Behavioral Ecology Advance Access published online on September 14, 2007
Behavioral Ecology, doi:10.1093/beheco/arm082
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Male age predicts extrapair and total fertilization success in the socially monogamous coal tit
a Institute for Evolutionary Biology and Ecology, University of Bonn, An der Immenburg 1, D-53121 Bonn, Germany b Institute of Avian Research "Vogelwarte Helgoland", An der Vogelwarte 21, D-26386 Wilhelmshaven, Germany c Institute for Evolution and Biodiversity, University of Münster, Hüfferstr. 1, D-48149 Münster, Germany
Address correspondence to T. Schmoll. E-mail: tschmoll{at}evolution.uni-bonn.de.
Received 4 June 2007; revised 18 July 2007; accepted 1 August 2007.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONFUNDINGREFERENCES |
|---|
An important step to a better understanding of the adaptive significance of extrapair mating behavior in socially monogamous species is to uncover the identity of extrapair sires (EPS). Here, we used a combination of multilocus DNA fingerprinting and microsatellite analysis to identify EPS in the socially monogamous coal tit (Periparus ater), a passerine bird with high rates of extrapair paternity. We then analyzed how fertilization success was related to male age in 2 consecutive first brood periods based on knowledge of the exact age of the majority of territorial males. EPS were significantly older compared with the males they cuckolded. Furthermore, extrapair and, as a consequence, also total fertilization success were positively related to male age, while within-pair success was not. Interestingly, fertilization success did not increase linearly with male age but leveled off for older age classes and was most parsimoniously described by the inverse term of male age. Results were consistent over the 2 years, while the demography of the study population differed with respect to the age distribution of territorial males. Furthermore, we also show that individual males increased their extrapair fertilization (EPF) success with age indicating that cross-sectional analyses were not confounded by cohort effects. Together with the results from other species, these findings suggest that male age (or a strong correlate thereof) is a major determinant of EPF success in several socially monogamous bird species.
Key words: extrapair paternity, fertilization success, male age, mating preference, Parus ater, Periparus ater.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONFUNDINGREFERENCES |
|---|
An ever-increasing number of molecular studies on genetic parentage in wild bird populations demonstrate that the occurrence of extrapair paternity (EPP) is the rule rather than an exception, including strictly socially monogamous bird species (reviewed in Griffith et al. 2002
; Westneat and Stewart 2003). The evolutionary causes and consequences of extrapair mating behavior, however, are not yet well understood (e.g., Griffith et al. 2002; Westneat and Stewart 2003).
When studying extrapair mating systems in socially monogamous species, straightforward and important questions are How do the cuckolders (extrapair sires, EPS) differ from the males they cuckold? and Which traits affect variation in extrapair fertilization (EPF) success between males? An important step to a better understanding of extrapair mating behavior therefore is to uncover the identity of EPS in order to test for phenotypic and genotypic correlates of male fertilization success. A number of such correlates have been identified so far including sexually selected ornamental traits (plumage coloration and song), social dominance status, body size, and body condition (cf. Table 3 in Griffith et al. 2002). The most consistently observed predictor of EPF success across socially monogamous bird species, however, is likely to be male age (cf. Table 3 in Griffith et al. 2002). Older males are usually more successful (Griffith et al. 2002), although even within the same species the evidence may be mixed, for example, in the blue tit Cyanistes caeruleus (Kempenaers et al. 1997; Delhey et al. 2003; Charmantier et al. 2004).
Older males may have an advantage in male–male competition over EPFs simply because they are more experienced (Weatherhead and Boag 1995), for example, when they have to trade off mate guarding their own females against extrapair advertising directed toward other females (Johnsen et al. 2003). Furthermore, life-history trade-offs may favor males that delay their investment in extrapair mating, which would also lead to a positive relationship between age and EPF success.
A female mating preference for older males may have evolved through indirect selection given a positive correlation of age (or longevity) with fitness (Brooks and Kemp 2001; but see Hansen and Price 1995 for an alternative model). Older males have demonstrated a certain ability to survive, and this ability may—at least in part—be caused by a superior genetic quality (in terms of viability) compared with nonsurvivors. Mating with an older male thus increases a female's probability to mate with a high-quality male and therefore selects for the mating preference for older males. A female extrapair mating preference for older males would be particularly supportive of the idea that females gain genetic viability benefits (Brooks and Kemp 2001) because potentially confounding direct benefits are thought to be of minor, if any, importance (Griffith et al. 2002).
Here, we analyzed how extrapair, within-pair, and total fertilization success are related to male age in a nest-box population of the coal tit (Periparus ater, formerly Parus ater), a socially monogamous passerine with high levels of EPP (Lubjuhn et al. 1999; Dietrich et al. 2004a). We used a combination of multilocus DNA fingerprinting and microsatellite genotyping to assign extrapair offspring (EPO) to EPS with high reliability. We draw on a strong database including an exceptionally detailed knowledge of the age of individual males. In most studies up to date, male age was only rarely exactly known, and hence, most analyses relied on age determination based on field measurements (e.g., via plumage characteristics or capture history) and were restricted to comparisons between groups of yearling versus older (e.g., Delhey et al. 2003) or newly captured versus recaptured males (e.g., Bouwman et al. 2007). The exact relationship between male age and fertilization success and the reliable estimation of effect sizes are therefore largely unexplored in the literature. In contrast, in our study population the majority of adult breeding birds had been banded as nestlings and their exact age was therefore known. Many studies also used cross-sectional approaches, which are at a considerable risk to be confounded by cohort effects. Furthermore, analyses based on data from a single year may be limited in their general explanatory power due to unexplained year effects frequently observed in the study of natural populations. In order to avoid these shortcomings, we used data from 2 consecutive first brood periods allowing for longitudinal analyses within individual males as well as between-year comparisons.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONFUNDINGREFERENCES |
|---|
Study species, study population, and general field methods
Coal tits are small, territorial, altricial, cavity-nesting passerine birds with biparental care (Glutz von Blotzheim and Bauer 1993
). They are socially monogamous (Glutz von Blotzheim and Bauer 1993) but show comparatively high rates of EPP (see appendix in Griffith et al. 2002). Within the well-investigated Paridae, these are the highest rates recorded so far (Lubjuhn et al. 1999; Dietrich et al. 2004a).
We studied an established nest-box population of coal tits in a mixed coniferous forest near Lingen/Emsland (Lower Saxony, Germany, 52°27'N, 7°15'E) in 2000 and 2001. The 325-ha study area contained about 560 nest-boxes, harboring 132 coal tit breeding pairs in 2000 and 184 pairs in 2001. During the breeding seasons (April–July), all nest-boxes were monitored at least weekly in order to record breeding phenology, parameters of reproductive performance (clutch and brood size, hatching, and fledging success), and the identity of adult birds. Adults captured while feeding nestlings 10- to 14-days old were regarded the social (i.e., putative) parents of the respective broods. Both adults and nestlings were banded with uniquely numbered metal rings of the Institute of Avian Research "Vogelwarte Helgoland" (Wilhelmshaven, Germany). Blood samples (approximately 50 µl) were taken from the ulnar vein under license (no. 509f-42502-46), diluted in 250 µl of APS buffer (Arctander 1988), and stored at –20 °C until further use. Blood sampling had no detectable effect on fledging success or local recruitment probability into the study population (Schmoll et al. 2004).
Ageing and age measures
In the field, adult birds were classified as yearlings (i.e., in second calendar year, hereafter 2Y) or older (i.e., in at least third calendar year, hereafter 3Y+) according to plumage characteristics. In contrast to 3Y+ birds, 2Y coal tits show a distinct molt limit within their greater coverts in spring with old, outworn feathers from the juvenile plumage in the distal part of the wing contrasting in color and brightness with a variable number of new feathers in the proximal part of the wing (Jenni and Winkler 1994). Only very few 2Y birds (about 2.5%) have molted either all or no greater coverts at all and thus lack a molt limit (Jenni and Winkler 1994 and our own observations).
We used 2 age measures as predictor variables in subsequent statistical analyses, age categories (2Y vs. 3Y+ males), as well as exact age in years. Age was exactly known for the majority of the breeding individuals (56.9% and 64.0% in 2000 and 2001, respectively), as these had been banded as nestlings in the study area. When age had been misclassified in the field for individuals of this group (4.8% of 335 adult birds), data were corrected accordingly. Remaining individuals, first captured and banded as adults, were classified as 2Y or 3Y+ based on the plumage characteristics described above (no plumage characteristics are known to reliably age 3Y+ birds more exactly, Jenni and Winkler 1994). Thus, for individuals estimated as 3Y+ this estimate reflects a measure of their minimum age and therefore tends to underestimate their actual age.
Assignment of genetic parentage
We used a combination of multilocus DNA fingerprinting and microsatellite analysis to 1) exclude genetic parentage for the putative (i.e., social) parents and 2) assign EPO to their genetic fathers (EPS). Existing multilocus data on parentage exclusion (see Dietrich 2001) as well as assignment of EPO to EPS based on a multilocus procedure (see Schmoll, Janzon, et al. 2003) were cross-checked and complemented using 4 polymorphic microsatellite markers (see Schmoll et al. 2007). The resulting parentage assignments were thus supported by 2 independent sets of molecular genetic markers for more than 99% of offspring analyzed (a few samples failed to be genotyped by one of the methods). Because details of all procedures have been described elsewhere, only the fundamental methods are outlined briefly here.
Multilocus DNA fingerprinting analysis
DNA was isolated according to a modified standard protocol (Lubjuhn and Sauer 1999) and digested with the restriction enzyme Hae III. After separation by horizontal agarose gel electrophoresis, gels were dried followed by in-gel hybridization using the 32P-labeled oligonucleotide (CA)8. The banding patterns were visualized by scanning with a phosphoimager (Storm 860, Amersham Biosciences, Freiburg, Germany). Parentage exclusion gels always contained a brood's nestlings along with its social (i.e., putative) parents. Banding patterns were highly informative and analyzed according to Westneat (1990) using the image-editing software Adobe Photoshop 5.5. The putative fathers were excluded from genetic parentage if 2 or more novel fragments (i.e., distinct fragments neither attributable to the social mother's nor to the social father's banding pattern) were present in the banding pattern of the focal nestling and if band sharing between offspring and putative father was low whereas band sharing between offspring and putative mother was high (Lubjuhn et al. 1999; Dietrich 2001). The probability of falsely assigning one putative parent to an offspring was as low as 1.1 x 10–5 (for details, see Dietrich 2001).
We assigned EPO to EPS by means of standardized across-gel comparisons of multilocus DNA fingerprints using scanned gels and standard image-editing software according to Schmoll, Janzon, et al. (2003). In brief, in addition to the DNA fingerprint gels prepared for parentage exclusion (see above) we ran and processed—under identical conditions—5 and 8 gels containing DNA samples of all potential EPS (referred to as PEPS gels) representing all 107 and 174 territorial males blood sampled in the study population in 2000 and 2001, respectively. At least one male's DNA was run onto both of 2 flanking PEPS gels to produce a virtual overlap of these gels. Meanwhile, patterns of diagnostic paternal fragments within an EPO's banding pattern (i.e., novel fragments that must have been inherited from an EPS, see above) had been marked by symbols within parentage exclusion gel files. These symbol patterns were first copied onto the DNA fingerprint banding pattern of the respective brood's social father using the software's "copy" option. Then the pattern of symbol marks was pasted—combined with the underlying social father's banding pattern itself—into the PEPS gel file that contained a copy of the social father's banding pattern. The crucial point in the procedure is that now the social father's DNA fingerprint could be aligned along a copy of itself that had been run on another gel (the PEPS gel). It could therefore be adjusted to the specific running conditions of the respective PEPS gel by using the software's "distort" option. Together with the social father's banding pattern, the pattern of symbol marks representing diagnostic paternal fragments to be matched by a potential EPS had thus also been adjusted to the specific conditions of the PEPS gel. This procedure leaves the symbol marks in the same gel-specific positions in which fragments of the true genetic father must be expected. PEPS gels were then screened for potential EPS males by shuttling the pattern of symbol marks over them. If no candidate males were encountered on the first PEPS gel (i.e., the PEPS gel containing territorial neighbors of the focal brood's social male), the remaining PEPS gel files were screened in the same way with males run onto both the 2 flanking PEPS gels (see above) allowing further adjustments to the specific conditions of the respective gels. Backed up by conventional DNA fingerprinting in cases of doubt, this procedure allows the screening and identification of EPS (see Schmoll, Janzon, et al. 2003).
Microsatellite analysis
We used the 4 polymorphic microsatellite loci: PAT MP 2-43 (Otter et al. 1998), Pdoµ5 (Griffith et al. 1999), Pocc6 (Bensch et al. 1997), and Mcyµ4 (Double et al. 1997) to cross-check and complement the results obtained from multilocus DNA fingerprinting analysis (see Schmoll et al. 2007). Loci were amplified by polymerase chain reaction (PCR) according to optimized standard protocols (for details on PCR conditions see Stiels 2004; Mund 2005). PCR products were separated on an ABI PRISM® 377 DNA sequencer (Applied Biosystems, Foster City, CA) and allele sizes were scored using GENESCAN 2.1 (Applied Biosystems). We used CERVUS 3.0 (Marshall et al. 1998; Kalinowski et al. 2007) to calculate population genetic parameters and exclusion probabilities (see Schmoll et al. 2007). The combined exclusion probability of the 4 loci (i.e., the probability that a randomly chosen male will not possess an offspring's paternal alleles given that the genotype of the mother is known, Jamieson 1994) amounted to 0.999, and the probability of chance inclusion (i.e., the probability of a male matching an offspring's genotype purely by chance given that the genotype of the mother is known, Jeffreys et al. 1992) was low (mean ± standard deviation [SD]: 1.6 x 10–3 ± 2.6 x 10–3, range: 1.8 x 10–2–1.1 x 10–6 based on 773 nestling–mother dyads).
Nestlings were regarded within-pair offspring (WPO) if there was a complete match with their putative parents' genotypes. Only a single nestling in 2001 did not match the putative mother's genotype, and exclusion of genetic maternity was also supported by the multilocus approach in this case (this nestling was excluded from further analysis). For all remaining offspring, social maternity was thus assumed to reflect genetic maternity. Accordingly, nestlings were regarded EPO if their genotypes did not match those of the putative father at one or more loci. An EPO was assigned to an EPS if there was a complete match of their respective genotypes, and also the multilocus procedure (see above) supported the respective assignment.
Statistical analysis
The software packages R 2.3.1 (R Development Core Team 2006) and SPSS v. 12.0 (SPSS Inc, Chicago, IL) were used. Statistical tests were 2-tailed in all cases, and the null hypothesis was rejected at P < 0.05.
We used generalized linear models (GLMs) to test for effects of male age on fertilization success. Parsimonious description of the data was obtained by fitting minimal adequate models by stepwise removing terms from the maximal model as long as this caused no significant (P < 0.05) decrease in model fit (Crawley 2005). Minimal adequate models of different fundamental structure (i.e., models that were not nested) were compared by means of the Akaike Information Criterion (AIC), and the minimal adequate model with the lowest AIC score was accepted as describing the data in the most parsimonious way.
Data on male EPF success (i.e., the number of EPO assigned to individual males) were strongly overdispersed with an excess of zero counts and were thus analyzed by fitting GLMs with a negative binomial error structure and a log-link function (R function glm.nb in library MASS, Venables and Ripley 2002). Maximal models were fitted and then modified using the R function glm.convert (library MASS, Venables and Ripley 2002) to keep the estimated theta parameter of the negative binomial distribution fixed for subsequent model simplification. Within-pair fertilization success of territorial males (i.e., the number of genetic offspring in their own broods) as well as total fertilization success (i.e., the total number of genetic offspring as the sum of EPO and WPO) were analyzed by fitting GLMs with a Poisson error structure and a log-link function. All blood-sampled territorial males were included in analyses of EPF success. However, because a few broods failed to be genotyped, within-pair and total fertilization success could not be established for all territorial males so that sample sizes differ slightly.
We analyze and present data separately for 2 consecutive first brood periods because some birds were captured in both years and because the demography of the study population, namely the age distributions for territorial males, differed significantly between years (Kolmogorov–Smirnov 2-sample test: Z = 1.95, N2000 = 107, N2001 = 174, P < 0.001). On average, territorial males tended to be older in 2000 compared with 2001 (
2000 ± standard error = 2.4 ± 0.2 and 2001 = 2.0 ± 0.1 years; Mann–Whitney U test: U = 8147.5, N2000 = 107, N2001 = 174, P = 0.063). Excluding males for which only an estimate of their minimum age was available yielded the same result (data not shown). Furthermore, also the distribution of territorial males across age categories (2Y vs. 3Y+ males) differed significantly between years (likelihood ratio test: G = 16.53, degrees of freedom [df] = 1, P < 0.001) with younger males being overrepresented in 2001 compared with 2000.
In a few analyses, marginally nonsignificant results were obtained in the year with lower sample size (in 2000). In these cases, combined error probabilities for the analyses of both years were computed according to Sokal and Rohlf (1995).
For a subsample of males data on fertilization success were available for both years and allowed for longitudinal analyses of the change in fertilization success for individual males. Brood sizes at the time of sampling and hence the mean number of analyzed nestlings per brood differed significantly between years (2000 ± SD = 8.3 ± 1.7 and 2001 ± SD = 7.1 ± 1.4 analyzed nestlings; Mann–Whitney U test: U = 4045.5, N2000 = 94, N2001 = 164, P < 0.001). To control for potentially confounding effects of between-year variation in brood sizes, we standardized measures of fertilization success by dividing them through the sample means of the respective years. Thus, EPF success was expressed as the number of EPO assigned to individual males divided by the mean number of potentially fertilizable EPO per territorial male (2000: 210 EPO/107 males = 1.96; 2001: 338 EPO/174 males = 1.94). Within-pair fertilization success was expressed as the number of WPO assigned to individual males divided by the mean number of potentially fertilizable WPO nestlings per territorial male for which within-pair fertilization success had been established (2000: 568 WPO/94 males = 6.04; 2001: 1166 WPO/164 males = 5.05). Total fertilization success was expressed as the number of genetic offspring assigned to individual males divided by the mean number of potentially fertilizable nestlings per territorial male for which within-pair and therefore also total fertilization success had been established (2000: 778 nestlings/94 males = 8.28; 2001: 1166 nestlings/164 males = 7.11). Pairwise comparisons were then performed based on standardized values.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONFUNDINGREFERENCES |
|---|
Frequency of EPP
Relevant information on the frequency of EPP is summarized in Table 1 (for a more detailed analysis of the fundamental patterns of EPP in the study population see Dietrich et al. 2004a
). Neither was there a difference in the proportion of broods with EPP or the proportion of EPO between years (likelihood ratio tests: G = 0.01, df = 1, P = 1.0 and G = 0.90, df = 1, P = 0.36) nor did the mean number of EPO per analyzed brood differ (Mann–Whitney U test: U = 7576.5, N2000 = 94, N2001 = 164, P = 0.82).
|
Identification of EPS and paternity assignment
Relevant information regarding the identification of EPS is summarized in Table 2. The proportion of EPO to which an EPS could be assigned did not differ between years (G = 0.71, df = 1, P = 0.41), and the same was evident for total paternity assignment rates (G = 1.50, df = 1, P = 0.25).
|
Age of EPS versus cuckolded males
Pairwise comparisons revealed that EPS were significantly older than the males they cuckolded in both years (Table 3A and Figure 1). As the exact age of part of the older (i.e., 3Y+) individuals tends to be underestimated (see Methods), the true effect sizes are likely to be even larger than reported in Table 3A. Indeed, when restricting the samples of the Wilcoxon's signed-rank tests to cases for which the exact age of both males was known (i.e., both males had been banded as nestlings), the differences remained significant despite substantial reductions in sample sizes and effect sizes increased (Table 3B). Furthermore, the same general pattern was also evident when age categories (2Y vs. 3Y+ males) were used as measures of age (sign tests, 2000: N = 45, P = 0.08; 2001: N = 67, P < 0.001; combined error probability: P < 0.001).
|
|
Male EPF success
Older males were more likely than younger ones to achieve at least one EPF (logistic regressions, 2000:
2 = 2.81, df = 1, N = 107, P = 0.094; 2001: 2 = 12.09, df = 1, N = 174, P < 0.001; combined error probability: P < 0.001), and this effect was even more pronounced when age category (2Y vs. 3Y+ males) was used as the predictor variable (logistic regressions, 2000: 2 = 6.06, df = 1, N = 107, P = 0.014; 2001: 2 = 24.53, df = 1, N = 174, P < 0.001). Furthermore, the number of EPO assigned to individual males was positively related to age (Spearman rank correlations, 2000: rS = 0.25, N = 107, P = 0.011; 2001: rS = 0.38, N = 174, P < 0.001), although not in a linear manner. EPF success leveled off for older males and was most parsimoniously described by the inverse term of male age in both the years (Table 4, Figure 2). These relationships were statistically indistinguishable between years (interaction age–1 x year, GLM: 2 = 0.02, df = 1, P = 1.0), and the percentage of the total deviance explained by the inverse term of male age amounted to 8.8% in 2000 and 15.9% in 2001, respectively. In line with these patterns, there were highly significant differences in the number of EPO sired by 2Y versus 3Y+ males (Mann–Whitney U tests, 2000—mean ± SD number of EPO for 2Y males: 0.3 ± 1.4; 3Y+ males: 1.6 ± 3.3, U = 745.5, N1 = 26, N2 = 81, P = 0.007; 2001—2Y males: 0.3 ± 1.2; 3Y+ males: 2.1 ± 3.0, U = 2345.5, N1 = 84, N2 = 90, P < 0.001). For a subsample of 54 males, data on EPF success were available for the first brood periods of both years, allowing for a longitudinal analysis of the change in EPF success between years. Males significantly increased their EPF success between years (Wilcoxon's signed-rank test on standardized values: Z = –2.32, N = 54, P = 0.019, see Figure 3). Males aged 2Y in 2000 contributed more to this effect than males aged 3Y+ in 2000 (2Y males: Z = –1.84, N = 15, P = 0.074, mean increase between years: 1.10 EPO; 3Y+ males: Z = –1.49, N = 39, P = 0.14, mean increase between years: 0.34 EPO).
|
|
|
Male within-pair fertilization success
There were no effects of male age or its inverse or squared terms on within-pair fertilization success in either year (GLMs: all
2 < 3.2, all df = 1, all P > 0.07). For a subsample of 50 males, data on within-pair fertilization success were available for the first brood periods of both the years, allowing for a longitudinal analysis. There was no difference in within-pair fertilization success between years for individual males (paired t-test on standardized values: t49 = –0.161, P = 0.87).
Male total fertilization success
Total fertilization success was positively related to male age (Spearman rank correlations: 2000: rS = 0.26, N = 94, P = 0.012; 2001: rS = 0.24, N = 164, P = 0.002), although again not in a linear manner. Similar to EPF success also total success leveled off for older males and was most parsimoniously described by the inverse term of male age in both years (Table 5, Figure 4). These relationships were statistically indistinguishable between years (interaction age–1 x year, GLM: 2 = 0.23, df = 1, P = 0.63), and the percentage of the total deviance explained by the inverse term of male age amounted to 5.6% in 2000 and 6.9% in 2001, respectively. In line with these patterns, there were significant differences in total fertilization success between the age categories of 2Y and 3Y+ males (Mann–Whitney U tests, 2000—mean ± SD number of genetic offspring for 2Y males: 5.9 ± 3.9; 3Y+ males: 7.9 ± 4.0, U = 534.0, N1 = 20, N2 = 74, P = 0.056; 2001—2Y males: 5.4 ± 2.6; 3Y+ males: 7.3 ± 3.9, U = 2372.0, N1 = 79, N2 = 85, P < 0.001; combined error probability: P < 0.001). Finally, for a subsample of 50 males, data on total fertilization success were available for the first brood periods of both the years, allowing for a longitudinal analysis. As individual males' EPF success significantly increased between years (see above) whereas within-pair fertilization success was unaffected (see above), we analyzed whether the increase in EPF success would translate into an increase also in total fertilization success. We found that individual males showed a trend to increase their total fertilization success between years (paired t-test on standardized values: t49 = –1.775, P = 0.08, see Figure 5). Males aged 2Y in 2000 contributed more to this trend than males aged 3Y+ in 2000 (2Y males: t13 = –1.487, P = 0.16, mean increase between years: 0.24 genetic offspring; 3Y+ males: t34 = –1.172, P = 0.25, mean increase between years: 0.12 genetic offspring).
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONFUNDINGREFERENCES |
|---|
We have shown that EPS in a population of the socially monogamous coal tit were significantly older than the males they cuckolded. In line with this, extrapair and, as a consequence, also total fertilization success were positively related to male age in cross-sectional analyses. The observed patterns of age-related fertilization success were very similar (indeed, statistically indistinguishable) in 2 consecutive first brood periods, within which the study population differed significantly in the age distribution of territorial males. Thus, despite differences in demography, there were no obvious year effects as they have frequently been evident in other studies of natural populations. Furthermore, individual males increased their EPF success in longitudinal analyses, suggesting that potentially confounding cohort effects have not affected the results obtained from the cross-sectional analyses. We infer that age-related male fertilization success represents a general phenomenon in the study population.
Fertilization success was not a linear function of male age but leveled off for older age classes (Figures 2 and 4), suggesting that it is mainly the difference between 2Y versus 3Y+ males that is of crucial importance. In line with this, also the difference between these age groups was highly significant and rather pronounced. 2Y males rarely sired any EPO at all, and 3Y+ males were, on average, between 5 and 7 times as successful as 2Y males. Most studies up to date could provide only this latter type of analysis due to the challenge of determining age exactly in the field (e.g., Delhey et al. 2003; Bouwman et al. 2007). Our results suggest that comparisons between these age groups capture the essence of age-related fertilization success and allow meaningful inference and thus seem justified if establishing the exact age for a large fraction of males is not feasible. However, although the inverse term of male age best explained variation in fertilization success, also a quadratic model proved significant. Because estimates for the oldest age classes (
5Y+) were based on rather small samples as reflected in wide confidence intervals for predicted values (Figures 2 and 4), we cannot exclude the possibility that fertilization success could decrease again for these age classes, for example, as a consequence of senescence (see also Hansen and Price 1995).
Our results, demonstrating an advantage of older males in gaining EPFs, are in line with the findings of most of the few other studies in socially monogamous bird species with comparable extrapair mating systems (cf. Table 3 in Griffith et al. 2002; Delhey et al. 2003; Kleven, Marthinsen, et al. 2006; Bouwman et al. 2007; but see e.g., Charmantier et al. 2004). They underscore that male age, or a strong correlate thereof, is indeed an important determinant of EPF success in socially monogamous bird species. However, we would like to emphasize that male age explained only a fairly small proportion of the total variation in fertilization success (up to a maximum of about 16%) in our study population. Thus, further yet unidentified determinants of male EPF success like song (e.g., Forstmeier et al. 2002), sexual plumage ornamentation (Sheldon and Ellegren 1999), genotype (Masters et al. 2003), social rank (Otter et al. 1998), or experience (Weatherhead and Boag 1995) may explain additional variation in male success independent of age in our system and need to be addressed in future studies. In fact, if these yet unidentified traits were themselves more or less strongly correlated with age, they could have confounded the true relationship between male age and success in the present study. For example, in black-capped chickadees Poecile atricapillus, females show a tendency to prefer those males as extrapair copulation partners that are ranking high in the social hierarchy relative to their social mates (Smith 1988; Otter et al. 1998, see also Mennill et al. 2002). While male age is by far the strongest predictor of an individual's social rank (Schubert et al. 2007), the preference for rank was largely independent of male age (Otter et al. 1998). Experimental manipulation of age-related traits is required to disentangle their effects from age effects per se.
Age-related male EPF success could either result from a female mating preference for older males or from older males being more successful in competing for and securing EPFs (or from a combination of these mechanisms). A female (extrapair) mating preference for older males may be indirectly selected through genetic benefits in terms of increased offspring viability (age-based indicator mechanism of genetic quality, Brooks and Kemp 2001). If male age is used by females as an honest signal of expected male genetic contribution to offspring viability and females mate extra-pair to secure genetic benefits, we would also expect that older males suffer less from cuckoldry in their own nests and hence within-pair fertilization success should be positively related to male age too. In contrast to this straightforward prediction, this was not the case, a pattern that was also evident in other species with similar extrapair mating systems (e.g., Johnsen et al. 2001; Kleven, Marthinsen, et al. 2006). Furthermore, an open-ended mating preference might also be predicted, and males should be the more attractive as extrapair partners and, thus, the more successful the older they are. This was not the case either as EPF success leveled off for older age classes. However, such continually diminishing returns in EPF success with increasing age could also result from older males being indeed more attractive but less fertile (Brooks and Kemp 2001; see also Hansen and Price 1995). An age-based indicator model predicts not only that older males are more attractive as extra-pair copulation partners but also, like any indirect benefit model, a higher genetic quality of EPO compared with their within-pair maternal half-siblings (cf. Schmoll, Dietrich, et al. 2003; Schmoll et al. 2005). In the case of the age-based indicator model, the difference in fitness-related traits between half-siblings should correspond to the age difference of cuckolded versus extrapair male and this prediction could be subjected to a more conclusive test of the age-based indicator model of extrapair mating in future studies (for a similar approach based on an attractive secondary sexual plumage trait see Sheldon et al. 1997). Finally, older males may signal their genetic quality more reliably than younger males (Proulx et al. 2002), and females could also actively choose older males to secure the direct benefit of sufficient sperm supply, given that functional fertility is positively related to male age (Sheldon 1994).
Several recent studies requested that a more active role of males in pursuing extrapair copulations should be more strongly considered (e.g., Westneat and Stewart 2003) and the observed patterns of age-related fertilization success could well be due to older males being more sophisticated in securing extrapair copulations or fertilizations under intense sperm competition. Older males may simply be more experienced and skilled in obtaining EPFs (Weatherhead and Boag 1995) than younger and especially 2Y males. For example, if breeding is rather synchronous, conflicting demands may arise of extrapair mating activities like advertising or performing extraterritorial forays on the one hand and mate guarding the social female to secure within-pair paternity on the other. An individual trade-off between these conflicting demands may be achieved more favorably by experienced than inexperienced males (e.g., Johnsen et al. 2003; Kleven, Marthinsen, et al. 2006). Age-related fertilization success may also reflect male life-history trade-offs if younger and particularly 2Y males are unable to invest as heavily into EPF effort as older males. This could relate to extrapair signaling and advertising, pursuing extraterritorial forays (which must probably be traded against mate guarding the own female), or to sperm production, storage, and use under intense sperm competition (when allocation of a limited sperm supply to extrapair matings must probably be traded against allocation to within-pair matings). In line with this idea, testes size (e.g., Merilä and Sheldon 1999; Graves 2004; Laskemoen 2005), the size of the seminal glomera (Laskemoen 2005), or cloacal protuberances (Bouwman et al. 2007) were larger in older males in a number of passerine species. Establishing lifetime patterns of resource allocation to extrapair mating effort along with its fitness consequences is required to further elucidate this potential explanation.
There are broader implications of the finding that male age is an important correlate of EPF success. In nonexperimental approaches that establish phenotypic correlates of (extrapair) fertilization success, male age is likely to confound the true relationship between trait and male success if trait expression itself is age dependent, which makes it difficult to draw firm conclusions (e.g., Kleven, Jacobsen, et al. 2006; Bouwman et al. 2007). Given that male age seems to be one of the most consistently reported correlates of EPF success (see above), it is mandatory to record and account for male age to illuminate whether traits under scrutiny have any potential to signal individual differences of male quality also independent of age. Another implication refers to the estimation of how strongly EPP increases variation in male fitness and thus the opportunity for sexual selection to operate and produce sexually dimorphic phenotypes in socially monogamous species (e.g., Whittingham and Dunn 2005). If EPF success is indeed primarily a function of age, not accounting for this will tend to overestimate the opportunity of sexual selection because part of the variance in male fitness attributed to EPFs will in fact be attributable to male age. Establishing lifetime within-pair and extrapair, and thereby total, fertilization success is necessary to obtain an unbiased estimate. Alternatively, estimates could be done for samples splitted into cohorts or at least age classes. Finally, studies that aim to demonstrate ecological and social constraints on the expression of extrapair mating behavior may wish to take into account the age of social males and of the males potentially available for extrapair copulations (cf. Westneat and Mays 2005).
|
FUNDING |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONFUNDINGREFERENCES |
|---|
Deutsche Forschungsgemeinschaft (Lu 572/2-4).
|
ACKNOWLEDGEMENTS |
|---|
We thank Sabrina Bleidissel, Maria Orland, Christiane Wallnisch, and Darius Stiels for their help in the laboratory; Julia Delingat, Julia Eggert, Jorg Welcker, and especially Volker Janzon and Doris Winkel for support in the field; and Karin and Herbert Körner for their hospitality during field work. Katharina Förster kindly assayed microsatellite markers for their cross-amplificability in coal tits. Helpful comments of an anonymous reviewer improved the manuscript.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONFUNDINGREFERENCES |
|---|
Arctander P. Comparative studies on avian DNA restriction fragment length polymorphism analysis: Convenient procedures based on blood samples from live birds. J Ornithol (1988) 129:205–216.[CrossRef]
Bensch S, Price T, Kohn J. Isolation and characterization of microsatellite loci in a Phylloscopus warbler. Mol Ecol (1997) 6:91–92.[CrossRef][Medline]
Bouwman KM, Van Dijk RE, Wijmenga JJ, Komdeur J. Older male reed buntings are more successful at gaining extrapair fertilizations. Anim Behav (2007) 73:15–27.[CrossRef][Web of Science]
Brooks R, Kemp DJ. Can older males deliver the good genes? Trends Ecol Evol (2001) 16:308–313.[CrossRef][Medline]
Charmantier A, Blondel J, Perret P, Lambrechts MM. Do extra-pair paternities provide genetic benefits for female blue tits Parus caeruleus? J Avian Biol (2004) 35:524–532.[CrossRef]
Crawley MJ. Statistics—An Introduction using R. (2005) Chichester: John Wiley & Sons.
Delhey K, Johnsen A, Peters A, Andersson S, Kempenaers B. Paternity analysis reveals opposing selection pressures on crown coloration in the blue tit (Parus caeruleus). Proc R Soc Lond B (2003) 270:2057–2063.[Medline]
Dietrich V. Zum Auftreten alternativer Fortpflanzungsstrategien in einer Lingener Population der Tannenmeise (Parus ater) (Diploma thesis). (2001) Braunschweig (Germany): TU Braunschweig.
Dietrich V, Schmoll T, Winkel W, Epplen JT, Lubjuhn T. Pair identity —an important factor concerning variation in extra-pair paternity in the coal tit (Parus ater). Behaviour (2004a) 141:817–835.[CrossRef]
Dietrich V, Schmoll T, Winkel W, Epplen JT, Lubjuhn T. Cuckoldry and recapture probability of adult males are not related in the socially monogamous coal tit (Parus ater). J. Ornithol. (2004b) 145:327–333.[CrossRef]
Double MC, Dawson D, Burke T, Cockburn A. Finding the fathers in the least faithful bird: A microsatellite-based genotyping system for the superb fairy-wren Malurus cyaneus. Mol Ecol (1997) 6:691–693.[CrossRef]
Forstmeier W, Kempenaers B, Meyer A, Leisler B. A novel song parameter correlates with extra-pair paternity and reflects male longevity. Proc R Soc Lond B (2002) 269:1479–1485.[Medline]
Glutz von Blotzheim UN, Bauer KM. Handbuch der Vögel Mitteleuropas, 13/I, Passeriformes (Part 4). (1993) Wiesbaden (Germany): Aula.
Graves GR. Testicular volume and asymmetry are age-dependent in black-throated blue warblers (Dendroica caerulescens). Auk (2004) 121:473–485.[CrossRef][Web of Science]
Griffith SC, Owens IPF, Thuman KA. Extra pair paternity in birds: a review of interspecific variation and adaptive function. Mol Ecol (2002) 11:2195–2212.[CrossRef][Medline]
Griffith SC, Stewart IRK, Dawson DA, Owens IPF, Burke T. Contrasting levels of extra-pair paternity in mainland and island populations of the house sparrow (Passer domesticus): is there an ‘island effect’? Biol J Linn Soc (1999) 68:303–316.[CrossRef][Web of Science]
Hansen TF, Price DK. Good genes and old age: Do old mates provide superior genes? J Evol Biol (1995) 8:759–778.[CrossRef][Web of Science]
Jamieson A. The effectiveness of using co-dominant polymorphic allelic series for (1) checking pedigrees and (2) distinguishing full-sib pair members. Anim Genet (1994) 25:37–44.[Medline]
Jeffreys AJ, Allen MJ, Hagelberg E, Sonnberg A. Identification of the Skeletal Remains of Mengele, Josef by DNA Analysis. Forensic Sci Int (1992) 56:65–76.[CrossRef][Web of Science][Medline]
Jenni L, Winkler R. Moult and Ageing of European Passerines. (1994) London: Academic Press.
Johnsen A, Lifjeld JT, Andersson S, Örnborg J, Amundsen T. Male characteristics and fertilisation success in bluethroats. Behaviour (2001) 138:1371–1390.[CrossRef]
Johnsen A, Lifjeld JT, Krokene C. Age-related variation in mate-guarding intensity in the bluethroat (Luscinia s. svecica). Ethology (2003) 109:147–158.[CrossRef][Web of Science]
Kalinowski ST, Taper ML, Marshall TC. Revising how the computer program CERVUS accommodates genotyping error increases success in paternity assignment. Mol Ecol (2007) 16:1099–1106.[CrossRef][Medline]
Kempenaers B, Verheyen GR, Dhondt AA. Extrapair paternity in the blue tit (Parus caeruleus): female choice, male characteristics, and offspring quality. Behav Ecol (1997) 8:481–492.
Kleven O, Marthinsen G, Lifjeld JT. Male extraterritorial forays, age and paternity in the socially monogamous reed bunting (Emberiza schoeniclus). J Ornithol (2006) 147:468–473.[CrossRef]
Kleven O, Jacobsen F, Izadnegahdar R, Robertson RJ, Lifjeld JT. Male tail streamer length predicts fertilization success in the North American barn swallow (Hirundo rustica erythrogaster). Behav Ecol Sociobiol (2006) 59:412–418.[CrossRef][Web of Science]
Laskemoen T. Variation in testis size and sperm morphology in the bluethroat, Luscinia svecica svecica (Master thesis). (2005) Oslo (Norway): University of Oslo.
Lubjuhn T, Gerken T, Brün J, Epplen JT. High frequency of extra-pair paternity in the Coal Tit. J Avian Biol (1999) 30:229–233.[CrossRef]
Lubjuhn T, Sauer KP. DNA fingerprinting and profiling in behavioural ecology. In: DNA profiling and DNA fingerprinting—Epplen JT, Lubjuhn T, eds. (1999) Basel: Birkhäuser. 39–52.
Marshall TC, Slate J, Kruuk LEB, Pemberton JM. Statistical confidence for likelihood-based paternity inference in natural populations. Mol Ecol (1998) 7:639–655.[CrossRef][Medline]
Masters BS, Hicks BG, Johnson LS, Erb LA. Genotype and extra-pair paternity in the house wren: a rare-male effect? Proc R Soc Lond B (2003) 270:1393–1397.[Medline]
Mennill DJ, Ratcliffe LM, Boag PT. Female eavesdropping on male song contests in songbirds. Science (2002) 296:873–873.
Merilä J, Sheldon BC. Testis size variation in the greenfinch Carduelis chloris: relevance for some recent models of sexual selection. Behav Ecol Sociobiol (1999) 45:115–123.[CrossRef][Web of Science]
Mund V. Fremdvaterschaften bei der sozial monogamen Tannenmeise (Parus ater): Was macht Fremdkopulationspartner für Weibchen attraktiv? (Diploma thesis). (2005) Bonn (Germany): University of Bonn.
Otter K, Ratcliffe L, Michaud D, Boag PT. Do female black-capped chickadees prefer high-ranking males as extra-pair partners? Behav Ecol Sociobiol (1998) 43:25–36.[CrossRef][Web of Science]
Proulx SR, Day T, Rowe L. Older males signal more reliably. Proc R Soc Lond B (2002) 269:2291–2299.[Medline]
R Development Core Team. R: A language and environment for statistical computing [Internet]. (2006) Vienna (Austria): R Foundation for Statistical Computing. ISBN 3-900051-07-0. Available from: http://www.R-project.org.
Schmoll T, Janzon V, Epplen JT, Lubjuhn T. Extra-pair sires as identified by means of standardized across-gel comparisons in multilocus DNA fingerprints. Electrophoresis (2003) 24:2758–2763.[CrossRef][Web of Science][Medline]
Schmoll T, Dietrich V, Winkel W, Epplen JT, Lubjuhn T. Long-term fitness consequences of female extra-pair matings in a socially monogamous passerine. Proc R Soc Lond B (2003) 270:259–264.[Medline]
Schmoll T, Dietrich V, Winkel W, Epplen JT, Schurr F, Lubjuhn T. Paternal genetic effects on offspring fitness are context dependent within the extrapair mating system of a socially monogamous passerine. Evolution (2005) 59:645–657.[Web of Science][Medline]
Schmoll T, Dietrich V, Winkel W, Lubjuhn T. Blood-sampling does not affect fledging success and fledgling local recruitment in coal tits (Parus ater). J Ornithol (2004) 145:79–80.[CrossRef]
Schmoll T, Schurr F, Winkel W, Epplen JT, Lubjuhn T. Polyandry in coal tits (Parus ater): Fitness consequences of putting eggs into multiple genetic baskets. J Evol Biol (2007) 20:1115–1125.[CrossRef][Web of Science][Medline]
Schubert KA, Mennill DJ, Ramsay SM, Otter KA, Boag PT, Ratcliffe LM. Variation in social rank acquisition influences lifetime reproductive success in black-capped chickadees. Biol J Linn Soc (2007) 90:85–95.[CrossRef][Web of Science]
Sheldon BC. Male phenotype, fertility, and the pursuit of extra-pair copulations by female birds. Proc R Soc Lond B (1994) 257:25–30.
Sheldon BC, Ellegren H. Sexual selection resulting from extrapair paternity in collared flycatchers. Anim Behav (1999) 57:285–298.[CrossRef][Web of Science][Medline]
Sheldon BC, Merilä J, Qvarnström A, Gustafsson L, Ellegren H. Paternal genetic contribution to offspring condition predicted by size of male secondary sexual character. Proc R Soc Lond B (1997) 264:297–302.
Smith SM. Extra-pair copulations in Black-capped Chickadees: the role of the female. Behaviour (1988) 107:15–23.
Sokal RR, Rohlf FJ. Biometry. (1995) 3rd ed. New York: W.H. Freeman.
Stiels D. Etablierung von Mikrosatellitensystemen zur Elternschaftsanalyse bei Tannenmeisen (Parus ater) (Diploma thesis). (2004) Bonn (Germany): University of Bonn.
Venables WN, Ripley BD. Modern applied statistics with S-PLUS. (2002) 4th ed. New York: Springer.
Weatherhead PJ, Boag PT. Pair and extra-pair mating success relative to male quality in red-winged blackbirds. Behav Ecol Sociobiol (1995) 37:81–91.[CrossRef][Web of Science]
Westneat DF. Genetic parentage in the Indigo Bunting: A study using DNA fingerprinting. Behav Ecol Sociobiol (1990) 27:67–76.[Web of Science]
Westneat DF, Mays HL. Tests of spatial and temporal factors influencing extra-pair paternity in red-winged blackbirds. Mol Ecol (2005) 14:2155–2167.[CrossRef][Medline]
Westneat DF, Stewart IRK. Extra-pair paternity in birds: causes, correlates, and conflict. Ann Rev Ecol Evol Syst (2003) 34:365–396.[CrossRef]
Whittingham LA, Dunn PO. Effects of extra-pair and within-pair reproductive success on the opportunity for selection in birds. Behav Ecol (2005) 16:138–144.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  T. Schmoll, F. M Schurr, W. Winkel, J. T Epplen, and T. Lubjuhn Lifespan, lifetime reproductive performance and paternity loss of within-pair and extra-pair offspring in the coal tit Periparus ater Proc R Soc B, January 22, 2009; 276(1655): 337 - 345. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||