JOURNAL OF AVIAN BIOLOGY 35: 524 /532, 2004
Do extra-pair paternities provide genetic benefits for female blue tits Parus caeruleus? Anne Charmantier, Jacques Blondel, Philippe Perret and Marcel M. Lambrechts
Charmantier, A., Blondel, J., Perret, P. and Lambrechts, M. M. 2004. Do extra-pair paternities provide genetic benefits for female blue tits Parus caeruleus ? / J. Avian Biol. 35: 524 /532. A large body of theories on extra-pair paternity (EPP) in birds has proposed four main ‘‘genetic’’ hypotheses to explain this behaviour: the ‘‘good genes’’ hypothesis, the genetic diversity hypothesis, the genetic compatibility hypothesis and the fertility insurance hypothesis. Empirical tests have been scarce, mainly because high sample sizes are difficult to collect. We have tested these hypotheses in three Mediterranean populations of blue tits Parus caeruleus in which 50 /68% of the broods contained extra-pair young. Results showed that the distribution of extra-pair young among broods was not random, and that survival to fledging of extra-pair young was higher than that of their within-pair sibs. These results support the idea of genetic effects benefiting extra-pair young. However, comparison of cuckolded and cuckolding males showed no significant difference in their body size, age, survival or relatedness with their paired females, and offspring morphometrics did not differ between extra-pair and within-pair young. We conclude that none of the genetic hypotheses can explain fully the high level of extra-pair paternity, at least in our populations of Mediterranean blue tits. We suggest that direct ecological benefits of EPP for females should be tested more often in correlative as well as experimental approaches. A. Charmantier (correspondence), J. Blondel, P. Perret and M. M. Lambrechts, Centre d’Ecologie Fonctionnelle Evolutive, CNRS, 1919 Route de Mende, F-34293 Montpellier Cedex 5, France. Present address of A. Charmantier: Edward Grey Institute, Department of Zoology, University of Oxford, Oxford OX1 3PS, UK. E-mail:
[email protected]
Extra-pair copulation (EPC), responsible for genetic polygamy, is a common and widespread reproductive strategy in socially monogamous birds (Birkhead and Møller 1992, Griffith 2000, Møller and Cuervo 2000, Hasselquist and Sherman 2001). Recent advances in molecular genetic techniques revealed that EPC often results in extra-pair paternity (EPP, Westneat et al. 1990, Birkhead and Møller 1992). It is straightforward to understand how males can enhance their reproductive success by fertilizing extra-pair females and thus fathering offspring at the expense of the paternal care provided by another male. Yet it seems that males are not the only decision-maker for EPC to happen and that females may play the major role (Birkhead and Møller 1993, Venier et al. 1993, Caizergues and Lambrechts 1999). In addition, females may apply cryptic female choice to
decide which sperm fertilises the eggs after copulation (Birkhead and Møller 1993). Thus female birds would play an active role in soliciting EPC, and males should provide or expose direct or indirect benefits to convince females to accept copulation and egg fertilisation (Birkhead and Møller 1993, Kempenaers and Dhondt 1993). At least seven benefits have been discussed in the literature to explain why female birds would solicit or accept EPC (Westneat et al. 1990, Birkhead and Møller 1992, Kempenaers and Dhondt 1993, Petrie and Kempenaers 1998, Jennions and Petrie 2000, Griffith et al. 2002). A first group of ‘‘ecological’’ hypotheses refers to immediate benefits improving brood fitness. For instance, females could accept EPC with unpaired males
# JOURNAL OF AVIAN BIOLOGY
524
JOURNAL OF AVIAN BIOLOGY 35:6 (2004)
in exchange for help with resource provisioning of the chicks, or nest defence. However, little empirical evidence has given support for direct benefits hypotheses (but see Gray 1997). A second group of four ‘‘genetic’’ hypotheses deals with indirect benefits whereby females accepting EPC would enhance the average genetic quality of the brood (reviews in Kempenaers and Dhondt 1993, Jennions and Petrie 2000, Griffith et al. 2002). First, the ‘‘good genes’’ hypothesis proposes that females can assess the genetic quality of the males from phenotypic criteria, such as body size, body condition, behaviour (song, display), or colours. Provided that there is enough variation in these phenotypic traits (Petrie et al. 1998), females should seek extra-pair copulation with the male expressing the highest genetic quality. Males advertising high genetic quality are predicted to gain extra-pair paternities at the expense of males that express a lower genetic quality. Furthermore, extra-pair offspring should have enhanced growth, fecundity or survival compared to within-pair offspring (Kokko 2001). The ‘‘good genes’’ hypothesis has received recent support from intra-specific studies (Sheldon et al. 1997, Yezerinac and Weatherhead 1997) especially in a Belgian population of blue tits (Parus caeruleus, Kempenaers et al. 1992, Kempenaers et al. 1997) where cuckolded males produce fewer offspring, survive less well, have shorter average songs and smaller tarsi than males that do not loose paternity. The same study showed that in nests with mixed paternity (n/ 165), extra-pair offspring were heavier than within-pair offspring, and survived better in cases of partial brood mortality (n /10). However, other field studies in tit species did not found any support for this hypothesis (Krokene et al. 1998, Strohbach et al. 1998). Second, in an alternative genetic benefit hypothesis, EPP maximises genetic diversity among offspring (Williams 1975, Westneat et al. 1990), increasing the probability that some offspring will survive, especially in unpredictable and fluctuating environments favouring certain genotypes in some conditions and other genotypes in other circumstances. This hypothesis predicts higher average fitness in broods with several genetic fathers than in broods with one genetic father, no difference in average fitness between extra-pair and within-pair offspring, high rates of EPP in all broods, and often more than two extra-pair fathers in broods with mixed paternity. To this day, the only evidence for an effect of genetic diversity on EPP is a comparative analysis of 432 bird studies showing a positive correlation between proportion of polymorphic loci and EPP rate (Petrie et al. 1998). Third, the so called ‘‘genetic compatibility’’ hypothesis (Kempenaers et al. 1999) supposes that EPC increases heterozygosity in offspring and therefore increases hatching or survival rates in offspring, especially in JOURNAL OF AVIAN BIOLOGY 35:6 (2004)
populations with inbreeding depression. It predicts higher EPP rates in pairs with higher genetic relatedness, and higher relatedness between the social pair partners than the extra-pair partners. Such a positive relationship between social pair relatedness and occurrence of EPP has been reported in three species of shorebird (Blomqvist et al. 2002). Recently, a blue tit study has given more support for the compatibility hypothesis, with extra-pair offspring showing higher heterozygosity than their within-pair mates, in an Austrian population with inbreeding depression (Foerster et al. 2003). Fourth, the ‘‘fertility insurance’’ hypothesis (Wetton and Parkin 1991, Krokene et al. 1998) predicts no average fitness differences between extra-pair and within-pair offspring. Extra-pair young should be randomly distributed among broods (Kempenaers and Dhondt 1993), except for infertile males’ broods in which all offspring are extra-pair. This hypothesis also predicts higher hatching success in broods with extrapair young in case of temporary sperm depletion (Wetton and Parkin 1991). According to Krokene et al. (1998) this hypothesis may partly explain why female birds seek EPC, despite the fact that infertile males are rarely reported. Many studies tested these different hypotheses at the intra-specific level. However, as Griffith et al. (2002) pointed out, good empirical tests require a large body of information on individual fitness and quality of females, males and offspring, thus very few studies have collected enough data of this type to discriminate between the various hypotheses (but see Kempenaers et al. 1997, Sheldon et al. 1997, Krokene et al. 1998, Strohbach et al. 1998, Johnsen et al. 2000). In addition, these hypotheses are not mutually exclusive. For example females can seek EPC for ‘‘good genes’’ which may at the same time lead to fertility enhancement. In an attempt to provide new insights concerning this behaviour, we used extra-pair paternity data from three Mediterranean populations of blue tits where EPP rates have been reported as high (Charmantier and Blondel 2003) to test the four main genetic hypotheses. For this purpose, we used morphological, physiological and lifehistory traits of offspring and breeding males. We compared cuckolded males with cuckolding males, extra-pair young (EPY) with within-pair young (WPY) and social partners with extra-pair partners to test predictions from each genetic benefit hypothesis.
Material and methods Study sites and field methods The study was carried out in 2000, 2001 and 2002 in three populations of blue tit breeding in nestboxes and occupying habitats differing in growth conditions of the chicks (mainland-Rouvie`re 438 40?N, 038 40?E, 525
deciduous oakwood, Corsica-Pirio 428 31?N, 088 46?E, evergreen oakwod and Corsica-Muro 428 33?N, 088 55?E, deciduous oakwood, see Lambrechts et al. (1999b) for description of the study sites). Table 1 summarizes the sample dates and sizes for each population. All nestboxes were monitored throughout the breeding season using protocols that have been applied since 1976, to determine laying date, clutch size, hatching date, hatching success and fledging success. Hatching success is the number of hatchlings over the number of eggs laid and fledging success is the number of fledglings over the number of hatchlings. Blood samples were collected from chicks 5 to 9 days after hatching. Parents were captured when chicks were between 10 and 15 days old. Tarsus length was measured to the nearest 1/10 mm with a calliper, flattened wing lengths to the nearest mm with a ruler and body mass to the nearest 1/10 g with a Pesola spring balance. A condition index was estimated as the residual from a linear regression of body mass on tarsus length. Chick tarsus and body mass were measured at day 15. Offspring recruitment, i.e. survival from chick to adulthood, was determined by the capture of breeding pairs in nestboxes in succeeding years. Recruitment rate was very low: 1.7%, 4.9% and 6.9% in Pirio, Muro and Rouvie`re, respectively. We estimated overwinter survival of adult males through subsequent recaptures in the following breeding season. In Pirio and Muro, haematocrit levels were determined in 90% and 63% of the chicks respectively. At day 15, 20 mg of blood were collected from the brachial vein into a microtube and centrifugated (3 min at 13 000 rotation per minute) to determine the percentage of the volume of erythrocytes in the total blood sample.
Genetic analysis DNA was extracted from blood samples using the QIAamp blood kit (Qiagene). 89.2% of broods were sexed (87.8% of offspring) by PCR amplification of the P2 and P8 introns of the CHD1 gene (Griffiths et al. 1998). The PCR products were separated on 3% agarose gel run at 150 W for 45 minutes and visualised with ethidium bromide under UV.
Four to seven microsatellite loci were used for paternity exclusion and paternity assignment (POCC1, POCC6, Bensch et al. 1997, Phtr3, Fridolfsson et al. 1997, Pca3, Pca7, Pca8 and Pca9; Dawson et al. 2000). Loci were amplified by polymerase chain reaction (PCR) following the conditions described by Bensch (1997), Dawson (2000) and Fridolfsson (1997). PCR products were run on an ABI 310 machine, and scored using the GENESCAN and GENOTYPER software packages. Paternity was excluded when there were at least two mismatches between the father’s and the offspring’s genotypes. Estimation of probabilities of exclusion and paternity assignments were done using the software CERVUS (Marshall et al. 1998). All EPY and all males were genotyped on at least six microsatellite loci, and the high polymorphism made it possible to identify genetic fathers (when captured) without ambiguity in genotypes. In Pirio, Muro and Rouvie`re respectively, 11%, 10% and 8% of eggs were unhatched. We estimated from observation of their content that at least 60% of unhatched eggs showed no sign of development. Only 11 dead embryos were successfully genotyped, but none of them was extra-pair. As fertility of unhatched eggs was not always checked and not all embryos were genotyped, our results include only hatched chicks. We measured the genetic similarity between the two parents of social and genetic pairs by estimating Queller and Goodnight’s relatedness coefficient (R) using the software Relatedness 5.0 (Queller and Goodnight 1989).
Statistical analyses All analyses were performed using SAS 8.02 (SAS Institute 1992). Non-parametric tests were used whenever distributions differed from normality. The analyses were done separately for the three populations and then data were pooled with study sites as independent variables. Results for single populations were only given when they differed from the overall analysis. Distribution of EPY among broods The randomness of EPY distribution among broods in each year and study site was checked using the multivariate hypergeometric distribution following
Table 1. Sample sizes and extra-pair paternity rates in three Mediterranean populations of blue tits Parus caeruleus. Island Pirio Data collection Number of broods genotyped % of broods with mixed paternity % of broods with at least one EP father identified Number of chicks genotyped % EPY
526
2000 /2001 50 68.0% 50% 288 25.4%
Mainland Muro 2001 30 50.0% 53.3% 205 18.2%
Rouvie`re 2000 /2002 97 53.6% 69.2% 839 16.1%
JOURNAL OF AVIAN BIOLOGY 35:6 (2004)
Male comparisons We compared the phenotypic characteristics (tarsus length, body mass, condition index, wing length and age) and relatedness with the female, of cuckolded and uncuckolded males using t-tests with each characteristic as a response variable. Rate of overwinter survival was compared using a x2-test. We used phenotypic male traits for which we know they have a genetic basis (Charmantier et al. 2004) and which have been used by many authors as indicators of male quality (Kempenaers et al. 1997, Krokene et al. 1998, Strohbach et al. 1998, Leech et al. 2001). We repeated the same type of analysis comparing cuckolded males with their cuckolding males when identified. When extra-pair broods were fathered by more than one male, we used the mean values of the extra-pair fathers.
Brood performance comparison We compared the performance of broods with mixed and non-mixed paternity using GLMs with laying date, hatching success, fledging success, mean offspring body mass at 15 days, mean tarsus length at 15 days, mean condition index and mean haematocrit level as response variables.
Comparison of EPY with WPY Offspring performance was measured through body mass and tarsus length at 15 days, condition index, survival to fledging, haematocrit level and recruitment as breeders the next year. We used a generalized linear mixed model (PROC MIXED, SAS Institute 1992) with morphological or physiological traits as the response variable, brood identity as a random effect (Krackow and Tkadlec 2001), and tested for effects of EPP. This comparison within broods controls for territory quality and social parent quality (Kempenaers et al. 1997). Sex was included as a fixed effect in all models. For all comparisons, we performed power analyses with GPower version 2.0 (Faul and Erdfelder 1992) choosing a type I error of 0.05 and a effect size of 0.5 (Cohen 1977) which is less than half the effect size corresponding to the effects of ‘‘good genes’’ found in Belgian blue tits (Kempenaers et al. 1992). JOURNAL OF AVIAN BIOLOGY 35:6 (2004)
Results Paternity exclusion and paternity assignment The microsatellite loci showed high levels of heterozygosity and the minimum combined probability of exclusion was 0.99 (see Charmantier and Blondel 2003 for details). Annual proportion of broods containing EPY over the three populations ranged from 0.40 to 0.68, with overall means of 0.68, 0.50 and 0.54 in Pirio, Muro and Rouvie`re respectively (Table 1). Proportion of EPY in broods was respectively 0.25, 0.18 and 0.16. There was generally a low proportion of EPY in the broods (1 /3 EPY), except for two nests in Pirio in 2001, two nests in Muro in 2001 and one nest in Rouvie`re in 2002 where all offspring were sired by an extra-pair male. Overall, 51.8% of EPP were assigned, amounting to 60.4% of the mixed paternity broods in which at least one extra-pair father was identified (see Table 1 for details). Unless stated otherwise, all 177 broods were included in the analysis. For analyses comparing cuckolded males to their cuckolding males, the number of broods where at least one extra-pair male was identified amounted to 61.
Distribution of EPY among broods For each year and population, EPY were not distributed randomly over the broods (x2, all PB/0.001), except in Pirio in 2000 (x225 /6.17, P/0.95). The non-randomness was due to more broods than expected by the hypergeometric distribution containing either many or no EPY (Fig. 1, shows the distribution in Rouvie`re in 2002).
Male comparisons The t-tests on male morphometrics, age and relatedness with females did not show any significant differences between cuckolded and non-cuckolded males (Table 2), 0.45 0.4
Proportion of broods
Neuha¨user et al. (2001). The expected frequencies were calculated using a SAS macro program obtained online at http://www.bioinf.uni-hannover.de/ /neuhaus/ macro.sas. Comparing observed and expected frequencies was done by using x2 tests.
Observed frequency
0.35
Expected frequency
0.3 0.25 0.2 0.15 0.1 0.05 0 0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Proportion of EPY in the brood
Fig. 1. Comparison of observed and expected distributions of extra-pair young (EPY) over broods in the Rouvie`re population of blue tits in 2002. The expected data follow a multivariate hypergeometric distribution computed following Neuha¨user et al. (2001).
527
Table 2. Comparison of male characteristics between cuckolded and non-cuckolded males. R is Queller and Goodnight’s relatedness coefficient. Mean (SE) of non-cuckolded males (n/101)
Mean (SE) of cuckolded males (n/76) Wing length (mm) Body mass (g) Tarsus length (mm) Condition index Age (years) Relatedness with the female (R) Real number of chicks recruited per brood
65.6 (0.29) 10.40 (0.1) 16.63 (0.06) 0.103 (0.072) 2.2 (0.2) 0.006 (0.214) 0.28 (0.06)
as well as between cuckolded and their cuckolding males (Table 3). Comparing recruitment rates (after paternity assignment) did not reveal any difference either. The power for these tests to detect an effect size of 0.5 ranged between 0.95 and 0.99. Overwinter survival rate was higher for non-cuckolded males than cuckolded males (46% against 36%, n /101, 76) although the difference was not significant (x21 / 1.95, P/0.163). The same trend occurred between cuckolding and cuckolded males (44% against 33%, n/61, x21 /1.69, P/0.194).
Brood performance comparisons None of the brood performance variables were affected by cuckoldry (F1,176 for laying date, hatching success, fledging success; F1,165 for morphological measures at 15 days and condition index and F1,66 for haematocrit; all P/0.132).
Comparison of EPY with WPY Morphological measures of 15 days chicks did not significantly differ between EPY and WPY, nor did haematocrit level (Table 4). In Rouvie`re, there was evidence for higher recruitment of EPY than WPY
Table 3. Pairwise comparisons of cuckolded males vs. their cuckolding males. Comparing cuckolded with cuckolding males (n/61)
Wing length (mm) Body mass (g) Tarsus length (mm) Condition index Age (years) Relatedness with the female (R)
528
t-test
P-value
/0.42 /0.21 /0.17 /0.07 /0.32 /0.38
0.677 0.838 0.863 0.947 0.722 0.705
66.2 (0.3) 10.45 (0.1) 16.74 (0.06) 0.112 (0.087) 2.4 (0.1) 0.012 (0.222) 0.43 (0.09)
Comparing cuckolded with non-cuckolded males t-test
P-value
/1.37 /0.35 /1.3 0.00 /0.91 /0.19 /1.47
0.174 0.730 0.194 0.945 0.362 0.849 0.1442
(11% and 6% of recruitment for EPY and WPY, F1,741 / 4.4, P /0.037), but this trend was not significant when all populations were pooled. Survival to fledging was significantly higher for EPY in the overall analysis (Table 4). The 101 mixed broods conferred a power of 0.99 to detect an effect size of 0.5.
Discussion Tests on males and offspring showed: (a) non-random distribution of EPY among broods with a high number of broods containing either no EPY or many of them, and (b) higher survival of EPY compared to their within-pair sibs. This provides some evidence that females choose to engage in EPC to enhance the fitness of their broods. However, we did not find any difference for male morphometrics, overwinter survival and relatedness to their socially paired female neither between cuckolded and cuckolding, nor between WPY and EPY. The main advantage of our study compared to previous ones is the statistical power thanks to: (i) a high number of broods, (ii) three populations involved, and (iii) the possibility to compare offspring performances as well as male performances. Indeed, we have gathered the three types of data which Griffith et al. (2002) considered as essential to distinguish between the main explanations: the distribution of EPY among broods, the distribution of EPY among males and the comparison between EPY and WPY. Therefore, it was possible to test six predictions on male and offspring performances in order to discriminate between the four main genetic hypotheses explaining why females would benefit from accepting or soliciting EPC. Table 5 summarizes the different predictions arising from the hypotheses and the main results of our study for each prediction. One must keep in mind that if an effect was predicted but not detected, then it implies that either the hypothesis is false or we lack the power necessary to detect it. This lack of power can result from insufficient sample size or from a lack of investigations on other relevant fitness measures, e.g. phenotypic cues used by JOURNAL OF AVIAN BIOLOGY 35:6 (2004)
Table 4. Comparison of EPY and WPY performances in broods with mixed paternity (n /101). Differences were tested using a GLMM, sex was included in all models as a fixed effect. Mean (SE) of EPY (n/245) Body mass (g) Tarsus length (mm) Condition index Haematocrit level Survival to fledging Recruitment
10.3 (0.1) 16.40 (0.04) 0.069 (0.058) 43.08 (0.533) 0.95 (0.01) 0.07 (0.02)
Mean (SE) of WPY (n/1087)
F
ddl
P
10.4 (0.0) 16.36 (0.03) 0.171 (0.027) 42.94 (0.266) 0.90 (0.01) 0.05 (0.01)
2.82 3.49 0.18 2.65 4.55 0.59
1,536 1,536 1,533 1,195 1,581 1,581
0.094 0.062 0.675 0.105 0.033 0.441
females to assess the genetic quality of males may not be correlated with morphological size. When a predicted effect is significantly revealed, then the corresponding hypothesis will be validated, which does not mean that the other hypotheses should be rejected because several of them are not necessarily mutually exclusive. In the light of this logic, the results summarized in Table 5 do not validate any of the genetic benefit hypotheses. Higher performance of EPY compared to WPY is evidence for a genetic effect, yet the results do not enable discrimination between the different hypotheses.
Good genes hypothesis The ‘‘good genes’’ hypothesis has indisputably received more empirical support than any other hypothesis (reviews in Westneat et al. 1990, Birkhead and Møller 1992, Kempenaers and Dhondt 1993, Griffith et al. 2002). Interestingly, although there was a trend for those males which lost paternity to be smaller, lighter, in worse
condition, younger and to survive less over winter, than those which did not loose paternity, this trend was not significant. This contrasts with the results found in Belgian and Austrian blue tit populations (Kempenaers et al. 1992, Foerster et al. 2003). Detecting a ‘‘good genes’’ effect such as the one in Kempenaers’ Belgian population necessitates a minimum sample size of 32 broods, which is consistently inferior to our sample size of 101 broods with mixed paternity. Although we could assign quite a large number of EPP, we did not find any evidence that extra-pair males were of higher genetic quality than the males they cuckolded. This means that either the ‘‘good genes’’ hypothesis does not explain the high rates of EPP in our populations, or that female choice is based on other phenotypic male traits than those we measured. In particular, ultraviolet reflectance of crown plumage has recently been revealed in this species as an important factor for female choice of within-pair and extra-pair partners (Delhey et al. 2003, Johnsen et al. 2003). EPY did not differ from WPY in morphometrics and in haematocrit level, but there was some evidence for
Table 5. Predictions of the four main genetic hypotheses explaining the evolutionary occurrence of EPP and results from the study of three populations of blue tits. R: Relatedness coefficient (Queller and Goodnight 1989), WPY: within-pair young and EPY: extrapair young. Higher Cuckolded males Cuckolded males R between social Random distribution of EPY of higher genetic of higher genetic partners higher performance of than in extra broods with quality than quality than among broods? pair? mixed paternity? their cuckolding non-cuckolded males? males? Good genes
Genetic diversity Genetic compatibility Fertility insurance Results in our populations
no: more broods with high or low levels of EPY than expected yes no: high levels of EPY when R between social parents is high yes, expect for some broods with only EPY no: more broods with high or low levels of EPY than expected
JOURNAL OF AVIAN BIOLOGY 35:6 (2004)
EPY fitter than WPY?
yes
yes
no
yes
yes
no
no
no
yes
no
no
no
yes
yes
yes
no
no
no
yes: higher no hatching success
no: for morphometric measures and overwinter survival
no: for morphometric measures and overwinter survival
no
no
yes: higher survival for EPY than for WPY
529
higher survival to fledging of EPY. In Rouvie`re, data showed higher recruitment rates of EPY than WPY, but this result was not confirmed in the other populations nor in the overall analysis. This result in Rouvie`re could perhaps be explained by a higher local recruitment rate (6.9%) in this population compared to the other two (1.7% and 4.9%), presumably as a result of the particular location of this study area which is quite isolated. In blue tit populations, like in most passerines, offspring settle far from their natal habitat, resulting in low local recruitment, not necessarily reflecting total recruitment (Kempenaers et al. 1997, Lambrechts et al. 1999a). This makes it difficult to assess the genetic benefits.
Genetic diversity hypothesis The main prediction derived from the genetic diversity hypothesis is that the proportion of EPP in a population should be negatively correlated with the genetic variability of the population (Petrie et al. 1998). Although tests at the intra-population level are difficult, we expect random distribution of EPY among males and females (see Table 5, Kempenaers and Dhondt 1993) but this was not the case in our populations. We have shown previously that at the intra-specific level, rate of EPP did not correlate with genetic diversity estimated from expected heterozygosity (Charmantier and Blondel 2003). If the benefits from genetic diversity in broods had played a role in promoting the evolutionary occurrence of EPC, detecting this effect would be difficult because the only predicted effect is that, on the long-term, broods with mixed paternities should have a higher fitness than broods without EPY, e.g. higher mean survival or higher mean offspring production in mixed paternity broods.
Genetic compatibility hypothesis Thanks to the high polymorphism of the microsatellite markers, we were able to identify the genetic fathers of 51.8% of the EPY. In 60.4% of the broods with mixed paternity, at least one extra-pair male could be identified, which allowed us to verify in these broods that relatedness between the social parents did not determine the presence of EPY in the brood nor the choice of the extra-pair mate. Hence, our study does not confirm the recent discoveries in an Austrian blue tit population (Foerster et al. 2003) nor the accumulation of evidence in favour of the genetic compatibility hypothesis and inbreeding avoidance by females, in various bird species (Kempenaers et al. 1999, Blomqvist 2002). 530
Fertility insurance hypothesis Non random distribution of EPY among broods runs counter to the fertility insurance hypothesis in which EPP should not depend on male attractiveness or quality. However, this hypothesis is difficult to test. Besides the fact that there is still ambiguity in the literature on the predictions derived from the fertility insurance hypothesis, e.g. concerning the potential awareness of females on the fertility status of their male (Kempenaers and Dhondt 1993, Griffith et al. 2002), egg fertility can only be determined with confidence after examination of the perivitelline layers around the ovum (Kempenaers et al. 1999). In the tree swallow Tachycineta bicolor, analysis of the perivitelline layers of eight unhatched eggs showed that they were in fact fertile and had suffered from embryo mortality (Kempenaers et al. 1999). Our study does not include information on the presence of sperm and holes in the perivitelline layers, which means that fertility must be interpreted as the combination of egg fertilization and viability. In our populations, 2.8% of 177 broods had total loss of paternity, which is similar to what has been found in other blue tit studies, e.g. 2.1% in a Belgian population (Kempenaers et al. 1997) and 2.1% in a Norwegian population (Krokene et al. 1998). Although many passerine studies of EPP detect similar levels of broods with total loss of paternity, direct evidence for the fertility insurance hypothesis remains scarce (Wetton and Parkin 1991, Krokene et al. 1998). In these few studies, alternative interpretations were still possible (Lifjeld 1994). Once again, we conclude that no evidence was found in our study populations to support the hypothesis that females engage in EPCs to improve hatching success in their broods. In conclusion, none of the genetic hypotheses is fully supported by our data. This could be because of methodological reasons (e.g. local recruitment not reflecting total recruitment in the breeding population) or biological reasons (e.g. females use multiple traits to judge males and these traits are not necessarily correlated). However, it could also be that other mechanisms are likely to explain the high occurrence of EPP in our populations. In particular, direct benefits could play an important role in the Mediterranean populations, where environmental constraints are high. Investigation on female home range expansion would give useful information to test whether females seek EPC in order to enlarge and improve resource acquisition. Acknowledgements / We thank all researchers and students who have participated in the field work and data collection. Thank you to C. Debain, F. Di Giusto and P. Sourrouille for help during the genetic analyses, to Markus Neuha¨user for his help on statistics and to Bart Kempenaers for useful comments on the manuscript. Two anonymous reviewers provided useful comments to improve the manuscript. A. C. was supported by a grant from the ‘‘Ministe`re Franc¸ais de l’Education Nationale, de l’Enseignement et de la Recherche’’. M. L. was supported by JOURNAL OF AVIAN BIOLOGY 35:6 (2004)
a European Network grant (EKVL-CT-1999-0017) and funding from the ‘‘Bureau des ressources ge´ne´tiques’’. Ringing materials were supplied by CRBPO.
References Bensch, S., Price, T. and Khon, J. 1997. Isolation and characterization of microsatellite loci in a Phylloscopus warbler. / Mol. Ecol. 6: 91 /92. Birkhead, T. and Møller, A. P. 1992. Sperm competition in birds: evolutionary causes and consequences. / Academic Press, London. Birkhead, T. and Møller, A. P. 1993. Female control of paternity. / Trends Ecol. Evol. 8: 100 /104. Blomqvist, D. 2002. Genetic parentage and mate guarding in the Artic-breeding western sandpiper. / Auk 119: 228 /233. Blomqvist, D., Andersson, M., Ku¨pper, C., Cuthill, I. C., Kis, J., Lanctot, R., Sandercock, B. K., Sze´kely, T., Wallander, J. and Kempenaers, B. 2002. Genetic similarity between mates and exra-pair parentage in three species of shorebirds. / Nature 419: 613 /615. Caizergues, A. and Lambrechts, M. M. 1999. Male ‘‘macho’’ mammals exploiting females versus male ‘‘Don Juan’’ birds exploited by females: the opposite-sex exploitation (OSEX) theory. / Ecol. Lett. 2: 204 /206. Charmantier, A. and Blondel, J. 2003. A contrast in extra-pair paternity levels on mainland and island populations of Mediterranean blue tits. / Ethology 109: 351 /364. Charmantier, A., Kruuk, L. E. B., Blondel, J. and Lambrechts, M. M. 2004. Testing for microevolution in body size in three blue tit populations. / J. Evol. Biol. 17: 732 /743. Cohen, J. 1977. Statistical power analysis for the behavioral sciences (revised edition). / Academic Press, New York. Dawson, D. A., Hanotte, O., Greig, C., Stewart, I. R. K. and Burke, T. 2000. Polymorphic microsatellites in the blue tit Parus caeruleus and their cross-species utility in 20 songbird families. / Mol. Ecol. 9: 1919 /1952. Delhey, K., Johnsen, A., Peters, A., Andersson, S. and Kempenaers, B. 2003. Paternity analysis reveals opposing selection pressures on crown coloration in the blue tit (Parus caeruleus ). / Proc. R. Soc. Lond. B 270: 2057 /2063. Faul, F. and Erdfelder, E. 1992. GPower: a priori, post-hoc, and compromise analyses for MS-DS (Computer Program). / FRG, Bonn University, Bonn. Foerster, K., Delhey, K., Johnson, A., Lifjeld, J. T. and Kempenaers, B. 2003. Females increase offspring heterozygosity and fitness through extra-pair matings. / Nature 425: 714 /717. Fridolfsson, A.-K., Gyllensten, U. B. and Jakobsson, S. 1997. Microsatellite markers for paternity testing in the willow warbler Phylloscopus trochilus : high frequency of extra-pair young in an island population. / Hereditas 126: 127 /132. Gray, E. M. 1997. Female red-winged blackbirds accrue material benefits from copulating with extra-pair males. / Anim. Behav. 53: 625 /639. Griffith, S. C. 2000. High fidelity on islands: a comparative study of extrapair paternity in passerine birds. / Behav. Ecol. 11: 265 /273. Griffith, S. C., Owens, I. P. F. and Thuman, K. A. 2002. Extra pair paternity in birds: a review of interspecific variation and adaptive function. / Mol. Ecol. 11: 2195 /2212. Griffiths, R., Double, M. C., Orr, K. and Dawson, R. J. G. 1998. A DNA test to sex most birds. / Mol. Ecol. 7: 1071 /1075. Hasselquist, D. and Sherman, P. W. 2001. Social mating systems and extrapair fertilizations in passerine birds. / Behav. Ecol. 12: 457 /466. Jennions, M. D. and Petrie, M. 2000. Why do females mate multiply? A review of the genetic benefits. / Biol. Rev. Cambridge Phil. Soc. 75: 21 /64. JOURNAL OF AVIAN BIOLOGY 35:6 (2004)
Johnsen, A., Andersen, V., Sunding, C. and Lifjeld, J. T. 2000. Female bluethroats enhance offspring immunocompetence through extra-pair copulations. / Nature 406: 296 /299. Johnsen, A., Delhey, K., Andersson, S. and Kempenaers, B. 2003. Plumage colour in nestling blue tits: sexual dichromatism, condition dependence and genetic effects. / Proc. R. Soc. Lond. B 270: 1263 /1270. Kempenaers, B., Congdon, B., Boag, P. and Robertson, R. J. 1999. Extrapair paternity and egg hatchability in tree swallows: evidence for the genetic compatibility hypothesis? / Behav. Ecol. 10: 304 /311. Kempenaers, B. and Dhondt, A. A. 1993. Why do females engage in extra-pair copulations? A review of hypothesis and their predictions. / Bel. J. Zool. 123: 93 /103. Kempenaers, B., Verheyen, G. R. and Dhondt, A. A. 1997. Extrapair paternity in the blue tit (Parus caeruleus ): female choice, male characteristics, and offspring quality. / Behav. Ecol. 8: 481 /492. Kempenaers, B., Verheyen, G. R., Van den Broeck, M., Burke, T., Van Broeckhoven, C. and Dhondt, A. A. 1992. Extrapair paternity results from female preference for highquality males in the blue tit. / Nature 357: 494 /496. Kokko, H. 2001. Fisherian and ‘‘good genes’’ benefits of mate choice: how (not) to distinguish between them. / Ecol. Lett. 4: 322 /326. Krackow, S. and Tkadlec, E. 2001. Analysis of brood sex ratios: implications of offspring clustering. / Behav. Ecol. Sociobiol. 50: 293 /301. Krokene, C., Rigstad, K., Dale, M. and Lifjeld, J. T. 1998. The function of extrapair paternity in blue tits and great tits: good genes or fertility insurance? / Behav. Ecol. 9: 649 /656. Lambrechts, M. M., Blondel, J., Caizergues, A., Dias, P. C., Pradel, R. and Thomas, D. W. 1999a. Will estimates of lifetime recruitment of breeding offspring on small-scale study plots help us to quantify processes underlying adaptation? / Oikos 86: 147 /151. Lambrechts, M. M., Perret, P., Maistre, M. and Blondel, J. 1999b. Do experiments with captive non-domesticated animals make sense without population field studies? A case study with blue tits’ breeding time. / Proc. R. Soc. Lond. B 266: 1311 /1315. Leech, D. I., Hartley, I. R., Stewart, I. R. K., Griffith, S. and 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. Lifjeld, J. T. 1994. Do female house sparrows copulate with extra-pair males to enhance their fertility? / J. Avian Biol. 25: 75 /76. Marshall, T. C., Slate, J., Kruuk, L. E. B. and Pemberton, J. M. 1998. Statistical confidence for likelihood-based paternity inference in natural populations. / Mol. Ecol. 7: 639 /655. Møller, A. P. and Cuervo, J. J. 2000. The evolution of paternity and paternal care in birds. / Behav. Ecol. 11: 472 /485. Neuha¨user, M., Forstmeier, W. and Bretz, F. 2001. The distribution of extra-pair young within and among broods / a technique to calculate deviations from randomness. / J. Avian Biol. 32: 358 /363. Petrie, M. and Kempenaers, B. 1998. Extra-pair paternity in birds: explaining variation between species and population. / Trends Ecol. Evol. 13: 52 /58. Petrie, M., Doums, C. and Møller, A. P. 1998. The degree of extra-pair paternity increases with genetic variability. / Proc. Natl. Acad. Sci. USA 95: 9390 /9395. Queller, D. C. and Goodnight, K. F. 1989. Estimating relatedness using genetic markers. / Evolution 43: 258 /275. SAS Institute, I. 1992. SAS user’s guide, statistics. / SAS Institute, Inc, Cary, NC. Sheldon, B. C., Merila¨, J., Qvarnstro¨m, A., Gustafsson, L. and Ellegren, H. 1997. Paternal genetic contribution to offspring condition predicted by size of male secondary sexual character. / Proc. R. Soc. Lond. B 264: 297 /302.
531
Strohbach, S., Curio, E., Bathen, A., Epplen, J. T. and Lubjuhn, T. 1998. Extrapair paternity in the great tit (Parus major ): a test of the ‘‘good genes’’ hypothesis. / Behav. Ecol. 9: 388 /396. Venier, L. A., Dunn, P. O., Lifjeld, J. T. and Robertson, R. J. 1993. Behavioural patterns of extra-pair copulation in tree swallows. / Anim. Behav. 45: 412 /415. Westneat, D. F., Sherman, P. W. and Morton, M. L. 1990. The ecology and evolution of extra-pair copulations in birds. / Curr. Ornithol. 7: 331 /369. Wetton, J. H. and Parkin, D. T. 1991. An association between fertility and cuckoldry in the house sparrow, Passer domesticus. / Proc. R. Soc. Lond. B 245: 227 /233.
532
Williams, G. 1975. Sex and evolution. / Princeton University Press, Princeton, New Jersey. Yezerinac, S. M. and Weatherhead, P. J. 1997. Extra-pair mating, male plumage coloration and sexual selection in yellow warblers (Dendroica petechia ). / Proc. R. Soc. Lond. B 264: 527 /532.
(Received 15 August 2003, revised 19 January 2004, accepted 9 February 2004.)
JOURNAL OF AVIAN BIOLOGY 35:6 (2004)
