Behavioral Ecology Advance Access originally published online on August 11, 2004
Behavioral Ecology 2005 16(1):41-47; doi:10.1093/beheco/arh117
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Offspring sex ratios correlate with pair–male condition in a cooperatively breeding fairy–wren
Dept of Biology, Queen's University, Kingston, Ontario K7L 3N6
Address correspondence to M. Rathburn. E-mail: rathburn{at}biology.queensu.ca.
Received 22 August 2003; revised 6 February 2004; accepted 2 May 2004.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
We examined sex allocation patterns in island and mainland populations of cooperatively breeding white-winged fairy-wrens. The marked differences in social structure between island and mainland populations, in addition to dramatic plumage variation among males both within and between populations, provided a unique situation in which we could investigate different predictions from sex allocation theory in a single species. First, we test the repayment (local resource enhancement) hypothesis by asking whether females biased offspring sex ratios in relation to the assistance they derived from helpers. Second, we test the male quality (attractiveness) hypothesis, which suggests that females mated to attractive high-quality males should bias offspring sex ratios in favor of males. Finally, we test the idea that females in good condition should bias offspring sex ratios toward males because they are able to allocate more resources to offspring, whereas females in poor condition should have increased benefits from producing more female offspring (Trivers-Willard hypothesis). We used molecular sexing techniques to assess total offspring sex ratios of 86 breeding pairs over 2 years. Both offspring and first brood sex ratios were correlated with the pair-male's body condition such that females increased the proportion of males in their brood in relation to the body condition (mass corrected for body size) of their social partner. This relation was both significant and remarkably similar in both years of our study and in both island and mainland populations. Although confidence of paternity can be low in this and other fairy-wren species, we show how this finding might be consistent with the male quality (attractiveness) hypothesis with respect to male condition. There was no support for the repayment hypothesis; the presence of helpers had no effect on offspring sex ratios. There was weak support for both the male quality (attractiveness) hypothesis with respect to plumage color and the maternal condition hypothesis, but their influence on offspring sex ratios was negligible after controlling for the effects of pair-male condition.
Key words: sex allocation, offspring sex ratio, parental condition, cooperative breeding, plumage coloration.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Parental manipulation of offspring sex ratio is favored when the long-term fitness advantages of producing sons versus daughters differ (Trivers and Willard, 1973
). Although it is clear that such sex ratio manipulation is possible in birds (Cockburn et al., 2002), and sometimes fits the expectations from theory (West and Sheldon, 2002), neither the processes leading to sex ratio adjustment nor the patterns that result from those processes are well understood in this taxon (Komdeur and Pen, 2002) for two reasons. First, until recently, reliable nondestructive sexing of nestlings was impossible in all but a few species; a problem now solved with the discovery of sex-specific microsatellite primers (Griffiths et al., 1998). Second, classical sex allocation theory does not always apply to birds because their life histories often violate several assumptions of that theory (Komdeur and Pen, 2002). Despite this, considerable progress has been made recently, documenting cases of sex ratio manipulation in birds and testing some specific hypotheses about the direction of an effect (West and Sheldon, 2002).
Many different factors can potentially influence variation in offspring sex ratios in birds, including parental quality (Trivers and Willard, 1973), local resource competition or enhancement (Emlen et al., 1986; Lessells and Avery, 1987), and the costs of reproduction (Stamps, 1990). Recent reviews provide an excellent summary of both the hypotheses and their support from studies of birds (Cockburn et al., 2002, Komdeur and Pen, 2002, West and Sheldon 2002). Although there is some excellent evidence for the influence of helpers, paternal attractiveness, and maternal condition on sex ratio manipulation in birds (see below), the relative importance of such factors in a single system has rarely been assessed (but see Koenig et al., 2001).
Here, we examine sex allocation patterns in island and mainland populations of cooperatively breeding white-winged fairy-wrens (Malurus leucopterus). Our study populations had markedly different social structure that allowed us to assess the relative influence of helpers on offspring sex ratio: most mainland pairs had one to five helpers at their nests, whereas island pairs had few, if any (Rathburn and Montgomerie, 2003). White-winged fairy-wren males also exhibit delayed plumage maturation, not obtaining full nuptial plumage until their fourth year. Brown-plumaged younger males are behaviorally subordinate to nuptial-plumaged (black or blue) males, and females prefer to copulate with nuptial-plumaged males (Rowley and Russell, 1995, 1997). These marked differences in social structure between island and mainland populations, in addition to the dramatic plumage differences among males, provided a unique situation in which we could investigate the effects of social system (helpers), plumage coloration (male attractiveness), and parental condition on patterns of offspring sex ratio adjustment in a single species.
Specifically, we test three predictions from theory (Cockburn et al., 2002). First, we test the repayment (local resource enhancement) hypothesis by asking whether females bias offspring sex ratios in relation to the assistance they derive from helpers (Emlen et al., 1986; Lessells and Avery, 1987). Among some cooperatively breeding birds, offspring sex ratio is often correlated with the number and sex of helpers-at-the-nest (Gowaty and Lennartz, 1985; Komdeur, 1996; Komdeur et al., 1997). Helpers are generally young from previous broods that provide care for their parents' offspring. By staying to help, these individuals enhance the fitness of the breeding pair and thus provide repayment for being allowed to remain on their natal territory (Emlen et al., 1986). Thus, it is in a female's best interest to adjust sex ratios favoring the helping sex (in this case males) when the presence of helpers increases her total reproductive success. By comparing mainland (many helpers) and island (few or no helpers) populations and assessing the influence of helper numbers on the mainland, we are able to address this question in two different ways.
Second, we test the male quality (attractiveness) hypothesis, which suggests that females mated to attractive high-quality males should overproduce males. Such a bias is expected when male reproductive success is correlated with male attractiveness or other measures of quality. If traits that make a male attractive to females are heritable, then the sons of attractive males will have higher reproductive value than do daughters. To date, however, studies on sex allocation in relation to paternal quality have produced mixed results. Females mated to more colorful males (Sheldon et al., 1999), larger males (Kölliker et al., 1999), and males with larger ornaments (Ellegren et al., 1996) have all been found to produce male-biased broods, but in some species, offspring sex ratio is apparently not related to paternal condition (Saino et al., 1999; Westerdahl et al., 1997).
We assess this hypothesis by looking at both the body condition (a measure of quality) and the plumage coloration (a measure of attractiveness) of the pair-male for each brood. Several recent studies have shown that male condition correlates with aspects of plumage ornamentation and parental quality. In this (Rathburn MK and Montgomerie R, unpublished data) and other fairy-wren species (Mulder et al., 1994), however, the confidence of paternity for the pair-male can sometimes be quite low, so this hypothesis is difficult to test directly without knowledge of the paternity of each offspring. We discuss the implications of this later. We assessed the plumage coloration of pair-males as either nuptial (older, preferred) or brown (younger) but did not measure any within-category plumage variation that might also correlate with quality.
Finally, because male fairy-wrens are polygynous, the variance in their reproductive success is higher than that of females (Mulder et al., 1994). Because females in good condition are able to allocate more resources to offspring, they should bias sex ratios toward males, whereas females in poor condition should have increased benefits from producing more female offspring (Trivers-Willard hypothesis). Recent work has shown that maternal condition is correlated with male-biased sex ratio in numerous bird species (Bradbury and Blakey, 1998; Nager et al., 1999; Whittingham and Dunn, 2000; Whittingham et al., 2002).
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
General field methods
Field work was conducted during the peak breeding seasons at two study sites in Western Australia during 2000 and 2001 (Rathburn and Montgomerie, 2003
): on Dirk Hartog Island (25°53' S, 113°06' E) from late July to late September and on the mainland near Lancelin (31°09, 115°26) from early October to late November.
Adults were trapped opportunistically on their territories by using mist-nets and provided with an Australian Bird and Bat Banding Scheme alloy band in addition to a unique combination of color bands for field identification. We took small blood samples from the brachial vein of each individual in addition to standard morphological measurements (Lowe, 1989). We also removed the central rectrix from the right side of the tail of each adult to determine feather growth rate.
We recorded the identity of the breeding pair and all helpers on each territory during daily behavioral observations of nonaggressive social interactions. In total, we located 104 nests on the island and 62 on the mainland. Nests were checked daily during nest building and egg laying, and then every 1–3 d thereafter until nest failure or fledging.
In Western Australia, white-winged fairy-wrens breed predominantly in the austral spring (July–November), but they do not have clearly defined breeding seasons and have been found breeding in every month, except June (Tidemann and Marples, 1987). Thus, we designate as "first broods" the first brood that we found for each pair during our field seasons, but these may not always have been the first broods in the July–November "breeding season" in the sense that this term is traditionally used. Of the 86 breeding females that we monitored in total (both sites, both years), 14 raised two broods in a season, and one other raised three broods. Each female remained paired with the same male for all broods within a season. Only five females (three on the island, two on the mainland) nested and successfully fledged offspring in both years of our study, but none of these females was paired to the same male in both years.
We used two different techniques to assess adult body condition: relative mass and feather growth rate. To determine an individual's relative mass, controlling for body size, we calculated the residuals from a regression of body mass on tarsus length (Brown, 1996). Because there was significant variation in body mass between both study sites and sexes (Rathburn and Montgomerie, 2003), we performed this analysis separately for all adult males and females captured at each site that were the putative parents of broods for which we analyzed offspring sex ratios (see below). We used untransformed mass and tarsus values for this analysis because the residuals from these regressions were more normally distributed (Shapiro-Wilks tests) than residuals from the log-log regressions typically used to calculate residuals as a measure of condition. None of the four mass on tarsus correlations was even close to significant (r = .04–0.15, p = .35–0.81, n = 35–41), so body mass alone would have been as good a measure of condition for the present study, and gives the same conclusions as we report for residual male mass.
Feather growth rate provides an index of a bird's nutritional condition during molt (Grubb, 1989). To determine this, we measured the distance spanned by up to 10 growth bars (i.e., 10 d of feather growth) on the fifth rectrix of each bird, following techniques described by Grubb (1989) and Hill and Montgomerie (1994), and calculated an average daily feather growth rate. In white-winged fairy-wrens, adults molt only once each year, in the austral autumn immediately after breeding (Rowley and Russell, 1997). Thus, feather growth rates are an index of condition long before the breeding season during which they were measured.
Sex determination
We determined the sex of 286 nestlings from 102 broods by using molecular techniques. To do this, we took a small blood sample from each nestling at 6 or more days of age. Then, DNA was extracted from these blood samples by using salt extraction (Hillis et al., 1996) and amplified under standard polymerase chain reaction (PCR) conditions (Ta = 50°C) with P8/P2 primers that amplify the sex-specific CHD gene (Griffiths et al., 1998). PCR products were separated by electrophoresis on an agarose gel, stained with ethidium bromide, and photographed under ultraviolet light. DNA from adults of known sex were run on each gel to ensure that we could accurately distinguish between the sexes.
Data analysis
We analyzed only the broods of females that were mated to nuptial-plumaged (n = 33) and brown-plumaged (n = 46) males, thus excluding broods of seven females mated to males in mottled plumage. Males in mottled plumage are probably intermediate in age and status between nuptial- and brown-plumaged males (Rowley and Russell, 1997) but were too few to analyze separately. Including these females in the analyses as if they were mated to either brown- or nuptial-plumaged males does not affect our conclusions.
To focus on sex ratios at egg laying, we did not include three broods in which one egg or nestling disappeared before blood sampling, possibly owing to predation. There were no other instances of nestling mortality, so sex ratio variation could not be attributed to differential mortality through biased investment in parental care. We did include data from five broods in which we were unable to sex one of the nestlings. To ensure that these missing data did not bias offspring sex ratios toward males, all analyses were repeated with all unsexed birds included as females, the nonhelping sex, but our conclusions were not affected.
Because all pairs remained together within years and no pairs remained together between years, we combined all nestlings of all broods of each breeding pair to determine overall offspring sex ratios per pair per breeding season. Restricting analyses to first broods only, and to one brood for each female, did not affect our conclusions but reduced statistical power. For the five females that bred in both years of our study, there was no correlation between their offspring sex ratios between years (r = .03, p = .80), suggesting that offspring sex ratios are not characteristic of individual females and may either vary at random or be adjusted by females to current conditions (e.g., mate characteristics, number of helpers, own condition).
We used binomial tests to compare observed sex ratios to those expected from a 1:1 distribution (Zar, 1999). To examine the relation between sex ratios and other variables, we constructed generalized linear models (GLMs) by using GLMstat (version 5.7.5; available at http://www.ozemail.com/kjbeath/glmstat.html). For each GLM, we used the number of male offspring per female per year as the response variable, and the total number of sexed nestlings per female per year as the denominator. The model was constructed with a binomial error structure and a logit link function (Wilson and Hardy, 2002). We used the change in deviance between our data and the null model to test the significance of each variable. So that our findings can be compared with those of other studies, we report the effect size correlation, rY1 or 
, which we calculated from the GLM chi-square statistic by using the formula in Rosenthal (1991). Because our data set comprised offspring from two different subspecies, with different evolutionary histories, we initially included study site as a predictor variable in every model. Our sample size of brood sex ratios is too small to permit a more comprehensive statistical analysis, such as path analysis, or a multiple regression analysis that included all possible predictor variables. Thus, we tested each hypothesis separately, using the appropriate predictor variables in each case. We then evaluated the relative importance of significant predictor variables from each of these separate hypothesis tests in a final GLM, including only those variables (and study site) as predictors.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Population sex ratios
Overall offspring sex ratio in broods of females mated to nuptial- and brown-plumaged males was almost exactly 1:1 (134 males/133 females) for both study sites combined. On the island, 48.3% of these nestlings were male (n = 122 nestlings), whereas 51.7% were male on the mainland (n = 145 nestlings). Neither of these overall offspring sex ratios was significantly different from 1:1 (binomial tests: island, p = .68; mainland, p = .65).
Mean offspring sex ratios (±SE) were also not significantly different from 50% male on the island (49.3 ± 5.5%; one sample t test, t = 0.13, p = .90, n = 44 pairs) or the mainland (54.1 ± 4.8%; t = 0.86, p = .39, n = 35 pairs). There was thus a slight tendency for island pairs to have female-biased offspring sex ratios on the island and male–biased sex ratios on the mainland, but the difference between study sites was not significant (t test, t = 0.65, p = .52, n = 44 island, 35 mainland pairs). There was also no significant difference in offspring sex ratio between a pair's first and second brood (paired t = 0.44, p = .67, n = 15 pairs) within a season.
Influence of helpers
To examine offspring sex ratios in relation to the number of helpers at a nest, we constructed a GLM with the number of helpers and location as predictor variables. We also included male plumage phenotype into this model to control for significant differences in helper numbers between nests of nuptial- and brown-plumaged males (Rathburn and Montgomerie, 2003).
The number of helpers at a nest had no effect on offspring sex ratios (Table 1). There were only two instances of helpers at the nests of brown-plumaged males on the island, so to ensure that the lack of variation in helper numbers among island broods was not unduly influencing our findings, we redid this analysis on mainland broods alone, in which helpers were common and their number highly variable among nests (Rathburn and Montgomerie, 2003). In the mainland population, however, there was no significant effect of the number of helpers on offspring sex ratios, controlling for the plumage color category (nuptial versus brown) of the pair-male (helpers, 
2 = 0.14, p = .71; plumage, 2 = 2.04, p = .15; whole model, 2 = 2.13, p = .35, df = 2,32).
|
Pair-male quality: plumage category
Next, we examined if the proportion of male nestlings differed between male plumage phenotypes, by constructing a GLM with only pair-male plumage coloration and study site as predictor variables. Females mated to nuptial-plumaged males had a tendency to produce a higher proportion of male nestlings compared with nestlings of females paired to brown-plumaged mates (
2 = 3.49, p = .06; effect size for male plumage category, = 0.21) (Figure 1), and there was no effect of study site in this model (2 = 0.15, p = .69; whole model, 2 = 3.79, p = .15, df = 2,76).
|
Pair-male quality: body condition
To investigate whether offspring sex ratios were related to male condition, we included study site, male plumage phenotype, male feather growth rate, and residual male mass as potential predictor variables in a GLM. There was no effect of male feather growth rate on offspring sex ratios, but females mated to heavier males produced a significantly higher proportion of male offspring (effect size for residual male mass,
= 0.39) (Table 2).
|
Because we captured males opportunistically through the day and through the breeding season, we examined these relations further to control for any potential variation in these variables that could have given a spurious correlation. There were weak correlations between residual male mass and both time of day (mainland, r = .12, p = .50, n = 35; island, r = .28, p = .08, n = 41) and day of the year (positive on the mainland: r = .34, p = .05, n =35, negative on the island, r = –.15, p = .36, n = 41), but none of these were significant after Bonferroni correction for multiple analyses (
= 0.013) and there was no suggestion that any of these relations was nonlinear. Adding both time of day and day of the year to this model (Table 2), and removing feather growth rate (which was far from significant), had no effect on our conclusions, as residual male mass remained a significant predictor of offspring sex ratios (p < .002), whereas day of the year and time of day were not (p > .35 in each case). Eight breeding males were captured in both years of our study, providing an opportunity to assess whether an individual's condition varied from year to year. A male's condition in 2001 was significantly correlated with his condition in 2000 (r = .74, p = .04, n = 8), suggesting that age or environment accounted for about half of the variation in condition from one year to the next.
Maternal condition
To test the prediction that females in better condition produce male-biased offspring sex ratios, we entered two indices of female condition (feather growth rate and residual female mass) along with study site as predictor variables in a GLM. We also included male plumage color as a predictor in this model to control for the relation between male color and offspring sex ratios, reported above. Offspring sex ratios were not related to female feather growth rate, but females in better condition tended to produce male-biased broods (effect size for residual female mass, = 0.22) (Table 3).
|
Overall effects
To determine the relative influence of the most important predictors from the preceding analyses on offspring sex ratios, we constructed a GLM with study site, male plumage color, residual female mass, and residual male mass as predictor variables. In this GLM, the only significant predictor of offspring sex ratios was residual male mass, such that females mated to males in better condition produced significantly male-biased offspring (effect size for residual male mass,
= 0.48) (Table 4). We also performed the above analysis restricting our data set to the first broods of each female. Among these first broods, females mated to males in better condition skewed offspring sex ratios favoring males (effect size for male condition, = 0.43) (Table 4b).
|
To determine if this significant relation was influenced by the nonsignificant predictor variables in the model, we used stepwise deletion to remove nonsignificant effects from the GLM. All predictors except residual male mass were thus removed as nonsignificant in this model. Thus, there was no change in our findings—females skewed offspring sex ratios toward males when paired with males in better condition regardless of location or male plumage color (
2 = 14.2, p = .0002, df = 1,74; effect size, = 0.43). Moreover, this pattern was significant in both years of our study (2000, 2 = 6.95, p = .008, df = 1,46; = 0.38; 2001, 2 = 8.30, p = .004, df = 1,26; = 0.54), and within both island (2 = 7.1, p = .008, df = 1,39; = 0.42) and mainland populations (2 = 7.3, p = .007, df = 1,33; = 0.46) (Figure 2).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
In contrast to most studies of cooperatively breeding birds, the number of helpers at the nest in white-winged fairy-wrens had no effect on sex allocation patterns. Overall, however, females mated to males in better body condition had male-biased offspring sex ratios in their broods. This pattern was consistent between years, suggesting that it is not simply an artifact of sampling (Griffith et al., 2003
). The effect of pair-male condition on offspring sex ratios in white-winged fairy-wrens is similar in magnitude (effect size, = 0.43) to that of other bird studies in which a significant effect of helper number (effect sizes = 0.36–0.56, mean = 0.48, n = 4 studies of three species) or mate quality (effect sizes = 0.35–0.45, mean = 0.40, n = 5 studies of three species) has been found (West and Sheldon, 2002).
Sex allocation theory clearly predicts that in a cooperatively breeding species with only one sex helping at the nest, females should bias their offspring sex ratio in favor of the helping sex (Gowaty and Lennartz, 1985; Komdeur et al., 1997)—a result that has been found in at least five studies so far (West and Sheldon, 2002). Thus, it is particularly surprisingly that there was no relation between the number or presence of helpers on offspring sex ratios in white-winged fairy-wrens (Table 1). Although helpers were present at the nest in this species, especially on the mainland, we do not know their relative contribution to providing care for the offspring, so it is not possible for us to evaluate the importance of helpers in enhancing a female's reproductive success.
If helpers do not enhance a female's fitness, then no relation between helper number and offspring sex ratio is expected. For example, in superb fairy-wrens (Malurus cyaneus), helpers do contribute parental care at the nest, but this care does not release females from any parental care and does not influence a female's overall reproductive success (Mulder et al., 1994). Instead, superb fairy-wren offspring are allowed to remain on their natal territory as helpers because the care they provide for nestlings mitigates the loss of parental care from the social father (Mulder et al., 1994). White-winged fairy-wren helpers appear to function in a similar manner, as Tidemann (1986) found that helpers usually contributed more parental care than the social father in this species.
The relation between maternal condition and offspring sex ratios in white-winged fairy-wrens was significant on its own (Table 3), but not when controlling for male condition (Table 4). Thus, maternal condition does not appear to influence offspring sex ratios in this species. Alternatively, female and male condition may be correlated such that variation in male condition explains much of the variation in female condition. However, this seems unlikely as there is only a weak correlation between male and female condition within pairs (r = .16, p = .17, n = 74). The effects of maternal condition on offspring sex ratios have been largely ignored owing to the difficulties in making clear predictions based on theory (West and Sheldon, 2002). Thus, some critical experiments are needed to determine whether the apparently weak effect that we have uncovered in white-winged fairy-wrens is real or simply the result of a correlation between female condition and other variables that actually influence offspring sex ratios.
Very little is known about the qualities that female white-winged fairy-wrens prefer in a partner, but we do know that they pair with nuptial-plumaged miles in preference to brown-plumaged males (Rowley and Russell, 1997), suggesting that plumage plays an important role in a female's mate choice decisions. Our finding that females bias offspring sex ratio in favor of males when they are mated to nuptial-plumaged males provides additional support for the idea that plumage coloration is an indicator of male attractiveness. Because white-winged fairy-wrens undergo delayed plumage maturation, nuptial coloration is positively related to male age and survival. Thus, offspring of females mated to nuptial-plumaged males would gain mating or survival benefits. Rowley and Russell (1997) also report that males can have nuptial plumage color in one year and then brown plumage in the following breeding season, further suggesting that plumage coloration may be a condition-dependent trait. As for maternal condition, the relation between male plumage color and offspring sex ratio was not significant when controlling for male condition, suggesting that this too is at best a weak effect.
Among monogamous species, females tend to produce male-biased offspring sex ratios when they are mated to males of relatively high quality or condition (Ellegren et al., 1996; Kölliker et al., 1999; Sheldon et al., 1999), but rarely has this relation been examined in a cooperatively-breeding species. In cooperatively-breeding acorn woodpeckers (Melanerpes formicivorus), Koenig et al. (2001) found no relation between male condition and offspring sex ratio, although they examined various factors correlated with condition (year, season, territory quality) rather than direct physical indicators of condition. We assume that body condition of male white-winged fairy-wrens during the breeding season is an index of male quality, but this may not be true in other bird species.
The relation that we observed between pair-male condition and sex allocation in white-winged fairy-wrens suggests that females are able to assess the quality of their social mate and manipulate sex ratios accordingly. Because it is unlikely that females use residual mass to assess male condition, we expect that they use male behavior and plumage characteristics as an indicator of a male's quality. White-winged fairy-wrens are nonmigratory and remain on their territories throughout the year (Rowley and Russell, 1995), so females have ample time to assess male behavior. Because of the nature of their social system, females can gain considerable information from daily interactions with males within and between their social groups. Females can also assess the behavior of both their partners and their neighbors through observations of male-male interactions, territory defense, foraging skill, sexual displays, petal carrying behavior, paternal care, and realized reproductive success.
Females may also use male plumage characteristics to assess male quality. Although we found some evidence that females may bias their offspring sex ratios in response to their social mate's plumage color category (Table 2), we did not assess the fine-scale characteristics of male plumage that may provide females with additional information on male quality within plumage color categories. Indeed, the timing of molt, the bright white scapular and wing feathers of nuptial-plumaged males, or the size and color of their long tail may serve as indicators of male quality. For example, in superb fairy-wrens, the timing of nuptial molt is a condition-dependent trait and females prefer to mate with males that molt into nuptial plumage earlier in the year (Mulder and Magrath, 1994). Recently, the spectral characteristics of plumage (reflectance, brightness, and hue) have also been shown to be reliable indicators of male quality (Doucet and Montgomerie, 2003; Keyser and Hill, 2000), and females are able to distinguish between males by using these plumage characteristics, preferentially mating with high-quality males (Bennett et al., 1996; Hunt et al., 1999; Johnsen et al., 1998).
Although the correlation between pair-male condition and offspring sex ratio is interesting, it does not constitute unequivocal support for the male quality (attractiveness) hypothesis, owing to the potentially high levels of extrapair paternity in this species. Females may well adjust their offspring sex ratio in relation to the pair-male's condition, but that male might not be the father of many of those offspring and thus would not pass on to them any genes that confer quality. We offer three potential explanations to reconcile this apparent discrepancy, and we note that no other studies of sex ratio allocation in birds have as yet controlled for extrapair paternity. First, it is possible that confidence of paternity for the pair-male is positively correlated with his body condition. Such a correlation has been found in other species (Møller et al., 2003) and would provide some indirect support for the male quality (attractiveness) hypothesis if the condition of extrapair males was unknown to females or random with respect to the pair-male. Thus, if extrapair males were of higher quality than was the pair-male, females should still bias their brood sex ratio toward males as predicted by the male quality (attractiveness) hypothesis.
Second, if extrapair males were of equal or higher quality than were the pair-males, then a correlation between pair-male condition and offspring sex ratio would still be expected. There is, for example, some evidence that female birds seek extrapair copulations from males of higher quality than their social partner (Graves et al. 1993, Hasselquist et al. 1996; Kempenaers et al. 1992).
Finally, it is possible that this correlation between pair-male condition and brood sex ratio is simply spurious, the result of a positive correlation between pair-male condition and some other aspect of fairy-wren biology that females are sensitive to when adjusting brood sex ratios. For example, pair-male condition might simply reflect territory quality, a factor that could be important for offspring development and survival. Because males do not achieve full-breeding plumage, and presumably maximal reproductive success, until their fourth year, females might be expected to bias their offspring sex toward males when the chances for male survival are highest. Clearly, more work will be needed to sort among these possibilities.
Recent studies of offspring sex ratios in birds have shown that there is a variety of social, environmental, and physical factors that can influence sex allocation patterns, but just as many studies find no relation between offspring sex ratios and these same variables (Cockburn et al., 2002). Some of these differences between species and populations may be attributable to differences in life histories (Komdeur and Pen, 2002), although in our study, we found no difference between island and mainland populations with clearly different evolutionary histories and life-history characteristics (Rathburn and Montgomerie, 2003). Despite that, the clear patterns of sex allocation that we describe in this article do not have a simple explanation.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Bennett ATD, Cuthill IC, Partridge JC, and Maier EJ, 1996. Ultraviolet vision and mate choice in zebra finches. Nature 380:433–435.[CrossRef]
Bradbury RB, and Blakey JK, 1998. Diet maternal condition and offspring sex ratio in the zebra finch. Proc R Soc Lond B 265:895–899.
Brown ME, 1996. Assessing body condition in birds. Curr Ornith 13:67–135.
Cockburn A, Legge S, and Double MC, 2002. Sex ratios in birds and mammals: can the hypotheses be disentangled? In: Sex ratios: concepts and research methods, (Hardy ICW, ed). Cambridge: Cambridge University Press; 266–286.
Doucet SM, and Montgomerie R, 2003. Multiple sexual ornaments in satin bowerbirds: ultraviolet plumage and bowers signal different aspects of male quality. Behav Ecol 14:503–506.
Ellegren H, Gustafsson L, and Sheldon BC, 1996. Sex ratio adjustment in relation to paternal attractiveness in a wild bird population. Proc Natl Acad Sci USA 93:11723–11738.
Emlen ST, Emlen JM, and Levin SA, 1986. Sex-ratio selection in species with helpers-at-the-nest. Am Nat 127:1–8.[CrossRef][Web of Science]
Gowaty PA, and Lennartz, MR, 1985. Sex ratios of nestling and fledgling red-cockaded woodpeckers (Picoides borealis) favor males. Am Nat 126:347–353.[CrossRef][Web of Science]
Graves J, Ortega-Ruano J, and Slater PJB, 1993. Extra-pair copulations and paternity in shags: do females choose better males? Proc R Soc Lond B 253:3–7.
Griffith SC, Örnborg J, Russell AF, Andersson S, and Sheldon BC, 2003. Correlations between ultraviolet coloration, overwinter survival and offspring sex ratio in the blue tit. J Evol Biol 16:1–10.[CrossRef][Web of Science][Medline]
Griffiths R, Double MC, Orr K, and Dawson RJD, 1998. A DNA test to sex most birds. Mol Ecol 7:1071–1075.[CrossRef][Medline]
Grubb TC Jr, 1989. Ptilochronology: feather growth bars as indicators of nutritional status. Auk 106:314–320.[Web of Science]
Hasselquist D, Bensch S, and von Schantz T, 1996. Correlation between male song repertoire, extra-pair paternity and offspring survival in the great reed warbler. Nature 381:3229–3232.
Hill GE, and Montgomerie R, 1994. Plumage colour signals nutritional condition in the house finch. Proc R Soc Lond B. 258:47–52.
Hillis DM, Moritz C, and Mable BK, 1996. Molecular systematics. Sunderland, Massachusetts: Sinauer Associates Inc.
Hunt S, Cuthill IC, Bennett ATD, and Griffiths R, 1999. Preferences for ultraviolet partners in the blue tit. Anim Behav 58:809–815.[CrossRef][Web of Science][Medline]
Johnsen A, Andersson S, Örnborg J, and Lifjeld JT, 1998. Ultraviolet plumage ornamentation affects social mate choice and sperm competition in bluethroats: a field experiment. Proc R Soc Lond B 265:1313–1318.
Kempenaers B, Verheyen GR, Van den Broeck M, Burke T, Van Broeckhoven C, and Dhondt AA, 1992. Extra-pair paternity results from female preference for high-quality males in the blue tit. Nature 357:494–496.[CrossRef]
Keyser AJ, and Hill GE, 2000. Structurally based plumage coloration is an honest signal of quality in male blue grosbeaks. Behav Ecol 11:202–209.
Koenig WD, Stanback MT, Haydock J, and Kraaijeveld-Smit F, 2001. Nestling sex ratio variation in the cooperatively breeding acorn woodpecker, Melanerpes formicivorus. Behav Ecol Sociobiol 49:357–365.[CrossRef][Web of Science]
Kölliker M, Heeb P, Werner I, Mateman AC, Lessells CM, and Richner H, 1999. Offspring sex ratio is related to male body size in the great tit, Parus major. Behav Ecol 10:68–72.
Komdeur J, 1996. Facultative sex ratio bias in the offspring of Seychelles warblers. Proc R Soc Lond B 263:661–666.
Komdeur J, Daan S, Tinbergen J, and Mateman AC, 1997. Extreme adaptive modification in sex ratio of the Seychelles warbler's eggs. Nature 385:522–525.[CrossRef]
Komdeur J, and Pen I, 2002. Adaptive sex allocation in birds: the complexities of linking theory and practice. Proc R Soc Lond B 357:373–380.
Lessells CM, and Avery MI, 1987. Sex-ratio selection in species with helpers at the nest: some extensions of the repayment model. Am Nat 129:610–620.[CrossRef][Web of Science]
Lowe KW, 1989. The Australian bird bander's manual: Australian bird and bat banding schemes. Canberra: Australian National Parks and Wildlife Service.
Møller AP, Brohede J, Cuervo JJ, de Lope F, and Primmer C, 2003. Extrapair paternity in relation to sexual ornamentation, arrival date, and condition in a migratory bird. Behav Ecol 14:707–712.
Mulder RA, Dunn PO, Cockburn A, Lazenby-Cohen KA, and Howell MJ, 1994. Helpers liberate female fairy-wrens from constraints on extra-pair mate choice. Proc R Soc Lond B 255:223–229.
Mulder RA, and Magrath MJL, 1994. Timing of prenuptial moult as a sexually selected indicator of male quality in superb fairy-wrens, Malurus cyaneus. Behav Ecol 5:393–400.
Nager RG, Monaghan P, Griffiths R, Houston DC, and Dawson R, 1999. Experimental demonstration that offspring sex ratio varies with maternal condition. Proc Natl Acad Sci USA 96:570––573.
Rathburn MK, and Montgomerie R, 2003. Breeding biology and social structure of white-winged fairy-wrens (Malurus leucopterus): Contrasts between island and mainland subspecies having different plumage phenotypes. Emu 103:295–306.[CrossRef][Web of Science]
Rosenthal, R, 1991. Meta-analytic procedures for social research. Newbury Park, California: Sage.
Rowley I, and Russell E, 1995. The breeding biology of the white-winged fairy-wren Malurus leucopterus leucontonus in a Western Australian coastal heathland. Emu 95:175–184.[Web of Science]
Rowley I, and Russell E, 1997. Fairy-wrens and grasswrens. Oxford: Oxford University Press.
Saino N, Ellegren H, and Møller AP, 1999. No evidence for adjustment of sex allocation in relation to paternal ornamentation and paternity in barn swallows. Mol Ecol 8:399–406.[CrossRef]
Sheldon BC, Andersson S, Griffith SC, Örnborg J, and Sendecka J, 1999. Ultraviolet colour variation influences blue tit sex ratios. Nature 402:874–877.[CrossRef]
Stamps, JA, 1990. When should avian parents differentially provision sons and daughters? Am Nat 135:671–685.[CrossRef][Web of Science]
Tidemann SC, 1986. Breeding in three species of fairy-wrens (Malurus): do helpers really help? Emu 86:131–138.[Web of Science]
Tidemann SC, and Marples TG, 1987. Periodicity of breeding behaviour of three species of fairy-wrens (Malurus spp) Emu 87:73–77.[Web of Science]
Trivers RL, and Willard DE, 1973. Natural selection of parental ability to vary the sex ratio of offspring. Science 179:90–92.
West SA, and Sheldon BC, 2002. Constraints in the evolution of sex ratio adjustment. Science 295:1685–1688.
Westerdahl H, Bensch S, Hansson B, Hasselquist D, and von Schantz T, 1997. Sex ratio variation among broods of great reed warblers Acrocephalus arundinaceus. Mol Ecol 6:543–548.[CrossRef][Web of Science]
Whittingham LA, and Dunn PO, 2000. Offspring sex ratios in tree swallows: females in better condition produce more sons. Mol Ecol 9:1123–1129.[CrossRef][Medline]
Whittingham LA, Valkenaar SM, Poirier NE, and Dunn PO, 2002. Maternal condition and nestling sex ratio in house wrens. Auk 119:125–131.[CrossRef][Web of Science]
Wilson K, and Hardy ICW, 2002. Statistical analysis of sex ratios: an introduction. In: Sex ratios: concepts and research methods, (Hardy ICW, ed). Cambridge: Cambridge University Press; 48–92.
Zar JH, 1999. Biostatistical analysis. Upper Saddle River, New Jersey: Prentice-Hall Inc.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  T. A.F Long and A. Pischedda Do female Drosophila melanogaster adaptively bias offspring sex ratios in relation to the age of their mate? Proc R Soc B, September 7, 2005; 272(1574): 1781 - 1787. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||