Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Extra-pair mate choice in the song sparrow (Melospiza melodia) 2009

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
ubc_2009_fall_ames_caroline.pdf [ 1.77MB ]
Metadata
JSON: 1.0068158.json
JSON-LD: 1.0068158+ld.json
RDF/XML (Pretty): 1.0068158.xml
RDF/JSON: 1.0068158+rdf.json
Turtle: 1.0068158+rdf-turtle.txt
N-Triples: 1.0068158+rdf-ntriples.txt
Citation
1.0068158.ris

Full Text

EXTRA-PAIR MATE CHOICE IN THE SONG SPARROW (Melospiza melodia) by Caroline Elizabeth Ames B.Sc., Simon Fraser University, 2004 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in The Faculty of Graduate Studies (Forestry) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) July 2009 © Caroline Elizabeth Ames, 2009 ABSTRACT Extra-pair paternity (EPP) is common in birds yet its adaptive significance remains unclear. Since the strategy of EPP is thought to carry costs, females are predicted to obtain indirect genetic benefits (e.g. ‘good genes’) or direct material benefits (e.g. fertility insurance) from pursuing extra-pair copulations (EPCs). Breeding synchrony may also influence the costs and benefits of EPP to males and females. I examine ‘good genes’ benefits of EPP and the effect ofbreeding synchrony on EPP in a socially monogamous population of song sparrows wherein 29% of 751 offspring were sired by extra-pair males. The good genes hypothesis predicts that females mate with extra-pair males that have higher expected fitness than their social mate in order to improve the fitness of extra-pair young (EPY) compared to within-pair maternal half-siblings. Using traits closely linked to lifetime reproductive success, I found no evidence that EPY were fitter than their maternal half-siblings or that extra-pair males were fitter than cuckolded males. However, I found that middle-aged males on average were 3.1 —4.7 times more likely to sire EPY than first-year males and 1.3 —2.0 times more likely to sire EPY than very old males. This is consistent with similar, well-established patterns of age-related variation in annual reproductive success in song sparrows, suggesting that male success in siring EPY is influenced by experience and ability, rather than quality. I found a significant negative relationship between breeding synchrony among neighbors and the proportion of EPY within broods of focal males. This result supports the ‘mate 11 guarding constraint’ hypothesis predicting that EPP decreases as synchrony increases because a larger proportion of males allocate time toward guarding their fertile social mate, instead of toward pursuing EPCs. However, I found that paternity loss was similar for males that sired EPY outside their social mate’s fertile period (40.4% of 57 males lost paternity) and for males that sired EPY during their mate’s fertile period (37.8% of 37 males lost paternity). This result suggests that mate guarding did not constrain males in the pursuit of EPCs; however, the exact timing of EPCs was unknown and may have influenced this result. 111 TABLE OF CONTENTS ABSTRACT.ii TABLE OF CONTENTS iv LIST OF TABLES vi LIST OF FIGURES vii ACKNOWLEDGEMENTS viii CO-AUTHORSHIP STATEMENT x INTRODUCTION 1 1.1 Extra-Pair Paternity in Passerines 1 1.2 Costs of pursuing EPCs 2 1.2.1 Females 2 1.2.2 Males 3 1.3 Benefits of pursuing EPCs 3 1.3.1 Females 3 1.3,2 Males 7 1.4 Ecological Factors 7 1.5 Study Species and Population 9 1.6 Thesis Overview 10 1.7 References 12 2 EXTRA-PAIR MATE CHOICE: A RETROSPECTIVE TEST OF THE GOOD GENES HYPOTHESIS 16 2.1 Introduction 16 2.2 Methods 20 2.2.1 Field Methods 20 2.2.2 Genetic Analysis and Paternity Assignment 21 2.2.3 Traits Related to Fitness 22 2.2.4 Statistical Analyses 23 2.3 Results 25 2.3.1 Overview 25 2.3.2 Within versus Extra-pair Young 26 2.3.3 Male Fitness and EPY Sired 26 2.3.4 Male Fitness and Paternity Loss 28 2.4 Discussion 29 2.4.1 Within versus Extra-Pair Young 29 2.4.2 Male Fitness and EPY Sired 31 2.4.3 Male Fitness and Paternity Loss 33 2.4.4 Conclusion 35 2.5 References 45 3 BREEDING SYNCHRONY, DENSITY, AND EXTRA-PAIR PATERNITY 49 3.1 Introduction 49 3.2 Methods 52 3.2.1 FieldMethods 52 3.2.2 Genetic Analysis and Paternity Assignment 53 3.2.3 Breeding Synchrony and Density 54 iv A6LsouaLJoJJj7 U.NOISSf13SIU‘IVUINIJD L9souaij>j 09tT10WdJuj-iupcpU1A11SUQUIpO1jtoo’j£V gçu.IoJa-.ixipuiAuoiqouiCg 8cMTA1AO1t’E 8cuoissnosiu LçSfUJAJOfl3puttinj-iixjosuosu1dmoD£E LcpsujpuAuoqouAgouoTwJo>jUTSSO]‘W’-’dE 9cM!MAQ 9cSflflS)J ççsosAjuVJg IA E9pu1SJiiputuosMoJJdsuos iojspooiquafliMjjokjtsuppuCuonpuAsuipaiqjodIqsuoIp1a>J 19puusj.iupuNuoSMoLrndS UOSJOJ9661OE66JUJOJJkuuidJ!d-Jxpu‘tsupuTpaJqjiooj‘Auonpu(s uIpaiqoj‘Auonpu/suipiquoiindod‘poudjq’jinuuuj,s 4v puu‘ouiuuojid Au.onpo1datwIojipu1puosios‘udsjtjjrnuopuisalJtpuuJ,%juosMonds uosjiuiiojJj1nuu1spooiqunjpAi.AdJJOuoii.iodoidqjodrqsuotp>j91ojqi pui‘ourniuojiod oAipnpoJdaIWp,JJpuijuosuos‘udsjijopuisjuosMonds uosopwpop3wqAjpnuuipansAJJO.Ioqurnuoqjodiqsuoijjc°iqi 117U1EJSJflpUJAJUO sMoJJ1dsuosojuipopjopn3prn.md-iuxojospiiijosuosuidmoopannjv 6CprnMsI1JPUN uosqis-jjq1itd-uT4pM‘ju.Ip3wJ13q4pusMoiudsuOSAdouuuoJ13d oAIpnpoJdalwiJT1pupuosispu‘uudsJIJ‘J1AIA.InSUIsua1JJIu 966104£661UIOJJA&IUO4S44MPUISI1UNU0 spooiqoindsuosjo14u33J3dpui(AJ)unoi(1nd-J4xJoiuaij11 LP1WISIJUPJAJU0SMOJJPdSUOSUISSUT404p4uaIS4I1JIl°IqL s1’1avL4OISrI 99pujsjJpuuosMoiI1dsuosojspooiq Uflfl!M.AJJOuoiiiodoidpunAuo.iqouAsjnoojuqdiqsuoiinp (JJAipdsJ‘p-n jund)9661O£661wojanpicnjoiuoniUTpooiqqon.iojpainsnw(spJi3Xdw) iCuonpuiCsuipoatqiojpun(s1oiiopoiitj)iCuoiqouAsuipiquoinfndodioit njnwpunpunjsinpunuosMoJinds uosjnmpnwAq(ijnnuunpJ1sAJJJ°ioqmnuuooqdiqsuoinpjFl S1flDU101SF! ACKNOWLEDGEMENTS I would like to thank my supervisor, Dr. Peter Arcese, for his guidance, constant support, and valuable suggestions throughout the course of this program. Thank you to my supervisory committee, Drs. Darren Irwin and Kathy Martin, for their insightful edits and comments during meetings and on a draft of this manuscript. A special thank you to Dr. Jane Reid for her constructive comments on a draft of this manuscript and whose knowledge and research has inspired my own research on extra-pair paternity in song sparrows. I would like to thank Katie O’Connor and Drs. Lukas F. Keller and Amy B. Man for genotyping and assigning paternity to the song sparrows of Mandarte Island. Thank you to the Mandarte Island field teams, past and present, for collecting the song sparrow data which have made this thesis possible. I also thank Drs. Valerie LeMay and Mark Drever for providing helpful reading material and suggestions on statistics. I also appreciate the feedback and encouragement I received from the Arcese and Martin labs. Thank you to Alaine Camfield, Nicola Freeman, Danika Kleiber, Michelle Martin, and Amy Wilson for their friendship throughout and helpful suggestions. I would like to express my gratitude to my family and friends who supported me along the way and who always had encouraging words for me. Many thanks to my longtime friend Anna Drake for the revealing discussions on statistical analyses and extra-pair paternity in birds. I would also like to send a heartfelt thank you to my parents, Steven and Lesley Ames, my sisters, Christina and Julie, and my aunt, Diane Taylor, for their continuous encouragement and support over the years. viii xrnqmnjoqsiuJo/(TiSJA!ufl041soouoiisaioijoluotulil3doU 041aIoJJdiqsJ11oqogto’>iospudTqsi1oqogoprnpipuwj(iisti) JiOuno343J3S0)JUu00u1:&IIU13S0OU00gP3ImMJqpopiAoJdsmrnpunj1uosa CO-AUTHORSHIP STATEMENT Dr. Peter Arcese provided the data for these studies. I designed the studies, analyzed the data, and prepared the manuscript with the assistance of Dr. Peter Arcese. x 1 INTRODUCTION 1.1 Extra-Pair Paternity in Passerines Extra-pair paternity (EPP) occurs in over 70% of the socially monogamous bird species surveyed, with an average of 11.1% of offspring being sired by extra-pair males and 18.7% of broods containing one or more extra-pair young (EPY) (Griffith et al. 2002). EPP occurs when a female mates with a male other than her social mate (an ‘extra-pair male’) and produces extra-pair young (EPY) as a consequence. The highest rates of EPP detected occur in the cooperatively breeding fairy wren (Malurus cyaneus), wherein 76% of all offspring are sired by extra-group males and 95% ofbroods contain at least one EPY (Mulder et al. 1994, Griffith et al. 2002). Among socially monogamous species, reed buntings (Emberiza schoenclus) display the highest levels of EPP, with 55% of offspring being sired by extra-pair males and 86% of broods containing one or more EPY (Dixon et al. 1994, Griffith et al. 2002). Given the prevalence of EPP within and among avian species, it is clear that describing the adaptive significance of EPP is necessary to understanding the evolution ofmating systems overall. A challenge to understanding the evolution of EPP is explaining the high level of interspecific variation in EPP among related species and across populations of the same species (Griffith et al. 2002, Westneat and Stewart 2003). For example, within the Hirundininae (i.e. aerial insectivores such as swallows and martins), tree swallows (Tachycineta bicolor) generally have higher EPP (54.0% of offspring; Griffith et al. 2002) than barn swallows (Hirundo rustica, 28.2% of offspring; Griffith et al. 2002). In another example, Griffith (2000) showed that the level of EPP for mainland populations 1 of passerines was, on average, 2.1 times higher than for island populations of the same species. To date, numerous studies have attempted to explain variation in EPP levels, but report a wide range of results, leaving the causes of variation in EPP levels poorly understood overall. One approach to understanding variation in EPP at the level of populations is to examine the complex set of interactions that occur between the traits of the female, her social mate, and the extra-pair male(s), and how ecological factors influence these interactions (Westneat and Stewart 2003). In order to accomplish this, Westneat and Stewart (2003) recommend that future researchers examine how variation in the traits of individuals and in the ecological conditions experienced by individuals can influence the potential costs and benefits of EPP. I take this approach in my thesis by investigating the potential causes of variation in EPP in a population of song sparrows (Melospiza melodia) resident on Mandarte Island, BC. Below, I briefly review the potential costs and benefits of pursuing extra-pair copulations (EPCs) in birds, and how variation in several ecological factors might influence these costs. I then summarize the results of my thesis research. 1.2 Costs of pursuing EPCs 1.2.1 Females Although EPP is relatively common in birds, the pursuit of EPCs by females is potentially costly, including the risk of reduced paternal investment, physical retaliation by the female’s social mate, investing in poor quality young, increased exposure to sexually transmitted diseases, and the time and energy costs of searching for and 2 assessing potential extra-pair mates (reviewed in Petrie and Kempenaers 1998). A major cost to females of pursuing EPCs is thought to be the withholding of parental care by males, but the evidence is mixed (reviewed in Whittingham and Dunn 2001, Sheldon 2002). 1.2.2 Males A major cost to males of pursuing extra-pair copulations (EPCs) is thought to be the potential loss of paternity in their own nests in the event that they cannot effectively guard their fertile social mate and pursue EPCs at the same time (e.g., Chuang-Dobbs et al. 2001). Two recent experimental studies show that males that are unable to mate guard or engage in other paternity guarding tactics have increased rates ofpaternity loss in their own nests (e.g. Komdeur et al. 2007, Johnsen et al. 2008). Other costs to males of pursuing EPCs include the risk of sexually transmitted diseases, sperm depletion, and potential trade-offs between the efforts invested in engaging in EPP versus parental investment in their own nest (reviewed in Petrie and Kempenaers 1998). 1.3 Benefits of pursuing EPCs 1.3.1 Females Because the pursuit of EPCs is thought to be costly to females, evolutionary theory predicts that the behavior must also entail compensatory benefits to favor its occurrence in populations. Behavioral observations in several species show that females often solicit EPCs during extra-territorial forays, adding to the view that females sometimes benefit from this behavior (reviewed in Westneat and Stewart 2003). Current hypotheses about 3 the potential fitness benefits that females may receive from mating with extra-pair males can be divided into those arguing direct versus indirect benefits, each of which is described in more detail below. Direct Benefits Females may obtain direct (material) benefits from extra-pair males that enhance their fecundity in a current year by increasing fertility (the ‘fertility insurance’ hypothesis; Wetton and Parkin 1991, Sheldon 1994), parental care (Blomqvist et al. 2005), nest defense (Gray 199Th), or access to breeding resources on the extra-pair male’s territory (Gray 199Th). In red-winged blackbirds (Agelaius phoeniceus), for example, females that obtained EPCs gained additional nest defense and foraging opportunities from extra- pair males (Gray 1 997b) and also hatched and fledged a greater proportion of young than females that did not obtain EPCs (Gray 1997a). In another example, female moustached warblers (Acrocephalus melanopogon) with EPY in their nest gained parental care from the extra-pair sires once chicks had hatched (Blomqvist et al. 2005). Several studies have also tested the ‘fertility insurance’ hypothesis for EPP by relating the hatching success of broods to the occurrence of EPP within broods, or to whether or not females obtained EPCs (e.g. Wetton and Parkin 1991, Gray 1997a, Whitekiller et al. 2000). However, as Griffith et al. (2002) point out, these studies cannot account for potential confounding effects such as female quality. For example, the positive relationship that Gray (1997a) found between hatching success and EPP is not necessarily an indication that females receive fertility insurance benefits from engaging in EPCs if high quality females are more likely to engage in EPCs than poor quality females. 4 Indirect Benefits Females pursuing EPCs may also accrue genetic benefits through offspring fitness, such as those related to ‘good genes’ (i.e. additive genetic benefits), genetic compatibility (i.e. non-additive benefits), or genetic diversity (reviewed in Griffith et al. 2002, Akçay and Roughgarden 2007). The ‘good genes’ hypothesis predicts that females that have social mates of poor intrinsic genetic quality will mate with extra-pair males of higher intrinsic genetic quality in order to obtain ‘good genes’ that improve the survival and/or future reproductive success of EPY compared to their within-pair maternal half siblings (Griffith et al. 2002, Akçay and Roughgarden 2007). For example, in great reed warblers (Acrocephalus arundinaceus), EPY were sired by extra-pair males that had larger song repertoires than the female’s social mate, and song repertoire size was positively correlated with post-fledging survival of offspring (Hasselquist et al. 1996). A key test of the good genes hypothesis is a fitness comparison between EPY and their within-pair maternal half sibs because this indicates differential paternal genetic contribution by controlling for maternal genetic contribution and rearing environment (Griffith et al. 2002, Akçay and Roughgarden 2007). I discuss the good genes hypothesis further in Chapter 2, where this hypothesis is explicitly tested. The genetic compatibility hypothesis predicts that females will obtain non-additive genetic benefits by mating with extra-pair males whose genome is more compatible to their own than their social mate’s (Griffith et al. 2002, Akçay and Roughgarden 2007); the resultant EPY will be fitter than their within-pair maternal half-siblings either due to .5 inbreeding avoidance (Tregenza and Wedell 2000) or due to increased viability through reduced intragenomic conflict (Zeh and Zeh 1997). A further prediction of the genetic compatibility hypothesis is that EPY will be fitter than their paternal half-sibs because it is the combination of the male and female genotypes that produce the fitness advantage of EPY rather than the male genotype alone, as would be predicted by the ‘good genes’ hypothesis (e.g. Johnsen et al. 2000). There is support for the genetic compatibility hypothesis in the literature. For example, Suter et al. (2007) found that in reed buntings, EPY were more heterozygous than their maternal haif-sibs because females were less genetically similar to the extra-pair male than to their social mate. Higher heterozygosity may have conferred a fitness advantage to EPY as they had higher fledgling survival than their maternal half sibs (Suter et al. 2007). In another example, Fossøy et al. (2007) also found that female bluethroats (Luscinia s. svecica) increased the heterozygosity of their offspring by mating with genetically dissimilar extra-pair males. Further, EPY expressed higher immunocompetence than both their maternal and paternal half-sibs. However, several studies also failed to find evidence that females obtain non-additive genetic benefits from extra-pair males (e.g. Kieven and Lifjeld 2005, Bouwman et al. 2006). The genetic diversity hypothesis predicts that females mate with multiple extra-pair males as a ‘bet-hedging’ strategy that increases diversity of offspring genomes in order to improve the chances of offspring survival and successful reproduction in unpredictable environments (Yasui 2001). This hypothesis has not been tested explicitly. 6 Sexually antagonistic coevolution It has also been suggested that EPP may not actually be adaptive for females, but may nevertheless result from females making the ‘best of a bad job’ if there is strong selection in males to achieve EPCs at the expense of female fitness (Westneat and Stewart 2003, Arnqvist and Kirkpatrick 2005). Few studies have tested this hypothesis explicitly. 1.3.2 Males The obvious benefit to males of obtaining EPCs is that they can increase their reproductive success without having to provide additional parental care for their extra- pair offspring (i.e. direct benefits). Males may also obtain indirect benefits such as good genes or more genetically compatible genes through EPP, although studies to date have not addressed this possibility. 1.4 Ecological Factors Ecological factors can be expected to affect the ability of individuals to obtain EPCs by influencing the availability of potential mates in space and time. One such factor is breeding synchrony which is expressed as the extent of overlap in female fertile periods (Kempenaers 1993). When breeding synchrony is high a large proportion of males in the population may have to choose between guarding their fertile social mate and pursuing EPCs with the many fertilizable females in the population. When mate guarding is important for preventing paternity loss and when the benefits ofprotecting paternity outweigh the benefits of pursuing EPCs, breeding synchrony is predicted to be negatively related to the level of EPP (the ‘mate-guarding constraint’ hypothesis; Birkhead and 7 Biggins 1987, Westneat et a!. 1990): as breeding synchrony increases, a larger proportion of males allocate their time and energy toward mate guarding and sexual activities with their fertile mate instead of toward pursuing EPCs. Alternatively, if males pursue EPCs instead of guarding their fertile social mate when breeding synchrony is high, then synchrony should be positively related to the level of EPP (the ‘mating opportunity’ hypothesis; Stutchbury and Morton 1995): a concentration of fertile females in space and time should cause a large number ofmales to simultaneously compete for EPCs. Females benefit from obtaining EPCs when synchrony is high because they have more opportunities to directly compare the quality of competing males. Breeding density is another ecological factor that could influence the level of EPP by increasing the encounter rate of potential extra-pair mates (Birkhead and Møller 1992). Habitat structure could affect the level of EPP by influencing a male’s ability to mate guard or a female’s ability to obtain EPCs. For example, in the great grey shrike (Lanius excubitor), individuals chose secretive locations, such as inside tree crowns or bushes, to engage in EPCs whereas open locations were used for within-pair copulations (Tryjanowski et al. 2007); thus, variation in habitat structure may also have the potential to affect the level of EPP. Resource distribution may also influence the level of EPP. For example, if resources are patchy, females may often be distant from their social mate, thus increasing their likelihood of encountering extra-pair males while unguarded (e.g. Reyer et a!. 1997). However, Hoi-Leitner et a!. (1999) showed that female serins (Serinus serinus) on food-supplemented territories were more likely to obtain EPCs than control females, perhaps because food supplementation allowed them to compensate for 8 any retaliatory withholding of male parental care. Other ecological factors that may influence the level of EPP include weather conditions (e.g. Johnsen and Lifjeld 2003) and the nature of the social environment (e.g. the quality of neighboring males; Estep et al. 2005). 1.5 Study Species and Population Song sparrows are socially monogamous passerines which, on Mandarte Island, exhibit genetic promiscuity with 29% of 751 offspring being sired by extra-pair males from 1993-96 (O’Connor et al. 2006). The level of EPP on Mandarte Island is similar to levels estimated in a mainland song sparrow population near Seattle, Washington (24.0% EPP; Hill 1999). The song sparrow population on Mandarte Island provides an ideal system for studying the adaptive significance of EPP. First, nearly all birds alive on the island from 1993-96 have been genotyped at 8 microsatellite by O’Connor et al. (2006) such that the paternity of most offspring and the identity of most extra-pair sires in the population are known. Second, the Mandarte Island population is a relatively closed system with very little emigration, and all nestlings and inmuigrants to the population have been individually color-banded; therefore, all individuals can be identified and monitored closely throughout their lives. Since the population has been monitored continuously since 1975 (Smith et al. 2006), detailed life history data have been collected for nearly all individuals. These include data on life span, lifetime reproductive success, survival to the next season, number of clutches per season, lay date, and age, among other traits. In contrast, many studies of EPP only sample a subset of the study population and are unable to identif’ many extra-pair sires. As a result, sample sizes are often small and 9 measures of EPP have a high degree of uncertainty. Further, few studies have long-term data of the quality collected on Mandarte Island, and lack the ability to track individuals after fledging or compile detailed life history data for individuals. Overall, therefore, the Mandarte song sparrow population offers a nearly ideal population for study. 1.6 Thesis Overview In order to examine the adaptive significance of EPP, I test the good genes hypothesis in Chapter 2 to determine if females that mate with extra-pair males increase the fitness of their offspring. Many studies that test the good genes hypothesis employ modest sample sizes and traits not clearly linked to individual fitness, or fail to test a key good genes prediction that extra-pair young (EPY) are fitter than their within-pair maternal half- siblings. I test this prediction using 751 genotyped offspring from 287 broods over four years (1993-96). The traits I use to compare fitness between EPY and within-pair young (WPY) are closely linked to lifetime reproductive success including life span, the number and proportion of successful social nest attempts produced in a lifetime, survival to independence, survival from independence to recruitment, and the number of independent and recruited genetic offspring (EPY and WPY) produced at age one. I also investigate the relationship of extra-pair mating success and paternity loss to male fitness and conduct >100 paired comparisons of extra-pair and cuckolded males. The traits I use to measure male fitness are closely linked to lifetime reproductive success but independent of extra-pair mating success and paternity loss; traits include life span, the number and proportion of successful social nest attempts produced in a lifetime, annual nest initiation date, the proportion of genetic offspring (EPY and WPY) recruited annually, and male 10 age. Male age is used as a fitness trait to test the good genes hypothesis because older males are predicted to have ‘proven’ their genetic viability by living a relatively long time (reviewed in Brooks and Kemp 2001). I did not use the number of genetic offspring (EPY and WPY) produced annually as a measure of male fitness because it is not independent of extra-pair mating success and paternity loss. In Chapter 3 I examine the effect ofbreeding synchrony and breeding density on the level of EPP within broods. As described above, breeding synchrony may be negatively related to the level of EPP (the ‘mate guarding constraint’ hypothesis; Birkhead and Biggins 1987, Westneat et al. 1990) or positively related to the level of EPP (the ‘mating opportunity’ hypothesis; Stutchbury and Morton 1995). In support of the ‘mate guarding constraint’ hypothesis, I found that breeding synchrony among neighbors was significantly negatively related to the proportion of EPY within broods of focal males. I tested the ‘mate guarding constraint’ hypothesis further by comparing the rate of paternity loss between males that sired EPY outside the fertile period of their social mate and males whose mate’s fertile period overlapped that of the extra-pair female. I also tested whether the level of EPP within broods was related to local breeding density (i.e. the number of neighboring males) and the interaction between density and breeding synchrony. In Chapter 4, I discuss the results ofmy work in the context of the large literature on the evolution of extra-pair paternity in birds. 11 1.7 References Akçay, E. & Roughgarden, J. 2007. Extra-pair paternity in birds: review of the genetic benefits. Evolutionary Ecology Research 9:855-868. Arnqvist, G. & Kirkpatrick, M. 2005. The evolution of infidelity in socially monogamous passerines: the strength of direct and indirect selection on extrapair copulation behaviour in females. American Naturalist 165:S26-S37. Birkhead, T.R. & Biggins, J.D. 1987. Reproductive synchrony and extrapair copulation in birds. Ethology 74:320-334. Birkhead, T.R. & Møller, A.P. 1992. Sperm Competition in Birds: evolutionary causes and consequences. Academic Press, London. Blomqvist, D.D. 2005. High frequency of extra-pair fertilisations in the moustached warbler, a songbird with a variable breeding system. Behaviour 142:1133-1148. Bouwman, K.M., Burke, T. and Komdeur, J. 2006. How female reed buntings benefit from extrapair mating behaviour: testing hypotheses through patterns of paternity in sequential broods. Molecular Ecology 15: 2589-2600. Brooks, R. & Kemp, D.J. 2001. Can older males deliver the good genes? Trends in Ecology and Evolution 16:308-313. Chuang-Dobbs, H.C., Webster, M.S. & Holmes, R.T. 2001. The effectiveness of mate guarding by male black-throated blue warbiers. Behavioral Ecology 12:541-546. Dixon, A., Ross, D., OMalley, S.L.C. and Burke, T. 1994. Paternal Investment Inversely Related to Degree of Extra-Pair Paternity in the Reed Bunting. Nature 371: 698- 700. Estep, L.K., Mays, H., Keyser, A.J., Ballentine, B. & Hill, G.E. 2005. Effects of breeding density and plumage coloration on mate guarding and cuckoldry in blue grosbeaks (Passerina caerulea). Canadian Journal of Zoology 83:1143-1148. Fossøy, F., Johnsen, A. and Lifjeld, J.T. 2007. Multiple genetic benefits of female promiscuity in a socially monogamous passerine. Evolution 62:145-156. • Gray, E.M. 1997a. Do female red-winged blackbirds benefit genetically from seeking extra-pair copulations? Animal Behaviour 53:605-623. 12 Gray, E.M. I 997b. Female red-winged blackbirds accrue material benefits from copulating with extra-pair males. Animal Behaviour 53:625-639. Griffith, S.C. 2000. High fidelity on islands: a comparative study of extrapair paternity in passerine birds. Behavioral Ecology 11:265-273. Griffith, S.C., Owens, I.P.F. & Thuman, K.A. 2002. Extra-pair paternity in birds: a review of interspecific variation and adaptive function. Molecular Ecology 11:2195-2212. Hasseiquist, D., Bensch, S., von Schantz, T. 1996. Correlation between male song repertoire, extra-pair paternity and offspring survival in the great reed warbler. Nature 381:229-232. Hill, C.E. 1999. Song and extra-pair mate choice in song sparrows. Ph.D. Thesis, Dept of Psychology, University of Washington, Seattle. Hoi-Leitner, M., Hoi, H., Romero-Pujante, M. and Valera, F. 1999. Female extra-pair behaviour and environmental quality in the serin (Serinus serinus): a test of the constrained female hypothesis. Proceedings of the Royal Society of London B 266:1021-1026. 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. & Lifjeld, J.T. 2003. Ecological constraints on extra-pair paternity in the blue throat. Oecologia 136:476-483. Johnsen, A., Pam, H., Fossøy, F., Kleven, 0., Laskemoen, T. and Lifjeld J.T. 2008. Is female promiscuity constrained by the presence of her social mate? An experiment with bluethroates Luscinia svecica. Behavioral Ecology and Sociobiology 62:1761-1767. Kempenaers, B. 1993. The use ofbreeding synchrony index. Omis Scandinavica 24:84. Kleven, 0. & Lifjeld, J.T. 2005. No evidence for increased offspring heterozygosity from extrapair mating in the reed bunting (Emberiza schoeniclus). Behavioral Ecology 16:561-565. Komdeur, J., Burke, T., and Richardson, D.S. 2007. Explicit experimental evidence for the effectiveness of proximity as mate-guarding behaviour in reducing extra-pair fertilization in the Seychelles warbler. Molecular Ecology 16:3679-3688. 13 Mulder, P.A., Dunn, P.O., Cockburn, R.A., Lazenby-Cohen, K.A. and Howell, M.J. 1994. Helpers liberate female fairy-wrens from constraints on extra-pair mate choice. Proceedings of the Royal Society of London B 255:223-229. O’Connor, K.D., Man, A.B., Arcese, P., Keller, L.F., Jeffrey, K.J., & Bruford, M.W. 2006. Extra-pair fertilization and effective population size in the song sparrow Melospiza melodia. The Journal of Avian Biology 37:572-578. Petrie, M. & Kempenaers, B. 1998. Extra-pair paternity in birds: explaining variation between species and populations. Trends in Ecology and Evolution 13:52-58. Reyer, H.-U., Bollmann, K., Schlapfer, A.R., Schymainda, A. and Kecack, G. 1997. Ecological determinants of extra-pair fertilizations and egg dumping in Alpine water pipits (Anthus spinoletta). Behavioral Ecology 8:534-543. Sheldon, B.C. 1994. Male phenotype, fertility, and the pursuit of extra-pair copulations by female birds. Proceedings of the Royal Society of London, Series B 257: 25-30. Sheldon, B.C. 2002. Relating paternity to paternal care. Philosophical Transactions of the Royal Society of London B 357:341-3 50. Smith, J.N.M., Keller, L.F., Man, A.B. & Arcese, P. 2006. Biology of small populations: the song sparrows of Mandarte Island. Oxford University Press, New York. Stutchbury, B.J. & Morton, E.S. 1995. The effect ofbreeding synchrony on extra-pair mating systems in songbirds. Behaviour 132:675-690. Suter, S.M., Keiser, M., Feignoux, R. and Meyer, D.R. 2007. Reed bunting females increase fitness through extra-pair mating with genetically dissimilar males. Proceedings of the Royal Society of London B 274:2865-2871. Tregenza, T. & Wedell, N. 2000. Genetic compatibility, mate choice and patterns of parentage: Invited review. Molecular Ecology 9:1013-1027. Tryjanowski, P., Antczak, M. and Hromada, M. 2007. More secluded places for extra-pair copulations in the great grey shrike Lanius excubitor. Behaviour 144:23-31. Westneat, D.F. & Stewart, I.R.K. 2003. Extra-pair paternity in birds: Causes, correlates, and conflict. Annual Review of Ecology Evolution and Systematics 34:365-396. Wetton, J.H. & Parkin, D.T. 1991. An association between fertility and cuckoldry in the house sparrow Passer domesticus. Proceedings of the Royal Society of London B 245:227-233. 14 Westneat, D.F., Schwagmeyer, P.L. and Mock, D.W. 2000. Badge size and extra-pair fertilizations in the house sparrow. Condor 102:342-348. Westneat, D.F., Sherman, P.W. & Morton, W.L. 1990. The ecology and evolution of extra-pair copulations in birds. In: Current Ornithology, vol. 7 (ed. D.M. Power) pp.331-369. Plenum Press, New York. Whittingham, L.A. & Dunn, P.O. 2001. Male parental care and paternity in birds. Current Ornithology 16:257-298. Yasui, Y. 2001. Female multiple mating as a genetic bet-hedging strategy when mate choice criteria are unreliable. Ecological Research 16:605-6 16. Zeh, J.A. & Zeh, D.W. 1997. The evolution of polyandry II. Post-copulatory defences against genetic incompatibility. Proceedings of the Royal Society of London B 264:69-75. 15 2 EXTRA-PAIR MATE CHOICE: A RETROSPECTIVE TEST OF THE GOOD GENES HYPOTHESIS1 2.1 Introduction The advent of molecular genetic techniques has revealed that extra-pair paternity (EPP) is taxonomically widespread and common in birds, a group previously thought mainly to practice monogamy (Griffith et al. 2002). Of the socially monogamous bird species surveyed to date, over 70% have some level of EPP with an average of 11.1% of offspring being extra-pair young (EPY) among socially monogamous species, and 18.7% ofbroods containing at least one EPY (Griffith et al. 2002). Most studies of EPP examine the potential genetic benefits of extra-pair mating to females because extra-pair copulations (EPCs) are thought to be costly to females, yet there are no obvious material benefits that extra-pair males provide to counteract these costs. Females pursuing EPCs may accrue genetic benefits through offspring fitness, such as those related to ‘good genes’ (i.e. additive genetic benefits), genetic compatibility (i.e. non-additive benefits), or genetic diversity (reviewed in Griffith et al. 2002, Akcay and Roughgarden 2007). For example, females that are constrained in their choice of social mate may mate with extra- pair males of higher intrinsic genetic quality than their social mate in order to obtain ‘good genes’ that improve the survival and/or future reproductive success of their offspring (Griffith et al. 2002, Akçay and Roughgarden 2007). 1 A version of this chapter will be submitted for publication. Ames, C.E. and Arcese, P.A. Extra-pair mate choice in the song sparrow (Melospiza melodia): a retrospective test of the good genes hypothesis. 16 Good genes models of female extra-pair mate choice, in particular, have received considerable attention in the literature and are widely debated. The good genes hypothesis has been tested in a variety of bird species, and while some studies have found support for the hypothesis (e.g. Sheldon et a!. 1997, Thusius et al. 2001), others have not (e.g. Augustin et al. 2007, Rosivall et al. 2009). Results also differ between studies of the same (e.g. blue tit (Parus caeruleus): Kempenaers et al. 1997, Deihey et al. 2007) or related species (Tachycineta bicolor; Whittingham and Dunn 2001, Hirundo rusitca; Hirundininae; Kieven et al. 2006a). Mixed results may be due partly to small sample size, because many tests have included less than 200 young, a recommended minimum for estimating population-level patterns of EPP (Griffith et al. 2002). Many studies also lack detailed life history data for individual birds, and thus test ‘good genes’ predictions using putative indexes of fitness, such as body condition (e.g. Sheldon et a!. 1997, Augustin et al. 2007, Rosivall et al. 2009), plumage (e.g. Thusius et a!. 2001, Kleven et al. 2006a, Deihey et al. 2007), immunocompetenee (e.g. Kieven and Lifjeld 2004, Garvin et a!. 2006), or social status (e.g. Otter et a!. 1998), in lieu of more robust indicators, such as seasonal and lifetime reproductive performance (e.g., Arcese 2003, Reid et al. 2005). It remains possible, therefore, that larger, more precise tests of the ‘good genes’ hypothesis will provide additional insight on the adaptive significance of extra-pair mating behavior. Here, I test several predictions of the good genes hypothesis in an individually-marked population of song sparrows, wherein all birds have been studied in detail since 1975, and nearly all birds alive in the population from 1993-1996 were genotyped at ?8 17 microsatellite loci (O’Connor et al. 2006, Smith et al. 2006). O’Connor et a!. (2006) used genetic paternity assignment to estimate that 29% of 751 offspring surviving to six days of age were sired by extra-pair males in this population. I test the key prediction of the good genes hypothesis, that EPY are fitter than their within-pair maternal haif-sibs (Griffith et al. 2002, Akçay and Roughgarden 2007). Differences in the fitness of EPY and within-pair young (WPY) in the same nest should indicate differences in paternal genetic contribution because EPY and WPY share maternal genes and a common rearing environment (Sheldon et al. 1997; but refer to Kempenaers (2009) for a review of how egg-order effects might influence any differences in fitness between EPY and their maternal haif-sibs). Although some studies have found evidence that EPY perform better than WPY (Sheldon et al. 1997, Charmantier et al. 2004, Garvin et a!. 2006, Bouwman et al. 2007, Suter et al. 2007) or that there is no difference between the maternal half-sibs (Whittingham and Dunn 2001, Schmoll et a!. 2003, Kleven and Lifjeld 2004, Augustin et al. 2007, Scbmoll et al. 2009, Rosivall et al. 2009), relatively few studies have tested the key good genes prediction that EPY should be fitter than their maternal haif-sibs (Griffith et al. 2002). In this study, I use a suite of traits linked to lifetime reproductive success in song sparrows to estimate individual fitness of EPY and WPY, including life span, lifetime number and proportion of successful social nest attempts produced, survival to independence, survival from independence to recruitment, and the number of independent and recruited genetic offspring (EPY and WPY) produced at age one (see Table 2.1 for trait definitions and rationale). This study is one of a few studies able to test the good genes hypothesis using robust indicators of fitness (see also Schmoll et al. 2003, 2005, • 2009). 18 The good genes hypothesis also predicts that females mated to less fit males should ‘trade up’ by pursuing extra-pair copulations (EPCs) with fitter males. I tested this prediction by directly comparing the fitness of extra-pair males to the males they cuckolded (Akcay and Roughgarden 2007). I also predicted that males that gain EPP should have higher fitness on average, and fewer EPY in their own nest, when compared to males that did not sire EPY (Griffith et al. 2002, Akçay and Roughgarden 2007). Similarly, I expected that males losing paternity in their own nest would be less fit than males not losing paternity (Griffith et al. 2002, Alcçay and Roughgarden 2007). The traits I use to measure male fitness are closely linked to lifetime reproductive success in song sparrows but independent of extra-pair mating success and paternity loss; traits include life span, the number and proportion of successful social nest attempts produced in a lifetime, annual nest initiation date, and the proportion of genetic offspring (EPY and WPY) recruited annually (see Table 2.1 for trait defmition and rationale). I also asked if male age influenced EPP because older males may have ‘proven’ their genetic viability by living a relatively long time (reviewed in Brooks and Kemp 2001). Finally, the good genes hypothesis assumes that males differ intrinsically in genetic quality. Thus, if females select males based on genetic quality, I also expected that male success at siring EPY and preventing paternity loss would both be repeatable from year to-year (cf. Lessells and Boag 1987). 19 2.2 Methods 2.2.1 Field Methods Mandarte Island is about 6 ha in size and lies 25 km northeast of Victoria, British Columbia, Canada (48° 38’ N, 123° 17’ W). Its resident, semi-isolated population of song sparrows has been studied intensively since 1975 (Smith et al. 2006). All sparrows on the island are uniquely marked as nestlings or immigrants. From 1993-1996, blood samples were taken from most adults and all offspring surviving to banding age (4-6 days post-hatch; henceforth referred to as ‘banded young’). Eggs and offspring dying prior to banding were excluded from analyses because their paternity is unknown. ‘Brood’ is defined as a nest containing at least one ‘banded young’. Survival and population size were estimated annually in April, when the entire population was counted (Smith et al. 2006). Briefly, all birds were monitored regularly each year from March to July, when females typically initiated 2-3 nesting attempts annually. Lay date (first egg of a clutch) was determined by direct observation or back-calculating from hatch date or chick age. Young fledge 9-11 days post-hatch and are cared for by both social parents to 24-28 days of age, when they become ‘independent young’. Offspring became ‘recruits’ to the population when they were known to have survived and remained on the island to 30 April of the following year. These data allow us to estimate seasonal and lifetime reproductive performance of all birds hatched or immigrating to the population (Reid et al. 2005, Smith et al. 2006). 20 2.2.2 Genetic Analysis and Paternity Assignment Genotyping procedures are described in detail in O’Connor et al. (2006) and outlined briefly here. From 1993-1996, blood samples were collected from the brachial vein of all 751 offspring that survived to six days post-hatch and 97% of 242 adults. Eight adults not genotyped from 1993-96 included two females, two socially mated territorial males, one unmated territorial male, and three unmated ‘floaters’. Eight loci were used to genotype all birds: MME1, MME2, MME3, MME7, MME8, and MME12 (Jeffrey et al. 2001), ESCU1 (Hanotte et al. 1994), and GF5 (Petren 1998). One additional locus (PSAP 335; Chan and Arcese 2002) was used in a small number of individuals to reduce uncertainty in paternity. Paternity assignment was conducted by O’Cormor et al. (2006) using maximum likelihood methods and program CERVUS (Marshall et al. 1998), and is described in detail in O’Connor et al. (2006). Briefly, all males one or more years old were considered as candidate sires of all offspring. A genotyping error rate of 3% was used for all simulations based on the mismatch frequency of mothers and offspring, and was reduced in the lab by repeatedly genotyping uncertain individuals. Due to the high average relatedness of sparrows on Mandarte Island, high probabilities of paternity ( 95%) were occasionally estimated for multiple closely related candidate sires. However, because a previous empirical study showed that 98% of extra-pair male song sparrows resided within one territory width of their extra-pair mates (C. Hill personal communication, Hill 1999), O’Connor et al. (2006) weighted raw paternity scores by the distance between the candidate sire and offspring’s territory centre and assigned paternity to the male with the highest distance-weighted LOD score. 21 2.2.3 Traits Related to Fitness The fitness-related traits I used to compare EPY to their within-pair maternal haif-sibs included life span, lifetime number and proportion of successful social nest attempts produced, survival to independence, survival from independence to recruitment, and the number of independent and recruited genetic offspring (EPY and WPY) produced at age one (see Table 2.1 for trait definitions and rationale). Traits used to analyze the distribution of EPP among males included life span, the number and proportion of successful social nest attempts produced in a lifetime, annual nest initiation date, the proportion of genetic offspring (EPY and WPY) recruited annually, and male age (see Table 2.1 for trait definition and rationale). I did not use the number of genetic offspring (EPY and WPY) produced annually as a measure of male fitness because it is not independent of extra-pair mating success and paternity loss. I was unable to measure realized lifetime reproductive success (i.e. the number of genetic EPY and WPY sired over a lifetime) for males because many males lived before 1993 or after 1996, when they may have gained or lost extra-pair paternity undetected by us. Therefore, traits measured over an individual’s lifetime (i.e. life span, the number and proportion of successful social nest attempts produced in a lifetime) were those that did not require genetic data to estimate. Traits measured annually, however, were estimated using the number of genetic offspring (EPY and WPY) produced. To account for interannual variation in the start of breeding, the annual nest initiation date (‘first lay date’) for each individual was standardized by year by calculating the z-score: z (x — IL)/u, where x is the observed value, and ii and u are the year-specific mean and standard 22 deviation of first lay dates in the population, respectively (e.g. Reid et al. 2005). Thus, positive z-scores denote individuals that bred later than average, and negative z-scores denote individuals that bred earlier than average. Male age (years) was treated as a categorical variable, immigrants were assumed to be age 1 on arrival in the population, and birds five years and older were pooled to maintain robust sample sizes (Smith 2006). Male age was used as a potential fitness trait of interest and was controlled for when relating other fitness traits to EPP because many of these traits are also linked to age (Smith et al. 2006). 2.2.4 Statistical Analyses All analyses were performed in SAS 9.1 (SAS Institute, 2003). I applied table-wise Bonferroni corrections to a values for each suite of traits used to test a given good genes prediction (a’ = a/n, where n is the number of traits; Sokal and Rohlf 1995) in order to reduce the type I error rate. Within versus Extra-Pair Young I used generalized linear mixed models (PROC GLIMMIX in SAS) with binary error structure and logit link function to test if offspring paternity (i.e. EPY or WPY) predicted survival from banding to independence, and from independence to recruitment. A Poisson error structure and log link were used to test if offspring paternity predicted reproductive success at age one, life span, and the number of successful nest attempts produced in a lifetime. A binomial error structure and logit link were used to test if offspring paternity predicted the proportion of successful nest attempts produced in a 23 lifetime. In these analyses I used offspring as the unit of analysis and included brood identity as a random factor to control for non-independence among maternal half-siblings (Charmantier et al. 2004). A separate analysis was conducted for each offspring trait. I included lay date, year of the study, age of the offspring’s mother and father, and a ‘lay date x year’ interaction term as covariates in all initial models, and then removed the terms sequentially using backward elimination (P > 0.10). Male Fitness and Repeatability I used generalized linear mixed models with a poisson error structure and log link to test if the number of EPY sired annually was predicted by male life span, the number and proportion of successful nest attempts produced in a lifetime, the first date on which nesting was initiated each season, the proportion of offspring recruited to the population annually, or male age. A poisson error structure was used to analyze the number of EPY sired annually because the data followed a poisson distribution. A binomial error structure and logit link function were used to test if the proportion of EPY a male had within broods annually (i.e. the total number of EPY within broods divided by the total number of banded young within broods) was predicted by the male traits listed above. Male identity was included as a random factor in these analyses and separate analyses were conducted for each trait. I included male age and year as covariates in all initial models but removed the terms by sequential backward elimination ifP> 0.10. Paired t tests were used to compare the traits of extra-pair males to those of the males they cuckolded. 24 I used the intraclass correlation coefficient (Lessells and Boag 1987) to estimate repeatability in the number of EPY sired annually across years and in the proportion of EPY a male had within broods annually across years. 2.3 Results 2.3.1 Overview From 1993-1996, 148 male song sparrows resided on Mandarte Island and contributed one (n = 66), two (n = 37), three (n 21), or four (n = 24) years of data on the number of EPY sired annually, totaling 299 male-years. Of the 299 male-years, 32.4% were from unmated males which on average sired 0.30 EPY annually (± 0.08 SE, range = 0 to 4, nmaleyears = 97), and 67.6% were from mated males which sired about 1.00 EPY annually (mean = 0.94 ± 0.10 SE, range 0 to 7, nmaleyears = 202). Eighty-nine male song sparrows contributed one (n = 34), two (n = 26), three (n = 20), or four (n = 9) years of data on the proportion of EPY within their social broods annually, totaling 182 male-years. On average, males had 1.59 social broods annually (± 0.04 SE, nmaleyears, flmales = 182, 89). Mean brood size at banding was 2.62 offspring (± 0.06 SE, nbroods = 287), but the mean number of WPY in a male’s brood was 29% less (1.86 ± 0.07 SE, flbroods = 287) because 29% of 751 nestlings were EPY (42% of287 broods contained at least one EPY; Table 2.2). Most EPY within a brood (83% of 121) were sired by one extra-pair male; the remaining 17% by two. 25 2.3.2 Within versus Extra-pair Young Contrary to a key prediction of the good genes hypothesis, I found no evidence that EPY were fitter than their maternal, within-pair half sibs (Table 2.3). EPY did not survive better to independence or recruitment than did their maternal half-sibs, and they also did not produce more genetic offspring (EPY and WPY) in their first year of breeding, live longer, or produce a larger number or proportion of successful social nest attempts over their lifetime. Also contrary to the good genes hypothesis, I found that in paired comparisons extra-pair males did not differ significantly from the males they cuckolded in life span, the number and proportion of successful social nest attempts produced in a lifetime, annual nest initiation date, the proportion of genetic offspring recruited to the population annually, or male age (Table 2.4). 2.3.3 Male Fitness and EPY Sired When analyzing all males in the population (i.e. mated and unmated males), I found that the number of EPY sired annually was positively related to male life span (F1,44 = 8.90, flrnaie-years, flmales = 299, 148, P = 0.003) and the number of successful social nest attempts produced in a lifetime (F1,44 = 7.22, nmajeyears, flmales = 299, 148, P = 0.008). In contrast, the number of EPY sired annually was not significantly related to the proportion of successful social nest attempts produced in a lifetime, annual nest initiation date, or the proportion of genetic offspring (EPY and WPY) recruited to the population annually (all P > 0.08). However, the number of EPY sired annually was related to male age (F1,44 = 9.70, flmale-years, flmales = 299, 148, P <0.00 1), with males two to four years old on average being 3.3—4.5 times more likely to sire EPY than first-year males, and 1.4— 1.9 times 26 more likely to sire EPY than males five years and older. It is possible, however, that including unmated males in the above analyses confused relationships between the number of EPY sired, life span, number of successful attempts, and age. This is because unmated males, the majority of which do not sire EPY (84.5% of 97 male-years), are mainly yearlings (Smith et al. 2006). Males that are unmated in their first year also display low lifetime reproductive success (Smith 1988). I therefore also conducted parallel analyses that included only mated males. When analyzing only mated males in the population, I found that the number of EPY sired annually was not significantly related to life span, the number and proportion of successful social nest attempts produced in a lifetime, the annual nest initiation date, or the proportion of genetic offspring (EPY and WPY) recruited to the population annually (Table 2.5). However, the number of EPY sired annually was significantly related to male age (Table 2.5, Figure 2.1): males two to four years old on average were 3.1-4.7 times more likely to sire EPY than first-year males, and 1.3-2.0 times more likely to sire EPY than males five years and older. Thus, male age appears to influence the number of EPY sired annually independent of male mating status. I also found that the number of EPY sired annually was not significantly related to the number or proportion of EPY a male had in his own nests annually (F1,88 2.87, nmale years, flmales 182, 89, P = 0.094, and F1,88 = 1.35, nma1eys, maJes 182, 89, P = 0.249, respectively). In fact, 50% of males siring one or more EPY were also cuckolded (nmaie years, flmales = 84, 53). However, I did find that extra-pair males, on average, gained more 27 paternity than they lost annually (mean difference = 0.89 ± 0.25 SE; paired t-test: t = 3.61, P <0.001, flrnale-years = 84). Males exhibited no repeatability in the number of EPY sired annually across years when all males in the population were analyzed (repeatability 0.01, flmale-years, males = 233, 82, F81,151 = 1.03, P = 0.428) and when only mated males were analyzed (repeatability = -0.02, nma1eyrs, flmales= 175, 64, F63111 = 0.95, P 0.590). 2.3.4 Male Fitness and Paternity Loss The proportion of EPY a male had within social broods annually was not significantly related to life span, the number and proportion of successful social nest attempts produced in a lifetime, annual nest initiation date, the proportion of genetic offspring (EPY and WPY) recruited to the population annually, or male age (Table 2.6). Males exhibited low but statistically significant repeatability in the proportion of EPY within their own broods annually across years (repeatability = 0.17, flmale-years, nmaj = 148, 55, F54,93= 1.55, P = 0.032). Repeatability in the proportion of EPY within broods may have been determined, in part, by the male’s social mate because males had the same mate in 51.6% of 93 consecutive first broods, and a minority of males (4 of 96 males) switched mates between breeding attempts within a year. 28 2.4 Discussion 2.4.1 Within versus Extra-Pair Young My results do not support a key prediction of the good genes hypothesis because I found no difference in survival, life span, or seasonal and lifetime reproductive performance between EPY and their within-pair maternal half-siblings. These results are consistent with several other studies in passerines (e.g. Whittingham and Dunn 2001, Schmoll et al. 2003, Kleven and Lifjeld 2004, Augustin et al. 2007, Schmoll et al. 2009, Rosivall et al. 2009). However, my results improve upon many earlier studies in that I examine a more complete and diverse set of seasonal and long-term fitness measures, as recommended by Griffith et al. (2002). It is possible that females do not benefit from EPP, and engage in EPCs, for example, in order to make the ‘best of a bad job’ if there is strong selection in males to achieve EPCs at the expense of female fitness (sexually antagonistic coevolution; Arnqvist and Kirkpatrick 2005). Nevertheless, I cannot rule out that good genes represent a benefit to female song sparrows that mate with extra-pair males. First, it is possible that the fitness-related traits I used (Table 2.1) did not accurately capture male quality, perhaps due to high environmental variance in food availability, population density, weather conditions, or predator/prey dynamics (e.g., Smith 1988, Arcese 2003). Although the traits I examined were closely linked to lifetime reproductive success based on the social mating system in song sparrows (Table 2.1), I was unable to measure realized lifetime reproductive success (i.e. the number of genetic EPY and WPY sired over a lifetime) for males or their sons. While no study of EPP to date has been able to measure realized lifetime reproductive success, it would be a more accurate fitness measure for determining the presence or absence of ‘good genes’ effects (Schmoll et al. 29 2009). A second possibility that prevents me from rejecting the good genes hypothesis as applied to mate choice in song sparrows is that the genetic benefits of extra-pair mating may simply be too small to detect even with large, detailed datasets such as this one, despite driving the evolution of mating behavior over long timeframes (e.g. Møller and Alatalo 1999). Although I was unable to detect any significant differences in fitness between EPY and their maternal half-sibs, I did find that EPY had, on average, marginally higher trait values than their maternal haif-sibs for four of the seven fitness traits analyzed (Table 2.3): survival from independence to recruitment, the number of independent and recruited genetic offspring produced at age one, and the proportion of successful social nest attempts produced in a lifetime. This suggests that there may have been a fitness difference between the haif-sibs that was too small to detect using conventional criterion for determining significance. If fitness differences between EPY and their maternal half-sibs were too small to detect, my current results imply that empirical field studies of good genes and related hypotheses may prove challenging in the absence of very large samples and more precise indexes of individual fitness than used here (Table 2.1). Third, there may have been sampling bias that masked fitness differences between EPY and their maternal haif-sibs. For example, genetic parentage was determined for offspring that survived to 4-6 days post-hatch when banding and blood collection were possible. However, from 1993-96, 11% of 959 eggs did not hatch and 12% of 854 hatchlings did not survive to banding; therefore, the genetic parentage of at least 206 offspring (21% of offspring) could not be determined, such that any 30 differences between EPY and WPY in hatch rate or in survival from hatch to banding would not have been detected. If, for example, EPY survived better to banding than their maternal haif-sibs, I would have been unable to detect this ‘good genes’ effect. Fourth, it remains possible that the genetic benefits of female extra-pair mate choice may be context-dependent and vary with environmental conditions (e.g. Schmoll et al. 2005). For example, in coal tits (Parus ater), the probability of offspring recruiting to the population is negatively related to hatch date; Schmoll et al. (2005) found that coal tit EPY that hatched relatively late in the season had a higher probability of recruiting locally than their maternal half-sibs while there was no difference in recruitment probability between EPY and WPY that hatched earlier in the season. 2.4.2 Male Fitness and EPY Sired In addition, I found no strong evidence that more fit males had higher extra-pair mating success. Among mated males, the number of EPY a male sired annually was not significantly related to life span, the lifetime number or proportion of successful social nest attempts, the annual nest initiation date, or the proportion of genetic offspring recruited to the population annually. Further, females did not appear to mate with extra- pair males that were more fit than their social mate, and males that sired more EPY did not have fewer EPY in their own nests. I also found that male success at siring EPY was not repeatable across years, indicating that the intrinsic quality of a male did not determine his success at siring EPY. This suggests that extrinsic factors, perhaps related to territory size, quality or location, the behavior of social mates, or social interactions with new or existing neighbors, introduce variation in a male’s investment or success in 31 EPP from year to year. However, male age was related to the number of EPY sired annually, such that first-year males and males aged five years and older (i.e. very old males) sired fewer EPY than males aged two to four years (i.e. middle-aged males). These observations are consistent with similar, well-established patterns of age-related variation in annual reproductive success in song sparrows, wherein first-year males experience the lowest reproductive success (Smith et al. 2006), are more likely to remain as non-territorial floaters or unmated territory holders than two and three-year olds (Arcese 1987, Smith and Arcese 1989), and where males five years and older experience declines in reproductive success as compared to middle-aged males (Smith et al. 2006). Similarly, Arcese (1987, 1989a,b) showed that two and three-year-old males were the most likely to retain territories in the face of challenges by non-territorial floaters, and were also more likely than younger and older males to engage in polygyny. Overall, these trends suggest that first-year males often lack the physical ability, experience or resources required to successfully gain EPP, and that older males suffer a reduced ability to gain EPP due to senescence (see also Keller et al. 2008). It is possible that middle-aged males are better at creating or exploiting extra-pair mating opportunities (e.g. Kieven et al. 2006b) or at providing extra pair females with direct benefits, such as nest defense against predators or additional foraging opportunities on their territory (e.g. Gray 1997). Detailed behavioral data on male and female extraterritorial forays and female resistance to EPCs are required to determine if EPCs are primarily male or female driven. Further, behavioral data indicating any direct benefits that females may be receiving from extra-pair males should 32 be investigated, for example, through observation of female extraterritorial forays during brood rearing, or response of males to predators near the extra-pair female’s nest. Male age is often an important predictor of extra-pair mating success, with many studies demonstrating that older males sire more EPY than younger males (e.g. Griffith et al. 2002, Kieven et al. 2006b, Bouwman et al. 2007, Schmoll et al. 2007). Although some evidence suggests that female birds prefer older males as extra-pair mates, perhaps because old age signals male viability and genetic quality (reviewed in Brooks and Kemp 2001), my results do not support this hypothesis because I demonstrate a decline in male extra-pair mating success for males aged five years and older (Figure 2.1), and extra-pair males were not older than the males they cuckolded. To my knowledge, no other study has measured a decline in extra-pair mating success in very old males, although Schmoll et al. (2007) demonstrated that extra-pair mating success in male coal tits increased between the ages of one and three years, then leveled off after the age of three. This result is potentially of interest to field ecologists because the exact age of every bird was known, enabling us to measure with greater precision age-related variation in EPP, whereas many studies coarsely divide males into ‘young’ and ‘old’ categories. I suggest that future studies use caution in interpreting results in the absence of detailed data on male age. 2.4.3 Male Fitness and Paternity Loss I also found that the proportion of EPY a male had within broods annually was not significantly related to life span, the number or proportion of successful social nest 33 attempts produced in a lifetime, annual nest initiation date, or the proportion of genetic offspring recruited to the population annually. Further, male age was not significantly related to the proportion of EPY a male had within broods, similar to several studies in passerines (e.g. Augustin et al. 2007, Bouwman et al. 2007, Neuman et al. 2007). However, males exhibited low but significant repeatability in the proportion ofpaternity lost from their own broods across years, suggesting that the level ofpaternity loss may have been an intrinsic trait of individual males. There may be several reasons why females consistently cuckold individual males if not to obtain ‘good genes’ benefits. For example, some males may be cuckolded if they are unable to provide their social mate with adequate nest defense or breeding resources, which their social mate might then have to obtain from extra-pair males. On the other hand, some males may lose paternity from their own nest if they do not adequately guard their social mate during the fertile period. For example, male white-throated sparrows (Zonotrichia albicollis) with a tan morph spend a larger proportion of time guarding their social mate from EPCs and, therefore, have fewer EPY in their own nests; however, males with a white morph aggressively pursue EPCs and subsequently have a higher proportion of EPY in their own nests because they spend less time guarding their social mate from males seeking EPCs (Tuttle 2003). It is also possible that repeatability in the proportion of EPY within broods may have been determined, in part, by the male’s social mate because a majority of males had the same mate in consecutive first broods and between breeding attempts within a year. For example, a study in coal tits (Parus ater) showed that pair identity was related to the proportion of EPY within broods, suggesting that interactions of characteristics of the male and his social mate might predict the proportion of EPY within broods (Dietrich 34 et al. 2004). In order to determine the relative role ofmales and females in determining the rate of EPP in song sparrows, studies are required that compare changes in the rate of EPP across consecutive broods when mate switching does and does not occur. 2.4.4 Conclusion In conclusion, I did not find support for the good genes hypothesis despite using large sample sizes and testing a diverse set of seasonal and long-term fitness measures. My results suggest that females may not obtain fitness benefits from EPP, and it is possible that females engage in EPCs as a result of sexually antagonistic coevolution. However, data on realized lifetime reproductive success of males and their sons are required to further assess potential ‘good genes’ benefits. I did find that age predicted a male’s success at siring EPY, with first-year and very old males siring fewer EPY than middle- aged males, a trend unique to studies to date. This result suggests that there may be age- related variation in the physical ability or experience of male song sparrows to sire EPY, rather than females preferring to mate with males that have ‘proven’ their viability. However, detailed behavioral studies are required to further investigate this possibility, particularly those that examine the exact age of individuals in relation to extra-pair mating success, rather than using general ‘young’ and ‘old’ age categories. My results also suggest that a broader range of hypotheses must be tested in the future to explain variation in EPP within species. To date, a vast majority of studies have focused on the potential indirect benefits of EPP, but have largely ignored the potential influence of direct benefits or ecological factors such as breeding synchrony. The mixed nature of my results, including a lack of repeatability in male extra-pair mating success, is most 35 consistent with the idea that variation in EPP results as a consequence of various constraints related to ecological or other extrinsic factors operating on the time budgets of individual birds with varying abilities or opportunities to engage in extra-pair mating as a means to increase lifetime reproductive success. If true, experimental studies that induce variation among individuals in the ability or opportunity to engage in extra-pair mating will be required to differentiate among hypotheses. 36 Ta bl e 2. 1 T ra its re la te d to fit ne ss in so n g sp ar ro w s o n M an da rte Is la nd Fo ca l T ra it M ea su re d in D ef in iti on Ti m es ca le R at io na le M al e (M ) o r O ffs pr in g (0 )? Li fe sp an M , 0 To ta ly ea rs al iv e o n M an da rte Li fe tim e M ajo rd et er m in an to f l ife tim e re pr od uc tiv e su cc es s in so n g sp ar ro w s (S mi th et al. 20 06 ). M ay in di ca te ge ne tic v ia bi lit y (B roo ks an d K em p 20 01 ). N um be r s u cc es sf ul M , 0 To ta ln u m be ro fs o ci al n es ta tte m pt s th at Li fe tim e N um be ro fb re ed in g at te m pt s o v er lif et im e hi gh ly so ci al n es ta tte m pt s pr od uc ed at le as to n e in de pe nd en t co rr el at ed w ith lif et im e re pr od uc tiv e su cc es s in so n g o ffs pr in g re ga rd le ss o fg en et ic pa te rn ity sp ar ro w s (S mi th 19 88 ). In de pe nd en to ffs pr in g cr ite rio n fo r‘ su cc es s’ be ca us e m al es as su m e m ajo rit y o f p ar en ta lc ar e fo ro ffs pr in g be tw ee n fle dg in g an d in de pe nd en ce (S mi th et al. 20 06 ). Pr op or tio n su cc es sf ul M , 0 To ta lp ro po rti on o fs o ci al n es t at te m pt s Li fe tim e In di ca te s ef fic ie nc y w ith w hi ch m al es ra ise d o ffs pr in g so ci al n es ta tte m pt s th at pr od uc ed at le as t o n e in de pe nd en t to in de pe nd en ce pe rn es ta tte m pt (se ea bo ve ). o ffs pr in g re ga rd le ss o fg en et ic pa te rn ity A nn ua ln es ti ni tia tio n M D at e o n w hi ch an in di vi du al o r its m at e Se as on Ea rly br ee de rs br ee d m o re fre qu en tly w ith in a se as o n da te la id its fir st eg g o f t he se as o n , an d pr od uc e a la rg er n u m be ro fo ffs pr in g th at be co m e st an da rd iz ed fo ry ea r( see M et ho ds ). su cc es sf ul br ee de rs (S mi th 19 88 , S m ith et al. 20 06 ). Pr op or tio n ge ne tic M Pr op or tio n o f i nd ep en de nt ge ne tic Se as on R ef le ct s di ffe re nc es in o v er -w in te r s u rv iv al o f o ffs pr in g re cr u ite d o ffs pr in g (E PY an d W PY ) r ec ru ite d o n o ffs pr in g an d ac co u n ts fo rm o st v ar ia tio n in lif et im e M an da rte Is la nd st ud y ar ea re pr od uc tiv e su cc es s (S mi th 19 88 ). A ge M N um be ro fy ea rs al iv e sin ce ye ar o f h at ch Se as on M ay in di ca te ge ne tic v ia bi lit y (B roo ks an d K em p 20 01 ). Su rv iv al to 0 W he th er o r n o ta n in di vi du al su rv iv ed to Se as on M ay in di ca te o ffs pr in g v ia bi lit y. in de pe nd en ce in de pe nd en ce Su rv iv al fro m 0 W he th er o r n o ta n in di vi du al su rv iv ed Se as on A n im po rta nt fa ct or in de te rm in in g lif et im e in de pe nd en ce to fro m in de pe nd en ce to re cr u itm en t re pr od uc tiv e su cc es s be ca us e th e m ajo rit yo fa du lts re cr u itm en t su rv iv e fo ro n ly o n e br ee di ng se as o n (S mi th 19 88 ). N um be r i nd ep en de nt 0 N um be ro fi nd ep en de nt an d re cr u ite d Se as on A n im po rta nt fa ct or in de te rm in in g lif et im e an d re cr u ite d ge ne tic ge ne tic o ffs pr in g (E PY an d W PY ) re pr od uc tiv e su cc es s (S mi th 19 88 ). o ffs pr in g pr od uc ed at pr od uc ed in th e fir st po ss ib le br ee di ng ag e o n e ye ar (ag ed o n e) Table 2.2 Percentage of extra-pair young (EPY) and percentage of song sparrow broods on Mandarte Island with at least one EPY from 1993 to 1996 Year Percentage of offspring Percentage of nests 1993 27.3% (48/176) 41.4% (29/70) 1994 27.0% (43/159) 42.4% (28/66) 1995 3 1.1% (70/225) 42.1% (32/76) 1996 30.4% (58/191) 42.7% (32/75) Total 29.2% (219/751) 42.2% (121/287) 38 Table 2.3 Differences in survival, life span, and seasonal and lifetime reproductive performance between EPY song sparrows and their maternal, within-pair haif-sibs on Mandarte Island. A-G represent final models from separate analyses. Estimates ± SE are on the logit scale (A,B,G; binary/binomial response) or log scale (C-F; Poisson response), and represent least-square means ± SE for ‘EPY’ and ‘WPY’, regression coefficient ± SE for ‘date of first egg’, and variances ± SE for random brood intercepts. noffspnng and nbroods indicate sample sizes of total offspring and individual broods, respectively. The table-wise Bonferroni corrected a-value for 7 tests of offspring traits is 0.007. Model Estimate ± SE df F P(n,) (A) Survival to independence EPY 751 (287) 0.550 ± 0.174 1,463 0.46 0.496 WPY 0.680 ± 0.126 social father’s age - 4,463 3.25 0.012 year - 3,463 19.29 <0.001 broodiD 0.386±0.182 - - (B) Survival from independence to recruitment EPY 471(234) -0.112±0.205 1,235 1.78 0.184 WPY -0.415±0.140 mother’s age - 4,235 2.20 0.069 date of first egg -0.022 ± 0.005 1,235 17.09 <0.001 year - 3,235 10.15 <0.001 brood ID negligible - - (C) Number independent genetic offspring produced at age one EPY 126(95) 0.108±0.191 1,30 0.10 0.754 WPY 0.038±0.134 year - 2,30 2.99 0.066 brood ID 0.483±0.175 - - 39 Model Estimate ± SE df F P(n) (D) Number recruited genetic offspring produced at age one EPY 126 (95) -0.986 ± 0.282 1,30 0.06 0.807 WPY -1.070±0.196 brood ID 0.364 ± 0.256 - - (E) Life span EPY 168(134) 0.722±0.109 1,33 1.83 0.185 WPY 0.893 ± 0.068 brood ID 0.100 ± 0.047 - - (F) Number successful social nest attempts in lifetime EPY 168 (134) 0.498 ± 0.144 1,33 0.75 0.392 WPY 0.640±0.100 broodiD 0.552±0.121 - - (G) Proportion successful social nest attempts in lifetime EPY 125(106) 0.700±0.179 1,18 1.39 0.254 WPY 0.455±0.110 broocliD 0.073±0.119 - - 40 Table 2.4 Paired comparisons of traits of extra-pair and cuckolded male song sparrows on Mandarte Island Means ± SE are presented. ‘n’ is the number of paired comparisons. t- and p-values are from paired t test. The table-wise Bonferroni corrected a-value for 6 tests of offspring traits is 0.008. Timescale Trait Social Male Extra-Pair Male n t P Lifetime Lifespan 4.271±0.173 4.674±0.190 129 1.58 0.117 Number successful social 5.109±0.260 5.124±0.282 129 0.04 0.968 nest attempts Proportion successful social 0.586 ± 0.018 0.587 ± 0.019 123 0.15 0.885 nest attempts Annual Annual nest initiation date -0.205 ± 0.072 -0.293 ± 0.067 110 -0.74 0.459 Proportion genetic offspring 0.402 ± 0.038 0.414 ± 0.033 93 0.23 0.8 16 recruited Age 2.620±0.107 2.682±0.103 129 0.45 0.656 41 T ab le 2. 5 R el at io ns hi p o f t he n u m be r o fE PY si re d a n n u a lly by m a te d m a le so n g sp ar ro w s o n M an da rt e Is la nd to lif e sp an , se a so n a la n d lif et im e re pr od uc tiv e pe rf or m an ce ,a n d ag e Th e n u m be ro fE PY sir ed by m al es (de pe nd en tv ar ia bl e) ha sb ee n re la te d to ea ch o ft he fit ne ss -re la te d tr ai ts (ex pla na tor yv ar ia bl es ) lis te d be lo w . V ar ia bl es re ta in ed in fin al m o de ls ar e in di ca te d in bo ld . Es tim at es ± SE ar e o n th e lo g sc al e, an d re pr es en tr eg re ss io n co ef fic ie nt ± SE fo rf itn es s- re la te d tr ai ts, an d v ar ia nc es ± SE fo rr an do m m al e in te rc ep ts. n in a1 e. .y e an d fl m al es ar e sa m pl e siz es o ft ot al m al e- ye ar o bs er va tio ns an d in di vi du al m al es ,r es pe ct iv el y. Th e ta bl e- w ise B on fe rro ni co rr ec te d cL -v alu e fo r6 te st s o fm al e tr ai ts is 0. 00 8. * de no te s v ar ia bl e sig ni fic an ce . Fi tn es s- Re la te d Tr ai t Li fe sp an N um be rs u cc es sf ul so ci al n es t a tte m pt s Pr op or tio n su cc es sf ul so ci al n es ta tte m pt s A nn ua ln es ti ni tia tio n da te Pr op or tio n ge ne tic o ffs pr in g re cr u ite d St at ist ic s fo rF itn es s- Re la te d Tr ai t (n in ai ) Es tim at e ± SE F P 20 2 (91 ) 0. 12 9 ± 0,0 55 5. 55 0. 02 0 20 2 (91 ) 0. 06 6 ± 0. 03 7 3. 18 0. 07 8 20 2 (91 ) - 0. 19 8 ± 0. 49 8 0. 16 0. 69 2 Ti m es ca le Li fe tim e A nn ua l M al e ag e Y ea r M al e id en tit y F P F P Es tim at e ± SE 4. 45 0. 00 2* 1. 99 0. 11 9 0. 42 1± 0. 15 0 4. 59 0. 00 2* 2. 07 0. 10 9 0. 43 4± 0. 15 1 5. 69 < 0. 00 1* 2. 28 0. 08 3 0. 46 1 ± 0. 15 4 20 1 (91 ) - 0. 29 1± 0. 12 6 3. 04 0. 08 4 3. 08 0. 01 9 1. 99 0. 12 0 0. 47 2± 0. 15 7 16 1 (78 ) 0. 05 9± 0. 24 3 0. 06 0. 80 9 3. 34 0. 01 4 0. 74 0. 53 4 0. 34 9± 0. 13 3 A ge 20 2 (91 ) - 5. 72 < 0. 00 1* - - 2. 27 0. 08 5 0. 45 1 ± 0. 15 1 Ta bl e 2. 6 R el at io ns hi p o ft he pr op or tio n o f E PY w ith in br oo ds a n n u a lly fo rm al e so n g sp ar ro w s o n M an da rte Is la nd to m al e lif e sp an ,s e a so n a la n d lif et im e re pr od uc tiv e pe rf or m an ce ,a n d ag e Th ep ro po rti on o fE PY w ith in br oo ds an n u al ly fo rm al es (de pe nd en tv ar ia bl e) ha sb ee n re la te d to ea ch o f t he fit ne ss -re la te d tra its (ex pla na tor yv ar ia bl es )l ist ed be lo w . V ar ia bl es re ta in ed in fin al m o de ls ar e in di ca te d in bo ld . Es tim at es ± SE ar e o n th e lo gi ts ca le, an d re pr es en tr eg re ss io n co ef fic ie nt ± SE fo rf oc al tra its ,a n d v ar ia nc es ± SE fo rr an do m m al e in te rc ep ts. rn a 1e -y ea rs an d flr na le s ar e sa m pl e siz es o ft ot al m al e- ye ar o bs er va tio ns an d in di vi du al m al es ,r es pe ct iv el y. Th e Bo nf er ro ni co rr ec te d a- v al ue fo r6 te st s o fm al e tra its is 0. 00 8. St at ist ic s fo rF itn es s- Re la te d flm ale -y ea rs Tr ai t M al e ag e Y ea r M al e id en tit y Ti m es ca le Fi tn es s- Re la te d Tr ai t (fl rna ies ) Es tim at e ± SE F P F P F P Es tim at e ± SE Li fe tim e Li fe sp an 18 2 (89 ) - 0. 20 6 ± 0. 07 9 6. 83 0. 01 0 3. 20 0. 01 7 0. 24 0. 86 7 0. 98 6 ± 0. 28 4 N um be r s u cc es sf ul so ci al 18 2 (89 ) - 0. 12 6 ± 0. 05 2 5. 96 0. 01 7 3. 19 0. 01 7 0. 39 0. 75 9 0. 97 5 ± 0. 28 6 n es ta tte m pt s Pr op or tio n su cc es sf ul so ci al 18 2 (89 ) - 0. 88 4 ± 0. 66 9 1. 75 0. 19 0 2. 57 0. 04 3 0. 22 0. 88 2 1. 09 9 ± 0.3 10 n es ta tte m pt s A nn ua l A nn ua l n es ti ni tia tio n da te 18 2 (89 ) - 0. 18 2 ± 0. 15 0 1. 47 0. 22 8 2. 66 0. 03 8 0. 27 0. 84 5 1. 16 0 ± 0.3 19 Pr op or tio n ge ne tic o ffs pr in g 15 4 (77 ) 0. 03 6 ± 0. 34 4 0. 01 0. 91 7 2. 40 0. 05 8 0. 61 0. 60 9 0. 88 9 ± 0. 30 3 re cr u ite d A ge 18 2 (89 ) - 2. 56 0. 04 4 - - 0. 18 0. 90 8 1. 12 2 ± 0.3 11 1717 (siioX)VItN £ 0 ‘-nCD C Tj I CD c•T -1 pusatdansuuisaInbs-4so (iooo>1‘16‘OZ =SaIWUSJICIWH‘ç= 1’wpucpus.iuprnpjuoso.uidsuos PPW£qAii1nU11uP.I!SAJ3JO.IqwnuI{JuAq3qdqsuo!JupUt 17I 2.5 References Akcay, E. & Roughgarden, J. 2007. Extra-pair paternity in birds: review of the genetic benefits. Evolutionary Ecology Research 9:855-868. Arcese, P. 1987. Age, intrusion pressure and defense against floaters by territorial male song sparrows. Animal Behaviour 35:773-784. Arcese, P. 1989a. Territory acquisition and loss in male song sparrows. Animal Behaviour 37:45-55. Arcese, P. 1989b. Intrasexual competition and the mating system of primarily monogamous birds: the case of the song sparrow. Animal Behaviour 38:96-111. Arcese, P. 2003. Individual quality, environment, and conservation. In: Animal Behavior and Wildlife Conservation (eds Festa-Bianchet, M. and Apollonio, M.) pp.271- 295. Island Press, Washington DC. Arnqvist, G. & Kirkpatrick, M. 2005. The evolution of infidelity in socially monogamous passerines: the strength of direct and indirect selection on extrapair copulation behaviour in females. American Naturalist 165:S26-S37. Augustin, J., Blomqvist, D., Szep, T., Szabo, Z.D. and Wagner, R.H. 2007. No evidence of genetic benefits from extra-pair fertilisations in female sand martins (Riparia riparia). Journal of Ornithology 148:189-198. Bouwman, K.M, Van Dijk, R.E., Wijmenga, J.J. & Komdeur, J. 2007. Older male reed buntings are more successful at gaining extra-pair fertilizations. Animal Behaviour 73:15-27. Brooks, R. & Kemp, D.J. 2001. Can older males deliver the good genes? Trends in Ecology and Evolution 16:308-3 13. Chan, Y. & Arcese, P. 2002. Subspeciflc differentiation and conservation of song sparrows (Melospiza melodia) in the San Fransisco Bay region inferred by microsatellite loci analysis. The Auk 119:641-651. Charmantier, A., Blondel, J., Perret, P. and Lambrechts, M.M. 2004. Do extra-pair paternities provide genetic benefits for female blue tits Parus caeruleus. Journal of Avian Biology 35:524-532. 45 Deihey, K., Peters, A., Johnsen, A. and Kempenaers, B. 2007. Fertilization success and UV ornamentation in blue tits Cyanistes caeruleus: correlational and experimental evidence. Behavioral Ecology 18: 399-409. Dietrich, V., Schmoll, T., Winkel, W., Epplen, J.T. & Lubjuhn, T. 2004. Pair identity — an important factor concerning variation in extra-pair paternity in the coal tit (Parus ater). Behaviour 141 :817-835. Garvin, J.C., Abroe, B., Pederson, M.C., Dunn, P.O. and Whittingham, L.A. 2006. Immune response of nestling warbiers varies with extra-pair paternity and temperature. Molecular Ecology 15:3833-3840. Gray, E.M. 1997. Female red-winged blackbirds accrue material benefits from copulating with extra-pair males. Animal Behaviour 53:625-639. Griffith, S.C., Owens, I.P.F. & Thuman, K.A. 2002. Extra-pair paternity in birds: a review of interspecific variation and adaptive function. Molecular Ecology 11:2195-2212. Hanotte, 0., Zanon, C., Pugh, A., Greig, C., Dixon, A. & Burke, T. 1994. Isolation and characterization of microsatellite loci in a passerine bird: the reed bunting Emberiza schoeniclus. Molecular Ecology 3:529-530. Hill, C.E. 1999. Song and extra-pair mate choice in song sparrows. Ph.D. Thesis, Dept of Psychology, University of Washington, Seattle. Jeffrey, K.J., Keller, L.F., Arcese, P. & Bruford, M.W. 2001. The development of microsatellite loci in the song sparrow, Melospiza melodia (Ayes), and genotyping error associated with good quality DNA. Molecular Ecology Notes 1:11-13. Keller, L.F., Reid, J.M. & Arcese, P. 2008. Testing evolutionary models of senescence in a natural population: age and inbreeding effects on fitness components in song sparrows. Proceedings of the Royal Society of London B 275:597-604. Kempenaers, B., Verheyren, G.R., & Dhondt, A.A. 1997. Extra-pair paternity in the blue tit Parus caeruleus: female choice, male characteristics, and offspring quality. Behavioral Ecology 8:481-492. Kempenaers, B. 2009. Behavioural Ecology: Cuckolder eggs come first. Current Biology 19:R364-R366. Kleven, 0., Jacobsen, F., Izadnegahdar, R., Robertson, R.J. and Lifjeld, J.T. 2006a. Mail tail streamer length predicts fertilization success in the North American barn swallow (Hirundo rustica erythrogaster). Behavioral Ecology and Sociobiology 59:412-418. • 46 Kieven, 0. and Lifjeld, J.T. 2004. Extrapair paternity and offspring immunocompetence in the reed bunting, Emberiza schoeniclus. Animal Behaviour 68:283-289. Kieven, 0., Marthinsen, G., & Lifjelt, J.T. 2006b. Male extra-territorial forays, age and paternity in the socially monogamous reed bunting (Emberiza schoeniclus). Journal of Ornithology. 147:468-473. Lessells, C.M. & Boag, P.T. 1987. Unrepeatable repeatabilities: A common mistake. The Auk 104:116-121. Marshall, T.C., Slate, J., Kruuk, L. & Pemberton, J.M. 1998. Statistical confidence for likelihood-based paternity inference in natural populations. Molecular Ecology 7:639-655. Møller, A.P. & Alatalo, R.V. 1999. Good-genes effects in sexual selection. Proceedings of the Royal Society of London B. 266:88-91. Neuman, C.R., Safran, R.J. & Lovette, I.J. 2007. Male tail streamer length does not predict apparent or genetic reproductive success in the North American barn swallows Hirundo rustica eiythrogaster. Journal of Avian Biology 38:28-36. O’Connor, K.D., Marr, A.B., Arcese, P., Keller, L.F., Jeffrey, K.J., & Bruford, M.W. 2006. Extra-pair fertilization and effective population size in the song sparrow Melospiza melodia. The Journal of Avian Biology 37:572-578. Otter, K., Ratcliffe, L., Michaud, D., & Boag, P.T. 1998. Do female black-capped chickadees prefer high-ranking males as extra-pair mates? Behavioural Ecology and Sociobiology 42:25-3 6. Petren, K. 1998. Microsatellite primers from Geospizafortis and cross-species amplification in Darwin’s finches. Molecular Ecology 7:1771-1788. Reid, J.M., Arcese, P., Cassidy, A.L.E.V., Hiebert, S.M., Smith, J.N.M., Stoddard, P.K., Marr, A.B. & Keller L.F. 2005. Fitness correlates of song repertoire size in free- living song sparrows (Melospiza melodia). American Naturalist 165:299-310. Rosivall, B., Szöllôsi, E., Hasseiquist, D. and Török, J. 2009. Effects of extrapair paternity and sex on nestling growth and condition in the collared flycatcher, Ficedula albicollis. Animal Behaviour 77:611-617. SAS Institute Inc. 2003. SAS version 9.1. Cary, NC, USA. Schmoll, T., Dietrich, V., Winkel, W., Epplen, J.T. & Lubjuhn, T. 2003. Long-term fitness consequences of female extra-pair matings in a socially monogamous passerine. Proceedings of the Royal Society Biological Sciences B 270:259-264. 47 Schmoll, T., Dietrich, V., Winkel, W., Epplen, J.T., Schurr, F. & Lubjuhn, T. 2005. Paternal genetic effects on offspring fitness are context dependent within the extra-pair mating system of a socially monogamous passerine. Evolution 59:645-657. Schmoll, T., Munci, V., Dietrich-Bischoff, V., Winkel, W. and Lubjuhn, T. 2007. Male age predicts extrapair and total fertilization success in the socially monogamous coal tit. Behavioral Ecology 18:1073-1081. Schmoll, T., Schurr, F.M., Winkel, W., Epplen, J.T. and Lubjuhn, T. 2009. Lifespan, lifetime reproductive performance and paternity loss of within-pair and extra-pair offspring in the coal tit Periparus ater. Proceedings of the Royal Society of London B 276:337-345. Sheldon, B.C., Merila, 3., Qvamstrom, A., Gustafsson, L. & Ellegren, H. 1997. Paternal genetic contribution of offspring condition predicted by size of male secondary sexual character. Proceedings of the Royal Society of London B. 264:297-302. Smith, J.N.M. 1988. Determinants of lifetime reproductive success in the song sparrow. In: Reproductive success: studies of individual variation in contrasting breeding systems (ed T.H. Clutton-Brock). pp.154-172. Chicago University Press, Chicago, Illinois, USA. Smith, J.N.M. & Arcese, P. 1989. How fit are floaters? Consequences of alternative territorial behaviors in a non-migratory sparrow. American Naturalist 133:830-845. Smith, J.N.M., Keller, L.F., Marr, A.B. & Arcese, P. 2006. Biology of small populations: the song sparrows of Mandarte Island. Oxford University Press, New York. Sokal, R.R. & Rohif, F.J. 1995. Biometry. W.H. Freeman, New York. Suter, S.M., Keiser, M., Feignoux, R. and Meyer, D.R. 2007. Reed bunting females increase fitness through extra-pair mating with genetically dissimilar males. Proceedings of the Royal Society B 274:2865-2871. Thusius, K.J., Peterson, K.A., Dunn, P.O. and Wbittingham, L.A. 2001. Male mask size is correlated with mating success in the common yellowthroat. Animal Behaviour 62:435-446. Tuttle, E.M. 2003. Alternative reproductive strategies in the white-throated sparrow: behavioral and genetic evidence. Behavioral Ecology 14:425-432. Whittingham, LA. & Dunn, P.O. 2001. Survival of extra-pair and within-pair young in tree swallows. Behavioral Ecology 12:496-500. 48 3 BREEDING SYNCHRONY, DENSITY, AND EXTRA- PAIR PATERNITY2 3.1 Introduction Extra-pair paternity (EPP) occurs in most bird species surveyed to date (Griffith et al. 2002), but the consequences ofbreeding synchrony and breeding density on mate availability and a male’s ability to engage in extra-pair matings remain unclear (Griffith et a!. 2002, Kokko and Rankin 2006). Breeding synchrony, expressed as the extent of overlap in female fertile periods in a population (Kempenaers 1993), may influence EPP through its effect on the spatial and temporal availability of potential mates. When breeding synchrony is high a large proportion of males in the population may have to choose between guarding their fertile social mate and pursuing EPCs with the many fertilizable females in the population. If mate guarding is important for preventing paternity loss and if males guard their fertile social mate instead ofpursuing EPCs, then breeding synchrony is predicted to be negatively related to the level of EPP (the ‘mate guarding constraint’ hypothesis; Birkhead and Biggins 1987, Westneat et al. 1990): as breeding synchrony increases, a larger proportion of males allocate their time and energy toward mate guarding and sexual activities with their fertile social mate instead of toward pursuing EPCs. EPP may also decline with increasing breeding synchrony simply due to a decrease in the ratio of sexually-active males to fertilizable females (i.e. the operational sex ratio; Westneat et al. 1990). Alternatively, if males pursue EPCs instead of guarding their fertile social mate when breeding synchrony is high, then synchrony should be 2A version of this chapter will be submitted for publication. Ames, C.E. and Arcese, P.A. Breeding synchrony and extra-pair paternity in the song sparrow (Melospiza melodia). 49 positively related to the level of EPP (the ‘mating opportunity’ hypothesis; Stutchbury and Morton 1995): a concentration of fertile females in space and time should cause a large number of males to simultaneously compete for EPCs. Females benefit from obtaining EPCs when synchrony is high because they have more opportunities to directly compare the quality of competing males. To date, studies of the effect of breeding synchrony on EPP in passerines have reported a mix ofpositive (e.g. Stutchbury et al. 1997, Chuang et a!. 1999, Perlut et al. 2008), negative (e.g. Saino et al. 1999, Thusius et al. 2001, van Dongen and Mulder 2009), and null results (e.g. Richardson and Burke 2001, Johnsen and Lifjeld 2003, Ant et al. 2004, Westneat and Mays 2005, Stewart et al. 2006, Albrecht et al. 2007). Several behavioral studies have also provided support for the ‘mate guarding constraint’ and ‘mating opportunity’ hypotheses (e.g. Chuang-Dobbs et al. 2001, van Dongen 2008). Breeding density may similarly influence the level of EPP via its effect on the spatial arrangement of potential mates. For example, positive relationships between breeding density and EPP might be expected if high density (i.e. having more neighbors) increases encounter rates and the number of extra-pair mating opportunities for males and females, reduces the efficiency ofmale mate guarding, or increases harassment of females by extra-pair males (Birkhead and Møller 1992, Charmantier and Perret 2004, Bouwman and Komdeur 2006, Augustin et a!. 2007; but see e.g. Chuang et al. 1999, Johnsen and Lifjeld 2003, Westneat and Mays 2005, Stewart et al. 2006). EPP may also vary as a function of breeding density and synchrony, with these variables acting in opposition (Thusius et al. 2001). This might result, for example, if pairs nesting in a locally dense 50 area nevertheless display low EPP due to a high degree of local breeding synchrony and mate guarding. I studied the effect ofbreeding synchrony and density in a completely color-banded population of song sparrows (Melospiza melodia) resident on Mandarte Island, BC, Canada. Nearly all birds in the population from 1993-1996 were genotyped at ?8 microsatellite loci, and genetic paternity assignment was used to estimate that 29% of 751 offspring surviving to six days of age were sired by extra-pair males in this population (O’Connor et al. 2006). Because the timing ofbreeding and spatial arrangement of territories was known with precision (e.g., Smith et al. 2006), I was able to develop and test several predictions related to the ‘mate guarding constraint’ and ‘mating opportunity’ hypotheses above. For example, I predicted that if a large proportion of male song sparrows guard their social mate instead ofpursuing EPCs when synchrony is high, then the level of EPP within broods should be negatively related to the degree of breeding synchrony between a focal female and her neighbors (i.e. the ‘mate guarding constraint’ hypothesis). Because prior results indicate that 95% of extra-pair young were sired by males with territories within 80 m of the focal nest (O’Connor et al. 2006), I followed Chuang et al. (1999) to estimate the effects of synchrony on EPP at the level of neighboring territories. I also predicted that EPP and breeding synchrony would be negatively related at the population level. Alternatively, ifmales pursue EPCs instead of guarding their social mate when synchrony is high, then the level of EPP within broods should be positively related to the degree of synchrony between a focal female and her territorial neighbors and at the population level (i.e. the ‘mating opportunity’ hypothesis). 51 Similarly, I predicted that, to the degree that high breeding density increases the number of extra-pair mating opportunities, it should also raise the level of EPP within broods. In addition, I tested for interactive relationships between breeding synchrony and density on EPP (e.g. Thusius et al. 2001). I also tested several additional predictions related to breeding synchrony and EPP at the individual level. For example, following the ‘mate guarding constraint’ hypothesis I predicted that males should sire EPY outside of their social mate’s fertile period more often than expected by chance. Also following to the ‘mate guarding constraint’ hypothesis, I predicted that males that succeed in siring EPY during their social mate’s fertile period will be more likely to lose paternity in their own nest than males that sire EPY outside their social mate’s fertile period. 3.2 Methods 3.2.1 Field Methods Mandarte Island is about 6 ha in size and lies 25 km northeast of Victoria, British Columbia, Canada (48° 38’ N, 123° 17’ W). Its resident, semi-isolated population of song sparrows has been studied continuously since 1975 (Smith et al. 2006). All sparrows on the island are uniquely marked as nestlings or, rarely, as immigrants. From 1993-96, blood samples were taken from most adults and all offspring surviving to banding age (4-6 days post-hatch; henceforth referred to as ‘banded young’). Eggs and offspring dying prior to banding were excluded from analyses because their paternity is unknown. ‘Brood’ is defined as a nest containing at least one ‘banded young’. Survival 52 and population size were estimated annually in April, when the entire population was enumerated (Smith et al. 2006). Territory boundaries and the locations of territorial individuals were mapped in April of each year. Briefly, all birds were monitored regularly each year from March to July, when females typically initiated 2-3 nesting attempts annually. Lay date (first egg of a clutch) was determined by direct observation or back-calculating from hatch date or chick age. Females lay one egg per day, averaging 3-4 eggs per clutch (range: 1-5 eggs). The fertile period, defined as the length of time females can store viable sperm in their reproductive tract, is unknown in song sparrows. Therefore, I followed Kempenaers (1993) and defined the fertile period as the period starting 5 days before the first egg in a clutch was laid and ending on the day the penultimate egg was laid. Incubation by the female begins on the day the penultimate egg is laid and lasts 12-13 days. Young fledge 9-11 days post-hatch and are cared for by both social parents to 24-28 days of age, when they become ‘independent young’. Offspring became ‘recruits’ to the population when they were known to have survived on the island to 30 April of the following year. These data allowed me to confidently determine the age of all individuals from 1993-96. 3.2.2 Genetic Analysis and Paternity Assignment Genotyping procedures are described in detail in O’Connor et al. (2006) and outlined briefly here. From 1993-1996, blood samples were collected from the brachial vein of all 751 offspring that survived to six days post-hatch and 97% of 242 adults. Eight adults not genotyped included two females, two socially mated territorial males, one unmated territorial male, and three unmated ‘floaters’ (Arcese 1987). Eight loci were used to 53 genotype all birds: MME1, MME2, MME3, MME7, MME8, and MME12 (Jeffrey et al. 2001), ESCU1 (Hanotte et a!. 1994), and GF5 (Petren 1998). One additional locus (PSAP 335; Chan and Arcese 2002) was used in a small number of individuals to reduce uncertainty in paternity. Paternity assignment used maximum likelihood methods and program CERVUS (Marshall et al., 1998) and is described in detail in O’Connor et a!. (2006). Briefly, all males one or more years old were considered as candidate sires of all offspring. A genotyping error rate of 3% was used for all simulations based on the mismatch frequency of mothers and offspring, and was reduced in the lab by repeatedly genotyping uncertain individuals. Due to the high average relatedness of sparrows on Mandarte Island, high probabilities of paternity ( 95%) were occasionally estimated for closely related candidate sires. However, because a previous study showed that 98% of extra-pair male song sparrows resided within one territory width of their extra-pair mates (C. Hill personal communication, Hill 1999), O’Connor et a!. (2006) weighted raw paternity scores by the distance between the candidate sire and offspring’s territory centre and assigned paternity to the male with the highest distance-weighted LOD score. 3.2.3 Breeding Synchrony and Density I calculated a breeding synchrony index for each brood on Mandarte Island from 1993- 96, following Kempenaers (1993). The breeding synchrony index was calculated as the average percentage of females that were fertile on a given day of the focal female’s fertile period (Kempenaers 1993). The index ranges from 0% to 100%. An index of 0% indicates that there are no breeding females that have fertile periods that overlap the fertile period of the focal female. An index of 100% indicates that all breeding females 54 have fertile periods that overlap the focal female’s fertile period on each day. Synchrony was estimated at the population level (i.e. population synchrony) using all nests on the island, and on the local level (i.e. local synchrony) using nests on territories within 80 m of the focal territory because O’Connor et al. (2006) found that this was the distance within which >95% of extra-pair males resided. I defined local breeding density as the number of male territories within 80 m of the centre of a focal territory, including territories of mated and unmated males (range: 4 to 18 neighboring territories). 3.2.4 Statistical Analyses All analyses were performed in SAS 9.1 (SAS Institute, 2003). I used linear mixed models (PROC MIXED) with restricted maximum likelihood (REML) to analyze annual trends in population synchrony, local synchrony, and local breeding density. Pair identity (i.e. the identity of the male and female pair) was included as a random factor. The residuals of all models were normally-distributed. I used generalized linear mixed models (PROC GLIMMIX) to analyze the proportion of EPY within broods (binomial error structure and logit link), where the number of EPY within a brood was the response numerator and the total number of offspring within a brood was the response denominator. Pair identity was included as a random factor. Nonparametric tests were used for all other analyses as indicated. In order to determine if extra-pair males sire EPY outside of their social mate’s fertile period more often than expected by chance, I calculated the number of extra-pair males whose social mate’s fertile period overlapped the fertile period of the cuckolded male’s 55 social mate (i.e. the extra-pair female) and the number of extra-pair males whose social mate’s fertile period did not overlap the extra-pair female’s fertile period. Next, I compared this observed frequency distribution (n = 121) to an expected frequency distribution (n = 1023). In order to generate the expected frequency distribution I calculated the number of males within an 80 m radius of the extra-pair male (i.e. neighboring males) whose social mate’s fertile period overlapped the fertile period of the extra-pair male’s social mate, and the number of neighboring males whose social mate’s fertile period did not overlap the fertile period of the extra-pair male’s mate; cuckolded males were excluded from calculation of the expected frequency distribution. 3.3 Results 3.3.1 Overview From 1993 to 1996, 29% of 751 offspring were sired by extra-pair males and 42% of 287 broods contained at least one EPY. During the study period, individuals experienced a wide range of ecological conditions with population synchrony ranging from 0.9% to 40.8%, local synchrony from 0.0% to 70.8%, and local breeding density from 4 to 18 occupied territories (Table 3.1). From 1993-96, overall population density did not vary significantly annually, ranging from 71 to 82 males, and from 41 to 52 females. Although population and local synchrony varied with lay date (Figure 3.1), the proportion of EPY within broods did not vary with lay date (F1,47 = 0.64, nb1npairs = 287, 139, P = 0.425; I did not control for year since it was unrelated to the proportion of EPY within broods (Table 3.1)). 56 3.3.2 Paternity Loss in Relation to Synchrony and Density To estimate the effect of breeding synchrony and local breeding density on the proportion of EPY within broods, I constructed a model that included population synchrony, local synchrony, local breeding density, and the interaction of density and each of the synchrony measures, and then reduced this model by removing non-significant predictors sequentially. With interactions between density and synchrony removed (Table 3.2), the remaining model suggested that the proportion of EPY within broods was significantly negatively related to local synchrony (Figure 3.2) but not significantly related to population synchrony or local breeding density (Table 3.2). 3.3.3 Comparisons of Extra-Pair and Cuckolded Males Most extra-pair males sired EPY outside the fertile period of their social mate (57.9% of 121 males), while the remaining males (42.1 % of 121 males) had mates whose fertile period overlapped that of the extra-pair female, similar to rates expected by chance (Log- likelihood ratio test, G 2.17, df= 1, P = 0.141). On average, the fertile periods of social and extra-pair females overlapped by 22.1% (± 2.8% SE, range = 0 to 100%, n = 121) or 1.64 days (± 0.21 days SE, range = 0 to 8 days, n 121). Contrary to my prediction under the ‘mate guarding constraint’ hypothesis, paternity loss was similar for males that sired EPY outside their social mate’s fertile period (40.4% of these 57 males lost paternity) and for males whose mate’s fertile period overlapped that of the extra-pair female (37.8% of these 37 males lost paternity; Log-likelihood ratio test, G = 0.06, df= 1, n 94, P = 0.807). The proportion of paternity lost by extra-pair males was also not 57 significantly related to the percentage overlap in the fertile periods of his social and extra- pair mates (F1,9 = 0.15, flbroods, npairs = 94, 74, P = 0.701). 3.4 Discussion 3.4.1 Overview The average level of breeding synchrony in this study population (19.0 — 25.6%) was relatively low compared to 21 passerine species examined by Stutchbury and Morton (1995; mean = 32.6% , range = 8 — 73%) but similar to levels reported in several recent studies (e.g. Thusius et al. 2001, Ant et al. 2004). I found that EPP within broods was related to local but not population synchrony, similar to fmdings by Chuang Ct al. (1999) in the black-throated blue warbler (Dendroica caerulescens). Chuang et al. (1999) argued that local synchrony may be a more biologically relevant determinant of the level of EPP than population synchrony if extra-pair males obtain EPCs mainly from females on neighboring territories. This is the case on Mandarte Island, where >95% of extra-pair male song sparrows had territories within 80 m of nests in which they sired EPY (O’Connor et al. 2006). My findings emphasize the need to assess synchrony at the level of local territories, especially in species where extra-pair males are often close neighbors. 3.4.2 Local Synchrony and Extra-Pair Paternity The negative relationship I found between local synchrony and the proportion of EPY within broods suggests that male song sparrows were constrained in their ability to obtain EPCs during periods of relatively high synchrony, perhaps due to the demands of mate guarding. However, I found that males were not more likely to sire EPY outside their 58 social mate’s fertile period than expected by chance, which did not support the ‘mate guarding constraint’ hypothesis. Similarly, extra-pair males whose mate’s fertile period overlapped that of the extra-pair female were not more likely to be cuckolded than males that sired EPY outside their social mate’s fertile period. I may not have been able to detect these potential costs of EPP to extra-pair males in individual-level analyses despite finding a negative relationship between EPP and local synchrony overall because I did not know the exact timing of the EPC in relation to the fertile period of the extra-pair male’s social mate. If the overlap in fertile period between the social female and the extra-pair female was incomplete, males may have actually sired EPY during the several days when the fertile periods did not overlap, when mate guarding may not have been necessary. For example, if the fertile period of the extra-pair male’s social mate and the extra-pair female overlap by two days when the length of the extra-pair female’s fertile period is seven days, then the extra-pair male could have sired EPY during the five days when his mate was not fertile but when the extra-pair female was fertile. In this case, the fertile period of the social female and the extra-pair female would have been counted as ‘overlapping’ in analyses despite that, in reality, the social female was not fertile when the extra-pair male sired EPY. Quantitative data on the timing of EPCs and the time males spend mate guarding in song sparrows are required to investigate this further. Of the few studies reporting quantitative data on the time males spend mate guarding in relation to breeding synchrony, Chuang-Dobbs et al. (2001) showed that when synchrony was high, male black-throated blue warbiers reduced the time spent in mate guarding, presumably in an attempt to gain EPCs with fertile females on neighboring territories. 59 Van Dongen (2008) found that when synchrony was low, male golden whistlers (Pachycephala pectoralis) were more aggressive toward intruding males, presumably because the risk of cuckoldry was greater. Male golden whistlers also increased mate guarding in response to territorial intrusions when synchrony was low, but not when it was high (van Dongen 2008). Although I have not quantified mate guarding in male song sparrows on Mandarte Island, casual observations suggest that western male song sparrows are similar to those in eastern NA, who follow their mates closely and reduce dramatically time devoted to singing during their mate’s fertile period (Arcese et al. 2002, Turner and Barber 2004). These observations are consistent with the idea that mate guarding may conflict with a male’s ability to obtain EPCs. 3.4.3 Local Breeding Density and Extra-Pair Paternity I found that local breeding density was not related to EPP in this study population, similar to studies in the house sparrow (Stewart et al. 2006), red-winged blackbird (Westneat and Mays 2005), bluethroat (Luscinia s. svecica; Johnsen and Lifjeld 2003), and black throated blue warbler (Chuang et al. 1999). Furthermore, I found no interactive effects of breeding density and synchrony on EPP. However, at least three factors potentially complicate the interpretation of my results. First, males may adjust mate guarding in response to the perceived risk of cuckoldry, resulting in similar levels of EPP within broods at high and low densities. For example, in Seychelles warblers (Acrocephalus sechellensis) males increased mate guarding in response to an experimental increase in the number of neighboring males (Komdeur 2001), and increased mate guarding by males reduced the occurrence of EPP within broods (Komdeur et al. 2007). Second, 60 relationships between EPP and density may be confounded by male quality. For example, high quality males in areas of high female density may elect to invest more time in obtaining EPCs, whereas lower quality males nesting in high density areas may increase mate guarding effort to reduce the risk of cuckoldry. Strategies may also differ depending on the number and quality of neighboring males (e.g. Estep et al. 2005). A third possibility is that females have a dominant role in EPP and are highly selective of potential extra-pair mates. In such cases, females mated to poor quality or genetically incompatible males may seek EPCs from superior males to enhance offspring fitness (reviewed by Akcay and Roughgarden 2007). Female song sparrows have been observed soliciting EPCs during their fertile period (Arcese et al. 2002), showing that females sometimes evade their males during the fertile period. Definitive descriptions of the relationship between EPP and breeding density and synchrony are therefore likely to require that detailed behavioral studies take place concurrently with studies of genetic paternity in territorial birds. The results from my study indicate that behavioral studies in song sparrows are required to determine whether EPCs are primarily pursued by males or females. Data are also required on the time males spend mate guarding in relation to the level ofbreeding synchrony and density, and on the time males spend pursuing EPCs during their mate’s fertile period and at varying levels of breeding density. Further, male quality should be examined in relation to whether males pursue EPCs or mate guard at varying levels of breeding synchrony and density. These data may help clarify the results from my study. 61 Ta bl e 3. 1 M ea n a n n u a lf er til e pe rio d, po pu la tio n br ee di ng sy nc hr on y, lo ca lb re ed in g sy nc hr on y, lo ca lb re ed in g de ns ity ,a n d e x tr a- pa ir pa te rn ity fr om 19 93 to 19 96 fo r s o n g sp ar ro w s o n M an da rte Is la nd . A ll br oo ds w ith in ea ch ye ar w er e u se d to ca lc ul at e th e v ar ia bl es in th is ta bl e. Po pu la tio n sy nc hr on y v ar ie d am o n g ye ar s ( F 3,2 6 8 = 8. 46 , n b ro ,n p a = 28 7, 13 9, P < 0. 00 1) w he re as lo ca ls yn ch ro ny di d n o t( F 3,2 6 8 = 0. 27 , f lb ro od s,n pa irs = 28 7, 13 9, P 0. 84 9). Lo ca lb re ed in g de ns ity v ar ie d am o n g ye ar s ( F 3,2 1 4 = 14 .6 8, flb ro od s, fl pa ir s = 28 7, 13 9, P < 0. 00 1) . Th er e w as n o an n u al v ar ia tio n in th e pe rc en ta ge o f o ffs pr in g th at w er e EP Y (L og -li ke lth oo dr at io te st , G = 1. 20 ,d f= 3, P = 0. 75 4) o r in th e pe rc en ta ge o f n es ts th at co n ta in ed at le as t o n e EP Y (L og -li ke lih oo dr at io te st , G = 0, 02 ,d f= 3, P = 0. 99 9). Fe rti le pe rio d Lo ca lb re ed in g de ns ity (no . Ex tra -p ai r Po pu la tio n sy nc hr on y (% ) Lo ca ls yn ch ro ny (% ) (da ys ) n ei gh bo rin g te rr ito rie s) pa te rn ity br oo ds % o f % o f Y ea r (il fer til e M ea n ± SE M ea n ± SE M m . M ax . M ea n ± SE M m . M ax . M ea n ± SE M m . M ax . o ffs pr in g n es ts fe m al es ) 19 93 70 (4 8) 7. 56 ± 0. 08 21 .9 ±0 .9 % 3. 4% 39 .0 % 22 .2 ± 1. 7% 0. 0% 62 .5 % 11 .4 7± 0. 35 4 17 27 .3 % 41 .4 % 19 94 66 (4 5) 7. 03 ± 0. 10 22 .0 ± 1. 0% 2. 0% 36 .9 % 21 .5 ± 1. 6% 0. 0% 70 .8 % 11 .8 8± 0. 31 5 16 27 .0 % 42 .4 % 19 95 76 (4 7) 7. 46 ±0 .0 7 25 .6 ± 1. 2% 0. 9% 40 ,8 % 22 .3 ± 1. 7% 0. 0% 69 .0 % 12 .5 7 ± 0. 29 7 16 31 .1 % 42 .1 % 19 96 75 (47 ) 7. 29 ± 0. 08 19 .0 ± 0. 7% 4. 1% 30 .0 % 20 .5 ± 1. 5% 0. 0% 51 .4 % 13 .5 5 ± 0. 34 5 18 30 .4 % 42 .7 % LJ £9 - -£wOF9cZ1cjjrnj cgoc00OLO+0900X3ISUp)Auo.npu(snoo’j 66Oiootvo+oo-xXuospui(suoT1ndoj 66600009W0F0000- OWOt’Z6ZL60F6Z-i(uoiipuicsjooi 090009E£OIXuoiipui(suondoj WBSTq’A AJ[JOUO!lJOdOlJ IuuTssouop*siudIpo6E1qpnqTIuoospooiq L81SUMOZSIdUiUSjospoituosidaiuiiwdwoprnuiojSU11Auosaidai sumsaIt{M‘jjIrua,doxosjqiuiojosiojuosuTogpo uoTssal2aLusaidaissuwt3spowpugoqoiqoiu1onpo1uTa1Xqppnojio sqUu1ApuTwip.IoJsoTsugpi°qinpawoipu!10P0WI’uinpouirnalsJquU puiqsalJtpuII,4JuosMo.ucdsuos.IoJ spoo.iquqjjo&suppuuuoitpuAsupa1qjodqsuoupj Fi gu re 3. 1 Po pu la tio n br ee di ng sy nc hr on y (fi lle dc irc le s) a n d lo ca lb re ed in g sy nc hr on y (em pty ci rc le s) m e a su re d fo re ac h br oo d in re la tio n to la y da te fr om 19 93 to 19 96 (p an el a- d, re sp ec tiv el y) 80 - 80 - a b , — 70 - 70 - 0 0 60 - 0 0 60 - 50 - 0 0 0 0 50 - 0 0 0 0 40 - 0 U 0 0 0 3 0 0 0 0 - • • 0 0 0 30 - 0 ° . 0 • 0 0 c 0 0 • Q S • E ‘ 0 - ° ° . . 4 • 20 - • 0 • • • . o 0 . 0 0 0 0 0 0 o • . . 10 - 0 0 0 0 0 0 0 0 10 - 0 0 0 0 0 • 0 0 0 0- 0 0 0 0 0- 0 I I — _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 80 90 10 0 11 0 12 0 13 0 14 0 15 0 16 0 17 0 18 0 19 0 80 90 10 0 11 0 12 0 13 0 14 0 15 0 16 0 17 0 18 0 19 0 D at e o ff irs t e gg D at e o f f irs t e gg 80 80 d C 70 0 70 0 0 60 0 60 50 50 0 C 0 - 40 40 80 0 0 0 0 0 0 0 0 3 0 30 0 0 0 o : 20 0 • s :‘ 20 • I 0 • 0 0 0 0 10 0 • • • c c ? ° • 10 0 0 0 0 0 0 . U • • 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 I I I I I I I I I I I I I 80 90 10 0 11 0 12 0 13 0 14 0 15 0 16 0 17 0 18 0 19 0 80 90 10 0 11 0 12 0 13 0 14 0 15 0 16 0 17 0 18 0 19 0 D at e o ff irs te gg D at e o ff irs t e gg 99 (%)iuoiqoutCs oc<oc-o<ot’-o<o-oz<o-oi<oi-o<0 -1•0 6t7tT I74., TCl) ct178LI-1< - co purnqiojpusaidan(spooiqjo.iqwnu)szisjduusputsSUN puus.Iupu1jAJuoso.indsuosiojspooiqmqJ AJ3JOUO!J.IOdO.TdpuuAuoitpusiuoiuMpqdqsuoipqa>jz 3.5 References Akçay, E. & Roughgarden, J. 2007. Extra-pair paternity in birds: review of the genetic benefits. Evolutionary Ecology Research 9:855-868. Albrecht, T., Schnitzer, J., Kreisinger, J., Exnerová, A., Bryja, J. & Munclinger, P. 2007. Extrapair paternity and the opportunity for sexual selection in long-distant migratory passerines. Behavioral Ecology 18:477-486. Arcese, P. 1987. Age, intrusion pressure and defence against floaters by territorial male song sparrows. Animal Behaviour 35:773-784. Arcese, P. 1989a. Territory acquisition and loss in male song sparrows. Animal Behaviour 37:45-55. Arcese, P. 1989b. Intrasexual competition and the mating system ofprimarily monogamous birds: the case of the song sparrow. Animal Behaviour 38:96-111. Arcese, P., Sogge, M.K., Marr, A.B. and Patten, M.A. 2002. Song sparrow (Melospiza melodia). The Birds of North America Online (ed A. Poole). Cornell Lab of Ornithology, Ithaca, NY. Ant, D., Hansson, B., Bensch, S., von Schantz, T. & Hasseiquist, D. 2004. Breeding synchrony does not affect extra-pair paternity in great reed warblers. Behaviour 141:863-880. Augustin, J., Blomqvist, D., Szép, T., Szabó, Z.D. & Wagner, R.H. 2007. No evidence of genetic benefits from extra-pair fertilisations in female sand martins (Riparia riparia). Journal of Ornithology 148:189-198. Birkhead, T.R. & Biggins, J.D. 1987. Reproductive synchrony and extrapair copulation in birds. Ethology 74:320-334. Birkhead, T.R. & Møller, A.P. 1992. Sperm Competition in Birds: evolutionary causes and consequences. Academic Press, London. Bouwman, K.M. & Komdeur, J. 2006. Weather conditions affect levels of extra-pair paternity in the reed bunting Emberiza schoeniclus. Journal of Avian Biology 37:238-244. 67 Chan, Y. & Arcese, P. 2002. Subspecific differentiation and conservation of song sparrows (Melospiza melodia) in the San Fransisco Bay region inferred by microsatellite loci analysis. The Auk 119:641-651. Charmantier, A. & Perret, P. 2004. Manipulation of nest-box density affects extra-pair paternity in a population of blue tits (Parus caeruleus). Behavioral Ecology and Sociobiology 56:360-365. Chuang, H.C., Webster, M.S. & Holmes, R.T. 1999. Extra-pair paternity and local synchrony in the black-throated blue warbler. Auk 116:726-736. Chuang-Dobbs, H.C., Webster, M.S. & Holmes, R.T. 2001. The effectiveness of mate guarding by male black-throated blue warblers. Behavioral Ecology 12:541-546. Estep, L.K., Mays, H., Keyser, A.J., Ballentine, B. & Hill, G.E. 2005. Effects ofbreeding density and plumage coloration on mate guarding and cuckoldry in blue grosbeaks (Passerina caerulea). Canadian Journal of Zoology 83:1143-1148. Griffith, S.C., Owens, I.P.F. & Thuman, K.A. 2002. Extra-pair paternity in birds: a review of interspecific variation and adaptive function. Molecular Ecology 11:2195-2212. Hanotte, 0., Zanon, C., Pugh, A., Greig, C., Dixon, A. & Burke, T. 1994. Isolation and characterization of microsatellite loci in a passerine bird: the reed bunting Emberiza schoeniclus. Molecular Ecology 3:529-530. Hill, C.E. 1999. Song and extra-pair mate choice in song sparrows. Ph.D. Thesis, Dept of Psychology, University of Washington, Seattle. Jeffrey, K.J., Keller, L.F., Arcese, P. & Bruford, M.W. 2001. The development of microsatellite loci in the song sparrow, Melospiza melodia (Ayes), and genotyping error associated with good quality DNA. Molecular Ecology Notes 1:11—13. Johnsen, A. & Liljeld, J.T. 2003. Ecological constraints on extra-pair paternity in the blue throat. Oecologia 136:476-483. Kempenaers, B. 1993. The use ofbreeding synchrony index. Ornis Scandinavica 24:84. Kokko, H. & Rankin, D.J. 2006. Lonely hearts or sex in the city? Density-dependent effects in mating systems. Philosophical Transactions of the Royal Society B 361:319-334. Komdeur, J. 2001. Mate guarding in the Seychelles warbler is energetically costly and adjusted to paternity risk. Proceedings of the Royal Society of London B 268:2103-2111. 68 Komdeur, J., Burke, T. & Richardson, D.S. 2007. Explicit experimental evidence for the effectiveness of proximity as mate-guarding behaviour in reducing extra-pair fertilization in the Seychelles warbler. Molecular Ecology 16:3679-3688. Marshall, T.C., Slate, J., Kruuk, L. & Pemberton, J.M. 1998. Statistical confidence for likelihood-based paternity inference in natural populations. Molecular Ecology 7:639-655. O’Connor, K.D., Marr, A.B., Arcese, P., Keller, L.F., Jeffrey, K.J., & Bruford, M.W. 2006. Extra-pair fertilization and effective population size in the song sparrow Melospiza melodia. The Journal of Avian Biology 37:572-578. Perlut, N.G., Freeman-Gallant, C.R., Strong, A.M., Donovan, T.M., Kilpatrick, C.W. & Zalik, N.J. 2008. Agricultural management affects evolutionary processes in a migratory songbird. Molecular Ecology 17:1248-1255. Petren, K. 1998. Microsatellite primers from Geospiza fortis and cross-species amplification in Darwin’s finches. Molecular Ecology 7:1771-1788. Richardson, D.S. & Burke, T. 2001. Extrapair paternity and variance in reproductive success related to breeding density in Bullock’s orioles. Animal Behaviour 62:519-525. Saino, N., Primmer, C.R., Ellegren, H. & Møller, A.P. 1999. Breeding synchrony and paternity in the barn swallow (Hirundo rustica). Behavioral Ecology and Sociobiology 45:2 1 1-218. SAS Institute Inc. 2003. SAS version 9.1. Cary, NC, USA. Smith, J.N.M. & Arcese, P. 1989. How fit are floaters? Consequences of alternative territorial behaviors in a non-migratory sparrow. American Naturalist 133:830-845. Smith, J.N.M., Keller, L.F., Marr, A.B. & Arcese, P. 2006. Biology of small populations: the song sparrows of Mandarte Island. Oxford University Press, New York. Stewart, I.R.K., Hanschu, R.D., Burke, T. & Westneat, D.F. 2006. Tests of ecological, phenotypic, and genetic correlates of extra-pair paternity in the house sparrow. Condor 108:399-413. Stutchbury, B.J. & Morton, E.S. 1995. The effect of breeding synchrony on extra-pair mating systems in songbirds. Behaviour 132:675-690. 69 OL ‘L9I-ccJ:I7TiCojoojP-I°NspTq)pTqpuTM-pJU!Aiu.indiind-nix utoun{Juisiopnjjnioduipunpnndsjossojcooz‘YH‘Ni’u 6001ot:uquopqsqqnjojoq3Tu.IJjoJnu.inof(s11vopad vvqda9(ElpvJ)SJJTSIqMupjoJossonsuu.nmjind-nixpun‘Auonpu(s uipoaiqjnwj‘uoinuownuioJdt1jnN6OOV‘pJn2’UJ7A‘UUOUUA ct’c-Lc:c6 u1JnqosussiMJmn(s’.z 1vJopadv qda3tüpvJ)sJopsiqM.upoUtuOitpUiCS uipatq4JMCrnAuoissinjnuop.upunuipJnnOJA[soozur‘uuocEunA 9817-Egt’:cEAoIotflU1flAJOI1Tfl0fT!1-’J1”-’J iouopvJpoautvz.zdsoapijsonndsuosmooiv‘1q1n‘.aj‘tunj 69-E9:TooojI°”4H uouiuiooqiU’kisuppunAuo.upu 1csuipaqXqpoungui si(iuuoindi,ndniixtOOiV1‘mnqun!u.vI‘uosiJ‘ud‘uunu‘f)J‘snisnqj 9Z1-6t1:017AoJoiqotDogpunXooojInJOTAnIpH P°P°°HUISSOOflSUOiinZqI1JJiind-niixjosinIauo3L66T3 9‘II’1‘Wf‘1PI“V8‘J°1‘TU‘J-’°P°M‘HM‘1d!J‘1’\ffH‘(‘nqqoin-ig 4 GENERAL DISCUSSION The adaptive significance of EPP is unclear despite a very large volume of research on the topic (reviews in Griffith et al. 2002, Westneat and Stewart 2003, Akcay and Roughgarden 2007). There is a need for long-term studies that test hypotheses related to EPP using large sample sizes, genetic paternity data from most individuals in the population, and detailed data on individual traits. My thesis addresses this need by examining the potential adaptive significance of EPP in an island population of song sparrows for which the paternity of nearly all offspring and the identity of nearly all extra-pair sires in the population were known from 1993-96. Further, this population has been studied in detail continuously since 1975 (Smith et al. 2006), thereby allowing me to examine a variety of individual life history traits and demographic variables in relation to EPP. In Chapter 2, I tested the good genes hypothesis to determine if females obtained fitness benefits from extra-pair males. In Chapter 3, I investigated the influence of two ecological factors on the level of EPP, namely breeding synchrony and breeding density. EPP occurs when a female mates with an extra-pair male and produces EPY, for which her social mate often provides parental care. The loss of paternity from the male’s own nest and the large amount of care he potentially invests in offspring that are not his own has the potential to create significant sexual conflict between male and female mates. For example, males may employ several retaliatory tactics when they detect that their mate has obtained EPCs, including withholding of parental care or engaging in physical punishment. Given the potential costs of these male tactics, it has been suggested that females must also benefit from obtaining EPCs in order for this behavior to have evolved. 71 For example, in several species females have been observed to pursue EPCs during extraterritorial forays during their fertile periods, indicating that EPP may not simply be the result of coercion by extra-pair males. However, there is little consensus in the literature on whether or not females benefit from EPP. Hypotheses for how females may benefit from EPP are broadly divided into direct benefits which increase female reproductive success in a current season (e.g. fertility insurance, access to breeding resources on the extra-pair male’s territory) and indirect benefits which improve the fitness of the female’s offspring (e.g. good genes, genetic compatibility) (see Chapter 1). The costs and benefits associated with EPP may also be altered by ecological factors (e.g. breeding synchrony, breeding density) which may change the availability of potential extra-pair mates in space and time. Given the prevalence of EPP in avian mating systems (reviewed in Griffith et al. 2002), understanding the adaptive significance of EPP is necessary to understanding the evolution ofmating systems overall. In my thesis I used a population of socially monogamous song sparrows resident on Mandarte Island to test hypotheses of the adaptive significance of EPP. O’Connor et al. (2006) previously found that 29% of 751 offspring in this population were sired by extra pair males from 1993-96. I first tested the good genes hypothesis which predicts that females mated to males of low fitness should mate with extra-pair males of higher fitness in order to improve the fitness of extra-pair offspring compared to their within-pair maternal half-siblings. A potential flaw ofmany studies that test indirect benefits 72 hypotheses is that they use small sample sizes and do not compare the fitness of EPY to that of their within-pair maternal half-sibs. Further, many studies use traits that are not clearly linked to fitness such as body size and condition and plumage ornamentation. In my thesis I directly compared the fitness of EPY to their within-pair haif-sibs, using traits based on the long-term data that were closely linked to fitness. These traits included life span, the number and proportion of successful social nest attempts produced in a lifetime, survival to independence and recruitment, and the number of independent and recruited genetic offspring (EPY and within-pair young [WPY]) produced as yearlings. However, I was unable to detect any differences between EPY and their maternal haif-sibs, suggesting that female song sparrows in this population do not mate with extra-pair males to obtain ‘good genes’. Further, I found no difference in fitness between extra-pair males and the males they cuckolded. These results are consistent with several other studies with relatively large sample sizes (e.g. Whittingham and Dunn 2001, Schmoll et al. 2003, Bouwman et al. 2007, Schmoll et al. 2009). As well, Akçay and Roughgarden (2007) reviewed the literature on EPP and concluded that evidence is equivocal on whether or not females engage in EPCs to obtain ‘good genes’. Although I used a relatively large data set (n 751 offspring) and tested traits closely linked to fitness, I did not demonstrate a fitness benefit of EPP to females. It is possible that females do not obtain fitness benefits from EPP, but engage in EPCs to make the ‘best of a bad job’ when there is strong selection in males to achieve EPCs (sexually antagonistic coevolution; Arnqvist and Kirkpatrick 2005). However, it is also possible that the traits I used did not accurately capture fitness. For example, Hunt et al. (2004) argue that the number of grand-offspring produced is a better measure of fitness than lifetime offspring production. 73 I did not measure the number of grand-offspring produced for male offspring because genetic data were available from 1993-96 whereas many offspring continued to reproduce after 1996. Future studies should attempt to estimate the number of grand- offspring produced, where possible, to more accurately assess fitness differences between EPY and their maternal half-sibs. Furthermore, it would be interesting to test the genetic compatibility hypothesis (i.e. that females reduce inbreeding by engaging in EPCs) in this population of song sparrows, given that inbreeding has been shown to reduce fitness (Keller 1998). Because Keller and colleagues are currently obtaining precise estimates of relatedness by correcting the social pedigree for the Mandarte population using genetic material from essentially all birds hatched on the island since 1993, more detailed analyses of the relation between genetic compatibility and EPP may soon be possible. I also related male fitness to the number of EPY sired, and to the proportion of paternity lost from a male’s own nest. I did not find evidence that fitter males sired more EPY annually as none of the fitness-related traits I tested were significantly related to the number of EPY sired by males. Further, there was no repeatability in the number of EPY sired annually by males across years, thus providing further evidence against a ‘good genes’ model. Age-related effects on a male’s ability to sire EPY may have caused the lack of repeatability in the number of EPY sired annually by males across years. I also found that none of the fitness-related traits I tested were significantly related to the proportion of paternity lost by males annually. However, I did find that there was weak but significant repeatability in the proportion ofpaternity lost by males across years, suggesting that the level of paternity loss may have been an intrinsic trait of individual 74 males. To explain why females consistently cuckold individual males if not to obtain ‘good genes’, future studies might consider testing additional hypotheses related to EPP, for example, that females engage in EPCs to obtain direct benefits from extra-pair males, such as those related to defense against potential predators, or providing food to fledged young (Janssen et al. 2008). Another possibility is that the male’s social mate partly detennines the proportion of EPY within broods as the majority ofmales had the same mate within and across years (e.g. Dietrich et al. 2004). Future studies might aim to examine changes in the rate of EPP within broods across consecutive nesting attempts when pairs stay together versus when mate switching occurs. One of the most interesting results from my thesis was that male success at siring EPY was significantly related to male age: the number of EPY sired by males increased up to the age of four years, and then declined in males aged five years and older. This is in contrast to the hypothesis that older males are preferred by females for their ‘proven’ viability genes (reviewed in Brooks and Kemp 2001). While many studies have demonstrated a positive relationship between male age and extra-pair mating success (e.g. Griffith et al. 2002, Bouwman et al. 2007, Schmoll et al. 2007), my result appears to be the first to show a decline in extra-pair mating success in old age (note, however, that Schmoll et al. (2007) reports that extra-pair mating success leveled off in male coal tits three years and older). The rise and then decline in extra-pair mating success in song sparrows suggests that success is related to both experience and physical ability in male song sparrows, a suggestion that is in line with several other studies of age-related performance in this population (e.g. Smith et al. 2006). My ability to detect a decline in 75 extra-pair mating success may have been due to the detailed nature of the Mandarte data set, where the exact age of every individual in the population was known with precision because all birds were individually color-banded and tracked throughout their lives. By contrast, most other studies of EPP in birds divide males into coarse ‘young’ and ‘old’ age classes by necessity. In Chapter 3, I tested the effect ofbreeding synchrony and breeding density on the level of EPP in song sparrows. I found that the proportion of EPY within a male’s nest was negatively related to breeding synchrony among neighbors, thus providing support for the ‘mate guarding constraint’ hypothesis. However, I was unable to support this hypothesis in individual-level analyses. For example, males were not more likely to sire EPY outside the fertile period of their social mate than expected by chance. Further, extra-pair males whose mate’s fertile period overlapped that of the extra-pair female were not more likely to be cuckolded than males that sired EPY outside their social mate’s fertile period. Most studies find that EPP and population and local synchrony are unrelated (e.g. Johnsen and Lifjeld 2003, Ant et al. 2004, Westneat and Mays 2005, Stewart et al. 2006, Albrecht et al. 2007), however, there are several studies that have found that EPP and synchrony are negatively related (e.g. Thusius et al. 2001, Van Dongen and Mulder 2009). To test this hypothesis further, I recommend that future studies obtain quantitative data on mate guarding in song sparrows, including the relationship between the amount of time a male spends mate guarding and the level ofbreeding synchrony on adjacent territories. If mate guarding does limit a male’s ability to engage in EPCs, then when synchrony is low I would expect to observe males guarding their mates more intensely 76 during the fertile period and neighboring males performing extra-territorial intrusions at a higher frequency. To date, few studies have obtained quantitative data on mate guarding in relation to synchrony and EPP (e.g. Chuang-Dobbs et al. 2001, van Dongen 2008). I also found that breeding density and the level of EPP were unrelated at the local and population level in each of four years, similar to other studies (e.g. Johnsen and Lifjeld 2003, Westneat and Mays 2005, Stewart et al. 2006). However, I may not have detected a relationship between density and EPP if, for example, males adjust mate guarding in response to the perceived risk of cuckoldry, resulting in similar levels of EPP within broods at high and low densities. In order to test this hypothesis, quantitative data on mate guarding would be required to determine if males nesting in high density areas mate guard more intensely than males nesting in low density areas. Another possibility is that females are highly selective in their choice of extra-pair mate and do not necessarily engage in EPCs even when nesting on a territory surrounded by a high density of neighboring males. Ideally, testing this hypothesis would involve radio-tracking females to determine if they pursue EPCs during extraterritorial forays, determining whether females gain fitness benefits from EPCs, and identifiing the traits of males that females engage in EPCs with. In conclusion, I have examined hypotheses related to the adaptive significance of EPP in song sparrows: that females mate with extra-pair males to improve the fitness of EPY relative to within-pair maternal half-sibs, and that breeding synchrony and density influence the frequency of EPP. In order to further this particular field of study, 77 8L sMoJ.ndsuosj1UJUOUIIJJUTUOTPWAU iCijd(iuiAuoJq3u/suipaTqpoojjooijrn&TTprnnopujuiuoSTUJIpU1 puiisnssnsaiNJOsoipnjsarn4njojuisiCsoupTAo1d JJTMpU‘suoipJusqoqnsuqituJOJanuJs(spsoppuizis spujsjPU1N‘ppgUT1OIJOD011{nogJ!pU04J0O.I1?P?p1oTAPqqA!P4!U1flb qnoqjypuiiid-.iixpui‘ouwjinoosiq‘Juijouoiaiut jopodsipJoiArnqiSAUTsTpns1tqpuouuuooal(ooz)1uM1spu3US\ 4.1 References Akcay, E. & Roughgarden, J. 2007. Extra-pair paternity in birds: review of the genetic benefits. Evolutionary Ecology Research 9:855-868. Arlt, D., Hansson, B., Bensch, S., von Schantz, T. & Hasseiquist, D. 2004. Breeding synchrony does not affect extra-pair paternity in great reed warblers. Behaviour 141:863-880. Arnqvist, G. & Kirkpatrick, M. 2005. The evolution of infidelity in socially monogamous passerines: the strength of direct and indirect selection on extrapair copulation behaviour in females. American Naturalist 165:S26-S37. Bouwman, K.M, Van Dijk, R.E., Wijmenga, J.J. & Komdeur, J. 2007. Older male reed buntings are more successful at gaining extra-pair fertilizations. Animal Behaviour 73:15-27. Brooks, R. & Kemp, D.J. 2001. Can older males deliver the good genes? Trends in Ecology and Evolution 16:308-313. Chuang-Dobbs, H.C., Webster, M.S. & Holmes, R.T. 2001. The effectiveness ofmate guarding by male black-throated blue warblers. Behavioral Ecology 12:54 1-546. Dietrich, V., Schmoll, T., Winkel, W., Epplen, J.T. & Lubjuhn, T. 2004. Pair identity — an important factor concerning variation in extra-pair paternity in the coal tit (Parus ater). Behaviour 141 :817-835. Griffith, S.C., Owens, I.P.F. & Thuman, K.A. 2002. Extra-pair paternity in birds: a review of interspecific variation and adaptive function. Molecular Ecology 11:2195-2212. Hunt, J., Bussière, L.F., Jennions, M.D. and Brooks, R. 2004. What is genetic quality? Trends in Ecology and Evolution 19:329-333. Janssen, M.S., P. Arcese, Sloan, M.H. and K.J., Jewell. 2008. Polyandry and Sex Ratio in the Song Sparrow. Wilson Journal of Ornithology 120:395-398. Johnsen, A. & Lifjeld, J.T. 2003. Ecological constraints on extra-pair paternity in the blue throat. Oecologia 136:476-483. Keller, L.F. 1988. Inbreeding and its fitness effects in an insular population of song sparrows (Melospiza melodia). Evolution 52:240-250. 79 O’Connor, K.D., Marr, A.B., Arcese, P., Keller, L.F., Jeffrey, K.J., & Bruford, M.W. 2006. Extra-pair fertilization and effective population size in the song sparrow Melospiza melodia. The Journal of Avian Biology 37:572-578. Schmoll, T., Dietrich, V., Winkel, W., Epplen, J.T. & Lubjuhn, T. 2003. Long-term fitness consequences of female extra-pair matings in a socially monogamous passerine. Proceedings of the Royal Society Biological Sciences B 270:259-264. Schmoll, T., Mund, V., Dietrich-Bischoff, V., Winkel, W. and Lubjuhn, T. 2007. Male age predicts extrapair and total fertilization success in the socially monogamous coal tit. Behavioral Ecology 18:1073-1081. Schmoll, T., Schurr, F.M., Winkel, W., Epplen, J.T. and Lubjuhn, T. 2009. Lifespan, lifetime reproductive performance and paternity loss of within-pair and extra-pair offspring in the coal tit Periparus ater. Proceedings of the Royal Society of London B 276:337-345. Smith, J.N.M., Keller, L.F., Marr, A.B. & Arcese, P. 2006. Biology of small populations: the song sparrows of Mandarte Island. Oxford University Press, New York. Thusius, K.J., Dunn, P.O., Peterson, K.A. & Whittingham, L.A. 2001. Extrapair paternity is influenced by breeding synchrony and density in the common yellowthroat. Behavioral Ecology 12:633-639. van Dongen, W.F.D. 2008. Mate guarding and territorial aggression vary with breeding synchrony in golden whistlers (Pachycephala pectoralis). Naturwissenschaften 95:537-545. van Dongen, W.F.D. & Mulder, R.A. 2009. Multiple ornamentation, female breeding synchrony, and extra-pair mating success of golden whistlers (Pachycephala pectoralis). Journal of Ornithology. Published online: 10 February 2009. Westneat, D.F. & Mays, H.L. 2005. Tests of spatial and temporal factors influencing extra-pair paternity in red-winged blackbirds. Molecular Ecology 14:2155-2 167. Westneat, D.F. & Stewart, I.R.K. 2003. Extra-pair paternity in birds: Causes, correlates, and conflict. Annual Review of Ecology Evolution and Systematics 34:365-396. Whittingham, LA. & Dunn, P.O. 2001. Survival of extra-pair and within-pair young in tree swallows. Behavioral Ecology 12:496-500. 80

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
United States 12 0
China 7 13
Japan 3 0
Canada 1 0
City Views Downloads
Washington 8 0
Beijing 7 0
Tokyo 3 0
Mountain View 2 0
Montreal 1 0
Sunnyvale 1 0
Ashburn 1 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}

Share

Share to:

Comment

Related Items