Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Extra-pair mate choice in the song sparrow (Melospiza melodia) Ames, Caroline Elizabeth 2009-11-10

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

Item Metadata

Download

Media
24-ubc_2009_fall_ames_caroline.pdf [ 1.77MB ]
Metadata
JSON: 24-1.0068158.json
JSON-LD: 24-1.0068158-ld.json
RDF/XML (Pretty): 24-1.0068158-rdf.xml
RDF/JSON: 24-1.0068158-rdf.json
Turtle: 24-1.0068158-turtle.txt
N-Triples: 24-1.0068158-rdf-ntriples.txt
Original Record: 24-1.0068158-source.json
Full Text
24-1.0068158-fulltext.txt
Citation
24-1.0068158.ris

Full Text

EXTRA-PAIR MATE CHOICE IN THE SONG SPARROW(Melospiza melodia)byCaroline Elizabeth AmesB.Sc., Simon Fraser University, 2004A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMASTER OF SCIENCEinThe Faculty of Graduate Studies(Forestry)THE UNIVERSITY OF BRITISH COLUMBIA(Vancouver)July 2009© Caroline Elizabeth Ames, 2009ABSTRACTExtra-pair paternity (EPP) is common in birds yet its adaptive significance remainsunclear. Since the strategy of EPP is thought to carry costs, females are predicted toobtain indirect genetic benefits (e.g. ‘good genes’) or direct material benefits (e.g.fertilityinsurance) from pursuing extra-pair copulations (EPCs). Breeding synchrony may alsoinfluence the costs and benefits of EPP to males and females. I examine ‘goodgenes’benefits of EPP and the effect ofbreeding synchrony on EPP in a socially monogamouspopulation of song sparrows wherein 29% of 751 offspring were sired by extra-pairmales.The good genes hypothesis predicts that females mate with extra-pair males that havehigher expected fitness than their social mate in order to improve the fitness of extra-pairyoung (EPY) compared to within-pair maternal half-siblings. Using traits closelylinkedto lifetime reproductive success, I found no evidence that EPY were fitter than theirmaternal 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 tosire EPY than first-year males and 1.3 —2.0 times more likely to sire EPY than very oldmales. This is consistent with similar, well-established patterns of age-related variationin annual reproductive success in song sparrows, suggesting that male success in siringEPY is influenced by experience and ability, rather than quality.I found a significant negative relationship between breeding synchrony amongneighborsand the proportion of EPY within broods of focal males. This result supportsthe ‘mate11guarding constraint’ hypothesis predicting that EPP decreases as synchrony increasesbecause a larger proportion of males allocate time toward guarding their fertile socialmate, instead of toward pursuing EPCs. However, I found that paternity loss was similarfor males that sired EPY outside their social mate’s fertile period (40.4% of 57 males lostpaternity) and for males that sired EPY during their mate’s fertile period (37.8% of 37males lost paternity). This result suggests that mate guarding did not constrain males inthe pursuit of EPCs; however, the exact timing of EPCs was unknown and may haveinfluenced this result.111TABLE OFCONTENTSABSTRACT.iiTABLE OF CONTENTSivLIST OF TABLESviLIST OF FIGURESviiACKNOWLEDGEMENTSviiiCO-AUTHORSHIP STATEMENTxINTRODUCTION11.1 Extra-Pair Paternityin Passerines11.2 Costs of pursuingEPCs21.2.1 Females21.2.2 Males31.3 Benefits of pursuingEPCs31.3.1 Females31.3,2 Males71.4 Ecological Factors71.5 Study Species andPopulation91.6 Thesis Overview101.7 References122 EXTRA-PAIRMATE CHOICE: A RETROSPECTIVETEST OF THE GOODGENES HYPOTHESIS162.1 Introduction162.2 Methods202.2.1 Field Methods202.2.2 Genetic Analysisand Paternity Assignment212.2.3 Traits Relatedto Fitness222.2.4 Statistical Analyses232.3 Results252.3.1 Overview252.3.2 Within versus Extra-pairYoung262.3.3 Male Fitness andEPY Sired262.3.4 Male Fitness and PaternityLoss282.4 Discussion292.4.1 Within versus Extra-PairYoung292.4.2 Male Fitness andEPY Sired312.4.3 Male Fitness andPaternity Loss332.4.4 Conclusion352.5 References453 BREEDING SYNCHRONY, DENSITY,AND EXTRA-PAIR PATERNITY493.1 Introduction493.2 Methods523.2.1 FieldMethods523.2.2 Genetic Analysis andPaternity Assignment533.2.3 Breeding Synchronyand Density54ivA6LsouaLJoJJj7U.NOISSf13SIU‘IVUINIJDL9souaij>j09tT10WdJuj-iupcpU1A11SUQUIpO1jtoo’j£Vgçu.IoJa-.ixipuiAuoiqouiCg8cMTA1AO1t’E8cuoissnosiuLçSfUJAJOfl3puttinj-iixjosuosu1dmoD£ELcpsujpuAuoqouAgouoTwJo>jUTSSO]‘W’-’dE9cM!MAQ9cSflflS)JççsosAjuVJgIAE9pu1SJiiputuosMoJJdsuosiojspooiquafliMjjokjtsuppuCuonpuAsuipaiqjodIqsuoIp1a>J19puusj.iupuNuoSMoLrndSUOSJOJ9661OE66JUJOJJkuuidJ!d-Jxpu‘tsupuTpaJqjiooj‘Auonpu(suIpaiqoj‘Auonpu/suipiquoiindod‘poudjq’jinuuuj,s 4vpuu‘ouiuuojidAu.onpo1datwIojipu1puosios‘udsjtjjrnuopuisalJtpuuJ,%juosMondsuosjiuiiojJj1nuu1spooiqunjpAi.AdJJOuoii.iodoidqjodrqsuotp>j91ojqipui‘ourniuojiodoAipnpoJdaIWp,JJpuijuosuos‘udsjijopuisjuosMondsuosopwpop3wqAjpnuuipansAJJO.Ioqurnuoqjodiqsuoijjc°iqi117U1EJSJflpUJAJUOsMoJJ1dsuosojuipopjopn3prn.md-iuxojospiiijosuosuidmoopannjv6CprnMsI1JPUNuosqis-jjq1itd-uT4pM‘ju.Ip3wJ13q4pusMoiudsuOSAdouuuoJ13doAIpnpoJdalwiJT1pupuosispu‘uudsJIJ‘J1AIA.InSUIsua1JJIu966104£661UIOJJA&IUO4S44MPUISI1UNU0spooiqoindsuosjo14u33J3dpui(AJ)unoi(1nd-J4xJoiuaij11LP1WISIJUPJAJU0SMOJJPdSUOSUISSUT404p4uaIS4I1JIl°IqLs1’1avL4OISrI99pujsjJpuuosMoiI1dsuosojspooiqUflfl!M.AJJOuoiiiodoidpunAuo.iqouAsjnoojuqdiqsuoiinp(JJAipdsJ‘p-njund)9661O£661wojanpicnjoiuoniUTpooiqqon.iojpainsnw(spJi3Xdw)iCuonpuiCsuipoatqiojpun(s1oiiopoiitj)iCuoiqouAsuipiquoinfndodioitnjnwpunpunjsinpunuosMoJindsuosjnmpnwAq(ijnnuunpJ1sAJJJ°ioqmnuuooqdiqsuoinpjFlS1flDU101SF!ACKNOWLEDGEMENTSI 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 mysupervisory committee, Drs. Darren Irwin and Kathy Martin, for their insightful edits andcomments 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 whoseknowledge and research has inspired my own research on extra-pair paternity in songsparrows. 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 songsparrow data which have made this thesis possible. I also thank Drs. Valerie LeMay andMark Drever for providing helpful reading material and suggestions on statistics. I alsoappreciate the feedback and encouragement I received from the Arcese and Martin labs.Thank you to Alaine Camfield, Nicola Freeman, Danika Kleiber, Michelle Martin, andAmy Wilson for their friendship throughout and helpful suggestions.I would like to express my gratitude to my family and friends who supported me alongthe way and who always had encouraging words for me. Many thanks to my longtimefriend Anna Drake for the revealing discussions on statistical analyses and extra-pairpaternity in birds. I would also like to send a heartfelt thank you to my parents, Stevenand Lesley Ames, my sisters, Christina and Julie, and my aunt, Diane Taylor, for theircontinuous encouragement and support over the years.viiixrnqmnjoqsiuJo/(TiSJA!ufl041soouoiisaioijoluotulil3doU041aIoJJdiqsJ11oqogto’>iospudTqsi1oqogoprnpipuwj(iisti)JiOuno343J3S0)JUu00u1:&IIU13S0OU00gP3ImMJqpopiAoJdsmrnpunj1uosaCO-AUTHORSHIP STATEMENTDr. Peter Arcese provided the data for these studies. I designed the studies, analyzed thedata, and prepared the manuscript with the assistance of Dr. Peter Arcese.x1 INTRODUCTION1.1 Extra-Pair Paternityin PasserinesExtra-pair paternity (EPP) occursin over 70% of the sociallymonogamous bird speciessurveyed, with an average of11.1% of offspring being siredby extra-pair males and18.7% of broods containing one or moreextra-pair young (EPY) (Griffith etal. 2002).EPP occurs when a femalemates with a male other than hersocial mate (an ‘extra-pairmale’) and produces extra-pairyoung (EPY) as a consequence.The highest rates of EPPdetected occur in the cooperativelybreeding fairy wren (Maluruscyaneus), wherein 76%of all offspring are sired byextra-group males and 95% ofbroods contain at least oneEPY (Mulder et al. 1994, Griffithet al. 2002). Among sociallymonogamous species,reed buntings (Emberiza schoenclus)display the highest levelsof EPP, with 55% ofoffspring being siredby extra-pair males and 86%of broods containing one ormore EPY(Dixon et al. 1994, Griffith etal. 2002). Given the prevalenceof EPP within and amongavian species, it is clearthat describing the adaptive significanceof EPP is necessary tounderstanding the evolutionofmating systems overall.A challenge to understandingthe evolution of EPP is explainingthe high level ofinterspecific variationin EPP among related species andacross populations of the samespecies (Griffith et al. 2002,Westneat and Stewart 2003). Forexample, within theHirundininae (i.e. aerialinsectivores such as swallowsand martins), tree swallows(Tachycineta bicolor) generallyhave higher EPP (54.0% of offspring;Griffith et al.2002) than barn swallows(Hirundo rustica, 28.2% of offspring;Griffith et al. 2002). Inanother example, Griffith(2000) showed that the level ofEPP for mainland populations1of passerines was, on average,2.1 times higher than forisland populations of thesamespecies. To date, numerousstudies have attemptedto explain variation inEPP levels, butreport a wide range ofresults, leaving the causesof variation in EPP levelspoorlyunderstood overall.One approach to understandingvariation in EPP atthe level of populationsis to examinethe complex set of interactionsthat occur between thetraits of the female,her socialmate, and the extra-pair male(s),and how ecological factorsinfluence these interactions(Westneat and Stewart2003). In order to accomplishthis, Westneat and Stewart(2003)recommend that future researchersexamine how variationin the traits of individualsandin the ecological conditions experiencedby individuals can influencethe potential costsand benefits of EPP.I take this approach inmy thesis by investigatingthe potentialcauses of variationin EPP in a population ofsong sparrows (Melospiza melodia)residenton Mandarte Island, BC. Below,I briefly review the potentialcosts and benefitsofpursuing extra-pair copulations(EPCs) in birds,and how variation in severalecologicalfactors might influencethese costs. I then summarizethe results of my thesis research.1.2 Costs of pursuingEPCs1.2.1 FemalesAlthough EPP is relativelycommon in birds, the pursuitof EPCs by females ispotentially costly, includingthe risk of reduced paternalinvestment, physicalretaliationby the female’s socialmate, investing in poor qualityyoung, increased exposuretosexually transmitted diseases,and the time and energycosts of searchingfor and2assessing potential extra-pair mates (reviewedin Petrie and Kempenaers 1998).A majorcost to females of pursuing EPCs is thoughtto be the withholdingof parental care bymales, but the evidence is mixed (reviewedin Whittingham and Dunn 2001,Sheldon2002).1.2.2 MalesA major cost to males of pursuingextra-pair copulations (EPCs) isthought to be thepotential loss of paternity in their own nestsin the event that they cannot effectivelyguard their fertile social mate andpursue EPCs at the same time(e.g., Chuang-Dobbs etal. 2001). Two recent experimentalstudies show that males thatare unable to mate guardor engage in other paternity guarding tacticshave increased rates ofpaternityloss in theirown nests (e.g. Komdeur et al. 2007, Johnsenet al. 2008). Other costs to malesofpursuing EPCs include the riskof sexually transmitted diseases,sperm depletion, andpotential trade-offs betweenthe efforts invested in engaging in EPPversus parentalinvestment in their own nest (reviewedin Petrie and Kempenaers 1998).1.3 Benefits of pursuing EPCs1.3.1 FemalesBecause the pursuit of EPCs is thoughtto be costly to females, evolutionarytheorypredicts that the behaviormust also entail compensatory benefits tofavor its occurrencein populations. Behavioral observationsin several species show that femalesoften solicitEPCs during extra-territorial forays,adding to the view that femalessometimes benefitfrom this behavior (reviewed inWestneat and Stewart 2003). Currenthypotheses about3the potential fitnessbenefits that femalesmay receive from matingwith extra-pairmalescan be divided intothose arguing directversus indirect benefits,each of whichisdescribed in more detailbelow.Direct BenefitsFemales may obtain direct(material) benefitsfrom extra-pair malesthat enhance theirfecundity in a current yearby increasing fertility(the ‘fertility insurance’hypothesis;Wetton and Parkin1991, Sheldon 1994),parental care (Blomqvistet al. 2005), nestdefense (Gray199Th), or access to breedingresources on the extra-pairmale’s territory(Gray 199Th). In red-wingedblackbirds (Agelaiusphoeniceus), for example,femalesthat obtained EPCsgained additional nestdefense and foragingopportunities fromextra-pair males (Gray 1 997b)and also hatched andfledged a greater proportionof young thanfemales that did not obtainEPCs (Gray1997a). In anotherexample, female moustachedwarblers (Acrocephalusmelanopogon)with EPY in theirnest gained parentalcare fromthe extra-pair siresonce chicks hadhatched (Blomqvistet al. 2005). Severalstudies havealso tested the ‘fertilityinsurance’ hypothesisfor EPP by relatingthe hatching successofbroods to the occurrenceof EPP within broods,or to whether or notfemales obtainedEPCs (e.g. Wettonand Parkin 1991, Gray1997a, Whitekilleret al. 2000). However,asGriffith et al. (2002) pointout, these studies cannotaccount for potentialconfoundingeffects such asfemale quality. For example,the positive relationshipthat Gray (1997a)found between hatchingsuccess and EPPis not necessarily anindication that femalesreceive fertilityinsurance benefits fromengaging in EPCsifhigh quality femalesaremore likely to engagein EPCs than poor qualityfemales.4Indirect BenefitsFemales pursuing EPCs may also accrue genetic benefits through offspring fitness, suchas 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 andRoughgarden 2007). The ‘good genes’ hypothesis predicts that females that have socialmates of poor intrinsic genetic quality will mate with extra-pair males of higher intrinsicgenetic quality in order to obtain ‘good genes’ that improve the survival and/or futurereproductive 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 songrepertoires than the female’s social mate, and song repertoire size was positivelycorrelated with post-fledging survival of offspring (Hasselquist et al. 1996). A key test ofthe good genes hypothesis is a fitness comparison between EPY and their within-pairmaternal half sibs because this indicates differential paternal genetic contribution bycontrolling for maternal genetic contribution and rearing environment (Griffith et al.2002, Akçay and Roughgarden 2007). I discuss the good genes hypothesis further inChapter 2, where this hypothesis is explicitly tested.The genetic compatibility hypothesis predicts that females will obtain non-additivegenetic benefits by mating with extra-pair males whose genome is more compatible totheir 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.5inbreeding avoidance (Tregenza and Wedell 2000) or due to increased viability throughreduced intragenomic conflict (Zeh and Zeh 1997). A further prediction of the geneticcompatibility hypothesis is that EPY will be fitter than their paternal half-sibs because itis the combination of the male and female genotypes that produce the fitness advantageof 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 compatibilityhypothesis 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 lessgenetically similar to the extra-pair male than to their social mate. Higher heterozygositymay have conferred a fitness advantage to EPY as they had higher fledgling survival thantheir maternal half sibs (Suter et al. 2007). In another example, Fossøy et al. (2007) alsofound that female bluethroats (Luscinia s. svecica) increased the heterozygosity of theiroffspring by mating with genetically dissimilar extra-pair males. Further, EPY expressedhigher immunocompetence than both their maternal and paternal half-sibs. However,several studies also failed to find evidence that females obtain non-additive geneticbenefits 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-pairmales as a ‘bet-hedging’ strategy that increases diversity of offspring genomes in order toimprove the chances of offspring survival and successful reproduction in unpredictableenvironments (Yasui 2001). This hypothesis has not been tested explicitly.6Sexually antagonistic coevolutionIt has also been suggested that EPP may not actually be adaptive for females, but maynevertheless result from females making the ‘best of a bad job’ if there is strong selectionin 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 MalesThe obvious benefit to males of obtaining EPCs is that they can increase theirreproductive 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 goodgenes or more genetically compatible genes through EPP, although studies to date havenot addressed this possibility.1.4 Ecological FactorsEcological factors can be expected to affect the ability of individuals to obtain EPCs byinfluencing the availability of potential mates in space and time. One such factor isbreeding 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 thepopulation may have to choose between guarding their fertile social mate and pursuingEPCs with the many fertilizable females in the population. When mate guarding isimportant for preventing paternity loss and when the benefits ofprotecting paternityoutweigh the benefits of pursuing EPCs, breeding synchrony is predicted to be negativelyrelated to the level of EPP (the ‘mate-guarding constraint’ hypothesis; Birkhead and7Biggins 1987, Westneat et a!. 1990): as breeding synchrony increases, a larger proportionof males allocate their time and energy toward mate guarding and sexual activities withtheir fertile mate instead of toward pursuing EPCs. Alternatively, if males pursue EPCsinstead of guarding their fertile social mate when breeding synchrony is high, thensynchrony should be positively related to the level of EPP (the ‘mating opportunity’hypothesis; Stutchbury and Morton 1995): a concentration of fertile females in space andtime should cause a large number ofmales to simultaneously compete for EPCs.Females benefit from obtaining EPCs when synchrony is high because they have moreopportunities to directly compare the quality of competing males.Breeding density is another ecological factor that could influence the level of EPP byincreasing 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 mateguard or a female’s ability to obtain EPCs. For example, in the great grey shrike (Laniusexcubitor), individuals chose secretive locations, such as inside tree crowns or bushes, toengage in EPCs whereas open locations were used for within-pair copulations(Tryjanowski et al. 2007); thus, variation in habitat structure may also have the potentialto affect the level of EPP. Resource distribution may also influence the level of EPP.For example, ifresources 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 thancontrol females, perhaps because food supplementation allowed them to compensate for8any retaliatory withholding of male parental care. Other ecological factors that mayinfluence the level of EPP include weather conditions (e.g. Johnsen and Lifjeld 2003) andthe nature of the social environment (e.g. the quality of neighboring males; Estep et al.2005).1.5 Study Species and PopulationSong sparrows are socially monogamous passerines which, on Mandarte Island, exhibitgenetic promiscuity with 29% of 751 offspring being sired by extra-pair males from1993-96 (O’Connor et al. 2006). The level of EPP on Mandarte Island is similar to levelsestimated in a mainland song sparrow population near Seattle, Washington (24.0% EPP;Hill 1999). The song sparrow population on Mandarte Island provides an ideal systemfor studying the adaptive significance of EPP. First, nearly all birds alive on the islandfrom 1993-96 have been genotyped at 8 microsatellite by O’Connor et al. (2006) suchthat the paternity of most offspring and the identity of most extra-pair sires in thepopulation are known. Second, the Mandarte Island population is a relatively closedsystem with very little emigration, and all nestlings and inmuigrants to the populationhave been individually color-banded; therefore, all individuals can be identified andmonitored closely throughout their lives. Since the population has been monitoredcontinuously since 1975 (Smith et al. 2006), detailed life history data have been collectedfor 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 othertraits. In contrast, many studies of EPP only sample a subset of the study population andare unable to identif’ many extra-pair sires. As a result, sample sizes are often small and9measures of EPP have a high degree of uncertainty. Further, few studies have long-termdata of the quality collected on Mandarte Island, and lack the ability to track individualsafter fledging or compile detailed life history data for individuals. Overall, therefore, theMandarte song sparrow population offers a nearly ideal population for study.1.6 Thesis OverviewIn order to examine the adaptive significance of EPP, I test the good genes hypothesis inChapter 2 to determine if females that mate with extra-pair males increase the fitness oftheir offspring. Many studies that test the good genes hypothesis employ modest samplesizes and traits not clearly linked to individual fitness, or fail to test a key good genesprediction 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 fouryears (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 numberand proportion of successful social nest attempts produced in a lifetime, survival toindependence, survival from independence to recruitment, and the number ofindependentand recruited genetic offspring (EPY and WPY) produced at age one. I also investigatethe relationship of extra-pair mating success and paternity loss to male fitness andconduct >100 paired comparisons of extra-pair and cuckolded males. The traits I use tomeasure male fitness are closely linked to lifetime reproductive success but independentof extra-pair mating success and paternity loss; traits include life span, the number andproportion of successful social nest attempts produced in a lifetime, annual nest initiationdate, the proportion of genetic offspring (EPY and WPY) recruited annually, and male10age. Male age is used as a fitness trait to test the good genes hypothesis because oldermales are predicted to have ‘proven’ their genetic viability by living a relatively longtime (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 notindependent of extra-pair mating success and paternity loss.In Chapter 3 I examine the effect ofbreeding synchrony and breeding density on the levelof EPP within broods. As described above, breeding synchrony may be negativelyrelated to the level of EPP (the ‘mate guarding constraint’ hypothesis; Birkhead andBiggins 1987, Westneat et al. 1990) or positively related to the level of EPP (the ‘matingopportunity’ hypothesis; Stutchbury and Morton 1995). In support of the ‘mate guardingconstraint’ hypothesis, I found that breeding synchrony among neighbors wassignificantly negatively related to the proportion of EPY within broods of focal males. Itested the ‘mate guarding constraint’ hypothesis further by comparing the rate ofpaternity loss between males that sired EPY outside the fertile period of their social mateand males whose mate’s fertile period overlapped that of the extra-pair female. I alsotested 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 breedingsynchrony. In Chapter 4, I discuss the results ofmy work in the context of the largeliterature on the evolution of extra-pair paternity in birds.111.7 ReferencesAkçay, E. & Roughgarden, J. 2007. Extra-pair paternity in birds: review of the geneticbenefits. Evolutionary Ecology Research 9:855-868.Arnqvist, G. & Kirkpatrick, M. 2005. The evolution of infidelity in socially monogamouspasserines: the strength of direct and indirect selection on extrapair copulationbehaviour in females. American Naturalist 165:S26-S37.Birkhead, T.R. & Biggins, J.D. 1987. Reproductive synchrony and extrapair copulation inbirds. Ethology 74:320-334.Birkhead, T.R. & Møller, A.P. 1992. Sperm Competition in Birds: evolutionary causesand consequences. Academic Press, London.Blomqvist, D.D. 2005. High frequency of extra-pair fertilisations in the moustachedwarbler, a songbird with a variable breeding system. Behaviour 142:1133-1148.Bouwman, K.M., Burke, T. and Komdeur, J. 2006. How female reed buntings benefitfrom extrapair mating behaviour: testing hypotheses through patterns of paternityin sequential broods. Molecular Ecology 15: 2589-2600.Brooks, R. & Kemp, D.J. 2001. Can older males deliver the good genes? Trends inEcology and Evolution 16:308-313.Chuang-Dobbs, H.C., Webster, M.S. & Holmes, R.T. 2001. The effectiveness of mateguarding 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 InverselyRelated 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 breedingdensity and plumage coloration on mate guarding and cuckoldry in bluegrosbeaks (Passerina caerulea). Canadian Journal of Zoology 83:1143-1148.Fossøy, F., Johnsen, A. and Lifjeld, J.T. 2007. Multiple genetic benefits of femalepromiscuity in a socially monogamous passerine. Evolution 62:145-156.• Gray, E.M. 1997a. Do female red-winged blackbirds benefit genetically from seekingextra-pair copulations? Animal Behaviour 53:605-623.12Gray, E.M. I 997b. Female red-winged blackbirds accrue material benefits fromcopulating with extra-pair males. Animal Behaviour 53:625-639.Griffith, S.C. 2000. High fidelity on islands: a comparative study of extrapair paternity inpasserine birds. Behavioral Ecology 11:265-273.Griffith, S.C., Owens, I.P.F. & Thuman, K.A. 2002. Extra-pair paternity in birds: areview of interspecific variation and adaptive function. Molecular Ecology11:2195-2212.Hasseiquist, D., Bensch, S., von Schantz, T. 1996. Correlation between male songrepertoire, 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, Deptof Psychology, University of Washington, Seattle.Hoi-Leitner, M., Hoi, H., Romero-Pujante, M. and Valera, F. 1999. Female extra-pairbehaviour and environmental quality in the serin (Serinus serinus): a test of theconstrained female hypothesis. Proceedings of the Royal Society of London B266:1021-1026.Johnsen, A., Andersen, V., Sunding, C. and Lifjeld, J.T. 2000. Female bluethroatsenhance offspring immunocompetence through extra-pair copulations. Nature406:296-299.Johnsen, A. & Lifjeld, J.T. 2003. Ecological constraints on extra-pair paternity in the bluethroat. Oecologia 136:476-483.Johnsen, A., Pam, H., Fossøy, F., Kleven, 0., Laskemoen, T. and Lifjeld J.T. 2008. Isfemale promiscuity constrained by the presence of her social mate? Anexperiment with bluethroates Luscinia svecica. Behavioral Ecology andSociobiology 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 fromextrapair mating in the reed bunting (Emberiza schoeniclus). Behavioral Ecology16:561-565.Komdeur, J., Burke, T., and Richardson, D.S. 2007. Explicit experimental evidence forthe effectiveness of proximity as mate-guarding behaviour in reducing extra-pairfertilization in the Seychelles warbler. Molecular Ecology 16:3679-3688.13Mulder, 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 matechoice. 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 sparrowMelospiza melodia. The Journal of Avian Biology 37:572-578.Petrie, M. & Kempenaers, B. 1998. Extra-pair paternity in birds: explaining variationbetween 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 Alpinewater pipits (Anthus spinoletta). Behavioral Ecology 8:534-543.Sheldon, B.C. 1994. Male phenotype, fertility, and the pursuit of extra-pair copulationsby 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 ofthe 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-pairmating systems in songbirds. Behaviour 132:675-690.Suter, S.M., Keiser, M., Feignoux, R. and Meyer, D.R. 2007. Reed bunting femalesincrease 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 ofparentage: Invited review. Molecular Ecology 9:1013-1027.Tryjanowski, P., Antczak, M. and Hromada, M. 2007. More secluded places for extra-paircopulations 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 thehouse sparrow Passer domesticus. Proceedings of the Royal Society of London B245:227-233.14Westneat, D.F., Schwagmeyer, P.L. and Mock, D.W. 2000. Badge size and extra-pairfertilizations in the house sparrow. Condor 102:342-348.Westneat, D.F., Sherman, P.W. & Morton, W.L. 1990. The ecology and evolution ofextra-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 matechoice criteria are unreliable. Ecological Research 16:605-6 16.Zeh, J.A. & Zeh, D.W. 1997. The evolution of polyandry II. Post-copulatory defencesagainst genetic incompatibility. Proceedings of the Royal Society of London B264:69-75.152 EXTRA-PAIR MATE CHOICE: A RETROSPECTIVETEST OF THE GOOD GENES HYPOTHESIS12.1 IntroductionThe advent of molecular genetic techniques has revealed that extra-pair paternity (EPP) istaxonomically widespread and common in birds, a group previously thought mainly topractice monogamy (Griffith et al. 2002). Of the socially monogamous bird speciessurveyed to date, over 70% have some level of EPP with an average of 11.1% ofoffspring 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 EPPexamine the potential genetic benefits of extra-pair mating to females because extra-paircopulations (EPCs) are thought to be costly to females, yet there are no obvious materialbenefits that extra-pair males provide to counteract these costs. Females pursuing EPCsmay accrue genetic benefits through offspring fitness, such as those related to ‘goodgenes’ (i.e. additive genetic benefits), genetic compatibility (i.e. non-additive benefits), orgenetic diversity (reviewed in Griffith et al. 2002, Akcay and Roughgarden 2007). Forexample, 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 theiroffspring (Griffith et al. 2002, Akçay and Roughgarden 2007).1A version of this chapter will be submitted for publication. Ames, C.E. and Arcese, P.A. Extra-pair matechoice in the song sparrow (Melospiza melodia): a retrospective test of the good genes hypothesis.16Good genes models of female extra-pair mate choice, in particular, have receivedconsiderable attention in the literature and are widely debated. The good geneshypothesis has been tested in a variety of bird species, and while some studies have foundsupport 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 thesame (e.g. blue tit (Parus caeruleus): Kempenaers et al. 1997, Deihey et al. 2007) orrelated species (Tachycineta bicolor; Whittingham and Dunn 2001, Hirundo rusitca;Hirundininae; Kieven et al. 2006a). Mixed results may be due partly to small samplesize, because many tests have included less than 200 young, a recommended minimumfor estimating population-level patterns of EPP (Griffith et al. 2002). Many studies alsolack detailed life history data for individual birds, and thus test ‘good genes’ predictionsusing 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 etal. 2006a, Deihey et al. 2007), immunocompetenee (e.g. Kieven and Lifjeld 2004, Garvinet a!. 2006), or social status (e.g. Otter et a!. 1998), in lieu of more robust indicators, suchas 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-pairmating behavior.Here, I test several predictions of the good genes hypothesis in an individually-markedpopulation of song sparrows, wherein all birds have been studied in detail since 1975, andnearly all birds alive in the population from 1993-1996 were genotyped at ?817microsatellite loci (O’Connor et al. 2006, Smith et al. 2006). O’Connor et a!. (2006) usedgenetic paternity assignment to estimate that 29% of 751 offspring surviving to six daysof age were sired by extra-pair males in this population. I test the key prediction of thegood 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 EPYand within-pair young (WPY) in the same nest should indicate differences in paternalgenetic contribution because EPY and WPY share maternal genes and a common rearingenvironment (Sheldon et al. 1997; but refer to Kempenaers (2009) for a review of howegg-order effects might influence any differences in fitness between EPY and theirmaternal haif-sibs). Although some studies have found evidence that EPY perform betterthan WPY (Sheldon et al. 1997, Charmantier et al. 2004, Garvin et a!. 2006, Bouwman etal. 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 etal. 2007, Scbmoll et al. 2009, Rosivall et al. 2009), relatively few studies have tested thekey good genes prediction that EPY should be fitter than their maternal haif-sibs (Griffithet al. 2002). In this study, I use a suite of traits linked to lifetime reproductive success insong sparrows to estimate individual fitness of EPY and WPY, including life span,lifetime number and proportion of successful social nest attempts produced, survival toindependence, survival from independence to recruitment, and the number of independentand recruited genetic offspring (EPY and WPY) produced at age one (see Table 2.1 fortrait definitions and rationale). This study is one of a few studies able to test the goodgenes hypothesis using robust indicators of fitness (see also Schmoll et al. 2003, 2005,• 2009).18The good genes hypothesis also predicts that females mated to less fit males should ‘tradeup’ by pursuing extra-pair copulations (EPCs) with fitter males. I tested this predictionby directly comparing the fitness of extra-pair males to the males they cuckolded (Akcayand Roughgarden 2007). I also predicted that males that gain EPP should have higherfitness on average, and fewer EPY in their own nest, when compared to males that didnot sire EPY (Griffith et al. 2002, Akçay and Roughgarden 2007). Similarly, I expectedthat males losing paternity in their own nest would be less fit than males not losingpaternity (Griffith et al. 2002, Alcçay and Roughgarden 2007). The traits I use to measuremale fitness are closely linked to lifetime reproductive success in song sparrows butindependent of extra-pair mating success and paternity loss; traits include life span, thenumber and proportion of successful social nest attempts produced in a lifetime, annualnest initiation date, and the proportion of genetic offspring (EPY and WPY) recruitedannually (see Table 2.1 for trait defmition and rationale). I also asked if male ageinfluenced EPP because older males may have ‘proven’ their genetic viability by living arelatively long time (reviewed in Brooks and Kemp 2001).Finally, the good genes hypothesis assumes that males differ intrinsically in geneticquality. Thus, if females select males based on genetic quality, I also expected that malesuccess at siring EPY and preventing paternity loss would both be repeatable from yearto-year (cf. Lessells and Boag 1987).192.2 Methods2.2.1 Field MethodsMandarte Island is about 6 ha in size and lies 25 km northeast of Victoria, BritishColumbia, Canada (48° 38’ N, 123° 17’ W). Its resident, semi-isolated population ofsong sparrows has been studied intensively since 1975 (Smith et al. 2006). All sparrowson the island are uniquely marked as nestlings or immigrants. From 1993-1996, bloodsamples were taken from most adults and all offspring surviving to banding age (4-6 dayspost-hatch; henceforth referred to as ‘banded young’). Eggs and offspring dying prior tobanding were excluded from analyses because their paternity is unknown. ‘Brood’ isdefined as a nest containing at least one ‘banded young’. Survival and population sizewere 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, whenfemales 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 daysof age, when they become ‘independent young’. Offspring became ‘recruits’ to thepopulation when they were known to have survived and remained on the island to 30April of the following year. These data allow us to estimate seasonal and lifetimereproductive performance of all birds hatched or immigrating to the population (Reid etal. 2005, Smith et al. 2006).202.2.2 Genetic Analysis and PaternityAssignmentGenotyping procedures are described indetail in O’Connor et al. (2006) and outlinedbriefly here. From 1993-1996, blood samples werecollected from the brachial vein of all751 offspring that survived to six dayspost-hatch and 97% of 242 adults. Eightadultsnot genotyped from 1993-96 included two females, twosocially mated territorial males,one unmated territorial male, and three unmated ‘floaters’.Eight loci were used togenotype all birds: MME1, MME2, MME3,MME7, MME8, and MME12 (Jeffrey etal.2001), ESCU1 (Hanotte et al. 1994), andGF5 (Petren 1998). One additional locus(PSAP 335; Chan and Arcese 2002) was usedin a small number of individuals to reduceuncertainty in paternity. Paternity assignment wasconducted by O’Cormor et al. (2006)using maximum likelihood methods and programCERVUS (Marshall et al. 1998), and isdescribed in detail in O’Connor et al. (2006).Briefly, all males one or more years oldwere considered as candidate sires of all offspring. Agenotyping error rate of 3% wasused for all simulations based on the mismatch frequencyof mothers and offspring, andwas reduced in the lab by repeatedly genotyping uncertainindividuals. Due to the highaverage relatedness of sparrows on Mandarte Island,high probabilities of paternity (95%) were occasionally estimated for multiple closely related candidatesires. However,because a previous empirical study showed that98% of extra-pair male song sparrowsresided within one territory width of their extra-pair mates(C. Hill personalcommunication, Hill 1999), O’Connor et al. (2006) weighted rawpaternity scores by thedistance between the candidate sire and offspring’s territorycentre and assigned paternityto the male with the highest distance-weighted LOD score.212.2.3 Traits Related to FitnessThe fitness-related traits I used to compare EPY to their within-pair maternal haif-sibsincluded life span, lifetime number and proportion of successful socialnest attemptsproduced, survival to independence, survival from independence to recruitment, and thenumber of independent and recruited genetic offspring (EPY and WPY) produced at ageone (see Table 2.1 for trait definitions and rationale). Traits used to analyze thedistribution of EPP among males included life span, the number and proportion ofsuccessful social nest attempts produced in a lifetime, annual nest initiation date, theproportion of genetic offspring (EPY and WPY) recruited annually, and male age (seeTable 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 notindependent of extra-pair mating success and paternity loss.I was unable to measure realized lifetime reproductive success (i.e. the number of geneticEPY and WPY sired over a lifetime) for males because many males lived before 1993 orafter 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 andproportion of successful social nest attempts produced in a lifetime) were those that didnot require genetic data to estimate. Traits measured annually, however, were estimatedusing the number of genetic offspring (EPY and WPY) produced. To account forinterannual variation in the start of breeding, the annual nest initiation date (‘first laydate’) 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 standard22deviation 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-scoresdenote individuals that bred earlier than average. Male age (years) was treated as acategorical 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 whenrelating other fitness traits to EPP because many of these traits are also linked to age(Smith et al. 2006).2.2.4 Statistical AnalysesAll analyses were performed in SAS 9.1 (SAS Institute, 2003). I applied table-wiseBonferroni corrections to a values for each suite of traits used to test a given good genesprediction (a’ = a/n, where n is the number of traits; Sokal and Rohlf 1995) in order toreduce the type I error rate.Within versus Extra-Pair YoungI used generalized linear mixed models (PROC GLIMMIX in SAS) with binary errorstructure and logit link function to test if offspring paternity (i.e. EPY or WPY) predictedsurvival from banding to independence, and from independence to recruitment. APoisson error structure and log link were used to test if offspring paternity predictedreproductive success at age one, life span, and the number of successful nest attemptsproduced in a lifetime. A binomial error structure and logit link were used to test ifoffspring paternity predicted the proportion of successful nest attempts produced in a23lifetime. In these analysesI used offspring as the unitof analysis and included broodidentity as a random factorto control for non-independenceamong maternal half-siblings(Charmantier et al. 2004).A separate analysis was conductedfor each offspring trait. Iincluded lay date, year ofthe study, age of the offspring’smother and father,and a ‘laydate x year’ interactionterm as covariates in all initialmodels, and then removedtheterms sequentially using backwardelimination (P > 0.10).Male Fitness and RepeatabilityI used generalized linearmixed models with a poissonerror structure and log linkto testif the number of EPY siredannually was predictedby male life span, the numberandproportion of successfulnest attempts producedin a lifetime, the firstdate on whichnesting was initiated eachseason, the proportion ofoffspring recruited to the populationannually, or male age.A poisson error structure wasused to analyze the numberof EPYsired annually because thedata followed a poisson distribution.A binomial errorstructure and logit link functionwere used to test if the proportionof EPY a male hadwithin broods annually (i.e.the total number of EPY withinbroods divided by thetotalnumber of banded young withinbroods) was predictedby the male traits listed above.Male identity was includedas a random factor in theseanalyses and separate analyseswere conducted for eachtrait. I included male ageand year as covariates in allinitialmodels but removed theterms by sequential backwardelimination ifP> 0.10. Pairedttests were used to comparethe traits of extra-pair malesto those of the males theycuckolded.24I used the intraclass correlation coefficient (Lessells and Boag 1987) to estimaterepeatability in the number of EPY sired annually across years and in the proportion ofEPY a male had within broods annually across years.2.3 Results2.3.1 OverviewFrom 1993-1996, 148 male song sparrows resided on Mandarte Island and contributedone (n = 66), two (n = 37), three (n 21), or four (n = 24) years of data on the number ofEPY sired annually, totaling 299 male-years. Of the 299 male-years, 32.4% were fromunmated 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), orfour (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±0.07SE,flbroods= 287) because 29% of 751 nestlings were EPY (42% of287 broods containedat least one EPY; Table 2.2). Most EPY within a brood (83% of 121) were sired by oneextra-pair male; the remaining 17% by two.252.3.2 Within versus Extra-pairYoungContrary to a key prediction ofthe good genes hypothesis,I found no evidence that EPYwere fitter than theirmaternal, within-pair half sibs(Table 2.3). EPY did not survivebetter to independenceor recruitment than did their maternalhalf-sibs, and they alsodidnot produce more geneticoffspring (EPY and WPY) intheir first year of breeding, livelonger, or produce a largernumber or proportion of successfulsocial nest attempts overtheir lifetime. Also contraryto the good genes hypothesis,I found that in pairedcomparisons extra-pair malesdid not differ significantly fromthe males they cuckoldedin life span, the number andproportion of successful socialnest attempts producedin alifetime, annual nest initiationdate, the proportionof genetic offspring recruitedto thepopulation annually,or male age (Table 2.4).2.3.3 Male Fitness and EPYSiredWhen analyzing all males inthe population (i.e. mated andunmated males), I found thatthe number of EPY sired annuallywas positively related to malelife span (F1,44 = 8.90,flrnaie-years, flmales = 299, 148, P = 0.003) and the number of successfulsocial nest attemptsproduced in a lifetime (F1,144 = 7.22,nmajeyears,flmales= 299, 148, P = 0.008). In contrast,the number of EPY sired annually wasnot significantly related tothe proportion ofsuccessful social nest attemptsproduced in a lifetime, annualnest initiation date,or theproportion of genetic offspring (EPYand WPY) recruited to the populationannually (allP>0.08). However, the number of EPYsired annually was relatedto male age (F1,144 =9.70,flmale-years, flmales= 299, 148, P <0.00 1),with males two to four yearsold on averagebeing 3.3—4.5 times more likelyto sire EPY than first-year males,and 1.4— 1.9 times26more likely to sire EPY than males five years and older. It is possible, however, thatincluding unmated males in the above analyses confused relationships between thenumber of EPY sired, life span, number of successful attempts, and age. This is becauseunmated males, the majority of which do not sire EPY (84.5% of 97 male-years), aremainly yearlings (Smith et al. 2006). Males that are unmated in their first year alsodisplay low lifetime reproductive success (Smith 1988). I therefore also conductedparallel analyses that included only mated males.When analyzing only mated males in the population, I found that the number of EPYsired annually was not significantly related to life span, the number and proportion ofsuccessful social nest attempts produced in a lifetime, the annual nest initiation date, orthe 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 tomale age (Table 2.5, Figure 2.1): males two to four years old on average were 3.1-4.7times more likely to sire EPY than first-year males, and 1.3-2.0 times more likely to sireEPY than males five years and older. Thus, male age appears to influence the number ofEPY sired annually independent of male mating status.I also found that the number of EPY sired annually was not significantly related to thenumber or proportion of EPY a male had in his own nests annually (F1,88 2.87,nmaleyears, flmales182, 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(nmaieyears, flmales= 84, 53). However, I did find that extra-pair males, on average, gained more27paternity 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 whenall 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, F63,111 = 0.95, P 0.590).2.3.4 Male Fitness and Paternity LossThe proportion of EPY a male had within social broods annually was not significantlyrelated to life span, the number and proportion of successful social nest attemptsproduced 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). Malesexhibited low but statistically significant repeatability in the proportion of EPY withintheir 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 havebeen determined, in part, by the male’s social mate because males had the same mate in51.6% of 93 consecutive first broods, and a minority of males (4 of 96 males) switchedmates between breeding attempts within a year.282.4 Discussion2.4.1 Within versus Extra-Pair YoungMy results do not support a key prediction of the good genes hypothesis because I foundno difference in survival, life span, or seasonal and lifetime reproductive performancebetween EPY and their within-pair maternal half-siblings. These results are consistentwith 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 morecomplete and diverse set of seasonal and long-term fitness measures, as recommended byGriffith et al. (2002). It is possible that females do not benefit from EPP, and engage inEPCs, for example, in order to make the ‘best of a bad job’ ifthere is strong selection inmales to achieve EPCs at the expense of female fitness (sexually antagonisticcoevolution; Arnqvist and Kirkpatrick 2005). Nevertheless, I cannot rule out that goodgenes 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 capturemale quality, perhaps due to high environmental variance in food availability, populationdensity, weather conditions, or predator/prey dynamics (e.g., Smith 1988, Arcese 2003).Although the traits I examined were closely linked to lifetime reproductive success basedon the social mating system in song sparrows (Table 2.1), I was unable to measurerealized lifetime reproductive success (i.e. the number of genetic EPY and WPY siredover a lifetime) for males or their sons. While no study of EPP to date has been able tomeasure realized lifetime reproductive success, it would be a more accurate fitnessmeasure for determining the presence or absence of ‘good genes’ effects (Schmoll et al.292009).A second possibility that prevents me from rejecting the good genes hypothesis asapplied to mate choice in song sparrows is that the genetic benefits of extra-pair matingmay 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 andAlatalo 1999). Although I was unable to detect any significant differences in fitnessbetween 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 fitnesstraits analyzed (Table 2.3): survival from independence to recruitment, the number ofindependent and recruited genetic offspring produced at age one, and the proportion ofsuccessful social nest attempts produced in a lifetime. This suggests that there may havebeen a fitness difference between the haif-sibs that was too small to detect usingconventional criterion for determining significance. If fitness differences between EPYand their maternal half-sibs were too small to detect, my current results imply thatempirical field studies of good genes and related hypotheses may prove challenging inthe absence of very large samples and more precise indexes of individual fitness thanused here (Table 2.1). Third, there may have been sampling bias that masked fitnessdifferences between EPY and their maternal haif-sibs. For example, genetic parentagewas determined for offspring that survived to 4-6 days post-hatch when banding andblood collection were possible. However, from 1993-96, 11% of 959 eggs did not hatchand 12% of 854 hatchlings did not survive to banding; therefore, the genetic parentage ofat least 206 offspring (21% of offspring) could not be determined, such that any30differences between EPY and WPY in hatch rate or in survival from hatch to bandingwould not have been detected. If, for example, EPY survived better to banding than theirmaternal haif-sibs, I would have been unable to detect this ‘good genes’ effect. Fourth, itremains possible that the genetic benefits of female extra-pair mate choice may becontext-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 thepopulation is negatively related to hatch date; Schmoll et al. (2005) found that coal titEPY that hatched relatively late in the season had a higher probability of recruitinglocally than their maternal half-sibs while there was no difference in recruitmentprobability between EPY and WPY that hatched earlier in the season.2.4.2 Male Fitness and EPY SiredIn addition, I found no strong evidence that more fit males had higher extra-pair matingsuccess. Among mated males, the number of EPY a male sired annually was notsignificantly related to life span, the lifetime number or proportion of successful socialnest attempts, the annual nest initiation date, or the proportion of genetic offspringrecruited 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 didnot have fewer EPY in their own nests. I also found that male success at siring EPY wasnot repeatable across years, indicating that the intrinsic quality of a male did notdetermine his success at siring EPY. This suggests that extrinsic factors, perhaps relatedto territory size, quality or location, the behavior of social mates, or social interactionswith new or existing neighbors, introduce variation in a male’s investment or success in31EPP from year to year.However, male age was related to the number of EPY sired annually, such that first-yearmales and males aged five years and older (i.e. very old males) sired fewer EPY thanmales aged two to four years (i.e. middle-aged males). These observations are consistentwith similar, well-established patterns of age-related variation in annual reproductivesuccess in song sparrows, wherein first-year males experience the lowest reproductivesuccess (Smith et al. 2006), are more likely to remain as non-territorial floaters orunmated territory holders than two and three-year olds (Arcese 1987, Smith and Arcese1989), and where males five years and older experience declines in reproductive successas 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 theface of challenges by non-territorial floaters, and were also more likely than younger andolder males to engage in polygyny. Overall, these trends suggest that first-year malesoften 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 alsoKeller et al. 2008). It is possible that middle-aged males are better at creating orexploiting extra-pair mating opportunities (e.g. Kieven et al. 2006b) or at providing extrapair females with direct benefits, such as nest defense against predators or additionalforaging opportunities on their territory (e.g. Gray 1997). Detailed behavioral data onmale and female extraterritorial forays and female resistance to EPCs are required todetermine if EPCs are primarily male or female driven. Further, behavioral dataindicating any direct benefits that females may be receiving from extra-pair males should32be investigated, for example, through observation of female extraterritorial forays duringbrood 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 studiesdemonstrating 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 someevidence suggests that female birds prefer older males as extra-pair mates, perhapsbecause old age signals male viability and genetic quality (reviewed in Brooks and Kemp2001), my results do not support this hypothesis because I demonstrate a decline in maleextra-pair mating success for males aged five years and older (Figure 2.1), and extra-pairmales were not older than the males they cuckolded. To my knowledge, no other studyhas measured a decline in extra-pair mating success in very old males, although Schmollet al. (2007) demonstrated that extra-pair mating success in male coal tits increasedbetween the ages of one and three years, then leveled off after the age of three. Thisresult is potentially of interest to field ecologists because the exact age of every bird wasknown, enabling us to measure with greater precision age-related variation in EPP,whereas many studies coarsely divide males into ‘young’ and ‘old’ categories. I suggestthat future studies use caution in interpreting results in the absence of detailed data onmale age.2.4.3 Male Fitness and Paternity LossI also found that the proportion of EPY a male had within broods annually was notsignificantly related to life span, the number or proportion of successful social nest33attempts produced ina lifetime, annual nestinitiation date,or the proportion of geneticoffspring recruited tothe population annually.Further, male age wasnot significantlyrelated to the proportionof EPY a male had withinbroods, similarto several studies inpasserines (e.g. Augustinet al. 2007, Bouwmanet al. 2007, Neumanet al. 2007).However, males exhibitedlow but significant repeatabilityin the proportionofpaternitylost from their ownbroods across years, suggestingthat the level ofpaternity loss mayhave been an intrinsictrait of individual males.There may be severalreasons whyfemales consistentlycuckold individual malesif not to obtain ‘goodgenes’ benefits. Forexample, some males maybe cuckolded iftheyare unable to providetheir social matewith adequate nest defenseor breeding resources,which their socialmate might thenhave to obtain fromextra-pair males. Onthe other hand,some males may losepaternityfrom their own nest if theydo not adequatelyguard their social mateduring the fertileperiod. For example, malewhite-throatedsparrows (Zonotrichiaalbicollis) witha tanmorph spend a largerproportion of timeguarding their socialmate from EPCsand,therefore, have fewerEPY in their own nests;however, males witha white morphaggressively pursueEPCs and subsequentlyhave a higher proportionof EPY in their ownnests because they spendless time guarding theirsocial mate from malesseeking EPCs(Tuttle 2003). It isalso possible that repeatabilityin the proportionof EPY within broodsmay have been determined,in part, by the male’ssocial mate becausea majority of maleshad the same mate in consecutivefirst broods and betweenbreeding attempts withinayear. For example,a study in coal tits (Parusater) showed that pairidentity was relatedto the proportion of EPY withinbroods, suggesting thatinteractions of characteristicsofthe male and his social matemight predict the proportionof EPY within broods(Dietrich34et al. 2004). In order to determine the relative roleofmales and females in determiningthe rate of EPP in song sparrows, studies are requiredthat compare changes in the rate ofEPP across consecutive broods when mate switchingdoes and does not occur.2.4.4 ConclusionIn conclusion, I did not find support for the good genes hypothesisdespite using largesample sizes and testing a diverse set of seasonal and long-termfitness measures. Myresults suggest that females may not obtain fitnessbenefits from EPP, and it is possiblethat females engage in EPCs as a result of sexuallyantagonistic coevolution. However,data on realized lifetime reproductive success of malesand their sons are required tofurther assess potential ‘good genes’ benefits. I did findthat age predicted a male’ssuccess at siring EPY, with first-year and very old malessiring fewer EPY than middle-aged males, a trend unique to studies to date. This result suggeststhat there may be age-related variation in the physical ability or experience of male songsparrows to sire EPY,rather than females preferring to mate with males that have ‘proven’their viability.However, detailed behavioral studies are required tofurther investigate this possibility,particularly those that examine the exact age ofindividualsin relation to extra-pairmating success, rather than using general ‘young’ and‘old’ age categories. My resultsalso suggest that a broader rangeof hypotheses must be tested in the future to explainvariation in EPP within species. To date, avast majority of studies have focused on thepotential indirect benefits of EPP, but have largely ignoredthe potential influence ofdirect benefits or ecological factors such as breeding synchrony. Themixed nature of myresults, including a lack of repeatability in male extra-pairmating success, is most35consistent with the idea that variationin EPP results as a consequence of variousconstraints related to ecological orother extrinsic factors operating on the timebudgets ofindividual birds with varying abilities or opportunitiesto engage in extra-pair mating as ameans to increase lifetime reproductive success.If true, experimental studies that inducevariation among individuals in the abilityor opportunity to engage in extra-pair matingwill be required to differentiate among hypotheses.36Table2.1TraitsrelatedtofitnessinsongsparrowsonMandarteIslandFocalTraitMeasuredinDefinitionTimescaleRationaleMale(M)orOffspring (0)?LifespanM,0TotalyearsaliveonMandarteLifetimeMajordeterminantoflifetimereproductivesuccessinsongsparrows(Smithetal.2006).Mayindicategeneticviability(BrooksandKemp2001).NumbersuccessfulM,0TotalnumberofsocialnestattemptsthatLifetimeNumberofbreedingattemptsoverlifetimehighlysocialnestattemptsproducedatleastoneindependentcorrelatedwithlifetimereproductivesuccessinsongoffspringregardlessofgeneticpaternitysparrows(Smith1988).Independentoffspringcriterionfor‘success’becausemalesassumemajorityofparentalcareforoffspringbetweenfledgingandindependence(Smithetal.2006).ProportionsuccessfulM,0TotalproportionofsocialnestattemptsLifetimeIndicatesefficiencywithwhichmalesraisedoffspringsocialnestattemptsthatproducedatleastoneindependenttoindependencepernestattempt(seeabove).offspringregardlessofgeneticpaternityAnnualnestinitiationMDateonwhichanindividualoritsmateSeasonEarlybreedersbreedmorefrequentlywithinaseasondatelaiditsfirsteggoftheseason,andproducealargernumberofoffspringthatbecomestandardizedforyear(seeMethods).successfulbreeders(Smith1988,Smithetal.2006).ProportiongeneticMProportionofindependentgeneticSeasonReflectsdifferencesinover-wintersurvivalofoffspringrecruitedoffspring(EPYandWPY)recruitedonoffspringandaccountsformostvariationinlifetimeMandarteIslandstudyareareproductivesuccess(Smith1988).AgeMNumberofyearsalivesinceyearofhatchSeasonMayindicategeneticviability(BrooksandKemp2001).Survivalto0WhetherornotanindividualsurvivedtoSeasonMayindicateoffspringviability.independenceindependenceSurvivalfrom0WhetherornotanindividualsurvivedSeasonAnimportantfactorindetermininglifetimeindependencetofromindependencetorecruitmentreproductivesuccessbecausethemajorityofadultsrecruitmentsurviveforonlyonebreedingseason(Smith1988).Numberindependent0NumberofindependentandrecruitedSeasonAnimportantfactorindetermininglifetimeandrecruitedgeneticgeneticoffspring(EPYandWPY)reproductivesuccess(Smith1988).offspringproducedatproducedinthefirstpossiblebreedingageoneyear(agedone)Table 2.2 Percentage of extra-pair young (EPY) and percentage of song sparrowbroods on Mandarte Island with at least one EPY from 1993 to 1996Year Percentage of offspring Percentage of nests1993 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)38Table 2.3 Differences in survival, life span, and seasonal and lifetime reproductiveperformance between EPY song sparrows and their maternal, within-pair haif-sibson 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 representleast-square means ± SE for ‘EPY’ and ‘WPY’, regression coefficient ± SE for ‘date offirst egg’, and variances ± SE for random brood intercepts.noffspnng and nbroods indicatesample sizes of total offspring and individual broods, respectively. The table-wiseBonferroni corrected a-value for 7 tests of offspring traits is 0.007.Model Estimate ± SE df F P(n,)(A) Survival to independenceEPY 751 (287) 0.550 ± 0.174 1,463 0.46 0.496WPY 0.680 ± 0.126social father’s age- 4,4633.25 0.012year- 3,46319.29 <0.001broodiD 0.386±0.182 - -(B) Survival from independence torecruitmentEPY 471(234) -0.112±0.2051,2351.78 0.184WPY -0.415±0.140mother’s age- 4,2352.20 0.069date of first egg -0.022 ± 0.0051,23517.09 <0.001year- 3,23510.15 <0.001brood ID negligible - -(C) Number independent geneticoffspring produced at age oneEPY 126(95) 0.108±0.191 1,30 0.10 0.754WPY 0.038±0.134year - 2,30 2.99 0.066brood ID 0.483±0.175 - -39Model Estimate± SE df F P(n)(D) Number recruited geneticoffspring produced at age oneEPY 126 (95) -0.986 ± 0.282 1,300.06 0.807WPY -1.070±0.196brood ID 0.364 ± 0.256- -(E) Life spanEPY 168(134) 0.722±0.109 1,331.83 0.185WPY 0.893 ± 0.068brood ID 0.100 ±0.047 - -(F) Number successful social nestattempts in lifetimeEPY 168 (134) 0.498 ±0.144 1,33 0.75 0.392WPY 0.640±0.100broodiD 0.552±0.121- -(G) Proportion successful social nestattempts in lifetimeEPY 125(106) 0.700±0.179 1,181.39 0.254WPY 0.455±0.110broocliD0.073±0.119 - -40Table 2.4 Paired comparisons of traits of extra-pair and cuckolded male songsparrows on Mandarte IslandMeans ± SE are presented. ‘n’ is the number of paired comparisons. t- and p-values arefrom paired t test. The table-wise Bonferroni corrected a-value for 6 tests of offspringtraits is 0.008.Timescale Trait Social Male Extra-Pair Male n t PLifetime Lifespan 4.271±0.173 4.674±0.190 129 1.58 0.117Number successful social5.109±0.260 5.124±0.282 129 0.04 0.968nest attemptsProportion successful social0.586 ± 0.018 0.587 ± 0.019 123 0.15 0.885nest attemptsAnnual Annual nest initiation date -0.205 ± 0.072 -0.293 ± 0.067 110 -0.74 0.459Proportion genetic offspring0.402 ± 0.038 0.414 ± 0.033 93 0.23 0.8 16recruitedAge 2.620±0.107 2.682±0.103 129 0.45 0.65641Table2.5RelationshipofthenumberofEPYsiredannuallybymatedmalesongsparrowsonMandarteIslandtolifespan,seasonalandlifetimereproductiveperformance,andageThenumberofEPYsiredbymales(dependentvariable)hasbeenrelatedtoeachofthefitness-relatedtraits(explanatoryvariables)listedbelow.Variablesretainedinfinalmodelsareindicatedinbold.Estimates±SEareonthelogscale,andrepresentregressioncoefficient±SEforfitness-relatedtraits,andvariances±SEforrandommaleintercepts.nina1e..yeandflmalesaresamplesizesoftotalmale-yearobservationsandindividualmales,respectively.Thetable-wiseBonferronicorrectedcL-valuefor6testsofmaletraitsis0.008.*denotesvariablesignificance.Fitness-RelatedTraitLifespanNumbersuccessfulsocialnestattemptsProportionsuccessfulsocialnestattemptsAnnualnestinitiationdateProportiongeneticoffspringrecruitedStatisticsforFitness-RelatedTrait(ninai)Estimate±SEFP202(91)0.129±0,0555.550.020202(91)0.066±0.0373.180.078202(91)-0.198±0.4980.160.692Timescale Lifetime AnnualMaleageYearMaleidentityFPFPEstimate±SE4.450.002*1.990.1190.421±0.1504.590.002*2.070.1090.434±0.1515.69<0.001*2.280.0830.461±0.154201(91)-0.291±0.1263.040.0843.080.0191.990.1200.472±0.157161(78)0.059±0.2430.060.8093.340.0140.740.5340.349±0.133Age202(91)-5.72<0.001*--2.270.0850.451±0.151Table2.6RelationshipoftheproportionofEPYwithinbroodsannuallyformalesongsparrowsonMandarteIslandtomalelifespan,seasonalandlifetimereproductiveperformance,andageTheproportionofEPYwithinbroodsannuallyformales(dependentvariable)hasbeenrelatedtoeachofthefitness-relatedtraits(explanatoryvariables)listedbelow.Variablesretainedinfinalmodelsareindicatedinbold.Estimates±SEareonthelogitscale,andrepresentregressioncoefficient±SEforfocaltraits,andvariances±SEforrandommaleintercepts.rna1e-yearsandflrnalesaresamplesizesoftotalmale-yearobservationsandindividualmales,respectively.TheBonferronicorrecteda-valuefor6testsofmaletraitsis0.008.StatisticsforFitness-Relatedflmale-yearsTraitMaleageYearMaleidentityTimescaleFitness-RelatedTrait(flrnaies)Estimate±SEFPFPFPEstimate±SELifetimeLifespan182(89)-0.206±0.0796.830.0103.200.0170.240.8670.986±0.284Numbersuccessfulsocial182(89)-0.126±0.0525.960.0173.190.0170.390.7590.975±0.286nestattemptsProportionsuccessfulsocial182(89)-0.884±0.6691.750.1902.570.0430.220.8821.099±0.310nestattemptsAnnualAnnualnestinitiationdate182(89)-0.182±0.1501.470.2282.660.0380.270.8451.160±0.319Proportiongeneticoffspring154(77)0.036±0.3440.010.9172.400.0580.610.6090.889±0.303recruited Age182(89)-2.560.044--0.180.9081.122±0.3111717(siioX)VItN£0‘-nCDCTjICDc•T-1pusatdansuuisaInbs-4so(iooo>1‘16‘OZ=SaIWUSJICIWH‘ç=1’wpucpus.iuprnpjuoso.uidsuosPPW£qAii1nU11uP.I!SAJ3JO.IqwnuI{JuAq3qdqsuo!JupUt17I2.5 ReferencesAkcay, E. & Roughgarden, J. 2007. Extra-pair paternity in birds: reviewof the geneticbenefits. Evolutionary Ecology Research 9:855-868.Arcese, P. 1987. Age, intrusion pressure and defense against floatersby territorial malesong sparrows. Animal Behaviour 35:773-784.Arcese, P. 1989a. Territory acquisition and loss in male song sparrows.AnimalBehaviour 37:45-55.Arcese, P. 1989b. Intrasexual competition and the mating system of primarilymonogamous birds: the case of the song sparrow. AnimalBehaviour 38:96-111.Arcese, P. 2003. Individual quality, environment, and conservation.In: Animal Behaviorand Wildlife Conservation (eds Festa-Bianchet, M.and Apollonio, M.)pp.271-295. Island Press, Washington DC.Arnqvist, G. & Kirkpatrick, M. 2005. The evolutionof infidelity in socially monogamouspasserines: the strength of direct and indirect selectionon extrapair copulationbehaviour in females. American Naturalist 165:S26-S37.Augustin, J., Blomqvist, D., Szep, T., Szabo, Z.D.and Wagner, R.H. 2007. No evidenceof genetic benefits from extra-pair fertilisations in female sandmartins (Ripariariparia). Journal of Ornithology 148:189-198.Bouwman, K.M, Van Dijk, R.E., Wijmenga, J.J. &Komdeur, J. 2007. Older male reedbuntings are more successful at gaining extra-pair fertilizations.AnimalBehaviour 73:15-27.Brooks, R. & Kemp, D.J. 2001. Can older males deliverthe good genes? Trends inEcology and Evolution 16:308-3 13.Chan, Y. & Arcese, P. 2002. Subspeciflc differentiation and conservationof songsparrows (Melospiza melodia) in the San Fransisco Bay regioninferred bymicrosatellite loci analysis. The Auk 119:641-651.Charmantier, A., Blondel, J., Perret, P. and Lambrechts,M.M. 2004. Do extra-pairpaternities provide genetic benefits for female blue tits Paruscaeruleus. Journalof Avian Biology 35:524-532.45Deihey, K., Peters, A., Johnsen, A. and Kempenaers, B. 2007. Fertilization success andUV ornamentation in blue tits Cyanistes caeruleus: correlational and experimentalevidence. 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 andtemperature. Molecular Ecology 15:3833-3840.Gray, E.M. 1997. Female red-winged blackbirds accrue material benefits from copulatingwith extra-pair males. Animal Behaviour 53:625-639.Griffith, S.C., Owens, I.P.F. & Thuman, K.A. 2002. Extra-pair paternity in birds: areview of interspecific variation and adaptive function. Molecular Ecology11:2195-2212.Hanotte, 0., Zanon, C., Pugh, A., Greig, C., Dixon, A. & Burke, T. 1994. Isolation andcharacterization of microsatellite loci in a passerine bird: the reed buntingEmberiza schoeniclus. Molecular Ecology 3:529-530.Hill, C.E. 1999. Song and extra-pair mate choice in song sparrows. Ph.D. Thesis, Dept ofPsychology, University of Washington, Seattle.Jeffrey, K.J., Keller, L.F., Arcese, P. & Bruford, M.W. 2001. The development ofmicrosatellite loci in the song sparrow, Melospiza melodia (Ayes), andgenotyping error associated with good quality DNA. Molecular Ecology Notes1:11-13.Keller, L.F., Reid, J.M. & Arcese, P. 2008. Testing evolutionary models of senescence ina natural population: age and inbreeding effects on fitness components in songsparrows. 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 bluetit Parus caeruleus: female choice, male characteristics, and offspring quality.Behavioral Ecology 8:481-492.Kempenaers, B. 2009. Behavioural Ecology: Cuckolder eggs come first. Current Biology19:R364-R366.Kleven, 0., Jacobsen, F., Izadnegahdar, R., Robertson, R.J. and Lifjeld, J.T. 2006a. Mailtail streamer length predicts fertilization success in the North American barnswallow (Hirundo rustica erythrogaster). Behavioral Ecology and Sociobiology59:412-418.• 46Kieven, 0. and Lifjeld, J.T. 2004. Extrapair paternity and offspring immunocompetencein the reed bunting, Emberiza schoeniclus. Animal Behaviour 68:283-289.Kieven, 0., Marthinsen, G., & Lifjelt, J.T. 2006b. Male extra-territorial forays, age andpaternity 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. TheAuk 104:116-121.Marshall, T.C., Slate, J., Kruuk, L. & Pemberton, J.M. 1998. Statistical confidence forlikelihood-based paternity inference in natural populations. Molecular Ecology7:639-655.Møller, A.P. & Alatalo, R.V. 1999. Good-genes effects in sexual selection. Proceedingsof the Royal Society of London B. 266:88-91.Neuman, C.R., Safran, R.J. & Lovette, I.J. 2007. Male tail streamer length does notpredict apparent or genetic reproductive success in the North American barnswallows 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 sparrowMelospiza melodia. The Journal of Avian Biology 37:572-578.Otter, K., Ratcliffe, L., Michaud, D., & Boag, P.T. 1998. Do female black-cappedchickadees prefer high-ranking males as extra-pair mates? Behavioural Ecologyand Sociobiology 42:25-3 6.Petren, K. 1998. Microsatellite primers from Geospizafortis and cross-speciesamplification 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 paternityand 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 fitnessconsequences of female extra-pair matings in a socially monogamous passerine.Proceedings of the Royal Society Biological Sciences B 270:259-264.47Schmoll, T., Dietrich, V., Winkel,W., Epplen, J.T., Schurr, F.& Lubjuhn, T. 2005. Paternalgenetic effects on offspring fitnessare context dependent withinthe extra-pair matingsystem of a socially monogamouspasserine. Evolution 59:645-657.Schmoll, T., Munci, V., Dietrich-Bischoff,V., Winkel, W. and Lubjuhn,T. 2007. Male agepredicts extrapair and total fertilizationsuccess in the socially monogamouscoal tit.Behavioral Ecology 18:1073-1081.Schmoll, T., Schurr, F.M.,Winkel, W., Epplen, J.T. and Lubjuhn,T. 2009. Lifespan,lifetime reproductive performanceand paternity loss of within-pairand extra-pairoffspring in the coal tit Periparusater. Proceedings of the Royal Societyof London B276:337-345.Sheldon, B.C., Merila, 3., Qvamstrom,A., Gustafsson, L. & Ellegren, H. 1997.Paternalgenetic contribution of offspringcondition predicted by sizeof male secondarysexual character. Proceedings ofthe Royal Society of London B.264:297-302.Smith, J.N.M. 1988. Determinantsof lifetime reproductive successin the song sparrow. In:Reproductive success: studies of individualvariation in contrastingbreeding 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? Consequencesof alternativeterritorial behaviors in a non-migratorysparrow. American Naturalist133:830-845.Smith, J.N.M., Keller, L.F., Marr,A.B. & Arcese, P. 2006. Biologyof small populations:the song sparrows of Mandarte Island.Oxford University Press,New York.Sokal, R.R. & Rohif, F.J. 1995. Biometry.W.H. Freeman, NewYork.Suter, S.M., Keiser, M., Feignoux,R. and Meyer, D.R. 2007. Reedbunting femalesincrease fitness through extra-pairmating with genetically dissimilarmales.Proceedings of the Royal SocietyB 274:2865-2871.Thusius, K.J., Peterson,K.A., Dunn, P.O. and Wbittingham,L.A. 2001. Male mask size iscorrelated with mating success inthe common yellowthroat. AnimalBehaviour62:435-446.Tuttle, E.M. 2003. Alternativereproductive strategies inthe white-throated sparrow: behavioraland genetic evidence. BehavioralEcology 14:425-432.Whittingham, LA. & Dunn, P.O.2001. Survival of extra-pair and within-pairyoung in treeswallows. Behavioral Ecology 12:496-500.483 BREEDING SYNCHRONY, DENSITY, AND EXTRA-PAIR PATERNITY23.1 IntroductionExtra-pair paternity (EPP) occurs in most bird species surveyed to date (Griffith et al.2002), but the consequences ofbreeding synchrony and breeding density on mateavailability and a male’s ability to engage in extra-pair matings remain unclear (Griffithet a!. 2002, Kokko and Rankin 2006). Breeding synchrony, expressed as the extent ofoverlap in female fertile periods in a population (Kempenaers 1993), may influence EPPthrough its effect on the spatial and temporal availability of potential mates. Whenbreeding synchrony is high a large proportion of males in the population may have tochoose between guarding their fertile social mate and pursuing EPCs with the manyfertilizable females in the population. If mate guarding is important for preventingpaternity loss and if males guard their fertile social mate instead ofpursuing EPCs, thenbreeding synchrony is predicted to be negatively related to the level of EPP (the ‘mateguarding constraint’ hypothesis; Birkhead and Biggins 1987, Westneat et al. 1990): asbreeding synchrony increases, a larger proportion of males allocate their time and energytoward mate guarding and sexual activities with their fertile social mate instead of towardpursuing EPCs. EPP may also decline with increasing breeding synchrony simply due toa decrease in the ratio of sexually-active males to fertilizable females (i.e. the operationalsex ratio; Westneat et al. 1990). Alternatively, if males pursue EPCs instead of guardingtheir fertile social mate when breeding synchrony is high, then synchrony should be2A version of this chapter will be submitted for publication. Ames, C.E. and Arcese, P.A. Breedingsynchrony and extra-pair paternity in the song sparrow (Melospiza melodia).49positively related to the level of EPP (the ‘mating opportunity’ hypothesis; Stutchburyand Morton 1995): a concentration of fertile females in space and time should cause alarge number of males to simultaneously compete for EPCs. Females benefit fromobtaining EPCs when synchrony is high because they have more opportunities to directlycompare the quality of competing males. To date, studies of the effect of breedingsynchrony 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 etal. 2001, van Dongen and Mulder 2009), and null results (e.g. Richardson and Burke2001, 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-Dobbsetal. 2001, van Dongen 2008).Breeding density may similarly influence the level of EPP via its effect on the spatialarrangement of potential mates. For example, positive relationships between breedingdensity and EPP might be expected if high density (i.e. having more neighbors) increasesencounter rates and the number of extra-pair mating opportunities for males and females,reduces the efficiency ofmale mate guarding, or increases harassment of females byextra-pair males (Birkhead and Møller 1992, Charmantier and Perret 2004, Bouwmanand Komdeur 2006, Augustin et a!. 2007; but see e.g. Chuang et al. 1999, Johnsen andLifjeld 2003, Westneat and Mays 2005, Stewart et al. 2006). EPP may also vary as afunction 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 dense50area nevertheless display low EPP due to a high degree of local breeding synchrony andmate guarding.I studied the effect ofbreeding synchrony and density in a completely color-bandedpopulation of song sparrows (Melospiza melodia) resident on Mandarte Island, BC,Canada. Nearly all birds in the population from 1993-1996 were genotyped at ?8microsatellite loci, and genetic paternity assignment was used to estimate that 29% of 751offspring 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 ofterritories was known with precision (e.g., Smith et al. 2006), I was able to develop andtest several predictions related to the ‘mate guarding constraint’ and ‘mating opportunity’hypotheses above. For example, I predicted that if a large proportion of male songsparrows guard their social mate instead ofpursuing EPCs when synchrony is high, thenthe level of EPP within broods should be negatively related to the degree of breedingsynchrony 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 bymales with territories within 80 m of the focal nest (O’Connor et al. 2006), I followedChuang et al. (1999) to estimate the effects of synchrony on EPP at the level ofneighboring territories. I also predicted that EPP and breeding synchrony would benegatively related at the population level. Alternatively, ifmales pursue EPCs instead ofguarding their social mate when synchrony is high, then the level of EPP within broodsshould be positively related to the degree of synchrony between a focal female and herterritorial neighbors and at the population level (i.e. the ‘mating opportunity’ hypothesis).51Similarly, I predicted that, to the degree that high breeding density increases the numberof extra-pair mating opportunities, it should also raise the level of EPP within broods. Inaddition, I tested for interactive relationships between breeding synchrony and density onEPP (e.g. Thusius et al. 2001).I also tested several additional predictions related to breeding synchrony and EPP at theindividual level. For example, following the ‘mate guarding constraint’ hypothesis Ipredicted that males should sire EPY outside of their social mate’s fertile period moreoften 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’sfertile period will be more likely to lose paternity in their own nest than males that sireEPY outside their social mate’s fertile period.3.2 Methods3.2.1 Field MethodsMandarte Island is about 6 ha in size and lies 25 km northeast of Victoria, BritishColumbia, Canada (48° 38’ N, 123° 17’ W). Its resident, semi-isolated population ofsong sparrows has been studied continuously since 1975 (Smith et al. 2006). Allsparrows on the island are uniquely marked as nestlings or, rarely, as immigrants. From1993-96, blood samples were taken from most adults and all offspring surviving tobanding age (4-6 days post-hatch; henceforth referred to as ‘banded young’). Eggs andoffspring dying prior to banding were excluded from analyses because their paternity isunknown. ‘Brood’ is defined as a nest containing at least one ‘banded young’. Survival52and population size were estimated annually in April, when the entire population wasenumerated (Smith et al. 2006). Territory boundaries and the locations of territorialindividuals were mapped in April of each year. Briefly, all birds were monitoredregularly each year from March to July, when females typically initiated 2-3 nestingattempts annually. Lay date (first egg of a clutch) was determined by direct observationor back-calculating from hatch date or chick age. Females lay one egg per day, averaging3-4 eggs per clutch (range: 1-5 eggs). The fertile period, defined as the length of timefemales 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 periodstarting 5 days before the first egg in a clutch was laid and ending on the day thepenultimate egg was laid. Incubation by the female begins on the day the penultimateegg is laid and lasts 12-13 days. Young fledge 9-11 days post-hatch and are cared for byboth 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 onthe island to 30 April of the following year. These data allowed me to confidentlydetermine the age of all individuals from 1993-96.3.2.2 Genetic Analysis and Paternity AssignmentGenotyping procedures are described in detail in O’Connor et al. (2006) and outlinedbriefly here. From 1993-1996, blood samples were collected from the brachial vein of all751 offspring that survived to six days post-hatch and 97% of 242 adults. Eight adultsnot genotyped included two females, two socially mated territorial males, one unmatedterritorial male, and three unmated ‘floaters’ (Arcese 1987). Eight loci were used to53genotype 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 reduceuncertainty in paternity. Paternity assignment used maximum likelihood methods andprogram 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 alloffspring. A genotyping error rate of 3% was used for all simulations based on themismatch frequency of mothers and offspring, and was reduced in the lab by repeatedlygenotyping uncertain individuals. Due to the high average relatedness of sparrows onMandarte Island, high probabilities of paternity(95%) were occasionally estimated forclosely related candidate sires. However, because a previous study showed that 98% ofextra-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 rawpaternity scores by the distance between the candidate sire and offspring’s territory centreand assigned paternity to the male with the highest distance-weighted LOD score.3.2.3 Breeding Synchrony and DensityI calculated a breeding synchrony index for each brood on Mandarte Island from 1993-96, following Kempenaers (1993). The breeding synchrony index was calculated as theaverage percentage of females that were fertile on a given day of the focal female’s fertileperiod (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 thefertile period of the focal female. An index of 100% indicates that all breeding females54have fertile periods that overlap the focal female’s fertile period on each day. Synchronywas estimated at the population level (i.e. population synchrony) using all nests on theisland, and on the local level (i.e. local synchrony) using nests on territories within 80 mof the focal territory because O’Connor et al. (2006) found that this was the distancewithin which >95% of extra-pair males resided. I defined local breeding density as thenumber of male territories within 80 m of the centre of a focal territory, includingterritories of mated and unmated males (range: 4 to 18 neighboring territories).3.2.4 Statistical AnalysesAll analyses were performed in SAS 9.1 (SAS Institute, 2003). I used linear mixedmodels (PROC MIXED) with restricted maximum likelihood (REML) to analyze annualtrends 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. Theresiduals of all models were normally-distributed. I used generalized linear mixedmodels (PROC GLIMMIX) to analyze the proportion of EPY within broods (binomialerror structure and logit link), where the number of EPY within a brood was the responsenumerator and the total number of offspring within a brood was the responsedenominator. Pair identity was included as a random factor. Nonparametric tests wereused for all other analyses as indicated.In order to determine if extra-pair males sire EPY outside of their social mate’s fertileperiod more often than expected by chance, I calculated the number of extra-pair maleswhose social mate’s fertile period overlapped the fertile period of the cuckolded male’s55social mate (i.e. the extra-pair female) and the number of extra-pair males whose socialmate’s fertile period did not overlap the extra-pair female’s fertile period. Next, Icompared this observed frequency distribution (n = 121) to an expected frequencydistribution (n = 1023). In order to generate the expected frequency distribution Icalculated 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 theextra-pair male’s social mate, and the number of neighboring males whose social mate’sfertile period did not overlap the fertile period of the extra-pair male’s mate; cuckoldedmales were excluded from calculation of the expected frequency distribution.3.3 Results3.3.1 OverviewFrom 1993 to 1996, 29% of 751 offspring were sired by extra-pair males and 42% of 287broods contained at least one EPY. During the study period, individuals experienced awide range of ecological conditions with population synchrony ranging from 0.9% to40.8%, local synchrony from 0.0% to 70.8%, and local breeding density from 4 to 18occupied territories (Table 3.1). From 1993-96, overall population density did not varysignificantly 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 proportionof EPY within broods did not vary with lay date (F1,147 = 0.64,nb1,npairs = 287, 139, P= 0.425; I did not control for year since it was unrelated to the proportion of EPY withinbroods (Table 3.1)).563.3.2 Paternity Loss in Relation to Synchrony and DensityTo estimate the effect of breeding synchrony and local breeding density on the proportionof EPY within broods, I constructed a model that included population synchrony, localsynchrony, local breeding density, and the interaction of density and each of thesynchrony measures, and then reduced this model by removing non-significant predictorssequentially. With interactions between density and synchrony removed (Table 3.2), theremaining model suggested that the proportion of EPY within broods was significantlynegatively related to local synchrony (Figure 3.2) but not significantly related topopulation synchrony or local breeding density (Table 3.2).3.3.3 Comparisons of Extra-Pair and Cuckolded MalesMost extra-pair males sired EPY outside the fertile period of their social mate (57.9% of121 males), while the remaining males (42.1 % of 121 males) had mates whose fertileperiod 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 socialand extra-pair females overlapped by 22.1%(±2.8% SE, range = 0 to 100%, n = 121) or1.64 days(±0.21 days SE, range = 0 to 8 days, n 121). Contrary to my predictionunder the ‘mate guarding constraint’ hypothesis, paternity loss was similar for males thatsired EPY outside their social mate’s fertile period (40.4% of these 57 males lostpaternity) and for males whose mate’s fertile period overlapped that of the extra-pairfemale (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 not57significantly related to the percentage overlap in the fertile periods of his social and extra-pair mates (F1,19 = 0.15,flbroods, npairs= 94, 74, P = 0.701).3.4 Discussion3.4.1 OverviewThe average level of breeding synchrony in this study population (19.0 — 25.6%) wasrelatively 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 recentstudies (e.g. Thusius et al. 2001, Ant et al. 2004). I found that EPP within broods wasrelated 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 levelof EPP than population synchrony if extra-pair males obtain EPCs mainly from femaleson neighboring territories. This is the case on Mandarte Island, where >95% of extra-pairmale 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 levelof local territories, especially in species where extra-pair males are often close neighbors.3.4.2 Local Synchrony and Extra-Pair PaternityThe negative relationship I found between local synchrony and the proportion of EPYwithin broods suggests that male song sparrows were constrained in their ability to obtainEPCs during periods of relatively high synchrony, perhaps due to the demands of mateguarding. However, I found that males were not more likely to sire EPY outside their58social mate’s fertile period than expected by chance, which did not support the ‘mateguarding constraint’ hypothesis. Similarly, extra-pair males whose mate’s fertile periodoverlapped that of the extra-pair female were not more likely to be cuckolded than malesthat sired EPY outside their social mate’s fertile period. I may not have been able todetect these potential costs of EPP to extra-pair males in individual-level analyses despitefinding a negative relationship between EPP and local synchrony overall because I didnot know the exact timing of the EPC in relation to the fertile period of the extra-pairmale’s social mate. If the overlap in fertile period between the social female and theextra-pair female was incomplete, males may have actually sired EPY during the severaldays when the fertile periods did not overlap, when mate guarding may not have beennecessary. For example, if the fertile period of the extra-pair male’s social mate and theextra-pair female overlap by two days when the length of the extra-pair female’s fertileperiod is seven days, then the extra-pair male could have sired EPY during the five dayswhen his mate was not fertile but when the extra-pair female was fertile. In this case, thefertile 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 whenthe extra-pair male sired EPY. Quantitative data on the timing of EPCs and the timemales 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 inrelation to breeding synchrony, Chuang-Dobbs et al. (2001) showed that when synchronywas 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.59Van Dongen (2008) found that when synchrony was low, male golden whistlers(Pachycephala pectoralis) were more aggressive toward intruding males, presumablybecause the risk of cuckoldry was greater. Male golden whistlers also increased mateguarding in response to territorial intrusions when synchrony was low, but not when itwas high (van Dongen 2008). Although I have not quantified mate guarding in malesong sparrows on Mandarte Island, casual observations suggest that western male songsparrows are similar to those in eastern NA, who follow their mates closely and reducedramatically 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 mateguarding may conflict with a male’s ability to obtain EPCs.3.4.3 Local Breeding Density and Extra-Pair PaternityI found that local breeding density was not related to EPP in this study population, similarto studies in the house sparrow (Stewart et al. 2006), red-winged blackbird (Westneat andMays 2005), bluethroat (Luscinia s. svecica; Johnsen and Lifjeld 2003), and blackthroated blue warbler (Chuang et al. 1999). Furthermore, I found no interactive effects ofbreeding density and synchrony on EPP. However, at least three factors potentiallycomplicate the interpretation of my results. First, males may adjust mate guarding inresponse to the perceived risk of cuckoldry, resulting in similar levels of EPP withinbroods at high and low densities. For example, in Seychelles warblers (Acrocephalussechellensis) males increased mate guarding in response to an experimental increase inthe number of neighboring males (Komdeur 2001), and increased mate guarding bymales reduced the occurrence of EPP within broods (Komdeur et al. 2007). Second,60relationships between EPPand density may be confoundedby male quality. Forexample, high quality malesin areas of high femaledensity may electto invest more timein obtaining EPCs, whereas lowerquality males nesting in highdensity areas mayincrease mate guardingeffort to reduce the risk ofcuckoldry. Strategiesmay also differdepending on the numberand quality of neighboringmales (e.g. Estep et al. 2005).Athird possibility is that femaleshave a dominant role in EPPand are highly selective ofpotential extra-pair mates.In such cases, females matedto poor quality or geneticallyincompatible males mayseek EPCs from superior malesto enhance offspring fitness(reviewed by Akcayand Roughgarden 2007). Femalesong sparrows havebeen observedsoliciting EPCs during theirfertile period (Arceseet al. 2002), showing thatfemalessometimes evade their malesduring the fertile period.Definitive descriptions oftherelationship betweenEPP and breeding densityand synchrony are thereforelikely torequire that detailedbehavioral studies takeplace concurrently with studiesof geneticpaternity in territorial birds.The results frommy study indicate that behavioralstudies in song sparrows arerequiredto determine whether EPCs areprimarily pursued bymales or females. Dataare alsorequired on the timemales spend mate guardingin relation to the level ofbreedingsynchrony and density,and on the time malesspend pursuing EPCs duringtheir mate’sfertile period andat varying levels of breedingdensity. Further, malequality should beexamined in relation to whethermales pursue EPCs ormate guard at varying levels ofbreeding synchronyand density. These datamay help clarify the resultsfrom my study.61Table3.1Meanannualfertileperiod,populationbreedingsynchrony,localbreedingsynchrony,localbreedingdensity,andextra-pairpaternityfrom1993to1996forsongsparrowsonMandarteIsland.Allbroodswithineachyearwereusedtocalculatethevariablesinthistable.Populationsynchronyvariedamongyears(F3,268=8.46,nbro,npa=287,139,P<0.001)whereaslocalsynchronydidnot(F3,268=0.27,flbroods,npairs=287,139,P0.849).Localbreedingdensityvariedamongyears(F3,214=14.68,flbroods,flpairs=287,139,P<0.001).TherewasnoannualvariationinthepercentageofoffspringthatwereEPY(Log-likelthoodratiotest,G=1.20,df=3,P=0.754)orinthepercentageofneststhatcontainedatleastoneEPY(Log-likelihoodratiotest,G=0,02,df=3,P=0.999).FertileperiodLocalbreedingdensity(no.Extra-pairPopulationsynchrony(%)Localsynchrony(%)(days)neighboringterritories)paternitybroods%of%ofYear(ilfertileMean±SEMean±SEMm.Max.Mean±SEMm.Max.Mean±SEMm.Max.offspringnestsfemales)199370(48)7.56±0.0821.9±0.9%3.4%39.0%22.2±1.7%0.0%62.5%11.47±0.3541727.3%41.4%199466(45)7.03±0.1022.0±1.0%2.0%36.9%21.5±1.6%0.0%70.8%11.88±0.3151627.0%42.4%199576(47)7.46±0.0725.6±1.2%0.9%40,8%22.3±1.7%0.0%69.0%12.57±0.2971631.1%42.1%199675(47)7.29±0.0819.0±0.7%4.1%30.0%20.5±1.5%0.0%51.4%13.55±0.3451830.4%42.7%LJ£9--£wOF9cZ1cjjrnjcgoc00OLO+0900X3ISUp)Auo.npu(snoo’j66Oiootvo+oo-xXuospui(suoT1ndoj66600009W0F0000-OWOt’Z6ZL60F6Z-i(uoiipuicsjooi090009E£OIXuoiipui(suondojWBSTq’AAJ[JOUO!lJOdOlJIuuTssouop*siudIpo6E1qpnqTIuoospooiqL81SUMOZSIdUiUSjospoituosidaiuiiwdwoprnuiojSU11AuosaidaisumsaIt{M‘jjIrua,doxosjqiuiojosiojuosuTogpouoTssal2aLusaidaissuwt3spowpugoqoiqoiu1onpo1uTa1XqppnojiosqUu1ApuTwip.IoJsoTsugpi°qinpawoipu!10P0WI’uinpouirnalsJquUpuiqsalJtpuII,4JuosMo.ucdsuos.IoJspoo.iquqjjo&suppuuuoitpuAsupa1qjodqsuoupjFigure3.1Populationbreedingsynchrony(filledcircles)andlocalbreedingsynchrony(emptycircles)measuredforeachbroodinrelationtolaydatefrom1993to1996(panela-d,respectively)80-80-ab,—70-70-0060-0060-50-000050-000040-0U00030000-••00030-0°.0•00c00•QS•E‘0-°°..4•20-•0•••.o0.000000o•..10-0000000010-00000•0000-00000-0II—__________________________________________________________80901001101201301401501601701801908090100110120130140150160170180190DateoffirsteggDateoffirstegg8080dC70070006006050500C0-404080000000003030000o:200•s:‘20•I0•0000100•••cc?°•10000000.U••00001000000000000IIIIIIIIIIIII80901001101201301401501601701801908090100110120130140150160170180190DateoffirsteggDateoffirstegg99(%)iuoiqoutCsoc<oc-o<ot’-o<o-oz<o-oi<oi-o<0-1•06t7tTI74.,TCl)ct178LI-1<-copurnqiojpusaidan(spooiqjo.iqwnu)szisjduusputsSUNpuus.Iupu1jAJuoso.indsuosiojspooiqmqJAJ3JOUO!J.IOdO.TdpuuAuoitpusiuoiuMpqdqsuoipqa>jz3.5 ReferencesAkçay, E. & Roughgarden, J. 2007. Extra-pair paternity in birds: review of the geneticbenefits. 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-distantmigratory passerines. Behavioral Ecology 18:477-486.Arcese, P. 1987. Age, intrusion pressure and defence against floaters by territorial malesong sparrows. Animal Behaviour 35:773-784.Arcese, P. 1989a. Territory acquisition and loss in male song sparrows. AnimalBehaviour 37:45-55.Arcese, P. 1989b. Intrasexual competition and the mating system ofprimarilymonogamous birds: the case ofthe song sparrow. Animal Behaviour 38:96-111.Arcese, P., Sogge, M.K., Marr, A.B. and Patten, M.A. 2002. Song sparrow (Melospizamelodia). The Birds of North America Online (ed A. Poole). Cornell Lab ofOrnithology, Ithaca, NY.Ant, D., Hansson, B., Bensch, S., von Schantz, T. & Hasseiquist, D. 2004. Breedingsynchrony does not affect extra-pair paternity in great reed warblers. Behaviour141:863-880.Augustin, J., Blomqvist, D., Szép, T., Szabó, Z.D. & Wagner, R.H. 2007. No evidence ofgenetic benefits from extra-pair fertilisations in female sand martins (Ripariariparia). Journal of Ornithology 148:189-198.Birkhead, T.R. & Biggins, J.D. 1987. Reproductive synchrony and extrapair copulationinbirds. Ethology 74:320-334.Birkhead, T.R. & Møller, A.P. 1992. Sperm Competition in Birds: evolutionary causesand consequences. Academic Press, London.Bouwman, K.M. & Komdeur, J. 2006. Weather conditions affect levels of extra-pairpaternity in the reed bunting Emberiza schoeniclus. Journal of Avian Biology37:238-244.67Chan, Y. & Arcese, P. 2002. Subspecific differentiationand conservation of songsparrows (Melospiza melodia) in the SanFransisco Bay region inferred bymicrosatellite loci analysis. The Auk 119:641-651.Charmantier, A. & Perret, P. 2004. Manipulation ofnest-box density affects extra-pairpaternity in a population of blue tits (Parus caeruleus).Behavioral Ecology andSociobiology 56:360-365.Chuang, H.C., Webster, M.S. & Holmes,R.T. 1999. Extra-pair paternity and localsynchrony in the black-throated blue warbler.Auk 116:726-736.Chuang-Dobbs, H.C., Webster, M.S.& Holmes, R.T. 2001. The effectiveness of mateguarding 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 ofbreedingdensity and plumage coloration on mateguarding and cuckoldry in bluegrosbeaks (Passerina caerulea). CanadianJournal of Zoology 83:1143-1148.Griffith, S.C., Owens, I.P.F. & Thuman, K.A. 2002.Extra-pair paternity in birds: areview of interspecific variation and adaptive function.Molecular Ecology11:2195-2212.Hanotte, 0., Zanon, C., Pugh, A., Greig,C., Dixon, A. & Burke, T. 1994. Isolationandcharacterization of microsatellite loci in apasserine bird: the reed buntingEmberiza schoeniclus. Molecular Ecology3:529-530.Hill, C.E. 1999. Song and extra-pair matechoice in song sparrows. Ph.D. Thesis, DeptofPsychology, University of Washington, Seattle.Jeffrey, K.J., Keller, L.F., Arcese, P. & Bruford,M.W. 2001. The development ofmicrosatellite loci in the song sparrow, Melospizamelodia (Ayes), andgenotyping error associated with goodquality DNA. Molecular Ecology Notes1:11—13.Johnsen, A. & Liljeld, J.T. 2003.Ecological constraints on extra-pairpaternity in the bluethroat. Oecologia 136:476-483.Kempenaers, B. 1993. The use ofbreedingsynchrony index. Ornis Scandinavica 24:84.Kokko, H. & Rankin, D.J. 2006. Lonely hearts or sexin the city? Density-dependenteffects in mating systems. PhilosophicalTransactions of the Royal Society B361:319-334.Komdeur, J. 2001. Mate guardingin the Seychelles warbler is energetically costlyandadjusted to paternity risk. Proceedings ofthe Royal Society of London B268:2103-2111.68Komdeur, J., Burke, T. & Richardson, D.S. 2007. Explicit experimental evidence fortheeffectiveness of proximity as mate-guarding behaviour in reducing extra-pairfertilization in the Seychelles warbler. Molecular Ecology 16:3679-3688.Marshall, T.C., Slate, J., Kruuk, L. & Pemberton, J.M. 1998. Statistical confidenceforlikelihood-based paternity inference in natural populations. Molecular Ecology7:639-655.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 sparrowMelospiza 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 amigratory songbird. Molecular Ecology 17:1248-1255.Petren, K. 1998. Microsatellite primers from Geospiza fortis and cross-speciesamplification in Darwin’s finches. Molecular Ecology 7:1771-1788.Richardson, D.S. & Burke, T. 2001. Extrapair paternity and variance in reproductivesuccess related to breeding density in Bullock’s orioles. Animal Behaviour62:519-525.Saino, N., Primmer, C.R., Ellegren, H. & Møller, A.P. 1999. Breeding synchrony andpaternity in the barn swallow (Hirundo rustica). Behavioral Ecology andSociobiology 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 alternativeterritorial 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-pairmating systems in songbirds. Behaviour 132:675-690.69OL‘L9I-ccJ:I7TiCojoojP-I°NspTq)pTqpuTM-pJU!Aiu.indiind-nixutoun{Juisiopnjjnioduipunpnndsjossojcooz‘YH‘Ni’u6001ot:uquopqsqqnjojoq3Tu.IJjoJnu.inof(s 11 vopadvvqda9(ElpvJ)SJJTSIqMupjoJossonsuu.nmjind-nixpun‘Auonpu(suipoaiqjnwj‘uoinuownuioJdt1jnN6OOV‘pJn2’UJ7A‘UUOUUAct’c-Lc:c6u1JnqosussiMJmn(s’.z 1 vJopadv 1 vqda3tüpvJ)sJopsiqM.upoUtuOitpUiCSuipatq4JMCrnAuoissinjnuop.upunuipJnnOJA[soozur‘uuocEunA9817-Egt’:cEAoIotflU1flAJOI1Tfl0fT!1-’J1”-’JiouopvJpoautvz.zdsoapijsonndsuosmooiv‘1q1n‘.aj‘tunj69-E9:TooojI°”4HuouiuiooqiU’kisuppunAuo.upu 1 csuipaqXqpounguisi(iuuoindi,ndniixtOOiV1‘mnqun!u.vI‘uosiJ‘ud‘uunu‘f)J‘snisnqj9Z1-6t1:017AoJoiqotDogpunXooojInJOTAnIpHP°P°°HUISSOOflSUOiinZqI1JJiind-niixjosinIauo3L66T39‘II’1‘Wf‘1PI“V8‘J°1‘TU‘J-’°P°M‘HM‘1d!J‘1’\ffH‘(‘nqqoin-ig4 GENERAL DISCUSSIONThe adaptive significance of EPP is unclear despite a very large volume of research onthe topic (reviews in Griffith et al. 2002, Westneat and Stewart 2003, Akcay andRoughgarden 2007). There is a need for long-term studies that test hypotheses related toEPP using large sample sizes, genetic paternity data from most individuals in thepopulation, and detailed data on individual traits. My thesis addresses this need byexamining the potential adaptive significance of EPP in an island population of songsparrows for which the paternity of nearly all offspring and the identity of nearly allextra-pair sires in the population were known from 1993-96. Further, this population hasbeen studied in detail continuously since 1975 (Smith et al. 2006), thereby allowing me toexamine a variety of individual life history traits and demographic variables in relation toEPP. In Chapter 2, I tested the good genes hypothesis to determine if females obtainedfitness benefits from extra-pair males. In Chapter 3, I investigated the influence of twoecological 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 whichher social mate often provides parental care. The loss of paternity from the male’s ownnest and the large amount of care he potentially invests in offspring that are not his ownhas the potential to create significant sexual conflict between male and female mates. Forexample, males may employ several retaliatory tactics when they detect that their matehas obtained EPCs, including withholding of parental care or engaging in physicalpunishment. Given the potential costs of these male tactics, it has been suggested thatfemales must also benefit from obtaining EPCs in order for this behavior to have evolved.71For example, in several species females have been observed to pursue EPCs duringextraterritorial forays during their fertile periods, indicating that EPP may not simply bethe result of coercion by extra-pair males. However, there is little consensus in theliterature on whether or not females benefit from EPP.Hypotheses for how females may benefit from EPP are broadly divided into directbenefits which increase female reproductive success in a current season (e.g. fertilityinsurance, access to breeding resources on the extra-pair male’s territory) and indirectbenefits which improve the fitness of the female’s offspring (e.g. good genes, geneticcompatibility) (see Chapter 1). The costs and benefits associated with EPP may also bealtered by ecological factors (e.g. breeding synchrony, breeding density) which maychange the availability of potential extra-pair mates in space and time. Given theprevalence of EPP in avian mating systems (reviewed in Griffith et al. 2002),understanding the adaptive significance of EPP is necessary to understanding theevolution ofmating systems overall.In my thesis I used a population of socially monogamous song sparrows resident onMandarte 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 extrapair males from 1993-96. I first tested the good genes hypothesis which predicts thatfemales mated to males of low fitness should mate with extra-pair males of higher fitnessin order to improve the fitness of extra-pair offspring compared to their within-pairmaternal half-siblings. A potential flaw ofmany studies that test indirect benefits72hypotheses is that they use small sample sizes and do not compare the fitness of EPY tothat of their within-pair maternal half-sibs. Further, many studies use traits that are notclearly linked to fitness such as body size and condition and plumage ornamentation. Inmy thesis I directly compared the fitness of EPY to their within-pair haif-sibs, using traitsbased on the long-term data that were closely linked to fitness. These traits included lifespan, the number and proportion of successful social nest attempts produced in a lifetime,survival to independence and recruitment, and the number of independent and recruitedgenetic 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 malesto obtain ‘good genes’. Further, I found no difference in fitness between extra-pair malesand the males they cuckolded. These results are consistent with several other studieswith 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 ornot females engage in EPCs to obtain ‘good genes’. Although I used a relatively largedata set (n 751 offspring) and tested traits closely linked to fitness, I did notdemonstrate a fitness benefit of EPP to females. It is possible that females do not obtainfitness benefits from EPP, but engage in EPCs to make the ‘best of a bad job’ when thereis strong selection in males to achieve EPCs (sexually antagonistic coevolution; Arnqvistand Kirkpatrick 2005). However, it is also possible that the traits I used did notaccurately capture fitness. For example, Hunt et al. (2004) argue that the number ofgrand-offspring produced is a better measure of fitness than lifetime offspring production.73I did not measure the number of grand-offspring produced for male offspringbecausegenetic data were available from 1993-96 whereas many offspring continuedtoreproduce after 1996. Future studies should attempt to estimatethe number of grand-offspring produced, where possible, to more accurately assess fitness differencesbetweenEPY and their maternal half-sibs. Furthermore, it wouldbe interesting to test the geneticcompatibility hypothesis (i.e. that females reduce inbreedingby engaging in EPCs) in thispopulation of song sparrows, given that inbreeding has been shown to reduce fitness(Keller 1998). Because Keller and colleagues are currentlyobtaining precise estimates ofrelatedness by correcting the social pedigree for the Mandarte populationusing geneticmaterial from essentially all birds hatched on the island since 1993, more detailedanalyses of the relation between genetic compatibility and EPPmay soon be possible.I also related male fitness to the number of EPY sired, andto the proportion of paternitylost from a male’s own nest. I did not find evidencethat fitter males sired more EPYannually as none of the fitness-related traits I tested were significantly relatedto thenumber of EPY sired by males. Further, there was no repeatabilityin the number of EPYsired annually by males across years, thus providing further evidence againsta ‘goodgenes’ model. Age-related effects on a male’s ability to sireEPY may have caused thelack of repeatability in the number of EPY sired annuallyby males across years. I alsofound that none of the fitness-related traits I testedwere significantly related to theproportion of paternity lost by males annually. However,I did find that there was weakbut significant repeatability in the proportion ofpaternitylost by males across years,suggesting that the level of paternity loss may havebeen an intrinsic trait of individual74males. 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 fledgedyoung (Janssen et al. 2008). Another possibility is that the male’s social mate partlydetennines the proportion of EPY within broods as the majority ofmales had the samemate within and across years (e.g. Dietrich et al. 2004). Future studies might aim toexamine changes in the rate of EPP within broods across consecutive nesting attemptswhen pairs stay together versus when mate switching occurs.One of the most interesting results from my thesis was that male success at siring EPYwas significantly related to male age: the number of EPY sired by males increased up tothe age of four years, and then declined in males aged five years and older. This is incontrast to the hypothesis that older males are preferred by females for their ‘proven’viability genes (reviewed in Brooks and Kemp 2001). While many studies havedemonstrated 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 tobe the first to show a decline in extra-pair mating success in old age (note, however, thatSchmoll et al. (2007) reports that extra-pair mating success leveled off in male coal titsthree years and older). The rise and then decline in extra-pair mating success in songsparrows suggests that success is related to both experience and physical ability in malesong sparrows, a suggestion that is in line with several other studies of age-relatedperformance in this population (e.g. Smith et al. 2006). My ability to detect a decline in75extra-pair mating success may have been due to the detailed nature of the Mandartedataset, where the exact age of every individual in the population was known with precisionbecause all birds were individually color-banded and tracked throughout their lives. Bycontrast, 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 levelof EPP in song sparrows. I found that the proportion of EPY within a male’s nest wasnegatively related to breeding synchrony among neighbors, thus providing support for the‘mate guarding constraint’ hypothesis. However, I was unable to support this hypothesisin individual-level analyses. For example, males were not more likely to sire EPYoutside the fertile period of their social mate than expected by chance. Further, extra-pairmales whose mate’s fertile period overlapped that ofthe extra-pair female were not morelikely 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 andsynchrony are negatively related (e.g. Thusius et al. 2001, Van Dongen and Mulder2009). To test this hypothesis further, I recommend that future studies obtain quantitativedata on mate guarding in song sparrows, including the relationship between the amountoftime a male spends mate guarding and the level ofbreeding synchrony on adjacentterritories. If mate guarding does limit a male’s ability to engage in EPCs, then whensynchrony is low I would expect to observe males guarding their mates more intensely76during the fertile period and neighboring males performing extra-territorial intrusions at ahigher frequency. To date, few studies have obtained quantitative data on mate guardingin 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 andpopulation level in each of four years, similar to other studies (e.g. Johnsen and Lifjeld2003, Westneat and Mays 2005, Stewart et al. 2006). However, I may not have detecteda relationship between density and EPP if, for example, males adjust mate guarding inresponse to the perceived risk of cuckoldry, resulting in similar levels of EPP withinbroods at high and low densities. In order to test this hypothesis, quantitative data onmate guarding would be required to determine if males nesting in high density areas mateguard more intensely than males nesting in low density areas. Another possibility is thatfemales are highly selective in their choice of extra-pair mate and do not necessarilyengage in EPCs even when nesting on a territory surrounded by a high density ofneighboring males. Ideally, testing this hypothesis would involve radio-tracking femalesto determine if they pursue EPCs during extraterritorial forays, determining whetherfemales gain fitness benefits from EPCs, and identifiing the traits of males that femalesengage in EPCs with.In conclusion, I have examined hypotheses related to the adaptive significance of EPP insong sparrows: that females mate with extra-pair males to improve the fitness of EPYrelative to within-pair maternal half-sibs, and that breeding synchrony and densityinfluence the frequency of EPP. In order to further this particular field of study,778LsMoJ.ndsuosj1UJUOUIIJJUTUOTPWAUiCijd(iuiAuoJq3u/suipaTqpoojjooijrn&TTprnnopujuiuoSTUJIpU1puiisnssnsaiNJOsoipnjsarn4njojuisiCsoupTAo1dJJTMpU‘suoipJusqoqnsuqituJOJanuJs(spsoppuizisspujsjPU1N‘ppgUT1OIJOD011{nogJ!pU04J0O.I1?P?p1oTAPqqA!P4!U1flbqnoqjypuiiid-.iixpui‘ouwjinoosiq‘JuijouoiaiutjopodsipJoiArnqiSAUTsTpns1tqpuouuuooal(ooz)1uM1spu3US\4.1 ReferencesAkcay, E. & Roughgarden, J. 2007. Extra-pairpaternity in birds: review of the geneticbenefits. Evolutionary Ecology Research9:855-868.Arlt, D., Hansson, B., Bensch, S., von Schantz, T.& Hasseiquist, D. 2004. Breedingsynchrony does not affect extra-pair paternityin great reed warblers. Behaviour141:863-880.Arnqvist, G. & Kirkpatrick, M.2005. The evolution ofinfidelity in sociallymonogamouspasserines: the strength of directand indirect selection on extrapair copulationbehaviour in females. American Naturalist165:S26-S37.Bouwman, K.M, Van Dijk, R.E.,Wijmenga, J.J. & Komdeur, J. 2007. Older malereedbuntings are more successful at gaining extra-pairfertilizations. AnimalBehaviour 73:15-27.Brooks, R. & Kemp, D.J. 2001.Can older males deliver the good genes? TrendsinEcology and Evolution 16:308-313.Chuang-Dobbs, H.C., Webster, M.S. & Holmes,R.T. 2001. The effectiveness ofmateguarding 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 variationin extra-pair paternity in the coaltit(Parus ater). Behaviour 141 :817-835.Griffith, S.C., Owens, I.P.F. & Thuman, K.A. 2002.Extra-pair paternity in birds: areview of interspecific variationand adaptive function. Molecular Ecology11:2195-2212.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. andK.J., Jewell. 2008. Polyandry and Sex Ratiointhe Song Sparrow. Wilson Journal ofOrnithology 120:395-398.Johnsen, A. & Lifjeld, J.T. 2003. Ecologicalconstraints on extra-pair paternity in thebluethroat. Oecologia 136:476-483.Keller, L.F. 1988. Inbreeding and its fitnesseffects in an insular populationof songsparrows (Melospiza melodia). Evolution52:240-250.79O’Connor, K.D., Marr, A.B., Arcese, P., Keller, L.F., Jeffrey,K.J., & Bruford, M.W.2006. Extra-pair fertilization and effective populationsize in the song sparrowMelospiza melodia. The Journal of Avian Biology 37:572-578.Schmoll, T., Dietrich, V., Winkel, W., Epplen, J.T. & Lubjuhn, T. 2003.Long-term fitnessconsequences of female extra-pair matings in a sociallymonogamous 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 agepredicts extrapair and total fertilization success in thesocially 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 paternityloss of within-pair and extra-pairoffspring in the coal tit Periparus ater. Proceedings ofthe Royal Society of LondonB 276:337-345.Smith, J.N.M., Keller, L.F., Marr, A.B. & Arcese, P. 2006. Biologyof small populations:the song sparrows of Mandarte Island. Oxford UniversityPress, New York.Thusius, K.J., Dunn, P.O., Peterson, K.A. & Whittingham,L.A. 2001. Extrapair paternity isinfluenced by breeding synchrony and density in thecommon yellowthroat.Behavioral Ecology 12:633-639.van Dongen, W.F.D. 2008. Mate guarding and territorial aggression varywith breedingsynchrony in golden whistlers (Pachycephala pectoralis).Naturwissenschaften95:537-545.van Dongen, W.F.D. & Mulder, R.A. 2009. Multiple ornamentation,female breedingsynchrony, and extra-pair mating success ofgolden whistlers (Pachycephalapectoralis). Journal of Ornithology. Published online:10 February 2009.Westneat, D.F. & Mays, H.L. 2005. Tests of spatial andtemporal factors influencingextra-pair paternity in red-winged blackbirds. MolecularEcology 14:2155-2 167.Westneat, D.F. & Stewart, I.R.K. 2003. Extra-pair paternityin birds: Causes, correlates,and conflict. Annual Reviewof Ecology Evolution and Systematics 34:365-396.Whittingham, LA. & Dunn, P.O. 2001. Survivalof extra-pair and within-pair young in treeswallows. Behavioral Ecology 12:496-500.80

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0068158/manifest

Comment

Related Items