Funct. Ecol. 22, 303 –310. (doi:10.1111/j. evolution and design of animal communication expression in the bluefin killifish, Lucania
1365-2435.2007.01363.x) systems. Phil. Trans. R. Soc. Lond. B 340, 215– 225. goodei: a real-time PCR study. J. Comp. Physiol.
12. Caine NG, Mundy NI. 2000 Demonstration of a (doi:10.1098/rstb.1993.0060) A 190, 147–154. (doi:10.1007/s00359-003-
foraging advantage for trichromatic marmosets 27. Endler JA. 1990 On the measurement and 0478-z)
(Callithrix geoffroyi) dependent on food colour. classification of color in studies of animal color 42. Long KD, Houde AE. 1989 Orange spots as a
Proc. R. Soc. Lond. B 267, 439–444. (doi:10.1098/ patterns. Biol. J. Linn. Soc. 41, 315 –352. (doi:10. visual cue for female mate choice in the
rspb.2000.1019) 1111/j.1095-8312.1990.tb00839.x) guppy (Poecilia reticulata). Ethology 82,
13. Weiss MR. 1997 Innate colour preferences and 28. Endler JA. 1991 Variation in the appearance of 316–324. (doi:10.1111/j.1439-0310.1989.
flexible colour learning in the pipevine swallowtail. guppy color patterns to guppies and their predators tb00511.x)
Anim. Behav. 53, 1043 –1052. (doi:10.1006/anbe. under different visual conditions. Vision Res. 31, 43. Endler JA. 1995 Multiple-trait coevolution and
1996.0357) 587 –608. (doi:10.1016/0042-6989(91)90109-I) environmental gradients in guppies. Trends Ecol.
14. Kovach J. 1980 Mendelian units of inheritance 29. Endler JA, Mielke PW. 2005 Comparing entire Evol. 10, 22 –29. (doi:10.1016/S0169-5347(00)
control color preferences in quail chicks (Coturnix colour patterns as birds see them. Biol. J. Linn. Soc. 88956-9)
44. Brooks R, Endler JA. 2001 Female guppies agree to development of a colour-based mate preference and 47. Schaerer S, Neumeyer C. 1996 Motion detection in 9
differ: phenotypic and genetic variation in mate- the sensory bias to which it is genetically linked. goldfish investigated with the optomotor response

rspb.royalsocietypublishing.org
choice behavior and the consequences for sexual Proc. R. Soc. B 272, 2181–2188. (doi:10.1098/rspb. is ‘color blind’. Vision Res. 36, 4025–4034. (doi:10.
selection. Evolution 55, 1644 –1655. (doi:10.1111/j. 2005.3197) 1016/S0042-6989(96)00149-6)
0014-3820.2001.tb00684.x) 46. Blows MW, Hoffmann AA. 2005 A reassessment of 48. Krauss A, Neumeyer C. 2003 Wavelength
45. Grether GF, Kolluru GR, Rodd FH, de la Cerda J, genetic limits to evolutionary change. Ecology 86, dependence of the optomotor response in zebrafish
Shimazaki K. 2005 Carotenoid availability affects the 1371 –1384. (doi:10.1890/04-1209) (Danio rerio). Vision Res. 43, 1275–1284.

Proc. R. Soc. B 282: 20143108