Behavioral Ecology Vol. 13 No. 4: 503-510
© 2002 International Society for Behavioral Ecology
Egg sex ratio and paternal traits: using within-individual comparisons
a Institute of Cell, Animal and Population Biology, University of Edinburgh, West Mains Road, Edinburgh EH9 3JT, UK b Department of Animal Ecology, Evolutionary Biology Centre, Uppsala Universitet, Norbyvägen 18D, Uppsala, S-752 36, Sweden c Zoologische Museum, Universität, Zürich-Irchel, Winterthurerstrasse 190, CH-8057 Zürich, Switzerland
Address correspondence to K.R. Oddie, who is now at the Centre d'Ecologie Fonctionelle et Evolutive, UPR 9056 CNRS, 1919 Route de Mende, F-34293, Montpellier CEDEX 5, France. E-mail: oddie{at}cefe.cnrs-mop.fr .
Received 20 March 2001; revised 11 October 2001; accepted 11 October 2001.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Empirical studies of sex ratios in birds have been limited due to difficulties in determining offspring sex. Since molecular sexing techniques removed this constraint, the last 5 years has seen a great increase in studies of clutch sex ratio manipulation by female birds. Typically these studies investigate variation in clutch sex ratios across individuals in relation to environmental characteristics or parental traits, and often they find no relationships. In this study we also found that clutch sex ratios did not vary in relation to a number of biological and environmental factors for 238 great tit Parus major nests. However, interesting sex ratio biases were revealed when variation in clutch sex ratios was analyzed within individual females breeding in successive years. There was a significant positive relationship between the change in sex ratio of a female's clutch from one year to the next and the relative body condition of her partner. Females mating with males of higher body condition in year x + 1 produced relatively male-biased sex ratios, and the opposite was true for females mated with lower condition males. Within-individual analysis also allowed investigations of sex ratio in relation to partner change. There was no change in sex ratios of females pairing with the same male; however, females pairing with a new male produced clutches significantly more female biased. Comparisons of clutch sex ratios within individuals may be a powerful method for detecting sex ratio variation, and perhaps female birds may indeed manipulate egg sex but require personal contextual experience for such decisions.
Key words: body condition, great tits, offspring sex ratios, Parus major.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Natural selection favors individuals that modify investment in male and female offspring when fitness benefits from producing each sex differ (e.g., Charnov, 1982
;
Fisher, 1930). In offspring
sex ratios are representative of the division of resources between sons and
daughters, then male-biased sex ratios are expected when the reproductive
value of male offspring exceeds that of females and vice versa. Manipulations
of the proportions of males and females produced are possible either at the
egg stage or through differential investment in offspring during the period of
care. Maternal control of egg sex ratios has been convincingly demonstrated in
Hymenoptera (e.g., Charnov and Bull,
1977; Herre,
1987), yet investigations of egg sex ratios among birds have
yielded disappointing results, despite the potential for egg sex control
because of female heterogamety in this taxon. The interest in egg sex ratio manipulation in birds follows the development of molecular techniques for sexing birds, which previously presented an obstacle to sex ratio studies. These techniques have been mainly used to search for correlational evidence of sex ratio variation in relation to a number of breeding gradients or traits. The premise for such studies is that the reproductive values of male and female offspring may vary with breeding conditions, and hence females benefit from sex-biased investment according to their particular breeding situation. These conditions include timing of breeding, intensity of brood competition (reflected in clutch sizes, brood sizes, or hatching asynchrony), male quality or attractiveness, female quality, environmental or territory quality, presence or absence of helper offspring, and brood status. Although there is clearly a need for experimental manipulations in further investigations of avian sex allocation, observations of natural sex ratio variation and sex ratio skews can offer a view of natural investment patterns of parents in male and female offspring and the degree to which we can expect sex ratios to be biased in the wild. Furthermore, any particularly consistent, significant trends within species, genera, or even the class should be recognizable.
To data however, evidence for such egg sex ratio biases has been variable.
Hatchling sex ratios have been associated with diverse ecological factors or
traits in wild bird populations. Sex ratios have been found to vary in
relation to resource abundance (Appleby et
al., 1997; Komdeur et al.,
1997; Korpimaki et al.,
2000), timing of breeding
(Daan et al., 1996;
Dijkstra et al., 1990;
Howe, 1977;
Lessells et al., 1996;
Sheldon et al., 1999;
Weatherhead, 1983;
Zijlstra et al., 1992), clutch
size (Lessells et al., 1996),
hatching asynchrony (Lessells et al.,
1996), brood status (Nishiumi,
1998; Patterson and Emlen,
1980; Westerdahl et al.,
2000), paternal traits
(Ellegren et al., 1996;
Kölliker et al., 1999;
Sheldon et al., 1999;
Svensson and Nilsson, 1996;
Westerdahl et al., 1997),
maternal traits (Blank and Nolan,
1983; Gowaty and Lennarz, 1985;
Heg et al., 2000;
Nager et al., 1999;
Whittingham and Dunn, 2000),
harem size (Nishiumi, 1998),
and helping activity (Gowaty and Lennartz,
1985; Ligon and Ligon,
1990). In laboratory studies, sex ratio variation has been
associated with parental attractiveness (Burley,
1981,
1986) and diet and maternal
quality (Bradbury and Blakey,
1998; Kilner,
1998). Many studies, however, report no significant sex ratio
biases in relation to a number of variables, even when such associations are
sometimes expected (European starling Sturnus vulgaris:
Bradbury et al., 1997; lesser
snow goose Anser c. caerulescens:
Cooch et al., 1997;
Harmsen and Cooke, 1983; corn
bunting Milaria calandra: Hartley
et al., 1999; western bluebird Sialia mexicana:
Koenig and Dickinson, 1996;
blue tit Parus caruleus: Leech et al., in press; yellowhammer
Emberiza citrinella: Pagliani et
al., 1999; yellow-headed blackbird Xanthocephalus
xanthocephalus: Patterson and Emlen,
1980; bluethroat Luscinia svecica:
Questiau et al., 2000; great
tit Parus major: Radford and
Blakey, 2000; barn swallow Hirundo rustica:
Saino et al., 1999). It is
impossible to estimate how many more studies remain unpublished due to
preferential publication of significant results. One study that does uncover a
strong primary sex ratio bias in Tengmalm's owl (Aegolius funereus)
broods can offer no explanation for their observation
(Hörnfeldt et al., 2000).
Consistent sex ratio trends do not appear to be emerging, though this may be
due to inappropriate assumptions on which our expectations of sex ratio biases
are based. At present it is difficult to make any generalizations about causes
of avian sex ratio variation or about the adaptive nature of skews.
Within species, reports of sex ratio variation have also been inconsistent.
In two Swedish blue tit populations, clutch sex ratios have been shown to vary
with paternal sexual traits and probability of survival
(Sheldon et al., 1999;
Svensson and Nilsson, 1996).
However, no effect of paternal quality or extrapair paternity on clutch sex
ratio has been found in a British population of the same species (Leech et
al., in press). Similarly, in great tits, a positive relationship between
hatchling sex ratio and male body size has been demonstrated in one population
(Kölliker et al., 1999)
but not in two others (Lessells CM, personal communication;
Radford and Blakey, 2000).
Furthermore, hatchling sex ratio biases within the same populations have
sometimes been inconsistent, with significant sex ratio biases in some years
but not others (Koenig and Dickinson,
1996; Korpimaki et al.,
2000; Radford and Blakey,
2000).
When studies involve data collected over more than 1 year, data are
typically analyzed for all years pooled, with year as a factor in a general
linear model (e.g., see Radford and
Blakey, 2000, who analyzed years both separately and pooled). We
suggest that looking for differences within recaptured breeding individuals
across years may be a fruitful alternative approach to analyzing sex ratio
variation in wild populations. A within-individual analysis permits
identification of factors influencing clutch sex ratio having removed nuisance
variables specific to individual birds. Furthermore, as female birds are
expected to bias sex ratios according to their particular breeding
circumstances, analyzing variation within individual females might present a
more powerful method for detecting egg sex ratio skews in relation to the
particular breeding environment experienced. This method may provide valuable
insight as to individual decisions concerning investment in young of different
sexes, yet only two published studies have examined the variance in sex ratio
due to variation of a factor within an individual. Westerdahl et al.
(2000) examined primary sex
ratio variation within female great reed warblers (Acrocephalus
arundinaceus) breeding in different years and found that individual
females had a higher proportion of sons in their brood when they were of
primary rather than secondary breeding status. Komdeur et al.
(1997) revealed a strong shift
in sex ratios of individual female Seychelles warblers (Acrocephalus
seychellensis) translocated to territories of different quality.
Analysis of sex ratio variation across different breeding attempts also
allows investigation of the effect of mate swapping (enforced or chosen),
which cannot otherwise be examined. Females might be expected to increase the
proportion of sons in their brood if male offspring have higher reproductive
values than their sisters do when fathered by a high-quality male. Given that
high-quality males are more likely than poorer conspecifics to survive to
breed the next year, and also that female great tits may remain faithful to
their mate if he is of particularly high quality
(Lindén, 1991), the
following consequences for clutch sex ratios can be envisaged. (1) We expect
females who mate with the same male to benefit from retaining their
high-quality male and produce a higher proportion of sons in their clutches in
year 2, but sex ratios of females who change partners not to change over the 2
years. (2) If females are changing males as an active strategy to upgrade
their partner, we expect sex ratios of such divorcing females to increase from
year 1 to year 2, but those of females who retain their partners to remain
constant.
For an island population of great tits, we first analyzed egg sex ratio variation in the traditional between-individual manner and then used a within-individual approach from a set of birds that were caught breeding in more than 1 year. We also explored whether clutches of different sex ratios are of different value to females by looking for associations between sex ratio and hatching success and sex ratio and nest desertion. We encourage similar analyses in other bird populations.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Field work
Data were collected from great tits breeding in nest-boxes in 14 separate woodlands on the Swedish island of Gotland (57°10' N, 18°20' E) between 1997 and 1999. For each nest we recorded lay date (first egg), clutch size, hatch data (first egg), and the number of eggs hatching. Parents were caught and ringed (if not already ringed) while provisioning broods between 8 and 14 days after hatching. We measured parental tarsus length (to nearest 0.5 mm), wing length (to nearest 0.5 mm), and mass (to nearest 0.5 g), and we recorded age as either 1 year old or 2+ years. We calculated body condition index of adults as the residual from a linear regression of body mass on tarsus length. Measurements of male birds fitted a linear regression of mass on tarsus length better than mass on tarsus length cubed. There was little difference in fit, however, and we used residuals from mass on simple tarsus length to indicate body condition in analyses. Blood or tissue samples were required from each egg/offspring for sex determination. A 2-10 µl blood sample was taken from 1- or 2-day old nestlings by puncturing the brachial vain and collecting in a capillary, which was then stored in SET buffer at 4°C (1997) or 98% ethanol (1998 and 1999). We collected unhatched eggs 4 days or more after the hatch date of the first egg. (Only on rare occasions do eggs hatch more than 2 days after the first hatch date.) Blastocysts (seen as white spots on the yolk surface) and embryos were dissected out immediately and stored in the same way as blood samples.
Molecular sexing
We determined clutch sex ratios using a polymerase chain reaction
(PCR)-based molecular technique from DNA extracted from blood samples or
embryonic tissue from unhatched eggs. We used primers P2 and P8
(Griffiths et al., 1998) to
amplify introns within the CHD1 gene. PCR products were run on 6%
polyacrylamide gel for between 1 and 3 h at 75 W and visualized using silver
staining (Promega, 1996).
Female nestlings/embryos possessed two different-length copies of the PCR
products: CHD1-W from the W chromosome and CHD1-Z from the Z chromosome. Males
possessed only one copy, CHD1-Z, because males are homogametic. We sexed 74
individual adults phenotypically, and molecular sex matched in all cases.
Analysis between individuals in 3 years
We analyzed clutch sex ratio variation in relation to a number of factors
from pooled breeding data from 238 nests, collected over 3 years
(Table 1). We had data from 276
nests, but 38 individuals breeding in more than 1 year were included only once
to avoid pseudoreplication. In each such case the breeding attempt that was
included was selected randomly. We measured sex ratio as the proportion of
males in a clutch. Because of non-normally distributed error variance and
unequal sample sizes, we analyzed the proportional data with a general linear
model analysis of deviance, assuming binomial errors, and a logit link
function. The response variable was the number of males in a clutch, with the
number of eggs sexed as the binomial denominator. Using clutch size as the
denominator would lead to overrepresentation of females as not all eggs were
sexed, and those we failed to sex would be categorized as "not
male" in the analyses. Analyses presented here were weighted according
to the amount of information we had for each clutch (i.e., the proportion of a
clutch sexed [total sexed/clutch size]). Results did not differ if analyses
were repeated on clutches only with complete sex ratio data.
|
A model including several predictor variables and their second-order
interactions with year was first fitted to the data. The significance of a
term in the model was determined by assessing the change in deviance after
removal of that term, using a chi-square test with appropriate degrees of
freedom (Crawley, 1993).
Reported in this paper are changes in deviance values after removal of each
variable alone rather than sequentially because of the presence of missing
cases within the whole data set (see Table
1). A new general linear model was made for each factor, still
complete with all other terms and interactions, but excluding cases with
missing values for the factor in question. If cases with missing values are
not excluded in analyses, the change in degrees of freedom when the variable
in question is dropped from the model is inflated. Results did not differ when
variables were removed first from the model or later after stepwise exclusion
of other variables. In the latter analyses clutch size was never included
simultaneously with other factors in models because number of eggs sexed had
been included as the binomial denominator and was not independent of clutch
size.
We calculated a heterogeneity factor (HF), the ratio of residual deviance
to the residual degrees of freedom, to examine the data for overdispersion. A
value of HF < 1 indicates the variance in the data to be less than that
expected for a binomial distribution, and HF > 1 indicates more variance
than that expected. Here HF = 0.995, and so fitting the data to a binomial
model was justified, and scaling analyses by the HF did not change results
(Krackow and Tkadlek, 2001).
Analyses were carried out using the statistical package GLMStat
(Beath, 2000).
Associated questions: analyses with sex ratio as independent
variable
We examined the effect of clutch sex ratio on hatching success and nest
desertion to investigate whether females with different clutch sex ratios
invest differently in those clutches. To test whether hatching success varied
significantly with clutch sex ratio, we analyzed the proportion of eggs
hatching per clutch with a GLM with binomial errors and logit link. The number
of eggs hatching was used as the response variable, with clutch size as the
binomial denominator. Sex ratio was included singly as a predictor variable
and whether the removal of this term caused a significant increase in deviance
was assessed with an F test. Model deviance was scaled by the HF
(2.81) because the data were overdispersed, hence the application of an
F rather than a chi-square test. To test whether nest desertion was
associated with clutch sex ratio, we carried out a logistic regression
analysis of brood desertion in relation to the proportion of males in a
clutch. We included year, lay date, and clutch size as potential predictor
variables.
Analysis within individuals across 2 years
Due to the fact that females, being heterogametic, have the potential to
control clutch sex ratios rather than males, the following analyses were
performed on 23 female birds where sex ratios over 2 years were known. Data
for individual males where brood sex ratio was known over 2 years is presented
for comparison (n = 23). Birds were identified which nested in 1998
following 1997 and in 1999 following 1998. First we determined whether clutch
sex ratio was repeatable within individuals across years, using a simple
regression of sex ratio in the second year against the first. Sex ratio data
for females' clutches in year 2 were first arcsine square-root transformed
because of their proportional nature (Shapiro Wilks W = 0.928;
p =.030). Other sex ratio data were not transformed (females year 1:
Shapiro Wilks W = 0.964, p =.426; males year 1: Shapiro
Wilks W = 0.982, p =.089; males year 2: Shapiro Wilks
W = 0.975, p =.709). To find the variance in clutch sex
ratio due to individual birds, we also ran a general linear model with
binomial error structure and logit link including bird ID as a factor, with
number of males as the response variable and number of eggs sexed as the
binomial denominator. The change in deviance when bird ID was removed from the
model indicated the proportion of the variance attributable to between
individual differences.
We analyzed the change in egg sex ratio from one year to the next for
females, in relation to the change in their own body condition, laying date
and clutch size, with linear regressions. Not enough data existed to test for
effects of age, although it is possible that breeding experience of females
may influence clutch sex ratios, especially comparing first-time breeders with
others (e.g., Blank and Nolan,
1983; Heg et al.,
2000). In the same way, we analyzed the change in females' sex
ratio in relation to the change in their partners' quality relative to that of
their mate the previous year. Analyses were repeated for males. The
independent variable change in sex ratio (sex ratio year 2-sex ratio year 1)
was not transformed because data were distributed normally (females: Shapiro
Wilks W = 0.976, p =.841; males: Shapiro Wilks W =
0.975, p =.792).
Over two breeding attempts, a female may mate with either the same male in both years (same pair) or change her partner (new pair). We compared the clutch sex ratios of same-pair and new-pair females. We repeated these analyses for males. Retaining the male or mating with a new partner may be an active female choice (i.e., divorce) or a passive decision through survival of the male to the next breeding season (i.e., widowed). In this population, over a 3-year period, 24 new-pair females were caught. Of these 24, in only 3 cases was the original male caught breeding elsewhere in year 2. Admittedly, the other 21 males may have been breeding in natural holes, but these data strongly suggest that females mate with a new partner because of male overwinter mortality rather than choice. In this case we expect females mating with a male in good condition (assumed to be of high quality) to produce higher proportions of male eggs in year 2, whereas females pairing with a different male are not expected to change clutch sex ratios (scenario 1 presented in Introduction). We also examined whether there was a difference in the clutch sex ratio of females in the first breeding attempt according to their future mate category, same pair or new pair.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Analysis between individuals in 3 years
A total of 238 nests with known sex ratios were included in the analysis. Of these, 139 nests had complete sex ratio data (i.e., proportion of eggs sexed = 1). If data were missing, it was usually not more than 1 egg per nest; for 204 nests (86%), the proportion of eggs sexed was > 0.8. In total we sexed 2034 individual eggs and nestlings. Overall population sex ratios analyzed at the level of the nestling showed no deviation from a 1:1 sex ratio in any year (Table 2). For the 3 years pooled, the overall sex ratio of broods showed binomial distribution of the sex ratio (residual deviance/residual degrees of freedom = 236/237
1).
|
Analysis of the proportion of males per clutch for all 3 years pooled provided no evidence of systematic biases in sex ratio in relation to any factors included in a general linear model with binomial errors. Table 1 lists variables fitted to the model and the change in deviance and related p value when each term was removed individually from the full model. No variables explained significant variation in the proportion of males in a brood, whether they were removed first from the model or later after stepwise exclusion of other variables which contributed least to the variance.
Associated questions: analyses with sex ratio as independent
variable
The proportion of eggs hatching did not depend on the proportion of male
eggs in the clutch (
D = 2.122, p >.050, df = 1).
Neither was there any indication of the proportion of male eggs affecting the
likelihood of nest desertion (
21 = 0.323,
p =.570, n = 238).
Analysis within individuals across 2 years
There were 51 cases of female birds nesting in one year being recaptured in
a subsequent breeding attempt, and 44 cases of male breeding recaptures. Of
these, sex ratios of broods in both years were known for 23 females and 23
males (not necessarily paired). One female and two males were caught breeding
in all 3 years. In these cases one breeding attempt was selected at random for
each individual to be included in the analysis to avoid pseudo-replication in
the data set. Not all data (e.g., exact lay dates, adult measures) were
available for breeding attempts in both years; therefore sample sizes in
analyses may not always equal 23.
A simple linear regression of egg sex ratio of the second breeding attempt
against the first suggested that sex ratio was not repeatable across years
either for females (F1,20 = 1.094, p =.308) or
males (F1,20 = 0.117, p =.736). We examined the
variance in sex ratios due to individual birds in a general linear model with
binomial error structure, essentially testing whether differences within an
individual were smaller than differences between individuals. An insignificant
proportion of the variance in sex ratio was accounted for by between-subject
differences (females: D = 24.41, df = 22, p
>.30, HF = 1.3; males: D = 25.29, df = 22,
p >.20, HF = 1.2).
The following results involve analyses of the change in sex ratio within individual birds (i.e., sex ratio year 2-sex ratio year 1) with respect to the change in predictor variables from one year to the next. The change in egg sex ratio of female birds was positively related to the change in body condition of the male to which she was mated (F1,9 = 9.698, p =.012; Table 3 and Figure 1a). Applying a Bonferroni correction for multiple comparisons gives a new p statistical threshold of.013, and our result remains significant below this corrected value (Table 3). Females did not alter the sex ratio they produced in response to changes in any other variables—either to their own body condition or with respect to clutch size or laying date (Table 3). In comparison to females, clutch sex ratios of males breeding over successive years did not change consistently with regard to their mate's quality (Figure 1b).
|
|
Given this result, we returned to the first analysis between all
individuals across all years to test for a relationship between male body
condition and clutch sex ratio, but restricting analyses to only females aged
2 years or more (i.e., not their first breeding attempt). Among these 50
females there was no relationship between absolute male body condition and sex
ratio (D = 0.08, df = 1, p >.70, HF =
1.04). It appears that females respond to male body condition relative to
previous breeding attempts; at least we could detect no change in sex ratio in
relation to absolute male body condition, even for experienced breeders
only.
Finally, we compared the sex ratios of same-pair and new-pair males and females. Same-pair females showed no change in their clutch sex ratio (t = 0.265 p > t = 0.398), but those pairing with a new male showed a decrease in the proportion of males in their clutch (t = -2.103, p < t = 0.034; Figure 2a). The difference between the proportion of males in broods of same-pair and new-pair females was not statistically significant (t = 1.498, p =.156, n = 20). Clutch sex ratios of same-pair and new-pair males did not differ (t = -0.590, p =.563, n = 20; Figure 2b). There was no difference in clutch sex ratios in the first breeding attempt between same-pair and new-pair females (t = -1.037, p =.309, n = 29).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Despite abundant theories and recent interest in egg sex ratio studies in birds, empirical evidence of consistent sex ratio biasing by females is elusive. Identifying factors responsible for sex ratio biases through a correlative approach across individuals is a commonly used method for identifying relationships between variables and primary sex ratio variation. Like many other investigations, we found no correlative evidence of primary sex ratio biasing by great tits over 3 years, from a large data set of nearly 240 broods. Brood sex ratio did not vary in relation to several environmental variables or parental traits (Table 1).
Previous investigations of heterogeneity in great tit brood sex ratios at
the egg stage have been published from three different populations. In a Dutch
population, brood sex ratios increased with hatching date and hatching
asynchrony, whereas they decreased with increasing clutch size
(Lessells et al., 1996). The
proportion of sons in broods of a Swiss population of tits increased
significantly with increasing male tarsus length, and there was a similar,
though nonsignificant, trend with increasing breast stripe size
(Kölliker et al., 1999).
(Male breast band stripe was not measured in the present study because of time
constraints of fieldwork.) Great tit females prefer males with larger breast
stripes (Norris, 1990), and
male tarsus length correlates with breeding success
(Blakey, 1994;
Verboven and Mateman, 1997).
The relationship between sex ratio and male traits was interpreted as a female
response to either male genetic quality or body-size related territory quality
(or both). This relationship with tarsus length was not apparent in the
present study, despite a much larger sample size (n = 173 compared to
n = 57), nor in the Dutch population.
A third great tit study suggested one potential explanation for the
discrepancy in these findings. Radford and Blakey
(2000) found significant
predictors of brood sex ratio from a correlational analysis based on 5 years
of breeding data from British great tits. These included lay date (1993), male
age (1998), male tarsus length (1991), female tarsus length (1991 and 1998),
and female condition (1991). However, no relationships were consistent across
years, and no variables predicted sex ratios when data for all years were
combined (Radford and Blakey,
2000). They argued for evaluation of breeding data from several
years in order to identify consistent sex ratio biasing and confirm whether
sex ratio manipulation is truly a female breeding strategy. The previous sex
ratio biases reported from the Swiss and Dutch populations examined sex ratio
biases over 1 year, and one of these relationships has proved unrepeatable in
subsequent years (Lessells CM, personal communication).
These previous studies have all examined sex ratio variation among individuals. As we expect sex-ratio adjustment to be performed by a female in response to her particular situation, adjustment is unlikely to be detected by averaging all individuals of a population because they face various microenvironments. Fine-tuned sex-ratio adjustment in response to particular factors is more likely to be detected when examining variation within female individuals, as done here. By using a within-individual approach, we found interesting results for a sample of great tits in which no variables explained significant variance in observed sex ratios using a traditional analysis with year as an explanatory variable.
We found that females mated to males of better condition relative to their
last breeding attempt tended to increase their clutch sex ratios, and females
mated with males in relatively worse condition adjusted their brood sex ratios
negatively (Figure 1). This
relationship between sex ratio and mate's body condition was evident for
recaptured females but was absent for recaptured males because in birds the
female is heterogametic and thus responsible for egg sex ratios. Our results
suggest bidirectional female control of sex ratios. They are complementary to
those found by Kölliker et al. and from blue tits
(Sheldon et al., 1999;
Svensson and Nilsson, 1996),
where females were found to adjust sex ratios to increase the proportion of
males in their brood with increasing male quality. This could be considered
adaptive if high-quality (good body condition) males father high-quality sons
relative to daughters. Body condition has been shown to be moderately to
highly heritable in the closely related blue tit
(Merilä et al., 1999),
although how this varies with offspring sex is not known. Heritability
estimates are determined from regressions of offspring measures on measures
from the same-sex parent (i.e., male—father regressions and
female—mother regressions; Falconer,
1981). Male—father regressions have revealed high
heritability values for body weight (van Noorwijk et al., 1980) among great
tits and similarly high heritability of both weight and tarsus length in
another passerine with similar breeding ecology
(Gustafsson, 1986). With a
significant heritable component of body condition, females mating with males
in better condition could potentially increase grand-offspring production
through rearing more male offspring in their broods.
If all females modify sex ratios in respect to their partner's condition, one might expect a positive relationship between absolute male body condition and clutch sex ratios, which was not evident from a much larger data set (n = 164), even when the analysis was restricted to females aged 2 years or more (n = 50). One explanation for this discrepancy may be that females are responding to relative body condition based on that of previous mating events, and without a benchmark females are unable to respond to partner condition. Female control of the sex ratio could then only be expected to evolve if a significant proportion of the female population bred in more than 1 year. In this population, between 35% and 46% of female birds breed in more than 1 year (Oddie and Reim, unpublished data).
There was further evidence of the importance of relative mate quality from
analyses of partner fidelity. Same-pair females tended to have higher clutch
sex ratios relative to the last breeding attempt than new-pair females,
although this result is not statistically significant. Same-pair females
showed no change in sex ratio from one year to the next, but new-pair females
produced a higher proportion of female eggs. Clutch sex ratios in the first
breeding attempt did not differ between same-pair and new-pair females. This
further indicates that the differing sex ratios of same-pair and new-pair
females is a result of decreasing sex ratios among new-pair females, rather
than an increase in sex ratio among same-pair females. The result differs from
our prediction of increasing sex ratios among same-pair females and constant
sex ratios among new-pair females; however, new-pair females do still show a
decrease in sex ratio compared to same-pair females. In great tits, newly
formed pairs have lower breeding success than birds previously breeding
together (Perrins and McCleery,
1985), and pairs that enjoy high reproductive success together are
more likely to remain together
(Lindén, 1991). Females
with a new partner may be less sure of his parenting abilities (e.g., food
provisioning) than those paired to the same male, and consequently produce
more of the relatively smaller sex (females; e.g.,
Oddie, 2000;
Perrins, 1963), which require
less food. Alternatively, new-pair females may produce more female offspring
in their next clutch because of intrinsic male qualities (i.e., they are
mating with inferior males). Perhaps in great tits maintenance of the sex
ratio in a second year of breeding is a luxury only afforded by those who mate
with the same partner, and the default tactic is to decrease the sex ratio
with a new partner.
Together these results suggest that investigating sex ratio variation
within individuals over different breeding attempts may provide a more
powerful means of identifying factors causing sex ratio skews in birds than a
conventional across-individuals approach. These analyses have two advantages.
First, within-individual analyses of sex ratios control for any unexplained
variation due to individual birds. Second, they allow detection of sex ratio
changes relative to previous breeding experiences, if a previous breeding
attempt is a prerequisite for sex ratio biasing. Two previous studies have
examined repeatability of sex ratios of individual females
(Appleby et al., 1997;
Westerdahl et al., 1997), and
two others have found sex ratios to vary within individuals according to
breeding territory (Komdeur et al.,
1997) and breeding status
(Westerdahl et al., 2000). No
studies have considered within-individual variation in relation to more than
one determining factor. We found that great tit clutch sex ratios of
individual birds were not repeatable across years, suggesting that the
proportion of males in a brood is not fixed for each female but varies between
breeding attempts. The fact that we also found clutch sex ratio variation
among male birds between years may lead us to question any conclusions that
could be drawn. At least this variation allows that birds may potentially
adjust sex ratios according to environmental or mate characteristics.
One significant limitation of within-individual comparisons is that sample
sizes will inevitably be limited due to low numbers of birds caught breeding
in successive years. Data presented here suggest that females may adjust egg
sex ratios in relation to mate quality (see also Burley,
1981,
1986;
Ellegren et al., 1996;
Kölliker et al., 1999;
Sheldon et al., 1999;
Svensson and Nilsson, 1996).
It would be interesting to repeat these investigations using within-individual
comparisons of primary sex ratios from much larger data sets. We hope this
study will encourage further within-individual analyses. Furthermore, we
advocate accompanying experimental approaches to demonstrate causal
relationships in sex ratio studies—for example, experimental
manipulation of partner choice (e.g., partner removal experiments) and its
effects on relative sex ratio. However, whether studies are experimental or
correlational, we believe that a new within-individual approach could reveal
egg sex ratio biases, as females respond to their particular breeding
conditions.
|
ACKNOWLEDGEMENTS |
|---|
This work was carried out under Swedish ringing license no. 576 with Natural Environment Research Council funding to K.O. We thank local landowners of Burgsvik for land access, and other researchers on Gotland for help gathering data, in particular Andy Russell. Thanks to Simon Griffith and Richard Griffiths for lab advice, to Bart kempenaers and Loeske Kruuk for discussion, and Wolf Blanckenhorn, Ben Sheldon, Marcel Lambrechts, and two anonymous referees for comments on the manuscript.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Appleby BM, Petty SJ, Blakey JK, Rainey P, Macdonald DW, 1997. Does variation of sex ratio enhance reproductive success of offspring in tawny owls (Strix aluco)? Proc R Soc Lond B 264: 1111-1116.
Beath KJ, 2000. GLMStat User Manual version 5.2. Available at
http://www.ozemail.com.au/
kjbeath/glmstat.html
.
Blakey JK, 1994. Genetic evidence of extra-pair fertilisations in a monogamous passerine, the great tit Parus major. Ibis 136: 457-462.
Blank JL, Nolan V Jr., 1983. Offspring sex ratio in
red-winged blackbirds is dependent on maternal age. Proc Natl Acad Sci
USA 80:
6141-6145.
Bradbury RB, Blakey JK, 1998. Diet, maternal condition, and offspring sex ratio in the zebra finch, Poephila guttata. Proc R Soc Lond B 265: 895-899.
Bradbury RB, Cotton PA, Wright J., Griffiths R, 1997. Nestling sex ratio in the European starling Sturnus vulgaris. J Avian Biol 28: 255-258.
Burley N, 1981. Sex ratio manipulation and selection
for attractiveness. Science 211:
721-722.
Burley N, 1986. Sex-ratio manipulation in color-banded populations of zebra finches. Evolution 40: 1191-1206.[ISI]
Charnov EL, 1982. The theory of sex allocation. Monographs in population biology. Princeton, New Jersey: Princeton University Press.
Charnov EL, Bull JJ, 1977. When is sex environmentally determined? Nature 266: 828-830.[Medline]
Cooch E, Lank D, Robertson R, Cooke F, 1997. Effects of parental age and environmental change on offspring sex ratio in a precocial bird. J Anim Ecol 66: 189-202.
Crawley M, 1993. GLIM for ecologists. Oxford: Blackwell.
Daan S, Dijkstra C, Weissing FJ, 1996. An evolutionary
explanation for seasonal trends in avian sex ratios. Behav Ecol
7: 426-430.
Dijkstra C, Daan S, Buker JB, 1990. Adaptive seasonal variation in the sex ratio of kestrel broods. Funct Ecol 4: 143-147.
Ellegren H, Gustafsson L, Sheldon BC, 1996. Sex ratio
adjustment in relation to paternal attractiveness in a wild bird population.
Proc Natl Acad Sci USA 93:
11723-11728.
Falconer DS, 1981. Introduction to quantitative genetics. London: Longman.
Fisher RA, 1930. The genetical theory of natural selection. Oxford: Oxford University Press.
Gowaty PA, Lennartz MR, 1985. Sex-ratios of nestling and fledgling red-cockaded woodpeckers. Am Nat 126: 347-353.
Griffiths R, Double MC, Orr K, Dawson RJG, 1998. A DNA test to sex most birds. Mol Ecol 7: 1071-1075.[Medline]
Gustafsson L, 1986. Lifetime reproductive success and heritability: empirical support for Fisher's fundamental theorem. Am Nat 128: 761-764.
Harmsen R, Cooke F, 1983. Binomial sex-ratio distribution in the lesser snow goose: a theoretical enigma. Am Nat 121: 1-8.
Hartley IR, Griffith SC, Wilson K, Shepherd M, Burke T, 1999. Nestling sex ratios in the polygynously breeding corn bunting Milaria calandra. J Avian Biol 30: 7-14.
Heg D, Dingemanse NJ, Lessells CM, Mateman AC, 2000. Parental correlates of offspring sex ratio in Eurasian oystercatchers. Auk 117: 980-986.
Herre EA, 1987. Optimality, plasticity and selective regime in fig wasp sex ratios. Nature 329: 627-629.
Hörnfeldt B, Hipkiss T, Fridolfsson A-K, Eklund U, Ellegren H, 2000. Sex ratio and fledging success of supplementary-fed Tengmalm's owl broods. Mol Ecol 9: 187-92.[Medline]
Howe HF, 1977. Sex-ratio adjustment in the common
grackle. Science 198:
744-746.
Kilner R, 1998. Primary and secondary sex ratio manipulation by zebra finches. Anim Behav 56: 155-164.[ISI][Medline]
Koenig WD, Dickinson JL, 1996. Nestling sex ratio variation in western bluebirds. Auk 113: 902-910.
Kölliker M, Heeb P, Werner I, Mateman AC, Lessells CM, 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, Daan S, Tinbergen J, Mateman C, 1997. Extreme adaptive modification in sex ratio of the Seychelles warbler's eggs. Nature 385: 522-525.
Korpimaki E, May CA, Parkin DT, Wetton JH, Wiehn J, 2000. Environmental- and parental condition-related variation in sex ratio of kestrel broods. J Avian Biol 31: 128-134.
Krackow S, Tkadlek E, 2001. Analysis of brood sex ratios: implications of offspring clustering. Behav Ecol Sociobiol 50: 293-301.
Leech DI, Hartley IR, Stewart IRK, Griffith SC, Burke T,
2001. No effect of parental quality or extrapair paternity on
brood sex ratio in the blue tit (Parus caeruleus). Behav
Ecol 12:
674-680.
Lessells CM, Mateman AC, Visser J, 1996. Great tit hatchling sex ratios. J Avian Biol 27: 135-142.
Ligon JD, Ligon SH, 1990. Female-biased sex ratio at hatching in the green woodhoopoe. Auk 107: 765-771.
Lindén M, 1991. Divorce in great tits—chance or choice? An experimental approach. Am Nat 138: 1039-1048.
Merilä J, Przybylo R, Sheldon BC, 1999. Genetic variation and natural selection on blue tit body condition in different environments. Genet Res 73: 165-176.
Nager R, Monaghan P, Griffiths R, Houston DC, Dawson R,
1999. Exper imental demonstration that offspring sex ratio varies
with maternal condition. Proc Natl Acad Sci USA
96: 570-573.
Nishiumi I, 1998. Brood sex ratio is dependent on female mating status in polygynous great reed warblers. Behav Ecol Sociobiol 44: 9-14.
Norris KJ, 1990. Female choice and the evolution of the conspicuous plumage coloration of monogamous male great tits. Behav Ecol Sociobiol 26: 129-138.
Oddie KR, 2000. Size matters: competition between male and female great tit offspring. J Anim Ecol 69: 1-11.
Pagliani AC, Lee PLM, Bradbury RB, 1999. Molecular determination of sex ratio in yellowhammer Emberiza citrinella offspring. J Avian Biol 30: 239-244.
Patterson CB, Merritt Emlen J, 1980. Variation in nestling sex ratios of the yellow-headed blackbird. Am Nat 115: 743-747.
Perrins CM, 1963. Some factors influencing brood size and populations in great tits (PhD thesis). Oxford: Oxford University.
Perrins CM, McCleery RH, 1985. The effect of age and pair bond on the breeding success of great tits Parus major. Ibis 12: 306-315.
Promega, 1996. GenePrint STR systems technical manual notice of protocol change. Madison, Wisconsin: Promega.
Questiau S, Escaravage N, Eybert MC, Taberlet P, 2000. Nestling sex ratios in a population of Bluethroats Luscinia svecica inferred from AFLP (TM) analysis. J Avian Biol 31: 8-14.
Radford AN, Blakey JK, 2000. Is variation in brood sex
ratios adaptive in the great tit (Parus major)? Behav
Ecol 11:
294-298.
Saino N, Ellegren H, 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.
Sheldon BC, Andersson S, Griffith SC, Ornborg J, Sendecka J, 1999. Ultraviolet colour variation influences blue tit sex ratios. Nature 402: 874-877.
Svensson E, Nilsson J-A, 1996. Mate quality affects offspring sex ratio in blue tits. Proc R Soc Lond B 263: 357-361.
van Noordwijk AJ, Balen JH, Scharloo W, 1980. Heritability of ecologically important traits in the great tit Parus major. Ardea 68: 193-203.
Verboven N, Mateman AC, 1997. Low frequency of extra-pair fertilisations in the great tit (Parus major) revealed by DNA fingerprinting. J Avian Biol 28: 231-239.
Weatherhead PJ, 1983. Secondary sex ratio adjustment in red-winged blackbirds (Agelius phoeniceus). Behav Ecol Sociobiol 12: 57-61.
Westerdahl H, Bensch S, Hansson B, Hasselquist D, von Schantz T, 1997. Sex ratio variation among broods of great reed warblers Acrocephalus arundinaceus. Mol Ecol 6: 543-548.
Westerdahl H, Bensch S, Hansson B, Hasselquist D, von Schantz T, 2000. Brood sex ratios, female harem status and resources for nestling provisioning in the great reed warbler (Acrocephalus arundinaceus). Behav Ecol Sociobiol 47: 312-318.
Whittingham LA, Dunn P, 2000. Offspring sex ratios in tree swallows: females in better condition produce more sons. Mol Ecol 9: 1123-1129.[Medline]
Zijlstra M, Daan S, Bruinenberg-Rinsma J, 1992. Seasonal variation in the sex ratio of marsh harrier Circus aeruginosus broods. Funct Ecol 6: 553-559.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  P. Korsten, C. M. Lessells, A. C. Mateman, M. van der Velde, and J. Komdeur Primary sex ratio adjustment to experimentally reduced male UV attractiveness in blue tits Behav. Ecol., July 1, 2006; 17(4): 539 - 546. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Dreiss, M. Richard, F. Moyen, J. White, A.P. Moller, and E. Danchin Sex ratio and male sexual characters in a population of blue tits, Parus caeruleus Behav. Ecol., January 1, 2006; 17(1): 13 - 19. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A.N. Rutstein, H.E. Gorman, K.E. Arnold, L. Gilbert, K.J. Orr, A. Adam, R. Nager, and J.A. Graves Sex allocation in response to paternal attractiveness in the zebra finch Behav. Ecol., July 1, 2005; 16(4): 763 - 769. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||