Butterfly visual characteristics and ontogeny of responses to butterflies by a specialized tropical bird

The responses of two adult and three hand-reared, naive young rufous-tailed jacamars (Galbula ruficauda) to local butterflies were studied in feeding experiments. Four behavioural characteristics distinguish jacamars from other less specialized avian predators: (1) Exposed to butterflies for the first time, naive young jacamars would attack butterflies without showing signs of inhibition. Unacceptable butterflies, once captured, were taste-rejected quickly, and most survived the sampling. The few presumably unacceptable butterflies consumed by the birds were not observed to cause vomiting. (2) After gaining some familiarity with butterflies, young birds, like the adults, developed a reluctance to attack. They visually rejected certain classes of butterflies, often failing to attack them during an entire four-hour feeding trial. However, occasional attacks were made on butterflies in these ‘rejected’ classes. When this did occur, the insects proved to be actually easier to catch than those that were more often attacked. Once captured, however, the majority of these butterflies were taste-rejected. (3) For a given butterfly species, most individuals were either consumed or rejected. Thus, each species could be clearly classified as either acceptable or unacceptable to the jacamars. This consistency in jacamar responses resulted in a bimodal acceptability distribution of sympatric butterflies. (4) Young jacamars were capable of rapid associative learning and their responses were closely associated with butterfly visual characteristics in which colour pattern, flight behaviour, and morphology were also closely correlated. Thus, a single butterfly morphological parameter termed body shape (body length/thoracic diameter ratio) can adequately predict the feeding responses of jacamars. Visually detectable traits associated with butterflies possessing chemical defences may represent a balance between the need to signal unambiguously to specialized and/or experienced predators and the need to escape attacks by generalized and/or opportunistic predators. Since the proportion of specialized predators is higher in the tropical rainforest than in other habitat types, we expect greater divergence of morphological and behavioural characteristics between palatable and unpalatable butterflies in rainforest habitats.     

Are European White Butterflies Aposematic?

It has been suggested that the white coloration of Pieridae butterflies is a warning signal and therefore all white Pieridae could profit from a mimetic resemblance. We tested whether green-veined white (Pieris napi) and orange-tip (Anthocharis cardamines) butterflies benefit from white coloration. We compared their relative acceptability to wild, adult pied flycatchers (Ficedula hypoleuca) by offering live A. cardamines and P. napi together with two non-aposematic butterflies on the tray attached to birds' nesting boxes. Experienced predators equally attacked white and non-white butterflies, and the order of attack among the Pieridae was random. If anything, there was a slight indication that the female A. cardamines was the least favoured prey. Since birds did not avoid white coloration, we compared the palatability of these two species against known palatable and unpalatable butterflies by presenting them to great tits (Parus major). Pieris brassicae, which has been earlier described as unpalatable, was also included in the palatability test. However, there were no significant differences in the palatability of the butterflies to birds, and even P. brassicae was apparently palatable to the great tits. Our results do not unambiguously support the hypothesis that the white coloration of Pieridae would signal unpalatability. Nevertheless, in our last experiment, pied flycatchers often rejected or left untouched free flying P. napi and A. cardamines. This suggests that other features in a more natural situation, such as the agile flight pattern or odours might still make them unprofitable to birds. This revised version was published online in July 2006 with corrections to the Cover Date.     

Palatablility and escaping ability in Neotropical butterflies: tests with wild kingbirds (Tyrannus melancholicus, Tyrannidae)

The palatability and the ability of neotropical butterflies to escape after being detected, attacked and captured by wild kingbirds ( Tyrannus melancholicus Vieillot), was investigated by the release of 668 individuals of 98 butterfly species close to the birds, during their usual feeding activities. Most of the butterflies were attacked and eaten. Only the troidine swallowtails ( Parities and Battus ; Papilionidae) were consistently rejected on taste and elicited aversive behaviours in birds. Most other aposematic and/or mimetic species in the gehera Danaus and Lycorea (Danainae), Dione, Eueides and Heliconius (Heliconiinae), Hypothyris, Mechanitis and Melinaea (Ithomfinae), Biblis, Callicore and Diaethria (Limenitidinae) were generally eaten. Cryptic and non-mimetic species were always attacked and, if captured, they were also eaten. All Apaturinae, Charaxinae, Nymphalinae, Hesperidae, most Limenitidinae, Heliconiinag ( Agraulis, Dryas, Dryadula and Philaethria ) and Papilionidae ( Eurytides, Heraclides and Protesilaus ) were in this group. Results indicate that the learning process in kingbirds may demand a large mortality in prey populations, even among species generally accepted as unpalatable and aposematic. They also support the assertion that escaping ability and unpalatability evolved in butterflies as alternative strategies to avoid predation by birds. Mimetic relationships among several species are discussed. Evidence for the evolution of aposematism not related to unpalatability, but to escaping ability, was found for two hard-to-catch Morpho species.     

The responses of wild jacamars (Galbula ruficauda,Galbulidae) to aposematic,aposematic and cryptic,and cryptic butterflies in central Brazil

1. This article reports the responses of wild, adult jacamars to butterflies with distinct coloration types in central Brazil. Fully aposematic species, i.e. those exhibiting bright and/or contrasting colours on both wing surfaces (= A/A), were predominantly sight‐rejected by birds and, with one exception, the few butterflies attacked and captured were taste‐rejected afterwards. 2. Aposematic and cryptic butterflies, i.e. those exhibiting bright and/or contrasting colours on the upper and cryptic colours on the underwings (= A/C) were sight‐rejected while flying, when they show their conspicuous colours to predators. This suggests that birds associate butterfly colours with their difficulty of capture, as in the case of Morpho and several Coliadinae species. These butterflies, however, were heavily attacked at rest, when they are cryptic. 3, Fully cryptic butterflies, i.e. those exhibiting cryptic colours on both wing surfaces (= C/C) did not elicit sight rejections by birds. Comparisons involving the number of attacks and the capture success of flying and resting individuals showed no significant differences in species more frequently observed like some cracker butterflies (Hamadryas feronia and H. februa) and Taygetis laches. Compared with the A/C Coliadinae, these butterflies showed a lesser, although not significantly different, ability to escape while flying, but a greater and significantly different ability to escape while at rest. 4, A hunting tactic of jacamars, which consists of following flying A/C and C/C butterflies on sight, and waiting until they perch to locate and attack them, is described for the first time.     

Predation,thermoregulation, and wing color in pierid butterflies

Summary This paper explores two hypotheses about the relationships among predation, thermoregulation, and wing color in butterflies: First, that butterflies are susceptible to predation during thermally marginal periods (e.g., cool weather) when effective thermoregulation and flight are not possible; second, that Pieris butterflies are relatively unpalatable to visual predators, supporting the idea that the white wing pigment of Pieris represents aposematic coloration. Field experiments with Pieris and Colias in 1984 and 1985 demonstrate that substantial predation may occur during the morning period before butterflies are able to actively fly. Circumstantial evidence is presented to suggest that at least some of the predation is by small, cursorial mammals. Feeding experiments in the field using Grey Jays as predators indicate that Pieris napi and P. occidentalis are less palatable than other sympatric butterflies, including confamial Colias alexandra. These and previous results suggest that Pieris are edible but less preferred as prey by birds, and that the degree of palatibility may vary among Pieris species. The relatively low palatability of these Pieris is consistent with the hypothesis that their white pigmentation represents aposematic coloration; however, the cues by which potential bird predators might discriminate against Pieris have not been established.     

Sexual and seasonal differences in the frequency of beak marks on the wings of two Lethe butterflies

To assess bird predation pressure on butterflies, I investigated beak marks on the wings of two Lethe butterflies for 3 years in secondary temperate forests. If bird predation had significant effects on average longevity of butterflies, and if the number of specimens preyed upon was proportionate to the number of beak-marked specimens, the beak mark frequency would be negatively correlated with average longevity of a butterfly. Bird predation pressure is generally thought to influence average longevity of butterflies. Therefore, if there is a negative correlation between beak mark frequency and average longevity, bird predation pressure would be reflected in beak mark frequency. Beak mark frequency was negatively correlated with longevity in Lethe diana (Butler), the more abundant of the two species; thus, the beak mark frequency was considered to be a suitable index of bird predation pressure on the butterflies investigated in this study. In both Lethe species, beak mark frequency was higher in females than in males. Because female butterflies have a relatively smaller thorax and flight muscles and a larger abdomen that contains eggs, they are presumably weaker or less agile fliers than males, and are probably attacked more easily by birds. In autumn, butterflies were heavily attacked by birds irrespective of sex and species. Because the numbers of lepidopteran larvae, which are the preferred prey of many birds, decreased in autumn, birds were thought to shift their diets to alternative prey such as adult butterflies.     

BIRD PREDATION AS A SELECTIVE AGENT IN A BUTTERFLY POPULATION

In a population of the checkerspot butterfly, Euphydryas chalcedona, the detached wings of 309 individuals that had been attacked and eaten by birds were collected during a single flight season. During this time period a representative sample of 296 live butterflies in this population was photographed. Comparison of sex ratio and coloration of those butterflies that had been attacked with those that had not showed, first, that birds attacked slightly more females than males; and second, that among males, which are extremely variable in the amount of red on the forewing, birds attacked the less red individuals.     

Low Ability of Great Tits to Discriminate Similarly Inconspicuous Edible and Inedible Prey

Numerous studies have shown warning coloration to facilitate the discrimination of edible and inedible prey. However, inedible insect species may possess cryptic coloration as well. It has been shown that some other visual features (especially characteristic body shape) are sufficient for the recognition of some insect taxa (e.g. ladybirds, ants, wasps). We tested the ability of wild‐caught great tits (Parus major) to discriminate between the identically coloured edible (roach – Blaptica dubia) and the inedible (firebug – Pyrrhocoris apterus) as prey according to subtle peripheral visual traits (shape of legs and antennae, body posture, and means of locomotion). Both prey species were offered either simultaneously or alternately. The ability of the birds to learn was tested by means of fourteen trial repetitions in two sessions. In general, great tits were not able to learn to discriminate between firebugs and roaches by subtle shape cues alone during the two sessions. However, multivariate analysis of individual bird behaviour showed that they adopted one of three different attitudes to the presented prey. Most of the birds never attacked any or always attacked both prey. In addition, a small proportion of the birds was able to discriminate between the two prey types and attacked only roaches. Nevertheless, firebugs survived most of the attacks, which suggests that in case of chemically protected prey, the evolution of conspicuous coloration is not always the best/only option.     

The conceptual and practical implications of interpreting diet breadth mechanistically in generalist predatory insects

Seasonal polyphenism in animal colour patterns indicates that temporal variation in selection pressures maintains phenotypic plasticity. Spring generation of the polyphenic European map butterfly Araschnia levana has an orange–black fritillary‐like pattern whilst individuals of the summer generation are black with white bands across the wings. What selects for the colour difference is unknown. Because predation is a major selection pressure for insect coloration, we first tested whether map butterfly coloration could have a warning function (i.e. whether the butterflies are unpalatable to birds). In a following field experiment with butterfly dummies we tested whether the spring form is better protected than the summer form from predators in the spring, and vice versa in the summer. The butterflies were palatable to birds (blue tits Cyanistes caeruleus) and in the field the spring and summer form dummies were attacked equally irrespective of season. Therefore, we found no evidence that the map butterfly is warning‐coloured or that seasonal polyphenism is an adaptation to avian predation. Because insect coloration has multiple functions and map butterfly coloration is linked to morphology, life history and development it is likely that the interplay of several selection pressures explains the evolution of colour polyphenism. © 2012 The Linnean Society of London, Biological Journal of the Linnean Society, 2012, ?? , ??–??.     

Cryptic differences in colour among Müllerian mimics: how can the visual capacities of predators and prey shape the evolution of wing colours?

Antagonistic interactions between predators and prey often lead to co‐evolution. In the case of toxic prey, aposematic colours act as warning signals for predators and play a protective role. Evolutionary convergence in colour patterns among toxic prey evolves due to positive density‐dependent selection and the benefits of mutual resemblance in spreading the mortality cost of educating predators over a larger prey assemblage. Comimetic species evolve highly similar colour patterns, but such convergence may interfere with intraspecific signalling and recognition in the prey community, especially for species involved in polymorphic mimicry. Using spectrophotometry measures, we investigated the variation in wing coloration among comimetic butterflies from distantly related lineages. We focused on seven morphs of the polymorphic species Heliconius numata and the seven corresponding comimetic species from the genus Melinaea. Significant differences in the yellow, orange and black patches of the wing were detected between genera. Perceptions of these cryptic differences by bird and butterfly observers were then estimated using models of animal vision based on physiological data. Our results showed that the most strikingly perceived differences were obtained for the contrast of yellow against a black background. The capacity to discriminate between comimetic genera based on this colour contrast was also evaluated to be higher for butterflies than for birds, suggesting that this variation in colour, likely undetectable to birds, might be used by butterflies for distinguishing mating partners without losing the benefits of mimicry. The evolution of wing colour in mimetic butterflies might thus be shaped by the opposite selective pressures exerted by predation and species recognition.     

Warning Signal Efficacy: Assessing the Effects of Color,Iridescence, and Time of Day in the Field

Warning coloration deters predators from attacking distasteful or toxic prey. Signal features that influence warning color effectiveness are not well understood, and in particular, we know very little about how effective short‐wavelength and iridescent colors are as warning color elements in nature and how warning signal effectiveness changes throughout the day. We tested the effect of these factors on predation risk in nature using specimens of the distasteful pipevine swallowtail butterfly, Battus philenor. B. philenor adults display both iridescent blue and diffusely reflecting orange components in their warning signal. We painted B. philenor wings to create five different model types: all‐black, only‐iridescent‐blue, only‐orange, iridescent‐blue‐and‐orange (intact signal), and matte‐blue‐and‐orange. We placed 25 models in each of 14 replicate field sites for 72 h and checked for attacks three times each day. Model type affected the likelihood of attack; only‐orange models were, the only model attacked significantly less than the all‐black model. Iridescence did not enhance or decrease warning signal effectiveness in our experiment because matte‐blue‐and‐orange models were attacked at the same rate as iridescent‐blue‐and‐orange models. Time of day did not differentially affect model type. Video recordings of attacks revealed that insectivorous birds were responsible. The results of this experiment, when taken with previous work, indicate that the response to blue warning coloration is likely dependent on predator experience and context, but that iridescence per se does not affect warning signals in a natural context.     

Avoidance of an aposematically coloured butterfly by wild birds in a tropical forest

1. Birds are considered to be the primary selective agents for warning colouration in butterflies, and select for aposematic mimicry by learning to avoid brightly coloured prey after unpleasant experiences. It has long been thought that bright colouration plays an important role in promoting the avoidance of distasteful prey by birds. 2. The hypothesis that warning colouration facilitates memorability and promotes predator avoidance was tested by means of a field experiment using distasteful model butterflies. Artificial butterflies with a Heliconius colour pattern unknown to local birds were generated using bird vision models, either coloured or achromatic, and hung in tree branches in a tropical forest. Two sequential trials were conducted at each site to test avoidance by naïve and experienced predators. 3. There was a significant reduction in predation in the second trial. Also, coloured models were attacked less than achromatic models. Specifically, coloured butterflies were attacked significantly less in the second trial, but there was no significant decrease in predation on achromatic models. 4. The present results imply an important role for colour in enhancing aversion of aposematic butterflies. It has also been demonstrated that previous experience of distasteful prey can lead to enhanced avoidance in subsequent trials, supporting mimicry theory.     

Species‐dependent microarchitectural traits of iridescent scales in the triad taxa of Ornithoptera birdwing butterflies

Ornithoptera birdwing butterflies have blue, green, or orange iridescent scales in different species or subspecies. To understand the species‐ or subspecies‐dependent scale color differences, we performed comparative morphometric analyses of iridescent scales from three closely related taxa: O. priamus priamus (green), O. priamus urvillianus (blue), and O. croesus (orange). The three types of Ornithoptera wings exhibited reversible color changes to longer wavelengths with different kinetics upon immersion in methanol, suggesting that their color differences are at least partly based on differences in the size of air cavities made by nanostructures. Cover scales of all three color types were visually semi‐transparent glass scales that exhibited color when placed on a dark background. The dorsoventral differences in coloration were observed in single scales, suggesting the optical importance of scale surfaces. Scanning electron microscopy of cover scales in cross section revealed that all color types exhibited finely sculpted tapered ridges and thick, irregular basal multilayers containing tandemly clustered granular objects and air cavities. Scale thickness, ridge height, and multilayer thickness were significantly different among the three color types, and granular object size was significantly different between orange scales and blue and green scales. We conclude that each of the three taxa of Ornithoptera butterflies possesses unique quantitative size values on tapered ridges and irregular multilayers with granular objects and air cavities to express unique structural color. These species‐ or subspecies‐dependent structural colors might have evolved via quantitative shifts in these microarchitectural traits rather than via changes in the basic developmental or architectural plan for color expression.     

The significance of wing pattern diversity in the Lycaenidae: mate discrimination by two recently diverged species

Closely related species of lycaenid butterflies are determinable, in part, by subtle differences in wing pattern. We found that female wing patterns can act as an effective mate‐recognition signal in some populations of two recently diverged species. In field experiments, we observed that males from a Lycaeides idas population and an alpine population of L. melissa preferentially initiate courtship with conspecific females. A morphometric study indicated that at least two wing pattern elements were important for distinguishing the two species: hindwing spots and orange crescent‐shaped pattern elements called aurorae. We deceived male L. idas into initiating courtship with computer generated paper models of heterospecific females when these pattern elements were manipulated, indicating that the wing pattern elements that define the diversity of this group can be effective mate recognition signals.     

Chromosome‐level reference genome assembly and gene editing of the dead‐leaf butterfly Kallima inachus

The leaf resemblance of Kallima (Nymphalidae) butterflies is an important ecological adaptive mechanism that increases their survival. However, the genetic mechanism underlying ecological adaptation remains unclear owing to a dearth of genomic information. Here, we determined the karyotype (n = 31) of the dead‐leaf butterfly Kallima inachus, and generated a high‐quality, chromosome‐level assembly (568.92 Mb; contig N50: 19.20 Mb). We also identified candidate Z and W chromosomes. To our knowledge, this is the first study to report on these aspects of this species. In the assembled genome, 15,309 protein‐coding genes and 49.86% repeat elements were annotated. Phylogenetic analysis showed that K. inachus diverged from Melitaea cinxia (no leaf resemblance), both of which are in Nymphalinae, around 40 million years ago. Demographic analysis indicated that the effective population size of K. inachus decreased during the last interglacial period in the Pleistocene. The wings of adults with the pigmentary gene ebony knocked out using CRISPR/Cas9 showed phenotypes in which the orange dorsal region and entire ventral surface darkened, suggesting its vital role in the ecological adaption of dead‐leaf butterflies. Our results provide important genome resources for investigating the genetic mechanism underlying protective resemblance in dead‐leaf butterflies and insights into the molecular basis of protective coloration.     

The role of size and colour pattern in protection of developmental stages of the red firebug (Pyrrhocoris apterus) against avian predators

We investigated how predator/prey body‐size ratio and prey colour pattern affected efficacy of prey warning signals. We used great and blue tits (Parus major and Cyanistes caeruleus), comprising closely related and ecologically similar bird species differing in body size, as experimental predators. Two larval instars and adults of the unpalatable red firebug (Pyrrhocoris apterus), differing in body size and/or coloration, were used as prey. We showed that prey body size did not influence whether a predator attacked the prey or not during the first encounter. However, smaller prey were attacked, killed, and eaten more frequently in repetitive encounters. We assumed that body size influences the predator through the amount of repellent chemicals better than through the amount of optical warning signal. The larger predator attacked, killed and ate all forms of firebug more often than the smaller one. The difference between both predators was more pronounced in less protected forms of firebug (chemically as well as optically). Colour pattern also substantially affected the willingness of predators to attack the prey. Larval red–black coloration did not provide a full‐value warning signal, although a similarly conspicuous red‐black coloration of the adults reliably protected them. © 2010 The Linnean Society of London, Biological Journal of the Linnean Society, 2010, 100 , 890–898.     

Benefit by contrast: an experiment with live aposematic prey

Aposematic prey often have a coloration that contrasts withthe background. One beneficial effect of such conspicuous colorationis that it produces faster and more durable avoidance by predators.Another suggested benefit is that prey that contrast with thebackground are more quickly discerned and recognized as unpalatableby experienced predators. To further investigate the effectsof prey contrast on predator behavior, I conducted an experimentwith young chicks (Gallus gallus domesticus) as predators onlive aposematic and nonaposematic prey. Birds with prior experienceof both prey types were allowed into an arena with both palatableprey and aposematic prey on backgrounds that either closelymatched or contrasted with the coloration of the aposematicprey. Also, the time a bird had available to decide to attacka prey was manipulated by including a competing chick or not.The experienced birds showed greater attack latencies for aposematicprey on more contrasting backgrounds, and aposematic prey werealso attacked to a greater extent when on a matching background.The presence of a competitor generated similar effects, wherebirds in high competition attacked more and faster comparedto birds subjected to lower degree of competition, but therewas no interaction between competition and contrast. Thus,the experiment provides evidence that prey contrast againstthe background may produce better recognition and avoidance,independently of predator viewing time.     

The deflection hypothesis: eyespots on the margins of butterfly wings do not influence predation by lizards

Feeding experiments with lizards are used to examine the function of small eyespot markings found along the wing margins of many butterfly species. Such eyespots are frequently suggested to function by deflecting the attacks of vertebrate predators away from the vulnerable body towards the wing margins, which can tear easily; the eyespots are considered to mislead predators and to act as targets for their attacks. Such misdirected attacks give the butterflies a chance to evade capture, albeit sometimes losing pieces of wing tissue. As a model prey species, we used fruit-feeding individuals of the tropical butterfly Bicyclus anynana that were attacked in a standard way in laboratory cages by the anolis lizard, Anolis carolinensis . We also manipulated the butterflies' wing patterns by pasting eyespots on different parts of the wings to examine the deflection hypothesis in more detail. Our results indicate no influence on the lizard attacks either of the presence of eyespots, or of their position on the wings. The lizards attacked butterflies in a highly stereotyped manner both when the prey were presented on matching or on contrasting backgrounds. We thus found no support for the deflection hypothesis for attacks by insectivorous lizards. Indeed, our only support to date has been obtained for naïve flycatcher birds, but even this requires further corroboration. Although effective deflection may occur rather infrequently, except perhaps under certain ecological conditions such as high-density feeding of butterflies on fallen fruit, it may still be sufficiently consistent over time to have contributed to shaping the evolution of marginal eyespot patterns.  © 2007 The Linnean Society of London, Biological Journal of the Linnean Society , 2007, 92 , 661–667.     

Wing coloration and reflectance in Morpho butterflies as related to reproductive behavior and escape from avian predators

Summary Different species of large tropical butterflies belonging to the genus Morpho vary dramatically in both the amount of blue color on their wings and the associated irridescence (reflectance). This paper discusses how such a morphological properties may be related to both courtship behavior and effective means of reducing predation (especially by birds) in the low density adult populations. Essentially, the hypothesis is advanced that territorial species of Morpho can afford to possess very conspicuous wing coloration that may facilitate courtship interactions, in addition to spacing the territorial male population over the suitable habitat. While territoriality may be favored by natural selection, such behavior can only evolve if the species involved possess effective means of reducing predation, because territoriality in morphos is an extremely predictable form of behavior, toward which predators can easily orient. Two alternate hypotheses are advanced to account for low predation in a territorial morpho, Morpho amathonte, a species in which males are very bright and showy. The first hypothesis, which is more consistent with traditional ideas on the function of bright colors in morpho wings, maintains that predators learn quickly to avoid these butterflies as prey, since they are very difficult to catch. The second hypothesis, suggests that conspicuous territorial male morphos actually employ pursuit-stimuli to invite birds to attack and be subsequently unsuccessful.     

INTERSEXUAL COMPARISON OF MIMETIC PROTECTION IN THE BLACK SWALLOWTAIL BUTTERFLY,PAPILIO POLYXENES: EXPERIMENTS WITH CAPTIVE BLUE JAY PREDATORS

The black swallowtail butterfly (Papilio polyxenes asterius Stoll), is commonly assumed to exhibit female-limited Batesian mimicry of the aposematic pipevine swallowtail (Battus philenor [L.]), since the dorsal wing surfaces of P. polyxenes females, but not males, resemble those of the model. However, the ventral wing surface is monomorphic and closely resembles that of the model in both sexes. Thus both sexes of P. polyxenes should benefit from mimicry during periods of ventral surface exposure, such as during overnight roosting and other times of high predatory risk. Eight blue jays (Cyanocitta cristata L.) were offered ventrally and dorsally exposed butterfly prey items in an outdoor aviary. Model-conditioned birds refused male and female P. polyxenes equally when the butterflies were presented ventrally. However, significantly more males than females were attacked when the dimorphic dorsum was visible. Both sexes are thus similarly protected when the ventral wing surface is displayed during roosting. The high degree of bird-to-bird variability in response to P. polyxenes mimics suggests that there is a spectrum in ability or willingness of predators to discriminate among mimics of varying similarity to the model. Sexual dimorphism of the dorsal surface of P. polyxenes wings may reflect sexual selection favoring males that are recognizable as satisfactory mates or intrasexual competitors.