Behavioral Ecology Vol. 15 No. 3: 380-389
Behavioral Ecology vol. 15 no. 3 © International Society for Behavioral Ecology 2004; all rights reserved
Mutual ornamentation, sexual selection, and social dominance in the black swan
a Department of Zoology, University of Melbourne, Melbourne, VIC 3010, Australia, b Ballarat Field Naturalists, PO Box 328W, Ballarat West 3350, Australia, and c Centre for Ecological and Evolutionary Studies, University of Groningen, PO Box 14, 9750 AA Haren, The Netherlands
Address correspondence to K. Kraaijeveld, who is now at the Department of Biology, Galton Laboratory, Wolfson House, University College London, 4 Stephenson Way, London NW1 2HE, UK. E-mail: k.kraaijeveld{at}ucl.ac.uk.
Received 9 January 2003; revised 2 May 2003; accepted 19 June 2003.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
We investigated the adaptive significance of a sexually monomorphic ornament in the black swan Cygnus atratus. Both sexes grow curled feathers on their wings (range 7–22 curled feathers per wing), which are displayed prominently in a range of social interactions. The number of curled feathers increased until the birds reached sexual maturity (at 2 years of age) but did not vary with age thereafter. We found evidence for both sexual and social functions of the ornament. Paired, mature individuals of both sexes had higher numbers of curled feathers than unpaired, mature birds, and individuals paired assortatively with respect to curled feather number, suggesting the feathers may be involved in mutual sexual selection. More ornamented individuals were dominant in agonistic interactions with birds of the same sex and pairing status. Highly ornamented pairs were also more likely to maintain extended tenancy of preferred cygnet feeding areas, which resulted in improved offspring survival. The curled feathers thus appear to function as a signal of social dominance, which is highly correlated with reproductive success and is therefore a reliable signal of parental quality in mate choice.
Key words: black swans, Cygnus atratus, dominance, mutual sexual selection, ornamentation, signaling, social selection.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Empirical studies of sexual selection during the past few decades have focused predominantly on organisms with sexually dimorphic traits, in which trait elaboration is typically restricted to males. There is now abundant evidence that male–male competition and female choice can lead to extravagant ornamentation among males (Andersson, 1994
). Yet in many species, ornamental traits occur in both sexes and are elaborated to a similar degree in males and females. Darwin (1871) proposed that such mutual ornamentation was a genetically correlated consequence of intense sexual selection on males. Under this scenario, traits selected for in males are also expressed in females simply because they also carry the genes for these traits (Lande, 1980, 1987; Lande and Arnold, 1985). However, empirical evidence in support of a genetic correlation in a plumage trait is limited to one study (Roulin et al., 2001). Furthermore, recent phylogenetic work has demonstrated that evolutionary transitions between dimorphism and monomorphism have occurred frequently and in both directions (Wiens, 2001), suggesting that such changes are relatively unconstrained by genetic correlations.
An alternative explanation for similar ornamentation in males and females is that sexual selection acts on both sexes. This idea was first proposed by Huxley (1914), who coined the term "mutual sexual selection." Recent models have demonstrated that mutual mate choice resulting in mutual ornamentation is a theoretically plausible process (Johnstone, 1997; Johnstone et al., 1996; Kokko and Johnstone, 2002). Experimental studies of crested auklets Aethia cristatella have provided evidence for both mutual mate choice (Jones and Hunter, 1993) and competition over access to mates in both sexes (Jones and Hunter, 1999).
Although ornamentation is typically associated with the process of sexual selection, there are also numerous examples of ornamental signals for which a sexual role is unlikely (Zahavi, 1991). Examples from avian studies include bright coloration restricted to juveniles (which may induce parents to feed the young; Lyon et al., 1994), song and displays of passerines on winter territories far from the breeding grounds (Zahavi, 1972), and extravagant group displays that function in social circumstances unrelated to sex (Zahavi, 1990). Perhaps the best studied examples of avian signaling in which a sexual context is absent or indirect are the badges of status, plumage signals of dominance in social competition over resources such as food and shelter (Butcher and Rohwer, 1989; Savalli, 1995; Whitfield, 1987).
The same signals that are used in competition over access to mates may be used in competition over access to other resources, and there is therefore no need to assume that such signals will be selected by different selection processes (Zahavi, 1991). However, because sexual selection is defined in a strict sexual context (Andersson, 1994; Darwin, 1871), the evolution of signals, both sexual and nonsexual, may be better understood in a wider context, such as signal selection (Zahavi, 1991) or social selection (Crook, 1972; Tanaka, 1996; West-Eberhard, 1983, 1984, 1991). Signal selection and social selection refer to the same process and can be defined as "the selective force that arises when a signal influences the fitness of signallers or both the fitness of signallers and receivers" (Tanaka, 1996: 512).
West-Eberhard (1983) suggested that bright coloration in both sexes of many birds may be attributed to the fact that, in such species, both sexes perform aggressive displays. So far, the only avian ornament that has been shown to be involved in aggressive displays in both sexes is the crest of the crested auklet (Jones and Hunter, 1999), but the same has been suspected in monochromatic species of hummingbirds (Bleiweiss, 1992). For such aggressive behavior to be adaptive, it must have a positive effect on survival or reproductive success in both sexes. This has not been demonstrated for any mutually ornamented species.
In this study we considered both potential sexual and nonsexual functions for a feather ornament found in both sexes of the black swan Cygnus atratus. Males and females have curled ornamental feathers on their wings, which are prominently displayed during agonistic encounters between individuals and as part of the triumph ceremony (Kraaijeveld and Mulder, 2002). These curled feathers have no obvious advantage in terms of natural selection and thus qualify as ornaments (Amundsen, 2000). Our aims were to (1) document variation in the number of curled feathers among and within individuals in relation to sex, age, body size, and condition, (2) test for a relation between ornamentation and social pair formation, and (3) assess whether the number of curled feathers is related to social dominance, access to feeding sites, and reproductive success. To our knowledge, this is the first study to investigate the relation between ornamentation and reproductive success in a species with an ornament that is expressed equally in males and females.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Study species
Black swans are large (4–7 kg), long-lived, and socially monogamous waterfowl that occupy shallow wetlands, often at high densities (Frith, 1982
). Both sexes have black plumage with conspicuous black curled tertials and wing coverts. Both sexes participate in incubation (Brugger and Taborsky, 1994) and brood defense. Black swans have a low divorce rate (6%; Kraaijeveld, unpublished data) and once formed, pairs tend to remain together. In captivity, black swans may breed successfully from 2 years of age (Braithwaite, 1981). The proportion of adults breeding in any year is highly variable, and only one successful nest is produced per pair per season.
Capture and measurement of birds
This study was conducted between 1999 and 2001 at Lake Wendouree, Ballarat, Australia (37°33'S, 143°49'E), a permanent lake of 238 ha supporting a population of about 170 black swans year-round. We lured the swans from the water with bread morsels and captured them by hand. Adults were caught during the breeding seasons (July–November) of 1999–2001 and during the prebreeding seasons (May–June) of 2000 and 2001. We fitted adults with metal bands (Australian Bird and Bat Banding Scheme) and individually colored and numbered plastic leg bands. The plastic bands could be read with binoculars from a distance of up to 30 m (when the bird was on land or up-ending to feed on submerged aquatic plants) or when the bird was in clear water within 5 m of the shore. We censused all banded individuals twice weekly during the breeding season (July–November) and monthly during the remainder of the year by walking around the perimeter of the lake (6 km) and marking the location of banded individuals on a map.
For each captured adult, we measured tarsus length (with callipers to the nearest 1 mm), bill and total head length (callipers; nearest 0.1 mm), wing length (butt-ended ruler; nearest 1 cm) body mass (spring balance; nearest 100 g), and the number of curled feathers on each wing (described below).
Although most adults can be sexed on behavior, size, and cloacal inspection, we confirmed the sex of all individuals using a molecular technique (Griffith et al., 1998). Captured birds were assigned to one of three age classes. First-year black swans are easily recognized by their gray plumage, dull bill color, and black tips of their primaries (Marchant and Higgins, 1990). Most second-year individuals retain black tips in some of their primaries (10–16 known 2-year-old birds had black tips of some primaries, compared to 0 of 30 adults known to be 
3 years old). Individuals in black plumage with pure white primaries were at least 2 years or older and were classed as adults.
During the breeding season, we searched all available breeding habitat in a canoe once per week and plotted the location of nests on a map. We recorded the identity of the attending adults at each nest and the clutch size. Cygnets stayed in the nest for about 24 h after hatching and were then taken to the shore by their parents. Some cygnets were caught on the nest, but most were caught on shore, generally within the first week after hatching. We rounded up a brood together with their parents and caught all the cygnets simultaneously. At least one parent was also caught and retained while the cygnets were processed to ensure that cygnets were not separated from their parents. Cygnets were too small to be banded when first caught, so they were marked with a unique combination of small holes in the webs between their toes with a standard leather punch (see Braithwaite, 1981). These holes closed up as the cygnets grew to adult size, and there was no indication that this procedure had any harmful effect on the cygnets' health. The scars left by these punches were used for subsequent individual identification. We recaptured cygnets surviving to 8 months and subjected them to the same procedures as adults. Cygnets that had disappeared before this point were assumed to have died. This assumption is supported by a study of black swans in New Zealand in which 95–100% of ringed black swans recovered during their first year of life (n = 1792) were recovered on their natal lake (Williams, 1977).
All procedures used during this study were approved by the Animal Experimentation Ethics Committee of the University of Melbourne.
Ornamentation
We scored the degree of ornamentation for captured birds by counting the number of curled wing-feathers (tertials and wing coverts). A feather was defined as curled when part of at least one of the vanes reached above the central feather-shaft (rachis) when viewed from the side (Figure 1a). To determine whether this method was repeatable, we compared the feather counts of 22 individuals (13 males and 9 females) captured both in May and June 2000. The intraclass correlation coefficient (rI; Zar, 1999) was positive and significantly different from zero (n = 22, rI =.89, F = 16.88, p <.001), indicating that variation within individuals was significantly smaller than variation between individuals. The rI value is high compared to those reported in similar studies of feather ornaments (e.g., 0.41, 0.45, 0.52, 0.58, and 0.74 for different ornaments of least auklet Aethia pusilla; Jones and Montgomerie, 1991).
|
We assessed possible changes in feather number with age in two ways. We first calculated mean feather counts for different age classes using individuals of known age (banded as nestlings or aged using the criteria listed above). Each individual contributed to this analysis only once. For individuals that were measured twice, we used the most recent measurement to maximize the sample size for older age classes. We were most interested in older age classes because first-year birds were excluded from all further analyses a priori. Some 2-year-old individuals with all-white primaries may have been falsely classified as adults. To examine whether this was likely to bias our results, we inspected known second-year birds for a correlation between the number of primaries with black spots and the number of curled feathers. We then compared repeated measurements from the same individuals across years.
We examined whether ornamentation was correlated with the body size or body condition of the individual. Tarsus length remains fixed when a bird reaches adult size and is considered the best univariate measure of body size (Rising and Somers, 1989). Therefore, the residuals from the regression of body mass on tarsus length were used as an index for body condition (Packard and Boardman, 1987).
Pairing status
If curled feathers are used in mate choice, we would expect to find systematic differences in ornamentation between paired and unpaired individuals, and possibly a correlation between the degree of ornamentation of members of a social pair. We identified pairs from census data. The male and female of a social pair maintain close proximity throughout the year, and social pairs were thus easily recognizable. Individuals were classed as paired if they attempted to breed together or if they were seen together on consecutive census dates over at least 3 months. This eliminates the possibility of mistaking trial liaisons (Choudhury and Black, 1993) between unpaired birds for long-term pairing because such associations never lasted for more than 3 weeks.
Social interactions
Black swans raise their wings and display the curled feathers in a range of social contexts, but especially during agonistic interactions. We identified the individuals involved and quantified the outcomes of spontaneous agonistic interactions as well as those induced by feeding during the prebreeding seasons (April–July) of 2000 and 2001. Agonistic interactions included threat displays (during which competing individuals raised their wings; Figure 1b), chases, and pecks. We included only interactions with an unambiguous winner in the analysis. An individual was considered to have won the interaction if its opponent flew off or walked away with lowered wings. The observations were conducted blind to the sex, pairing status, and number of curled feathers for each individual, which were determined retrospectively from capture records. To avoid pseudoreplication, we included each dyad into the analysis only once. In all cases for which multiple interactions between the same dyad were observed (n = 32), the same bird was the winner in all interactions.
In addition to opportunistic quantification of agonistic encounters, we conducted focal observations on pairs during the nonbreeding season (May–July 2000) and when they were accompanied by dependent cygnets during the breeding season (July–November 1999 and 2000). Each observation lasted 1 h, and each pair was observed once. We recorded the number of agonistic interactions in which one or both of the participants raised their wings; whether the male, female, or both individuals were involved in the interaction; and the duration for which individuals kept their wings raised. Individuals often raised their wings in anticipation of an interaction, so we report the total amount of time spent in threat display posture during the observation, rather than per interaction.
Residency and cygnet foraging
After hatching, parents led the precocial cygnets to feeding grounds on lawns surrounding the lake or shallow water close to the shore. Frequent agonistic interactions occurred between families over access to these feeding sites. We assessed the success of pairs in competing over these feeding areas by comparing the duration of their tenancy of these areas. Certain pairs were consistently seen at the same locations, while others were seen in different locations each time. We assumed that pairs with prolonged tenancy of particular patches were behaviorally dominant to those that continually moved between sites, and we classified pairs according to their tenure. Pairs that were found along the same 500-m stretch of foreshore for 3 or more subsequent weeks during the cygnet-raising period were classed as resident. The remaining pairs were classed as itinerant.
We examined whether residency was related to ornamentation and reproductive success. We also estimated the interruption cost of agonistic encounters between families on the cygnets' opportunity to feed by conducting focal observations on dependent cygnets. Each observation lasted 1.5 h, during which we recorded the proportion of time the cygnet spent feeding. In families with more than one cygnet, we recorded the time during which at least one of the cygnets was feeding, as the others would also have had the opportunity to feed during this time.
Statistical analyses
We only compared measurements of different parameters if they had been obtained in the same season. Morphological measurements other than count data that did not deviate from normality were analyzed using parametric tests. We analyzed count data (curled feathers, clutch size and brood size) using nonparametric tests. The distributions of frequency and length of agonistic interactions deviated from normality and were also analyzed using nonparametric tests. Because residency status is an attribute of a pair, we used mid-pair values (calculated by averaging the male and female measurements) for analysis of traits in relation to residency. We analyzed residency status using binary logistic regression to test whether the effects of brood size and ornamentation were independent of each other. The test statistic in these analyses is the ratio of the coefficient over its standard error (t ratio; Systat, 1997). Cygnet survival to independence was analyzed using a generalized linear model with binomial error distribution and a logit link function. This analysis used the number of surviving cygnets per brood as the dependent variable and initial brood size as the binomial denominator. We assessed the significance of the independent variables from the change in deviance when the variable was removed from the model, which approximates a chi-square distribution (Beath, 2001). We used SYSTAT 7 (Systat, 1997) for all analyses, except generalized linear models, which were performed in GLMStat (Beath, 2001). Means are presented ± standard deviations. For non-normally distributed data, we present the median and the quartile deviation (Zar, 1999).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Curled feathers in relation to age and sex
The mean numbers of curled feathers for known-age (including those aged using the criteria described in the Methods) individuals of both sexes are illustrated in Figure 2. There was no significant correlation between the number of curled feathers and the number of primaries with black spots for either sex (males, rs = –.28, n = 10, p =.56; females, rs = –.20, n = 6, p =.90), and there is therefore no reason to assume that any second-year individuals with all-white primaries had unusually small or large numbers of curled feathers. The number of curled feathers differed significantly between females of different age classes (Kruskall-Wallis Z = 34.76, n = 99, p =.001; Figure 2). Second-year females had more curled feathers than first-year females but fewer than adult females (post hoc tests; Figure 2). In contrast, we found no differences between males of different ages (Kruskall-Wallis Z = 3.50, n = 87, p =.17; Figure 2). However, the sample for first-year males was small and highly variable, which may have masked effects in this age group. Males in their first and second year had more curled feathers than females of equivalent age (Figure 2). However, there was no detectable sexual dimorphism in curled feather number among adults (Table 1, Figure 2).
|
|
The lack of sexual dimorphism in curled feather number is remarkable, given that significant sexual dimorphism exists in other morphological traits (Table 1). Males were 7–10% larger than females in all structural measurements and 26% heavier (Table 1). The curled feather ornament showed more variability than other morphological traits, with a coefficient of variation of 22–24%, compared with 3–10% for other traits (Table 1).
Repeated measurements on the same individuals in consecutive years showed that the number of curled feathers increased by 40–68% between the first and second year in all individuals of both sexes (Figure 3a; Wilcoxon Z = 2.67, n = 9, p =.008). However, there was no clear pattern of increase for males that were measured in both their second and third year (Figure 3a; Wilcoxon Z = –0.11, n = 6, p =.92; no information for third-year females available), suggesting that the number of curled feathers does not increase with age once individuals reach maturity. The differences in numbers of curled feathers between second and third year were <10% for all three individuals (Figure 3a). Furthermore, there was no indication that the number of curled feathers increased in older birds (Figure 3a; Wilcoxon Z = 1.25, n = 31, p =.21). Of 31 adults with changed curled feather number between years, 16 increased and 11 decreased (
= 0.03, p =.86). Adults counted at intervals of 1.5 or 2 years showed more variability (Figure 3b). However, there was no consistent trend with age (Wilcoxon Z = 0.98, n = 35, p =.33); 21 birds showed an increase in curled feather number, whereas 13 showed a decrease ( = 1.88, p =.17). Finally, the two oldest individuals in our study population (a 10-year-old male and 11-year-old female) did not have unusually large numbers of curled feathers (33 and 16, respectively).
|
Curled feathers, body size, and condition
In males, the number of curled feathers was positively correlated with body condition (Spearman rank correlation rs =.35, n = 67, p =.009), but not with tarsus length (rs =.08, n = 67, p =.66). In females, neither correlation was significant (rs = –.09, n = 80, p =.59 and rs = –.07, n =80, p =.57, respectively). Individual changes in the number of curled feathers between years did not correspond to changes in body mass over the same period for adults of either sex (Spearman rank correlation; males: rs = –.04, n = 30, p =.88; females: rs = –.30, n = 30, p =.22).
Curled feathers and social pair formation
Paired male and female swans had significantly more curled feathers than adult (>2 years old) unpaired birds of the same sex (Table 2). Other morphological traits that might affect pairing status (body size, body mass, and body condition) showed no significant differences for either sex after Bonferroni correction (Table 2). The number of curled feathers was also positively correlated between members of a social pair (Figure 4).
|
|
Social competition: agonistic interactions, residency status, and reproductive success
Details of 153 interactions between individuals with different numbers of curled feathers were recorded (Table 3). A log-linear model showed that sex, pairing status, and the number of curled feathers all significantly influenced the outcome of agonistic interactions between adults. Males were more likely to win interactions than females (Table 3;
= 55.17, p =.0001). Paired birds were more likely to win than unpaired birds (Table 3; = 44.74, p =.0001). The individual with the largest number of curled feathers was more likely to win than the individual with fewer curled feathers (Table 3; = 15.32, p =.004). None of the interaction terms was significant (all p >.30). Of 40 interactions matched for both sex and pairing status, 28 (70%) were won by the individual with the largest number of curled feathers (Figure 5).
|
|
We examined the contribution of paired males and females to agonistic interactions by comparing the frequency of agonistic displays and the time spent in threat display posture during the prebreeding and cygnet-rearing periods. Males were involved in more agonistic interactions and spent more time in threat display posture than females during the prebreeding period (Table 4). During the cygnet-rearing period, only the time spent in threat display posture differed between the sexes (Table 4). The frequency and duration of threat displays did not differ between the prebreeding and cygnet-rearing periods for either sex (Mann-Whitney U tests, all p >.15).
|
Resident pairs had significantly more curled feathers than itinerant pairs but did not differ in body size or mass (Table 5). Large broods are known to enhance the dominance status of breeding pairs in geese (Loonen et al., 1999
). However, the average number of curled feathers of the pair remained a significant predictor for residency when brood size was controlled for in a binary logistic model (t ratio = 2.00, n = 22, p =.04). Addition of tarsus length and body condition reduced the overall significance of the model.
|
Clutch sizes of resident and itinerant pairs did not differ significantly, but resident pairs had significantly larger broods than itinerant pairs (Table 5). This could be due to differences in hatching success and/or early cygnet mortality. Cygnet survival to 8 months of age was more than twice as high for resident pairs compared to itinerant pairs (Table 5). Although resident pairs had higher numbers of curled feathers and higher cygnet survival, this did not result in a statistically significant correlation between number of curled feathers and cygnet survival (Figure 6). In a generalized linear model, residency status had a near-significant effect on cygnet survival (
deviance = 3.30, n = 20, p =.07), but the number of parental curled feathers did not (deviance = 0.18, n = 20, p =.67). When only residency status was included, the sample size was larger and the effect highly significant (deviance = 15.94, n = 64, p <.0001).
|
Cygnets of resident families spent more time feeding than those of itinerant families (resident: 58.19 ± 2.21 min/1.5h, n = 26; itinerant: 45.91 ± 8.16 min/1.5h, n = 6). In a backward stepwise general linear model, residency status and time of day significantly affected the cygnets' feeding time (F1,29 = 5.57, p =.03 and F1,29 = 6.10, p =.02, respectively, n = 32). Date, brood size, and the interaction between time of day and residency status did not significantly affect the cygnets' feeding time (all p >.58). Cygnets of resident pairs also tended to be interrupted less often than those of itinerant pairs (resident: 4.90 ± 2.45, n = 26, itinerant: 6.00 ± 1.27, n = 6; Mann-Whitney U test U = 174, p =.08).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Black swans are sexually dimorphic for most morphological traits, but the curled wing feathers used by both male and female swans in social displays are a genuinely monomorphic ornamental trait, expressed equally in both sexes. The curled feathers also show more variation between individuals than other morphological traits in the black swan, a pattern typical of ornamental traits in birds (Alatalo et al., 1988
; Blanco and De la Puente, 2002; Jones and Montgomerie, 1992; Jones et al., 2000). We found that the degree of feather elaboration (measured as the number of curled feathers) correlated with pairing status, social dominance, and, indirectly, reproductive success. Ornamented individuals thus are socially dominant and may be preferred as mating partners.
The full complement of curled feathers is acquired by the second year of age in males, which coincides with the age of first breeding in captive swans (Braithwaite, 1981). Second-year females had fewer curled feathers than adult females and may thus take slightly longer to reach full adult ornamentation. However, ornamentation did not increase with age in adult (3 years and older) birds. Young male barnacle geese Branta leucopsis with more adultlike plumage have more trial partners and higher social ranking than those with more juvenile plumage (Van der Jeugd and Blaakmeer 2001). However, we found no indication that second-year black swans that had retained more juvenile plumage characters (black spots in primaries) also had fewer curled feathers.
Social pair formation
Paired individuals of both sexes had significantly more curled feathers than unpaired individuals, suggesting that this ornament may play a role in mutual mate choice. Our finding that the number of curled feathers is correlated between the members of a social pair further strengthens this view. Little is known about courtship and the process of pair formation in this species. In geese, the triumph ceremony is known to play an important role in pair formation (Bigot et al., 1995; Choudhury and Black, 1993), and there is anecdotal evidence that the same is true for the black swan (Kraaijeveld and Mulder, 2002). The curled feathers feature prominently in social displays such as the triumph ceremony in this species and could conceivably be used in mate assessment. Pairing in the closely related mute swan C. olor is known to be assortative with respect to age (Perrins and McCleery, 1997), and this may also be true for the black swan. However, there was no evidence for a strong effect of age on the number of curled feathers. Thus, age-assortative mating is unlikely to explain the correlation between pairing and ornamentation. Nevertheless, other factors such as similar environmental conditions experienced by members of a pair (e.g., food availability) may lead to similar ornamentation and require experimental evaluation.
The direction of sexual selection depends on the relative roles of the sexes in parental care (Kokko and Monaghan, 2001; Parker and Simmons, 1996; Trivers, 1972). Generally, the sex providing less care should be more active in competition over mates (Trivers, 1972). In the black swan, males are more active in competition over feeding areas, which is an important part of the parental effort. Males also provide most of the incubation effort (Brugger and Taborsky, 1994). Competition among females over high-quality males might thus be more intense than competition among males over females, as has been described in another mutually ornamented species, the moorhen Gallinula chloropus (Petrie, 1983).
Barnacle geese compare up to six potential partners before a stable pair bond is formed (Choudhury and Black, 1993; Van der Jeugd and Blaakmeer, 2001), and a similar system may operate in the black swan. Although individuals may continue comparing potential partners even after pair formation has taken place (Black, 1996), low divorce rate (6% in our study population; Kraaijeveld, unpublished data) and high survival should limit the number of realized opportunities for mate choice. It therefore appears likely that social factors in addition to sexual competition also select for mutual ornamentation in the black swan.
Social dominance
The ornamental feathers are most conspicuously displayed during agonistic displays and during the triumph ceremony. Both function as threat signals toward conspecifics (Kraaijeveld and Mulder, 2002). Our finding that individuals with many curled feathers are dominant over same-sex individuals with fewer curled feathers in agonistic interactions suggests that the curled feathers play a role in social competition over resources. In addition to potential mates, such resources may include food and nesting space. In our swan population, prolonged access to suitable feeding habitat for the cygnets ("residency") is clearly an important resource that directly affects the survival prospects of cygnets.
Resident pairs had significantly more curled feathers than itinerant pairs, probably because curled feathers signal competitive ability and thus improve an individual's chances of gaining access to preferred areas and maintaining tenancy of these areas for a prolonged period. Competitive ability may depend on several factors, including body size, body condition, age, breeding experience, and motivation. We found a correlation between the number of curled feathers and body condition in males, suggesting possible condition dependence of the ornament. Correlations between body condition and the expression of a mutual ornament have been reported in one or both sexes for a number of species (magpie Pica pica: Blanco and De la Puente, 2002; inca tern Larosterna inca: Velando et al., 2001; crested auklet: Jones et al., 2000). However, we failed to obtain the same result when repeated measurements of the same individual were considered, and we therefore interpret this result cautiously.
Males are more active than females in social competition over resources other than mates. The role of social competition over nonsexual resources in the maintenance of the curled feather ornament may therefore be stronger in males than in females. As with the potential sex bias in sexual selection, this suggestion parallels findings in the moorhen, in which males compete over territories (Petrie, 1984). Both sexes in the moorhen use the red frontal shield as an indication of dominance (Petrie, 1984, 1988). In crested auklets, males are also the more aggressive sex, whereas females may be more active in defending mates (Jones and Hunter, 1999). Given these apparently different balances of selective forces between males and females, it is puzzling why ornament expression remains the same in males and females of black swans, moorhens, and crested auklets.
Status signaling may be particularly advantageous in high population densities, where individuals frequently encounter unfamiliar individuals (Rohwer, 1975; Whitfield, 1987). It is interesting to note in this respect that the black swan is both the only species in its genus with a feather ornament and also the only species that regularly breeds in colonies (Scott and the Wildfowl Trust, 1972). However, several other swan species have elaborate bill ornaments (e.g., black-necked swan C. melanocoryphus; mute swan) and threat displays that involve lifting of the wings (mute swan; Johnsgard, 1965), and the absence of feather ornaments in these species might reflect a lack of selectable variation.
Reproductive success
For social competition to lead to selection, it must have an effect on the fitness of competing individuals (Tanaka, 1996). Cygnet survival of pairs that are able to maintain residency in a patch of habitat on the lake foreshore was more than twice as high as that of itinerant pairs. One likely factor contributing to this improved survival is our finding that cygnets of resident pairs spent more time feeding than those of itinerant pairs. Similar benefits accrue to juvenile barnacle geese and bewick's swans C. bewicki with dominant parents (Black and Owen, 1989; Scott, 1980). Aggressive parents may also be able to provide better protection against predators. Nevertheless, the causality may also be reversed; in barnacle geese, larger families are more dominant (Loonen et al., 1999), and high cygnet survival may help parents to maintain residency. In black swans, however, the number of curled feathers was a better predictor of residency than was brood size.
Honesty of the signal
The prominence of the ornament in agonistic displays, the correlation between within-sex social dominance and ornamentation, and the association between ornamentation and residency status all suggest that the display feathers function as a badge of status. Such signals allow dominant individuals access to resources, while minimizing their aggressive interactions with subordinate individuals (Dawkins and Krebs, 1978; Rohwer, 1975; Rohwer and Rohwer, 1978). For such a signaling system to be evolutionarily stable and prevent poor-quality individuals from cheating, it must be costly (Pomiankowski et al., 1991; Zahavi and Zahavi, 1997). The physiological or energetic costs of growing and maintaining the display feathers (extra resources allocated to make the feather curled, aerodynamic costs, etc.) are likely to be trivial. However, the ornament could carry a substantial social cost if dominant individuals frequently test the signaling system ( Johnstone and Norris, 1993; Møller, 1987; Whitfield, 1987). Two recent studies found evidence for such dominance testing (Molles and Vehrencamp, 2001; Parker and Ligon, 2002), but the generality of these findings remains to be established. Alternatively, high levels of ornament expression may depend high levels of circulating testosterone, which may incur secondary costs such as high metabolic rate (Buchanan et al., 2001), high ectoparasite load (Poiani et al., 2000), and low immunocompetence (Evans et al., 2000).
Interactions between selective forces
Mate choice, competition over mates, and competition over other resources are not mutually exclusive processes, and their effects are likely to be additive (West-Eberhard, 1983). In species with biparental care, an individual's own attractiveness as a mate is expected to affect its access to potential mates, which should result in assortative pairing (the differential access hypothesis; Burley, 1986). It seems likely that as biparental care evolved from female-only care, males displayed increasing selectivity of mates (Burley and Johnson, 2002). In the black swan, cooperative parental defense of cygnet feeding areas has a positive effect on offspring survival and therefore on fitness of both members of the pair. Breeding adults thus gain fitness benefits from pairing with a socially dominant partner, and this may explain why ornament-assortative pairing occurs in the black swan. In crested auklets, individuals with longer crests are dominant over individuals with shorter crests and are also preferred as mates (Jones and Hunter, 1999). Nonsexual competition (e.g., over nesting sites and chick defense), is also likely to be important in crested auklets (Fraser et al., 2002; Jones and Hunter, 1999). Nevertheless, the relative importance of sexual versus nonsexual competition in maintaining a mutual ornament may differ across species.
Our study demonstrates reproductive benefits related to the expression of a mutual ornament. The proximate mechanism through which this benefit is achieved is the signaling of social dominance. Individuals benefit from pairing with a socially dominant partner, and the curled feathers therefore play a role in mate choice. The evolution and maintenance of the ornament is therefore not simply the consequence of sexual selection but should be considered in the broader context of social selection.
|
ACKNOWLEDGEMENTS |
|---|
We thank Tania Billing, Brian, Kevin, Joan Andrews, and Peter Carew for their help in the field. Femmie Kraaijeveld-Smit and Therésa Jones provided helpful comments on earlier drafts of this paper. We also thank the Department of Natural Resources and Environment and the Ballarat City Council for permission to conduct this work at Lake Wendouree. This research was supported by grants from the Holsworth Wildlife Research Fund and the Stuart Leslie Bird Research fund.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Alatalo RV, Höglund J, Lundberg A, 1988. Patterns of variation in tail ornament size in birds. Biol J Linn Soc 34:363-374.[CrossRef]
Amundsen T, 2000. Female ornaments: genetically correlated or sexually selected? In: Animal signals: signalling and signal design in animal communication (Espmark Y, Amundsen T, Rosenqvist G, eds). Trondheim, Norway: Tapir Academic Press; pp. 133–154.
Andersson M, 1994. Sexual selection. Princeton, New Jersey: Princeton University Press.
Beath KJ, 2001. GLMStat user manual, version 5.5. Available at: www.ozemail.com.au/
kjbeath/glmstat.html.
Bigot E, Hausberger M, Black JM, 1995. Exuberant youth: the example of triumph ceremonies in barnacle geese (Branta leucopsis). Ethol Ecol Evol 7:79-85.
Black JM, 1996. Partnerships in birds; the study of monogamy. Oxford: Oxford University Press.
Black JM, Owen M, 1989. Agonistic behaviour in barnacle goose flocks: assessment, investment and reproductive success. Anim Behav 37:199-209.[CrossRef]
Blanco G, De la Puente J, 2002. Multiple elements of the black-billed magpie's tail correlate with variable honest information on quality in different age/sex classes. Anim Behav 63:217-225.[CrossRef]
Bleiweiss R, 1992. Widespread polychromatism in female sunangel hummingbirds (Heliangelus: Trochilidae). Biol J Linn Soc 45:291-314.[CrossRef]
Braithwaite LW, 1981. Ecological studies of the Black Swan II. Colour and plumage changes, growth rates, sexual maturation and timing and frequency of breeding. Aust Wildl Res 8:121-133.
Brugger C, Taborsky M, 1994. Male incubation and its effect on reproductive success in the black swan. Ethology 96:138-146.[Web of Science]
Buchanan KL, Evans MR, Goldsmith AR, Bryant DM, Rowe LV, 2001. Testosterone influences basal metabolic rate in male house sparrows: a new cost of dominance signalling? Proc R Soc Lond B 268:1337-1344.[Medline]
Burley N, 1986. Sexual selection for aesthetic traits in species with biparental care. Am Nat 127:415-445.[CrossRef][Web of Science]
Burley NT, Johnson K, 2002. The evolution of avian parental care. Phil Trans R Soc Lond B 357:241-250.
Butcher GS, Rohwer S, 1989. The evolution of conspicuous and distinctive coloration for communication in birds. Curr Ornithol 6:51-108.
Choudhury S, Black JM, 1993. Mate-selection behaviour and sampling strategies in geese. Anim Behav 46:747-757.[CrossRef][Web of Science]
Crook JH, 1972. Sexual selection, dimorphism, and social organization in the primates. In Sexual selection and the descent of man 1871–1971 (Campbell B, ed). Chicago: Aldine; 231–281.
Darwin C, 1871. The descent of man, and selection in relation to sex. London: John Murray.
Dawkins R, Krebs JR, 1978. Animal signals: information or manipulation? In: Behavioural ecology: and evolutionary approach (Krebs JR, Davies NB, eds).Oxford: Blackwell; pp. 282–309.
Evans MR, Goldsmith AR, Norris SRA, 2000. The effects of testosterone on antibody production and plumage coloration in male house sparrows (Passer domesticus). Behav Ecol Sociobiol 47:156-163.[CrossRef][Web of Science]
Fraser GS, Jones IL, Hunter FM, 2002. Male-female differences in parental care in monogamous crested auklets. Condor 104:413-423.[CrossRef][Web of Science]
Frith HJ, 1982. Waterfowl in Australia. Sydney: Angus & Robertson.
Griffiths R, Double MC, Orr K, Dawson RJG, 1998. A DNA test to sex most birds. Mol Ecol 7:1071-1075.[CrossRef][Medline]
Huxley JS, 1914. The courtship habits of the great crested grebe (Podiceps cristatus); with an addition to the theory of sexual selection. Proc Zool Soc 35:491-562.
Johnsgard PA, 1965. Handbook of waterfowl behaviour. London: Constable & Co.
Johnstone RA, 1997. The tactics of mutual mate choice and competitive search. Behav Ecol Sociobiol 40:51-59.
Johnstone RA, Norris K, 1993. Badges of status and the cost of aggression. Behav Ecol Sociobiol 32:127-134.[CrossRef][Web of Science]
Johnstone RA, Reynolds JD, Deutsch JC, 1996. Mutual mate choice and sex differences in choosiness. Evolution 50:1382-1391.[CrossRef][Web of Science]
Jones IL, Hunter FM, 1993. Mutual sexual selection in a monogamous seabird. Nature 362:238-239.[CrossRef]
Jones IL, Hunter FM, 1999. Experimental evidence for mutual inter- and intrasexual selection favouring a crested auklet ornament. Anim Behav 57:521-528.[CrossRef][Web of Science][Medline]
Jones IL, Hunter FM, Fraser G, 2000. Patterns of variation in ornaments of crested auklets Aethia cristatella. J Avian Biol 31:119-127.[CrossRef]
Jones IL, Montgomerie R, 1991. Mating and remating of least auklets (Aethia pusilla) relative to ornamental traits. Behav Ecol 2:249-257.
Jones IL, Montgomerie R, 1992. Least auklet ornaments: do they function as quality indicators? Behav Ecol Socicobiol 30:43-52.
Kokko H, Johnstone RA, 2002. Why is mutual mate choice not the norm? Operational sex ratios, sex roles and the evolution of sexually dimorphic and monomorphic signalling. Phil Trans R Soc Lond B 357:319-330.
Kokko H, Monaghan P, 2001. Predicting the direction of sexual selection. Ecol Lett 4:159-165.
Kraaijeveld K, Mulder RA, 2002. The function of triumph ceremonies in the black swan. Behaviour 139:45-54.[CrossRef][Web of Science]
Lande R, 1980. Sexual dimorphism, sexual selection, and adaptation in polygenic characters. Evolution 34:292-305.[CrossRef][Web of Science]
Lande R, 1987. Genetic correlations between the sexes in the evolution of sexual dimorphism and mating preferences. In: Sexual selection: testing the alternatives (Bradbury JW, Andersson MB, eds). Chichester, UK: Wiley; pp. 83–95.
Lande R, Arnold SJ, 1985. Evolution of mating preference and sexual dimorphism. J Theor Biol 117:651-664.[CrossRef][Web of Science][Medline]
Loonen MJJE, Bruinzeel LW, Black JM, Drent RH, 1999. The benefit of large broods in barnacle geese: a study using natural and experimental manipulations. J Anim Ecol 68:753-768.[CrossRef]
Lyon BE, Eadle JM, Hamilton LD, 1994. Parental choice selects for ornamental plumage in American coot chicks. Nature 371:240-243.[CrossRef]
Marchant S, Higgins PJ, 1990. Handbook of Australian, New Zealand and Antarctic birds, vol. I: Ratites to ducks. Oxford: Oxford University Press.
Møller AP, 1987. Social control of deception among status signalling house sparrows Passer domesticus. Behav Ecol Sociobiol 23:21-26.
Molles LE, Vehrencamp SL, 2001. Songbird cheaters pay a retaliation cost: evidence for auditory conventional signals. Proc R Soc Lond B 268:2013-2019.[Medline]
Packard GC, Boardman TJ, 1987. The misuse of ratios to scale physiological data that vary allometrically with body size. In: New directions in ecological physiology, vol 1. (Feder ME, Bennet AF, Burggren WW, Huey RB, eds). Cambridge: Cambridge University Press; 216-239.
Parker GA, Simmons LW, 1996. Parental investment and the control of sexual selection: predicting the direction of sexual selection. Proc R Soc Lond B 263:315-321.
Parker TH, Ligon JD, 2002. Dominant male red junglefowl (Gallus gallus) test the dominance status of other males. Behav Ecol Sociobiol 3:20-24.[CrossRef]
Perrins CM, McCleery RH, 1997. Pairing behaviour in a colony of mute swans Cygnus olor. Wildfowl 47:31-41.
Petrie M, 1983. Female Moorhens compete for small fat males. Science 220:413-415.
Petrie M, 1984. Territory size in the moorhen (Gallinula chloropus): an outcome of RHP asymmetry between neighbours. Anim Behav 32:861-870.[CrossRef]
Petrie M, 1988. Intraspecific variation in structures that display competitive ability: large animals invest relatively more. Anim Behav 36:1174-1179.[CrossRef]
Poiani A, Goldsmith AR, Evans MR, 2000. Ectoparasites of house sparrows (Passer domesticus): an experimental test of the immunocompetence handicap hypothesis and a new model. Behav Ecol Sociobiol 47:230-242.[CrossRef][Web of Science]
Pomiankowksi A, Iwasa Y, Nee S, 1991. The evolution of costly mate preferences. 1. Fisher and biased mutation. Evolution 45:1422-1430.[CrossRef][Web of Science]
Rice WR, 1989. Analyzing tables of statistical tests. Evolution 43:223-225.[CrossRef][Web of Science]
Rising JD, Somers KM, 1989. The measurement of overall body size in birds. Auk 106:666-674.[Web of Science]
Roulin A, Dijkstra C, Riols C, Ducrest AL, 2001. Female- and male-specific signals of quality in the barn owl. J Evol Biol 14:255-266.
Rohwer SA, 1975. The social significance of avian winter plumage variability. Evolution 29:593-610.[CrossRef][Web of Science]
Rohwer SA, Rohwer FC, 1978. Status signalling in Harris' sparrows; experimental deception achieved. Anim Behav 26:1012-1022.[CrossRef][Web of Science]
Savalli UM, 1995. The evolution of bird coloration and plumage elaboration. Curr Ornithol 12:141-190.
Scott DK, 1980. Functional aspects of prolonged parental care in Bewick's swans. Anim Behav 28:938-952.[CrossRef][Web of Science]
Scott P, the Wildfowl Trust, 1972. The swans. Boston: Houghton Mifflin.
Systat, 1997. SYSTAT for windows, statistics, version 7 ed. Evanston, Illinois: SYSTAT.
Tanaka Y, 1996. Social selection and the evolution of animal signals. Evolution 50:512-523.[CrossRef][Web of Science]
Trivers RL, 1972. Parental investment and sexual selection. In: Sexual selection and the descent of man. (Campbell B, ed). Chicago: Aldine; pp. 136–179.
Van der Jeugd HP, Blaakmeer KB, 2001. Teenage love: the importance of trial liaisons, subadult plumage and early pairing in barnacle geese. Anim Behav 62:1075-1083.[CrossRef]
Velando A, Lessells CM, Marquez JC, 2001. The function of female and male ornaments in the Inca tern: evidence for links between ornament expression and both adult condition and reproductive performance. J Avian Biol 32:311-318.[CrossRef]
West-Eberhard MJ, 1983. Sexual selection, social competition, and speciation. Q Rev Biol 58:155-183.[CrossRef][Web of Science]
West-Eberhard MJ, 1984. Sexual selection, competitive communication and species-specific signals in insects. In: Insect communication (Lewis T, ed). London: Academic Press; 283–324.
West-Eberhard MJ, 1991. Sexual selection and social behavior. In: Man and beast revisited. (Robinson MH, Tiger L, eds). Washington, DC: Smithsonian Institution Press; pp. 159–172.
Whitfield DP, 1987. Plumage variability, status signalling and individual recognition in avian flocks. Trends Ecol Evol 2:13-18.
Wiens JJ, 2001. Widespread loss of sexually selected traits: how the peacock lost its spots. Trends Ecol Evol 16:517-523.[CrossRef]
Williams M, 1977. Locations of recoveries of black swans, Cygnus atratus Latham, banded at Lake Whangape and Lake Ellesmere, New Zealand. Aust Wildl Res 4:289-299.
Zahavi A, 1972. The social behaviour of the white wagtail, Motacilla a. alba, wintering in Israel. Ibis 113:203-211.[CrossRef][Web of Science]
Zahavi A, 1990. Arabian babblers: the quest for social status in a cooperative breeder. In: Cooperative breeding in birds (Stacey P, Koenig W, eds). Cambridge: Cambridge University Press; pp. 103–130.
Zahavi A, 1991. On the definition of sexual selection, Fisher's model, and the evolution of waste and of signals in general. Anim Behav 42:501-503.[CrossRef]
Zahavi A, Zahavi A, 1997. The handicap principle. Oxford: Oxford University Press.
Zar JH, 1999. Biostatistical analysis, 4th ed. Englewood Cliffs, New Jersey: Prentice Hall.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  R. Chan, D. Stuart-Fox, and T. S. Jessop Why are females ornamented? A test of the courtship stimulation and courtship rejection hypotheses Behav. Ecol., November 1, 2009; 20(6): 1334 - 1342. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. G. Murphy, D. Hernandez-Mucino, M. Osorio-Beristain, R. Montgomerie, and K. E. Omland Carotenoid-based status signaling by females in the tropical streak-backed oriole Behav. Ecol., September 1, 2009; 20(5): 1000 - 1006. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. M. Aguilar, R. Maia, E. S.A. Santos, and R. H. Macedo Parasite levels in blue-black grassquits correlate with male displays but not female mate preference Behav. Ecol., March 1, 2008; 19(2): 292 - 301. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. A. Tibbetts and T. R. Curtis Rearing conditions influence quality signals but not individual identity signals in Polistes wasps Behav. Ecol., May 1, 2007; 18(3): 602 - 607. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Komdeur, M. Oorebeek, T. van Overveld, and I. C. Cuthill Mutual ornamentation, age, and reproductive performance in the European starling Behav. Ecol., July 1, 2005; 16(4): 805 - 817. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||