S29.4: Variation in avian sexual dichromatism in relation to phylogeny and ecology

Sexual dichromatism is thought to have evolved in response to selection pressures that differ between males and females. In turn, variation in sexual selection pressures may be influenced by ecological conditions. Variation in predation, parasitism, or the distribution and abundance of resources can shift the balance between the benefits of ornamental plumage and the cost of maintaining such traits; such environmental conditions can act on male and female plumage with varying degrees of independence. Ecological factors may affect the expression of condition-dependent traits in different environments. Thus, diversity of ecological conditions often leads to extensive intra- and interspecific variability in sexual dichromatism.

While correlation between sexual dichromatism and ecological factors has been thoroughly documented (Anderson 1994), major questions remain. First, it is unclear why some groups of bird species show extensive variation in sexual dichromatism while other groups, often apparently subjects to similar variation in ecological conditions, are remarkably conservative in their sexual ornamentation and degree of dichromatism. Second, given high genetic correlation between sexes that is often found in morphological traits (e.g., Lande 1980; Björklund 1996), we need more information on how fast sexual dichromatism can evolve following an ecological change and whether taxa or trait groups differ in their ability to evolve dichromatism. Third, it is unclear to what degree ancestral dimorphic traits (such as pigment type and pigmentation distribution) and ontogenetic sequences of plumage traits may ‘set the stage’ or bias the evolution of derived dimorphic traits, and whether such constraints differ between species groups. Finally, the role of sexual selection versus other selective forces, and the roles of various mechanisms of sexual selection in the production of sexual dichromatism are highly debated issues. To address these questions, comparative analyses of sexual dichromatism in relation to ecological pressures should be accompanied by reconstruction of possible phylogenetic pathways of change leading to dichromatism. In this review, we will illustrate this approach by first briefly reviewing a series of studies that documented ecological correlates of variation in sexual dichromatism while statistically accounting for phylogeny. We then will show how reconstruction of evolutionary transformations in sexual traits may allow tests of the importance of processes and mechanisms behind the evolution of sexual dichromatism.

Strong associations between sexual dichromatism, latitude of breeding, and migratory tendencies is one of the most frequently documented ecological patterns of sexual dichromatism. Higher latitude, migratory, and geographically widespread bird species are more sexually dimorphic than lower latitude, resident species, or species with limited geographic distribution (e.g., Mayr 1942; Grant 1965; Hamilton 1961; Bailey 1978; Scott & Clutton-Brock 1989; Fitzpatrick 1994; Peterson 1996; Omland 1997; Price 1998).

The examples in this section focus on three major explanations for these patterns: (1) geographical variation in patterns of sexual and natural selection pressures, (i.e., duration of mate sampling period and the importance of species recognition), (2) operation of non-selective factors, such as genetic drift, in resident, small, and isolated populations, that may be more common at lower latitudes, and (3) a combination of (1) and (2). For example, if intensity of sexual selection is influenced by the amount of genetic variation in populations, then reduced genetic diversity in small populations could reduce intensity of sexual selection and thus sexual ornamentation. These examples illustrate two points. First, knowledge of ancestral state of sexual dichromatism, sex-biased transitions in plumage brightness, and relative frequency of sexual dichromatism transformations across lineages allow us to derive unique testable explanations for latitudinal pattens of sexual dichromatism, and thus to distinguish among competing hypotheses (i.e., selective and non-selective evolutionary factors). Second, intraspecific studies that examine variation in sexual dichromatism in relation to population size (i.e., mainland versus island populations of a species), migratory tendencies (i.e., recently established urban resident populations of a migratory species) may be most informative for understanding of mechanisms behind the interspecific pattern.

Hamilton (1961) documented that in Parulidae and Icteridae, species at lower latitudes were less sexually dimorphic than their relatives at higher latitudes, a pattern that he largely attributed to a decrease in female brightness at higher latitudes. Noting that low-altitude species are more sedentary and maintain longer pair bonds than high-latitude species, Hamilton suggested that duller colouration of females may reduce intra-sexual aggression at the time of pair formation and increased sexual dichromatism could facilitate accurate species and mate recognition. Both of these processes may contribute to the rapid reestablishment of territories and pair bonds favoured by short northern breeding season (Hamilton 1961). Bailey (1978) investigated latitudinal variation in colouration across 787 passerine species in North and Central America and also found that sexual dichromatism is more pronounced in high-latitude species, but that in dichromatic species female are brighter at higher latitudes. The particular factors behind this pattern remain to be examined. Several studies corroborated Hamilton’s (1961) idea that greater dichromatism may be associated with a reduced mate-sampling period. Resident species and species that mate while in winter flocks may have more opportunities and a longer time to evaluate and compare potential mates based on actual performance. On the contrary, migratory species or species with high frequency of extra-pair fertilisations may have to base their mating decisions mostly on morphological traits such as bright plumage, often in the absence of direct comparison among males. These differences in traits used in mate selection decisions, and differences in the information content of condition-dependent traits in a given environment (Slagsvold & Lifjeld 1997) may account for greater sexual dichromatism in migratory species, found even when variation in geographical factors is statistically controlled (Fitzpatrick 1994; Badyaev 1997a).

Alternatively, latitudinal variation in sexual dichromatism could be explained by geographical variation in patterns of natural selection, such as latitudinal differences in the types and kinds of predation (i.e., by latitudinal variation in nest and adult predation, Martin 1995, see 1.3). If latitudinal patterns are mostly produced by changes in female brightness (i.e., females become duller at higher latitudes) we can predict positive correlation between risk of mortality and latitude. Alternatively, background-matching as a predation-avoidance strategy may favour brighter colours for both sexes at lower latitudes (i.e., in the tropics Bailey 1978). Crucial to understanding the role of natural selection in shaping latitudinal gradient in sexual dichromatism is the phylogenetic information on whether latitudinal transitions in plumage brightness are sex-biased.

Dichromatic taxa tend to have wider geographic distributions than monomorphic taxa (e.g., Price 1998). However, interaction between species’ dispersal and competitive abilities, and degree of sexual dichromatism is unclear. Species subject to strong sexual selection may be less ecologically plastic and have high extinction rates (McLain 1993; McLain et al. 1995; Sorci et al. 1998). Theory suggests that energy allocated towards sexual ornamentation may be unavailable for other traits associated with an organism’s ability to track environmental changes, and strong local selection that favours location-specific partitioning of resources may limit species dispersal ability and, thus occupied range (Kirkpatrick & Barton 1997). However, sexually-dimorphic species tend to have wider geographical distributions (Price 1998), and may have greater physiological tolerances compared to monomorphic species (Badyaev & Ghalambor 1998).

Sexual dichromatism may be associated with the ability to tolerate environmental fluctuations when secondary sexual traits under current selection indicate adaptive abilities of an individual, such as an ability to tolerate energetic demands of long migrations and ability to select good quality wintering habitats (Fitzpatrick 1994). Fitzpatrick (1994) suggested that this mechanism is responsible for an association between migratory tendency and sexual dichromatism. She suggested that if sexual dichromatism indicates migratory abilities, then on a macroevolutionary level, a shift from migratory to resident status (such as in island populations) should be followed by a transition between sexual dichromatism and monomorphism because of the loss of genetic variation in migratory genes and their limited usefulness in indicating phenotypic quality (Fitzpatrick 1994). She suggested that short mate-selection period of migratory species and strong selection for mate and species recognition may favour sexual dichromatism (see also Hamilton 1961). Under this hypotheses, gain in sexual dichromatism following the transition from resident to migratory status is as likely as loss of sexual dichromatism in resident population (Fitzpatrick 1994).

Crucial to the understanding of relative importance of selective and non-selective factors in latitudinal variation in sexual dichromatism is knowledge of the ancestral state of sexual dimorphism in a taxa. For example, sexual dichromatism in Anatidae is most common in species that have wide geographic distribution, breed at higher latitudes, and occur on mainland, while monochromatism prevails among non-migratory, southern species that have restricted, isolated ranges, and often occupy oceanic islands (Scott & Clutton-Brock 1989; Omland 1997 and references therein). Using phylogenetic reconstruction of sexual dichromatism, Omland (1997) showed that sexual dichromatism is an ancestral stage in dabbling ducks. Widespread and migratory species, when settling on islands and becoming isolated, may form monochromatic populations because of genetic drift and inbreeding common in small populations (Peterson 1996; Omland 1997; Burke et al. 1998). Both, the genetic drift and selective explanation hypotheses predict equal gain and loss in sexual dichromatism state following shifts in migratory tendencies (Fitzpatrick 1994). However, given the complex and integrated nature of sex-limited traits (e.g., complex colour patterns), genetic drift alone would likely lead to biases in the direction of sexual dichromatism loss (e.g., Omland 1997).

Variation in sexual dichromatism may be influenced by interaction of selective and non-selective processes. For example, genetic drift and inbreeding in small island populations lead to low levels of genetic variation. In turn, reduced genetic variation in sexual traits and decreased among-individual variance may lower intensity of sexual selection and reduce sexual ornamentation (Burke et al. 1998; Petrie & Kempenaers 1998).

Variation in sexual selection arising from variance in male reproductive success and parental investment can exert strong selection on sexual dimorphism (Payne 1984; Kirkpatrick & Ryan 1991; Williams 1992; Anderson 1994; Owens & Bennett 1997). For example, interspecific variation in the extent to which each sex contributes to parental care may influence sexual dimorphism because of the possible effects of parental investment on sexual selection (Trivers 1972). Thus, variation in ecological determinants of parental investment should cause variation in sexual dimorphism (Anderson 1994). Strong selection on male plumage ornamentation resulting from high variance in male reproductive success should push male morphology farther from a what is optimal under natural selection. If females obtain less benefit from plumage ornamentation then the result is increased sexual dichromatism. Across mating systems, variance in male reproductive success is expected to be higher in polygynous than in monogamous species and thus we predict greater sexual dichromatism in polygynous mating systems.

However, while close association between mating systems and ecological conditions is well established in birds (e.g., Owens & Bennett 1997 and references therein), direct association between mating systems and sexual dichromatism is rarely found. The examples in this section address this apparent paradox and illustrate three important points. First, the expected association between mating system and sexual dichromatism is often documented only when mating systems are clearly defined and sexual dichromatism is partitioned into components such as carotenoid-, melanin, or structurally based dichromatism (Payne 1984; Scott & Clutton- Brock 1989; Møller & Birkhead 1994; Owens & Bennett 1994, 1995, 1997, 1998; Møller & Cuervo 1998). Second, phylogenetic information on sex-biased transitions in plumage brightness is very useful in drawing our attention to what exactly needs to be explained - change in male colouration, change in female colouration, or both. Recognition of the need to know which sex is changing appearance facilitated generation of testable hypotheses of association between mating systems, plumage brightness, and sexual dichromatism (Scott & Clutton-Brock 1989; Jones & Hunter 1993; Irwin 1994; Badyaev 1997a; Owens & Bennett 1997 1998; Burns 1998). Furthermore, a hierarchical approach to phylogenetic studies of life histories and mating system allows studies of temporal concordance between changes in plumage versus changes in mating systems (Owens & Bennett 1995, 1997). Finally, knowledge of transition sequences in mating system and ornamentation in both sexes were instrumental in advancing our understanding sexual dichromatism variation in lekking species.

In one of the first studies of association between mating system and sexual dichromatism, Crook (1964) showed that the monogamous weavers (Ploceidae) were monomorphic, while polygynous species were dichromatic. He attributed the pattern to the distribution of food and nesting habitat (Crook 1964). However, most recent studies have found that the association between mating system and dichromatism is not straightforward. Indeed, most passerines are sexually dimorphic regardless of their social mating system; polygynous European passerines are not more often sexually dimorphic in plumage than monogamous species (Møller 1986). Despite their polygynous mating system, many species of hummingbirds (Trochilidae) are monomorphic with female colouration showing the most variation (Bleiweiss 1992).

In one of the few studies that documented association between mating systems and sexual dichromatism, Scott and Clutton-Brock (1989) examined variation in plumage in 146 species of Anatidae. They carefully defined mating systems based on frequency of pairing, duration of pair bond, and partitioning of parental care, and found that sexual dichromatism was greater in species with frequent pair formations and distinct parental roles. Variation in male plumage brightness was most strongly correlated with frequency of pairing, paternal care, and nest dispersion (i.e., potentially with mating opportunities), while female brightness varied the most with nest placement and nesting habitat features (i.e., with predation risk) (Scott & Clutton-Brock 1989). These results corroborated Kear’s (1970) findings that in the majority of monochromatic species of waterfowl both sexes shared parental duties, while in most dimorphic species females raised the young alone. Similarly, in passerines, males of monochromatic species were more likely to participate in nest building (Soler et al. 1998), and share incubation with females than males of dichromatic species (Verner & Willson 1969). Extensive paternal care is associated with both - reduced mating opportunities for males, and greater predation risk. To distinguish between roles of mortality and mating opportunities in the association between sexual dichromatism and mating systems, it is necessary to know whether variation is due to male or female colouration changes. Owens and Bennett (1994) documented that adult mortality closely covaried with parental care, but not with sexual dichromatism across 37 Palearctic bird species (but see below). Their results suggested that the often-documented association between sexual dichromatism and parental care may be caused not by mortality due to parental care, but by variation in mating opportunities among species with different amount of paternal care. Among socially monogamous passerines, variation in male plumage brightness was associated with differences in the frequency of extra-pair paternity; species with greater levels of extra-pair paternity had brighter males and greater sexual dichromatism (Møller & Birkhead 1994).

Owens & Hartley (1998) partitioned overall sexual dimorphism into components across 73 bird species and found that different types of dimorphism are affected by different selection pressures. Sexual dimorphism in size was strongly associated with variation in social mating system and parental roles (see also Björklund 1990, 1991; Webster 1992), while sexual dichromatism in plumage was most closely associated with levels of extra-pair paternity (see also Møller & Birkhead 1994), and more weakly with sex differences in parental care (e.g., Verner & Willson 1969, see below). Given distinct patterns of covariation among different types of dimorphism, it is interesting to examine evolutionary lability of various types of dimorphism. For example, frequency of extra-pair paternity often varies widely among different populations of the same species (Petrie & Kempenaers 1998). This variation may be more easily reflected in the evolutionary labile types of traits, such as carotenoid- based colours (Hill 1996a; Gray 1996; Hill & Brawner 1998; Badyaev & Hill, 1999). On the contrary, body size and dimorphism in ornamentation may be more phylogenetically constrained and morphologically integrated, and therefore vary only with most fundamental distinctions among mating systems.

In a series of comparative studies Owens and Bennett (1995, 1997, 1998) showed that patterns of diversification in mating systems and life history strategies is strongly historically nested. They argued that phylogenetically distant taxa may have converged on similar mating systems despite of different evolutionary histories. Thus, phylogeny and current selection may differentially contribute to variation in mating system across species. Ancient evolutionary events, such as ancestral changes in partitioning of parental care, nesting, and feeding habits may bias the predicted response of the lineage to current ecological conditions (Owens & Bennett 1997). This historical bias of taxa in adapting only a certain range of mating patterns, could also limit variation in sexual dichromatism, and more importantly, account for lack of contemporary association between sexual dichromatism and mating systems. Experimental manipulation of current selection pressures would induce predictably different changes in a mating system of certain taxa (such as propensity to desert mates if local mate availability is increased), depending of evolutionary history of the taxa (Owens & Bennett 1997).

Ecological determinants of paternal care are expected to cause variation in sexual dimorphism (Anderson 1994). Male parental investment differs with variation in ecological factors such as climate or resource (e.g., foraging or nesting sites) distribution. For example, colder nest microclimate and spatial separation of nesting and feeding resources (as found at high elevations) was commonly associated with greater male care (Badyaev 1997a; Badyaev & Martin, unpubl. manuscript). Thus, in monogamous species, the intensity of sexual selection should covary with ecological factors associated with the elevation of a species' breeding. This association was documented across 126 extant species of Cardueline finches; species occupying lower elevations were more sexually dimorphic in plumage than species at higher elevations, and the altitudinal variation was largely due to increased brightness of male plumage at lower elevations (Badyaev 1997a). Given that altitudinal variation in sexual dichromatism was mostly contributed by changes in male plumage, further hypotheses and tests of potential cost of greater paternal care at high elevations, greater costs of bright male plumage production (i.e., diet and molt) and maintenance (i.e. variation in predation) were advanced (Badyaev 1997 ab; Badyaev & Martin, unpubl. manuscript).

Irwin (1994) examined variation in sexual dichromatism across Icterinae and reported that sexual dichromatism covaried with mating system and that polygynous species were more sexually dichromatic. Irwin found that association between sexual dichromatism and mating system was due largely to changes in female plumage; female colouration was more evolutionary labile than male colouration. Irwin suggested that variation in sexual dichromatism in Icterinae results from social selection on females rather than sexual selection on males. Selection on female by males to display brighter plumage should be greater in monogamous systems (Moreau 1960; Irwin 1994). This selection and more intensive female-female interactions may account for association between female plumage brightness, sexual dichromatism, and the mating system (Johnson 1988; Trail 1990; Bleiweiss 1992; Hill 1993a; Irwin 1994). These studies emphasized the importance of distinguishing between monomorphism when both sexes are bright and monomorphism where both sexes are dull. ‘Dull’ monomorphism could arise from monogamous mating systems where mates have the extended opportunity to evaluate each other’s relative quality based on performance and direct comparisons (e.g., mating while in winter flocks), and where selection pressures are similar between sexes. Examples could include monomorphism of non-migratory species and high-elevation species (Fitzpatrick 1994; Badyaev 1997a). ‘Bright’ monomorphism could result from similar selection pressures acting on the sexes and should be prevalent in monogamous mating systems with short mate-sampling periods (Jones & Hunter 1993; Irwin 1994).

Sexual dichromatism should be strongly associated with lek breeding, because variance in male reproductive success and hence sexual selection is assumed to be a very strong force in this mating system (Darwin 1871; Payne 1984; Kirkpatrick 1987). However a series of studies documented that lekking species are not more likely to be sexually dimorphic in plumage (e.g., Payne 1984; Höglund 1989). Studies of association between lekking and sexual dichromatism illustrate two points. First, it is important to know the sequence of transitions, i.e. whether shift to or from lekking behaviour precedes the change in sexual dichromatism. For example, if it is suggested that sexual dichromatism has evolved as a result of transition to lekking, it needs to be shown that shift to lekking resulted in sex-biased selection on plumage colouration. Second, examination of current selection in both sexes is needed to generate hypotheses about predicted patterns of colour variation in relation to lekking. Third, phylogenetic information about ancestral state of sexual dichromatism and plumage brightness in both sexes is most useful. For example, transition between monomorphic dull to monomorphic bright states is expected under strong correlated response of female characteristics to selection on males prior to evolution of sex-limited variation (Lande 1980). Increased risk of predation on leks may explain changes to monomorphic dull from sexually dimorphic or monomorphic bright as a result of transition to lekking (Bleiweiss 1997). Bleiweiss (1997) examined covariation of sexual dichromatism and plumage brightness with occurrence of lekking behaviour across 415 bird species. Analysis of evolutionary transitions of plumage brightness in both sexes allowed him to conclude that in addition to sexual selection, predation risks and foraging behaviours associated with lekking are likely to constrain plumage variation among lekking species (Bleiweiss 1997).

One explanation for sexual dichromatism is that it evolved through differential signaling of sexes to predators and selection for less conspicuous females (Wallace 1889; Baker & Parker 1979; Butcher & Rohwer 1989; Götmark 1992, 1993; Götmark et al. 1997; reviewed in Götmark 1998). The hypotheses of association between mortality and sexual dichromatism have been tested in two ways. First, researchers examined across-taxa variation in sex differences in mortality looking for evidence for sexual ornamentation cost. These studies tested the costs of sexual selection without the confounding effects of intraspecific variation in individual quality. However, most of the studies in this group have focused on variation in adult mortality, while dimorphism-induced variation in juvenile mortality is largely unexamined (e.g., Owens & Bennett 1994). Second, researchers have attempted to isolate factors or behaviours that affect mortality associated with sexual dichromatism. These studies looked for correlations between predation and display and mate-selection behaviours, participation in parental care (i.e., incubation, nestling provisioning), and plumage brightness and dichromatism. The inference from these studies is greatly strengthened by examining changes in sexual dichromatism as a consequence of changes in nesting or displaying habits, or by applying a hierarchical approach to changes in sexual dichromatism, life history strategies, and mating systems.

Sexual dichromatism in birds is generally thought to arise from sexual selection favouring conspicuous colouration in males, although natural selection (e.g., predation) is thought to ultimately limit conspicuousness (Darwin 1871; Fisher 1930; Hingston 1933; Kirkpatrick et al. 1990; Promislow et al. 1992, 1994; Götmark et al. 1997). Alternatively, bright colouration may be favoured by predation because it advertises that a prey is unprofitable and degree of sexual dichromatism may be a direct function of the difference between the sexes in their profitability to a predator (Cott 1946; Baker & Parker 1979; Butcher & Rohwer 1989; Götmark 1992, 1993, 1994, 1998). Promislow et al. (1992, 1994) have examined variation in sex-specific mortality schedules as consequences of the costs of sexual ornamentation in passerines and waterfowl. They suggested that female mortality may constrain the upper limit of sexual dichromatism in species by limiting the maximum mortality rate of males. In turn the brightness of males could be further constrained by additional mortality associated with bright plumage and more intensive sexual competition (Promislow et al. 1992, 1994; Promislow 1996). Similarly, Götmark et al. (1997) showed that predation on adult chaffinches (Fringilla coelebs) exerts greater pressure on female colouration than on male colouration, and could ultimately lead to variation in sexual dichromatism. In cardueline finches, variation in sexual dichromatism and plumage brightness in both sexes closely corresponded to variation in life history traits; sexual dichromatism was negatively correlated with fecundity because the association between plumage brightness and fecundity was different for males and females. Male plumage brightness was negatively correlated with clutch size and numbers of broods, but female brightness was positively correlated with clutch size across finches (Badyaev 1997b).

Examining variation in sex-specific costs of plumage brightness along an altitudinal gradient, Badyaev (1997b) found that the association between sexual ornamentation and fecundity was more similar between sexes in high-elevation species than in low-elevation species. Associations among plumage brightness and life history traits changed more with altitude for males than females, which is consistent with higher altitudinal variation in male plumage brightness in finches (Badyaev 1997a). Monomorphism of high elevation species may be caused by more similar selection pressures caused by equal sharing of parental care between sexes at higher altitudes. Badyaev & Martin (unpubl. manuscript) suggested that elevational variation in sexual dichromatism (Badyaev 1997a) is due to both higher adult mortality at lower elevations and reduced juvenile mortality at higher elevations (Badyaev 1997bc). While low elevations favour increased and more elaborated sexual ornamentation, development of such traits commonly results in reduced juvenile survival (Owens & Bennett 1994). Thus, prevalence of monomorphism across high elevation species could contribute to higher juvenile survival (Badyaev 1997c).

While a number of studies clearly established the relationship between sexual dichromatism and mortality, two problems persist: (1) the specific factors (e.g., variation in mating and parental behaviours) behind this relationship remain to be examined, and (2) studies that would allow directional hypotheses of causality are needed. Below I will review some factors that may mediate an association between sexual dichromatism and mortality.

If nest predation constrains brightness (i.e., Wallace 1889; Baker & Parker 1979; Shutler & Weatherhead 1990; Johnson 1991), female brightness should vary with nest predation, particularly in species where only the female incubates eggs and broods young. In contrast, male brightness may not vary as strongly with nest predation because of the reduced time males spend at the nest. Sexual dichromatism has been argued to vary inversely with nest predation (Scott & Clutton-Brock 1989; Shutler & Weatherhead 1990; Johnson 1991), although the association was never directly examined. By separately examining male and female plumage across Parulidae and Carduelinae, Martin & Badyaev (1996) found that female plumage brightness varied among nest heights. They found that female plumage brightness was negatively correlated with nest predation and the pattern of males and females was distinctly different. These results suggested that nest predation may place greater constraints on female than male plumage brightness, at least in taxa where only females incubate eggs and brood young. Martin & Badyaev (1996) found that female plumage patterns vary at least partly independently of male patterns, emphasizing the need to consider both female and male plumage variation in tests of plumage dimorphism. In warblers and finches, sexual dichromatism differed between ground- and off-ground-nesting species, but the relationship between plumage dimorphism and nest predation was positive rather than negative (Shutler & Weatherhead 1990; Johnson 1991). Moreover, differences in sexual dichromatism between ground- and off-ground-nesting birds result only partially from decreased male brightness (Shutler & Weatherhead 1990; Johnson 1991) and is contributed mostly by the increase in female brightness in ground-nesting birds related to their reduced risk of nest predation as compared to shrub-nesters (Martin & Badyaev 1996). Effects of nest predation on sexual dichromatism are most evident when one separately examines sexual dichromatism in different body parts. For example, dichromatism of rump but not breast strongly covaried with nest placement across Carduelinae (Badyaev 1997a). Variation in parasite prevalence across nesting and foraging strata could also contribute to vertical stratification of sexual dichromatism and plumage brightness (Hamilton & Zuk 1982; Garvin & Remzen 1997)

The importance of current variation in nesting biology to variation in sexual dichromatism was challenged by the view that mortality variation is almost entirely due to ancient evolutionary events, and that current variation in nesting and feeding habits is largely irrelevant to current avian life history strategies (Owens & Bennett 1995). If ancient and hierarchically-nested evolutionary diversifications (e.g., changes in nest placement) were associated with changes in sexual dichromatism, we should see concordant and similarly historically-nested patterns of divergence in sexual dichromatism. However, other studies suggested that large-scale diversification in life histories are produced by more recent ecological changes (e.g, Martin & Clobert 1996). These examples illustrate that to properly test the association between nesting and foraging habits and sexual dichromatism one must examine historical transitions in sexual dichromatism and plumage brightness in relation to changes in nesting strata or parental behaviour (e.g., Owens & Bennett 1994, 1997).

Ecological change associated with exploitation of new habitats is often accompanied by changes in mate recognition traits. These novel traits may evolve as a result of either preexisting sensory biases within lineages or characteristics of new environments that make certain traits more easily perceived (Schluter & Price 1993; Price 1998). Physical characteristics, such as abrasiveness, UV radiation, and thermoregulation requirements could ultimately constrain sexual dichromatism by favouring certain pigmentation and patterns of colouration (Burtt 1989). The examples in this section emphasize three points. First, comparative studies need to show that colour patterns are indeed preceded by habitat shifts (e.g., Marchetti 1993). Second, it needs to be shown that divergence into different habitats promotes divergence in sexually selected traits (Schluter & Price 1993; Barraclough et al. 1995; Price 1998; Møller & Cuervo 1998). Finally, interactions among habitat characteristics, display behaviour, and plumage colouration have to be examined on a macroevolutionary scale (Endler & Théry 1996; Irwin 1996).

Physical features of habitats may favour certain plumage pigmentation and thereby constrain distribution of other types of pigments or structural colours. For example, birds living in more abrasive environments have more melanin in their plumage (Burtt 1989) and the body surfaces that are more vulnerable to wear and abrasion have a higher proportion of melanin pigmentation (Fitzpatrick 1998). In turn, the presence of melanin may affect distribution of structural colours (reviewed in Prum 1998) and carotenoid-based pigmentation (references in Savalli 1995). High concentration of pigments may protect birds from UV radiation; higher sexual dichromatism in tanagers breeding at higher elevations (Brush 1970) may be attributed to this factor. Marchetti (1993) has shown that colour, colour patterns, and sexual dichromatism across species can be related adaptively to the light environment, where bright species occupy dark habitats. Götmark and Hohlfält (1995) found that male and female pied flycatchers (Ficedula hypoleuca) are about equally difficult to detect and therefore male plumage may be an example of disruptive crypsis.

Price (1996) examined variation in sexual dichromatism across finch species, and found that drier and more open habitats had a lower proportion of dichromatic species than did closed and moist habitats. Similarly, cardueline finches dwelling in closed habitats were more sexually dimorphic in plumage than related species in open habitats (Badyaev 1997a). However, plumage dichromatism in finches is associated with solitary nesting and most open habitat species are semi-colonial (Badyaev 1997a). Thus, habitat influences may be confounded by effects of nest dispersion. Price (1996) suggested that finches in the closed habitats may breed at higher densities and thus have increased potential for extra-pair paternity (Møller & Birkhead 1993, but see Westneat & Sherman 1997). The association between habitat type an sexual dichromatism, documented by Price (1996) could be confounded by differences in latitudinal distribution of finches, migratory tendencies, and differences in predation and parasitism risk (see 1.3).

Endler & Théry (1996) reported extremely high degree of ambient light specificity in display behaviours in three lekking tropical species. However, it is unclear whether such behaviours follow existing colouration patterns to maximize its function, or the colouration patterns evolve as a result of the light environment or display behaviours (Endler & Théry 1996). Phylogenetic analysis of transition sequence among light environments, behaviour, and plumage colouration patterns will further our understanding of the roles behaviour, ecology, and phylogenetic constraints play in the evolution of colouration patterns and sexual dichromatism.

Differences among habitats and geographical locations in food composition may influence sexual dichromatism Abbot et al. (1977), especially in the in diet-dependent components of sexual dichromatism (Hill 1993b, 1994b). For example, geographical variation in intensity of red colouration among populations of House Finches Carpodacus mexicanus was influenced by local access to carotenoids (Hill 1993b). However, female House Finches from all locations preferentially paired with brighter males (Hill 1994b).

Utilization of phylogenetic methods together with a consideration of the source of plumage colouration (melanin, carotenoid, or structural) as well as developmental constraints and pathways allow testing of the signal content of plumage colours and a better understanding of the roles that developmental and phylogenetic constraints play in evolution of sexual dichromatism. The examples in this section illustrate two points. First, different components of sexual dichromatism (i.e., carotenoid or melanin-based pigmentation, and structural colouration) have different evolutionary lability and distinct signal functions in behavioural interactions. Consequently, sexual dichromatism in these different types of colouration shows distinct patterns of covariation with selection pressures. Second, phylogenetic information is essential for an understanding of the sequence of transitions in complex sexual traits (i.e., colour patterns and colour combinations) and hence constraints on plumage colour patterns. Finally, traits may differ in the amount of sex-limited genetic variance or the information they provide in a given environment (Møller & Pomiankowski 1993; Marchetti 1998). These differences may cause biased evolution of sexual dichromatism in such traits. Phylogenetic methods allow us to understand the direction and magnitude of change in these traits by reconstructing their ancestral states.

In a series of comparative studies Hill (Hill 1994a, 1996a; Badyaev & Hill 1999) suggested that because carotenoid-based plumage colouration is more dependent on condition and less constrained developmentally than is melanin-based colouration, variation in sexual dichromatism should be driven more by changes in carotenoid-based colouration between males and females than by changes in melanin-based colouration. Badyaev and Hill (1999) examined this hypothesis and found that across all cardueline species (1) carotenoid-derived colouration has changed more frequently than melanin-based colouration, (2) in both sexes, increase in carotenoid-based colouration, but not in melanin-based colouration, was strongly associated with increase in sexual dichromatism, and (3) sexual dichromatism in carotenoid-based colouration contributed more to overall dichromatism than sexual dichromatism in melanin-based plumage. These results corroborated previous findings that in finches, the degree of sexual dichromatism of carotenoid-based plumage colouration increased with plumage redness, but not with amount of black pigmentation (Hill 1996a).

These findings supported the results of Gray’s (1996) analyses of male plumage variation across all North American passerines. Gray found that the amount of carotenoid pigmentation in male plumage was positively associated with overall dichromatism, while the amount of melanin and structural colouration in male plumage was not related to overall dichromatism. Analyzing patterns of variation across different clades of passerines, Gray (1996) noted that carotenoids appear to be used as ornamental signals by granivorous and insectivorous taxa (for which they are present in the diet but not overly abundant), but not used by frugivorous (for which they are overly abundant in the diet) or carnivorous taxa (for which they are rare in the diet). Consequently, Owens & Hartley (1998) found that carotenoid-, melanin- and structurally-derived sexual dichromatism do not show similar patterns of covariation with social mating systems, parental roles, and ecological conditions.

Phylogenetic methods were instrumental in revealing the roles of developmental constraints in the expression of colours, and especially of colour patterns. The similarity of colouration patterns and pigment distribution across a wide range of avian species implies common developmental mechanisms and constraints. In their comprehensive study of the evolution of colours and colour patterns in Phylloscopus warblers, Price & Pavelka (1996) showed that component elements of melanin-distribution patterns were repeatedly gained and lost during evolution. Price & Pavelka (1996) suggested that once evolved in some distant ancestor, the pattern of colouration may persist in a lineage (even when not expressed in the current phenotype), and can quickly reappear after loss given favourable selection pressures. Moreover, further selection may bias evolution of other components of the phenotype in the context of the patterns already present (i.e., overlay of different pigments, display postures emphasising colouration pattern, etc.) (Price & Pavelka 1996). Thus, identification of evolutionary sequences of colouration patterns is essential to the study of sexual dichromatism (Price & Pavelka 1996).

Schluter & Price (1993) noted that selection for sexual dimorphism will favour traits with a greater amount of sex-limited genetic variance, greater relevance to current condition, or easier detection. Therefore, under certain conditions, these traits (such as songs, various displays) will be more likely to invade a sexually dichromatic population, and thus bias evolution and establishment of other sexually dimorphic traits. For example, predation may limit variation in sexual dichromatism in Parulinae warblers, and song complexity may replace plumage characteristics as the target of sexual selection (Shutler & Weatherhead 1990). Similarly, Bailey (1978) suggested that structural colours are favoured by selection in the tropics because structural colours are easily changed by behavioural displays depending on variable light conditions in closed and dark tropical habitats (see also Endler & Théry 1996 and references therein).

Phylogenetic analyses of sexual dichromatism variation allow the identification of taxa groups that (1) retained sexual dichromatism after the termination of selective forces that caused then, and (2) show no variation in sexual dichromatism despite changes in selective pressures assumed to cause variation in sexual dichromatism (e.g., Sheldon & Whittingham 1997). Such biases in sexual dichromatism variation could be a result of phylogenetic constraints (McKitrick 1993; Miles & Dunham 1993). Potential causes of such biases in evolution of sexually dimorphic traits could include: reduced additive genetic variance and limited phenotypic variation, close genetic covariance among components of sexual dichromatism, phenotypic plasticity that could reduce selection pressures on sexual dichromatism (such as behavioural modification of displays), stabilising selection in which a trait is maintained by selection against alternative phenotypes, and pleiotropy (see reviews in Miles & Dunham 1993; Edwards & Naeem 1989; McKitrick 1993; Leroi et al. 1994). Sheldon and Whittingham (1997) noted that phylogenetic methods may be used to distinguish sexual dichromatism variation due to current stabilising selection (i.e. selection based on current ecological conditions) from phylogenetic conservatism caused by other evolutionary forces (Miles & Dunham 1993; Leroi et al.1994).

Phylogenetic inferences about origin of sexual dichromatism

Sexual dichromatism arises from sex-limited expression of genes or from selection acting on traits with sex-limited or sex-biased genetic variation (Lande 1980). However, once sex-limitation is established, variation in sexual dichromatism can be affected by both non-selective and selective factors (Anderson 1994). The examples in this section illustrate that phylogenetic methods can be used to distinguish among variation in sexual dichromatism produced by various evolutionary processes (e.g., Sheldon & Whittingham 1997)

Sexual dichromatism can evolve if there is sex-limited genetic variation or if sex-limited expression of some genes is favoured by sex-biased selection pressures. The sources of sex-linked variation could range from mutations on sex-chromosomes (Hutt 1949) to sex-limited expression of genes (Lande 1980). Several studies suggested that expression of sex-limited gene effects (such as colour or specific pattern) may be dependent on sex-specific hormonal balance (references in Owens & Short 1995). In their review, Owens & Short (1995) provided evidence that expression of secondary sexual colours in males is controlled by the absence of estrogen rather than the presence of testosterone. Thus, if sexual dichromatism is determined by sex-limited expression that is pleiotropically mediated (i.e., through hormonal balance), we can predict (1) easier and faster loss than gain of male secondary sexual colours, and (2) more frequent phylogenetic transition from dichromatism to monochromatism than from monochromatism to dichromatism (e.g., Price & Birch 1996; Omland 1997; see 2.3). If sexual dichromatism results from mutations on sex chromosomes that are magnified by selection favouring dichromatism, no directional biases between loss and gain of sexual dichromatism are expected.

Once sex-limitation is established, genetic drift, selection, and genetic interactions could influence the evolution of sexual dichromatism. On a macroevolutionary scale, genetic drift is not expected to produce consistent convergences of sexual dichromatism with other factors (e.g., ecological conditions) across lineages (Leroi et al., 1994; Sheldon & Whittingham 1997). On the contrary, if sexual dichromatism evolved in response to selection, change in sexual dichromatism should follow a certain sequence, e.g., transitions in environments or behaviours should be followed by transitions in traits. Pleiotropic interactions should produce multiple and simultaneous effects on sexually-dimorphic traits (Sheldon & Whittingham 1997 and references therein).

The evolution of sexual dichromatism requires sufficient additive genetic variance for a response to selection. Initial response to change in selection pressures may be limited because of high genetic correlation between the sexes (Lande 1980). One way to explore whether the amount of additive genetic variance biases evolution of sexual dichromatism is to examine the relative frequency of changes between monomorphism and dimorphism, as well as the variance in rates of evolution of male and female plumage traits (e.g., Price & Birch 1996). The examples discussed in this section suggest that the evolution of sexual dichromatism is largely unconstrained by the lack of genetic variance and that evolutionary losses of sexual dichromatism are more likely that gains. It is suggested that genetic drift and inbreeding in small parapatric populations, combined with biases towards loss of sex-limited and complex characters have probably caused repeated loss of sexual dichromatism in birds (Peterson 1996; Omland 1997; Price 1998).

Price and Birch (1996) estimated the frequency of evolutionary transitions in dichromatism across 5,298 passerines and found that (1) sexual dichromatism evolved independently and numerous times, indicating that the evolution of sexual dichromatism was largely unconstrained by an absence of genetic variance, and (2) transitions from sexual dichromatism to monomorphism were more likely than transition from monomorphism to sexual dichromatism. Omland (1997) reached similar conclusions in his study of Anatidae. He showed that (1) sexual dichromatism is an ancestral trait, and (2) evolution of sexual dichromatism was biased towards loss of dichromatism. Similarly, Burns (1998) found that tanagers (Thraupidae) descended from an ancestor that was dichromatic with colourful males and dull females. These findings corroborated Peterson (1996) study where he examined geographical variation in sexual dichromatism in 158 species of birds representing 43 families and concluded that sexual monomorphism with bright males and dull females is a likely ancestral stage in birds.

Several phylogenetic studies addressed whether transition of dichromatism states are due to male or female evolution. Examining the relative frequency of bright and dull monomorphism, and sexual dichromatism, Peterson (1996) concluded that the evolution of female plumage contributed to the evolution of sexual dichromatism as often as did evolution of male plumage. Changes in male and female plumage contributed equally to variation in sexual dichromatism, and males were five times more likely to lose bright plumage than to gain it, while in females the trend was the opposite (Peterson 1996). The fact that loss of sexual dichromatism occurs in both directions (to ‘dull’ and to ‘bright’ monomorphism) makes it less likely that selection can explain the majority of cases, leading Peterson (1996) to propose genetic drift as a potential evolutionary force behind variation in sexual dichromatism (see also Björklund 1990, 1991). Björklund (1991) documented that in two lineages of blackbirds, sexual dichromatism resulted from a loss of female brightness rather than a gain in male brightness. Similarly, Irwin (1994) found that changes in female plumage were more frequent than changes in male plumage, and that females were brighter in monogamous than in polygynous mating systems in Icterinae. These results corroborated Moreau (1960) observation that an association between plumage brightness and mating systems is mostly due to variation in female plumage. Using phylogenetic reconstructions, Burns (1998) found that tanagers (Thraupidae), descended from an ancestor that was dichromatic with colourful males and dull females. Transitions in sexual dichromatism where only males or only females changed were more common than transitions where both sexes changed. Female plumage brightness changed at least twice more often than male plumage (Burns 1998).

Phylogenetic analyses provide a powerful way of testing predictions of different sexual selection mechanisms. The examples in this section illustrate two points. First, hypotheses of sexual selection mechanisms can be tested by experimental examination of the congruence between current male phenotype and current female preferences, as well as by examining the general concordance between phenotypic appearance and current ecological conditions. Second, different selection models make distinct predictions of diversification patterns, hierarchical complexity, and convergence among lineages, thus allowing strong inferences about sexual selection mechanisms.

Hill (1994b) proposed that in the absence of changes in female preferences or viability costs, the sensory exploitation and the runaway models (reviewed in Anderson 1994) cannot account for reduction in sexually selected trait. Specifically, in the runaway model of sexual selection male appearance closely covaries with female preference, while under indicator models, females display preferences for extreme development of traits, while males are constrained (i.e., by physiological and energetics costs) in the ability to develop more elaborated ornaments (Hill 1994a, 1996a). Examining these predictions in geographic variation in male appearance and female preference across subspecies of the House Finch, Hill (1994a) concluded that the models of ‘non-adaptive’ (e.g., runaway) mate choice can be rejected. In the study of the delayed attainment of ornamental breeding plumage by young males (i.e. delayed plumage maturation), Hill (1996b) documented that selection acting on physiological trade-offs (size and colour of the patch in the House Finch) could cause concordant evolution of the expression of the ornamental trait in adults and the developmental speed at which the trait is acquired.

Recent studies of bowerbirds (Ptilonorhynchidae) by Kusmierski et al. (1997) and manakins (Pipridae) by Prum (1994, 1997) showed that patterns of trait distribution and differential evolutionary lability of traits could be used to uncover mechanism of selection operating within a lineage. In the runaway model, drift along equilibria lines between male trait and female preference produces periods of rapid evolution resulting in large-scale diversifications and elaboration of male secondary sexual traits (Lande 1980, Kirkpatrick 1987). Thus, the runaway model predicts (1) rapid differentiation in secondary sexual traits and evolution of multiple secondary sexual traits among lineages with little convergence between lineages, and (2) historically-nested distribution of traits that are shared among lineages within a clade (Prum 1997). The quality-indicator models of sexual selection predict different historical patterns. Because quality indicators are costly, selection on such traits would ultimately result in reduced genetic variance in these traits (reviewed in Anderson 1994). Thus, evolution of multiple indicator traits is strongly constrained because evolution of a new indicator would favour elimination of previous ones (Hill 1994a, 1996a; Iwasa & Pomiankowski 1994). Consequently, indicator models predict sequential evolution of increasingly informative and increasingly constrained sets of traits within lineages (Hill 1994a, 1996a; Prum 1997). The ‘chase-away’ process of sexual selection (Holland & Rice 1998) also predicts sequential evolution of more exaggerated traits, but that evolution should be accompanied by selection for retention of existing traits. Sensory bias models predict frequent convergence in traits across lineages that share wide and similar preexisting biases (Anderson 1994; Hill 1994a; Irwin 1996). The sensory drive hypothesis predicts strong convergence of preferences and traits across lineages with similar ecological conditions (Hill 1994a; Prum 1997). Similarly, if sexual traits evolve to minimize costs associated with mate sampling and selection, strong convergences in sexual traits among lineages that share similar ecological condition are expected (Schluter & Price 1993; Prum 1997; Price 1998). Finally, direct selection for species recognition should favour uniqueness of displays, and selects against shared traits among lineages, thus resulting in decreased trait diversity and reduced hierarchical structure within a lineage (Hamilton 1961; Grant 1965; Grant & Grant 1997; Prum 1997; Price 1998).

Kusmierski et al. (1997) found that in bowerbirds, sexually dimorphic plumage characters were extremely labile and, aside from few constraints on fundamental levels of display and plumage patterns, sexual dichromatism appeared to be largely unconstrained. This pattern of plumage variation was most consistent with the predictions of runaway models of sexual selection. Prum (1997) tested predictions of various models of sexual selection on display traits in manakins. He found that (1) diversity of manakin traits was explosive, indicating that evolution of these traits is largely unconstrained. Patterns of diversity and hierarchical structure of these displays within lineages was most consistent with the predictions of runaway and sensory bias mechanisms (Prum 1997, see also Irwin 1996) and also may be consistent with phylogenetic predictions of the ‘chase-away’ model of sexual selection (Holland & Rice 1998).

Abbott, I. Abbott, L.K. & Grant, P.R. 1977. Comparative ecology of Galapagos ground finches (Geospiza Gould): evaluation of the importance of floristic diversity and interspecific competition. Ecological Monographs 47: 151-184.

Anderson, M. 1994. Sexual selection. Princeton: Princeton University Press.

Badyaev, A.V. 1997a. Altitudinal variation in sexual dimorphism: A new pattern and possible underlying causes. Behavioral Ecology 9: 675-690

Badyaev, A.V. 1997b. Covariation between sexually selected and life history traits: An example with Cardueline finches. Oikos 80: 128-138.

Badyaev, A.V. 1997c. Avian life history variation along altitudinal gradients: An example with Cardueline finches. Oecologia 111: 357-364

Badyaev, A.V. & Ghalambor, C.K. 1998. Does a trade-off exist between sexual ornamentation and ecological plasticity? Sexual dichromatism and occupied elevational range in finches. Oikos 82: 319-325.

Badyaev, A.V. & Hill. G.E. 1999. Evolution of sexual dichromatism: Contribution of carotenoid- versus melanin-based coloration. Biological Journal of the Linnean Society: in press.

Bailey, S.F. 1978. Latitudinal gradients in colours and patterns of passerine birds. Condor 80: 72-381.

Baker, R.R. & Parker G.A. 1979. The evolution of bird colouration. Philosophical Transactions of the Royal Society B 287: 63-130.

Barraclough, T.G., Harvey, P.H. & Lee, S. 1995. Sexual selection and taxonomic diversity in passerine birds. Proceedings of the Royal Society of London, Series B 259: 211-215.

Björklund, M. 1990. A phylogenetic interpretation of sexual dimorphism in body size and ornament in relation to mating system in birds. Journal of Evolutionary Biology 3: 171-183.

Björklund, M. 1991. Evolution, phylogeny, sexual dimorphism and mating system in the grackles (Quiscalus spp.: Icterinae). Evolution 45: 608-621.

Björklund, M. 1996. The importance of evolutionary constraints in ecological time scales. Evolutionary Ecology 10: 423-431.

Bleiweiss, R. 1992. Widespread polychromatism in female sunangel hummingbirds (Heliangelus: Trochilidae). Biological Journal of the Linnean Society 45: 291-314.

Bleiweiss, R. 1997. Covariation of sexual dichromatism and plumage colours in lekking and non-lekking birds; A comparative analysis. Evolutionary Ecology 11: 217-235.

Brush, A.H. 1970. Pigments in hybrid, variant, and melanic tanagers (Birds). Comparative Biochem Physiology 36: 785-793.

Butcher, G.S. & Rohwer, S. 1989. The evolution of conspicuous and distinctive colouration for communication in birds. Current Ornithology 6: 51-108.

Burke, T., Griffith, S.C. & Smith, P. 1998. Identifying the factors driving extra-pair paternity in passerine birds: island populations are more monogamous, but is the reason genetic? Proceedings of the Seventh International Behavioral Ecology Congress. Pacific Grove. California.

Burns, K.J. 1998. A phylogenetic perspective on the evolution of sexual dichromatism in tanagers (Thraupidae): The role of female versus male plumage. Evolution: In press.

Burtt, E.H. Jr. 1986. An analysis of physical, physiological, and optical aspects of avian colouration with emphasis on wood-warblers. Ornithological Monographs 38: 1-122.

Cott, J.B. 1947. The edibility of birds: illustrated by five years' experiments and observations (1941-1946) on the food preferences of the hornet, cat and man; and considered with special reference to the theories of adaptive colouration. Proceedings of the Zoological Society of London 116: 371-524.

Crook, J.H. 1964. The evolution of social organization and visual communication in the weaver birds (Ploceinae). Behaviour suppliment 10.

Darwin, C. 1871. The descent of man, and selection in relation to sex. John Murray. London.

Edwards, S.V. & Naeem, S. 1993. The phylogenetic component of cooperative breeding in perching birds. American Naturalist 141: 754-789.

Endler, J.A. & Théry, M. 1996. Interacting effects of lek placement, display behavior, ambient light, and colour patterns in three neotropical forest-dwelling birds. American Naturalist 148: 421-452.

Fisher, R.A. 1930. The genetical theory of natural selection. Dover, New York.

Fitzpatrick, S. 1994. Colourful migratory birds: evidence for a mechanism other than parasite resistance for the maintenance of ‘good genes’ sexual selection. Proceedings of the Royal Society of London, Series B 257: 155-160.

Fitzpatrick, S. 1998. Birds’ tails as signaling devices: Markings, shape, length, and feather quality. American Naturalist 151: In press.

Gavrin, M.C. & Remsen, J.V. 1997. An alternative hypothesis for heavier parasite load of brightly coloured birds: exposure at the nest. Auk 114: 179-191.

Götmark, F. 1992. Anti-predator effect of conspicuous plumage in a male bird. Animal Behaviour 44: 51-55.

Götmark, F. 1993. Conspicuous colouration in male birds is favoured by predation in some species and disfavoured in others. Proceedings of the Royal Society of London, Series B 253: 143-146.

Götmark, F. 1994. Are bright birds distasteful? A reanalysis of H.B. Cott’s data on the edibility of birds. Journal of Avian Biology 25: 184-197.

Götmark, F. 1998. The importance of non-reproductive functions of bird colouration, especially anti-predator adaptations. In: Adams, N.J. & Slotow, R.H. (eds) Proc.22 Int. Ornithol. Congr., Durban.

Götmark, F. & Hohlfält, A. 1995. Bright male plumage and predation risk in passerine birds: are males easier to detect than females? Oikos 74: 475-484.

Götmark, F., Post, P., Olsson, A. J. & Himmelmann, D. 1997. Natural selection and sexual dimorphism: sex-biased sparrowhawk predation favours crypsis in female chaffinches. Oikos 80: 540-548.

Grant, P.R. 1965. Plumage and evolution of birds on islands. Systematics Zoology 14: 47-52.

Grant, P.R. & Grant, B.R. 1997. Genetics and the origin of bird species. Proceedings of the National Academy of Sciences USA 94: 7768-7775.

Gray, D.A. 1996. Carotenoids and sexual dichromatism in North American passerine birds. American Naturalist 148: 453-480.

Hamilton, T.H. 1961. On the functions and causes of sexual dimorphism in breeding plumage characteristics of North American species of warblers and orioles. American Naturalist 45: 121-123.

Hamilton, W.D. & Zuk, M. 1982. Heritable true fitness and bright birds: A role for parasites? Science 218: 384-387.

Hill, G. E. 1993a. Male mate choice and the evolution of female plumage colouration in the House Finch. Evolution 47: 1515-1525.

Hill, G.E. 1993b. Geographic variation in the carotenoid plumage pigmentation of male house finches (Carpodacus mexicanus). Biological Journal of Linnean Society 49:63-86.

Hill, G.E. 1994a. Trait elaboration via adaptive mate choice: sexual conflict in the evolution of signals of male quality. Ethology, Ecology, and Evolution 6: 351-370.

Hill, G.E. 1994b. Geographic variation in male ornamentation and female preferences in the house finch: a comparative test of models of sexual selection. Behavioral Ecology 5: 64-73.

Hill, G.E. 1996a. Redness as a measure of the production cost of ornamental colouration. Ethology, Ecology, and Evolution 8: 157-175.

Hill, G.E. 1996b. Subadult plumage in the house finch and tests of models for the evolution of delayed plumage maturation. Auk 113: 858-874.

Hill, G.E. & Brawner, W.R.I. 1998. Melanin-based plumage colouration in the house finch is unaffected by coccidial infection. Proceedings of the Royal Society of London. Series B 265: 1105-1109.

Hingston, R.W.G. 1933.The meaning of animal colour and adornment. London: Edward Arnold.

Höglund, J. 1989. Size and plumage dimorphism in lek-breeding birds: A comparative analysis. American Naturalist 134: 72-87.

Holland, B. & Rice, W.R. 1998. Chase-away sexual selection: Antagonistic seduction versus resistance. Evolution 52: 1-7.

Hutt, F.B. 1949. Genetics of the fowl. McGraw-Hill, New York.

Irwin, R.E. 1994. The evolution of plumage dichromatism in the New World blackbirds: social selection on female brightness? American Naturalist 144:890-907.

Irwin, R.E. 1996. The phylogenetic content of avian courtship display and song evolution. In: Martins, E.P. (ed.) Phylogenies and the comparative method in animal behavior: 234-252

Iwasa, Y. & Pomiankowski, A. 1994. The evolution of mate preferences for multiple sexual ornaments Evolution 48: 853-867.

Johnson, K. 1988. Sexual selection in pinyon jays II: Male choice and female-female competition. Animal Behaviour 36: 1048-1053.

Johnson, S.G. 1991. Effects of predation, parasites, and phylogeny on the evolution of bright colouration in North American male passerines. Evolutionary Ecology 5: 52-62.

Jones, I. L. & Hunter, F.M. 1993. Mutual sexual selection in a monogamous seabird. Nature, Lond. 362: 238-239.

Kear, J. 1970. Adaptive radiation of parental care in waterfowl. In: J.H. Crook (ed.), Social behavior in birds and mammals: essays on the social ethology of animals and man. Academic Press NY: 357-392

Kirkpatrick, M. 1987. Sexual selection by female choice in polygynous animals. Annual Reviews in Ecology and Systematics 18: 43-70.

Kirkpatrick, M. & Barton, N.H. 1997. Evolution of a species' range. American Naturalist 150: 1-23.

Kirkpatrick, M. & Ryan, M.J. 1991. The evolution of mating preferences and the paradox of lek. Nature 350: 33-38.

Kirkpatrick, M., Price, T. & Arnold, S.J. 1990. The Darwin-Fisher theory of sexual selection in monogamous birds. Evolution 44: 180-193.

Kusmierski, R., Borgia, G., Uy, A. & Crozier, R.H. 1997. Labile evolution of display traits in bowerbirds indicates reduced effects of phylogenetic constraints. Proceedings of the Royal Society of London, Series B 264: 307-311.

Lande, R. 1980. Sexual dimorphism, sexual selection and adaptation in polygenic characters. Evolution 34: 292-305.

Leroi, A.M., Rose, M.R. & Lauder, G.V. 1994 What does the comparative method reveal about adaptation? American Naturalist 143: 381-402.

Marchetti, K. 1993. Dark habitats and bright birds illustrate the role of the environment in species divergence. Nature 362: 149-152.

Marchetti, K. 1998. The evolution of multiple male traits in the yellow-browed leaf warbler. Animal Behaviour 55: 361-376.

Martin, T.E. 1995. Avian life history evolution in relation to nest sites, nest predation and food. Ecological Monographs 65: 101-127.

Martin, T.E. & Badyaev, A.V. 1996. Sexual dichromatism relative to nest height and nest predation: contributions of females versus males. Evolution 50: 2454-2460.

Martin, T.E. & Clobert, J. 1996. Nest predation and avian life history evolution in Europe versus North America: a possible role of humans. American Naturalist 147: 1028-1046.

Mayr, E. 1942. Systematics and the origin of species. Dover. New York.

McKitrick, M.C. 1993. Phylogenetic constraints in evolutionary theory: Has it any explanatory power? Annu Rev Ecol Syst 24: 307-330.

McLain, D.K. 1993. Cope's rules, sexual selection, and the loss of ecological plasticity. Oikos 68: 490-500.

McLain, D.K. Moulton, M.P. & Redfearn, T.P. 1995. Sexual selection and the risk of extinction of introduced birds on oceanic islands. Oikos 74: 27-34.

Miles, D.B. & Dunham, A.E. 1993. Historical perspectives in ecology and evolutionary biology: The use of phylogenetic comparative analyses. Annual Reviews in Ecology and Systematics 24: 587-619.

Møller, A.P. 1986. Mating systems among European passerines: A review. Ibis 128: 234-250.

Møller, A.P. & Cuervo, J.J. 1998. Speciation and feather ornamentation in birds. Evolution 52: 859-869.

Møller, A.P. & Birkhead, T.R. 1993. Cuckoldry and sociality; a comparative study of birds. American Naturalist 142: 118-140.

Møller, A.P. & Birkhead, T.R. 1994. The evolution of plumage brightness in birds is related to extra-pair paternity. Evolution 48: 1089-1100.

Møller, A.P. & Pomiankowski, A. 1993. Why have birds got multiple ornaments? Behavioral Ecology and Sociobiology 32: 167-176.

Moreau, R.E. 1960. Conspectus and classification of the Ploceinae weaverbirds. Ibis 102: 298-321.

Omland, K.E. 1997. Examining two standard assumptions of ancestral reconstructions: Repeated loss of dichromatism in dabbling ducks (Anatini). Evolution 51: 636-1646.

Owens, I. P.F. & Bennett, P.M. 1994. Mortality costs of parental care and sexual dimorphism in birds. Proceedings of the Royal Society of London, Series B 257: 1-8.

Owens, I. P.F. & Bennett, P.M. 1995. Ancient ecological diversification explains life-history variation among living birds. Proceedings of the Royal Society of London, Series B 261: 227-232.

Owens, I. P.F. & Bennett, P.M. 1997. Variation in mating systems among birds: ecological basis revealed by hierarchical comparative analysis of mate desertion. Proceedings of the Royal Society of London, Series B 264: 1103-1110

Owens, I.P.F. & Hartley, I.R. 1998. Sexual dimorphism in birds: why are there so many different forms of dimorphism? Proceedings of the Royal Society of London, Series B 265: 397-407

Owens, I.P.F. & Short, R.V. 1995. Hormonal basis of sexual dimorphism in birds: implications for new theories of sexual selection. Trends in Ecology and Evolution 10: 44-47.

Payne, R. B. 1984.Sexual selection, lek and arena behavior, and sexual size dimorphism in birds. Ornithological Monographs 33: 1-51.

Peterson, A.T. 1996. Geographic variation in sexual dichromatism in birds. Bulletin of British Ornithological Club 116: 156-172.

Petrie, M. & Kempenaers, B. 1998. Extra-pair paternity in birds: explaining variation between species and populations. Trends in Ecology and Evolution 13: 52- 58.

Price, T. 1996. An association of habitat with colour dimorphism in finches. Auk 113: 256-257

Price, T. 1998. Sexual selection and natural selection in bird speciation. Philosophical Transactions of the Royal Society London Series B 353: 251-260.

Price, T. & Birch, G.L. 1996. Repeated evolution of sexual colour dimorphism in passerine birds. Auk 113: 842-848.

Price, T. & Pavelka, M. 1996. Evolution of a colour pattern: history, development, and selection. Journal of evolutionary Biology 9: 451-470.

Promislow, D.E.L., Montgomerie, R.D. & Martin T.E. 1992. Mortality costs of sexual dimorphism in birds. Proceedings of the Royal Society of London, Series B 250: 143-150.

Promislow, D.E.L., Montgomerie, R.D. & Martin, T.E. 1994. Sexual selection and survival in North American waterfowl. Evolution 48: 2045-2050.

Prum, R. 1994. Phylogenetic analysis of the evolution of display behavior in the neotropical manakins (Aves: Pipridae). Ethology 84: 202-231.

Prum, R. 1997. Phylogenetic tests of alternative intersexual selection mechanisms: trait macroevolution in a polygynous clade (Aves: Pipridae). American Naturalist 149: 668-692.

Prum, R. 1998. Structural colours of birds: Anatomy, production mechanisms, and evolution. In: Adams, N.J. & Slotow, R.H. (eds) Proc.22 Int. Ornithol. Congr., Durban.

Savalli, U.M. 1995. The evolution of bird coloration and plumage elaboration: A review of hypothesis. Current Ornithology 12: 141- 190.

Schluter, D. & Price, T. 1993. Honesty, perception and population divergence in sexually selected traits. Proceedings of the Royal Society of London, Series B 253, 117-122.

Scott, D.K. & Clutton-Brock, T.H. 1989. Mating systems, parasites and plumage dimorphism in waterfowl. Behavioral Ecology and Sociobiology 26: 261-273.

Sheldon, F.H. & Whittingham, L.A. 1997. Phylogeny in studies of bird ecology, behavior, and morphology. In: Mindell, D.P (ed.) Avian Molecular Evolution and Systematics. Academic Press: 279-299.

Shutler, D. & Weatherhead, P. J. 1990. Targets of sexual selection: song and plumage of wood warblers. Evolution 44: 1967-1977.

Slagsvold, T., Dale, S. & Kruszewicz, A. 1995. Predation favours cryptic colouration in breeding male pied flycatchers. Animal Behaviour 50: 1109-1121.

Slagsvold, T. & Lifjeld, J. 1997. Incomplete knowledge of male quality may explain variation in extra-pair paternity in birds. Behaviour 134: 353-371.

Soler, J.J., Møller, A.P. & Soler, M. 1998. Nest building, sexual selection and parental investment. Evolutionary Ecology 12: 427-441.

Sorci, G., Møller, A.P. & Clobert, J. 1998. Plumage dichromatism of birds predicts introduction success in New Zealand. Journal of Animal Ecology 67: 263-269.

Trail, P.W. 1990. Why should lek-breeders be monomorphic? Evolution 44: 1837-1852.

Trivers, R.L. 1972. Parental investment and sexual selection. In: Cambell, B. (ed.). Sexual Selection and the Descent of Man, 1871-1971. London; Heinemann: 136-179.

Verner, J. & Willson, M.F. 1969. Mating systems, sexual dimorphism and the role of male North American passerine birds in the nesting cycle. Ornithological Monographs 9: 1-76

Wallace, A.R. 1889. Darwinism. Macmillan, London.

Webster, M.S. 1992. Sexual dimorphism, mating system and body size in New World blackbirds (Icterinae). Evolution 46: 1621-1641.

Westneat, D.F. & Sherman, P.W. 1997. Density and extra-pair fertilizations in birds: a comparative analysis. Behavioral Ecology and Sociobiology 41: 205-215.