Behavioral Ecology Advance Access originally published online on July 8, 2008
Behavioral Ecology 2008 19(5):1012-1017; doi:10.1093/beheco/arn069
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Being conspicuous and defended: selective benefits for the individual
Centre for Behaviour and Evolution, Institute of Neuroscience, Newcastle University, Henry Wellcome Building, Framlington Place, Newcastle upon Tyne, NE2 4HH, UK
Address correspondence to C. Halpin. E-mail: christina.halpin{at}ncl.ac.uk.
Received 12 March 2008; revised 17 April 2008; accepted 26 April 2008.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONFUNDINGREFERENCES |
|---|
Aposematic insects conspicuously advertise their unprofitability to potential predators. However, when these prey initially evolved, they were likely to have been rare and presumably at a greater risk of being detected and killed by naive predators. Both kin and individual selection theories have been used in attempts to explain this apparent paradox, with much of the empirical research supporting kin selection–based theories. Here, we experimentally test how chemical defence levels in prey and avian color biases influence the probability of a rare conspicuous morph having an initial survival advantage. We used newly hatched domestic chicks (Gallus gallus domesticus) foraging on green and purple prey, on a green or purple background, to model the evolutionary scenario of a rare conspicuous morph arising in a population of already defended cryptic prey. Defended prey were produced by spraying them with quinine solution, which the birds readily detect and can learn to avoid. Although attack rates were initially similar for both defended prey types, the chicks only learned to avoid defended prey when they were conspicuous, not when they were cryptic. In addition, defended conspicuous prey were more likely to be rejected on attack than defended cryptic prey, even when first encountered by a predator. These data suggest that there could be a selective advantage for a rare conspicuous morph to arise in a population of cryptic defended prey due to increased avoidance learning and taste-rejection in naive predators. Our findings also suggest that being a non-preferred color and/or highly defended will increase the probability of this evolutionary scenario.
Key words: aposematism, avoidance learning, color bias, predation, receiver psychology, taste-rejection, warning signal.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONFUNDINGREFERENCES |
|---|
Insects that use chemical defences in some way are often brightly colored and/or have distinct markings to advertise the fact that they are unprofitable to potential predators and are referred to as being "aposematic" (Cott 1940
; Friedlander 1976; Atkins 1980; Guilford 1988). Aposematic insects increase their chances of survival because predators are able to learn to avoid conspicuous defended prey more quickly than cryptic ones, as well as remember associations with conspicuous prey for longer and make fewer recognition errors (Gittleman and Harvey 1980; Guilford 1986; Roper and Redston 1987). Initially, however, warning coloration would likely increase an individual's risk of being detected and attacked by inexperienced predators (Gittleman and Harvey 1980; Guilford 1985, 1988; Roper and Wistow 1986). It would also likely take time and a certain number of encounters between predators and novel aposematic prey before the predators learned to avoid them (Guilford 1985). So the question is how a novel conspicuous morph could evolve and survive long enough to spread in a population in the first place (Härlin C and Härlin M 2003; Speed and Ruxton 2005; Tullberg et al. 2005).
Many of the previous studies attempting to shed light on this question have concentrated on looking for benefits of being aposematic when gregarious (Gamberale and Tullberg 1996, 1998; Riipi et al. 2001). However, many aposematic insect species are known to be solitary (Tullberg and Hunter 1996; Tullberg et al. 2000) and a small number of studies have suggested that warning coloration can impose a selective benefit to the individual (Wiklund and Järvi 1982; Sillen-Tullberg 1985; Engen et al. 1986; Evans 1987). Many prey secrete their defences on attack (e.g., Eisner and Meinwald 1966; Pasteels et al. 1983; de Jong et al. 1991), and avian predators are able to taste and discriminate among defended prey on the basis of their level of defence (Wiklund and Järvi 1982; Sillen-Tullberg 1985; Gamberale-Stille and Guilford 2004; Skelhorn and Rowe 2006a, 2006d). There is some evidence that aposematic insects are more likely to be tasted and rejected by predators than cryptic defended insects (Wiklund and Järvi 1982; Sillen-Tullberg 1985; Leimar et al. 1986) and many insects can survive a predatory attack if rejected (Wiklund and Järvi 1982; Sillen-Tullberg 1985; DeVries 2002). The predatory behavior of birds foraging on rare conspicuous prey with externally detectable toxins may therefore provide a route by which costly conspicuous signals can evolve by individual selection.
Although the pathway by which aposematism initially evolved is still debated, the most widely accepted theory is that warning coloration evolved in cryptic insects that were already defended in some way and that these in turn evolved from cryptic undefended forms (Guilford 1988; Marples et al. 2005). Perhaps surprisingly, only 2 experimental studies have looked at the initial evolution of conspicuousness within a defended cryptic prey population (Alatalo and Mappes 1996; Halpin et al. 2008). We have previously shown that a rare aposematic prey can be favored above an equally defended, but more abundant, cryptic prey type (Halpin et al. 2008). This was due to increased visual discrimination and avoidance learning, as well as predators being more likely to taste and reject a defended prey item that was conspicuous compared with one that was cryptic. However, our results also suggested that unlearned color biases play a role in the degree of learned avoidance (Halpin et al. 2008), raising the interesting questions of how much the color and the level of chemical defence influence the probability of a novel conspicuous morph having an initial survival advantage. Here, we test our prediction that predator color biases and prey defence levels will alter the survival chances of rare conspicuous defended morphs arising in a population of equally defended cryptic population. Crucially, we examine not only the avoidance learning rates but also the survival probabilities of the defended prey attacked, comparing the rejection rates postattack of conspicuous and cryptic defended prey. Our study explicitly addresses the question of whether or not conspicuousness, and/or rarity, per se affects the survival of equally defended prey. In addition, it is the first study to consider avian color biases in this context and how these might have an impact on selection for insect defences.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONFUNDINGREFERENCES |
|---|
Subjects and housing
Fifty-five domestic chicks (Gallus gallus domesticus) were hatched in the laboratory and housed in cages measuring 100 x 50 x 50 cm. We used domestic chicks because we can control their foraging experience immediately on hatching, ensuring that our results do not derive from previous learned experiences. The chicks were maintained at a room temperature of 25 °C and kept on a constant 14:10 h light:dark cycle, using fluorescent lights with no UV component. Water was available ad libitum, as were unmanipulated (brown) chick starter crumbs, except during training and experimental periods when food deprivation was necessary (see below). The experiment was carried out in accordance with the Association for the Study of Animal Behaviour and Home Office guidelines for animal research. Chicks were donated to free-range farms at the end of the experiment.
Artificial prey
Our experiment required crumbs that were either cryptic or conspicuous and defended or undefended. To create mildly and moderately defended prey, 600 g of chick crumbs were sprayed with 400 mL of 1% quinine sulfate solution and another 600 g were sprayed with 400 mL of 4% quinine sulfate solution. Undefended crumbs were sprayed with the equivalent amount of water to control for any change in texture. The crumbs were then spread out in a single layer and left to dry for 24 h after which they were sieved to ensure they were of similar size. Crumbs were then colored in order that we could manipulate their conspicuousness. To create cryptic and conspicuous crumbs, 300 g of the mildly defended crumbs and 300 g of the moderately defended crumbs were sprayed either with 180 mL of a solution of green food dye (Sugarflair, Benfleet, UK) (2.4 mL dye/180 mL water) or with 180 mL of a solution of purple food dye (Squires Kitchen, Farnham, UK) (3.9 mL dye/180 mL water). Undefended crumbs were dyed in the same way. We used green and purple dyes to color the crumbs because chicks have previously shown a preference for purple crumbs over green, and this allowed us to look at the effect of defence level on color biases (Halpin et al. 2008). Dye concentrations were chosen that, to our eyes, gave both crumb types a similar level of color saturation. All crumbs were again left to dry for 24 h and sieved one final time.
Experimental arena
The experimental arena was the same as in Halpin et al. (2008) with a buddy area at one end of the cage to contain 2 "buddy chicks" during training and experimental trials. The floor of the experimental arena was covered with either green- or purple-colored laminated paper (Halpin et al. 2008), which were colored using the same concentrations of food dyes as for the crumbs. A grid was drawn onto each background enabling us to identify individual crumbs (Halpin et al. 2008).
Training
Forty-four chicks were trained to forage for brown chick starter crumbs in the test arena, with the remaining 11 being used as buddy chicks. Half the experimental chicks were trained on a purple background, and the other half were trained on a green background with the training procedure following that used in Halpin et al. (2008). Four chicks that did not eat readily in the arena after training were not included in the experiment.
Experimental trials
Individual chicks were tested on the same color floor that they had experienced during training, but now chicks received a choice between purple and green crumbs. There were 4 experimental groups; 2 groups were tested on the green background, making the green crumbs cryptic and the purple crumbs conspicuous, while the other 2 groups were tested on the purple background where purple crumbs were cryptic and green crumbs were conspicuous. Each background color treatment was further divided by giving chicks either mildly or moderately defended crumbs (see Table 1 for summary of experimental groups).
|
In each experimental trial, all chicks were given a choice between 20 undefended cryptic prey, 15 defended cryptic prey, and 5 defended conspicuous prey (Halpin et al. 2008
). This experimental design enabled us to directly compare attack- and taste-rejection rates for defended cryptic and conspicuous prey, in a setting that mimics the evolutionary scenario where a distinct conspicuous morph arises in an already defended cryptic population. Each background color treatment was further divided by giving chicks either mildly or moderately defended crumbs with all defended crumbs within an experimental group being equally defended. The experimental groups were named according to the color and defence level of the conspicuous prey they were given: the purple mildly defended group received mildly defended purple crumbs on a green background and the purple moderately defended group received moderately defended purple crumbs on a green background. The equivalent groups on the purple background were named green mildly defended group and green moderately defended group, respectively. Crumbs were individually placed within the grid on the colored backgrounds using randomized maps that were created in Microsoft Excel. In each trial, the chicks were allowed to attack 16 crumbs before being removed from the arena. They received 2 such trials per day for 5 consecutive days.
Data analysis
First, we calculated the attack risk for each prey type by dividing the number of each prey type attacked by the number of that type that was available. To determine whether defended conspicuous prey were attacked more than the equally defended cryptic prey, we calculated the overall attack risk for each defended prey in our 4 groups across all 10 trials. The probability of being taste-rejected was calculated as the proportion of attacked prey that were not eaten. This was calculated for each prey type in each experimental group.
To test whether birds' responses to each prey type differed among groups, we used analyses of variance (ANOVAs) and independent t-tests. However, comparisons between prey types required us to use repeated measures ANOVAs. This is because each bird experienced a number of prey types giving us a within-subject design. The repeated measures ANOVA used prey type as the repeated measure, with defence level and conspicuous prey color as independent variables. This allowed us to test for differences in birds' responses to each prey type both between and within experimental groups (see Halpin et al. 2008). In addition, because the within-subject attack data for the different prey types were not independent of each other, using a standard ANOVA was not appropriate. All assumptions for ANOVA were met, and where appropriate, P values were corrected for multiple comparisons using a Bonferroni correction.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONFUNDINGREFERENCES |
|---|
Defence level and visual avoidance learning of conspicuous prey
Calculating the attack risk for each prey type in each trial gave us a measure of learned avoidance based on visual features of the prey, which could be compared among the prey types regardless of their different frequencies. There were clear differences in the attack risk across trials between our 4 groups (see Figure 1). As in our previous study, chicks did not appear to learn to avoid conspicuous defended prey when they were purple and only mildly defended (Figure 1c). Statistical analyses confirmed that there was a significant interaction between conspicuous prey color and the proportions of each prey type attacked (repeated measures ANOVA, F1,29 = 24.858, P < 0.001). There was also a significant interaction between prey type and defence level (F1,29 = 57.962, P < 0.001) and between the defence level and the conspicuous prey color (F1,29 = 28.991, P < 0.001). Given these interactions and the clear difference between the purple mildly defended group and the others, all further analyses were conducted on the purple and green experimental groups separately (Halpin et al. 2008
).
|
When green prey were conspicuous, they had a significantly lower attack risk than the equally defended purple cryptic prey (F1,14 = 141.8, P < 0.001; see Figure 1a,b), and as expected, moderately defended prey had an overall lower attack risk than mildly defended prey (F1,14 = 9.09, P < 0.01). Post hoc independent t-tests revealed, however, that there was no difference in the attack risk between the moderately and the mildly defended cryptic prey (t = 0.220, degrees of freedom [df] = 14, P > 0.05), whereas moderately defended conspicuous prey had a significantly lower attack risk than the mildly defended conspicuous prey (t = 3.007, df = 14, P < 0.01).
When purple prey were conspicuous, there was a significant interaction between prey type and defence level (F1,15 = 19.43, P < 0.001). Post hoc paired t-tests revealed that conspicuous defended prey were at a lower attack risk than cryptic prey when they were moderately defended (t = 7.031, df = 7, P < 0.001), but not when they were mildly defended (t = 1.449, df = 8, P > 0.05). Similarly, comparing attack risks on defended prey between groups, there was no difference between the moderately and mildly defended cryptic prey (t = 1.941, df = 15, P > 0.05), but moderately defended conspicuous prey had a significantly lower attack risk than the mildly defended conspicuous prey (t = 4.371, df = 15, P < 0.001).
Taken together, these data show that the chicks learned to avoid both moderately and mildly defended conspicuous prey when they were green but only learned to avoid moderately and not mildly defended purple conspicuous prey. This supports our previous findings that chicks have a preference for purple prey even when sprayed with 1% quinine solution (Halpin et al. 2008) but extends this to show that increasing defence levels can overcome the preference.
Comparing taste-rejection rates for defended cryptic and conspicuous prey
Although chicks attacked 16 prey in each trial, not all prey were eaten. The chicks would sometimes taste prey and reject them, a behavior that could allow prey to survive in the wild. In order to see if there were differences between the defended cryptic and conspicuous prey, we compared the probability of being taste-rejected for these prey types.
In contrast to the attack data, the taste-rejection data in all groups showed a similar pattern (see Figure 2). We therefore combined our data from both color treatments in a repeated measures ANOVA (the proportional data were arcsine transformed to restore normality). In agreement with previous studies (Skelhorn and Rowe 2006a, 2006d), we found that moderately defended prey were rejected at more than mildly defended prey (F1,29 = 7.03, P < 0.05). We also found a significant interaction between prey type and the color of the conspicuous prey (F1,29 = 5.56, P < 0.05). This was due to the difference in the probability of being taste-rejected between conspicuous and cryptic defended prey being greater for groups where green was the conspicuous color, compared with where the conspicuous prey were purple. Post hoc t-tests revealed that significantly more conspicuous prey were tasted and rejected, at both defence levels, when green was the conspicuous color (green mildly defended: t = 4.292, df = 7, P = 0.004; green moderately defended: t = 3.934, df = 7, P = 0.006). However, although the results were qualitatively similar when conspicuous prey were purple (see Figure 2), we did not find any significant difference between the taste-rejection probabilities (purple mildly defended: t = 1.576, df = 8, P > 0.05; purple moderately defended: t = 1.778, df = 7, P > 0.05).
|
These data show that chicks can taste-reject conspicuous prey more than equally defended cryptic prey, but importantly the color and defence level of the conspicuous prey influence the probability of a prey being taste-rejected. However, these data are for all trials combined, which does not measure the initial survival advantage to conspicuous prey when encountered by predator for the very first time. To investigate whether or not conspicuousness affected taste-rejection behavior on first encounter, we analyzed the behavior of those chicks whose first attack in Trial 1 was on a defended prey. Across the 4 groups, there were 20 chicks in total that attacked a defended prey first. For all groups, although the numbers are small, the pattern was the same; there was a higher probability of taste-rejection for conspicuous defended prey compared with cryptic defended prey (see Table 2), a pattern also seen in other similar studies (Halpin C, Rowe C, unpublished data). There were not sufficient data to test each group separately, and we therefore pooled the data to look for a difference in taste-rejection between cryptic and conspicuous defended prey. Overall, the chicks did reject more defended prey when they were conspicuous compared with when they were cryptic (Fisher's Exact test: 0.88 vs. 0.33, respectively, P < 0.05), providing crucial evidence for an initial selective advantage for conspicuous prey by individual selection.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONFUNDINGREFERENCES |
|---|
We have found that rare defended conspicuous prey can benefit from a reduced rate of attack and an increased probability of being released on attack compared with an equally defended cryptic prey type. Indeed, there can be a selective advantage to rare conspicuous morphs by increased rejection even in the first attack. Notably, our study is the first to compare taste-rejection probabilities for equally defended conspicuous and cryptic prey in a situation where both the conspicuousness and the defence level of the prey are manipulated. Our data highlight the importance of postattack behavior in studies of the evolution of aposematism and suggest that aposematic prey could potentially evolve through individual selection.
However, the selective advantage potentially enjoyed by rare conspicuous prey clearly depends on the color and defence level of the prey. First, the color of the prey affects the degree to which predators learn to avoid conspicuous defended prey (Halpin et al. 2008). Chicks are known to have unlearned color biases (e.g., Roper and Cook 1989; Rowe and Guilford 1996; Marples and Roper 1997; Gamberale-Stille and Tullberg 2001; Jetz et al. 2001; Lindström et al. 2001; Rowe and Skelhorn 2005), and other recent experiments have found a bias against purple food in naive individuals (Prior and Wilzeck 2008; Halpin et al. 2008). Although adaptive color biases are known to be influential in foraging contexts across a wide range of other species (e.g., Stiles 1976; Weiss and Papaj 2003; Schmidt and Schaefer 2004; e.g., Raine and Chittka 2007), we cannot provide an explanation as to why chicks might prefer purple to green food. However, whatever the reason for this color bias, our data suggest that a conspicuous prey of a preferred color would require a greater level of defence in order to benefit from the same levels of avoidance learning in naive predators as a conspicuous prey of a less preferred color (in this case green). Producing and sequestering chemical defences is often costly, and the level of defence varies both within and between species (Ruxton et al. 2004); yet, studies specifically investigating the evolution of these defences in insects are surprisingly rare (e.g., Skelhorn and Rowe 2006c; Skelhorn and Ruxton 2008). The results from this study suggest that how much insects invest in costly chemical defences may be influenced by avian color biases and that color biases might directly influence selection on chemical defences. We do not know whether the unlearned biases that we see in avian predators today existed before the evolution of warning coloration or resulted from it (Schuler and Roper 1992; Guilford and Dawkins 1993; Rowe and Guilford 1999; Rowe and Skelhorn 2004), but if these biases did exist prior to the evolution of aposematic insects, it is likely that selection would favor insects of unpreferred colors which birds will learn to avoid even at relatively low levels of defence.
Second, in conjunction with the color biases, the level of defence was clearly important for the avoidance learning and postattack survival of conspicuous prey. As found in a previous experiment (Skelhorn and Rowe 2006c), the greater the defence level the greater the degree of avoidance learning: chicks learned to avoid green conspicuous defended prey faster when they were moderately compared with mildly defended and only learned to avoid purple conspicuous prey when they were moderately defended. In addition, the defence level was key in determining postattack survival, with mildly defended prey less likely to be rejected than moderately defended prey. This in itself has been found in previous experiments (Skelhorn and Rowe 2006a, 2006d); however, in this study, the probability of being taste-rejected was also influenced by the color of the prey. When comparing the level of taste-rejection for conspicuous and cryptic prey, this was much higher when green was the conspicuous color compared with purple. This again highlights the potential role of color biases in prey survival and suggests that it may not be just the defence level that determines prey taste-rejection rates.
There are arguments that could be made against our taste-rejection result. First, it could be that our differences in taste rejection between the conspicuous and cryptic prey is the result of receiver error because there were also undefended cryptic prey present. It is possible that defended cryptic prey were ingested because they were confused with the undefended cryptic prey type, whereas defended conspicuous prey had a more distinct signal. However, this is not supported by our finding that prey rejection of conspicuous prey was higher than cryptic prey even in the very first attack where chicks had no additional information about the palatability of other prey types. We do not know why the chicks were more likely to reject a conspicuous prey than an equally defended cryptic one, but it is possible that they are innately cautious toward conspicuously colored prey (Sillen-Tullberg 1985; Sherratt 2002; Exnerova et al. 2003) and/or that the chicks attacking a conspicuous prey perceived it as being worse tasting than the chicks attacking a cryptic prey (e.g., Skelhorn and Rowe 2006b). Second, it could be argued that the taste-rejection data in our study using artificial prey do not transfer to natural foraging events. However, both our own experiments with chicks (Halpin CG, unpublished data) and those with other predators have shown that predators can reject live prey post-attack at similar rates (e.g., Sillen-Tullberg 1985) and that prey can survive a predatory attack if they are rejected (Wiklund and Järvi 1982; Sillen-Tullberg 1985; Evans 1987). The survival chances of the prey will depend on the toughness of the external tegument or exoskeleton and also the handling by the predator (Cott 1940; Wiklund and Järvi 1982; Sillen-Tullberg 1985; Evans 1987). The degree to which prey can survive attacks will be crucial for the evolution of conspicuous coloration because conspicuous signals will attract visually hunting predators (Roper and Wistow 1986; Alatalo and Mappes 1996; Lindström 1999; Halpin et al. 2008). Conspicuous coloration will only evolve if the survival rates are improved, which our data suggest can potentially be the case through avoidance learning and changes in taste-rejection rates.
Overall, our data support the theory that the evolution of aposematism can evolve via individual selection. Our previous work has found an individual selective advantage for chemical defences to arise in a cryptic population, and these data demonstrate that a selective advantage can exist for rare conspicuous defended prey arising in an already defended cryptic population (Skelhorn and Rowe 2006d; Halpin et al. 2008). Although our experiment shows that chicks will readily attack novel conspicuous prey, it is important to remember that predators vary greatly in their foraging behavior, their susceptibility to defences, and in their ability to learn (Endler and Mappes 2004), and dietary conservatism by wild birds might only enhance the effect that we have measured here (Marples et al. 2005). The question remains of how the detectability of a conspicuous signal and the postattack survival of the prey combine to influence the evolution of prey defences and aposematic coloration; however, this remains an intriguing area for future research.
|
FUNDING |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONFUNDINGREFERENCES |
|---|
C.G.H. is supported by a BBSRC studentship, and J.S. by a Lloyds Tercentenary Foundation Fellowship. The work was supported by a BBSRC grant awarded to C.R. and J.S.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONFUNDINGREFERENCES |
|---|
Alatalo RV, Mappes J. Tracking the evolution of warning signals. Nature (1996) 382:708–710.[CrossRef][Web of Science]
Atkins MD. Introduction to insect behavior (1980) New York: Macmillan Publishing Co.
Cott HB. Adaptive coloration in animals (1940) London: Methuen & Co Ltd.
de Jong PW, Holloway GJ, Brakefield PM, Vos H. Chemical defense in the ladybird beetles (Coccinellidae). II. Amount of reflex fluid, the alkaloid adaline and individual variation in defense in 2-spot ladybirds (Adalia bipunctata). Chemoecology (1991) 2:15–19.[CrossRef]
DeVries PJ. Differential wing toughness in distasteful and palatable butterflies: direct evidence supports unpalatable theory. Biotropica (2002) 34:176–181.[Web of Science]
Eisner T, Meinwald J. Defensive secretion of arthropods. Science (1966) 153:1341–1350.
Endler JA, Mappes J. Predator mixes and the conspicuousness of aposematic signals. Am Nat (2004) 163:532–547.[CrossRef][Web of Science][Medline]
Engen S, Järvi T, Wiklund C. The evolution of aposematic coloration by individual selection: a life-span survival model. Oikos (1986) 46:397–403.[CrossRef][Web of Science]
Evans DL. Tough, harmless cryptics could evolve into tough, nasty aposematics: an individual selectionist model. Oikos (1987) 48:114–115.[CrossRef][Web of Science]
Exnerova A, Landova E, Stys P, Fuchs R, Prokopova M, Cehlarikova P. Reactions of passerine birds to aposematic and nonaposematic firebugs (Pyrrhocoris apterus; Heteroptera). Biol J Linn Soc (2003) 78:517–525.[CrossRef][Web of Science]
Friedlander CP. The biology of insects (1976) London: Hutchinson & Co Ltd.
Gamberale G, Tullberg BS. Evidence for a more effective signal in aggregated aposematic prey. Anim Behav (1996) 52:597–601.[CrossRef][Web of Science]
Gamberale G, Tullberg BS. Aposematism and gregariousness: the combined effect of group size and coloration on signal repellence. Proc R Soc Lond B Biol Sci (1998) 265:889–894.[CrossRef]
Gamberale-Stille G, Guilford T. Automimicry destabilizes aposematism: predator sample-and-reject behaviour may provide a solution. Proc R Soc Lond B Biol Sci (2004) 271:2621–2625.[Medline]
Gamberale-Stille G, Tullberg BS. Fruit or aposematic insect? Context-dependent colour preferences in domestic chicks. Proc R Soc Lond B Biol Sci (2001) 268:2525–2529.[Medline]
Gittleman JL, Harvey PH. Why are distasteful prey not cryptic? Nature (1980) 286:149–150.[CrossRef][Web of Science]
Guilford T. Is kin selection involved in the evolution of warning coloration? Oikos (1985) 45:31–36.[CrossRef][Web of Science]
Guilford T. How do ‘warning colours’ work? Conspicuousness may reduce recognition errors in experienced predators. Anim Behav (1986) 34:286–288.[CrossRef][Web of Science]
Guilford T. The evolution of conspicuous coloration. Am Nat (1988) 131((Suppl)):S7–S21.[CrossRef]
Guilford T, Dawkins MS. Receiver psychology and the design of animal signals. Trends Neurosci (1993) 16:430–436.[CrossRef][Web of Science][Medline]
Halpin CG, Skelhorn J, Rowe C. Naïve predators and selection for rare conspicuous defended prey: the initial evolution of aposematism revisited. Anim Behav (2008) 75:771–781.[CrossRef][Web of Science]
Härlin C, Härlin M. Towards a historization of aposematism. Evol Ecol (2003) 17:197–212.[CrossRef]
Jetz W, Rowe C, Guilford T. Non-warning odors trigger innate color aversions—as long as they are novel. Behav Ecol (2001) 12:134–139.
Leimar O, Enquist M, Silllen-Tullberg B. Evolutionary stability of aposematic coloration and prey unprofitability: a theoretical analysis. Am Nat (1986) 128:469–490.[CrossRef][Web of Science]
Lindström L. Experimental approaches to studying the initial evolution of conspicuous aposematic signalling. Evol Ecol (1999) 13:605–618.[CrossRef]
Lindström L, Rowe C, Guilford T. Pyrazine odour makes visually conspicuous prey aversive. Proc R Soc Lond B Biol Sci (2001) 268:159–162.[Medline]
Marples NM, Kelly DJ, Thomas RJ. Perspective: the evolution of warning coloration is not paradoxical. Evol Int J Org Evol (2005) 59:933–940.
Marples NM, Roper TJ. Response of domestic chicks to methyl anthranilate odour. Anim Behav (1997) 53:1263–1270.[CrossRef][Web of Science][Medline]
Pasteels JA, Gregoire JC, Rowell-Rahter M. The chemical ecology of defence in arthropods. Annu Rev Entomol (1983) 28:263–289.[CrossRef][Web of Science]
Prior H, Wilzeck C. Selective feeding in birds depends on combined processing in the left and right brain hemisphere. Neuropsychologia (2008) 46:233–240.[CrossRef][Web of Science][Medline]
Raine NE, Chittka L. The adaptive significance of sensory bias in a foraging context: floral colour preferences in the bumblebee. Bombus terrestris. PLoS ONE (2007) 2:e556.[CrossRef]
Riipi M, Alatalo RV, Lindstrom L, Mappes J. Multiple benefits of gregariousness cover detectability costs in aposematic aggregations. Nature (2001) 413:512–51. 4.[CrossRef][Web of Science][Medline]
Roper TJ, Cook SE. Responses of chicks to brightly coloured insect prey. Behaviour (1989) 110:276–293.[CrossRef]
Roper TJ, Redston S. Conspicuousness of distasteful prey affects the strength and durability of one-trial avoidance learning. Anim Behav (1987) 35:739–747.[CrossRef][Web of Science]
Roper TJ, Wistow R. Aposematic coloration and avoidance learning in chicks. Q J Exp Psychol (1986) 38B:141–149.[Web of Science]
Rowe C, Guilford T. Hidden colour aversions in domestic chicks triggered by pyrazine odours of insect warning displays. Nature (1996) 383:520–522.[CrossRef][Web of Science]
Rowe C, Guilford T. The evolution of multimodal warning displays. Evol Ecol (1999) 13:655–671.[CrossRef]
Rowe C, Skelhorn J. Avian psychology and communication. Proc R Soc Lond B Biol Sci (2004) 271:1435–1442.[CrossRef][Medline]
Rowe C, Skelhorn J. Colour biases are a question of taste. Anim Behav (2005) 69:587–594.[CrossRef][Web of Science]
Ruxton GD, Sherratt TN, Speed M. Avoiding attack: the evolutionary ecology of crypsis, mimicry and aposematism. (2004) New York: Oxford University Press.
Schmidt V, Schaefer HM. Unlearned preference for red may facilitate recognition of palatable food in young omnivorous birds. Evol Ecol Res (2004) 6:919–925.
Schuler W, Roper TJ. Responses to warning coloration in avian predators. Adv Stud Behav (1992) 21:111–146.[CrossRef]
Sherratt TN. The coevolution of warning signals. Proc Biol Sci (2002) 269:741–746.
Sillen-Tullberg B. Higher survival of an aposematic than of a cryptic form of a distasteful bug. Oecologia (1985) 67:411–415.[CrossRef][Web of Science]
Skelhorn J, Rowe C. Avian predators taste-reject aposematic prey on the basis of their chemical defence. Biol Lett (2006a) 2:348–350.
Skelhorn J, Rowe C. Do the multiple defense chemicals of visually distinct species enhance predator learning? Behav Ecol (2006b) 17:947–951.
Skelhorn J, Rowe C. Prey palatability influences predator learning and memory. Anim Behav (2006c) 71:1111–1118.[CrossRef][Web of Science]
Skelhorn J, Rowe C. Taste-rejection by predators can explain the evolution of unpalatability. Behav Ecol Sociobiol (2006d) 60:550–555.[CrossRef][Web of Science]
Skelhorn J, Ruxton GD. Ecological factors influencing the evolution of insects' chemical defenses. Behav Ecol (2008) 19:146–153.
Speed MP, Ruxton GD. Aposematism: what should our starting point be? Proc Biol Sci (2005) 272:431–438.
Stiles FG. Taste preferences, color preferences, and flower choice in hummingbirds. Condor (1976) 78:10–26.
Tullberg BS, Hunter AF. Evolution of larval gregariousness in relation to repellent defences and warning coloration in tree-feeding Macrolepidoptera: a phylogenetic analysis based on independent contrasts. Biol J Linn Soc (1996) 57:253–276.[CrossRef][Web of Science]
Tullberg BS, Leimar O, Gamberale-Stille G. Did aggregation favour the initial evolution of warning coloration? A novel world revisited. Anim Behav (2000) 59:281–287.[CrossRef][Web of Science][Medline]
Tullberg BS, Merilaita S, Wiklund C. Aposematism and crypsis combined as a result of distance dependence: functional versatility of the colour pattern in the swallowtail butterfly larva. Proc Biol Sci (2005) 272:1315–1321.
Weiss MR, Papaj DR. Colour learning in two behavioural contexts: how much can a butterfly keep in mind? Anim Behav (2003) 65:425–434.[CrossRef][Web of Science]
Wiklund C, Järvi T. Survival of distasteful insects after being attacked by naive birds: a reappraisal of the theory of aposematic coloration evolving through individual selection. Evolution (1982) 36:998–1002.[CrossRef][Web of Science]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||