SEARCH

SEARCH BY CITATION

Keywords:

  • vision;
  • brood parasitism;
  • calcium;
  • egg;
  • light environment;
  • maculation;
  • perception;
  • signalling

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. EGG COLORATION, AVIAN VISION, AND THE LIGHT ENVIRONMENT
  5. SELECTION FOR BLUE–GREEN CHROMA
  6. THE ROLE OF MACULATION
  7. CONCLUSIONS
  8. REFERENCES

Recent work suggests that the evolution of egg coloration may have been constrained in three important ways that have not yet been critically synthesized in any review. First, on account of birds being able to see in the ultraviolet spectrum, the interaction between the properties of avian vision and the light environment of nests imply different perceptions of egg coloration from those experienced by humans. Second, a new hypothesis to explain blue–green egg coloration interprets it as a sexually selected signal to males of the laying female's genetic quality. Third, evidence from taxa as divergent as sparrowhawks and great tits indicates that protoporphyrin pigments responsible for maculation (spotting patterns) have a structural function in compensating for eggshell thinning, as caused by calcium stress, and, more recently, dichlorodiphenyltrichloroethane. We consider this to be the most convincing explanation for the primary function of spotting, although an important secondary function might arise through the fact that individual patterns of maculation may allow birds to identify their own eggs, effectively serving as signatures in the face of inter- or intra-specific brood parasitism. These constraints or hypotheses are not mutually exclusive, and should not be taken to imply that one, but not other, agents of selection might apply to any one species. However, the sexually-selected eggshell coloration hypothesis is least plausible for hole-nesting birds because of the poor light quality available, although such species have been the focus of research in this area, and only a single experimental study has shown a link between egg coloration and male provisioning. Furthermore, the observed relationships between female phenotypic quality and egg traits do not necessarily imply that they have signalling functions. © 2010 The Linnean Society of London, Biological Journal of the Linnean Society, 2010, 100, 753–762.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. EGG COLORATION, AVIAN VISION, AND THE LIGHT ENVIRONMENT
  5. SELECTION FOR BLUE–GREEN CHROMA
  6. THE ROLE OF MACULATION
  7. CONCLUSIONS
  8. REFERENCES

Several hypotheses have been raised in an attempt to explain the huge variation in the appearance of avian eggs, both in background colour and in maculation (patterning): those posited include crypsis (Wallace, 1890; Lack, 1958); egg mimicry by brood parasites and rejection by present and past hosts (Newton, 1896); filtering solar radiation (Montevechi, 1976); and aposematism (Swynnerton, 1916). Underwood & Sealy's (2002) review explores the adaptive significance of egg coloration in terms of these hypotheses, and Kilner's (2006) study provides a comparative analysis showing how egg colour and patterning could have evolved from a presumed ancestral white and immaculate state, in response to various selective agents. Both reviews differ little in their conclusions, either from each other, or from the 19th Century ones of Wallace (1890) and Newton (1896), that crypsis and brood parasitism are the main selective agents determining eggshell pigmentation, although Underwood & Sealy (2002) prefer the latter explanation.

Pigmentation in avian eggs is derived from two primary sources. Brown pigmentation is attributed to protoporphyrin, whereas blue–green is derived from biliverdin (Kennedy & Vevers, 1976; Burley & Vadhera, 1989; Miksik et al., 1996), and the latter is present in only approximately half of avian species (Kennedy & Vevers, 1976). It has been suggested by Collias (1993) that these two ground colour pigments are under independent genetic control, with simultaneous expression of both in eggs resulting in olive coloration. Wang et al. (2009) compared the difference between the quality of eggshell pigments in blue-shelled eggs and brown-shelled eggs from the same population (Dongxiang in China) and analyzed the correlation between the quantity of protoporphyrin and biliverdin in both categories of eggshell. They found no difference between the total quantity of pigment in blue- and brown-shelled eggs but positive correlations between the quantity of protoporphyrin and biliverdin in blue and brown eggshells, suggesting that eggshell protoporphyrin and biliverdin are probably derived from the same precursor material.

Many birds lay eggs speckled with dark (usually red, brown, and black) protoporphyrin pigment spots or blotches (maculation), the intensity of which appears to be regulated by additional female-specific genes (Gosler, Barnett & Reynolds, 2000), although the genetic mechanisms controlling spotting may vary between species (Mahler et al., 2008). Avilés et al. (2007) argued that there may also be an environmental component involved in the expression of bird egg coloration. Working on a time-series of eggs collected over a period of 24 years and preserved in the Zoological Museum in Copenhagen, they found that reed warbler Acrocephalus scirpaceus eggs were brighter in years with higher rainfall, and tended to be bluer and greener in colder years. Common cuckoo Cuculus canorus eggs of the reed warbler gens, by contrast, were bluer and greener in wetter years. This may be explained by the finding of Mikhailov (1997) suggesting that, perhaps for phylogenetic reasons, cuckoo eggshell structure differs substantially from that of their hosts.

A body of recent work suggests that environmental factors constrain egg coloration in a number of important ways, and these constraints have not yet been critically synthesized in any review. We discuss these issues here.

EGG COLORATION, AVIAN VISION, AND THE LIGHT ENVIRONMENT

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. EGG COLORATION, AVIAN VISION, AND THE LIGHT ENVIRONMENT
  5. SELECTION FOR BLUE–GREEN CHROMA
  6. THE ROLE OF MACULATION
  7. CONCLUSIONS
  8. REFERENCES

Over the past 15 years, a growing body of work has investigated the ecological consequences of differences between avian and human vision. Cherry & Bennett (2001) have shown (using the red-chested cuckoo Cuculus solitarius and its hosts as an example) that avian perception of egg coloration (as estimated by spectrophotometric measurements across the bird-visible range) differs from that of humans, on account of birds being able to see in the ultraviolet (UV) spectrum in addition to the human-visible one. Consequently, brood parasite and host eggs which appear dissimilar to humans may appear similar to birds. More recently, Starling et al. (2006) have shown, using similar methodology, that cryptic gentes (female races) (previously not identified by humans) of cuckoo eggs exist. The implication of these findings is that the validity of experimental studies of egg coloration using artificial (as opposed to natural eggs) is drawn into question. If artificial eggs are used in experiments, then they should be measured spectrophotometrically and shown to have similar reflectance spectra to natural eggs.

Both of these studies are based on museum collections, although the importance of the light environment of the nest in the perception of egg coloration should be emphasized. Hewitson (1846) was the first to note that birds which nest in cavities tend to lay white eggs. Although aware that there was a systematic bias in this correlation caused by the fact that hole-nesting and laying of an unpigmented egg were both primitive conditions in birds (both being encountered most frequently amongst nonpasserine orders), Lack (1958) suggested that the laying of white eggs might be adaptive through their being more conspicuous in dimly-lit environments, such as deep holes. Taking this idea further, Vorobyev & Osorio (1998) developed a theoretical model of avian vision that predicts higher relative detectability of UV-reflective wavelengths at lower environmental luminosities.

More recently, this has been explored both comparatively and experimentally by Avilés, Soler & Perez-Contreras (2006), who first examined variation in egg coloration in 98 species of European passerines (measured using UV-visible reflectance spectrometry) in relation to nesting habits. Controlling for phylogeny, they found that eggs of hole-nesting species had higher UV reflectance than those nesting in open habitats. Second, they manipulated UV reflectance of experimental eggs which they introduced outside the nest-cup of the hole-nesting spotless starling Sturnus unicolor, and studied the recognition of these eggs. Ultraviolet-reflecting control eggs were more frequently recognized than non-UV reflecting experimental eggs. These results were not the result of experimental eggs being recognized by starlings as parasitic because, when parasitic eggs were detected, starlings recognized them as such and removed them. Thus, they are consistent with Lack's hypothesis.

Avilés (2008) compared the extent to which parasitic cuckoo eggs could be discriminated from eggs of two host species: the redstart Phoenicurus phoenicurus, which always nests in cavities; and the pied wagtail Motacilla alba, which can nest in crevices or holes. Using Vorobyev et al.'s (1998) discrimination model to simulate host retinal function and spectrophotometric measurements of museum egg collections from the south of Finland, he investigated both colour matching and the role of nest luminosity on host perception of matching between eggs of hosts and six different gentes of cuckoo eggs. Cuckoo eggs of the Phoenicurus gens showed better chromatic matching with redstart host eggs than did other cuckoo races and, in most cases, cannot be discriminated by hosts. However, under dim light conditions, achromatic differences between cuckoo and host eggs can be distinguished easily, whereas the proportion of cuckoo eggs discriminated by chromatic signals was only marginally affected. Thus, nest luminosity has far greater effects on achromatic than on chromatic matching, and influences host perception of matching between parasitic eggs and their own.

A similar approach has been applied to conspecific egg recognition by Cassey et al. (2008b), who demonstrated that perceptual modelling of avian visual discrimination can predict behavioural rejection responses to foreign eggs in the open nests of wild song thrushes Turdus philomelos. They used a photoreceptor noise-limited colour opponent model of visual perception (Kelber, Vorobyev & Osorio, 2003; Endler & Mielke, 2005) that estimates the difference between two colours with respect to the spectral sensitivities of all four avian single cone photoreceptors (ultraviolet, short wavelength, medium wavelength, and long wavelength), aiming to predict behavioural rates of experimental egg discrimination. Their visual modelling of experimental and natural egg coloration suggests that photon capture from the ultraviolet and short wavelength-sensitive cones elicits egg rejection decisions in song thrushes, with inter-clutch variation in egg coloration providing sufficient contrasts for detecting conspecific parasitism in this species.

Wallace (1890) suggested that the primary function of egg coloration was to provide crypsis to avoid predation, although the experimental evidence supporting this hypothesis has been equivocal (Underwood & Sealy, 2002). One possible reason for this is that the experimental protocols typically involve painting eggs and comparing predation rates on painted versus natural eggs. With but one exception, all the egg-predation experiments cited in their review use painted eggs. The exception, namely that of Bertram & Burger (1981) in their study of ostrich Struthio camelus eggs, used an artificial nest, admittedly only a simple scrape in leaves, but not a natural one. Westmoreland & Kiltie (2007) conducted experiments involving switching natural eggs in the nests of three species of Icterid blackbirds, and found that, in two cases, egg coloration did serve a cryptic function, although, in an earlier study (Westmoreland & Kiltie, 1996), they had found that intraspecific variation in clutch crypsis was unrelated to clutch survival in these same two species, and that in the third species (where no cryptic function could be shown), less-cryptic clutches actually had higher survival. Among species that build conspicuous nests, selection for egg crypsis should be reduced because visual predators, at least, detect nests prior to eggs. This hypothesis was tested by Westmoreland (2008) using nests of American robins Turdus migratorius, in which he replaced their eggs in sequential trials with those of three species that differed markedly in colour from the nests of red-winged blackbirds Agelaius phoeniceus, Brewer's blackbirds Euphagus cyanocephalus, and yellow-headed blackbirds Xanthocephalus xanthocephalus. The overall survival of the three clutch types was approximately equivalent but clutches of red-winged blackbird eggs, the most conspicuous egg type to humans, were discovered earlier by predators. Because the experimental design controlled for effects of nest characteristics, this difference in egg survival can be attributed to differences in eggshell pigmentation, suggesting that it plays a role in camouflage in conspicuous nests. Remarkably, to our knowledge, these studies are the only experiments undertaken to test this hypothesis using natural eggs and nests, and the hypothesis that egg coloration is primarily cryptic in function, which is often deemed to be obvious (at least for ground-nesting nonpasserines such as charadriiformes), remains weekly substantiated experimentally.

Hanley, Doucet & Dearborn (2010) pointed out that conspicuous eggs are taxonomically widespread, and have erected the blackmail hypothesis to explain this. It proposes that conspicuous egg coloration has the effect of coercing males into providing additional parental care under specific ecological conditions where there is a high risk of nest predation or brood parasitism. Although they have provided several examples consistent with this hypothesis, to date, it remains untested.

Crypsis and exposure to solar radiation can be counteracting selection pressures on egg coloration, as in the case of the ostrich eggs, which are often left unattended for several weeks before incubation commences. Magige, Moe & Røskaft (2008) evaluated the effect of colour experimentally on both surface and core egg temperatures by painting eggs brown and white, respectively, and comparing them with unpainted control eggs, which are cream in colour. Core temperatures of brown eggs (which were less visible to humans) exceeded 37.5 °C, which is the temperature at which embryo mortality starts to increase, although the experimental procedure of painting could have contributed to this rise through reducing eggshell porosity. Westmoreland, Schmitz & Burns (2007) avoided this problem by using naturally-pigmented eggs to measure the influence of egg coloration on heat gain in Brewer's, red-winged and yellow-headed blackbirds, either exposed to full sunlight or placed under a diffusing umbrella. Eggs in sunlight acquired heat more rapidly than those under the umbrella but heat gain did not vary with egg colour in either environment. Thus, the thermoregulation hypothesis was not supported, suggesting that ostriches may be an unusual case.

In summary, the nest environment, combined with specific properties of avian vision (Bennett & Cuthill, 1994), is likely to constrain egg coloration in important ways, which, until recently, have been ignored by studies of eggshell coloration, and merit further investigation. Endler & Théry (1996) have, for example, shown how forest light environments differ; these differences could well affect crypsis of eggs in forest-nesting birds. Langmore et al. (2009) have shown that the dark eggs of Gould's bronze-cuckoo Chalcites russatus are cryptic in the dark host nests which it parasitizes, whereas a congeneric bronze-cuckoo species parasitizing host nests with greater ambient light levels lays mimetic eggs.

SELECTION FOR BLUE–GREEN CHROMA

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. EGG COLORATION, AVIAN VISION, AND THE LIGHT ENVIRONMENT
  5. SELECTION FOR BLUE–GREEN CHROMA
  6. THE ROLE OF MACULATION
  7. CONCLUSIONS
  8. REFERENCES

Moreno & Osorno (2003) raised a new hypothesis to explain blue–green egg coloration, interpreting it as a sexually selected signal to the male of the laying female's genetic quality, aiming to induce a higher allocation of paternal care. Because the blue–green coloration of avian eggshells is produced by the pigment biliverdin, which is an antioxidant, they argue specifically that its deposition may signal antioxidant capacity. A comparative study on blue–green egg coloration in passerines by Soler et al. (2005) lends support to this hypothesis, although another study by Kilner (2006) found no evidence that blue–green eggs have evolved specifically to signal female quality.

Much of the support for this hypothesis is derived from a substantial body of work on the pied flycatcher Ficedula hypoleuca, which nests in holes. In this species, eggs become paler during the laying sequence, suggesting that biliverdin may be limiting or, alternatively, that females strategically invest a diminishing amount of biliverdin as they themselves decline in condition through laying. Males frequently visit the nest during the laying period, allowing them to observe their female's eggs before the start of incubation (Moreno et al., 2005). In descriptive studies, the intensity of egg coloration is related both to cell-mediated and humoral measures of female immunocompetence (Morales, Sanz & Moreno, 2006), and supplementation of female diets at laying results in higher measures of blue–green chroma (i.e. associated with higher biliverdin levels) in eggs (Moreno et al., 2006). Furthermore, the intensity of blue–green chroma reflects the amount of maternal antibodies in yolk: egg colour predicts both maternal condition and fledging success; and immunoglobulin levels in incubating females are positively associated with those in eggs (Morales et al., 2006).

Recent work by Soler et al. (2008) on spotless starlings Sturnus unicolor has further supported this idea by investigating the blue–green colour of their eggs using three different experiments. First, they found that manipulation of female body condition by removing wing feathers led to a reduction in the intensity of blue–green coloration, although this may just be a seasonal effect. Next, by replacing natural eggs with model ones, they showed that male starlings were prepared to adjust their subsequent provisioning effort on the basis of the intensity of the blue–green coloration of these eggs (hatchlings were later added to the nest in order to measure this). Lastly, they showed that the effect of experimental food supply, measured in terms of immunocompetence, was predicted by blue–green coloration of eggs, indicating that egg colour predicts the nutritional environment experienced by nestlings during development.

Krist & Grim (2007) tested the signalling hypothesis experimentally in the collared flycatcher Ficedula albicollis, which also lays blue–green eggs, by cross-fostering clutches between nests to disentangle the effects of female and/or territory quality and egg colour on paternal effort and nestling quality. They found that blue pigment appeared to be a limited resource for females but failed to find any evidence that either of two male parental effort parameters (i.e. feeding frequencies to nestlings and intensity of nest defence) were related to egg colour, although this may have been because the effect of the experimental manipulation on egg colour was not strong enough to elicit a response by males. Thus, although the evidence for an association between female quality and the intensity of blue–green coloration is strongly supported by evidence, that for males making reproductive decisions on this basis is less so.

The hypothesized mechanism is that blue–green egg colour signals female antioxidant capacity because biliverdin has antioxidant properties in the mother. Hanley, Heiber & Dearborn (2008) examined this mechanism, which is critical to this hypothesis, and found that female grey catbirds Dumetella carolinensis with higher total antioxidant capacity laid eggs with higher blue–green chroma; and that males in turn provided more care to nestlings from clutches with above average blue–green egg chroma. This shows a potential link between female antioxidant capacity and blue–green egg chroma but it remains unclear exactly what this relationship implies. Morales, Velando & Moreno (2008) examined the cost of egg pigmentation in terms of antioxidants, by inducing increased reproductive effort in female pied flycatchers through nest removal and measuring egg pigmentation and plasma antioxidant levels in comparison with a control group. Experimental females showed a negative association between egg colour and antioxidant levels, whereas there was no relationship for control birds, suggesting a trade-off between allocations to these traits. Furthermore, experimental females with more colourful eggs raised more fledglings (especially when breeding early), whereas controls did not show a relationship between egg colour and reproductive success. Females laying more colourful eggs could have shifted their allocation towards current reproduction at the expense of their own antioxidant defences. These results indicate that blue egg coloration is a life-history trait subject to trade-offs with other attributes, and could be a signal only under harsh breeding conditions. However, it is clear that interpreting variation in female antioxidant capacity will require a better understanding of the relative importance of dietary intake of antioxidants, oxidative stress, and the cost of depositing biliverdin into eggs (Hanley et al., 2008).

Reynolds, Martin & Cassey (2009) have pointed out that, in cavity-nesting birds, individuals entering cavities will take several minutes to adapt to lower ambient light levels. Combined with the finding reported by Avilés (2008) that chromatic differences in cavities are far less discernible than achromatic ones, it is unlikely that chromatic differences would be reliable signals of female quality in cavity-nesting species, which form the majority of those in which this hypothesis has been investigated. Because reflectance measurements do not account for what the receivers' visual sensory systems actually process, Cassey et al. (2008a) used a more sophisticated approach of modelling avian colour space, and calculated photon capture for both the four single cone photoreceptors; and the principal member of the double cone class, for eggs in clutches of two introduced European thrush species, Turdus merula and Turdus philomelos in New Zealand. They found that differences in the avian-perceived colours of individual eggs were not consistently correlated with different measures of maternal investment, including yolk carotenoid concentration. They point out that even a strong existing correlation between female phenotypic quality and egg traits does not necessarily imply that egg traits have signalling functions because female pigment investment may benefit developing embryos directly, without involving signalling to potential mates or conspecifics in general. Cassey et al. (2009) have also used their photoreceptor noise-limited colour opponent model of avian perception on museum collections to assess whether individual birds are able to discriminate between background colours of eggs in different conspecific clutches. Clutches from 46 species in the superfamily Muscicapoidea were measured at the British Natural History Museum, and most eggs were predicted not to be easily discriminable from those in other conspecific clutches in terms of the shells' background coloration. These findings make it unlikely that avian-visible egg colour polymorphism in this group could have evolved through selection at the intraspecific level.

The sexual signalling hypothesis is potentially most applicable to long-lived species such as seabirds, where equal sharing of parental care may enhance reproductive success, and matching responses by mates could reduce conflict over care. Morales, Torres & Velando (2010) found this to be the case in the blue-footed booby Sula nebouxii, in which they exchanged fresh eggs between nests of the same laying date, and recorded the subsequent parental incubation effort. Although blue egg colour did not affect male effort, original (but not exchanged) eggshell colour was correlated with pair matching in incubation. They suggest that blue egg coloration facilitates an equal sharing of incubation because this signal is functional only for a short time as the colour fades soon after laying. By contrast, in colonially-nesting ring-billed gulls Larus delawarensis, where males have ample opportunity to assess their mate's egg coloration relative to that of other females, Hanley & Doucet (2009) found little evidence that blue–green pigmentation was limiting to females and that the extent of blue–green egg coloration related to female or offspring quality, nor evidence for males providing more care to clutches with higher blue–green chroma.

One way in which female pigment investment can benefit embryos directly is in terms of protection from solar radiation, which has been explored by Lahti (2008). In Africa, village weavers Ploceus cucullatus are parasitized by the diederick cuckoo Chrysococcyx caprius. Introduced populations in Mauritius and in the Dominican Republic are not subject to brood parasitism and, here, unconstrained by selective pressures to differ from cuckoo eggs, they have evolved eggs with more intense blue–green chroma, which protect embryos from solar radiation. Even in Africa, tropical populations have more blue–green chroma than temperate ones, where solar radiation is less intense.

The most plausible selection pressure for blue–green coloration relates not to sexual selection but to brood parasitism. Polačikováet al. (2009) investigated the hypothesis that the common cuckoo selects open-nesting great reed warbler Acrocephalus arundinaceus host pairs of good phenotypic quality. They found that the likelihood of being parasitized increased with decreasing colour variability within clutches, with parasitized females allocating costly blue pigments to eggshells more equally than to unparasitized ones. Parasitized females also showed significantly better body condition, and the probability of being parasitized increased with increasing body condition.

THE ROLE OF MACULATION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. EGG COLORATION, AVIAN VISION, AND THE LIGHT ENVIRONMENT
  5. SELECTION FOR BLUE–GREEN CHROMA
  6. THE ROLE OF MACULATION
  7. CONCLUSIONS
  8. REFERENCES

Poultry scientists have long been interested in the possibility that eggshell pigmentation might increase eggshell strength (Solomon, 1987), although there has been some dispute as to whether relationships between colour and pigmentation are merely correlated with or genetically linked to colour, or are directly functionally related to pigmentation (Underwood & Sealy, 2002). Recent work by Gosler, Higham & Reynolds (2005) has shown a functional link, at least in respect of egg maculation. Evidence was presented indicating that protoporphyrin pigments serve a structural function related to eggshell thickness and calcium availability: eggshell maculation in the great tit Parus major increases with decreasing soil calcium levels; pigments demarcate thinner areas of shell; and both the pigment intensity and distribution are related to shell thickness. They suggest that maculation serves to strengthen the eggshell, and may also reduce eggshell permeability when large amounts of pigment are used. Maculation affects the rate of water loss from the egg during incubation (but not that of unincubated eggs), and experimental evidence indicates that the effect of female incubation behaviour on water loss compensates in some way for variation in eggshell characteristics: differences between females in the degree of such compensation are related to differences in clutch maculation (Higham & Gosler, 2006).

Working in the calcium-rich area of Wytham Woods near Oxford, A. G. Gosler & T. A. Wilkin (2009) have further found that, over a period of 20 years, spot spread has declined systematically in parallel with a decline in soil calcium availability. Pigment spread is a variable characteristic with no apparent genetic basis but is related to the availability of calcium, and is a strong indicator (positively related) of the thickness of unpigmented shell areas (Gosler et al., 2005; Gosler & Wilkin, 2009); the change in spread observed corresponds to a 6.5% decline in eggshell thickness over this period. The decline in soil calcium availability, which is believed to have resulted from calcium leaching as a result of acid precipitation, has been most pronounced in limestone areas, where the strongest decline in eggshell pigment spread has been observed. The study shows that pigmentation can be used as an indirect measure of eggshell thickness and as a bioassay of calcium availability.

Jagannath et al. (2008) studied the effect on eggshell pigmentation of the organochlorine insecticide dichlorodiphenyltrichloroethane (DDT) in the Eurasian sparrowhawk Accipiter nisus. Dichlorodiphenyldichloroethylene (DDE), the metabolite of DDT, is notorious for reducing the eggshell thickness of predatory birds. Sparrowhawk eggshells show both protoporphyrin spotting (either as a distinct layer within the shell, or as a superficial layer) and a blue–green biliverdin-based ground colour (Mikhailov, 1997). The presence of internalized protoporphyrin was associated strongly with both shell-thinning and DDE content, but DDE did not always cause eggshell thinning. Green coloration increased significantly with DDE, but not eggshell thickness, and the colour of eggs collected before 1900 was consistent with the environmental absence of DDT. Their results support the view that protoporphyrin compensates for shell-thinning and that biliverdin indicates female condition. The fact that a mechanism which compensates for calcium deficiency is found in two phylogenetically disparate taxa, and especially in one (sparrowhawk) whose diet is never calcium-deficient, also suggests that its evolution occurred relatively early in the avian clade, pre-dating the appearance of the passerines and should be a widespread phenomenon in Aves.

Martínez-de la Puente et al. (2007) broadened the sexually selected eggshell coloration hypothesis to include spotting with protoporphyrin, which is an intermediate metabolite of haem biosynthesis. They explored the relationship between maculation and several variables reflecting female health and condition in the eggs of blue tits Cyanistes caeruleus. Females laying more-spotted eggs displayed poorer body condition, higher cellular concentration of the stress protein HSP70 and marginally lower total immunoglobulin levels in their blood. These females in turn were paired with males with higher levels of HSP70 and lower concentrations of immunoglobulins. They regard higher concentrations of stress proteins as a function of protoporphyrin, which is known to induce oxidative stress, although the role of the birds' diet in this process remains unclear as calcium deficiency might underlie both the eggshell patterning and female condition.

Working on the same species, Sanz & Garcıa-Navas (2009) explored the consequences, both in terms of breeding success and parental effort, of eggshell pigmentation pattern. They assessed the effect of the distribution, size, and intensity of spots on eggshell physical properties and on breeding parameters, including nestling condition, investment of parents in offspring care, and reproductive output in both deciduous oak woodland and evergreen forest. Clutches with more widely distributed spots had thicker eggshells, a shorter incubation period, less mass loss per day, and higher hatching probability than those with concentrated spots. Eggs with larger and darker spots, by contrast, had only thicker eggshells and a shorter incubation period. They found a positive relationship between spot spread and male (but not female) provisioning rates. Spot darkness was positively related to female tarsus length (a measure of body-size), whereas spot spread was positively related to clutch size, male body mass, and nestling tarsus length. They conclude that spot spread is an indicator of clutch quality, but again, experimental evidence that increased male provisioning is a direct response to increased spot spread is lacking.

Garcıa-Navas (2010), also working on blue tits (which within the Paridae are distantly related to great tits, Gill, Slikas & Sheldon, 2005), provides experimental support for the association between protoporphyrin eggshell pigmentation and shell thinning. He did not observe a decrease in size and intensity of pigment spots in eggs from calcium supplemented nests, but found that provisioning of calcium-rich material during the egg-laying period led to a wider distribution of pigment spots and reduced the proportion of eggs with defective shells. Eggs from supplemented nests had thicker shells than did those from control nests. Eggshell thickness affects the probability of hatching, and there were fewer unhatched late eggs in supplemented nests compared to control nests.

Despite the evidence for a calcium-deficiency compensation mechanism provided by protoporphyrin, it must be noted that not all maculation serves this function. In their study of sparrowhawk eggshell pigmentation, shell-thinning and DDT, Jagganath et al (2008) noted that only spots or patches of pigment internalized within the shell appeared to serve this purpose. There were also darker superficial (i.e. on the eggshell surface) speckles and blotches of protoporphyrin, whose pattern showed no correlation with DDE content, and neither were they associated with localized shell-thinning; their occurrence may have contributed to crypsis. Similarly, Berg, McCormack & Smith (2009) examined maculation in Mexican jays Aphelocoma ultramarine, which exhibited high eggshell variability across a steep elevation gradient within the Sierra del Carmen mountain range, although they found no relationship between local soil calcium levels and spotting patterns. They did not report whether the pigment was distributed on (i.e. superficial) or through (internalized) the eggshell section. However, we might expect that the wide diet of this species, which includes small vertebrates and birds' eggs, should buffer the laying female against local calcium deficiencies as a result of variation in local geology.

Maculation serves as a signal in avian brood parasite-host systems. Both Avilés et al. (2006) and Cherry, Bennett & Moskát (2007a) suggest that cuckoos do not lay at random within a host population, but choose nests. In the latter study, this improves egg matching: naturally-parasitized common cuckoo eggs were more similar to great reed warbler A. arundinaceus host eggs, both in terms of maculation and background coloration. Furthermore, individual patterns of maculation may allow birds to identify their own eggs, effectively serving as signatures in the face of inter- or intra-specific brood parasitism (Swynnerton, 1918; Victoria, 1972; Cherry, Bennett & Moskát, 2007b). This does not appear to be the case when eggs lack spots. This idea is further supported by the findings of a recent study by López-de-Hierro & Moreno-Rueda (2010) on the role of egg spotting in conspecific brood parasitism in the house sparrow Passer domesticus. They found that experimental eggs were rejected when their spot patterns were modified, but not when background colour was changed. An earlier study by Siefferman (2006) found that although individual eastern bluebird Sialia sialis females produced clutches of immaculate eggs with unique coloration, this did not facilitate discrimination of conspecific parasitic eggs from host eggs.

CONCLUSIONS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. EGG COLORATION, AVIAN VISION, AND THE LIGHT ENVIRONMENT
  5. SELECTION FOR BLUE–GREEN CHROMA
  6. THE ROLE OF MACULATION
  7. CONCLUSIONS
  8. REFERENCES

The ubiquity of maculation among passerines suggests that allowing birds to recognize their own eggs in the face of either intra or inter-specific brood parasitism, or even just in dense breeding colonies (Birkhead, 1978), is more likely to be a secondary than a primary function. The most convincing explanation of the primary function of spotting is that it serves to strengthen the eggshell (Solomon, 1987; Gosler, 2006). The sexually-selected eggshell coloration hypothesis is plausible only in open-nesting birds, but only a single study provides experimental proof that coloration of eggs leads to increased male provisioning. Most importantly, it is unclear whether observed relationships between female phenotypic quality and egg traits imply that they have signalling functions because female pigment investment may benefit developing embryos directly. Furthermore, because it remains unclear how the male of a short-lived species benefits (either directly or genetically) by reducing his contribution to a brood produced with a mate in less than perfect health, the argument for the evolution of sexual selection of egg colour via the male's attention to his mate's condition requires further explanation. We agree with Reynolds et al. (2009) that plumage traits are likely to be a more reliable indicator of the quality of female birds than are egg coloration traits because they are both more conspicuous and visible for much longer periods.

Our review highlights the diversity of selective agents that have influenced the evolution of eggshell pigmentation in birds. No single factor (neither crypsis, intra or inter-specific brood parasitism, sexual selection, calcium-compensation or shell strengthening, nor female health, nor the physiological or physical requirements of the embryo developing within) can explain the diverse patterns of pigmentation presented across the class Aves. Eggshell patterns evolve under a multiplicity of selective pressures operating simultaneously, which contribute to embryo fitness in a number of different ways.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. EGG COLORATION, AVIAN VISION, AND THE LIGHT ENVIRONMENT
  5. SELECTION FOR BLUE–GREEN CHROMA
  6. THE ROLE OF MACULATION
  7. CONCLUSIONS
  8. REFERENCES