Evolution in reverse: clean air and the peppered moth

Between 1848 and 1895 the melanic form f carbonaria of the peppered moth increased in Manchester from 0 to 98%. The reverse process is now occurring in one locality in Memyside where a year-to-year survey has shown that the pale form f. typica has increased from 6 to 30% between 1959 and 1984. Supporting information also comes from two studies in N Wales. The Clean Air Acts and the consequent decline in air pollution levels are probably responsible but there are many ill–understood problems, not the least being how, and the extent to which, non-visual selection operates.     

A decline of melanism in the peppered moth Biston betularia in The Netherlands

Melanic forms of the peppered moth Biston betularia were well established in The Netherlands by the end of the 19th century, indeed the first records of the black carbonaria form in 1867 are only about 20 years later than in England. Analysis of extensive sampling data collected by B. J. Lempke for a period of several years beginning in 1969 shows that carbonaria was at a frequency of about 60 to 70% in most of the country where epiphyte communities on trees were reduced due to the effects of air pollution. The pale typica and the three intermediate insularia forms were each at similar, low frequencies. Only in the extreme north and south-east of The Netherlands where epiphyte floras were richer was carbonaria at a lower frequency of less than 40%. Samples collected from seven localities in 1988 show that carbonaria has dramatically declined to a frequency of less than 10%. In contrast to England, the fully black form is being replaced not only by typica but also by the darkest of the insularia phenotypes. The decline in melanism coincides with a period of decreasing levels of sulphur dioxide and of increasing species diversity of lichens on trees.     

Background selection by the peppered moth (Biston betularia Linn.): individual differences

The hypothesis that dimorphically coloured, cryptic moths select appropriate rest sites by comparing their body scales to substrate reflectance was tested using typical and melanic morphs of the peppered moth, Biston betularia (L.). Experiments designed to block the individual's inspection of its inherited colour phenotype do not support Kettlewell's contrast/conflict (self-inspection) hypothesis. Instead, tracking of marked moths over successive days revealed individual differences in rest-site selection which were not related to treatments, experience (imprinting), nor closely to a moth's inherited colour pattern. Differences between family broods indicate that some genetic bias in background selection exists. The production of artificially selected lines with consistent but opposing preferences will allow us to investigate the co-evolution of pattern and behaviour.     

A bird’s eye view of the peppered moth

Industrial melanism in Biston betularia is one of the best known examples of the role of natural selection in evolution and has received considerable scrutiny for many years. The rise in frequency of the dark form of the moth (carbonaria) and a decrease in the pale form (typica) was the result of differential predation by birds, the melanic form being more cyptic than typica in industrial areas where the tree bark was darkened by air pollution. One important aspect of early work evaluating the relative crypsis of the forms of B. betularia on tree trunks with different lichen flora was the reliance on human observers. Humans, however, do not have the same visual capabilities as birds. Birds have well‐developed ultraviolet (UV) vision, an important component of their colour processing system that affects many aspects of behaviour, including prey detection. We examined the UV characteristics of the two forms of B. betularia and a number of foliose and crustose lichens. In human visible light the speckled form typica appeared cyptic when seen against a background of foliose lichen, whereas the dark form carbonaria was conspicuous. Under UV light the situation was reversed. The foliose lichens absorbed UV and appeared dark as did carbonaria. Typica, however, reflected UV and was conspicuous. Against crustose lichens, typica was less visible than carbonaria in both visible and UV light. These findings are considered in relation to the distribution and recolonization of trees by lichens and the resting behaviour of B. betularia.     

Effect of forest fragmentation on bird species richness in Papua New Guinea

Tropical forests worldwide are being fragmented at a rapid rate, causing a tremendous loss of biodiversity. Determining the impacts of forest disturbance and fragmentation on tropical biotas is therefore a central goal of conservation biology. We focused on bird communities in the interior (>100 m from forest edge) of forest fragments (300, 600, and 1200 ha) in the lowlands of Papua New Guinea and compared them with those in continuous forest. We surveyed bird communities using point counts, mist‐netting, and random walks, and measured habitat and microclimate characteristics at each site. We also surveyed leaf‐dwelling arthropods, butterflies, and ants, and obtained diet samples from birds to examine food availability and food preferences. We recorded significantly fewer bird species per point in the 300‐ha forest fragment than in other study sites. Overall, we recorded 80, 84, and 88 species, respectively, in forest fragments, and 102 in continuous forest. Frugivores (especially large frugivores) and insectivores had lower species richness in forest fragments than continuous forest. Our results did not support the food scarcity hypothesis, that is, the decline of insectivorous birds in forest fragments is caused by an impoverished invertebrate prey base. We also found no significant differences among forest fragments and continuous forest in microclimates of forest interiors. Rather, we found that microhabitats preferred by sensitive birds (i.e., 30% of species with the strongest preferences for continuous forest) were less common in forest fragments (19%–31% of points) than in continuous forest (86% of points). Our results suggest that changes in microhabitats may make forest fragments unsuitable for sensitive species. However, limited dispersal capabilities could also make some species of birds less likely to disperse and occupy fragments. In addition, impoverished food resources, size of the forest fragment, or hunting pressure could contribute to the absence of large frugivorous birds in forest fragments. The forest fragments in our study, preserved as village‐based protected areas, were not large enough to sustain the bird communities found in continuous forest. However, because these fragments still contained numerous bird species, preservation of such areas can be an important component of management strategies to conserve rainforests and birds in Papua New Guinea.     

The understanding of industrial melanism in the peppered moth (Biston betularia) (Lepidoptera: Geometridae)

Melanic polymorphism in B. betularia has been extensively studied. Correlations between high melanic frequency and high levels of air pollution have been demonstrated. Kettlewell and others have shown that differential bird predation has an important effect on the maintenance of the polymorphism, and coefficients of visual selection have been obtained on the assumption that the moth habitually rests on tree trunks. Computer models based on these selective coefficients show that they are not sufficient accurately to explain observed melanic frequencies. Other non-visual selective factors and weak frequency-dependent selection have been invoked to improve fits. Analysis of the resting positions of moths recorded in the wild demonstrates that B. betularia does not usually rest in exposed positions on tree trunks, but rather rests on the underside of branches, on trunks in shaded positions just below major branch joints or on foliate twigs. The results of a pilot selection experiment, while agreeing qualitatively with Kettlewell's results, suggest that fitness estimates that assume trunk-resting are quantitively incorrect. The error is greatest for melanic moths in rural areas. It is suggested that visual selective coefficients based on a true assessment of the resting behaviour of the moths may considerably improve the fit between computer predictions and observed phenotype frequency distributions.     

Behavioural studies on the peppered moth Biston betularia and a discussion of the role of pollution and lichens in industrial melanism

We examined the resting behaviour during pairing and egg-laying of bred female Biston betularia (L.) after they were set free at wooded sites in Somerset and Cardiff. Moths were not released immediately on emergence but were held for three nights to overcome an initial period of flight activity. Females (N= 257) were placed on trunks, and on large and small branches of trees, especially oaks, over periods of several weeks in 1984, 1985 and 1986. Releases were not made within 50 m of any known survivor. One hundred and thirty-one pairings were observed. We also examined oviposition behaviour in a cage experiment using different types of branches. Our observations support Mikkola's earlier conclusion from cage experiments with male moths that the species rests predominantly on branches and shows an appropriately specialized resting attitude, demonstrated here in a series of photographs. Many moths will rest underneath, or on the side of, narrow branches in the canopy. Once released females had settled and paired they only moved quite short distances, but positioning and egg-laying were profoundly influenced by the presence of foliose lichens. Females of all phenotypes show a strong preference to oviposit beneath a thallus of foliose lichens. Their tendency to rest against or close to the lichen during the day could markedly influence their crypsis. In the absence of foliose lichens, eggs are laid in cracks in the bark. The survivorship of released females both in different pairing combinations and during their whole reproductive life-span is analysed. These preliminary data suggest that bird predation can give rise to differential mortality of different pairing combinations and of the phenotypes. This technique provides a powerful method of measuring the relative crypsis and differential mortality of phenotypes with living females in natural resting sites. The significance of our observations on resting behaviour is discussed with regard to the role of air pollution and epiphytes in industrial melanism. Particular attention is given to the causal mechanisms underlying recent declines in the frequency of carbonaria melanics in parts of Britain. One important factor is the new growth made by trees since the marked reduction in particulate air pollution in the 1960s. It is emphasized that we need to know much more about the interactions between pollution, epiphytes and resting backgrounds, especially in the canopy, before we can be confident of our understanding of the evolution of industrial melanism.     

Soft Shoulders Ahead: Spurious Signatures of Soft and Partial Selective Sweeps Result from Linked Hard Sweeps

Characterizing the nature of the adaptive process at the genetic level is a central goal for population genetics. In particular, we know little about the sources of adaptive substitution or about the number of adaptive variants currently segregating in nature. Historically, population geneticists have focused attention on the hard-sweep model of adaptation in which a de novo beneficial mutation arises and rapidly fixes in a population. Recently more attention has been given to soft-sweep models, in which alleles that were previously neutral, or nearly so, drift until such a time as the environment shifts and their selection coefficient changes to become beneficial. It remains an active and difficult problem, however, to tease apart the telltale signatures of hard vs. soft sweeps in genomic polymorphism data. Through extensive simulations of hard- and soft-sweep models, here we show that indeed the two might not be separable through the use of simple summary statistics. In particular, it seems that recombination in regions linked to, but distant from, sites of hard sweeps can create patterns of polymorphism that closely mirror what is expected to be found near soft sweeps. We find that a very similar situation arises when using haplotype-based statistics that are aimed at detecting partial or ongoing selective sweeps, such that it is difficult to distinguish the shoulder of a hard sweep from the center of a partial sweep. While knowing the location of the selected site mitigates this problem slightly, we show that stochasticity in signatures of natural selection will frequently cause the signal to reach its zenith far from this site and that this effect is more severe for soft sweeps; thus inferences of the target as well as the mode of positive selection may be inaccurate. In addition, both the time since a sweep ends and biologically realistic levels of allelic gene conversion lead to errors in the classification and identification of selective sweeps. This general problem of “soft shoulders” underscores the difficulty in differentiating soft and partial sweeps from hard-sweep scenarios in molecular population genomics data. The soft-shoulder effect also implies that the more common hard sweeps have been in recent evolutionary history, the more prevalent spurious signatures of soft or partial sweeps may appear in some genome-wide scans.     

31° South: The physiology of adaptation to arid conditions in a passerine bird

Arid environments provide ideal ground for investigating the mechanisms of adaptive evolution. High temperatures and low water availability are relentless stressors for many endotherms, including birds; yet birds persist in deserts. While physiological adaptation probably involves metabolic phenotypes, the underlying mechanisms (plasticity, genetics) are largely uncharacterized. To explore this, we took an intraspecific approach that focused on a species that is resident over a mesic to arid gradient, the Karoo scrub‐robin (Cercotrichas coryphaeus). Specifically, we integrated environmental (climatic and primary productivity), physiological (metabolic rates: a measure of energy expenditure), genotypic (genetic variation underlying the machinery of energy production) and microbiome (involved in processing food from where energy is retrieved) data, to infer the mechanism of physiological adaptation. We that found the variation in energetic physiology phenotypes and gut microbiome composition are associated with environmental features as well as with variation in genes underlying energy metabolic pathways. Specifically, we identified a small list of candidate adaptive genes, some of them with known ties to relevant physiology phenotypes. Together our results suggest that selective pressures on energetic physiology mediated by genes related to energy homeostasis and possibly microbiota composition may facilitate adaptation to local conditions and provide an explanation to the high avian intraspecific divergence observed in harsh environments.     

Population genomics and the causes of local differentiation

Exactly 50 years ago, a revolution in empirical population genetics began with the introduction of methods for detecting allelic variation using protein electrophoresis (Throckmorton 1962; Hubby 1963; Lewontin & Hubby 1966). These pioneering scientists showed that populations are chock‐full of genetic variation. This variation was a surprise that required a re‐thinking of evolutionary genetic heuristics. Understanding the causes for the maintenance of this variation became and remains a major area of research. In the process of addressing the causes, this same group of scientists documented geographical genetic structure (Prakash et al. 1969), spawning the continued accumulation of what is now a huge case study catalogue of geographical differentiation (e.g. Loveless & Hamrick 1984; Linhart & Grant 1996). Geographical differentiation is clearly quite common. Yet, a truly general understanding of the patterns in and causes of spatial genetic structure across the genome remains elusive. To what extent is spatial structure driven by drift and phylogeography vs. geographical differences in environmental sources of selection? What proportion of the genome participates? A general understanding requires range‐wide data on spatial patterning of variation across the entire genome. In this issue of Molecular Ecology, Lasky et al. (2012) make important strides towards addressing these issues, taking advantage of three contemporary revolutions in evolutionary biology. Two are technological: high‐throughput sequencing and burgeoning computational power. One is cultural: open access to data from the community of scientists and especially data sets that result from large collaborative efforts. Together, these developments may at last put answers within reach.     

A theoretical analysis of the morph frequency variation in the peppered moth over England and Wales

The migration-selection model for the spatial and temporal variation of morph frequencies of Biston betularia over England and Wales (Cook & Mani, 1980) has been extended to include effects due to non-visual selection. The parameters for non-visual selection were chosen from the recent determination by Mani (1980) and by Creed et al. (1980). The morph frequencies over England and Wales were obtained through computer simulation and the results were compared with data along the Manchester-Yorkshire, Central Wales-East Anglia and South Wales-London transects. Best fits to the data were obtained by using the non-visual selective values of Mani for carbonaria and modified values of Creed et al. for insularia. It is concluded that the observed polymorphism could be well explained through a balance of migration and visual and non-visual selections.     

Importance of colour in the reaction of passerine predators to aposematic prey: experiments with mutants of Pyrrhocoris apterus (Heteroptera)

Persistent questions concerning the warning coloration of unpalatable insects address whether the bright aposematic colour itself or its combination with a species-specific dark pattern is the key factor in their protection against insectivorous birds, and how chromatic polymorphism originates and is maintained in aposematics. In the present study, these questions were tested experimentally, using the birds Parus major , Parus caeruleus , Erithacus rubecula , and Sylvia atricapilla as predators, and chromatically polymorphic firebug Pyrrhocoris apterus : red wild form, white, yellow, and orange mutants (all four of them with the same black melanin pattern, the mutants differing in colour of pteridine pigments only) and the nonaposematic brown-painted wild form as prey. The results show that a specific colour is essential for the birds to recognize the specific aposematic prey; the melanin pattern is not sufficient. White mutants were no better protected than nonaposematic firebugs; red wild-type and orange mutants were equally well protected against all bird species; and the reaction of birds to yellow mutants was species-specific. An evolutionary scenario of 'recurrent recessive mutations' is formulated to explain the origin of colour polymorphism in some aposematics.  © 2006 The Linnean Society of London, Biological Journal of the Linnean Society , 2006, 88 , 143–153.     

Effect of the MC1R gene on sexual dimorphism in melanin‐based colorations

Variants of the melanocortin‐1 receptor (MC1R) gene result in abrupt, naturally selected colour morphs. These genetic variants may differentially affect sexual dimorphism if one morph is naturally selected in the two sexes but another morph is naturally or sexually selected only in one of the two sexes (e.g. to confer camouflage in reproductive females or confer mating advantage in males). Therefore, the balance between natural and sexual selections can differ between MC1R variants, as suggest studies showing interspecific correlations between sexual dimorphism and the rate of nonsynonymous vs. synonymous amino acid substitutions at the MC1R. Surprisingly, how MC1R is related to within‐species sexual dimorphism, and thereby to sex‐specific selection, has not yet been investigated. We tackled this issue in the barn owl (Tyto alba), a species showing pronounced variation in the degree of reddish pheomelanin‐based coloration and in the number and size of black feather spots. We found that a valine (V)‐to‐isoleucine (I) substitution at position 126 explains up to 30% of the variation in the three melanin‐based colour traits and in feather melanin content. Interestingly, MC1R genotypes also differed in the degree of sexual colour dimorphism, with individuals homozygous for the II MC1R variant being 2 times redder and 2.5 times less sexually dimorphic than homozygous individuals for the VV MC1R variant. These findings support that MC1R interacts with the expression of sexual dimorphism and suggest that a gene with major phenotypic effects and weakly influenced by variation in body condition can participate in sex‐specific selection processes.     

Spatial variation in the evolutionary potential and constraints of basal metabolic rate and body mass in a wild bird

An organism's energy budget is strongly related to resource consumption, performance, and fitness. Hence, understanding the evolution of key energetic traits, such as basal metabolic rate (BMR), in natural populations is central for understanding life-history evolution and ecological processes. Here we used quantitative genetic analyses to study evolutionary potential of BMR in two insular populations of the house sparrow (Passer domesticus). We obtained measurements of BMR and body mass (M_b) from 911 house sparrows on the islands of Leka and Vega along the coast of Norway. These two populations were the source populations for translocations to create an additional third, admixed ‘common garden’ population in 2012. With the use of a novel genetic group animal model concomitant with a genetically determined pedigree, we differentiate genetic and environmental sources of variation, thereby providing insight into the effects of spatial population structure on evolutionary potential. We found that the evolutionary potential of BMR was similar in the two source populations, whereas the Vega population had a somewhat higher evolutionary potential of M_b than the Leka population. BMR was genetically correlated with M_b in both populations, and the conditional evolutionary potential of BMR (independent of body mass) was 41% (Leka) and 53% (Vega) lower than unconditional estimates. Overall, our results show that there is potential for BMR to evolve independently of M_b, but that selection on BMR and/or M_b may have different evolutionary consequences in different populations of the same species.     

城市化对鸟类筑巢的影响研究综述

繁殖期筑巢是鸟类生活史的重要阶段, 是鸟类繁殖成功的关键保障。全球范围的城市化加速推进, 使城市中筑巢繁殖的鸟类面临挑战的同时又为其提供了特殊机遇。本文通过搜索现有文献, 利用Biblioshiny程序提取并整合关键词, 得到城市鸟类生态研究的热点领域, 分析了城市气候、食物资源、巢址资源、捕食压力、污染和人为干扰对鸟类筑巢的影响, 并对今后城市化对鸟类筑巢影响研究中亟需解决的问题进行了展望。城市化对鸟类筑巢期、巢址选择及巢材选择产生显著影响, 与栖息在村镇生境的鸟类相比, 在城市中繁殖的鸟类在筑巢时间、空间和巢材使用上出现变化。在城市中筑巢的鸟类到底是通过调整行为策略作出积极适应, 还是被动接受城市环境中的诸多负面干扰, 需要针对研究对象和特定的城市栖息生境进行及时评估, 而非泛泛之谈。要在研究结论基础上, 深入思考并提出城市化建设过程中有利于鸟类种群和群落保护的具体措施, 高效发挥公民科学作用以解决城市中的鸟类保护实践问题。     

Prediction of bird community composition based on point-occurrence data and inferential algorithms: a valuable tool in biodiversity assessments

Local biological communities are made up of species, each of which has its own particular relationship with the environment. To the extent that these autecological niches limit species’ distributions, and by extension community composition, models of species’ ecological niches can predict species composition at particular sites, or at least provide a null hypothesis of potential species composition in the absence of effects of species interactions. We developed distributional predictions (ecological niche models) for 89 species occurring in dry tropical forest in the Balsas Basin of south‐western Mexico using an interpolation technique, and predicted the species likely to occur at 8 sites across the region. Onsite field inventory data were then used to test the community predictions, all of which were statistically significant. These results suggest that inventory efforts can be made more efficient by development beforehand of hypotheses that focus onsite collecting and inventory.     

Spatially variable coevolution between a haemosporidian parasite and the MHC of a widely distributed passerine

The environment shapes host–parasite interactions, but how environmental variation affects the diversity and composition of parasite‐defense genes of hosts is unresolved. In vertebrates, the highly variable major histocompatibility complex (MHC) gene family plays an essential role in the adaptive immune system by recognizing pathogen infection and initiating the cellular immune response. Investigating MHC‐parasite associations across heterogeneous landscapes may elucidate the role of spatially fluctuating selection in the maintenance of high levels of genetic variation at the MHC. We studied patterns of association between an avian haemosporidian blood parasite and the MHC of rufous‐collared sparrows (Zonotrichia capensis) that inhabit environments with widely varying haemosporidian infection prevalence in the Peruvian Andes. MHC diversity peaked in populations with high infection prevalence, although intra‐individual MHC diversity was not associated with infection status. MHC nucleotide and protein sequences associated with infection absence tended to be rare, consistent with negative frequency‐dependent selection. We found an MHC variant associated with a ~26% decrease in infection probability at middle elevations (1501–3100 m) where prevalence was highest. Several other variants were associated with a significant increase in infection probability in low haemosporidian prevalence environments, which can be interpreted as susceptibility or quantitative resistance. Our study highlights important challenges in understanding MHC evolution in natural systems, but may point to a role of negative frequency‐dependent selection and fluctuating spatial selection in the evolution of Z. capensis MHC.     

Predators and the breeding bird: behavioral and reproductive flexibility under the risk of predation

A growing body of work suggests that breeding birds have a significant capacity to assess and respond, over ecological time, to changes in the risk of predation to both themselves and their eggs or nestlings. This review investigates the nature of this flexibility in the face of predation from both behavioural and reproductive perspectives, and also explores several directions for future research. Most available work addresses different aspects of nest predation. A substantial change in breeding location is perhaps the best documented response to nest predation, but such changes are not always observed and not necessarily the best strategy. Changes in nesting microhabitat (to more concealed locations) following predation are known to occur. Surprisingly little work addresses the proactive avoidance of areas with many nest predators, but such avoidance is probably widespread. Individual birds could conceivably adopt anti‐predator strategies based on the nest predators actually present in an area, but such effects have yet to be demonstrated. In fact, the ways in which birds assess the risk of nest predation is unclear. Nest defence in birds has historically received much attention, but little is known about how it interacts with other aspects of decision‐making by parents. Other studies concentrate on predation risk to adults. Some findings suggest that risk to adults themselves influences territory location, especially relative to raptor nests. An almost completely unexplored area concerns the sorts of social protection from predators that might exist during the breeding season. Flocking typical of the non‐breeding season appears unusual while breeding, but a mated pair may sometimes act as a “flock of two”. Opportunistic heterospecific sociality may exist, with heterospecific protector species associations more prevalent than currently appreciated. The dynamics of singing during the breeding season may also respond to variation in predation risk, but empirical research on this subject is limited. Furthermore, a few theoretical and empirical studies suggest that changes in predation risk also influence the behaviour of lekking males. The major influence of predators on avian life histories is undoubtedly expressed at a broad phylogenetic scale, but several studies hint at much flexibility on an ecological time scale. Some species may forgo breeding completely if the risk of nest predation is too high, and a few studies document smaller clutch sizes in response to an increase in nest predation. Recent evidence suggests that a female may produce smaller eggs rather than smaller clutches following an increase in nest predation risk. Such an increase may also influence decisions about intraspecific brood parasitism. There are no clear examples of changes in clutch/egg size with changes in risk experienced by adults, but parental responses to predators have clear consequences for offspring fitness. Changes in risk to adults may also influence body mass changes across the breeding season, although research here is sparse. The topics highlighted herein are all in need more empirical attention, and more experimental field work whenever feasible.     

The Park Grass Experiment and next‐generation approaches: local adaptation of sweet vernal grass revisited

Long‐term ecological experiments provide unique opportunities to observe the effects of natural selection. The Park Grass Experiment at Rothamsted Experiment Station in Hertfordshire, UK, is the longest running ecological experiment that incorporates fertilization treatments and has been ongoing since 1856. In the 1970s, local adaptation was observed in the grass Anthoxanthum odoratum to the elevated soil aluminium levels of the fertilized plots. Gould et al. ( 2014 ) have utilized this system to reevaluate the extent of local adaptation, first documented nearly 45 years ago (Snaydon 1970 ), and to use emerging molecular approaches to identify candidate genes for the adaptation. From their work, they identify several plausible candidate loci for aluminium tolerance. This work shows the power of long‐term field‐based trials in a scientific age concentrated on rapidly emerging molecular techniques often utilized in short, narrowly focused laboratory or controlled environment experiments. The current study clearly illustrates the benefits gained by combining these molecular approaches within long‐term monitoring experiments that can be regularly revisited in a changing world and used to address questions on evolutionary scales.     

Melanin coloration has temperature‐dependent effects on breeding performance that may maintain phenotypic variation in a passerine bird

Fluctuating selection pressure may maintain phenotypic variation because of different types of individuals being adapted to different environmental conditions. We show that the extensive variation in the coloration of male pied flycatchers (Ficedula hypoleuca) can be maintained through differences in the reproductive success of male phenotypes under different conditions. The effects of weather conditions on the relative success of different male phenotypes varied between different phases of breeding. The reproductive output of black males was the highest when it was cold during egg‐laying but warm during the nestling period, whereas the fledgling production of brown males was highest when it was continuously warm. In addition, male forehead and wing patch sizes had context‐dependent effects on timing of breeding and nestling mortality, respectively. These results indicate that environmental heterogeneity plays a role in maintaining phenotypic variation. As melanin‐based coloration is heritable, climate change may alter phenotype frequencies depending on the patterns of warming.