The evolution of the social and genetic mating system
of purple-crowned fairy-wrens Malurus coronatus
Dissertation
zur Erlangung des akademischen Grades des Doktors der Naturwissenschaften (Dr. rer. Nat.) an der Universität Konstanz Mathematisch-Naturwissenschaftliche Sektion Fachbereich Biologie
vorgelegt von
Sjouke Anne Kingma
Konstanz, 2011
Tag der mündlichen Prüfung: 20. July 2011
1. Referentin: Dr. Anne Peters 2. Referent: Prof. Dr. Martin Wikelski 3. Referent: Prof. Dr. Karl-Otto Rothhaupt
2
foar ús âlde lju
3
4
Purple-crowned fairy-wren, ‘Artist’s impression’ by Eke Mekkes
“Charming as are many of Australian birds, I think the present species is entitled to the palm for elegance and beauty, not only among the members of its own genus, numerous and beautiful as they really are, but among all other groups of birds yet discovered; the charm, too, is considerably enhanced by the great novelty in the style of its colouring; for in how few birds do we find the lovely lilac tint which encircles and adorns the head of the bird!”
John Gould in Birds of Australia , 1865
5
6 Table of contents
Summary ...... 11
Zusammenfassung (German summary)...... 13
Chapter 1. General introduction...... 17
Natural selection and reproductive strategies...... 19 The evolution of extra-pair mating in birds ...... 20 The evolution of cooperative breeding ...... 23 Extra-pair mating and cooperation ...... 26
Chapter 2. Radical loss of an extreme extra-pair mating system...... 31
Abstract...... 33 Introduction ...... 35 Methods...... 37 Study species ...... 37 Comparing Malurus species ...... 39 Paternity analyses in M. coronatus ...... 41 Results ...... 44 Extra-pair paternity...... 44 Adaptations to extra-pair mating: morphology and behaviour ...... 46 Comparing Malurus life-history and ecology ...... 48 Discussion ...... 49 Comparing Malurus life-history and ecology ...... 50 Conclusion ...... 53 Acknowledgements...... 53
Chapter 3. Female fairy-wrens synchronise egg laying to facilitate extra-pair mating for inbreeding avoidance...... 55
Abstract...... 57 Introduction ...... 59 Methods...... 61 Study species and fieldwork ...... 61 Paternity analyses...... 62 Relatedness...... 63 Analyses...... 63 Results ...... 66 Inbreeding depression and inbreeding avoidance by EG mating...... 66
7 Ecological constraints on extra-group mating...... 67 Synchronisation by incestuously mated females ...... 68 Discussion ...... 71 Extra-group mating for inbreeding avoidance ...... 71 Synchronisation to overcome constraints on EG mating...... 72 Acknowledgements ...... 74
Chapter 4. Multiple benefits drive helping behaviour in a cooperatively breeding bird: an integrated analysis ...... 77
Abstract...... 79 Introduction ...... 81 Methods...... 84 Study site and species ...... 84 Relatedness and parentage...... 84 Feeding watches ...... 85 Analyses...... 85 Results...... 88 Discussion ...... 93 Indirect and direct fitness benefits...... 93 Interacting hypotheses explain cooperation: implications ...... 95 Acknowledgements ...... 95
Chapter 5. Multiple benefits of cooperative breeding in purple- crowned fairy-wrens: a consequence of fidelity? ...... 97
Abstract...... 99 Introduction ...... 101 Methods...... 103 Study site and species ...... 103 Population and nest monitoring ...... 103 Group size ...... 104 Nest observations ...... 104 Statistical analyses ...... 105 Comparative study ...... 108 Results...... 111 Productivity ...... 111 Feeding rates ...... 111 Comparative study ...... 115 Discussion ...... 118 Benefits of cooperative breeding in M. coronatus ...... 118 The relative importance of compensatory and additive effects ...... 119 Benefits of cooperative breeding: a role for extra-group mating? ...... 120 Conclusions...... 120 Acknowledgements ...... 121
8 Chapter 6. No evidence for offspring sex-ratio adjustment to social or environmental conditions in cooperatively breeding purple-crowned fairy-wrens...... 123
Abstract...... 125 Introduction ...... 127 Methods...... 130 Study species and field work ...... 130 Nestling feeding watches ...... 131 Environmental conditions: food abundance, rainfall, and territory quality...... 131 Molecular sexing...... 132 Sex ratio...... 132 Statistical analyses ...... 133 Results ...... 136 Helper-repayment hypothesis...... 136 Helper-competition hypothesis ...... 136 Costly-sex and Trivers-Willard hypothesis ...... 137 Discussion ...... 139 Helper-repayment and competition...... 139 The costly sex, environmental conditions, and future benefits...... 140 Implications ...... 141 Acknowledgements...... 142
Chapter 7. General discussion...... 145
Natural selection and reproductive strategies in M. coronatus ...... 147 The evolution of extra-pair mating in Malurus coronatus ...... 147 The evolution of cooperative breeding in Malurus coronatus ...... 149 The monogamy hypothesis: extra-pair mating affects cooperation...... 149 Cooperative breeding and offspring sex ratio...... 150 Conclusions ...... 151
References...... 155
Erklärung ...... 187
Record of achievement...... 189
Acknowledgements ...... 191
Curriculum vitae ...... 195
List of publications ...... 199
9
10 Summary
In this thesis, I describe my research on the causes and consequence of extra- pair (EP) mating and cooperative breeding in a passerine bird, the purple- crowned fairy-wren, Malurus coronatus. These costly behaviours are common aspects of animal mating systems, but how these are evolutionarily adaptive has puzzled biologists for decades. Why do females copulate with males outside their pair-bond? Why do some individuals assist in raising other individuals’ offspring? These questions were addressed based on detailed field-observations, combined with molecular analyses. Overall, relatedness among individuals is shown to be important in determining fitness benefits of both EP mating and cooperative behaviour in M. coronatus , and therefore predicts whether individuals engage in particular behaviours. In turn, as EP mating reduces relatedness among individuals, this appears to have important implications for the evolution of cooperative behaviour in general.
Females in many bird species actively pursue EP mating, but understanding why they do so (or why not) has generally proven challenging. I show that EP mating functions mainly in inbreeding avoidance in M. coronatus : broods of females paired with a first-order relative had almost 15 times higher rates of EPP than broods of females with an unrelated partner. Inbreeding resulted in increased hatching failure of eggs, and therefore EP mating in this species constitutes an adaptive mechanism to avoid large costs of inbreeding. While EP mating is adaptive in M. coronatus , overall levels of EPP are considerably lower (4-6% of offspring) than in its highly promiscuous congeners (40-80%). This is surprising, because life-history features that determine costs or benefits of EP mating (e.g., the extent of paternal care and incestuous pairings) appear rather similar between the species. Possibly, M. coronatus females are restricted in EP mating by relatively low density and breeding synchrony that limit access to fertile EP males. Within M. coronatus , variation in such factors indeed explained part of the variation in EP mating, but females in an incestuous pairing actively overcame such constraints by synchronizing breeding with fertility of unrelated neighbouring males. Thus, EP mating behaviour is shaped by a complex interplay between benefits and constraints and the relative importance of those can be actively modulated by females. Moreover, minor changes in the
11 costs, benefits and/or constraints can have far-reaching effects, suggesting that EP mating can evolutionarily be a rather flexible behaviour.
Cooperative behaviour that does not yield benefits for survival or reproductive success is in contrast with natural selection theory, and how such seemingly altruistic behaviour can persist is a prevailing topic in evolutionary biology. I show that helpers in M. coronatus feed siblings more than unrelated nestlings. Because helpers improve reproductive success and breeder survival, this supports the prediction of the kin-selection theory that helpers gain inclusive fitness of increased gene-transfer through relatives. Additionally, direct benefits also appear important in explaining variation in helper investment: less related helpers feed nestlings more when they are more likely to inherit the breeder position, probably to augment groups with future helpers. This therefore provides one solution of the puzzle how cooperation among non-related individuals can remain evolutionarily stable.
Helpers in M. coronatus are mostly males and breeder females may therefore benefit most from producing sons. Nonetheless, I did not find evidence for the hypothesis that females should adjust offspring sex ratio to overproduce males. One possibility is that sex-bias in long-term fitness returns is too small to drive evolution of sex ratio adjustment mechanisms, and this requires more in depth investigation.
The finding that helpers in M. coronatus have substantial beneficial effects on breeder survival and reproductive success is in strong contrast with its promiscuous congeners, in which such effects are much smaller. Since kinship is a major driver of apparently altruistic behaviour, low rates of EPP could contribute to cooperative behaviour. In a comparative study among 37 bird species, I show that this is a general pattern as enhancement of reproductive success and breeder survival by helpers is more frequent in species with lower levels of EPP. This result likely explains why cooperative breeding is associated with relatively low rates of EPP in general, and strongly suggests that the genetic mating system plays an important role in the evolution of social systems in animals.
12 Zusammenfassung
German Summary
In dieser Dissertation beschreibe ich meine Forschungen zu Ursachen und Konsequenzen von Kopulationen außerhalb des Paarbundes (extra-pair mating, EP) und von kooperativer Jungenaufzucht an einer Singvogelart, dem Purpurkopf-Staffelschwanz (auch: Purpurrot-gekrönter Fee-Zaunkönig), Malurus coronatus. Solch kostenaufwendige Verhaltensweisen sind häufig Aspekte von Paarungssystemen der Tiere, doch inwiefern sie evolutionär adaptiv sein können, beschäftigt Biologen schon seit Jahrzehnten. Warum kopulieren Weibchen mit Männchen außerhalb des Paarbundes? Warum helfen einige Individuen bei der Aufzucht von Jungen anderer Individuen? Diese Fragen wurden anhand von detailierten Freilandbeobachtungen in Kombination mit molekularen Analysen bearbeitet. Die Ergebnisse deuten darauf hin, dass der Verwandtschaftsgrad zwischen Individuen die Fitnessgewinne sowohl durch EP-Verhalten als auch durch kooperatives Verhalten entscheidend mitbestimmt. Im Gegenzug reduzieren Fremdvaterschaften den Verwandtschaftsgrad zwischen Individuen, und dies hat wichtige Konsequenzen für die Evolution von Kooperationsverhalten im Allgemeinen.
Weibchen vieler Arten suchen aktiv EP-Kopulationen auf, aber Gründe für ein Vorkommen - oder für den Umfang - dieses Verhaltens sind bisher noch wenig verstanden. In meiner Arbeit zeige ich, dass Fremdvaterschaften bei M. coronatus hauptsächlich der Inzuchtvermeidung dienen: Bruten von Weibchen, die mit einem Verwandten ersten Grades verpaart waren, zeigten fast fünfzehnfach höhere EP-Raten als Bruten von Weibchen, die mit ihrem Partner nicht-verwandt waren. Weil Inzucht den Schlüpferfolg deutlich minderte, dient Fremdvaterschaft bei dieser Art als adaptiver Mechanismus der Vermeidung hoher Inzuchtkosten. Obwohl EP-Kopulationen bei M. coronatus adaptiv sind, ist die Rate der Fremdvaterschaft deutlich niedriger (4-6% des Nachwuchses) als bei seinen promisken Gattungsmitgliedern (40-80%). Dies ist insofern überraschend, als Aspekte des Lebenslaufs, die die Kosten und Nutzen von EP-Kopulationen mitbestimmen (z.B. Ausmaß der väterlichen Brutpflege und Inzucht), zwischen den Arten relativ ähnlich zu sein scheinen.
13 Möglicherweise haben Weibchen von M. coronatus weniger Gelegenheit zu EP-Kopulationen, weil eine niedrigere Brutdichte und Paarsynchronität den Zugang zu fertilen Männchen limitieren. Innerhalb von M. coronatus erklärte die Variation dieser Faktoren tatsächlich einen Teil der Variation an Fremdvaterschaften. Jedoch überwanden Weibchen, die mit einem Verwandten verpaart waren, solche Hindernisse aktiv, indem sie ihr Brutverhalten mit der fruchtbaren Phase von nicht-verwandten benachbarten Männchen synchronisierten. Folglich wird EP-Verhalten von einem komplexen Zusammenspiel von Vorteilen und Beschränkungen bestimmt, deren relative Bedeutung von den Weibchen aktiv moduliert werden kann. Dass geringfügige Veränderungen von Kosten, Vorteilen und Beschränkungen weitreichende Wirkungen haben können, weist darauf hin, dass EP-Verhalten evolutionär relativ flexibel sein kann.
Kooperatives Verhalten, das keine Vorteile für Überleben oder Fortpflanzungserfolg bedeutet, steht im Widerspruch zur Theorie der natürlichen Auslese. Wie anscheinend altruistisches Verhalten dennoch bestehen kann, ist ein vornehmliches Thema der Evolutionsbiologie. In meiner Arbeit zeige ich, dass Helfer bei M. coronatus ihre Geschwister häufiger als nicht-verwandte Nestjunge füttern. Weil Helfer den Fortpflanzungserfolg und das Überleben des Brutpaares erhöhen, bestätigt dieser Befund eine Vorhersage, die auf Verwandtenselektion (kin selection) basiert: Helfer erhöhen ihre Gesamtfitness (inclusive fitness) durch erhöhten Gentransfer durch Verwandte. Zusätzliche direkte Vorteile scheinen ebenfalls wichtig zu sein für die Erklärung von Variation in der Einsatzbereitschaft der Helfer. Helfer füttern weniger nah verwandte Nestjunge stärker, wenn sie gute Aussichten darauf haben, die Rolle des Brutvogels zu übernehmen, und verstärken so möglicherweise die Gruppe ihrer zukünftigen Helfer. Diese Beobachtungen zeigen somit einen Lösungsweg auf für das Rätsel, wie Kooperation zwischen nicht-verwandten Individuen evolutionär stabil bleiben kann.
Obwohl Helfer bei M. coronatus besonders häufig männlich sind und brütende Weibchen daher von Söhnen besonders profitieren, finde ich keinen Beleg für die Hypothese, dass Weibchen das Geschlechterverhältnis ihres Nachwuchses zugunsten einer höheren Zahl von Söhnen modulieren sollten. Eine mögliche Erklärung dafür ist, dass auf lange Sicht Fitness-Vorteile zu
14 gering sind, um die Evolution von Mechanismen zur Anpassung des Geschlechterverhältnisses voranzutreiben, doch dies bedarf näherer Untersuchung.
Der erhebliche Vorteil für Überleben und Fortpflanzungserfolg von Brutvögeln durch Helfer bei M. coronatus steht in starkem Kontrast zu den promisken Arten der Gattung Malurus , bei denen solche Effekte viel geringer sind. Weil scheinbar altruistisches Verhalten durch Verwandtschaft begünstigt wird, könnten niedrige Fremdvaterschaftsraten zur Ausprägung von Kooperationsverhalten beitragen. Anhand einer vergleichenden Studie an 37 Vogelarten zeige ich, dass sich Überleben und Fortpflanzungserfolg der Brutvögel durch Helfer bei Arten mit niedrigeren Fremdvaterschaftsraten generell viel deutlicher erhöhen. Im Einklang dazu berichtet eine neue Studie einen Zusammenhang von kooperativem Brutverhalten und relativ niedriger Fremdvaterschaftsrate. Meine Befunde zeigen dass das genetische Paarungssystem eine wichtige Rolle spielt für die Evolution von Sozialsystemen der Tiere.
15
16 Chapter 1
General introduction
17
18 Natural selection and reproductive strategies
Darwin’s theory of evolution by natural selection predicts that behavioural, physiological and morphological traits are selected in a way that they maximise individuals’ genetic contribution to future generations (Darwin 1859, Endler 1986, Fisher 1930, Williams 1966). Broadly, this implies that (heritable) traits should provide benefits for individuals’ survival or reproductive success, and those benefits should outweigh the associated costs for such traits to be evolutionarily stable (Maynard-Smith 1978). Hence, how costs and benefits shape certain traits forms one major component of current studies of evolutionary biology (Stearns 1989).
In this thesis, I will address the evolution of (female) extra-pair (EP) mating and cooperative breeding. Despite common occurrence and considerable research attention, the question whether and how these aspects of animal reproductive behaviour evolved under natural selection has puzzled evolutionary biologists for decades. The main reason for this is that the benefits of such behaviour seem small relative to the costs, or that the benefits may not be immediately clear in the first place (e.g., Arnqvist & Kirkpatrick 2005). I studied the costs and benefits of EP mating and cooperative breeding based on field- and molecular data on a passerine bird, the purple-crowned fairy-wren Malurus coronatus (see Box 3 and 4). The following questions are central in this study:
1. What determines whether females engage in extra-pair mating?
2. Why do some individuals assist in raising other individuals’ offspring?
Below I will explain the reasoning behind these questions in more detail. Although EP mating and cooperative breeding seem initially quite diverse topics, I will subsequently illustrate that there may exist a strong evolutionary link between these behaviours.
19 The evolution of extra-pair mating in birds
What determines whether females engage in extra-pair mating?
Individuals in most bird species form socially monogamous pairs (Cockburn 2006, Lack 1968). Nonetheless, some of the offspring in nearly all studied species are sired by EP males (Griffith et al. 2002). EP mating has been studied in more than 250 bird species (Cornwallis et al. 2010), but answering the question why individuals engage in copulations with individuals outside the pair bond has proven surprisingly challenging. Males can improve their reproductive output by mating promiscuously (sensu Bateman 1948; see Schlicht & Kempenaers 2011), but the evolution of female EP mating is harder to explain, and is currently one of the most debated topics in the study of avian mating systems (e.g., Arnold & Owens 2002, Arnqvist & Kirkpatrick 2005, Griffith et al. 2002, Jennions & Petrie 2000). Behavioural observations have revealed that extra-pair paternity (EPP) in birds often results from female pursuits, suggesting that EP mating should be beneficial for females (Currie et al. 1998, Double & Cockburn 2000, Kempenaers et al. 1997). Because females cannot improve rates of egg production and rarely gain direct benefits by mating EP, current hypotheses predict that they may improve offspring (genetic) quality by doing so (Griffith et al. 2002, Trivers 1972; see Box 1). Quality of offspring can be improved when these inherit relatively ‘good genes’ from an EP father (Hasselquist et al. 1996, Jennions & Petrie 2000) or when EP mating with genetically relatively dissimilar males results in more heterozygous offspring (i.e., as form of inbreeding avoidance; Blomquist et al. 2002, Foerster et al. 2003). Evidence for such indirect benefits is, however, mixed and inconclusive (see e.g., Akçay & Roughgarden 2007, Kempenaers 2007). In contrast, the costs of EP mating are well established (Box 1). In fact, Arnqvist and Kirkpatrick (2005) suggest that the selection pressure on EP mating should be negative, as their comparative analyses suggest that indirect benefits of EPP do not outweigh the costs of reduced paternal care. Arnqvist and Kirkpatrick therefore conclude that ‘EP copulations may be the result of antagonistic selection on loci influencing the outcome of male-female encounters and that EP copulation behaviour per se may not be adaptive for females but may reflect sexual conflict due to strong selection in males to achieve EP copulations’. Following Griffith (2007), I feel that this conclusion may be
20 premature, but the outcome of the analyses performed by Arnqvist and Kirkpatrick (2005) does suggest that we are far from a general understanding of how different costs and benefits shape the evolution of (female) EP mating. Many studies have aimed to attribute variation in levels of EPP within and across species to differences in a range of ecological characteristics (e.g., breeding density and synchrony; see Box 1) that may constrain females in having access to fertile EP males (Gowaty 1996; or alternative phrased by Petrie & Kempenaers 1998, that such factors enhance the cost of EP mating). In many species, it has been shown that there is a role for ecological factors in explaining variation in incidence of EPP, but general patterns are often less clear (see Westneat & Stewart 2003 for an overview). Instead, comparative study revealed that there is usually relatively limited inter-specific variation in levels of EPP within families or orders of birds (Arnold & Owens 2002; Box 1), suggesting that historical factors may play an important role in explaining variation in EP mating behaviour in general. Overall, the evolution of EP mating appears to be shaped by a complex interplay between several contemporary and historical factors that determine the magnitude of the costs, benefits, and constraints of EP mating simultaneously (see Box 1). Determining the relative role of those factors is therefore crucial to understand the evolution of EP mating in particular systems and in general. In chapters 2 and 3 , I study what factors determine EP mating in M. coronatus, using an inter- and intra-specific approach. The fairy-wren genus (see Box 3) is well known for among the highest rates of EPP known in birds (40-80% of offspring; Brooker et al. 1990, Griffith et al. 2002, Karubian 2002, Mulder et al. 1994, Webster et al. 2004). In chapter 2 , I report results from parentage analyses in M. coronatus . Subsequently, I test whether inter-specific variation in levels of EPP among the fairy-wrens (see Box 3) can be explained by differences in contemporary life history features that may determine differences in benefits, costs and constraints of EP mating (see Box 1). In chapter 3 , I subsequently explore what determines the variation in incidence of EPP within my study population of M. coronatus (see Box 4 for details on the study species and population) . Specifically, I test whether EP mating functions in inbreeding avoidance, and, especially when EP mating constitutes an apparent optimal strategy, whether females are constrained by ecological variables in doing so.
21
Box 1. Extra-pair mating: benefits, costs and constraints
The main hypothesised benefits, costs and constraints involved in female EP mating behaviour. Inter- and intra-specific evidence for most of these hypotheses is mixed, and there is no general consensus on the evolution of EP mating (see reviews in Petrie & Kempenaers 1998, Griffith et al. 2002, Arnold & Owens 2002, Westneat & Stewart 2003, Kempenaers 2007, Akçay & Roughgarden 2007).
Benefits Good genes Extra-pair males are of superior genetic quality (Petrie et al. 1998)
Compatible Extra-pair males are genetically more compatible/ genes/inbreeding dissimilar than the female’s social partner (Kempenaers avoidance 2007, Mays et al. 2008, Zeh & Zeh 1996)
Direct benefits Females acquire access to EP territories for foraging, or gain additional territory defense by EP males (Gray 1997)
By mating EP, females procure additional help in raising of their offspring (Rubenstein 2007a)
Costs Loss of partner's Males adjust their parental care to certainty of parentage investment (Birkhead & Møller 1992, Dixon et al. 1994)
Sexually Females mating with multiple males have increased risks transmittable of being exposed to sexually transmittable diseases disease (Sheldon 1993)
Constraints Mate guarding Males use behavioural strategies to prevent their female from EP mating (Birkhead 1979, Lifjeld et al. 1994)
Breeding density Low density constrains females access to (suitable) EP males (Bennett & Owens 2002)
Breeding Low synchrony causes females not to have access to synchrony fertile males or not being able to assess male quality (Stutchbury & Morton 1995)
High synchrony limits the number of available males, as these males are occupied in mate guarding or copulating (Birkhead & Biggins 1987)
Ancestral Phylogenetic There is a large phylogenetic component that explains effects correlation about 55% of the inter-specific variation in rates of EPP (Arnold & Owens 2002)
22 The evolution of cooperative breeding
Why do some individuals assist in raising other individuals’ offspring?
Cooperative breeding is broadly defined as the phenomenon that some individuals (helpers or subordinates) forego own reproduction to assist others in caring for their offspring (Brown 1987, Cockburn 1998). The breeding system is prevalent in many animal taxa, including fish (Taborsky 1984), mammals (Solomon & French 1996) and birds (Koenig & Dickinson 2004), and accounts for instance for around 9% of bird species (Cockburn 2006). Nonetheless, the seemingly altruistic nature of helping others is in contrast with predictions of natural selection theory that individuals should maximise their own reproductive success or survival (Emlen & Vehrenkamp 1983, Fisher 1930). Charles Darwin (1859) realised that explaining ‘self-sacrificing behaviour’ was one of the key problems of his natural selection theory. Over a century later, William Hamilton (1964) proposed the kin selection theory that predicts that cooperation can be indirectly selected for when this yields inclusive fitness benefits through enhanced transfer of (shared) genes of relatives. The kin selection theory has provided an important piece of the puzzle (see e.g., Clutton-Brock 2002, Griffin & West 2003), and even the solution in certain systems (e.g., explaining sterile workers in eusocial systems; Crozier & Pamilo 1996, Queller & Strassmann 1998), but kin selection does not provide a universal or exclusive explanation for the evolutionary maintenance of cooperative breeding in vertebrates (e.g., Dunn et al. 1995, Legge 2000a, Wright et al. 1999). Recently, several alternative explanations for helping have been proposed (summarised in Box 2). However, these alternative benefits are largely underrepresented in studies of cooperative behaviour, which is probably partly why it has been questioned whether (some of) these are of any importance at all (e.g., Wright 2007). In chapter 4 , I therefore test whether kin selection and/or direct benefits (Box 2) can explain the evolutionary maintenance of helping behaviour (specifically nestling feeding) in M. coronatus.
23
Box 2. The main proposed adaptive benefits of helping behaviour
Hypothesis Fitness benefits Principle
Kin selection theory Indirect Increasing productivity of relatives (Hamilton 1964) yields inclusive fitness benefits
Social Prestige Future direct By investing more, subordinates (Zahavi 1974) signal quality to potential future partners
Active group augmentation Future direct Subordinate increase recruitment (Kokko et al. 2001) and new recruits will in turn assist them when they obtain the breeding position
Passive group augmentation Future direct Increasing group size will passively (Kokko et al. 2001) lead to subordinates survival benefits (increased predator repellence, food finding efficiency etc.)
Parentage acquisition Current direct Subordinates obtain direct (Magrath & Whittingham 1997) reproductive success
Pay-to-stay Current direct Subordinates benefit from staying on (Kokko et al. 2002) the territory, but they must invest as ‘paying rent’ in order not be evicted by the other residents
Many of the proposed adaptive benefits of helping behaviour (see Box 2) require that subordinates increase reproductive success or survival of breeders, and in turn, breeders may tolerate additional individuals in their territory when those exert such fitness benefits (e.g., Kokko et al. 2002, Mulder & Langmore 1993, Reyer 1986). Helpers can benefit breeders for instance through enhanced accumulated care and subsequent higher reproductive output (additive effects; Cockburn 1998, Hatchwell 1999, Heinsohn 2004, Legge 2000b) or through lightening the breeders’ workload, resulting in improved breeder survival or increased rates of reproduction (compensatory effects; Cockburn 1998, Hatchwell 1999, Khan & Walters 2002). In chapter 5 , I determine the effect of helper nestling-feeding rates on overall feeding rates and reproductive success, and on breeder feeding rates and survival in M. coronatus .
24 As cooperative breeding can strongly affect parental fitness, cooperative breeding has become a popular system to study how parents can adopt strategies that maximise their fitness. For instance, individuals should be selected to overproduce the helping sex when helping behaviour is sex biased (Cockburn 1998) and incurs great fitness benefits (Fisher 1930, Griffin et al. 2005). In chapter 6 , I test whether females in M. coronatus adjust offspring sex ratio according the (sex-specific) fitness returns of helpers.
Box 3. The fairy-wrens
The Malurus genus comprises 12 species, of which nine (see illustrations) are endemic to Australia. Fairy-wrens are cooperatively breeding birds that are ‘notorious’ for their extreme levels of extra-pair paternity (40-80% of offspring are sired by a male outside the socially monogamous pair bond). Whether this also applies to purple-crowned fairy-wrens, and the reason for potential differences, is subject of chapter 2 and 3. In chapter 4, 5 and 6 the causes and consequences of cooperative breeding are addressed.
Figure 1.1. From left to right, top to bottom: White-winged fairy-wren ( M. leucopterus ), red-backed fairy-wren ( M. melanocephalus ), white-shouldered fairy-wren ( M. alboscapulatus ; from New- Guinea), variegated fairy-wren ( M. lamberti ), blue-breasted fairy-wren ( M. pulcherrimus ), lovely fairy-wren ( M. amabilis ), red-winged fairy-wren ( M. elegans ), splendid fairy-wren ( M. splendens ), purple-crowned fairy-wren ( M. coronatus ), superb fairy-wren ( M. cyaneus )
The illustrations are drawn by Peter Marsack, and are obtained from the book ‘The Fairy-wrens and Grass-wrens’ by Ian Rowley and Eleanor Russell (Oxford University Press, 1997).
25 Extra-pair mating and cooperation
Sex is an antisocial force in evolution – E.O. Wilson –
As outlined above, the kin selection theory predicts that cooperation in animal societies is driven by the gain of inclusive fitness benefits (Hamilton 1964, see also above). If so, it can be expected that multiple mating by females, which reduces relatedness among group members (Cornwallis et al. 2010), strongly affects the evolution of cooperative behaviour in animals (Boomsma 2007, 2009, Charnov 1981). Studies of this ‘monogamy hypothesis’ have only very recently emerged. Phylogenetic study shows that lifetime monogamy promoted the evolution of eusociality (sterile workers) in insects (Hughes et al. 2008). Additionally, Cornwallis et al. (2010) show, as proposed by Boomsma (2007), that cooperative breeding in birds is associated with relatively low levels of EPP. Thus, although restricted to limited examples, the genetic mating system seems to play an important role in the evolution of cooperation in animals. If helper feeding rates in cooperatively breeding birds generally depend on inclusive fitness benefits, it can be predicted that resulting benefits of cooperative breeding (i.e., enhanced reproductive success or breeder survival; see above) are related to levels of EPP across species. As the fairy-wren species differ to great extent in levels of EPP (see chapter 2) they offer the ideal system to test this prediction. Therefore, I explore whether benefits of cooperative breeding are different between the monogamous M. coronatus and its highly promiscuous congeners in chapter 5 . I subsequently use a comparative study among a wide range of cooperatively breeding bird species to test whether benefits of cooperative breeding are more prevalent in more faithful cooperatively breeding bird species in general.
26
Box 4. Natural history of Malurus coronatus and field work
The information provided in this box is a combination of findings from Rowley & Russell (1993 a, b, 1997), and published (Hall & Peters 2008, 2009, Kingma et al. 2009, 2010, 2011a,b) and unpublished data from our study.
Distribution and habitat
Purple-crowned fairy-wrens M. coronatus are small Australian passerine birds. Two subspecies have been distinguished, of which I studied M. coronatus coronatus , and I will refer to this subspecies for the remainder of this thesis. This subspecies is found in the Kimberly and the Victoria River regions in the north-west of Australia, (15-17 degrees latitude). These areas are characterised by a tropical savannah habitat with sparse vegetation (see Fig. 1.2). M. coronatus is strongly dependent on patches of dense vegetation along the rivers (Fig. 1.2), where it mainly occupies thickets of Pandanus aquaticus (see Fig. 1.3). All-purpose territories are stable year-round and, arranged in a linearly fashion, they include 50-400 m of river length, including both sides of the stream. The birds are usually found within 10-20 meters of the water’s edge, where they forage on a wide range of arthropods.
Figure 1.2. M. coronatus inhabit riparian vegetation in the north of Australia. This aerial picture shows the sparsely vegetated savannah with a stretch of dense vegetation along the river, and the research station in Mornington Wildlife Sanctuary. Photo by Michelle Hall / Australian Wildlife Conservancy.
Species description and social system
M. coronatus are sexually and seasonally dimorphic (see illustrations in Box 3 and Fig. 1.4), and therefore the sexes can easily be distinguished. Males and females undergo seasonal moult, alternating eclipse and respectively purple and dark grey crown plumage, whereas ear coverts remain black in males and rufous-brown in females year-round. Young subordinate birds’ plumage is often as in adults in eclipse plumage, but older subordinates can have similar breeding plumage as breeders. Nonetheless, breeder status can easily be determined by behavioural cues, the most obvious that only the breeding pair engage in duet-songs.
27
Figure 1.3. A cooperative group of M. coronatus with the subordinates / helpers around the breeder male and female in the middle. Photo by Michelle Hall / Australian Wildlife Conservancy.
In each territory, a male and female form a stable socially monogamous pair (annual divorce rate is 5-10% of pairs, and survival of adults is over 80%). Breeding pairs are in 40 to 70% of the cases accompanied by on average between one and two (up to nine) male and female subordinates (Fig. 1.3). These subordinates are most often retained offspring from previous broods, although in rare cases, birds from elsewhere settle as subordinate. Subordinates either disperse or they inherit the territory when the same-sex breeder disappears. Dispersal is most often short distance where subordinates take up a breeder vacancy nearby, but long-distance dispersal (over 60 km) has also been observed.
Breeding biology
In the tropical wet-dry climate in north Australia, rain falls mainly during the wet- season (December-March). Arthropod abundance in this period is highest, and possibly as a consequence, most breeding takes place in this period. However, the birds can breed year-round, with a second peak in the late dry season (August- October). In years with low rainfall, however, breeding outside the wet-season is absent or rare. Females can initiate multiple broods per year, and after brood loss or as soon as two months after fledging females can start nest-building again. Nests are built exclusively by the dominant female. Most nests are placed in the crown at the base of the Pandanus aquaticus leaves (between 0.2 and 6 m high), often within 5 m of the water (>90% of cases; Fig. 1.4).
Figure 1.4. M. coronatus build their dome-shaped nest at the base of the leaves of river pandanus ( Pandanus aquaticus ). Photo by Michelle Hall / Australian Wildlife Conservancy.
28
Eggs are laid around sunrise on consecutive days, and clutch size ranges from 1 to 4 eggs, most often 3. Eggs are incubated exclusively by the dominant female, and they hatch around 14 days after the last egg was laid. Nestlings usually leave the nest 12 or 13 days after hatching. Fledglings remain nutritionally dependent for at least two months after leaving the nest. Usually all group members feed the nestlings (and fledglings) but some subordinates may refrain from doing so. Although most pairs have helpers, pairs without subordinates can also successfully raise offspring.
Field work
We studied a population of M. coronatus resident on Annie Creek and the Adcock River in the Mornington Wildlife Sanctuary (17 o 31’S 126 o 6’E), an area managed by the Australian Wildlife Conservancy.
Adult M. coronatus were captured with mist-nets and offspring in the nest, and all were marked with a numbered metal band and a unique combination of coloured leg bands for individual recognition, under permit from the Australian Bird and Bat Banding Scheme and the Western Australia Department of Conservation and Land Management. Basic morphological measures as well as a small blood sample from all birds were taken. Between 38 and 55 territories were continuously followed from July 2005 onwards, documenting individual survival, and changes in partnerships and group composition. Birds were censused approximately once a week, and during census breeder females were observed for signs of nesting. Laying-, hatching- and fledgling dates, clutch size, hatching success, number of nestlings and number of fledglings were determined from regular nest checks.
Overall, in this population around 20% of clutches produced fledglings. Nest failed during all stages, and this was mainly due to predation and flooding, but also (rarer) by hatching failure, disappearance of a breeder, nest collapse, nestlings dying, and brood parasitism. Partial hatch failure sometimes occurred, and sometimes there was partial brood loss (predation or birds removing dead nestlings).
29
30 Chapter 2
Radical loss of an extreme extra-pair mating system
Sjouke A. Kingma , Michelle L. Hall, Gernot Segelbacher & Anne Peters
BMC Ecology (2009): 9, 15
31
32 Abstract
Mating outside the pair-bond is surprisingly common in socially monogamous birds, but rates of extra-pair paternity (EPP) vary widely between species. Although differences in life-history and contemporary ecological factors may explain some interspecific variation, evolutionary forces driving extra-pair (EP) mating remain largely obscure. Also, since there is a large phylogenetic component to the frequency of EPP, evolutionary inertia may contribute substantially to observed EP mating patterns. However, the relative importance of plasticity and phylogenetic constraints on the incidence of EP mating remains largely unknown. We here demonstrate very low levels of EPP (4.4% of offspring) in the purple-crowned fairy-wren Malurus coronatus , a member of the genus with the highest known levels of EPP in birds. In addition, we show absence of the suite of distinctive behavioural and morphological adaptations associated with EP mating that characterise other fairy-wrens. Phylogenetic parsimony implies that these characteristics were lost in one speciation event. Nonetheless, many life-history and breeding parameters that are hypothesised to drive interspecific variation in EPP are not different in the purple-crowned fairy-wren compared to its promiscuous congeners. Such radical loss of an extreme EP mating system with all associated adaptations from a lineage of biologically very similar species indicates that evolutionary inertia does not necessarily constrain interspecific variation in EPP Moreover, if apparently minor interspecific differences regularly cause large differences in EPP, this may be one reason why the evolution of EP mating is still poorly understood.
33
34 Introduction
Social monogamy is the most common mating system in birds (Lack 1968). However, application of molecular tools revealed that most socially monogamous species engage in extra-pair (EP) mating (> 75% of studied species; Griffith et al. 2002, Westneat & Stewart 2003). Rates of extra-pair paternity (EPP) are highly variable between species, ranging from none to almost all broods containing EP offspring (Arnold & Owens 2002, Griffith et al. 2002, Petrie & Kempenaers 1998, Westneat & Stewart 2003). The evolution of EP mating remains puzzling, largely because the generally proposed potential costs (reduced male care) and benefits (increased genetic quality of offspring) remain controversial (Arnqvist & Kirkpatrick 2005, Griffith et al. 2002). When explaining variation among closely related species and between populations of the same species, EP mating rates are hypothesised to be a result of evolutionarily plastic responses to contemporary ecological factors. For example, lower breeding synchrony could reduce alternative mating opportunities and depleted genetic variation could diminish the genetic benefits of pursuing EP mating (Griffith et al. 2002). However, phylogenetic constraints on EP mating are important, with over 50% of the interspecific variation being explained at the level of families or orders (Arnold & Owens 2002, Griffith et al. 2002), suggesting that contemporary ecological factors may (sometimes) play a secondary role (Arnold & Owens 2002). Indeed, results of comparative studies may vary depending on whether phylogeny is taken into account (e.g., Stutchbury & Morton 1995 vs. Westneat & Sherman 1997). Nonetheless, the relative importance of phylogenetic constraints on one hand and plasticity on the other in determining interspecific variation in EPP rates remains largely unknown. Here we examine the evolutionary flexibility of EPP rates by studying a member of a genus of birds that display an unusual mating system with extensive behavioural and morphological adaptations specialised for EP mating. Australian fairy-wrens (genus: Malurus ) are considered the least faithful socially monogamous birds since all three members of the genus for which paternity has been analyzed show exceptionally high rates of EPP (see Table 1 and Rowley and Russell (1997), with up to 95% of nests containing at least one EP offspring (e.g., Mulder et al. 1994). EP mating appears to be under female control, involving targeted pre-dawn forays to the territory of a preferred male, as shown in superb fairy-wrens ( M. cyaneus ; Cockburn et al.
35 2009, Double & Cockburn 2000). Male fairy-wrens invest heavily in mating competition as is evident from behaviour as well as morphology. Before breeding, males moult into colourful breeding plumage (Rowley & Russell 1997) and develop unusually large testes and cloacal protuberances (the site of sperm storage), important for sperm competition (Rowe & Pruett-Jones 2006). Moreover, males engage in frequent courtship of EP females starting months prior to breeding and continuing all through the breeding season. EP courtship involves males intruding onto nearby territories followed by ritualised display of their bright breeding plumage during which a flower petal may be presented to the visited female. Territorial intrusions by extra- group males and petal displays have been described for eight of the nine Australian Malurus species (Rowley & Russell 1997), and therefore high EPP has generally been expected for all fairy-wrens (e.g., Rowe & Pruett-Jones 2006, Rowley & Russell 1997, Webster et al. 2004). In this study, we examine rate of EPP in purple-crowned fairy-wrens (Malurus coronatus ). M. coronatus are riparian specialists rarely seen more than 20 m from the watercourse (Rowley & Russell 1993a, 1997). Year-round, groups vigorously defend a stretch of the stream that serves as their exclusive, all-purpose territory (Hall & Peters 2008). The cooperatively- breeding mating system of M. coronatus appears similar to other Malurus (Rowley & Russell 1997), but predictions of high rates of EPP for the species (e.g., Rowe & Pruett-Jones 2006, Rowley & Russell 1997, Webster et al. 2004) may be premature. Although the species is less studied than most other fairy- wrens, extra-territorial display by males has never been observed and M. coronatus pairs coordinate song to form 'duets' (Hall & Peters 2009, Rowley & Russell 1997), a feature generally related to low rates of EPP (Hall 2004). Here, we aim to establish the relative importance of phylogenetic constraints and evolutionary plasticity in response to ecological factors and life history, as determinants of extra-pair mating in M. coronatus . We quantify EPP rate and investigate male behavioural and morphological adaptations known to be important for EP advertisement and mating competition in other Malurus species. Additionally, we compare life-history and ecology of purple-crowned fairy-wrens with the other three species with known levels of EPP (superb, splendid M. splendens , and red-backed M. melanocephalus fairy-wrens). We consider attributes that have been hypothesised to affect costs and benefits of, and constraints on, EP mating (reviewed in Arnold & Owens 2002, Griffith et al. 2002, Petrie & Kempenaers 1998, Westneat & Stewart 2003); in
36 particular: social mating system (Hasselquist & Sherman 2001) including number of helpers (Mulder et al. 1994, Webster et al. 2004); clutch size and nesting success (Arnold & Owens 2002); divorce and mortality rates (Arnold & Owens 2002, Cézilly & Nager 1995, Wink & Dyrcz 1993); incidence of incestuous pairings (Brooker et al. 1990, Tarvin et al. 2005); importance of paternal care (Arnold & Owens 2002, Hoi-Leitner et al. 1999, Møller 2000, Møller & Birkhead 1993, Møller & Cuervo 2000); and breeding synchrony and density (Gowaty & Bridges 1991, Stutchbury 1998a, b, Stutchbury & Morton 1995).
Methods
Study species
We studied a colour-banded population of M. coronatus resident along Annie Creek and the Adcock River in the Australian Wildlife Conservancy's Mornington Wildlife Sanctuary (S17° 31' E126° 6') in Western Australia. Like other fairy-wrens, M. coronatus are small (~10-12 g), sedentary, cooperatively breeding passerine birds. The dominant pair, the male and female that sing duets (Hall & Peters 2009, Rowley & Russell 1997), form exclusive long-term pair bonds, and breed together. Subordinate birds are usually progeny from previous broods, and contribute to nestling feeding (Rowley & Russell 1997). During regular weekly censuses in August to November 2005 and April 2006 to April 2008, we noted which group members were present and searched for nest-building females, identifying any intruders, and all interactions, including display behaviour. Nests were checked during incubation to determine clutch size. At time of banding, a small blood sample was collected by brachial veni-puncture from nestlings (n = 164) and fledglings that were still dependent on their parents (n = 48), and stored in Queens- or Longmire's lysis buffer for paternity analyses (see below). In addition, material from unhatched eggs (n = 8) and dead nestlings (n = 7) was collected and stored in ethanol. We collected paternity data throughout the study period and area, covering the entire range of breeding synchrony and population density. For the interspecific comparison with other Malurus (see below), we collected the following breeding parameters for M. coronatus , adhering to methods published for the other species (see also below). We measured cloacal protuberance (CP) height (h), width (w) and length (l; measured as the
37 distance from the anterior to the posterior edge) to the nearest 0.1 mm of males captured in full breeding plumage and calculated volume following (Briskie 1993, Tuttle et al. 1996). Average clutch size and annual number of broods raised to fledging was calculated over the period April 2006-April 2008. We calculated average number of helpers using data from 39 territories in October 2007. Divorce rates were calculated annually (from 1 April) as the percentage of pairs that had switched mates twelve months later, while both individuals survived (see also below). Actual annual divorce rates may be higher than estimated by this method because it does not correct for pair bond duration and is based on pairing at certain arbitrarily chosen times (for detailed discussion see Cockburn et al. 2003). However, this commonly used metric provides a comparable index of divorce rates. Annual adult mortality rates were calculated as the percentage of dominant birds that were not present in our study populations one year later. We can assume these individuals died, as surveys in adjacent areas revealed that (long-distance) dispersal by dominant birds is rare (Hall ML, Kingma SA, Peters A, unpublished data). To calculate % incestuous pairings (parent-offspring and full-sib pairings) we included all pairs for which we knew the relatedness of the dominant male and female (n = 55). Incestuous pairings do not necessarily result in inbreeding because they frequently end in divorce (see Figure 4 in Cockburn et al. 2008 for M. cyaneus ). Nevertheless, the occurrence of incestuous pairings does indicate potential for inbreeding which could be avoided by EP mating. We collected nestling food provisioning rate of males without helpers (n = 34 nests of 25 males) by observing nests (with 1-4 nestlings between 4 and 10 days old) for 60 minutes and calculated feeding rates as number of provisioning trips per hour (Kingma et al. 2010, 2011b; chapter 4 and 5). Breeding synchrony was based on all broods over the study period and calculated using the formula (Kempenaers 1993, Stutchbury et al. 1998):
tp ∑ fi, p i=1 SI p = ×100 − tp(F 1) where SIp is the synchrony index for each female p (higher % means more females breeding in the population), fi,p the number of fertile females
38 (excluding female p) in the population on day i, and tp the number of fertile days for female p, (defined as 6 days before the start of laying till the day of the penultimate egg), and F the total number of dominant females present during female p's fertile period. For every day when at least one fertile female was present, the average for every fertile female was calculated. Purple-crowned fairy-wrens are riparian specialists with territories linearly arranged along the stream (see illustration in Rowley & Russell 1997 and Box 4 in chapter 1), so territory size was approximated as the length of the stream occupied by the group determined by behavioural observation, based on GPS coordinates, in October 2007.
Comparing Malurus species
Malurus phylogeny We used a phylogeny based on allozyme data (Christidis & Schodde 1997) showing three main clades (see Fig. 2.1), with M. coronatus as sister species to M. splendens and M. cyaneus all in one clade, confirmed by recent DNA analysis (Gardner J, Trueman J, personal communication). We included only the nine Australian Malurus species in this study, because there is very little information available about the three Papua New Guinean Malurus species (see Rowley & Russell 1997).
Extra-pair paternity Published EPP data are available for M. cyaneus, M. splendens , and M. melanocephalus (see Table 2.1). For M. splendens , we present rates of EPP for two different populations (Brooker et al. 1990, Webster et al. 2004), for M. melanocephalus we present two estimates from the same population (Karubian 2002, Webster et al. 2008), and for M. cyaneus we report the range of rates of EPP from one population collected over 15 years (Cockburn et al. 2003, Double & Cockburn 2000, Dunn & Cockburn 1996, 1999, Green et al. 2000, Mulder et al. 1994). Since extra-group (rather than extra-pair) paternity was usually reported in M. cyaneus , we calculated rates of EPP as the percentage of offspring sired by a male outside the group, and added 4.9% within-group offspring sired by subordinate males, as reported in Dunn & Cockburn (1999). Since the percentage of broods in which subordinate males sired offspring has not been reported, the value presented is % broods with extra-group young, and may hence be a slight underestimation of the %
39 broods with EP young. The percentage of offspring sired by within-group subordinates was not reported for M. melanocephalus , but like in other fairy- wrens, few offspring are sired by within-group subordinates (Webster MS, Karubian J, personal communication).
Reproductive organs and breeding plumage For the comparison of relative testis size in all Malurus (including M. coronatus ) we used the relevant specimen data used by Dunn et al. (2001). Their standardised protocol for comparative analysis included at least five breeding males for each species, and used only individuals with enlarged testes for tropical species with variable breeding seasons (for details see Dunn et al. 2001, Pitcher et al. 2005). We compiled data on CP size from original sources (Karubian 2002, Mulder & Cockburn 1993, Pruett-Jones & Tarvin 2001, Russell & Rowley 2000, Tuttle et al. 1996). To make data on M. coronatus comparable with data on other species, we omitted inclusion of data on males that were not in breeding condition, as could be indicated by non- breeding plumage (e.g., in subsets in Karubian 2002, Pruett-Jones & Tarvin 2001). Similarly, we excluded a study in which CP length was measured to the cloacal vent instead of to the anterior edge (Rowe et al. 2008), leading to smaller CP sizes. Males in all fairy-wren species are seasonally dichromatic, alternating dull non-breeding plumage with bright-coloured breeding plumage for several months of the year. Breeding plumage cover (% of the body with seasonally dimorphic plumage) of males of each species was estimated based on illustrations in Rowley & Russell (1997) and Higgins et al. (2001). Using The Gimp 2.2 (http://www.gimp.org), we manually selected the area covered by males' seasonal dimorphic plumage on the drawings and calculated the number of pixels (using a 'histogram'). A similar procedure was used to calculate number of pixels of the total bird and the percentage breeding plumage was calculated by: (number of pixels of seasonally dimorphic plumage/number of pixels total bird) * 100. Tail-feathers were not included. The average of three measures was used to determine mean percentage breeding plumage. The percentage presented in Fig. 2.1d, was obtained by averaging the two mean values based on both sources (Higgins et al. 2001, Rowley & Russell 1997). Repeatability for three measures of the remaining seven species was high (Lessels & Boag 1987) for the 3 repeated estimates within each source (Rowley & Russell 1997: repeatability = 99.5, F = 617.9, p
40 < 0.001; Higgins et al. 2001: repeatability = 99.9, F = 2555.2, p < 0.001) as well as for the 2 averages between sources (repeatability = 92.3, F = 25.0, p < 0.001); since breeding plumage covers 100% of the body in two species ( M. melanocephalus and M. leucopterus ), they were not included in the repeatability calculations.
Breeding biology Data about general breeding biology, life-history and ecology of all Malurus species for which EPP data are available (Table 2.1) was, if possible, acquired from the same studies reporting EPP (Brooker et al. 1990, Cockburn et al. 2003, Double & Cockburn 2000, Dunn & Cockburn 1996, 1999, Green et al. 2000, Karubian 2002, Mulder et al. 1994, Webster et al. 2004, 2008) and from a comprehensive review (Rowley & Russell 1997). The following data were not available from these sources and were obtained from other studies, where possible from the same population in which EPP was studied: group- (Karubian 2008) and territory-size in M. melanocephalus and M. splendens (Chan & Augusteyn 2003, Tibbetts & Pruett-Jones 1999), % inbreeding in M. cyaneus (Cockburn et al. 2003), number of annually raised broods (Cockburn et al. 2008, Rowley et al. 1991) and annual divorce rates in M. cyaneus and M. splendens (Mulder & Magrath 1994, Russell & Rowley 1993). Divorce rates in M. melanocephalus are low, but not yet quantified (Webster MS, Karubian J, personal communication). Comparable data on nestling feeding rate was only available for dominant M. cyaneus males without helpers (Green et al. 1995, Peters et al. 2002). We calculated breeding synchrony in M. cyaneus from original data on 87 females from the 1996 and 1997 breeding season (see Peters et al. 2001), following the same procedure as for M. coronatus (see above). Average territory length (diameter) in other fairy-wrens was calculated from area (in ha.) assuming a circular shape, to compare with territory size in M. coronatus .
Paternity analyses in M. coronatus
DNA Extraction and genotyping Total genomic DNA was extracted from blood samples of dominant and subordinate birds, offspring, and tissue samples of eggs using standard salt- extraction described in Richardson et al. (2001). The samples were genotyped by Ecogenics GmbH (Zurich, Switzerland) using a set of six microsatellite
41 loci, which were previously used for paternity analyses in other fairy-wrens (Mcy 1, Mcy 3, Mcy 4, Mcy 8, developed for M. cyaneus (Double et al. 1997), and Msp4, Msp6, developed for M. splendens (Webster et al. 2004); see references for genbank numbers). Four microsatellite loci were included in a PCR multiplex (Mcy 1, Mcy 3, Mcy 4, Msp6 with fluorescently labeled reverse primers). The other two loci (Mcy 8 and Msp4) were used in single PCR reactions. PCR amplifications were optimised for a 10 l reaction volume containing 2 l of DNA, 5 l master mix (Qiagen, Cat. No 206143 for multiplex and Cat. No 203445 for single PCR; containing Hotstar polymerase, PCR buffer, and dNTPs), 1.5 l double distilled water, and 0.3 M of forward and reverse primers each. The following thermo treatment was used on a TC- 412 Programmable Thermal Controller (Techne): 35 cycles with 94°C for 30 seconds, 50°C for 90 seconds, and 72°C for 60 seconds. Before the first cycle, a prolonged denaturation step (95°C for 15 min) was included to activate the Hotstar enzyme, and the last cycle was followed by a 30 min extension at 60°C. Genotyping was performed on an ABI PRISM 3100 Genetic Analyzer. The amplified PCR products (1.2 l) were mixed with 10 l formamide containing GENESCAN-500 (LIZ) Size Standard (Applied Biosystems), and the genotype was determined on an ABI PRISM ® 3100 Genetic Analyzer using GeneScan Analysis ® Software 3.7 and Genotyper ® 3.7 Software (Applied Biosystems). All 346 individuals included in the paternity analysis had four (n = 3), five (n = 25) or all six (n = 318) loci typed.
Determination of parentage In total, we genotyped 227 offspring from 104 broods. We used CERVUS v 3.0 software (Kalinowski et al. 2007, Marshall et al. 1998) to analyze paternity data. The expected (He) and observed (Ho) heterozygosity were calculated for each locus. We calculated heterozygosity (and parental exclusion probability, see below) using 137 dominant birds only, because genotypes of nestlings and subordinate birds (in most cases offspring from previous broods) were not independent. None of the loci deviated from Hardy-Weinberg equilibrium and we did not find evidence for null alleles. Number of alleles ranged from 2 to 17 per locus and heterozygosity was high (Table 2.2). We calculated for each locus the probability of maternal and paternal exclusion, i.e., the probability of
42 exclusion of a randomly chosen male or female as parent of the offspring, based on allele frequencies of dominant birds. In total across all 6 loci, we had 50 alleles and the probability of wrongly assigning a randomly chosen male as the sire was less than 1% (Table 2.2). For paternity analyses, we assumed the social mother to be the genetic mother of the offspring. This assumption was justified since social mother and offspring had no mismatches except for one mismatch at a single locus, probably due to a scoring error (as suggested by homozygosity at the maternal locus). We assigned paternity using the following conservatively chosen parameters in CERVUS: number of candidate fathers = 20, proportion of males sampled = 0.9083 and frequency of typing error = 0.01. First, we tested whether the social (expected) father was likely the genetic father, by examining whether there were mismatches between the social father and offspring. In total 217 of the 227 offspring matched all paternal alleles, and were assigned as true sire with confidence of > 95% in all cases. Of the ten remaining offspring (from 6 nests, 4 nests with one, and 2 nests with three offspring not sired by the dominant male) with a paternal mismatch, one had a mismatch at one allele, six at 2 alleles, two at 3 alleles and one at 4 alleles. In the single case of one locus mismatch, it was likely that a subordinate male from the same group was the sire of the offspring, rather than a mutation or scoring error, because the subordinate male had no mismatches. Three other EP-offspring could be assigned to subordinate males from the same social group as the dominant male (in total one subordinate male sired one offspring, and one subordinate male sired three); both subordinates were unrelated to the dominant female. For the other six offspring not sired by their social father, we identified five dominant and one subordinate male from nearby groups as the genetic fathers (these matched the offspring's paternal alleles at all loci). Thus, in total, out of 227 offspring and 104 broods, we identified six extra-group offspring in four broods, and four instances of paternity by a subordinate male within the group in two broods, and thus ten EP offspring in six broods. Paternity assignment may be particularly complex in cooperative species with high male philopatry, due to the presence of close relatives, which share some alleles with the expected father (Double et al. 1997, Richardson et al. 2001). To verify our assignment of the social male as the sire of the 217 offspring that had no mismatches with their social father (thus excluding the 10 EP offspring), we used a likelihood-based approach.
43 CERVUS calculates the likelihood for each male that he is the sire (LOD score) based on the offspring and maternal genotypes; the male with the highest LOD score is the most likely sire. We included all males (of at least 3 months old) within five territories as potential fathers and computed LOD scores for each candidate father. Using this method, we could confirm the social father to be the sire in 201 cases (he had the highest LOD score of all candidate males). For the remaining 16 cases, one (n = 12) or two (n = 4) other males showed higher LOD scores than the social father (despite no mismatches between offspring and social father). We estimated the probability of falsely accepting a social father as sire as Pep_offspring × Pep_mate = 0.27% (if one male had higher LOD) and 0.54% (if two males had higher
LOD). Pep_offspring is the unequivocal probability that a given offspring is not sired by its social father = 4.7% (10 EP offspring and 201 within-pair offspring unambiguously assigned by allelic exclusion as well as LOD scores of social sires, see above). Pep_mate is the probability that the female would have mated EP with the one or two males in the population that had higher
LOD than the social father by chance. We estimated Pep_mate by considering as potential sires all males older than 3 months within five territories distance from the focal territory, including the subordinate males within the focal territory, which were 17 alternative males in this subset. Thus, Pep_mate was on average 5.9% (for one male with higher LOD) and 11.8% (for two males with higher LOD) respectively.
Results
Extra-pair paternity
In sharp contrast to the Malurus species studied so far (see Table 2.1), paternity analysis of 227 offspring from 104 nests in M. coronatus revealed very low levels of EPP: 4.4% of offspring (95% confidence interval CI: 1.7- 7.1%) in 5.8% of broods (CI: 1.2-10.3%) resulted from extra-pair matings. These 10 EP offspring in 6 broods included 4 offspring sired by a subordinate male in 2 groups, thus corresponding to 6 extra-group offspring (2.6%, CI: 0.5- 4.7%) in 4 broods (3.8%; CI: 0.1-7.6%). Paternity analysis was based on 6 microsatellite loci (2 to 17 alleles per locus) with a maternal exclusion probability of 94.2% and a paternal exclusion probability of 99.1% (Table 2.2). Microsatellite heterozygosity was high (range: 0.467-0.917; see Table 2.2).
44
Table 2.1. A comparison of rates of extra-pair paternity, and attributes of life-history, ecology and breeding biology in four Malurus species for which rates of extra-pair paternity are known.
M. coronatus M. cyaneus M. splendens M. melanocephalus Extra-pair paternity % Broods containing EP offspring 5.8 83-95 55, 83 75, 63
% EP offspring 4.4 66-81 42, 73 56, 51
% Offspring sired by within-group subordinate 1.8 5 5, 8 some General biology
Number of helpers 1.3 0.9 0.6, 1.1 0.4
Number of male helpers 0.9 0.9 0.3, 0.6 0.4
Clutch size 3.0 3.2 2.9 2.8
Annual max # successful broods 2 3 2 > 1 % Annual pair-divorce 6.0 4.2 2.9 low
% Annual adult survival 75, 53; (male, female) 81, 81 67, 70 70, 59
% Incestuous pairings 16.4 15 21.3
Provisioning rate of males without helpers (feeds/h) 6.2 5.8, 6.4
All, Months breeding 1-2 peak(s) Aug-Feb Aug-Jan Oct-Feb
Average breeding synchrony (max) 11.5 (28.2) 18.1 (29.8)
Territory linearly arranged Yes No No No
Territory length in m 156 87 234, 237 262
45 Table 2.2. Number of alleles, observed (Ho) and expected (He) heterozygosity and exclusion probabilities of microsatellite loci used for paternity analyses in Malurus coronatus , based on 137 dominant birds.
Prob. of maternal Prob. of paternal Locus No of alleles Ho He exclusion exclusion Mcy 1 6 0.796 0.734 0.326 0.502 Mcy 3 9 0.810 0.830 0.494 0.666 Mcy 4 2 0.496 0.500 0.123 0.187 Mcy 8 17 0.917 0.912 0.688 0.816 Msp4 9 0.467 0.478 0.127 0.290 Msp6 7 0.684 0.704 0.292 0.469 Total 50 0.9423 0.9906
Adaptations to extra-pair mating: morphology and behaviour
In M. coronatus, we found none of the male morphological adaptations for EP mating that are characteristic of other members of the genus (Fig. 2.1). M. coronatus has much smaller reproductive organs during breeding (Fig. 2.1a, b): their testes are on average 10% of the size of testes of other Malurus (range: 6-18%; Fig. 2.1b) and cloacal protuberances (CP; the site of sperm storage) in breeding males (57.4 ± 3.6 mm 3, n = 40) are on average about half of the CP size of other fairy-wrens (28-87%; Fig. 2.1a). A subset of males captured when their female was fertile (in the week before egg-laying) had an average CP of 66.3 ± 4.2 mm 3 (n = 25), a peak still much smaller than average breeding CP size in other Malurus (see Fig. 2.1a). Interestingly, M. lamberti and M. pulcherrimus appear to have rather small cloacal protuberances (see Fig. 2.1a; Rowe et al. 2008), although all other typical EP mating adaptations have been recorded (Fig. 2.1). It would therefore be interesting to explore EPP rates in these two species.
46 Figure 2.1. Comparison of extra-pair mating adaptations in males of the nine Australian Malurus species. For each species we illustrate (a) mean (± SE) size of the cloacal protuberance (CP) of breeding males (SE in M. cyaneus and data on CP size in M. pulcherrimus and M. amabilis were not available (n.a.); numbers depict sample sizes), (b) mean testes size (as % of body mass), (c) male breeding plumage cover (% of the body seasonally covered by colourful plumage), and (d) whether male extra-territorial forays, display behaviour and petal presentations to extra-pair females, have been described (ticks) or not (crosses). Data on M. coronatus (dark bars) are from our study, except for testes size (Dunn et al. 2001), and references for other data are given in the 'Methods' section. The phylogeny was obtained from Christidis & Schodde (1997).
47 Comparing Malurus life-history and ecology
We compared M. coronatus with three congeners that have high EPP, concerning aspects of general breeding biology, life-history and ecology that have been proposed or shown to affect relative costs and benefits of and constraints on extra-pair mating (Table 2.1). This comparison showed that social mating system, number of helpers, clutch size, breeding success, rates of divorce, adult mortality, and incestuous pairing, rates of paternal care, and peak breeding synchrony are all very similar (Table 2.1). All fairy-wrens are cooperative breeders, with subordinate birds assisting the dominant (breeding) pair with offspring provisioning (Table 2.1). Groups average 3.3 birds (range 2-11; n = 39) in M. coronatus , including male and female subordinate birds, with sex of subordinates being skewed towards males (around two third of subordinates were males). This results in approximately the same number of males present per territory compared to M. cyaneus (both 1.9 male per territory), and slightly more than in M. splendens (1.3-1.6) and M. melanocephalus (1.4; Table 2.1). Clutch size was 3.0 ± 0.1 eggs per clutch on average (range: 1-4, n = 157 nests) intermediate to other fairy-wrens (range 2.8-3.2 eggs; Table 2.1). Like other fairy-wrens, M. coronatus can raise more than one successful brood annually (Table 2.1; Cockburn et al. 2008, Rowley et al. 1986). Pair bonds in all Malurus species are long-lasting for two reasons. First, divorce rates in Malurus are relatively low (Table 2.1). In our population, three of 25 surviving pairs from 1 April 2006 and none of 24 pairs from 1 April 2007 divorced (average annual divorce rate: 6.0%). This is similar to other Malurus species: in M. cyaneus 4.2% of 93 pairs surviving to one year later had divorced (Mulder & Magrath 1994), and in M. splendens , 10 pairs divorced in 348 pair-years (2.9%; Russell & Rowley 1993). Second, mortality rates in Malurus are generally low (Rowley & Russell 1997). In our M. coronatus population on average 19% of dominant males and females did not survive until one year later (Hall ML, Kingma SA, Peters A, personal oberavtion) resulting in comparable to slightly higher survival rates than in other Malurus species. We observed nine incestuous pairings of 55 pairings with known pedigree (16.4%; five mother-son, two father-daughter, and two full-sib pairings). This is similar to the potential for inbreeding in M. cyaneus and M. splendens (Table 2.1), where incestuous pairings occurred in 75 out of 500
48 pairings (15.0%; Cockburn et al. 2003) and 29 out of 136 pairings (21.3%; Rowley et al. 1986) respectively. Nestling feeding rates are similar in M. coronatus compared to M. cyaneus , the only congener with data available for comparison (Table 2.1). In M. coronatus , males in unassisted pairs have a provisioning rate of 6.2 feeds/hour (n = 25 males at 34 nests), which is intermediate between two measures from one population of M. cyaneus (5.8 and 6.4 feeds/hour; Green et al. 1995, Peters et al. 2002). The average breeding synchrony index of M. coronatus (based on 252 breeding attempts) was 36% lower than in M. cyaneus (Table 2.1), the only congener available for this comparison. This is because breeding in M. coronatus could occur in any month. However, there were typically peaks in spring (August - September) and in the wet-season (December - April), so peak synchrony was almost identical in both species (28.2 and 29.8% respectively). Breeding density is determined by territory size and arrangement. Territory size in M. coronatus (average length = 156.5 ± 14.3 m, range: 50- 376, n = 39 territories) is intermediate to other fairy-wrens (Table 2.1). However, in contrast to other fairy-wrens which have territories arranged in a mosaic (Table 2.1), territories in M. coronatus are linearly arranged along the stream and have therefore fewer neighbours.
Discussion
EPP is a defining feature of the mating system of Australian fairy-wrens (genus Malurus ), associated with striking behavioural and morphological adaptations. Although high infidelity has been assumed to occur in all fairy- wrens (e.g., Rowe & Pruett-Jones 2006, Rowley & Russell 1997, Webster et al. 2004), we here show a marked exception: M. coronatus' mating system rather approaches genetic monogamy, with EPP rates being even lower than the average 18.7% of broods observed in socially monogamous birds (Griffith et al. 2002). Although levels of EPP in Malurus fluctuate between species as well as within species (e.g., in M. splendens : between populations 55-83% (Webster et al. 2004, Brooker et al. 1990); between years 33-63% (Webster et al. 2004) of nests), all studied species show high levels of promiscuity, with even the lowest values of EPP being far higher than that of M. coronatus (Table 2.1).
49 Contrary to all other fairy-wrens, M. coronatus males show reduced sexual signalling (reduced breeding plumage and the absence of courtship display), in line with lower levels of EPP (Owens & Hartley 1998). Moreover, while other Malurus have some of the largest testes and cloacal protuberances known for birds (Rowe & Pruett-Jones 2006), M. coronatus has relatively small reproductive organs, suggestive of reduced sperm- competition (Briskie 1993, Pitcher et al. 2005). In fact, their testes are well below the average 1% of body mass for songbirds (Pitcher et al. 2005) and cloacal protuberances of breeding males are also smaller than expected for their body size (Briskie 1993). A phylogeny including 9 of 12 species in the Malurus genus (Christidis & Schodde 1997) suggests that M. coronatus is derived from an ancestor with high EPP and the adaptations to EP mating that characterise its sister species (Fig. 2.1). Thus, phylogenetic parsimony suggests that EP mating and the entire suite of adaptations were lost within a single speciation event (see also Box 1 in Wiens (2001) for an example).
Comparing Malurus life-history and ecology
Hypotheses attempting to explain interspecific variation in EPP usually consider differences in social mating system or contemporary ecology as important explanations that can affect potential benefits, costs and constraints of EP mating (Griffith et al. 2002). However, the radical loss of extreme levels of EPP in M. coronatus is not associated with any marked differences in relevant life-history parameters. Clutch size could affect EPP rates (Arnold & Owens 2002), but average clutch size is very close to on average around 3 eggs in all species (Table 2.1), including M. coronatus . Similarly, differences in annual fecundity could affect rates of EPP (Arnold & Owens 2002), but all Malurus species can annually raise more than one successful brood (Table 2.1). Potential (genetic) benefits of EPP and EPP rates can be dramatically affected by differences in mating system and opportunity to express social mate choice (Dixon et al. 1994, Hartley et al. 1993, see also Hasselquist & Sherman 2001). However, all fairy-wrens are socially monogamous with restricted opportunities for social mate choice since breeding vacancies arise rarely in these long-lived, year-round territorial species with low divorce rates (see Rowley & Russell 1997, and Table 2.1). Low rates of divorce and mortality in M. coronatus are associated with a low EPP rate, consistent with
50 the pattern found in comparative studies (Arnold & Owens 2002, Cézilly & Nager 1995, Wink & Dyrcz 1999). While divorce rates are consistently low in all Malurus species, mortality rates are slightly lower in M. coronatus than in its sister species (Table 2.1). However, mortality rates are in general very low in the (promiscuous) Malurus species (see Figure 2 in Arnold & Owens 2002), and, in fact, much lower than would be predicted from their high EPP rates (Arnold & Owens 2002). Thus, it seems unlikely that a relatively slight decrease in adult mortality was important for the near-disappearance of EP mating in M. coronatus . EP mating can act as a mechanism to reduce the costs of incest and inbreeding (Tarvin et al. 2005). Philopatry increases the risk of incest and inbreeding in general, as it does in other fairy-wrens (Brooker et al. 1990, Cockburn et al. 2003, 2008, Cockburn & Double 2008, Tarvin et al. 2005). The frequency of incestuous pairings in purple-crowned fairy-wrens does not differ compared to the sister species (Table 2.1), including M. splendens for which inbreeding avoidance has been proposed as one of the benefits of EPP (Brooker et al. 1990, Tarvin et al. 2005). Therefore, in combination with the fact that molecular genetic variability is rather high (see Table 2.2), opportunities for genetic benefits of EP mating seem as likely in M. coronatus as in other fairy-wrens. Some life-history attributes can constrain females in EP mating behaviour. For example, the importance and extent of male care can affect rates of EPP (Arnold & Owens 2002, Møller & Cuervo 2000). All fairy-wrens are cooperative breeders with, on average 1-2 subordinates (Table 2.1) assisting the dominant (breeding) pair. In M. cyaneus , helpers liberate females from constraints on EP mating by compensating for the reduction in male care associated with EP behaviours (Mulder et al. 1994). Likewise, in M. coronatus , parents can decrease their feeding rate of nestlings when subordinates are present (Kingma et al. 2010; chapter 5). Increased value of male care in M. coronatus compared to other Malurus seems unlikely since nestling starvation is uncommon (Hall ML, Kingma SA, Peters A, personal observation), and nestling feeding rates of males without helpers are similar in M. coronatus compared to M. cyaneus (Table 2.1). In addition, male paternity-guarding behaviour might limit EPP (Birkhead & Møller 1998). However, mate guarding is no more intense in M. coronatus than in M. cyaneus (see Hall & Peters 2009, for a detailed discussion), and in M. cyaneus
51 male mate-guarding does not prevent EP mating by females (Double & Cockburn 2000, Mulder 1997). Availability of suitable EP mating partners and hence EP mating opportunities might be reduced by asynchronous breeding (Stutchbury 1998a) and low density (Westneat & Sherman 1997). Although M. coronatus breeds more irregularly and has a longer breeding season than other Malurus species, peak synchrony in M. coronatus is similar to peak synchrony in M. cyaneus (Table 2.1), its sister species that has the highest rates of EPP reported in birds (Griffith et al. 2002). Breeding density (territory size) at first appears similar. However, spatial arrangement of territories is different in purple-crowned fairy-wrens (see below). It is difficult to demonstrate causation for a single evolutionary event, but a possibility is that the riparian specialisation of M. coronatus , resulting in a one-dimensional arrangement of territories, may constrain access to an adequate sample of EP mates. Extra-pair offspring in M. cyaneus are generally sired by males within one or two territories (maximally five; Double & Cockburn 2000). Territory length in M. coronatus is intermediate between the other fairy-wrens and the number of males per territory is similar (see Table 2.1). Therefore, if female purple-crowned fairy-wrens traveled similar distances in search of EP males, this linear arrangement implies that a female could choose from fewer potential sires. For example, for average group size and composition of M. coronatus and M. cyaneus (Table 2.1), within a two-territory radius there would be 7.6 males in 4 linearly arranged territories in M. coronatus , compared to 34.2 males in 18 territories in a honeycomb arrangement in M. cyaneus (see Figures in Cockburn et al. 2009, Rowley 1981, Rowley & Russell 1993a). Male advertisement would be subject to analogous constraints since male M. coronatus would have fewer females to display to within a similar flight distance. However, inter-specific evidence for breeding density effects remains poor (Arnold & Owens 2002, Bennett & Owens 2002, Griffith et al. 2002, Westneat & Sherman 1997,). Moreover although extremely low breeding density will necessarily limit EP mating (cf. Bjørnstad & Lifjeld 1997 and Gyllensten et al. 1990 for an example), it is not immediately obvious why a relatively reduced availability of EP mates should lead to virtual female fidelity and absence of male EP mating adaptations in M. coronatus . However, this hypothesis could be explored by examining whether differences in EP mate availability (population density) explain some
52 of the variation in Malurus' rates of EPP between study years, populations and species.
Conclusion
Although EP mating can have a strong phylogenetic component (Arnold & Owens 2002, Griffith et al. 2002), our results show that remarkable differences in EP mating system and associated adaptations can occur among species with very similar life-histories. This pattern suggests that phylogenetic constraints do not necessarily limit rapid evolutionary modification of EPP rate, and that EP mating and its adaptations can be evolutionary much more labile than is commonly assumed (Arnold & Owens 2002, Griffith et al. 2002). If inter-specific variation in EPP levels is commonly not related to descent, social mating system, or obvious contemporary ecological factors, this could well be one reason why general explanations for the evolution of EPP have remained elusive (Griffith et al. 2002, Petrie & Kempenaers 1998, Westneat & Stewart 2003).
Acknowledgements
We are grateful to S. Legge, S. Murphy and other staff at Mornington Wildlife Sanctuary, and our excellent team of field-assistants. We thank M. Double and M.S. Webster for testing their microsatellite primers on our M. coronatus samples, M.S. Webster, J. Karubian and J.L. Gardner for providing unpublished data, and P.O. Dunn for providing us with the comparative data on Malurus testes size. K. Delhey, M.S. Webster, B. Kempenaers, A. Cockburn, and anonymous reviewers provided helpful comments on an earlier version the manuscript. This research was funded by the 'Sonderprogramm zur Förderung hervorragender Wissenschaftlerinnen' of the Max Planck Society (to AP). All fieldwork was performed with permission from the Max Planck Institute for Ornithology Animal Ethics Committee, the Australian Wildlife Conservancy, the Western Australia Department of Conservation and Land Management (licence BB002178), and the Australian Bird and Bat Banding Scheme (authority 2073).
53
54 Chapter 3
Female fairy-wrens synchronise egg laying to facilitate extra-pair mating for inbreeding avoidance
Sjouke A. Kingma, Michelle L. Hall & Anne Peters
55
56 Abstract
Inbreeding has negative effects on reproductive performance in many animal species. Therefore, individuals are predicted to adopt strategies to avoid inbreeding. Inbreeding avoidance has become one popularly proposed benefit behind the evolution of multiple mating by females, although evidence for this hypothesis is mixed. In purple-crowned fairy-wrens ( Malurus coronatus ) overall levels of extra-group paternity (EGP) are low (6% of broods contain EG offspring), probably as a consequence of low breeding density and synchrony, and males are morphologically and behaviourally poorly adapted to EP mating. Nonetheless, we show that related pairs have EG offspring in 50% of broods, which is about 15 times higher than in broods of unrelated pairs (3% of broods with EGP). As hatching success of eggs is negatively correlated with genetic similarity of the parents, reduced by more than 20% in clutches produced by females paired with first-order relatives, this suggests that EG mating has partly evolved to avoid the negative effects of inbreeding in this species. We suggest that incestuously-paired females adopt an active strategy to overcome the constraints of EG mating: compared to females in unrelated pairs, females in incestuous pairs were significantly more likely to synchronise their breeding attempt with breeding of neighbors. Probably as a consequence, females in incestuous pairs mated EG with directly neighboring males whose social partner was around egg laying, whereas promiscuous females with an unrelated partner mated with less synchronously breeding EG males from further away. These results reveal that negative effects of inbreeding can be a sufficient selection pressure for the evolution of EP mating, and suggest that synchronisation can be an active pre-copulatory strategy by females mating EP.
57
58 Introduction
Inbreeding has been shown to negatively affect reproductive performance in many animal species (see Amos et al. 2001, Charlesworth & Charlesworth 1987, Keller & Waller 2002, Kempenaers 2007). Such effects include reduction in offspring survival, immune-system and reproductive success, and derive most likely from the expression of recessive deleterious alleles in inbred offspring with increased levels of homozygosity (Keller & Waller 2002, Lynch et al. 1995). Because of these implications, inbreeding has received considerable attention (Charlesworth & Charlesworth 1987, Keller & Waller 2002, Lynch & Walsh 1998,), with a conspicuous focus on whether and how individuals can avoid it (Pusey & Wolf 1996). As such, inbreeding avoidance has, as part of the ‘genetic compatibility hypothesis’ (Tregenza & Wedell 2000, Zeh & Zeh 1996, 1997), become one popularly proposed adaptive benefit behind the evolution of extra-pair (EP) mating by females (e.g., Griffith & Immler 2009, Jennions & Petrie 2000, Kempenaers 2007, Westneat & Stewart 2003). Despite a number of examples (e.g, Bishop et al. 2007, Blomqvist et al. 2002, Brooker et al. 1990, Cohas et al. 2008, Eimes et al. 2005, Foerster et al. 2003, Freeman-Gallant et al. 2006, Rubenstein 2007a, Suter et al. 2007) however, there is large interspecific variation in the role of inbreeding avoidance in explaining EP mating (see Akçay & Roughgarden 2007, Kempenaers 2007, Mays et al. 2008). Variation in the applicability of the inbreeding avoidance hypothesis across species is perhaps not surprising, as the incidence of EP mating is determined by a complex interplay between benefits, cost, and constraints (Arnold & Owens 2002, Griffith et al. 2002). While the idea that EP mating can be beneficial when inbreeding is costly is well defined, other proposed benefits (e.g., good genes; Jennions & Petrie 2000), costs (e.g. reduced male care; Arnqvist & Kirkpatrick 2005) and constraints (see Arnold & Owens 2002, Griffith et al. 2002, Stutchbury 1998a, Westneat & Stewart 2003) of EP mating have poorly or not been taken into account in the discussion about EP mating as a mechanism of inbreeding avoidance (but see e.g., Arnqvist & Kirkpatrick 2005, Foerster et al. 2003, Neff & Pitcher 2005, Oh & Badyev 2006, Tarvin et al. 2005, that take into account costs or other benefits of EP mating). This is surprising, because for instance ecological variables, like breeding density and synchrony, have been shown to affect the incidence of EP mating within and across species (see Arnold & Owens 2002, Stutchbury
59 1998a, Westneat & Stewart 2003). Thus, females may be constrained in opportunities for EP copulations, and this may explain why in some species or individuals inbreeding is not avoided by EP mating. Additionally, other reasons for EP mating may apply simultaneously (see e.g., Foerster et al. 2003, Neff & Pitcher 2005, Oh & Badyev 2006, Tarvin et al. 2005), thereby potentially masking the function of EP mating in inbreeding avoidance. Thus, it is crucial to at the same time consider the suite of factors determining the costs, benefits and constraints of EP mating to understand whether and how inbreeding can be avoided by EP mating. Understanding whether EP mating by females evolved to avoid inbreeding is further complicated by the fact that the mechanisms of increased fertilisation success of more compatible EP males are still unclear. For instance, differential fertilisation success of less-related males could be the result of an active behavioural adaptation facilitating preferential mating by females, or be driven by post-copulatory ‘cryptic female choice’ through sperm competition or selection (e.g., Birkhead & Biggins 1998). These mechanisms likely differ in how they can be affected by -or affect - factors that constrain EP mating: preferential mating may be achieved by adopting strategies to target specific (unrelated) males, whereas cryptic female choice can be achieved through ‘a genetically loaded raffle’ after (potentially more opportunistic) multiple mating (Griffith & Immler 2009, but see Denk et al. 2005). The notion that females may adopt active strategies to overcome constraints on EP mating has received surprisingly little attention and is, as far as we are aware, restricted to theoretical examples. As part of explaining colonisation in birds, Wagner (1993) proposed that increasing breeding density may be a way in which females can facilitate additional mating opportunities. Similarly, Fishman & Stone (2005, 2006) predict that females can affect breeding synchrony to get access to fertile males. Despite the fact that understanding such strategies would have great implications for our understanding of causes and consequences of EP mating, it is unclear whether individuals do adopt strategies to overcome constraints of EP mating. Here, we determine the costs of inbreeding, and the benefits and constraints of EP mating in purple-crowned fairy-wrens Malurus coronatus . Due to limited dispersal and restricted mating opportunities, some individuals engage in incestuous pairings in this species (Kingma et al. 2009; chapter 2). Breeding density and synchrony are very low, as the birds breed
60 year-round in linearly arranged territories. Potentially as a consequence of such constraints, levels of EP paternity (EPP) in M. coronatus are low, and males are poorly adapted to sperm competition (Kingma et al. 2009; chapter 2). Therefore, we investigate whether females adopt strategies to overcome the constraints on EP mating, by testing specifically whether females (especially those in incestuous pairings that are predicted to benefit most from EP mating) synchronise their fertile period with fertility of neighboring pairs.
Methods
Study species and fieldwork
We studied a M. coronatus population, resident along Annie Creek and the Adcock River at Mornington Wildlife Sanctuary in North-west Australia (17°31’S, 126°6’ E) from July 2005 to September 2010. M. coronatus live in socially monogamous pairs (breeders) that may be assisted by a number of (male and female) subordinates (Kingma et al. 2010, 2011a, b, Rowley & Russell 1993a, 1997). The species is restricted to riparian vegetation and territories are fixed year-round and linearly arranged along creeks and rivers. The birds can breed year-round (see Rowley & Russell 1993a, 1997). During weekly population censuses, we documented group size, and social status (dominant or subordinate) of each uniquely colour-banded group member based on behavioural cues (the most obvious that only the dominant pair sing duets; Hall & Peters 2008, 2009). In addition, nesting activity was monitored year-round during weekly observations by following the female for at least 20-30 minutes, and nests were checked regularly to determine laying date and number of eggs and nestlings. All adult birds in the population were caught using mistnets and a small blood sample was taken by venipuncture of the brachial vein. Nestlings were banded when they were around 7 days old, and a small blood sample was taken similarly as in adults. Unhatched eggs and dead nestlings were collected. Blood and tissue samples were stored in ethanol, Queen’s lysis buffer or Longmire’s buffer. Coordinates of territory boundaries (as determined from movement of the birds) were taken using Garmin eTrex handheld GPS units, and occasional changes in boundaries were recorded throughout the study.
61 Paternity analyses
We genotyped all individuals in our population (from blood samples, or tissue samples from dead nestlings and unhatched eggs with visible development) applying PCR techniques using six microsatellite loci (see chapter 2 (Kingma et al. 2009) for details). We included all individuals with at least five loci successfully typed (98% of 759 individuals; 13 of the 15 individuals with less than five loci typed were tissue samples from unhatched eggs). Based on a subset of 183 birds (only birds that had a dominance position during the study period, to avoid overrepresentation of alleles derived from successful pairs), heterozygosity over six loci was on average 0.701 (range = 0.470- 0.928), with in total 53 alleles (mean = 8.83 alleles per locus, range = 3-18), so that the probability of falsely assigning a male as father was less than 1%. None of the loci deviated from Hardy-Weinberg equilibrium and there was no indication of null-alleles. For paternity analyses we assumed the social mother to be the genetic mother of the offspring, which was justified as only one of the 510 offspring mismatched the mother at one locus (which was assumed a mutation, as some of the offspring’s subsequent own offspring shared this allele). Subsequently, we tested whether the social father showed allelic mismatches with the offspring, given the genotype of the mother. Of the 510 offspring (from 217 broods), 480 entirely matched the social father, and these offspring were assumed to be sired by the within-pair male (see also Kingma et al. 2009; chapter 2). For the remaining 30 offspring with one or more mismatches, we subsequently aimed to identify the true sire. For one offspring, the single mismatch with the social father was an allele that was unknown in the population, and because we assumed this to be a mutation or scoring error, this offspring was considered sired by its social father. For the remaining 29 EP offspring (from 15 broods; 5.7% EP of 510 offspring in 6.9% of 217 broods), we tested whether one or more males within 10 territories distance could sire the offspring (the farthest identified EP sire was from six territories away). Eighteen of those offspring could unambiguously be assigned to one male, whereas for 11 offspring, we identified more than one potential sire whose genotype did not have mismatches with the genotype of the offspring (given the mother). These 11 offspring, their mothers and the potential sires were genotyped using three additional loci: Cu 28 (Gibbs et al. 1999), Pte24 and Pte26 (Blackmore et al. 2006). This way, the sire for eight additional offspring could be assigned. For one additional offspring that
62 did not show mismatches with two males, we assumed that the sire was the male that sired another offspring in the same brood. In the two remaining cases (both from broods with only one genotyped offspring), two males did not show mismatches with the offspring at all 9 loci. One of those was excluded from all analyses that require the location of the EP sire to be known. The other case was included as both potential sires were from the same territory. Four EP offspring (from two broods) were sired by a subordinate male within the same group (within-group EP: 0.78% of 510 offspring in 0.92% of 217 broods; see also Kingma et al. 2009; chapter 2). Those were excluded from the analyses, as we used extra-group (EG) paternity, only taking into account whether offspring were sired by males outside the group.
Relatedness
Pairwise relatedness between breeders was calculated using Genalex 6.4 (Peakall & Smouse 2006) based on the 183 dominant birds (see above), using the most commonly used estimators: LR (Lynch & Ritland 1999) and QG (Queller & Goodnight 1989). Before calculating relatedness, we reconstructed missing alleles of five of the seven individuals in which one microsatellite primer failed to work, based on a combination of pedigree and molecular data. Reliability of LR and QG relatedness estimators were compared using the program IRel (Gonçalves da Silva & Russello 2010), and the LR estimator was determined as the most suitable estimator of relatedness for our population, and therefore used for further analyses. For some analyses (see below) and for graphical representation, we categorised pairs as incestuous (relatedness: 0.323-0.660; the lowest observed relatedness of a pair of known first-order relatives was 0.323), and non- incestuous (relatedness < 0.323). Of the eight pairs that we classified as ‘incestuous’, the four with known pedigree were all pairings between first- order relatives.
Analyses
Statistical tests were performed using R 11.1 (R Development Core Team 2010). For most analyses, we used a mixed modeling approach, correcting for replicated observations of the same breeding pair by including pair-identity as a random variable (unless stated otherwise). Non-normally distributed response variables (binomial, proportion, quasi-poisson data) were analyzed
63 using generalised linear mixed models (GLMM) using the lme4 package in R (Bates & Sarkar 2007), whereas normally distributed response variables were fitted using maximum likelihood (ML) models in the nlme package in R (Pinheiro & Bates 2000; Pinheiro et al. 2009). Explanatory variables were included in the initial model and we rejected the null hypothesis when P < 0.05. Parameter estimates ( β) are given as mean ± standard error of the mean. Model residuals were visually checked for homoscedasticity and normality. The effect of pair relatedness on hatchability of 646 eggs in 216 broods from 85 pairs was tested using a GLMM with hatching success of broods as the response variable and pair relatedness as (continuous) explanatory variable. We corrected for brood size by binding the number of hatched eggs and unhatched eggs per brood using the ‘cbind’ command and a binomial error structure. We included only complete clutches: these are nests found during laying or incubation, and broods that were found at a later stage containing three or four offspring, which we considered complete broods (see Kingma et al. 2011a; chapter 4). We only included broods that had opportunity to hatch: broods that survived at least until a few days after hatching of the first egg (thus broods with eggs or eggs and very young hatchlings that flooded or that fully or partially disappeared were excluded), and complete broods that failed to hatch despite incubating beyond the full term (n = 9 broods). Because eggs from many broods were not genotyped (due to early stage predation and difficulty genotyping unhatched eggs), we could not correct for paternity. Since highly related pairs were more likely to have EG paternity in their broods (see results), this creates systematic bias against our hypothesis that eggs of closely related pairs have higher hatching failure (i.e., we underestimate the cost of inbreeding). We tested whether the incidence of EGP in 217 broods of 85 pairs was determined by relatedness of the social pair (LR estimate) and breeding density. We expressed breeding density as (1) the number of nearby territories (within 386 meters; the maximum distance that we observed between a sire his EG offspring) and (2) the number of breeding neighbours, that is the the number of nearby territories with a female that laid her first egg between -28 and 21 days after the focal pair (the maximum observed number of days between the first egg of the brood of the sire and the first egg of the EG brood). Because the number of breeding neighbours is related to breeding density and relatedness of the pair, we included this variable in a
64 separate model to avoid problems with collinearity of explanatory variables. The sample size is slightly lower in the second analysis (174 broods from 78 pairs), because for some broods we did not know whether birds in nearby territories were breeding. We fitted these variables as explanatory variables (one model with relatedness and density, and one with number of breeding neighbours) in a GLMM with binomial error distribution, and whether a brood contained EGP as response variable. To test whether EG offspring were sired by an EG male that was more genetically dissimilar from the female than her social male, we used a pairwise Wilcoxon signed rank test comparing relatedness of the female to her social male and to the EG sire. Second, we tested whether the difference in relatedness of the female with her social male and with the sire (as the response variable) depended on her relatedness with her social male (explanatory variable), using a ML model. In total, 13 cases were included from 11 broods by nine females (in two broods two different sires were identified). We used several approaches to explore whether females in incestuous pairs synchronised their breeding attempt with neighbors. First, we calculated the number of days between the first egg of the focal brood and the most synchronous brood of the EG sire, by subtracting first egg dates of both broods. We tested whether the range of the number of days differed among incestuous and non-incestuous pairs, using Barlett’s test for homogeneity of variance. One brood of a non-incestuous pair was not included because we did not find a nest of the sire within 169 days around the focal pair (which implies that we missed a nesting attempt). Second, we tested whether the distance to the EG sire (in number of territories) was different between incestuous and non-incestuous pairings using a Wilcoxon signed rank test including 14 cases (in 12 broods from 9 females). We repeated this analysis using a GLMM with quasi-poisson error, number of territories to EG sire as the response variable, relatedness (whether or not incestuous pairing) as the explanatory variable and female identity as a random variable. This model does not provide a P-value, but we tested the influence of relatedness by comparing the null-model (without relatedness) with the model with relatedness using an ANOVA. Third, to determine whether females in incestuous pairs synchronise egg laying with (one of) the neighbors, we selected all broods of incestuous pairs (also those that were not included in paternity analyses), and the brood
65 of the neighbour that was temporally closest to the focal pair, and calculated the absolute number of days between the dates when the first eggs were laid in both broods (‘number of days synchronisation’). As the ‘control’ group, we selected all broods of pairs that were located two territories away from focal pairs to minimise spurious effects of spatial differences in timing of breeding, and calculated the number of days synchronisation with the temporally closest neighbor (similarly as for incestuous pairs). Note that direct neighbors could not be used because we tested whether incestuous pairs synchronised with those. We only used control broods when those were laid by non- incestuous pairs and their neighbors were both non-incestuous pairs. In addition, to rule out potential temporal effects of breeding synchrony, we used broods of control pairs only when these were laid in the period the incestuous pair existed. We determined for each brood whether it was synchronous within 5, 10 or 20 days around the temporally most synchronised nesting neighbor. For each level of synchrony, the proportion of synchronous nests for each pair (bound using the cbind (number of synchronous broods, number of non-synchronous broods command) was included as the response variable in a GLMM with binomial response and whether or not a pair was incestuous as the explanatory variable. Focal (incestuous) pair ID was included as a random variable to link the focal and control pairs.
Results
Inbreeding depression and inbreeding avoidance by EG mating
Hatching success of eggs was significantly negatively correlated with pair relatedness (Z = -2.191, n = 216 broods, p = 0.028) because eggs of more related pairs were less likely to hatch ( β = -2.318 ± 1.058; Fig. 3.1). The likelihood that a nest contained EGY increased significantly with increasing genetic similarity of the members of the pair (Z = 3.434, n = 217 broods by 85 pairs, p < 0.001, β = 4.888 ± 1.424; Fig. 3.2). Social mates of promiscuous females were genetically more similar to the female than the EG sires (mean relatedness ± SE of the social mate: 0.200 ± 0.096; EG sire: -0.027 ± 0.048), but this effect was not statistically significant (Wilcoxon Z = -1.328, n = 13, p = 0.184; Fig. 3.3). The difference in relatedness between the social male and the EG sire, was significantly affected by relatedness of the social pair (t = -7.701, p < 0.001, β = -1.134 ± 0.147) as only females with a highly
66 related male mated with genetically more dissimilar males (Fig 3.3). Model residuals identified one outlier (see Fig 3.3); when excluding this data point, the effect was slightly smaller (t = -15.445, p < 0.001, β = -1.059 ± 0.069).
Figure 3.1. Inbreeding reduces hatchability of eggs in M. coronatus , as eggs laid by females with social mates that were genetically more similar were less likely to hatch. Note that in many cases eggs could not be genotyped and this also includes eggs resulting from extra-group mating. Therefore, levels of hatchability of eggs from highly related pairs (that had highest rates of extra-group paternity) are probably even lower. Numbers reflect number of eggs, number of broods. Relatedness between pair members was categorised (low: -0.255-0, intermediate: 0-0.310, high: 0.323-0.660) for illustration purpose, but for analyses continuous values of relatedness were used. All pairs with known pedigree in the high relatedness category were pairings between first-order relatives.
1.0
441, 148 0.9 181, 60
0.8
0.7 22, 8 Proportion hatched eggs hatched Proportion 0.6
low intermediate high Female relatedness to social male
Ecological constraints on extra-group mating
The likelihood that broods contained EGY decreased with decreasing numbers of territories nearby (within the maximum distance for EG mating; Z = 2.308, n = 217 broods by 85 pairs, p = 0.021, β = 0.489 ± 0.212; Fig. 3.4). Likewise, the likelihood that broods contained EGY decreased with decreasing number of territories in which the first egg was laid between -21 and 28 days after the first egg of the focal pair, but this was not quite significant (Z = 1.815, n = 174 broods by 78 pairs, p = 0.07, β = 0.546 ± 0.301; Fig. 3.4).
67 Synchronisation by incestuously mated females
Among incestuous pairs the number of days between egg-laying of the focal pairs and the pair of the sire was significantly less variable (range = 2-13 days, n = 6) than among non-incestuous pairs (range: -21-28 days, n = 8; Bartlett's K 2 = 6.225, p = 0.013; Fig. 3.5a). EG mating by females in incestuous pairs was almost exclusively with direct neighbors (in 5 of 6 cases) whereas females in non-incestuous pairs rarely mated with direct neighbors (in 1 of 8 cases), so that the distance to EG sires was significantly smaller for incestuous pairs (Z = - 2.550, p = 0.011; Fig. 3.5b). Correcting for pair-identity in a GLMM, the result was very similar (t = -2.045, p = 0.049, β = 0.693 ± 0.247). Clutches from incestuous pairs were almost twice as likely to be laid within five days of their neighbors compared to clutches laid by non- incestuous pairs (Z = 2.196, p = 0.028, β = 1.659 ± 0.755; Fig. 3.6). The difference in local breeding synchronisation between incestuous and non- incestuous pairs was smaller and non-significant when using 10 and 20 days synchrony (10 days: Z = 0.821, p = 0.412, β = 0.585 ± 0.713; 20 days: Z = 1.094, p = 0.274, β = 1.063 ± 0.971; see Fig 3.6).
Figure 3.2. Broods of closely related pairs were more likely to contain extra-group offspring. Relatedness between pair members was categorised (low: -0.318-0, intermediate: 0-0.314, high: 0.428-0.660) for illustration purpose, but for analyses continuous values of relatedness were used. All pairs with known pedigree in the high relatedness category were pairings between first-order relatives.
0.6
12
0.4
0.2 Proportion broods Proportion
with extra-group offspring extra-group with 143 62 0.0 low intermediate high Female relatedness to social male
68 Figure 3.3. Extra-group mating functions in inbreeding avoidance: females in an incestuous pair (i.e., with highly related social partner) mated extra-group with less related males, whereas other females (i.e. not mated incestuously) did not, indicating that different strategies for extra-group mating exist.
0.8
0.6
0.4
0.2
0.0
Relatedness to female to Relatedness -0.2
-0.4 Social male Extra-group sires
Figure 3.4. The likelihood that broods contained extra-pair offspring depended on the number of territories nearby (black bars; within the maximum distance we observed to an extra-group sire), where broods in denser areas were more likely to contain extra-group offspring. A similar (but not quite significant; P = 0.07) tendency existed when using the number of territories with pairs that had eggs between -28 - 21 days after the focal pair (‘simultaneous nesting’; grey bars).
0.4 all simultaneously nesting 12
0.3
0.2
22 60
Proportion broods broods Proportion 0.1 66 with extra-group offspring extra-group with 99 43 41 48 0.0 0-3 4-6 7-9 10-12 Number of nearby territories
69 Figure 3.5. Incestuously mated females (relatedness ≥ 0.323 with social male) that mate extra-group lay their eggs consistently shortly after the sire’s mate (A) and mated with extra-group males that are often direct neighbors (B). On the other hand, promiscuous females with an unrelated social male were less synchronised with the mate of her EG sire (A) that inhabit territories further away (B). Numbers in (B) reflect number of cases.
40 A
20
0
-20
-40 Number of days between first eggs of first eggs between days of Number the social pair and extra-group sire sire pair extra-group and social pair the
3.5 B
8 3.0
2.5
2.0 the extra-group sire extra-group the
Number of territories to to territories of Number 1.5 6
1.0 non-incestuous incestuous Social pair
70 Figure 3.6. Precise breeding synchronisation by incestuously-mated female fairy-wrens. Incestuous pairs (relatedness ≥ 0.323; black bars) were significantly more likely to have their first egg within five days of the first egg of their neighbor than non-incestuous pairs (grey bars; p = 0.028). There is a similar tendency for 10 and 20 days, but the difference between incestuous and non-incestuous pairs is far from significant (p = 0.412, and p = 0.274, respectively).
1.0
Incestuous pairs (18 broods) n.s. 0.9 Non-incestuous pairs (26 broods)
n.s. 0.8 * 0.7
0.6
0.5 with pairs neighboring with 0.4 Proportion Proportion synchronised broods 0.3 5 days 10 days 20 days
Level of synchronisation
Discussion
Extra-group mating for inbreeding avoidance
Inbreeding has been shown to reduce reproductive performance in many animal species (Amos et al. 2001, Charlesworth & Charlesworth 1987, Keller & Waller 2002, Kempenaers 2007), indicating selection against breeding with relatives (Pusey & Wolf 1996). In M. coronatus , restricted dispersal and limited breeding positions limit optimal social mate choice, and some individuals engage in incestuous pairings. Although such pairs often divorce after some time (Hall ML, Kingma SA, Peters A, unpublished data), many do initially nest together. We tested here whether females in incestuous pairs can avoid inbreeding through copulations with genetically dissimilar males outside the cooperative group.
71 Inbreeding avoidance through EP mating with genetically dissimilar males has recently become one popular hypothesis to explain multiple mating by females (Akçay & Roughgarden 2007, Kempenaers 2007, Mays et al. 2008). In line with this hypothesis, we found that M. coronatus females that were paired to a first-order relative were nearly 15 times more likely to produce EG offspring than females with a less related social male (Fig. 3.2). Those EG males were less related to the female than her social male was (Fig. 3.3), and therefore it is likely that the resulting offspring are more heterozygous than they would have been if fertilised by the social male. Offspring are predicted to benefit from increased heterozygosity because recessive deleterious alleles are less likely to be expressed than in more homozygous inbred offspring (Keller & Waller 2002, Lynch et al. 1995). Although we did not have the data to show that hatchability of eggs was associated with heterozygosity, we do show that hatching success of M. coronatus eggs decreases by more than 20% with increasing genetic similarity of the social pair. Especially considering that we could not correct for parentage in our conservative analysis and rates of hatching failure in incestuous pairs are probably even higher, this implies that inbreeding exerts a substantial cost on parental fitness. Thus, EG mating is one way in which females can avoid the negative effects of inbreeding.
Synchronisation to overcome constraints on EG mating
Overall rates of EGP are much lower in M. coronatus than in its congeners (6% vs. 40 to 80% of offspring; see Kingma et al. 2009; chapter 2), probably as a result of ecological constraints that reduce the availability of potential EG partners (Kingma et al. 2009). Here, we show that the number of nearby males had a positive effect on the incidence of EGP (Fig. 3.4), confirming this reasoning. Additionally, M. coronatus males are poorly adapted to EP mating, with small testes and cloacal protuberances (Kingma et al. 2009; see Fig. 2.1). Breeding is irregular and can take place at any time of the year, and if male fertility tracks the fertile period of his social partner, this implies that availability of fertile males is even further reduced in M. coronatus . Reduced periods of fertility of males can be expected as the size of cloacal protuberances (an indicator of sperm production) has been shown to peak around egg laying of their partner in other species with low EPP (e.g., in bearded tits Panurus biarmicus ; Sax & Hoi 1998). Altogether, the fact that females can be constrained in pursuing EP copulations raises the question
72 how they nonetheless can achieve this when they greatly benefit from EG mating (i.e., to avoid inbreeding). We found several indications that female M. coronatus in incestuous pairs adopt active strategies to overcome the constraints of EG mating. First, incestuous pairs laid eggs when they were nearly twice as likely as non- incestuous pairs to have neighbors that were also around egg-laying (Fig. 3.6), suggesting that females synchronise breeding with fertility of neighboring males. Second, females with related social males that mated EG laid their eggs more synchronously with - and consistently shortly after - the female of the EG sire than females that had EGP for other (unidentified) reasons (Fig. 3.5a). Third, EG offspring of females with related males were (nearly always) sired by neighboring males, whereas EG offspring of females with an unrelated partner were sired by males from further away (Fig. 3.5b). Female song rates are reduced during the pre-laying period, and males invest heavily in mate-guarding by maintaining close proximity to their partner until she lays her first egg (Hall & Peters 2009). M. coronatus females seeking EP copulations could therefore potentially use reduced female song rates as a cue to the specific period of fertility of neighbors, and target neighboring males when they are unconstrained by mate-guarding. This suggests that females in incestuous pairs actively ensure the presence of fertile EG males to facilitate EG copulations. This result forms, to our knowledge, the first empirical indication that EP mating to avoid inbreeding can be achieved by females through adopting an active pre-copulatory behavioural strategy. Our results may have broad implications for the understanding of the evolution of promiscuity. The fact that female strategies may affect the correlation between ecological factors and the incidence of EPP should be taken into account when considering the relationship between those. Additionally, we suggest that EP mating for inbreeding avoidance may be a more active strategy than often assumed, particularly in view of the lack of evidence for post-copulatory sperm selection (Denk et al. 2005). Therefore, our results form a considerable contribution to the discussion whether EP mating and inbreeding avoidance derive through active or primarily passive mechanism (i.e., sperm competition).
73 Acknowledgements
We are extremely grateful to E. Fricke and G. Segelbacher for help with the molecular work, to the staff at the Mornington Wildlife Sanctuary and the Australian Wildlife Conservancy for logistical support, and to our team of field assistants for their hard work in the field. All fieldwork was performed with permission from the Max Planck Institute for Ornithology Animal Ethics Committee, the Australian Wildlife Conservancy, the Western Australia Department of Conservation and Land Management (Licences BB002178 and BB002311), and the Australian Bird and Bat Banding Scheme (Authority 2230 and 2073). The research was funded by the “Minerva Sonderprogramm zur Förderung hervorragender Wissenschaftlerinnen” of the Max Planck Society (to AP).
74
75
76 Chapter 4
Multiple benefits drive helping behaviour in a cooperatively breeding bird: an integrated analysis
Sjouke A. Kingma , Michelle L. Hall & Anne Peters
American Naturalist (2011): in press
C < R x Bi + Bd
Extension of Hamilton’s principle
Where C is the cost for the actor, R is the relatedness between actor and recipient of help, Bi is the fitness benefit to the recipient, and Bd is the fitness benefit to the actor.
78 Abstract
Several hypotheses exist to explain the seemingly altruistic helping behaviour of cooperative breeders, although the general utility of these hypotheses remains unclear. While the potential importance of inclusive fitness benefits (kin selection) is traditionally widely appreciated, it is increasingly recognised that direct benefits may be more important than assumed. We use an integrative two-step framework to assess support for current hypotheses in purple-crowned fairy-wrens, a species where subordinates vary in relatedness to breeders and helping increases productivity. After establishing that assumptions of pay-to-stay and social prestige hypotheses (predicting that helping functions as ‘paying rent’ to stay on the territory or as a signal of individual quality, respectively) were not met, and that parentage by subordinates is extremely rare, we tested whether subordinates adjusted nestling feeding rates following the predictions of the kin selection and group augmentation hypotheses. Benefits of kin selection result from investment in relatives, and group augmentation benefits accrue when subordinates invest more in their own future helpers, for example when they have a better chance of inheriting the breeding position. We found that subordinates fed siblings more than unrelated nestlings, indicating that kin selection could facilitate cooperation. Moreover, the effect of relatedness on feeding effort varied depending on the probability of inheriting a breeding position, suggesting that active group augmentation can explain investment by unrelated subordinates. This statistical interaction would have gone undetected had we not considered both factors simultaneously, illustrating that a focus on single hypotheses could lead to underestimation of their importance in explaining cooperative breeding.
79
80 Introduction
In cooperatively breeding species, subordinate individuals seem to act altruistically by helping to raise offspring that are not their own (Brown 1987, Cockburn 1998). This phenomenon has received much attention because it violates the prediction of natural selection theory that individuals should invest in their own reproduction (Darwin 1859, Fisher 1930). Kin selection theory provides one solution to this, proposing that subordinates can gain indirect inclusive fitness benefits if they increase production of relatives (Hamilton 1964, Maynard Smith 1964). Despite unambiguous evidence that kin selection explains helping behaviour in some cooperatively breeding vertebrates (Baglione et al. 2003, Covas et al. 2006, Emlen & Wrege 1988, Komdeur 1994, MacColl & Hatchwell 2004, Nam et al. 2010, Russell & Hatchwell 2001, Wright et al. 2010; see also Cockburn 1998, Griffin & West 2003), it is now clear that kin selection is not an exclusive or even a general explanation for helping behaviour (Cockburn 1998, Clutton-Brock 2002, 2009, Griffin & West 2002). For instance, in some species, subordinates are not necessarily related to the offspring they care for (e.g., Canestrari et al. 2005, Dickinson 2004, Dunn et al. 1995, Legge 2000a, Wright et al. 1999), and there is considerable inter-specific variation in the importance of indirect benefits for helping behaviour (e.g., Griffin & West 2003). Thus, kin selection alone cannot explain the evolutionary maintenance of cooperative breeding, and alternative or additional mechanisms must be involved (Griffin & West 2002). Several alternative, direct benefits for subordinates have been postulated (see e.g., Bergmüller et al. 2007, Brown 1987, Cockburn 1998; Clutton-Brock 2002, 2009, Emlen & Wrege 1989, Lehmann & Keller 2006; see Table 4.1 for an overview). (1) The social prestige hypothesis states that subordinates may improve the likelihood of inheriting a breeding position by using helping to signal their quality to potential partners (Nowak & Sigmund 1998, Roberts 1998, Zahavi 1974, 1995). (2) Additionally, subordinates that help may enlarge group size, and benefit subsequently from improved survival in larger groups (passive group augmentation hypothesis), or from help by the new recruits when the subordinate obtains the breeding position (active group augmentation hypothesis; Kokko et al. 2001). Besides those future benefits, helping may also yield more immediate benefits: (3) Subordinates that help may secure parentage (parentage acquisition hypothesis; Dickinson 2004, Magrath & Whittingham 1997) or (4) helping can
81 be a form of rent paid in order to be tolerated in the group and to stay on the territory (pay-to-stay hypothesis; Kokko et al. 2002). However, probably because they are not mutually exclusive (e.g., Sumner et al. 2010), the importance of different benefits driving cooperative breeding remains generally unclear (e.g., Clutton-Brock 2002, 2009; Dickinson & Hatchwell 2004; Heinsohn 2004; Lehmann & Keller 2006; West et al. 2006, 2007). Here, we use a two-step framework to test the utility of the main hypotheses proposed to explain the evolutionary maintenance of helping behaviour in cooperatively breeding species (Bergmüller et al. 2007; see Table 4.1) in purple-crowned fairy-wrens Malurus coronatus . First, we established whether the central assumptions of these hypotheses (Table 4.1) are met in M. coronatus . Second, we tested whether individual subordinates adjusted nestling feeding rates as predicted by the kin selection and group augmentation hypotheses (see Table 4.2 for predictions). In M. coronatus , subordinates are usually philopatric offspring from previous broods. They can vary in their relatedness to the breeding pair, not so much due to extra-pair paternity (which is low; Kingma et al. 2009; chapter 2 and 3) but mainly as a result of replacement of one or both of the breeders, or, rarely, because they settled as subordinates with unrelated breeders (Kingma et al. 2010; chapter 5). Therefore, putative indirect as well as future and direct fitness benefits derived from helping differ between individuals (see Dickinson 2004, and Table 4.1). Investment in nestling feeding is costly in this species (as indicated by a negative relationship between individual feeding rates and survival in breeders; Kingma et al. 2010; see also Fig. 5.4), so that selection is predicted to favour this behaviour to be adjusted to the benefits resulting from it. As different explanations may not be mutually exclusive (e.g., Sumner et al. 2010), we use an integrated analysis, simultaneously exploring the importance of these hypotheses explaining subordinate investment.
82 Table 4.1. The main current hypotheses for explaining seemingly altruistic helping behaviour in cooperative breeders, their assumptions, and whether these assumptions are met in M. coronatus .
Hypothesis Fitness Principle a Assumptions Assumptions benefits met in M. coronatus ?
Kin selection theory Indirect Increasing productivity Subordinates enhance Yes b (Hamilton 1964) of relatives yields reproductive success inclusive fitness benefits Offspring are related Yes b,c,d
Social Prestige Future direct By investing more, Subordinates inherit Yes d (Zahavi 1974) subordinates signal breeding position in resident quality to potential territory future partners No d Competition among same-sex individuals, and queuing thus unstable
Active group Future direct Subordinate increase Subordinates enhance Yes b augmentation recruitment and new recruitment (Kokko et al. 2001) recruits will in turn assist them when they Subordinates inherit Yes d obtain the breeding breeding position in resident position territory Yes d Queuing among same-sex individuals is stable Yes d New recruits stay and help
Passive group Future direct Increasing group size Subordinates enhance Yes b augmentation will passively lead to recruitment (Kokko et al. 2001) subordinates survival benefits (increased Living in larger groups is Yes? b,e predator repellence, beneficial (e.g., enhanced food finding efficiency survival) etc.) All subordinates ‘help’ No c
Parentage Current direct Subordinates obtain Subordinates sire offspring, Rarely c,d acquisition direct reproductive or lay eggs (Magrath & success Whittingham 1997)
Pay-to-stay Current direct Subordinates benefit Non-cooperating individuals Unclear f (Kokko et al. 2002) from staying on the get punished territory, but they must invest as ‘paying rent’ in All subordinates ‘help’, and No c order not be evicted by unrelated subordinates the other residents potentially more
a See also Bergmüller et al. (2007) for a detailed description of different hypotheses. b As described in Kingma et al. (2009); chapter 2. c As described in Kingma et al. (2010); chapter 5. d From this study e Breeder survival is increased in larger groups, probably due to load lightening in nestling feeding (Kingma et al. 2010; chapter 5). Whether survival of subordinates increases as an effect of larger groups is unclear at this stage. f Punishment of non-cooperating subordinates was not investigated
83 Methods
Study site and species
M. coronatus are small (~ 10-12 g) facultative cooperatively breeding passerine birds that inhabit riparian vegetation in the wet-dry tropics of north Australia (Rowley & Russell 1993a, 1997, Skroblin & Legge 2010). They live in socially monogamous pairs (breeders), with 40 to 70% of pairs having on average 1 to 2 male and/or female subordinates (range 0-9 subordinates; Rowley & Russell 1997, Kingma et al. 2009; chapter 2). We studied a population of M. coronatus along Annie Creek and the Adcock River at Mornington Wildlife Sanctuary in North-west Australia (17°31’S, 126°6’ E) from July 2005 to October 2009. Groups and territory boundaries are stable year-round, and during weekly population censuses, we documented group size and social status of each uniquely colour-banded group member. Each bird could be unambiguously assigned breeder or subordinate status from behavioural cues (the most obvious that only the dominant pair sing duets; Hall & Peters 2008, 2009). In addition, nesting activity (occurring year- round) was monitored during weekly observations by following the female, and once found, nests were checked regularly to determine laying date and number of nestlings. M. coronatus lay usually three eggs (range 1-4) and nestlings fledge 13-14 days after hatching (Hall ML, Kingma SA, Peters A, personal observation).
Relatedness and parentage
Relatedness of subordinates to the breeding pair was determined using social pedigree data. Because extra-pair mating (4.4% of offspring) as well as within-group paternity by subordinates (1.8% of offspring) are rare in M. coronatus (Kingma et al. 2009; chapter 2), the social relationship of subordinates to the breeders reliably reflects their relatedness to nestlings. The two subordinates that were a result of extra-pair mating were excluded from the analyses. For the nine subordinates present as adults at the onset of the study, we confirmed that they were offspring of (one of) the breeders using molecular parentage testing as described below. Subordinates were most often less related because of replacement of one (17 individuals) or both breeders (2 individuals), but 3 individuals were dispersers that settled as
84 subordinates with unrelated breeders, between one week and over a year before nestling feeding started. To investigate whether subordinate males or females gained direct reproductive success, we conducted molecular parentage analyses (see chapter 2 for details). Briefly, we genotyped all breeders, subordinates and nestlings using six microsatellite loci, with exclusion probabilities of falsely assigning the breeder female and male as parents of less than 6% and 1% respectively (see Table 2.2). If a breeder could be rejected as genetic parent based on allelic mismatches, we tested whether another bird in the population sired the offspring. In one case of an extra-pair offspring, a subordinate male in the group as well as a neighboring male showed no mismatches. This group and the neighboring male were subsequently genotyped using three additional loci (Cu 28, Pte24, Pte26; chapter 3) and only the within-territory subordinate male showed no mismatches with the offspring, so we assumed he was the sire.
Feeding watches
Between September 2006 and October 2009, 187 nest watches were conducted at 51 nests on 26 territories, recording the individual feeding rate for each group member, including 54 subordinates. Seventeen of these 54 subordinates were observed at more than one nest, resulting in an average of 1.4 observed nests per subordinate (range 1-4), and a total of 286 individual feeding watches. Group members could be identified unambiguously by their colour rings in more than 97% (4243 of 4373) of feeding visits. Exclusion of feeding watches where we did not identify all birds did not change the results. Feeding watches were conducted on four (in some cases five) consecutive days between four and ten days after hatching, alternating morning (am; starting between 5:50 and 10:35) and afternoon (pm; starting between 14:30 and 18:00) watches, with the first feeding watch randomly conducted in the morning or afternoon. In some cases, fewer than four watches per nest were conducted due to nest failure. The duration of feeding watches was generally 60 minutes (mean ± SE = 60 ± 0.3 min.; range: 36-90).
Analyses
To test whether assumptions of the different hypotheses regarding the mechanisms underlying helping behaviour are met in M. coronatus , we tested
85 if and to what extent potential benefits apply (see Table 4.1). We quantified the magnitude of these benefits for each subordinate, based on relatedness between subordinates and breeders for inclusion in the analyses of feeding rates (see below). To do so, each subordinate was assigned to a ‘relatedness class’: related to both breeders, to the opposite- sex breeder only, to the same- sex breeder only, or unrelated to both breeders (see Table 4.3). First, kin-selected benefits for subordinates were expressed as their relatedness to the breeders (see Table 4.3 and above). Second, we explored potential benefits accruing from group augmentation by quantifying how likely subordinates were to inherit the breeding position in their resident territory if the same-sex breeder in their group died or dispersed (Table 4.1). First, we established that queues for inheritance are stable (see Table 4.1), by showing that younger subordinates did not inherit the breeding position when an older same-sex subordinate was present. Subsequently, we calculated the proportion of all subordinates over the entire study period that inherited the resident breeding position when the same-sex breeder died or dispersed, according to their relatedness to the opposite-sex breeder, and this proportion was used as the ‘probability of breeding position inheritance’ for further analyses of nestling feeding rates (see below and Table 4.3). To statistically confirm that the probability that the oldest subordinate filled the vacancy depended on its relatedness to the remaining opposite-sex breeder and on its own sex, we fitted generalised linear mixed models, using the lme4 package (Bates & Sarkar 2007) in R 2.8.1 ( R Development Core Team 2008). We fitted a set of models with ‘sex’, ‘relatedness’ and ‘sex and relatedness’ as independent variables, and compared these models using Akaike Information Criterion corrected for number of parameters (AICc; Burnham & Anderson 2002) using the spreadsheet available at http://www.uvm.edu/~bmitchel/software.html. Territory identity was included as a random factor to correct for repeated observations across territories. Third, we tested whether, and in which relatedness class, subordinates could gain direct reproductive success. Benefits of gaining parentage and breeding position inheritance were highly correlated (only males unrelated to the female gained parentage, and these males were also most likely to inherit; see ‘Results’ and Table 4.3), but because within-group extra-pair paternity is very rare (only one nest included in the analyses of feeding rates
86 involved a subordinate that sired an offspring), we conducted further analyses (see below) without direct benefits of parentage. To test whether subordinates adjust nestling feeding rates as predicted by the kin selection and group augmentation hypotheses (Table 4.2), we examined whether relatedness and probability of breeding position inheritance affected feeding rates of subordinates (feeding trips per hour, square-root transformed to normalise model residuals). We used maximum likelihood models in the nlme package in R (Pinheiro & Bates 2000; Pinheiro et al. 2009). Only birds older than the youngest age of a subordinate that was unrelated to one of the breeders (274 days) were included, since all younger birds were related to both breeders. To correct for replicated observations of nests and birds, nest identity and bird identity nested in nest identity were included as random factors. We fitted and compared four different models (using AICc) with respectively ‘relatedness to nestlings’, ‘probability of breeding position inheritance’, both variables, and both variables plus their interaction, as independent variables (values as given in Table 4.3), with feeding rates as response variable. Subsequently, we investigated whether observed investment patterns corresponded to predicted investment strategies for each hypothesis (see Table 4.2). The models used to analyze feeding rates also included a set of four potentially confounding variables that we had identified as explaining variation in feeding rates in prior analyses (not including potential benefits). In these models, we tested whether feeding rates were affected by: (1) brood age, (2) brood size (number of nestlings), (3) group size (including only subordinates >145 days old, as this was the youngest age a juvenile was seen feeding), (4) whether an observation was done in the morning or afternoon, and (5) age of subordinates (divided as first-year birds, and older birds). We fitted all possible models with feeding rates as dependent and these five variables as independent variables; the model with all variables except ‘group size’ gave the lowest AICc value (all other combinations had an AICc value of at least 1.67 higher). Therefore the other four variables were included in all models exploring the effect of potential benefits of cooperative breeding.
87
Table 4.2. Subordinate investment strategies in M. coronatus .
Hypothesis a Predicted investment strategy Observed in M. coronatus b
Kin selection theory Subordinates feed related nestlings Yes , in interaction with more (Griffin & West 2003) probability of breeding position inheritance (see Fig. 4.1)
Active group augmentation Subordinates with better prospects of Yes, in interaction with inheritance feed more (Dickinson relatedness (see Fig. 4.1) 2004, Kokko et al. 2001)
Parentage acquisition Subordinates that can gain Yes, but very rarely, and parentage feed more (Dickinson statistically indistinguishable 2004, Magrath & Whittingham 1997) from active group augmentation (see ‘Discussion’) a Hypotheses for which all assumptions, as defined in Table 4.1, are met. b See Table 4.4 and Fig. 4.1
Results
Most subordinate M. coronatus help raise kin: individuals were related to one or both breeders in respectively 23 (30%) and 47 (62%) of the 76 cases where we conducted feeding watches. Subordinates in territories with older same-sex subordinates never inherited a vacant breeding position in their resident territory (n = six male and one female subordinates). Oldest subordinates in the group however, often inherited the breeding position: in 15 of 29 cases when the resident same-sex breeder disappeared (11 male, 18 female breeders), the subordinate took over the breeding position (Table 4.3). The probability of inheriting a vacant breeding position depended on subordinate’s sex (Table 4.3): male subordinates were more likely to fill a vacancy than females; estimate ± SE = 2.56 ± 1.01, Z = 2.54, p = 0.011). In addition, consistent with expectations of inbreeding avoidance, fewer subordinates inherited breeding positions when they were related to the opposite-sex breeder than when the opposite-sex breeder was unrelated (in 10 of 22 cases (= 45%) vs. 5 of 7 cases (= 71%) respectively; Table 4.3), but this effect was not statistically significant (estimate ± SE = -1.72 ± 1.08, Z = -1.60, p = 0.11). Although the model with sex and relatedness (AICc = 40.18) gave a better fit than the model with relatedness only (AICc = 45.64), it was only marginally better than the model with sex only (AICc = 40.31).
88 Table 4.3. Population averages of relatedness between subordinates and nestlings, parentage acquisition and breeding position inheritance for male and female subordinate purple-crowned fairy-wrens, depending on whether they are related to both, one, or none of the breeders.
a b c Predicted relatedness with offspring Parentage acquisition by subordinates Breeding position inheritance subordinate sex: male female male female male female a related breeders both c (older subordinate present) 0.50 0.50 0 (31) 0 (6) 0.00 (6) 0.00 (1)
both c (no older subordinate present) 0.50 0.50 0 (53) 0 (37)
} 0.78 (9) 0.23 (13) opposite-sex only 0.25 0.25 0 (9) 0 (5)
same-sex only 0.25 0.25 0.05 (21) 0 (5)
} 1.00 (2) 0.60 (5) none 0 0 0.11 (9) 0 (4) a Subordinates were defined as ‘related’, when the breeder was a first-order relative. Since subordinates probably infer relatedness through associative learning (Komdeur & Hatchwell 1999), they are most likely unaware of their relatedness with second-order related opposite-sex breeders, & such cases were not included (n = 3). Genetic relatedness values differ only marginally from the given expected relatedness due to low frequency of extra-pair paternity (not shown). b The proportion of broods where a subordinate achieved parentage. Numbers in parentheses are number of broods. c The proportion of subordinates that inherited the resident breeding position when the same-sex breeder disappeared, and the remaining opposite-sex breeder was either related or unrelated to the subordinate, using averages of all cases over the entire study period for each class (number of cases are given in parentheses). The distinction between subordinates in groups with and without older same-sex subordinates was only made for groups where subordinates were related to both breeders, as this was the only category with cases where older same-sex subordinates were present. Subordinate feeding rates varied with their relatedness to the nestlings, with higher feeding rates when nestlings were more related (Table 4.4, model 1). When included singly in the statistical model, estimates of the probability of breeding position inheritance did not significantly predict subordinate feeding rates (Table 4.4, model 2). However, when the positive effect of kin selection was corrected for, the effect size of estimates of breeding position inheritance on feeding rate was 56% higher than when kin-selected benefits were not corrected for, and the AICc value was lower than the model without ‘relatedness’ (Table 4.4, model 3). Similarly, effect sizes of kin-selected benefits increased when estimates of breeding position inheritance were included in the model (Table 4.4, model 1 vs. 3). However, a model with a significant interaction effect between probability of territory inheritance and ‘relatedness’ yielded the best fit to the data (lowest AICc value; Table 4.4, model 4). The probability of breeding position inheritance appears to be more important in explaining feeding rates when subordinates are related to only one or none of the breeders and kin-selected benefits are thus smaller or absent (see Fig. 4.1). Sample sizes for subordinates unrelated to both breeders were low (see Fig. 4.1c), but model 4 nevertheless gave the best fit to the data when these were excluded from the analyses and conclusions were unchanged (results not given). Since ‘subordinate age’ may to some extent be correlated with kin- selection and breeding position inheritance we repeated the four models without subordinate age. In all four models, the models without subordinate age had an AICc value of between 4 and 13 higher than the models with subordinate age but otherwise the conclusions did not change (results not shown).
Table 4.4. Statistical models testing the effect of relatedness to the nestlings and probability of breeding position inheritance on subordinates’ nestling feeding rates. Presented are statistical models with one (model 1 and 2) or both (model 3) potential benefits, and their interaction (model 4). Model 3, with both benefits, gives a better fit to the data than model 1 (as indicated by a lower AICc value) and model 2 with one of the benefits only. However, there is a significant interaction effect between both benefits, and the model including the interaction (model 4) gave the best fit to the data. For each level of each variable, effect size, 95% (lower,upper) confidence intervals (CI) and p-value are given per model.
model 1. kin-selected benefits model 2. breeding position inheritance model 3. both benefits model 4. interaction (AICc = 697.91) (AICc = 705.62) (AICc = 693.50) (AICc = 690.79) effect size 95% CI p effect size 95% CI p effect size 95% CI p effect size 95% CI p kin-selected benefits (K) (r = 0.25) a -0.211 (-0.585,0.163) 0.280 -0.429 (-0.826,-0.032) 0.046 -1.324 (-2.245,-0.404) 0.011 kin-selected benefits (K) (r = 0) a -1.070 (-1.644,-0.495) 0.001 -1.207 (-1.767,-0.646) <0.001 -3.270 (-5.105,-1.435) 0.002 breeding position inheritance (I) 0.367 (-0.058,0.793) 0.104 0.571 (0.138,1.004) 0.017 0.285 (-0.122,0.692) 0.186
K (r = 0.25) * I a 1.304 (0.280,2.327) 0.022
K (r = 0) * I a 3.016 (0.616,5.416) 0.024
age of subordinate b 0.685 (0.336,1.014) <0.001 0.484 (0.119,0.849) 0.016 0.685 (0.220,0.882) 0.004 0.586 (0.277,0.896) 0.002 brood size 0.154 (-0.065,0.372) 0.174 0.242 (0.006,0.477) 0.050 0.154 (-0.085,0.339) 0.246 0.123 (-0.103,0.349) 0.293 age of nestlings 0.184 (0.122,0.245) <0.001 0.184 (0.122,0.246) <0.001 0.184 (0.124,0.247) <0.001 0.184 (0.122,0.246) <0.001 morning/afternoon c 0.159 (-0.0004,0.312) 0.052 0.158 (-0.001,0.318) 0.053 0.159 (0.003,0.323) 0.047 0.161 (0.0003,0.321) 0.051
a Effect sizes relative to fully related subordinates (r = 0.5)
b Effect sizes of older subordinates relative to younger subordinates
c Effect sizes of afternoon observations relative to morning observations Figure 4.1. Kin selected benefits (relatedness of subordinates to breeders) and prospects of territory inheritance have an interacting effect on nestling feeding rates (feeds/h) by subordinate purple-crowned fairy-wrens: subordinates feed (B) half-related and (C) unrelated nestlings less than (A) full siblings, but only when prospects of inheriting the breeding position are relatively low. Raw data are illustrated and dot size reflects number of observations (ranging from n = 1 to n = 19), but appropriate statistical models correcting for confounding variables and replicated observations were used for analyses (see text).
Discussion
Determining the importance of direct and indirect fitness benefits in driving helping behaviour may reveal adaptive mechanisms underlying the evolution of cooperative breeding. Here, we first determined which assumptions of current hypotheses are met in M. coronatus . The pay-to-stay hypothesis (stating that subordinates must help to be allowed to stay in the group; Kokko et al. 2002) and the passive group augmentation hypothesis (that predicts that when helpers increase group size they can gain associated survival benefits; Kokko et al. 2001) appear not applicable, as these require all subordinates to contribute to raising the offspring (Table 4.1, 4.2), which is not the case in M. coronatus (Fig. 4.1; and see chapter 5). Additionally, since subordinates do not seem to compete for breeding positions (subordinates never inherit if older same-sex subordinates are present), the social prestige hypothesis, stating that subordinates can show their quality by helping and be more likely to inherit a vacant breeding position (Zahavi 1974) is unlikely to explain helping behaviour (Table 4.1, 4.2). We have however, shown that for two other hypotheses (specifically, kin-selection and active group- augmentation) assumptions are met, and that these are important in explaining nestling feeding rates by subordinate M. coronatus . Below, we outline the utility of each of these hypotheses in explaining the evolutionary maintenance of cooperative breeding in M. coronatus .
Indirect and direct fitness benefits
Kin selection provides one adaptive explanation for costly helping behaviour in M. coronatus . We show that subordinates adjust their effort according to their relatedness to breeders, feeding siblings more than less-related nestlings (Fig. 4.1). Nestling feeding is costly: breeders that have higher feeding rates have lower survival probability (Fig. 5.3), and younger subordinates feed less than older subordinates (Table 4.4). Nestling feeding by subordinates increases the production of fledglings and the survival of breeders (Kingma et al. 2010; see Figs. 5.2 and 5.3), and therefore they gain inclusive fitness benefits from helping related breeders (Table 4.1). This finding adds to the evidence that kin-selected indirect benefits may be an important mechanism driving helping behaviour in cooperatively breeding vertebrates (see Griffin & West 2003). Nonetheless, unrelated subordinates
93
do feed nestlings at high rates too, and therefore one or more additional mechanisms must drive helping behaviour. We showed that subordinates can inherit a breeding position on their resident territory, which is a prerequisite for obtaining future direct benefits (Table 4.1). Probability of breeding position inheritance significantly predicted subordinate feeding rates when correcting for relatedness: the likelihood of breeding position inheritance becomes more important in explaining helping behaviour as indirect kin-selected benefits get smaller. Thus, subordinates gaining future direct benefits may explain why unrelated subordinates also contribute substantially to nestling feeding (Table 4.2). The most likely explanation is presented by the ‘active group augmentation’ theory (Kokko et al. 2001). This theory predicts that the current nestlings will in turn help subordinates when these inherit the breeding position in the future. In M. coronatus , subordinates can inherit the resident territory and they increase productivity and thereby the production of future helpers (see Table 4.1). Moreover, dominance queues for inheritance are stable among subordinates in the group, facilitating predictable prospects of territory inheritance (Kokko et al. 2001). Thus, in addition to kin-selected benefits, future direct benefits can also contribute to the maintenance of cooperative breeding in M. coronatus. This interaction of different hypotheses in explaining helping behaviour illustrates that, without taking multiple mechanisms into account, the real importance of single hypotheses may remain concealed. This could be one reason why conclusions from studies on the benefits of helping behaviour are mixed, and both theories are debated (e.g., Bergmüller et al. 2007 vs. Wright 2007). Although the number of subordinate males that achieved paternity was very low, their own parentage may nonetheless be an incentive for cooperation (see e.g., Dickinson 2004). At this stage however, the role of the potential gain of parentage in helping behaviour cannot fully be resolved in this virtually monogamous species. In our analyses, it was not possible to distinguish potential parentage from breeding position inheritance benefits, as both benefits only occur when a subordinate male is unrelated to the opposite-sex breeder. Unrelated female subordinates fed nestlings less than unrelated males, consistent with females never gaining parentage but also with their lower probability of inheriting the breeding position. However, female subordinates fed nestlings more when unrelated only to the male
94
breeder than when unrelated only to the female breeder, consistent with investment being related to future (breeding position inheritance) benefits.
Interacting hypotheses explain cooperation: implications
Because potential mechanisms underlying cooperative behaviour may interact, they could be underestimated or even concealed when investigated in isolation (Sumner et al. 2010; this study; but see also Wright et al. (2010) for discussion on population-wide kin selection). In fact, if we had not taken indirect benefits into account, the importance of future benefits (most likely through group augmentation) would remain unnoticed. Although it remains difficult to quantify the relative selection pressures these benefits exert on helping behaviour and their long-term effects on future reproductive success and longevity, the observation that the importance of different benefits in explaining helping behaviour may be underestimated when considered in isolation may have important implications for our understanding of different mechanisms driving cooperative breeding. If our finding reflects a general pattern, and effects of indirect benefits are systematically underestimated, this could explain why on average only approximately 10% of variation in helping behaviour can be explained by indirect benefits (Griffin & West 2003). Similarly, if the effects of direct benefits are commonly concealed by interacting effects of other benefits, this could be one reason why the utility of hypotheses based on direct benefits remains unclear (Clutton-Brock 2002). Thus, the ongoing debate about the importance of direct and indirect benefits in the evolution of helping behaviour in cooperatively breeding vertebrates will possibly only be resolved when more studies simultaneously investigate multiple factors that could shape helper strategies.
Acknowledgements
We are grateful to S. Legge, S. Murphy and other staff at Mornington Wildlife Sanctuary, to the Australian Wildlife Conservancy for logistical support, to our team of field assistants for their hard work in the field, and to the Max Planck Society for funding (‘Minerva Sonderprogramm zur Förderung hervorragender Wissenschaftlerinnen’ to AP). K. Delhey, A. Qvarnström and two anonymous reviewers provided comments on the manuscript and statistical advice.
95
96
Chapter 5
Multiple benefits of cooperative breeding in purple- crowned fairy-wrens: a consequence of fidelity?
Sjouke A. Kingma , Michelle L. Hall, Elena Arriero & Anne Peters
Journal of Animal Ecology (2010): 79, 757-768
97
98
Abstract
Kin selection is one of the mechanisms that can explain apparent altruism by subordinate individuals in cooperatively breeding species, if subordinates boost the production of kin. We compared productivity and breeder survival in pairs with and without subordinates in a genetically monogamous cooperatively breeding bird, the purple-crowned fairy-wren Malurus coronatus . Additive effects of subordinate help increased productivity. Total feeding rates to the nest were increased by two or more subordinates, and fledgling production was greater in larger groups. Not all subordinates contributed to nestling feeding, and the effect of group size was greater when non-contributors were excluded from analyses, suggesting that increased fledgling production was a direct result of help. Compensatory effects of subordinate help improved breeder survival. Assisted breeders reduced their work-load by 20-30%, irrespective of the number of helpers. While re-nesting intervals were not affected by group size, reduced breeder feeding rates resulted in improved survival and breeders in larger groups survived better. Subordinates and nestlings are usually progeny of the breeding pair in this species, and benefits of cooperative breeding are very different from three congeners with extremely high levels of extra-group paternity. In these Malurus, fledgling production and survival of male breeders are not enhanced in larger groups. This is consistent with the expectation that kin-selected benefits vary with relatedness, and thus levels of extra-group paternity. We tested whether benefits of cooperative breeding in 37 avian species varied with levels of extra-group mating. Both direct and phylogenetically-controlled comparisons showed that improvement of (male) breeder survival and enhanced productivity are more likely when fidelity is higher, as predicted when investment of subordinates correlates with relatedness to offspring. This pattern highlights the importance of considering the genetic mating system for understanding the evolution of cooperative breeding.
99
100
Introduction
In cooperatively breeding species, subordinate individuals forego their own reproduction to assist the breeding pair in the rearing of usually non- descendant offspring (Brown 1987, Cockburn 1998, Stacey & Koenig 1990). The adaptive basis of cooperative breeding and such seemingly altruistic behaviour has been discussed extensively, but remains a highly debated topic to date (Cockburn 1998, Dickinson & Hatchwell 2004, Hatchwell 2009, Lehmann & Keller 2006). However, it has generally been recognised that for cooperative breeding to be evolutionarily stable, there must be benefits for both the subordinate individuals and the breeders that accommodate them. Determining such benefits is therefore crucial for our understanding of the evolution and maintenance of cooperative breeding. Several direct and indirect benefits of cooperative breeding have been proposed (see e.g., Cockburn 1998, Dickinson & Hatchwell 2004, Emlen & Wrege 1989). Breeders can benefit if subordinates enhance their reproductive output (Cockburn 1998, Hatchwell 1999) or survival (Crick 1992, Khan & Walters 2002). Subordinates can obtain direct benefits, like parentage (e.g., Hartley & Davies 1994, Li & Brown 2002, Richardson et al. 2002), or benefits of group living (e.g., Gaston 1978, Kokko et al. 2002). In addition, future benefits may be important, for instance when subordinates can inherit the territory (e.g., Ligon & Ligon 1978), and therefore subordinates may benefit from raising offspring that later become helpers (group augmentation; e.g., Kokko et al. 2001). Third, inclusive fitness benefits may apply when subordinates increase productivity of close relatives (indirect kin-selected benefits: Hamilton 1964). Many benefits of cooperative breeding depend on enhanced reproductive output. This can result directly from an improvement in the outcome of the current reproductive attempt (see Cockburn 1998, Hatchwell, 1999). Alternatively or additionally, subordinates can positively influence future reproduction through increased breeder survival (see Khan & Walters 2002) or shortened intervals between reproductive attempts (e.g., Brown & Brown 1981, Canestrari et al. 2008, Innes & Johnston 1996, Langen & Vehrencamp 1999, Woxvold & Magrath 2005). Which effects prevail (future or current benefits) may depend on whether assisted breeders decrease investment (compensatory effects; Crick 1992, Khan & Walters 2002) or
101
maintain similar effort, so that subordinate assistance is additive (Cockburn 1998, Hatchwell 1999, Heinsohn 2004, Legge 2000b). Subordinates can benefit from helping if they derive inclusive fitness from enhanced production of related individuals (Emlen 1995a, 1997, Emlen & Wrege 1989, Griffin & West 2003, Hamilton 1964; for other benefits see Cockburn 1998). As provisioning of care to dependent young is costly (Bryant 1988, Heinsohn & Legge 1999), subordinates are expected to adjust their contribution according to potential benefits and, assuming that kin-selection plays an important role, invest more in closely related offspring (Griffin & West 2003, Hamilton 1964, Komdeur 1994). Thus, the genetic mating system may modulate effects of group size: in species with high levels of mating outside the group, offspring are less related to subordinates, and consequently indirect benefits are lower. If subordinates adjust care accordingly, the effects of subordinates on productivity may depend on the level of extra-group paternity. Surprisingly however, such a role of extra- group mating in shaping benefits of cooperative breeding has, as far as we are aware, not been explicitly considered among cooperatively breeding vertebrates (see Cockburn 2004, Hatchwell 1999). Here, we investigate whether group size is related to current (offspring production) or future (breeder survival and re-nesting interval) productivity in purple-crowned fairy-wrens Malurus coronatus . In addition, we explore nestling feeding rates as a mechanism behind observed effects. M. coronatus is genetically nearly monogamous (Kingma et al. 2009), and a remarkable exception in the genus Malurus, which has some of the highest known levels of extra-pair paternity reported in birds (Brooker et al. 1990, Griffith et al. 2002, Karubian 2002, Kingma et al. 2009, Mulder et al. 1994, Webster et al. 2004, 2008; see also Table 2.1). Therefore, we explicitly compare benefits of cooperative breeding between Malurus species. Finally, we conduct a phylogenetically-controlled comparative analysis of the effects of subordinates in all cooperatively breeding birds with known levels of extra-group paternity, to establish whether the genetic mating system might play a role in the nature of benefits of cooperative breeding in general.
102
Methods
Study site and species
A population of between 33 and 48 M. coronatus groups was studied from March 2006 to September 2008 at Mornington Wildlife Sanctuary, in north- west Australia (S17°31’ E126°6’). The study site consists of about 8 km of riparian vegetation along Annie Creek and the Adcock River. M. coronatus are riparian specialists endemic to the northern tropics of Australia, and are restricted to creeks and rivers, rarely venturing more than 20 meters from the riverbed (Rowley & Russell 1997; Hall ML, Kingma SA, Peters A, personal observation). Territories are arranged linearly along rivers occupying both riverbanks, and are maintained throughout the year (Hall & Peters 2008, 2009). M. coronatus breed cooperatively, and live in socially monogamous pairs (breeders) with between 40 and 70% of pairs being accompanied by on average between 1 and 2 male and/or female subordinates (range 0-9 subordinates; Rowley & Russell 1997, see also chapter 2 and 4). Nesting occurs primarily in the wet season, but birds can breed opportunistically in any month (Rowley & Russell 1993a, 1997). Nests are built and eggs incubated exclusively by the breeder female. Clutch size is 1-4 eggs (median 3); incubation lasts around 14 days and the nestling period for about 13 days (Hall ML, Kingma SA, Peters A, personal observation). Subordinates assist the dominant pair in feeding the nestlings (Rowley & Russell 1993a).
Population and nest monitoring
All individuals in our study population were banded with a unique combination of three plastic colour bands and a numbered metal band. Sex was determined on the basis of sex-specific plumage characteristics (see Rowley & Russell 1997). Weekly population censuses documented group size, survival and social status of each group member, and nesting activity. A group member was defined as an individual within a given territory present for a minimum of two weeks. Each bird in a group could be unambiguously assigned dominant or subordinate status from behavioural cues (the most distinctive being that only the dominant pair participates in a song duet; Hall & Peters 2008, 2009).
103
Females were observed weekly for signs of nesting activity. Once found, nests were checked regularly to determine clutch size, hatching date, and number of nestlings. Nestlings were banded when they were approximately 7 days old (average banding age ± SE = 7.4 ± 0.13 days, range = 5-12, n = 111 broods). Number of fledglings was determined by inspecting the territory around the day when fledging was expected. The average (± SE) number of days between expected fledging date and determination of number of fledglings was 1.8 ± 0.22 days (n = 76, range = 0-9 days) and this interval did not predict the proportion of nestlings at banding that fledged, indicating that potential bias of early fledgling mortality is negligible (Generalised Linear Model with number of nestlings as binomial total, number of fledglings as response variable, and number of days after expected fledge date as explanatory variable: n = 68 broods, Wald = 0.40, p = 0.53).
Group size
Breeders were defined as the dominant breeding male and female, and ‘group size’ includes the breeding pair and all subordinates. Juveniles (subordinates less than 145 days old, the youngest age a subordinate bird was observed feeding nestlings) were not included in our measure of ‘group-size’. Some subordinate individuals did not contribute to nestling feeding (18 of 80 subordinates in 13 out of 39 nests with subordinates). Therefore the analyses were repeated with ‘number of helpers’, defined as individuals that contributed to nestling feeding in at least one nest watch (see below). Group- size (2, 3, and 4+) and ‘number of helpers’ (0, 1, 2+) were grouped into three categories since cases with more than two subordinates or helpers were rare (11 and 7 of 73 observed nests respectively).
Nest observations
Between September 2006 and September 2008, we conducted 266 nest watches at 73 nests on 39 territories, recording the total feeding rate (number of visits to the nest per hour) and the individual feeding rate for each group member. Group members could generally be unambiguously identified by their colour rings, except for 67 feeding visits (< 2% of 3,610 feeding visits recorded in total) in 38 nest watches of 25 nests in 19 territories. Generally four nest watches (mean ± SE: 3.63 ± 0.10; range: 1-5) were conducted on usually consecutive days between four and ten days after hatching,
104
alternating morning (am; 5:30 - 11:00) and afternoon (pm; 14:00 - 18:00), with the timing of the first nest watch randomly determined. The duration of nest watches was generally 60 minutes (mean ± SE = 60.1 ± 0.26; range: 30-90 minutes).
Statistical analyses
For analyses a mixed modeling approach in Genstat 11.0 was used. By fitting random factors in the model, mixed models allow correcting for hierarchical repeated sampling (Schall 1991). For continuous response variables, restricted maximum likelihood models (REML) were used. For binary and count (poisson) data, generalised linear mixed models (GLMMs) were used with logit link (with binomial totals of 1) and log-link functions respectively. All explanatory variables and relevant interactions were included initially, and the variable with the highest p-value was removed stepwise, leading to a final model with only significant terms (p < 0.05). Reported values of non- significant variables were obtained after re-inclusion in the final (significant) model. We present effect sizes, (back-transformed) means, and significance details as predicted by the full models in the text and tables, but in the figures we present raw data. To meet model assumptions, residuals of final REML models were checked for normality, and response variables were transformed when residuals were not normally distributed (see below). Random factors with a negative component of variance were omitted from the models. Where relevant, post-hoc tests were conducted by testing whether differences between categories were larger than the calculated least significant differences (LSD) at 5, 1 and 0.1% levels. As post-hoc tests were not available for GLMMs, categorical differences were assessed by restricting original analyses to the subset of data of relevant categories. Recently, it has been argued that penalised quasilikelihood parameter estimation, as implemented in Genstat, could be problematic for GLMMs with binomial or poisson data (Bolker et al. 2009). This was not the case for our analyses since repeating the statistical models using Laplace approximation for parameter estimation in the lme4 package (Bates & Sarkar 2006) in R 2.8.2 (R development Core Team 2009) yielded very similar results (data not shown).
105
Analyses of productivity We tested whether ‘group size’ affected the following variables: clutch size, number of fledglings, breeder survival, and re-nesting interval. Since group- size effects caused by helping may be better predicted by number of helpers or feeding rates (e.g., Davies & Hatchwell 1992), we repeated these analyses, replacing ‘group size’ with ‘number of helpers’ and then with ‘feeding rate’ (mean total feeding rate or mean individual feeding rate averaged per nest) where possible (see below). The effect of ‘group size’ and ‘number of helpers’ on clutch size and number of fledglings (poisson distribution) was tested using GLMMs with ‘Breeding pair identity (ID)’ as a random factor. Only nests that were at least partly successful (i.e. produced fledglings) and two nests that contained dead nestlings were included in the analyses of fledgling success; nests that failed completely due to flooding, nest collapse, disappearance of the entire brood, inter-specific brood parasitism or disappearance of one of the parents, were not included. Survival of breeders was expressed as likelihood of surviving until the next reproductive attempt. We obtained information on breeder survival from weekly territory censuses. We did not use a mark-recapture approach as ‘recapture rates’ were high: we resighted more than 95% of birds each week throughout the year. Moreover, including all nests in the study period, breeders survived in 440 of 474 (93%) cases until the next reproductive attempt. Our estimates of survival were unlikely to be confounded by dispersal of breeders out of the study site as we confirmed that this was rare with annual surveys of over 90% of suitable habitat within a 20 km radius of the study site (Hall ML, Kingma SA, Peters A, unpublished data). To avoid pseudoreplication of individuals with multiple nests, only the last nest of individuals that did not survive and the penultimate nest for those that did survive until a next nesting attempt were used. GLMM analyses with the random term ‘territory-ID’ were used for analyses with ‘group size’ as explanatory variable. Males and females were analyzed together by adding ‘sex’ as a factor, but survival was not different between males and females (Wald = 0.04, p = 0.84) and effects of group size on survival did not depend on breeder sex (interaction effect: Wald = 0.02, p = 0.99). Brood size (Wald = 0.06, p = 0.97) and the number of days individuals cared for nestlings/fledglings (with a maximum of 105 days when fledglings are independent; Wald = 1.36, p = 0.25) did not affect survival. For the
106
analyses of the effect of number of helpers on breeder survival the last (or penultimate; see above) nest for which feeding watches were conducted were used. Sample sizes when including ‘number of helpers’ and ‘feeding rates’ as explanatory variable were reduced to the subset of nests at which feeding watches were conducted. To avoid over-parameterised models, we included only ‘breeder sex’ as an additional explanatory variable in these GLMMs, but also in this model survival was not different between males and females (Wald = 0.92, p = 0.35). Re-nesting interval was expressed as the logarithm of number days between end of a nest (failure date or fledging date) and initiation (first egg) of the next nest per pair. Re-nesting interval was analysed using REML models with ‘Female-ID’ as a random term. ‘Female-ID’ revealed negative variance in the model with ‘number of helpers’ and the model with ‘feeding rate’ as the dependent variable, and was omitted from these models. Brood size did not affect re-nesting interval (in all models Wald < 3.37, p > 0.19), but the significant positive effect of the number of days the offspring survived (in all models Wald > 14.75, p < 0.001) was corrected for.
Analyses of feeding rates To investigate whether ‘group size’ and ‘number of helpers’ influence total feeding rate and individual feeding rates of dominant breeders (response terms, square-root transformed to normalise model residuals), REML models were used with control for random factors (‘territory-ID’ and ‘nest-ID within territory-ID’ in analyses of total feeding rate and ‘bird-ID’ and ‘nest-ID within bird-ID’ in analyses of breeder feeding rates; territory-ID showed negative variance in analyses of breeder feeding rates and was not included). Brood size and wind-intensity were corrected for in the model, as larger broods were fed more often overall and by individuals (in all models Wald > 39.41; p < 0.001), and feeding rates were lower when conditions were very windy, compared to no or light wind (in all models Wald > 8.90, p < 0.014). In addition, females were observed brooding younger nestlings mainly in the morning and, as brooding may affect feeding rates, we corrected for the significant interaction between whether a nest watch was done in the morning or afternoon and age of the brood (in all models Wald > 9.83, p < 0.003). In the analyses of breeder feeding rate, sex of the focal birds was added, but feeding rates were not different between males and females (in both models Wald < 0.92, p > 0.34), and neither did effects of group size or
107
number of helpers on breeder feeding rate depend on sex of the breeder (interaction effects; in both models Wald < 0.80, p > 0.67). Nest watches where one or more visiting birds could not be identified were omitted from these analyses.
Comparative study
A systematic search was conducted for literature providing levels of extra- group paternity in altricial cooperatively breeding bird species, using review papers (Arnold & Owens 2002, Griffith et al. 2002, Owens & Hartley 1998), and the Web of Science (keywords: “parentage” or “paternity” and “cooperative” or “birds”). Subsequently, for the species for which rates of EGP (for more than eight broods) are published, we investigated whether or not subordinates increased reproductive success and breeder survival, obtained from review papers (Cockburn 1998, Hatchwell 1999, Khan & Walters 2002, Legge 2000b), Web of Science (search on common and scientific species name), or directly from researchers. This resulted in a data-set of 37 species, widely distributed throughout the Aves (28 genera, 22 families; see Fig. 5.1). Although positive effects of subordinates may only become apparent after fledging in some species (e.g., Hatchwell et al. 2004), number of fledglings per nest was used as measure of reproductive success because these data are most widely available and therefore most suitable for standardised comparison between species. In addition, if effects of subordinates varied within species (for example with breeder female age (Magrath 2001) or brood size (Meade et al. 2010)), the overall effect was included (see footnotes in Table 5.2). Effects of subordinates on the benefits of cooperative breeding (increased reproductive success and male and female breeder survival) were compared between species with low (below median, 'faithful"; 18 species) and high (above median, 'unfaithful'; 19 species) rates of EGP using a Fisher exact test (Agresti 1992). Effects of subordinates may be similar among species that share common ancestry (Harvey & Pagel 1991). Therefore additionally phylogenetically informed statistical analyses (generalised least squares models) were conducted, controlling for phylogenetic associations among species, to examine the link between levels of EGP (as continuous variable) and the benefits of cooperative breeding (whether or not group size had a positive effect on number of fledglings per nest or breeder survival.
108
A phylogenetic tree was reconstructed (see Fig. 5.1) for 37 cooperatively breeding bird species included in the comparative study (see Table 5.2), using Hackett et al. (2008) for the major lineages, Jønsson & Fjeldså’s (2006) supertree for the passerine species, and Christidis & Schodde (1997) for the Malurus species. Because we used information from different sources to produce our phylogenetic hypothesis, all branch lengths were assumed equal for analyses. The program MESQUITE (Maddison & Maddison 1997) was used to construct the phylogenetic tree. Associations between extra-group paternity (EGP) and benefits of cooperative breeding were tested using phylogenetically informed generalised least squares models fitted by generalised least squares in R 2.9.2 (R development Core Team 2009), using the “ape” (Paradis et al. 2004) and “nlme” (Pinheiro et al. 2009) packages. Levels of EGP were negatively correlated with whether subordinates enhanced reproductive success (“ERS”; correlation: -0.189, t 37 = -2.454, p = 0.019). The presence of phylogenetic autocorrelation in the data was examined using Moran’s autocorrelation Index (Diniz-Filho 2001), but there was no significant phylogenetic autocorrelation in the data of EGP (Moran’s I for EGP: observed = -0.034 ± 0.01, expected = -0.028, p = 0.51, variance explained by phylogeny (var ) = 0.113) or ERS (Moran’s I for ERS: observed = -0.021 ± 0.01, expected = -0.028, p = 0.53, var = 0.177). We also found a significant correlation between extra- group paternity and whether subordinates enhanced male survival (-0.232, t 21 = -2.353, p = 0.029), and a similar non-significant trend was found for females
(-0.266, t 22 = -1.320, p = 0.20). Moran’s I function showed a significant phylogenetic autocorrelation in the data (Moran’s I for enhanced male survival: observed = -0.134 ± 0.02, expected = -0.05, p < 0.001, var : 0.595; Moran’s I for enhanced female survival: observed = -0.121 ± 0.02, expected = - 0.05, p < 0.001, var : 0.831). The significant autocorrelation index for the enhancement of (male and female) breeder survival shows phylogenetic inertia, indicating that part of the across-species variation in whether subordinates enhanced survival of breeders can be explained by phylogenetic associations among species. Nonetheless, for the enhancement of male survival the results obtained when controlling for phylogenetic associations among species are consistent with results of analyses when phylogeny was not controlled for. Therefore, the fact that species are non-independent from each other, does not affect our main conclusions. For the enhancement of female survival however,
109
phylogenetically controlled analyses do not show a significant correlation with EGP. Therefore the significant effect in the analyses where phylogeny was not controlled for may be the result of phylogenetic associations among species, and should be interpreted with caution.
Figure 5.1. Phylogenetic tree for 37 cooperatively breeding species for which rates of extra-group paternity and effects of subordinates on reproductive success and/or breeder survival are known (see Table 5.2).
110
Results
Productivity
Clutch size averaged (± SE) 2.9 ± 0.05 eggs, and was unrelated to group size (Table 5.1a). However, number of fledglings was significantly affected by group size (Table 5.1b). On average, groups with one subordinate (n = 22 nests) produced 0.04 more fledglings per nest than pairs (n = 52), while groups of four or more (n = 24) produced 0.77 more fledglings per nest. Number of fledglings was even more enhanced by number of helpers: number of fledglings per nest increased by 0.42 and 0.92 when pairs had one helper and two or more helpers respectively (see also Fig. 5.2a, Table 5.1b). The probability that breeders would survive until a subsequent nest increased from 55% for breeders in pairs, to 75% in groups with one subordinate, and 89% in groups with two or more subordinates (Fig. 5.3a, Table 5.1c). The pattern was similar for number of helpers (70%, 84% and 95% survival probability for breeders without, with one helper, and with two or more helpers respectively), though not significant (Table 5.1c), probably due to lower sample sizes (n = 25, 19, and 14 individuals for pairs, groups with one helper and groups with two or more helpers). Survival probability of breeders until a subsequent nesting attempt was negatively correlated with feeding rate (Fig. 5.4, Table 5.1c). Re-nesting interval was independent of group size, number of helpers, and breeder feeding rates (Table 5.1d).
Feeding rates
Nests with two or more helpers received significantly more food than nests of unassisted pairs or pairs with one helper (42% and 30% respectively; see Fig. 5.2b, Table 5.1e). Group size did not significantly affect total feeding rates, though differences between means followed a similar pattern (19% and 18% more feeding trips at nests of groups with two or more subordinates than groups with no or one subordinate respectively; Table 5.1e). Male and female breeders in groups fed at lower rates (on average about 20-30% less) than breeders in pairs (Fig. 5.3b, Table 5.1f), and the effect of number of helpers on breeder feeding rate was similar (Table 5.1f).
111
Table 5.1 . Overview of benefits of cooperative breeding in M. coronatus . Shown are statistical significance and effect sizes for group size and the number of helpers on measures of current productivity (clutch size (a), fledgling production (b)), future productivity (breeder survival (c), re-nesting interval (d)), and total (e) and breeder (f) feeding rates. In models of future productivity (c, d) we also tested for effects of breeder feeding rates. Group size comprises breeders and all subordinates, including individuals that did not contribute to nestling feeding, whereas number of helpers is the number of subordinates that were observed feeding nestlings (see Methods). Values are obtained from final statistical models after correcting for other significant variables (see ‘Statistical Analyses’ in the Methods section). Average effect sizes are given (± SE for continuous variables).
Wald p Effect sizes n a a. Clutch size Group size 0.12 0.94 3: -0.013; 4+: -0.013 164 c, 55 p Number of helpers 0.21 0.90 1: -0.025; 2+: -0.006 58 c, 43 p b. Number of fledglings Group size 13.47 0.002 3: 0.021; 4+: 0.373 98 n, 54 p Number of helpers 6.73 0.047 1: 0.238; 2+: 0.460 50 n, 40 p c. Breeder survival Group size 6.91 0.037 3: 0.906; 4+: 1.949 88 i Number of helpers 3.04 0.23 1: 0.798; 2+: 2.106 58 i Breeder feeding rate 8.72 0.005 X ± SE: -0.536 ± 0.182 58 i d. Re-nesting interval (log-transformed) Group size 2.46 0.30 3: -0.115; 4+: 0.040 96 n, 37 f Number of helpers 1.50 0.48 1: -0.029; 2+: 0.161 35 n, 22 f Breeder feeding rate 0.47 0.50 X ± SE: 0.037 ± 0.024 35 n, 22 f e. Total feeding rate (Sqrt-transformed) Group size 4.56 0.11 3: 0.001; 4+: 0.284 266 w, 73 n, 39 t Number of helpers 17.33 <0.001 1: 0.136; 2+: 0.586 266 w, 73 n, 39 t f. Breeder feeding rate (Sqrt- transformed) Group size 23.86 <0.001 3: -0.343; 4+: -0.386 456 w, 70 n, 39 t Number of helpers 10.52 0.001 1: -0.248; 2+: -0.356 456 w, 70 n, 39 t
a Abbreviations for sample sizes (n) are as follows: c = clutches, p = pairs, n = nests, i = individuals, f = females, w = nest watches, t = territories. Note that sample sizes are smaller in analyses with number of helpers and feeding rates than in analyses with group size, because feeding watches were not conducted on all nests.
112
Figure 5.2. Additive effects of helpers on productivity: (a) number of fledglings per nest, and (b) total feeding rates. The graph shows raw data (mean + SE), but statistical analyses controlled for additional variables in (a) GLMM and (b) REML models (see Method section). Numbers depict sample size as (a) number of nests and (b) number of nest watches, number of nests. Significant differences are indicated at the 5 (*), 1 (**) and 0.1% (***) level.
(a) * 3.0
2.5
2.0
1.5 28 13 9 Number of fledglings per nest
) 22 (b) *** -1 ** 20
18
16
14
12
Total feeding rate (visits h 10 144, 39 66, 20 56, 14
0 1 2+ Number of helpers
113
Figure 5.3. Compensatory effects of group size on breeders: (a) breeder survival probability until a subsequent nest, and (b) breeder feeding rate. Note that patterns are similar in analyses with ‘number of helpers’ in stead of ‘group size’ as independent variable, but for breeder survival the effect is not significant, probably due to small sample size (see Table 4.1c). The graph shows raw data (mean + SE), but statistical analyses controlled for additional variables in (a) GLMM and (b) REML models (see Method section). Numbers depict sample size as (a) number of birds, (b) number of nest watches, number of nests. Significant differences are indicated at the 5% (*) and 0.1% (***) level.
* 1.0 (a)
0.8
0.6
Survival probability Survival 0.4 4128 19
8 ***
) (b) *** -1
7
6
5 Feeding rate h (visits 4 228, 33 112, 19 116, 18 2 3 4+ Group size
114
Figure 5.4. Average feeding rates at the last or penultimate breeding attempt of breeders that subsequently disappeared and presumably died, or survived respectively. The graph shows raw data (mean + SE), but statistical analyses controlled for additional variables in (a) GLMM models (see Method section). Numbers represent number of birds.
7 ) )
-1 6
5 (visits h
4 13 45
Average breeder feeding rate Average breeder no yes
Survived until next nesting attempt
Comparative study
Subordinates increase productivity (number of fledglings per breeding attempt) in almost all (16 of 18; 89%) faithful species, but only in nine of the 19 unfaithful species (p = 0.013; see Table 5.2). Similarly, in the phylogenetic analysis, the effect of subordinates on reproductive success (increase in the number of fledglings per breeding attempt; correlation: -0.189, t 37 = -2.454, p = 0.019) are negatively correlated with rates of EGP across species. Thus, subordinates in species with low EGP are more likely to have a positive effect on reproductive success than subordinates in species with high EGP (see Table 5.2). In addition, there are clear differences in the enhancement of breeder survival, as survival of assisted male breeders is increased in seven of 12 (58%) faithful species, but in none of the unfaithful species (p = 0.007; Table 5.2). Similarly, when correcting for phylogeny, there is a significant negative correlation between probability that subordinates enhance male breeders’ survival and rates of EGP (correlation = -0.232, t 21 = -2.353, p = 0.029). A similar pattern is found for enhancement of female survival (analysis of enhancement of female survival in faithful (67%) vs. unfaithful (20%) species (p = 0.043)), though this is not statistically significant when correcting for phylogeny (correlation: -0.266, t 22 = -1.316, p = 0.20).
115
Table 5.2. The effect of subordinates in 37 cooperatively breeding bird species for which levels of extra-group paternity (EGP) are published. Listed are the level of EGP, whether subordinates increase the number of fledglings per nest (reproductive success, RS), and whether subordinates increase survival probability of male and female breeders. Species were divided in two categories, where species with below- and above-median levels of EGP were respectively classified as ‘low EGP’ and ‘high EGP’.
Increased breeder survival References
Species EGP a Increased RS b male female EGP RS and survival Arabian babbler ( Turdoides squamiceps ) 0 Yes ...... 1 42-46 Florida scrub-jay ( Aphelocoma coerulescens ) 0 Yes Yes Yes 2 47-51 apostlebird ( Struthidea cinerea ) 0 Yes No k No k 3 52 laughing kookaburra ( Dacelo novaeguineae ) 0 Yes h ...... 4 53, 54 bushtit ( Psaltriparus minimus ) 0 Yes ...... 5 55 sociable weaver ( Philetairus socius ) 0 Yes No Yes 6 56,57 white-winged chough ( Corcorax melanorhamphos ) 0 Yes Yes Yes 7 58-62 Galapagos hawk ( Buteo galapagoensis ) 0 Yes ...... 8 63-65 acorn woodpecker ( Melanerpes formicivorus ) 0 Yes Yes Yes l 9,10 66-69 Alpine accentor ( Prunella collaris ) 0 No i ...... 11 70 Henderson reed warbler ( Acrocephalus vaughani taiti ) 0 No ...... 12 12 bicolored wren ( Campylorhynchus griseus ) 2 Yes Yes Yes 13 71-73 dunnock ( Prunella modularis ) 2.2 Yes No No 14 74-79 red-cockaded woodpecker ( Picoides borealis ) 2.3 Yes Yes Yes 15, 16 80-84 stripe-backed wren ( Campylorhynchus nuchalis ) 2.9 Yes No No 17 73, 85 whitefronted bee-eater ( Merops bullockoides ) 3.1 c Yes No No 18 86-90 purple-crowned fairy-wren ( Malurus coronatus ) 3.8 Yes Yes Yes 19 91 European bee-eater ( Merops apiaster ) 5.3 Yes Yes l Yes 20 92 carrion crow ( Corvus corone ) 5.6 Yes ... … 21 93 noisy miner ( Manorina melanocephala ) 5.7 Yes ...... 22 94 bell miner ( Manorina melanophrys ) 7.7 No ...... 23 95 superb starling ( Lamprotornis superbus ) 9 Yes ...... 24 96 white-breasted thrasher ( Ramphocinclus brachyurus ) 10 Yes ...... 25 97 longtailed tit ( Aegithalos caudatus ) 10.5 No No m No 26 98-100 subdesert mesite ( Monias benschi ) 16.7 No ...... 27 101 grey-crowned babbler ( Pomatostomus temporalis ) 18.3 Yes ...... 28 102-105 brown jay ( Cyanocorax morio ) 21.6 Yes No No 29 106-109 American crow ( Corvus brachyrhynchos ) 21.7 Yes h No No 30 110,111 Tibetan ground tit ( Pseudopodoces humilis ) 23.4 No … … 31 31 white-browed scrubwren ( Sericornis frontalis ) 23.5 No j No No 32 112,113 White-throated magpie-jay ( Calocitta formosa ) 39.1 No No No 33 114,115 Seychelles warbler ( Acrocephalus seychellensis ) 40 Yes No n No n 34 116-118 western bluebird ( Sialia Mexicana ) 43.1 Yes No No 35 119 red-backed fairy-wren ( Malurus melanocephalus ) 62.7 d No … No 36 120 splendid fairy-wren ( Malurus splendens ) 80.6 e No No Yes 37 121-123 superb fairy-wren ( Malurus cyaneus ) 84.4 f No No Yes 38-40 124-128 Australian magpie ( Gymnorhina tibicen dorsalis ) 85.0 g No ...... 41 129-131
116
a Extra-group paternity (EGP), as the percentage of nests that contain at least one offspring sired by a male outside the cooperatively breeding group. We used extra-group instead of extra-pair paternity to avoid complications of shared reproduction within groups in species with more complex mating systems (see e.g., Cockburn 1998). b Increase in number of fledglings per nest. c It is unclear how many nests contain offspring sired by subordinate males and hence level of extra-pair paternity is given. d Only EPP given, and this value is probably an overestimation of the rate of EGP. However, as within-group EPP is < 2% (Webster et al. 2008), this overestimation is marginal. e EGP was not given for one population but EPP was also very high (Webster et al. 2004). f The average of three studies in the same population. g EGP was also high but slightly lower in another subspecies Gymnorhina tibicen tyrannical (Durrant & Hughes 2005). h There was a positive correlation between group size and number of fledglings. The author states, based on a pairwise comparison of reproductive success of groups that change in composition, that there is no positive effect. However, as Dickinson & Hatchwell (2004) outline, pairwise comparison likely underestimates true effects, and therefore we used the results of the correlative approach. i There was a positive effect of group size in another population for which rates of EGP were unknown (Nakamura 1998). j Subordinates have a positive effect on reproductive success of yearling females (Magrath 2001). k No distinction was made between adult subordinates and breeders, as both were included in analyses of the effects of group size on survival. l Clear pattern but not quite significant at the 5% level. m Subordinates increase breeder male survival when broods are large (Meade et al. 2010). n Negative effect of subordinates.
References 1. Lundy et al. 1998, 2. Quinn et al. 1999, 3. Woxvold & Mulder 2008, 4. Legge & Cockburn 2000, 5. Bruce et al. 1996, 6. Covas et al. 2006, 7. Heinsohn et al. 2000, 8. Faaborg et al. 1995, 9. Haydock et al. 2001, 10. Mumme et al. 1985, 11. Hartley et al. 1995, 12. Brooke & Hartley 1995, 13. Haydock et al. 1996, 14. Burke et al. 1989, 15. Haig et al. 1993, 16. Haig et al. 1994, 17. Rabenold et al. 1990, 18. Wrege & Emlen, 1987, 19. Kingma et al. 2009, 20. Jones et al. 1991, 21. Baglione et al. 2002, 22. Poldmaa et al. 1995, 23. Conrad et al. 1998, 24. Rubenstein, 2007a, 25. Temple et al. 2009, 26. Hatchwell et al. 2002, 27. Seddon et al. 2005, 28. Blackmore & Heinsohn 2008, 29. Williams 2004, 30. Townsend et al. 2009, 31. Du & Lu, 2009, 32. Whittingham et al. 1997, 33. Berg, 2005, 34. Richardson et al. 2001, 35. Dickinson & Akre 1998, 36. Webster et al. 2008, 37. Brooker et al. 1990, 38. Mulder & Magrath 1994, 39. Double & Cockburn 2000, 40. Green et al. 2000, 41. Hughes et al. 2003. 42. Wright 1998, 43. Zahavi, 1974, 44. Ridley 2007, 45. Brown 1975, 46. Zahavi 1990, 47. Franzreb 2007, 48. Mumme 1992, 49. Stallcup & Woolfenden 1978, 50. Woolfenden & Fitzpatrick 1990, 51. Woolfenden 1981, 52. Woxvold & Magrathm 2005, 53. Parry 1973, 54. Legge 2000b, 55. Sloane 1996, 56. Covas R, personal communication, 57. Covas et al. 2008, 58. Heinsohn R, personal communication, 59. Heinsohn 1995, 60. Beck & Heinsohn 2006, 61. Heinsohn 1992, 62. Boland et al. 1997, 63. Faaborg et al. 1980, 64. Faaborg 1986, 65. Faaborg & Bednarz 1990, 66. Koenig 1981, 67. Stacey 1979, 68. Koenig & Mumme 1987, 69. Koenig & Stacey 1990, 70. Davies et al. 1995, 71. Austad & Rabenold 1985, 72. Austad & Rabenold 1986, 73. Rabenold 1990, 74. Davies & Houston 1986, 75. Davies 1986, 76. Davies & Hatchwell 1992, 77. Davies 1990, 78. Davies 1992, 79. Houston & Davies 1985, 80. Khan & Walters 2002, 81. Lennartz et al. 1987, 82. Lennartz & Harlow 1979, 83. Neal et al. 1993, 84. Heppell et al. 1994, 85. Rabenold 1984, 86. Emlen & Wrege 1991, 87. Emlen & Wrege 1989, 88. Emlen 1990, 89. Emlen 1981, 90. Emlen 1990, 91. This study, 92. Lessells 1990, 93. Canestrari et al. 2008, 94. Dow & Whitmore 1990, 95. Poiani 1993, 96. Rubenstein 2007b, 97. Temple 2005, 98. McGowan et al. 2003, 99. Gaston 1973, 100. Hatchwell et al. 2004, 101. Seddon et al. 2003, 102. Blackmore & Heinsohn 2007, 103. Eguchi et al. 2007, 104. Brown et al. 1982, 105. Brown & Brown 1981, 106. Williams et al. 1994, 107. Williams 2000, 108. Williams & Hale 2006, 109. Lawton & Guindon 1981, 110. Caffrey 1999, 111. Caffrey 2000, 112. Magrath 2001, 113. Magrath & Yezerinac 1997, 114. Langen & Vehrencamp 1999, 115. Innes & Johnston 1996, 116. Komdeur 1994, 117. Komdeur 1996a, 118. Brouwer et al. 2006, 119. Dickinson et al. 1996, 120. Varian-Ramos et al. 2010, 121. Rowley & Russell 1997, 122. Russell & Rowley 1988, 123. Rowley & Russell 1990, 124. Cockburn et al. 2008, 125. Russell et al. 2007, 126. Green et al. 1995, 127. Nias 1986, 128. Nias & Ford 1992, 129. Finn & Hughes 2001, 130. Durrant 2004, 131. Veltman 1989.
117
Discussion
Our results show that subordinates contribute substantially to productivity and survival of breeders in M. coronatus . Larger groups fed nestlings at higher rates and produced more fledglings per nest, while breeders in groups reduced their work load, resulting in higher survivorship. Benefits of cooperative breeding appear therefore to differ substantially from closely related fairy-wren species. These findings and their implications are discussed in turn below.
Benefits of cooperative breeding in M. coronatus
Many explanations for the evolution and maintenance of cooperative breeding require a positive effect of subordinate individuals on reproductive success or survival of breeders (Brown 1987, Cockburn 1998, Emlen & Wrege 1989). In general, correlational investigation of effects of subordinates can be problematic (see Cockburn et al. 2008), but several lines of evidence suggest that the positive effects on productivity and survival presented here are a direct result of subordinate contributions: (1) at nests where more than one subordinate assisted, total feeding rates per nest were enhanced, resulting in increased fledgling production (Fig. 5.2), either through reduced nestling starvation or reduced partial predation associated with nest attendance. (2) Females did not increase productivity by adjusting clutch size to expected care (e.g., Davies & Hatchwell 1992, Woxvold & Magrath 2005). (3) Not all subordinates contributed to nestling feeding, and the effect size of number of contributors (helpers) was larger than the effect size of group size (Table 5.1b). This result suggests that increased fledgling production is a direct result of help with nestling feeding, rather than due to consistently high productivity of certain breeders and/or territories (e.g., Cockburn et al. 2008, Woxvold & Magrath 2005). (4) Breeders in groups reduced their effort by on average 20-30% compared to breeders in pairs, regardless of the number of subordinates (Fig. 5.3). If parental care is costly (Bryant 1988, Clutton-Brock 1991, Hatchwell 1999, Owens & Bennett 1994), this reduction in work load is likely to explain the positive correlation between group size and breeder survival (Fig. 5.3). Indeed, although the relation between number of actual contributors and survival was statistically not significant, probably due to reduced sample sizes, survivorship of male and female breeders was directly
118
negatively related to their own nestling feeding rate at the previous nest (Fig. 5.4). If kin selection is the selective force driving the evolution and maintenance of cooperative breeding, subordinates must increase the production of kin. This assumption is fulfilled in M. coronatus , as pair-bonds are long-lasting and extra-pair paternity is rare (4.4% of offspring; chapter 2), so in most cases subordinates helped their parents raise their siblings. Since subordinates also enhanced productivity (see above) in M. coronatus, kin- selected benefits may arise through increased production of full siblings (Fig. 5.2) as well as enhanced survival of parents (Fig. 5.3). Nonetheless, not all subordinates feed nestlings, and not all subordinates are related, so additional benefits for subordinates are possible (see Cockburn 1998): for instance, subordinate individuals may gain future benefits by enhancing group size (Kokko et al. 2001) or by ensuring a higher likelihood of obtaining a breeding position later (Ligon & Ligon 1978). Direct benefits may also play a role in helping behaviour in some cases, as subordinate male M. coronatus can (occasionally) obtain paternity in the nest (see chapter 2; Kingma et al. 2009), or subordinates may directly benefit from staying in the territory (Gaston 1978, Kokko et al. 2002).
The relative importance of compensatory and additive effects
In M. coronatus, patterns of feeding rates correspond to patterns of fledging success (Fig. 5.2), such that groups with two or more subordinates had higher feeding rates and greater reproductive success (additive effects) compared to pairs and groups with only one subordinate. Surprisingly, breeders reduced their feeding rates (compensatory effects) when accompanied by only one subordinate, thereby equalizing total feeding rate and fledging success. Why would breeders in small groups reduce feeding rates rather than attempt to improve nesting success? It could be that in the long-lived, multi-brooded M. coronatus , breeder investment strategies depend heavily on future reproduction (Ghalambor & Martin 2001, Trivers 1972), rather than on maximizing the outcome of the current brood, in contrast to species that have only one brood per year. It would therefore be interesting to explore whether the relative importance of current and future breeding (including number of annual breeding attempts) correlates with breeder investment strategies among cooperative breeders (see Hatchwell 1999, for such an approach with survival rates and nestling starvation).
119
Benefits of cooperative breeding: a role for extra-group mating?
The monogamous M. coronatus differs substantially from its promiscuous sister species M. melanocephalus, M. cyaneus and M. splendens in benefits of cooperative breeding. Despite long-term studies, no clear effects of subordinates on fledging success have been reported in these Malurus species. Additionally, male survival, investigated in two of these species, were unaffected by the presence of subordinates (see Table 5.2 for details). We propose that these inter-specific differences may result from differences in investment by subordinates associated with relatedness to the offspring, which is lower in the promiscuous species. Our comparative analyses of 37 cooperative breeders showed that, similar to the pattern observed among the four fairy-wren species, profound differences in productivity and male breeder survival were associated with levels of extra-group mating (Table 5.2). These reduced effects of subordinates in cooperatively breeding birds with greater levels of EGP (and hence lower levels of relatedness of subordinates to nestlings), are consistent with the predictions of kin-selection theory that subordinates adjust their investment to relatedness of the brood (Emlen 1997a, Griffin & West 2003, Hamilton 1964, Komdeur 1996a). Future studies that explore whether there are indeed inter-specific differences in subordinates investment due to differences in levels of EGP would therefore be worthwhile. The question remains why subordinates help in cooperatively breeding species with high rates of EGP. Indeed, the incidence of cooperative breeding in birds declines with increasing levels of extra-pair paternity (Boomsma 2007). Cooperative breeding in promiscuous species may be maintained if direct benefits also play a role (e.g., Richardson et al. 2002), or subordinates assist minimally in order to ‘pay-to-stay’ (e.g., Kokko et al. 2002). Direct benefits for subordinates, in addition or alternative to kin-selected benefits, are therefore interesting to investigate in more detail among cooperatively breeding species, and in promiscuous species particularly .
Conclusions
Our study showed that subordinates simultaneously increase overall productivity and enhance breeder survival in M. coronatus , and that these effects most likely were directly mediated by subordinate individuals’ helping behaviour. Levels of extra-group paternity are low, so most subordinates in
120
our population are philopatric offspring of the breeders and full siblings of the nestlings they are caring for. Therefore, indirect inclusive fitness benefits, arising through enhanced productivity of close kin and their parents’ increased survival, may be one reason why subordinate M. coronatus help. Benefits of cooperative breeding are far less clear in the promiscuous fairy- wrens, as in those species subordinates do not affect breeder male survival and nesting success. We showed, based on an extensive review (Table 5.2), that this pattern is rather general among species. In unfaithful species, the benefits of cooperative breeding are less prevalent than in faithful species, probably because of reduced investment by subordinates associated with lower potential for kin-selection. Since extra-group mating could significantly alter benefits of cooperative breeding for both breeders and subordinates, incorporating these ideas in the (theoretical) framework of cooperative breeding (Hatchwell 1999), may generate important insights into the evolution and maintenance of the breeding system.
Acknowledgements
We are grateful to S. Legge, S. Murphy and other staff at Mornington Wildlife Sanctuary, and to the Australian Wildlife Conservancy. In addition, we thank our excellent team of field assistants and in particular MSc students R. Klooster and J. Rijpstra. R. Covas and R. Heinsohn kindly provided unpublished data. Constructive comments from B. Hatchwell, the editors, and an anonymous referee improved the manuscript. This research was funded by the ‘Sonderprogramm zur Förderung hervorragender Wissenschaftlerinnen’ of the Max Planck Society (to AP). All fieldwork was performed with permission from the Max Planck Institute for Ornithology Animal Ethics Committee, the Australian Wildlife Conservancy, the Western Australia Department of Conservation and Land Management (licences BB002178 and BB002311), and the Australian Bird and Bat Banding Scheme (Authority 2230 and 2073).
121
122
Chapter 6
No evidence for offspring sex-ratio adjustment to social or environmental conditions in cooperatively breeding purple-crowned fairy-wrens
Sjouke A. Kingma, Michelle L. Hall & Anne Peters
Behavioral Ecology and Sociobiology (2011): in press
123
124
Abstract
When fitness returns or production costs vary between male and female offspring, selection is expected to favour females that adjust offspring sex ratio accordingly. However, to what extent vertebrates can do so is the subject of ongoing debate. Here, we explore primary sex ratios in 125 broods of cooperatively breeding purple-crowned fairy-wrens Malurus coronatus . We expected that females might adjust offspring sex ratio because this passerine species experiences considerable variation in social and environmental conditions. (1) However, although helpers substantially increase parental fitness, females (particularly in pairs and small groups) did not overproduce philopatric males ( helper-repayment hypothesis ). (2) Sex-ratio adjustment based on competition among individuals ( helper-competition hypothesis ) did not conceal helper-repayment effects or drive sex allocation on its own: while high quality territories can accommodate more birds, brood sex ratios were independent of territory quality, alone or in interaction with group size. (3) Additionally, males are larger than females and are possibly more costly to produce ( costly-sex hypothesis ), and (4) female offspring may benefit more from long-term effects of favourable conditions early in life ( Trivers-Willard hypothesis ). Nonetheless, large seasonal variation in food abundance was not associated with a consistent skew in primary sex ratios. Thus, overall, our results did not support the main hypotheses of adaptive sex-ratio adjustment in M. coronatus . We discuss that long-term differential costs and benefits may be insufficient to drive evolution of primary sex-ratio manipulation by M. coronatus females. More investigation is therefore needed to determine the general required sex-differences in long-term fitness returns for mechanisms of primary sex-ratio manipulation to evolve.
125
126
Introduction
Whether females can adaptively manipulate offspring sex ratio at conception is, despite long-standing interest (Charnov 1982, Fisher 1930, Frank 1990, Hamilton 1967, Seger & Stubblefield 2002, Trivers & Willard 1973, Williams 1979), still highly debated (e.g., Clutton-Brock 1986, Cockburn et al. 2002, Komdeur & Pen 2002, Krackow 1999, West et al. 2005). Theory predicts that selection should favour mothers that adaptively adjust the sex ratio of offspring to conditions that affect prospects of future survival or reproductive performance differently between the sexes (Charnov 1982, Leimar 1996, Trivers & Willard 1973). Additionally, females are expected to adjust offspring sex ratio when direct reproductive costs or benefits depend on the sex of their offspring, for instance when one sex is more costly to produce (Fisher 1930, Ligon & Ligon 1990, Patterson et al. 1980), or when, in cooperatively breeding species, helping behaviour is sex-biased (Emlen et al. 1986, Griffin et al. 2005, Komdeur 2004, Pen & Weissing 2000). However, despite clear theoretical predictions and some compelling empirical examples (e.g., Dijkstra et al. 1990, Kilner 1998, Komdeur et al. 1997, 2002, Legge et al. 2001, Ligon & Ligon 1990, Nager et al. 1999, Pike 2005, Sheldon et al. 1999, Thuman et al. 2003, Whittingham & Dunn 2000), the generality of adaptive sex allocation in vertebrates remains unclear, particularly in birds (see Clutton-Brock 1986, Cockburn & Double 2008, Cockburn et al. 2002, Frank 1990, Komdeur & Pen 2002, West & Sheldon 2002, Williams 1979,). Here, we explore the utility of four main hypotheses (see Table 6.1) in explaining primary sex ratio in purple-crowned fairy-wren Malurus coronatus broods. Because M. coronatus breeds under highly variable social and environmental conditions, we have a clear expectation that mechanisms exist to adaptively adjust primary sex ratio (Table 6.1). M. coronatus breed cooperatively, with breeding pairs being assisted in reproduction by one or more non-breeding subordinates (Kingma et al. 2009, 2010, Rowley & Russell 1997). Cooperative breeding has emerged as a model system to test adaptive sex-ratio manipulation in birds (Komdeur 2004) because parental fitness is often strongly dependent on the helping behaviour of subordinates (see Cockburn 1998, Hatchwell 1999, Kahn & Walters 2002, Kingma et al. 2010 for reviews on birds). As such behaviour is generally sex-biased (Cockburn 1998), the helper-repayment hypothesis (or local resource enhancement hypothesis ) predicts that females produce more offspring of the helping sex,
127
especially in unassisted pairs and small groups (Emlen et al. 1986, Gowaty & Lennartz 1985, Griffin et al. 2005, Lessells & Avery 1987, Pen & Weissing 2000). This hypothesis is mainly supported in species in which helpers have a strong effect on parental fitness (see Griffin et al. 2005). In M. coronatus , fitness effects of helpers are substantial: the presence of one helper enables parents to reduce costly nestling feeding and improve their survival, and additional helpers subsequently increase overall fledgling production (Kingma et al. 2010; see also Fig. 5.2 and 5.3). Since males stay longer in their natal group (Hall ML, Kingma SA, Peters A, unpublished data), M. coronatus females in pairs and small groups are expected to produce more male offspring. Alternatively, or additionally, when groups are large, females may reduce the costs of competition among group members by producing the dispersing sex, especially when resources in the territory are scarce ( helper- competition or local resource competition hypothesis , Clark 1978, Emlen 1997b, Julliard 2000). If this applies to M. coronatus , females in larger groups and relatively low-quality territories are expected to produce more female offspring. M. coronatus breed year-round in a tropical environment, where seasonal conditions (e.g., rainfall and food abundance) are highly variable (Shine & Brown 2008). If resource requirements differ between the sexes (costly-sex hypothesis ; Fisher 1930), or if future success depends more on early development in one sex than the other (like in species with sex- differences in reproductive skew; Trivers-Willard hypothesis ; Trivers & Willard 1973), offspring sex ratio should be biased in response to food abundance (e.g., Appleby et al. 1997, Kilner 1998, Ligon & Ligon 1990, Rubenstein 2007b). In M. coronatus , males are larger than females (Rowley & Russell 1997), so under good conditions sex ratios could be biased towards males. Alternatively, if the general pattern that reproductive skew among females is higher than among males in cooperative breeders (Hauber & Lacey 2005, Rubenstein & Lovette 2009) applies to M. coronatus , a skew towards females can be predicted when conditions are better (Rubenstein 2007b). Thus, if one or more of the above-mentioned hypotheses apply to M. coronatus , we expect brood sex ratios to vary with group size, territory quality, and/or seasonal conditions (see also Table 6.1).
128 Table 6.1. Summary of the predictions and assumptions of the main hypotheses for sex-ratio adjustment in cooperative breeders. For M. coronatus , we tested whether specific assumptions are met, identified species-specific predictions, and tested whether these predictions were supported. Prediction of sex-ratio adjustment specific to M. Evidence of adaptive sex-ratio Hypothesis Predictions Assumptions Assumptions met in M. coronatu s coronatus adjustment in M. coronatus a
Helper-repayment Bias towards offspring of the Helpers have a positive Yes, helpers increase breeder survival Sex ratio negatively correlated with No, sex ratio independent of number of (Gowaty & Lennartz helping sex, especially when no effect on breeder fitness and fledgling production (Kingma et number of subordinates b subordinates ( β=-0.01±0.08, Z=-0.2, p=0.85; 1985) helpers present (Griffin et al. 2005) al. 2010; Figs. 5.2 and 5.3) Fig. 6.2a)
Sex ratio negatively correlated with No, sex ratio independent of number of Helping and/or philopatry Yes, males more philopatric (more number of male subordinates, male ( β=0.002±0.09, Z=0.02, p =0.98) and is sex-biased likely to help) than females (this positively with number of female female ( β=-0.06±0.2, Z=-0.3, p=0.76) study) subordinates subordinates
Helper-competition Bias towards offspring of the Individuals compete for Presumed, territory quality is Territory quality (or territory quality No, sex ratio was independent of territory (Clark 1978) dispersing sex, when resources (limited) resources correlated with number of in interaction with number of quality (pandanus cover: β=0.0001±0.0001, are limited and/or there are subordinates (this study) subordinates already present) Z=0.9, p=0.39; Fig. 6.2b; territory size: β=- already subordinates present positively correlated with sex ratio 0.0009±0.001, Z =-0.6, p=0.57) and number Dispersal is sex-biased Yes, females disperse at a younger of subordinates (see above) age (Hall ML, Kingma SA, Peters A, unpublished data)
No, sex ratio independent of food Trivers-Willard Females in better condition Early conditions determine Unknown Seasonal food abundance (and abundance (β=0.01±0.07, Z=0.2, p=0.84; (Trivers & Willard produce offspring with highest adult success, and this number of helpers b) negatively Fig. 2c), pre-breeding rainfall 1973) variance in reproductive success effect is stronger in one of correlated with sex ratio (β=0.0002±0.0003, Z=0.6, p==0.53; Fig 2d), the sexes and number of helpers (0-1 helper: β=0.29±0.27, Z=1.1, p=0.28; 0-2+ helpers: Variance in reproductive Presumed: females in cooperative β=0.02±0.25, Z=0.03, p=0.92; 1-2+ helpers: success is larger in one sex breeders generally have higher β=-0.27±0.31, Z=-0.9, p=0.39) than the other reproductive skew than males (Hauber & Lacey 2005)
Costly-sex Bias towards the cheaper sex, or Sexes differ in size Yes, male nestlings on average 5% Seasonal food abundance (and No, sex ratio independent of food (Fisher 1930) sex ratio adjusted to resource heavier than females (this study) number of helpers) positively abundance , pre-breeding rainfall , and availability correlated with sex ratio number of helpers (see above and Fig. Larger offspring are more No evidence: proportion of male 6.2c,d) costly to produce nestlings did not affect feeding rates by the group or by breeder females (this study) a 125 broods from 61 females on 49 territories were included in all analyses of brood sex ratio, except for analyses with number of male and female subordinates (n = 123 broods from 61 females on 49 territories) and food abundance (n = 80 broods from 49 females on 41 territories).
b A distinction was made between subordinates (all non-breeders in the territory), and number of helpers (subordinates > 145 days), because birds younger than 145 days do not feed nestlings (Kingma et al. 2010, 2011b; chapter 4 and 5). Methods
Study species and field work
Between 2005 and 2010, we studied a population of M. coronatus along Annie Creek and the Adcock River at Mornington Wildlife Sanctuary in north-west Australia (17°31’S, 126°6’ E). This part of Australia is characterised by its tropical wet-dry climate (see e.g., Shine & Brown 2008). M. coronatus is insectivorous, and can breed year-round with a distinct peak in breeding activity during the wet-season (February-April), and a second peak in the late dry-season in good years (August/September-November; Hall & Peters 2009, Rowley & Russell 1993a, 1997). In each group, one male and female (breeders) form a socially monogamous pair, and between 40 and 70% of pairs are accompanied by on average between one and two (up to nine) subordinates of both sexes, of which most contribute to nestling feeding (Kingma et al. 2010, 2011b; chapter 4 and 5). Clutch size is generally three eggs, but can vary between 1 and 4 eggs (mean ± SE: 2.9 ± 0.05, n = 155), and incubation and nestling periods are both about 13-14 days (Hall ML, Kingma SA, Peters A, unpublished data). M. coronatus are riparian specialists with territories arranged linearly along the river, and birds rarely move further than about 20 meters from the dense riverside vegetation (Rowley & Russell 1993a). During year-round weekly censuses, we documented group size and social status of each uniquely colour-banded group member (Hall & Peters 2008, 2009, Kingma et al. 2009, 2010). In addition, nesting activity was monitored during each census by following the breeder female for at least 20- 30 minutes. Once found, nests were checked regularly to determine laying date, clutch size and number of nestlings. Territories are stable year-round, and boundaries are rather fixed so that these could easily be determined from movement patterns of birds in each group; changes were noted throughout the study. Nestlings were weighed to the nearest 0.05 g using a 10 or 30 g Pesola spring balance. Additionally, we took a small blood sample by venipuncture of the brachial vein when nestlings were 6-8 days old (rarely 4-5 or 9-11 days old). Dead nestlings and unhatched eggs were collected for later sex- determination (see below). Blood and tissue samples were stored in ethanol, Queen’s lysis buffer or Longmire’s buffer.
Nestling feeding watches
Details about nestling feeding watches are described in chapter 5 (Kingma et al. 2010). Briefly, we conducted 1 to 5 (mostly 4) feeding watches of approximately one hour, recording feeding rate to the nest for each individual group member. Feeding watches were conducted between approximately 4 and 10 days after hatching, alternating morning and afternoon.
Environmental conditions: food abundance, rainfall, and territory quality
Monthly arthropod abundance was determined using sweep-net collection in six territories from September 2007 onwards. On two afternoons in the middle of each month (three territories on each day), we conducted 3 sets of 15 sweeps in each territory, in high grass close to the creek or river, at the midpoint of the territories and about 10 m upstream and downstream. Arthropods were counted, and ‘monthly average number of arthropods per sweep’ was calculated as measure of monthly food abundance. Seasonal variation in food abundance is large, ranging from 0.3 to 5.8 arthropods per sweep per month (data collected over 33 months). Arthropod lengths were also measured (in mm); using monthly averages of summed length of all arthropods yielded very similar results (data not shown). Daily rainfall data were collected within the study area throughout the study period. For the months for which arthropod abundance data were available, we tested whether arthropod abundance correlated with cumulative rainfall over 1, 2, 4, 8, 12, 16, 20, 24, 28, and 32 weeks before egg- laying. Rainfall over the preceding 24 weeks (rs = 0.69, p < 0.001) best predicted arthropod abundance (1 week: rs = 0.05; 2 weeks: rs = -0.02; 4 weeks: rs = 0.10; 8 weeks: rs = 0.34; 12 weeks: rs = 0.49; 16 weeks: rs = 0.58; 20 weeks: rs = 0.65; 28 weeks: rs = 0.68; 32 weeks: rs = 0.65), and therefore cumulative rainfall over 24 weeks before egg-laying was used as measure of ‘pre-breeding rainfall’ in further analyses. Rainfall is seasonal, falling mostly between December and March each year (see also Shine & Brown 2008). Therefore, pre-breeding rainfall for the 125 nests included in the analyses of sex ratio (see below) was highly variable, ranging from 2.1 to 918.5 mm (see also Fig. 6.2d). We used the total cover of river pandanus ( Pandanus aquaticus ) for each territory and territory size as proxies for territory quality. Several observations show that M. coronatus is strongly dependent on river pandanus
131
vegetation (see also Rowley & Russell 1993a, 1997). First, on average 51% of time during the day is spent in pandanus (339 individual time budget observations covering in total 81 hours of observation on 102 birds; Kingma SA, Klooster RSL, Hall ML, Peters A, unpublished data). Second, all territories at our study site contain pandanus, and stretches of river without pandanus are unoccupied. Third, nests are almost exclusively (~ 95%) built in pandanus (Hall ML, Kingma SA, Peters A, unpublished data). Therefore, coordinates of all pandanus in the study area were mapped using Garmin eTrex handheld GPS units. For small patches, we estimated total length and width. For larger patches (i.e., longer than ~ 20 m), we calculated length from the GPS coordinates of start- and end-points, and estimated its width. Based on coordinates of territory boundaries, we calculated the total pandanus cover for each territory (in m 2). Pandanus can be distributed patchily, but also in large continuous stretches of over 100 meters long, and this distribution varies between territories, resulting in large variation in pandanus cover (range: 58-3,624 m 2 per territory; see also Fig. 6.2b). Territory size was expressed as length of the territory (range: 35 to 376 m), since width of the riparian strip does not vary greatly.
Molecular sexing
DNA was extracted from either egg-, tissue-, or blood-samples using extraction methods described in Richardson et al. (2001) or using DNeasy blood & tissue kits (Qiagen). Offspring were sexed using PCR-methods described in detail in Kahn et al. (1998). This method appeared reliable as only one of the 340 individuals (0.29%) that we could sex in the field based on plumage characteristics, was sexed incorrectly using molecular techniques.
Sex ratio
Sex ratios represent the proportion of males per brood. For analyses we used only broods for which all eggs could be sexed (i.e., primary sex ratio; hereafter ‘sex ratio’), to exclude potential effects of early sex-biased mortality. Broods that were first found when containing nestlings were only included when those contained three or four nestlings, because those broods almost certainly contained the complete clutches (four-egg clutches are relatively rare (i.e., ~ 16% of broods with known-clutch size; Hall ML, Kingma SA, Peters A, unpublished data), and we never found five-egg clutches; see also Rowley &
132
Russell 1993a). In total, 125 broods (372 offspring) from 61 females on 49 territories were included in the sex ratio analyses.
Statistical analyses
Statistical tests were performed using R 11.1 (R Development Core Team 2010). Null hypotheses were rejected when p < 0.05. Parameter estimates ( β) are given as mean ± standard error of the mean. To test whether overall sex ratio deviated from parity, we used an exact binomial test (e.g. Cordero et al. 2000, Legge et al. 2001). Additionally, we tested whether the observed frequency of nests with each possible number of males deviated from expected (using a χ2 test), based on the overall population sex ratio and binomial expectation (see Cockburn & Double 2008, Williams 1979), in broods with three eggs (n = 91; not enough broods were available with one (n = 1), two (n = 17) or four eggs (n = 16) for such a calculation). To explore hypotheses of adaptive sex-ratio manipulation, we used a two-step approach. First, we examined whether important assumptions of each hypothesis (i.e., sex-specific costs or benefits; as outlined in Table 6.1) were met in M. coronatus . Second, we examined whether offspring sex ratio varied with the corresponding factors, as predicted by these hypotheses (see Table 6.1 and below). To test which parameters (see below and Table 6.1) were associated with brood sex ratio, we fitted those variables as explanatory variables in generalised linear mixed models (GLMM) with a logit link function and binomial error, using the lme4 package (Bates & Sarkar 2007) in R. Sex ratio was set as the response variable, with brood size as the denominator (using the cbind command; see Wilson & Hardy 2002). Female identity nested within territory identity was always included as a random factor to correct for inclusion of multiple broods from most individual females and territories. In analyses with more than one explanatory variable, we included all variables and two-way interactions, and removed statistically non-significant terms (p > 0.05), starting with the variable with the highest p-value (and interactions were excluded before single variables). As none of the variables or interactions had a significant effect on sex ratios (see results), all presented results derived from models with the random terms and only a single explanatory variable, and interaction effects are not given.
133
Helper-repayment hypothesis To test the assumption of the helper-repayment hypothesis (see Table 6.1), we explored whether the incidence or extent of helping is sex-biased. First, we compared the number of male and female helpers at each nest (using a paired t-test). Only subordinates that were older than 145 days, defined as ‘helpers’, were included because only subordinates older than this age feed nestlings (Kingma et al. 2010, 2011b; chapter 4 and 5). Second, we tested whether nestling feeding rates differ between male and female helpers, by fitting a maximum likelihood model in the nlme package in R (Pinheiro & Bates 2000, Pinheiro et al. 2009). In total, feeding rates of 48 different helpers (32 males and 16 females) at 40 nests were determined, and 21 helpers were observed at more (subsequent) nests, resulting in a total of 256 individual feeding watches. Individual feeding rate (feeds per hour, square-root transformed to normalise model residuals) was included as the response variable, and sex of the subordinate as the explanatory variable. Only helpers that were related to both breeders were included in the analysis of feeding rates, because helpers reduced feeding rates when one or both breeders were replaced by unrelated breeders (Kingma et al. 2011b; chapter 4). Moreover, future helpers are most often (> 60% of all helpers) related to both breeders (since divorce rates, mortality rates and extra-pair paternity are low; Kingma et al. 2009, Table 2.1). We included nest identity and helper identity nested within nest identity as random factors, and we corrected for the significant effect of brood age (t = 4.3, p < 0.001), whether an observation was done in the morning or afternoon (t = 2.1, p = 0.041), and helper age category (more or less than a year old; t = 4.1, p < 0.001; see chapter 5 for details). To test the helper-repayment hypothesis, we fitted a GLMM with brood sex ratio as the dependent variable (see above), and total number of subordinates, number of male subordinates or number of female subordinates in the group as explanatory variables.
Helper-competition hypothesis To infer whether there is competition among birds in the group (a central assumption of the helper-competition hypothesis; Table 6.1), we first tested whether the number of subordinates at each nest correlated with territory quality. We fitted a GLMM with log link function and poisson error, with number of subordinates as the response variable, pandanus cover or territory size as explanatory variables, and territory identity as the random variable.
134
To test whether females adjusted brood sex ratios to the number of subordinates already present, and whether territories could accommodate more subordinates (based on territory quality, see below), we used GLMMs with sex ratio as the response variable (see above), and the explanatory variables: (1) pandanus cover or territory size, (2) the residuals of the regression between number of subordinates and pandanus cover or territory size, or (3) number of subordinates divided by m 2 pandanus, or by territory size.
Costly-sex and Trivers-Willard hypothesis We first investigated whether male nestlings are more costly to raise by comparing the body mass of male and female nestlings. In a paired t-test we compared average male and female nestling mass in each (mixed-sex) brood. As the effect was independent of brood size (not given), all mixed-sex broods were used. Second, we tested whether the number of male nestlings in three- egg broods affected total feeding rates of the group, or feeding rates of breeder females (163 feeding watches at 43 nests and 33 females in 30 different territories; sample size of other brood sizes were too small for similar analysis). We fitted maximum likelihood models with feeding rates (square- root transformed to normalise model residuals) as the response variable and number of male nestlings as the explanatory variable. For the analyses of group feeding rates, territory identity and nest identity nested within territory identity were included as random factors, and we corrected for the significant effects of wind intensity (the effect of strong wind as compared to no wind; t = -2.0, p = 0.043), number of helpers (significant difference between groups with no, and 2+ helpers: t = 3.2, p = 0.009), and the interaction between brood-age and whether an observation was done in morning or afternoon (t = -2.5, p = 0.013; see also Kingma et al. 2010; chapter 5). For the analyses of female feeding rates, female identity and nest identity nested within female identity were included as random factors, and we corrected for significant effect of number of helpers (0-1 helpers: t = -3.1, p = 0.018; 0-2+ helpers: t = -2.1, p = 0.069) and whether an observation was done in the morning or afternoon (t = 3.5, p < 0.001; the other variables had no significant effect on female feeding rates). To test whether brood sex ratio varies with variables that can be expected to affect nestling production costs or conditions (see Table 6.1), we fitted a GLMM with brood sex ratio as the dependent variable (see above),
135
and the number of helpers, and pre-breeding rainfall or arthropod abundance in the month of egg-laying (subset of 80 broods of 49 females on 41 different territories) as explanatory variables. Results were similar when using arthropod abundance in month of hatching, and therefore not given. Number of helpers was categorised as 0, 1, and 2+ helpers, as feeding rates are mainly enhanced when more than one helper is present, and parents reduce feeding rates when assisted by one or more helpers (Kingma et al. 2010; chapter 5).
Results
The overall primary offspring sex ratio was slightly male biased, as 52.7% of offspring were male (196 of 372), but this bias was not statistically significant (binomial test: p = 0.33; 95% confidence interval: 47.5-57.9%). The frequency distribution of brood sex ratios was not different from expected based on the overall offspring sex ratio (Fig. 6.1).
Helper-repayment hypothesis
Helping in M. coronatus is sex-biased, as more male (0.68 ± 0.10) than female (0.28 ± 0.05) helpers were present at the nests included in sex ratio analyses (paired-sample t-test: t = 3.9, n = 123 nests, p < 0.001). Male helpers (2.35 ± 0.19 feeds/h; 179 feeding watches) fed nestlings on average slightly more than female helpers (2.02 ± 0.29 feeds/h; 77 feeding watches), but this effect was not statistically significant (t = 1.0, p = 0.28, β = 0.18 ± 0.16). The total number of subordinates in the group did not correlate with brood sex ratio (Table 6.1 and Fig. 6.2a), nor did the number of male or female subordinates (Table 6.1).
Helper-competition hypothesis
Territories with more pandanus have more subordinates, as the number of subordinates increases on average by 0.50 ± 0.02 per 1000 m 2 pandanus (Z = 2.6, p = 0.010). Similarly, larger territories had on average more subordinates (Z = 2.3, p = 0.020, β = 0.38 ± 0.16 more subordinates per 100 m). Brood sex ratio was independent of pandanus cover or territory size (Table 6.1 and Fig. 6.2b). Similarly, sex ratio was not correlated with residual group size (obtained from the correlation between pandanus cover and
136
number of subordinates: Z = -0.3, p = 0.76, β = -0.04 ± 0.13; or the correlation between territory size and number of subordinates: z = -0.2, p = 0.83, β = -0.03 ± 0.12,). Similarly, the number of subordinates per m 2 pandanus (z = -0.6, p = 0.54, β = -15.37 ± 24.99) and number of subordinates per m territory (z = 0.7, p = 0.44, β = 5.08 ± 6.57) did not affect brood sex ratio.
Costly-sex and Trivers-Willard hypothesis
Male nestlings were on average 4.82 ± 1.10% (range: -18.44-62.07%) heavier than females in the same nest (t = 4.6, n = 109 broods, p < 0.001). When excluding three nests with a large difference between males and females, this pattern remained very similar (t = 4.3, n = 106, p < 0.001, average difference = 3.53 ± 0.83%, range: -18.44-24.19%). However, despite males being heavier, nestling feeding rates by groups ( β = -0.04 ± 0.08, Z = -0.5, p = 0.63) or breeder females (Z = -0.3, p = 0.78, β = -0.02 ± 0.07) were not higher in broods with more male nestlings. Brood sex ratio was independent of pre-breeding rainfall, monthly arthropod abundance at laying and hatching, and number of helpers (see Table 6.1 and Fig. 6.2).
Figure 6.1. Expected and observed frequency of each possible number of male offspring per brood in 91 three-egg broods, based on the average offspring population sex ratio of 0.527. Observed and expected frequency do not differ significantly ( χ23 = 2.1, p = 0.56), suggesting absence of precise control of primary sex ratios.
50 Expected Observed
40
30
Frequency 20
10
0 0 1 2 3
Number of males per brood
137
Figure. 6.2. Primary brood sex ratios are independent of (a) number of subordinates in the group, (b) territory quality (total m 2 pandanus cover), (c) seasonal food abundance (average monthly number of arthropods) and (d) pre-breeding rainfall (cumulative rainfall over 24 weeks before egg-laying (for details see Methods). Dot sizes in (a) reflect number of broods, ranging from one (smallest dots) to 16 (largest dot) broods.
a) b) 1.0
0.8
0.6
0.4
0.2
0.0 Sex-ratio (proportion males) Sex-ratio
0 2 4 6 0 1000 2000 3000 4000 Number of subordinates Territory quality (m 2 pandanus)
c) d) 1.0 1.0
0.8 0.8
0.6 0.6
0.4 0.4
0.2 0.2
0.0 0.0 Sex-ratio (proportion males) Sex-ratio (proportion males) Sex-ratio
0 1 2 3 4 5 6 0 200 400 600 800 1000 Food abundance Pre-breeding rainfall (mm) (mean number of arthropods)
138
Discussion
The observed patterns of primary offspring sex ratio did not support any of the tested hypotheses of adaptive sex-ratio adjustment in M. coronatus (Table 6.1). This is surprising because there are several potential benefits that could drive the evolution of sex-ratio manipulation, and sample sizes are relative large (see Ewen et al. 2004). Potential constraints on the evolution of sex- ratio manipulation, and lack of support for current hypotheses, are discussed below.
Helper-repayment and competition
In M. coronatus , females in pairs or small groups did not produce more males (Fig. 6.2a), though male helpers are more common than female helpers (on average around 0.7 male vs. 0.3 female helpers per group). This result is different from some cooperatively breeding species (e.g., Griffin et al. 2005, Komdeur et al. 1997), but very similar to three other fairy-wrens ( M. cyaneus , Cockburn & Double 2008; M. leucopterus, Rathburn & Montgomerie 2005; M. melanocephalus, Varian-Ramos 2010). In at least one of those species, M. melanocephalus, helpers provide no obvious fitness advantage (Varian-Ramos et al. 2010), consistent with the helper-repayment hypothesis (Griffin et al. 2005). Helper M. coronatus however, have considerable positive effects on fitness of breeders, perhaps much larger than in its congeners (see Kingma et al. 2010; Figs. 5.2 and 5.3). By assisting in nestling feeding, helpers enhance fledging success and enable breeders to reduce their feeding rate, which directly affects survival rates of breeders (Kingma et al. 2010; Fig. 5.4). Nonetheless, despite clear predictions that especially unassisted females should produce more males (Griffin et al. 2005), our results did not support the helper-repayment hypothesis. One reason why females do not produce more offspring of the more helpful sex may be that competition between group members outweighs the overall benefits (helper-competition hypothesis; Clark 1978, Komdeur et al. 1997, Julliard 2000; Table 6.1). We did however not find such a pattern: although groups are on average larger in better territories in M. coronatus , females did not produce more males when territories could probably maintain more subordinates, or produce more females when territories were presumably saturated (Fig. 6.2b).
139
Although helping behaviour is often sex-biased, fitness returns may differ only marginally between males and females. In M. coronatus , helping is male-biased whereas dispersal is female-biased, but females can still stay, help and compete for resources in their natal group. It is unclear what level of sex-bias in those features is needed for selection to favour sex-biased offspring production. In fact, especially in species with multiple breeding attempts per year, like M. coronatus , helping by females may be sufficient to reduce selection on sex-determination mechanisms. Hence, the magnitude of sex-bias in philopatry and helping behaviour is important to consider in empirical and comparative studies of sex allocation in cooperative breeders, to reveal the ultimate general utility of the helper-repayment hypothesis.
The costly sex, environmental conditions, and future benefits
As opposed to a large number of studies that show a strong effect of food abundance on sex ratio (e.g., Byholm et al. 2002, Clout et al. 2002, Kilner 1998, Korpimäki et al. 2000, Ligon & Ligon 1990, Rubenstein 2007b), we did not find evidence that variation in food abundance affects sex ratios in M. coronatus (Fig. 6.2c,d). This is surprising, considering that current theory predicts such a relationship (Table 6.1), especially in species that can breed under highly variable seasonal conditions (see Rubenstein 2007b). Male nestling M. coronatus are significantly larger, and therefore perhaps more costly to raise than females (e.g., Andersson et al. 1993, Fiala & Congdon 1983; but see below). Moreover, larger nestlings could be more sensitive to food shortage (Korpimäki et al. 2000, Torres & Drummond 1999), an additional reason why males should mainly be produced at peak food abundance. Nonetheless, our results did not support the costly-sex hypothesis (see Table 6.1). Nestling sexual size-dimorphism in M. coronatus (around 5%) could be too small to substantially increase production costs, so that selection on sex-ratio manipulation is limited (see also Rösner et al. 2009). Indeed, sex- ratio adjustment to seasonal food abundance is most obvious in species with large sexual dimorphism in body mass (e.g., 23% dimorphism in common grackles Quiscalus quiscula (Howe 1977), 24-42% in European kestrels Falco tinnunculus (Dijkstra et al. 1988, 1990), 47% in Peregrine falcons Falco peregrinus (Olsen & Cockburn 1991), ~20% in tawny owls Strix aluco (Appleby et al. 1997, Sunde et al. 2003, but see Wiebe & Bortolotti 1992: dimorphism 5-13% in American kestrels Falco sparverius ). In agreement with relatively low sexual dimorphism, we did not find evidence for clear sex-
140
differences in production costs in M. coronatus , as broods with more males were not fed more often. This, however, does not preclude the possibility that males may need more food at earlier nestling ages or require higher-quality food. Therefore, additional work is needed to determine what magnitude of sex-differences in production costs constitutes a sufficient selective force for sex-ratio manipulation (see Table 6.1). As an alternative, the Trivers-Willard hypothesis predicts that females should produce offspring of the sex with greatest variance in reproductive success under good conditions (Trivers & Willard 1973, Table 6.1). We did not find support for this hypothesis, as large variation in food abundance does not affect offspring sex ratio. However, we cannot exclude that a central assumption of this hypothesis is not upheld. Although females may benefit more than males from being larger or in better condition for dispersal and competition over breeding positions, and skew in cooperatively breeding birds is generally larger among females (Hauber & Lacey 2005, Rubenstein 2007b, Rubenstein & Lovette 2009), we cannot identify reproductive skew among males and females at this stage. In fact, although the population sex ratio is slightly male-biased (proportion males = 0.57) indicating higher female than male mortality, reproductive skew may be rather similar among sexes in this long-lived, genetically nearly monogamous species (Kingma et al. 2009; chapter 2). Thus, lack of skew in sex ratio could reflect the lack of sex differences in effects of early development on future success.
Implications
In M. coronatus , we did not find any support for the main hypotheses for adaptive primary sex-ratio manipulation, supporting the general view that these hypotheses are not universally applicable to birds (Charnov 1982, Clutton-Brock 1986, Cockburn et al. 2002, Komdeur & Pen 2002, Krackow 1999, West et al. 2005). Although effects of different selection pressures could be concealed when they act simultaneously in opposite direction, the most plausible explanation for this result is that the sex-bias in long-term benefits is too small for adaptive mechanisms of offspring sex-ratio adjustment to have evolved in M. coronatus . This idea may encourage future empirical and comparative studies to focus on the magnitude of long-term fitness consequences required to drive the evolution of sex-ratio adjustment (e.g., the magnitude of sex bias in helping or condition-dependent future success). Such an approach is conspicuously limited so far, especially in irregularly breeding
141
species (see e.g., Appleby et al. 1997, Bradbury et al. 1997, Komdeur & Pen 2002, Komdeur et al. 1997). Hence, negative results form important pieces of the puzzle of whether and how birds adjust sex ratios, and publishing such results is important to determine the general utility of different hypotheses of sex-ratio manipulation in birds (see e.g., Cockburn & Double 2008, Ewen et al. 2004, Griffin et al. 2005, West et al. 2002, West & Sheldon 2002).
Acknowledgements
We are extremely grateful to S. Legge, S. Murphy and other staff at Mornington Wildlife Sanctuary, to the Australian Wildlife Conservancy for logistical support, and to our team of field assistants for their hard work in the field. We thank E. Fricke for molecular sexing, J. Heathcote for collecting rainfall data, K. Delhey for discussion, and C. Brown and two anonymous reviewers for comments. All fieldwork was performed with permission from the Max Planck Institute for Ornithology Animal Ethics Committee, the Australian Wildlife Conservancy, the Western Australia Department of Conservation and Land Management (licences BB002178 and BB002311), and the Australian Bird and Bat Banding Scheme (Authority 2230 and 2073). The research was funded by the ‘Minerva Sonderprogramm zur Förderung hervorragender Wissenschaftlerinnen’ of the Max Planck Society (to AP).
142
143
144
Chapter 7
General conclusion
145
146
Natural selection and reproductive strategies in M. coronatus
Natural selection theory states that (heritable) behavioural, morphological, and physiological traits in animals have evolved as strategies that maximise individuals’ survival and/or reproductive success (Darwin 1859, Williams 1966, Endler 1986). In this light, the main topics of this thesis, extra-pair (EP) mating and cooperation among individuals have puzzled evolutionary biologists for decades: these (costly) behaviours are common in the animal kingdom, but why individuals do so is often unclear (e.g., Dugatkin 2002, Arnqvist & Kirkpatrick 2005). In this study, I aimed to unravel how different factors that may determine costs, benefits, and/or constraints of EP mating behaviour and cooperative breeding shape such behaviour in M. coronatus . Here I will, based on a summary of the main results, show how EP mating and cooperative breeding can be beneficial (and therefore be favoured by natural selection). Additionally, I will discuss that it is important to incorporate knowledge of both the social and genetic mating system when aiming to understand the evolution of reproductive behaviour.
The evolution of extra-pair mating in Malurus coronatus
What determines whether females engage in extra-pair mating?
Why females engage in EP mating has intrigued evolutionary biologists ever since molecular tools became available that could be used for accurate paternity analyses (Griffith et al. 2002). I used an inter- and intra-specific approach to test how factors that determine the costs, benefits and constraints affect the incidence of EP mating in M. coronatus. In chapter 2, I show that the socially monogamous fairy-wrens offer an ideal system to study the underlying components of EP mating: while all studied fairy-wren species are highly promiscuous (having 40-80% EP offspring; Table 2.1), EP mating in M. coronatus is rare (4-6%). Phylogenetic inference suggests that EP mating was ‘lost’ in M. coronatus . This is surprising as comparison of the life history of the different species appears that the species are rather similar, and that differences in costs and benefits are unlikely to explain the large inter-specific difference in levels of extra-pair paternity (EPP; see Table 2.1). This result suggests that evolutionary inertia does not necessarily apply to EP mating, and that genetic mating systems may be more flexible than
147
currently assumed (see Arnold & Owens 2002). Nonetheless, the question remains what contemporary variables drive such flexibility and explain reduced levels of EPP in M. coronatus . Relatively limited numbers of accessible fertile EP males resulting from low breeding density and synchrony (see Table 2.1) may be one reason for low levels of EPP in M. coronatus . In line with this, I showed in chapter 3 that variation in these ecological variables partly predicts whether broods contain EPP within my study population of M. coronatus , suggesting that constraints on EP mating do apply in this species. Nonetheless, females that obviously benefit from EP mating were not prevented from doing so. Consistent with many bird species (see Kempenaers 2007), high genetic similarity of the social parents has a negative impact on the hatchability of eggs in M. coronatus (Fig. 3.2). As predicted, this constitutes a strong selective force on EP mating, as females in incestuous pairs had EP offspring in half of the broods, sired by unrelated EP males. This rate of EPP is almost 15 times higher than rates of EPP by females with an unrelated partner, and comparable with rates of EPP of other fairy-wren species (Fig. 3.2). M. coronatus females in incestuous pairs appear to overcome the constraints of low breeding density and synchrony by synchronising their breeding attempt with direct neighbours, probably to ensure the fertility of targeted EP males (Fig. 3.6). The finding that females may actively affect factors that would otherwise constrain them in EP mating illustrates that the ‘constrained female hypothesis’ (Gowaty 1996) should be adopted with caution. In fact, because the relationship between ecological variables and the incidence of EPP may depend on the (variable) magnitude of benefits of EP mating, this may explain why inter- and intra-specific studies of the general role of ecological variables have yielded mixed results (see Westneat & Stewart 2003). In order to explain variation in EP mating and its evolution in general, further studies are needed that similarly explore how the complex interplay between the suite of benefits, costs and/or constraints determines EP mating within specific systems, and how individuals can actively affect this interplay. Additionally, to determine the role of single components (e.g., ecological factors) in EP mating, it would be worthwhile to perform experiments where other components (e.g., the benefits) of EP mating can be kept constant.
148
The evolution of cooperative breeding in Malurus coronatus
Why do some individuals assist in raising other individuals’ offspring?
The question how costly cooperative behaviour can be favoured by natural selection has evoked a lively research field, but there is no general consensus why helpers engage in seemingly altruistic helping in cooperatively breeding species. Helping behaviour in M. coronatus consists most obviously of nestling feeding, which I have shown is costly (as it reduces breeder survival and older helpers feed more often than young helpers). I show in chapter 4 and 5 that M. coronatus helpers are usually retained offspring, and therefore often help their parents. Because helpers improve the reproductive success (Fig 5.1) and survival of breeders (Fig. 5.2), indirect fitness benefits form one explanation for helping behaviour in this species. Further evidence for the importance of kin-selected benefits comes from the fact that helpers invest more in related offspring than in less related offspring (Fig. 4.1). Nonetheless, in contrast with the kin selection theory, also unrelated helpers invest in offspring care. I suggest that the magnitude of future direct benefits for M. coronatus helpers also determines who helps and how much: in an integrated analysis with kin- selected benefits, I show that helpers feed more when they are more likely to inherit a breeding position (Fig. 4.1), in which case helping can be explained as mechanism to augment the group with future helpers (Kokko et al. 2001). Thus, besides indirect benefits, also direct benefits are important in explaining helping behaviour in M. coronatus . The study presented in chapter 4 also reveals that specific benefits may remain concealed when investigated in isolation, because kin selection and group augmentation can explain variation in feeding rates in concert (see Table 4.2 and Fig. 4.1). I believe that once more studies adopt a similar integrated approach of multiple benefits, this will spur our understanding of how indirect and direct benefits, either alone or in concert, shape cooperative behaviour in animal communities.
The monogamy hypothesis: extra-pair mating affects cooperation
M. coronatus helpers have a strong effect on breeder fitness as they increase breeder survival rates and reproductive success (chapter 5). This result is in strong contrast with the promiscuous fairy-wrens, where helpers do not significantly improve reproductive success or survival of breeders (Table 5.2).
149
If kin selection is the main driving force behind cooperative behaviour, reduced feeding rates associated with reduced relatedness of helpers to offspring (due to EPP) may explain why benefits of cooperative breeding are less strong in the more promiscuous fairy-wrens . Consistent with this idea, comparative analyses among 37 cooperative species (see Table 5.2) reveals that genetic monogamy promotes the occurrence of benefits of cooperative breeding (increased reproductive success and/or survival) in general. This pattern may in turn explain why levels of EPP are negatively associated with the likelihood that avian species breed cooperatively or not (Cornwallis et al. 2010). These results show that the genetic mating system of birds can play an important role in explaining evolution of social mating systems. The monogamy-hypothesis therefore provides an important avenue to pursue in order to understand kin selection and the evolution of cooperative breeding, and more generally the bewildering diversity in behaviour and mating systems. Future studies should include comparative work addressing whether helper contribution increases with lower frequencies of EPP, to test whether this is indeed responsible for the pattern that the benefits of cooperative breeding are smaller in more promiscuous species. It could for instance be argued that levels of EPP are higher in species where nestling and breeder mortality are low and helpers can not really improve reproductive success or breeder survival (and may as a consequence have no helpers at all). Such study, combined with magnitudes of need for parental care across species (e.g., as determined by rates of nestling and breeder mortality) would allow to identify cause and effect of my results, and those by Cornwallis et al. (2010).
Cooperative breeding and offspring sex ratio
The fact that helpers can affect fitness of breeders has resulted in cooperatively breeding species becoming popular models for studies of adaptive offspring sex ratio adjustment: the helper-repayment hypothesis predicts that females should overproduce offspring of the helping sex (Gowaty & Lennartz 1985), especially when helpers exert strong fitness returns (Griffin et al. 2005). In M. coronatus helping behaviour is male-biased and breeders benefit substantially from having helpers, but I did not find that females adjust offspring sex ratio accordingly (see Fig. 6.2). It is, however, premature to assume that this forms evidence that the sex-allocation and
150
helper-repayment theories can be rejected. It could be that actual long-term sex differences in fitness returns are too small in M. coronatus (and other species) to drive the evolution of sex determining mechanisms. Detailed long- term examination and comparative work that combines sex-bias in helping behaviour and its relative fitness returns is needed to determine how and whether sex bias in fitness returns can drive the evolution of sex ratio manipulation across species.
Conclusions
In this thesis, I have addressed the causes and consequences of important components of reproductive behaviour in M. coronatus . I have shown that both EP mating and helping behaviour can be highly beneficial for individuals. It also became clear however, that understanding these behaviours may be complex because both strategies are affected by different costs, benefits and/or constraints that can act in concert, and which relative importance can differ under different circumstances. First, I have shown that EP mating is a way in which M. coronatus females can avoid the negative effects of inbreeding. Although ecological factors may generally constrain M. coronatus females to engage in EP mating, these constraints do not prevent females from EP mating when associated benefits are high (i.e., when they avoid inbreeding by doing so), as females actively adopt strategies to overcome such constraints. Thus, there exists a complex interplay between ecological variables and benefits on the evolution of EP mating, highlighting that investigating individual and species specific ratios between benefits and constraints (and probably also costs) may enable to understand how EP mating evolved. Second, I have shown that kin selection is one important explanation for seemingly altruistic helping behaviour in M. coronatus , as helpers provide more (costly) care to more related offspring. However, also direct benefits of helping, in the form of producing future helpers, are important. I show that when these benefits were not analysed in concert, the importance of single benefits would remain concealed or underestimated. Therefore, statements that direct benefits may be relatively unimportant (e.g., Wright 2007) seem premature, and will likely be refuted when more studies adopt a similar integrated analysis.
151
Overall, there appears a close evolutionary link between cooperative breeding and EP mating: the incidence of EPP reduces the adaptive benefits of cooperative breeding for helpers and, probably as a consequence of associated reduction in helper investment, also for breeders. I am confident that when future studies in evolutionary biology take the link between social and genetic mating systems into account, this will increase our understanding of how natural- and kin-selection can shape associated behaviours.
152
153
154
References
A
Agresti A (1992). A survey of exact inference for contingency tables. Statist. Sci. 7, 131-153. Akçay E, Roughgarden J (2007). Extra-pair paternity in birds: review of the genetic benefits. Evol. Ecol. Res. 9, 855-868. Amos W, Worthington Wilmer J, Fullard K, Burg TM, Croxall JP, Bloch D, Coulson T (2001). The influence of parental relatedness on reproductive success. Proc. R. Soc. Lond. B. 268, 2021-2027. Anderson DJ, Reeve J, Martinez Gomez JE, Weathers WW, Hutson S, Cunningham HV, Bird DM (1993). Sexual size dimorphism and food requirements of nestling birds. Can. J. Zool. 71, 2541-2545. Appleby BM, Petty SJ, Blakey JK, Rainey P, Macdonald DW (1997). Does variation of sex ratio enhance reproductive success of offspring in the tawny owl ( Strix aluco )? Proc. R. Soc. Lond. B. 264, 1111-1116. Arnold KE, Owens IPF (2002). Extra-pair paternity and egg dumping in birds: life history, parental care and the risk of retaliation. Proc. R. Soc. Lond. B. 269, 1263-1269. Arnqvist G, Kirkpatrick M (2005). The evolution of infidelity in socially monogamous passerines: the strength of direct and indirect selection on extrapair copulation behavior in females. Am. Nat. 165, S26-S37. Austad SN, Rabenold KN (1985). Reproductive enhancement by helpers and an experimental inquiry into its mechanism in the bicolored wren. Behav. Ecol. Sociobiol. 17, 19-27. Austad SN, Rabenold KN (1986). Demography and the evolution of cooperative breeding in the bicolored wren, Campylorhynchus griseus . Behaviour 97, 308-324.
B
Baglione V, Marcos JM, Canestrari D, Ekman J (2002). Direct fitness benefits of group living in a complex cooperative society of carrion crows, Corvus corone corone . Anim. Behav. 64, 887-893.
155
Baglione V, Canestrari D, Marcos JM, Ekman J (2003). Kin selection in cooperative alliances of carrion crows. Science 300, 1947-1949. Bateman AJ (1948). Intra-sexual selection in Drosophila. Heredity 2, 349- 368. Bates D, Sarkar D (2006). Lme4: linear mixed-effects models using S4 classes. R package version 0.995-2. http://www.r-project.org. Bates D, Sarkar D (2007). Lme4: linear mixed-effects models using S4 classes. R package version 0.9975-12, http://CRAN.R-project.org. Beck NR, Heinsohn R (2006). Group composition and reproductive success of cooperatively breeding white-winged choughs ( Corcorax melanorhamphos ) in urban and non-urban habitat. Austr. Ecol. 31, 588-596. Bennett PM, Owens IPF (2002). Evolutionary ecology of birds: life histories, mating systems and extinction. Oxford: Oxford University Press. Berg EC (2005). Parentage and reproductive success in the white-throated magpie-jay, Calocitta formosa , a cooperative breeder with female helpers. Anim. Behav. 70, 375-385. Berg EC, Eadie JM, Langen TA, Russell AF (2009). Reverse sex-biased philopatry in a cooperative bird: genetic consequences and a social cause. Molec. Ecol. 18, 3486-3499. Bergmüller R, Bshary R, Johnstone RA, Russell AF (2007). Integrating cooperative breeding and cooperation theory . Behav. Process. 76, 61-72. Birkhead TR (1979). Mate guarding in the magpie Pica pica . Anim. Behav. 27, 866-874. Birkhead TR, Biggins JD (1987). Reproductive synchrony and extra-pair copulation in birds. Ethology 74, 320-334. Birkhead TR, Biggins JD (1998). Sperm competition mechanisms in birds: models and data. Behav. Ecol. 9, 253-260. Birkhead TR, Møller AP (1992). Sperm competition in birds: evolutionary causes and consequences. London: Academic Press. Birkhead TR, Møller AP (1998). Sperm competition and sexual selection. London: Academic Press. Bishop JM, O’Ryan C, Jarvis JUM (2007). Social common mole-rats enhance outbreeding via extra-pair mating. Biol. Lett. 3, 176-179. Bjørnstad G, Lifjeld JT (1997). High frequency of extra-pair paternity in a dense and synchronous population of willow warblers Phylloscopus trochilus . J. Avian Biol. 28, 319-324.
156
Blackmore CJ, Heinsohn R (2007). Reproductive success and helper effects in the cooperatively breeding grey-crowned babbler. J. Zool. 273, 326-332. Blackmore CJ, Heinsohn R (2008). Variable mating strategies and incest avoidance in cooperatively breeding grey-crowned babblers. Anim. Behav. 75, 63-70. Blackmore CJ, Adcock GJ, Ebert D, Heinsohn R (2006). Microsatellite loci for population and behavioural studies of grey-crowned babblers (Pomatostomus temporalis : Aves). Molec. Ecol. Notes 6, 412-414. Blomqvist D, Andersson M, Küpper C, Cuthill IC, Kis J, Lanctot RB, Sandercock BK, Székely T, Wallander J, Kempenaers B (2002). Genetic similarity between mates and extra-pair parentage in three species of shorebirds. Nature 419, 613-615. Boland CRJ, Heinsohn R, Cockburn A (1997). Experimental manipulation of brood reduction and parental care in cooperatively breeding white- winged choughs. J. Anim. Ecol. 66, 683-691. Bolker B, Brooks ME, Clark CJ, Geange SW, Poulson JR, Stevens MHH, White J-SS (2009). Generalized linear mixed models: a practical guide for ecology and evolution. Trends Ecol. Evol. 24, 127-135. Boomsma JJ (2007). Kin selection versus sexual selection: why the ends do not meet. Curr. Biol. 17, R673-683. Boomsma JJ (2009). Lifetime monogamy and the evolution of eusociality. Philos. Trans R. Soc Lond. B. 364, 3191-3207 . Bradbury RB, Cutton PA, Wright J, Griffiths R (1997). Nestling sex ratio in the European starling Sturnus vulgaris . J. Avian Biol. 28, 255-258. Briskie JV (1993). Anatomical adaptations to sperm competition in Smith's longspurs and other polygynandrous passerines. Auk 110, 875-888. Brooke M, Hartley IR (1995). Nesting Henderson reed-warblers (Acrocephalus vaughani taiti ) studied by DNA fingerprinting: unrelated coalitions in a stable habitat? Auk 112, 77-86. Brooker MG, Rowley I, Adams M, Baverstock PR (1990). Promiscuity: an inbreeding avoidance mechanism in a socially monogamous species? Behav. Ecol. Sociobiol. 26, 191-199. Brouwer L, Richardson DS, Eikenaar C, Komdeur J (2006). The role of group size and environmental factors on survival in a cooperatively breeding tropical passerine. J. Anim. Ecol. 75, 1321-1329. Brown JL (1975). Helpers among Arabian babblers Turdoides squamiceps . Ibis 117, 243-244.
157
Brown JL (1987). Helping and communal breeding in birds: ecology and evolution. Princeton: Princeton University Press. Brown JL, Brown ER (1981). Kin selection and individual selection in babblers. In: Natural selection and social behavior. Alexander R, Tinkle D (eds). New York: Chiron Press, pp 244-256. Brown JL, Brown ER, Brown SD, Dow DD (1982). Helpers - effects of experimental removal on reproductive success. Science 215, 421-422. Bruce JP, Quinn JS, Sloane SA, White BN (1996). DNA fingerprinting reveals monogamy in the bushtit, a cooperatively breeding species. Auk 113, 511-516. Bryant DM (1988). Energy expenditure and body mass changes as measures of reproductive costs in birds. Funct. Ecol. 2, 23-34. Burke T, Davies NB, Bruford MW, Hatchwell BJ (1989). Parental care and mating-behaviour of polyandrous dunnocks Prunella modularis related to paternity by DNA fingerprinting. Nature 338, 249-251. Burnham KP, Anderson DR (2002). Model selection and multimodel inference: a practical information-theoretic approach. New York: Springer-Verlag. Byholm P, Brommer JE, Saurola P (2002). Scale and seasonal sex-ratio trends in northern goshawk Accipiter gentilis broods. J. Avian Biol. 33, 399-406.
C
Caffrey C (1999). Feeding rates and individual contributions to feeding at nests in cooperatively breeding western American crows. Auk 116, 836- 841. Caffrey C (2000). Correlates of reproductive success in cooperatively breeding western American crows: if helpers help, it's not by much. Condor 102, 333-341. Canestrari D, Marcos JM, Baglione V (2005). Effects of parentage and relatedness on the individual contributions to cooperative chick care in the carrion crow Corvus corone corone . Behav. Ecol. Sociobiol. 57, 422- 428. Canestrari D, Marcos JM, Baglione V (2008). Reproductive success increases with group size in cooperative carrion crows, Corvus corone corone . Anim. Behav. 75, 403-416.
158
Cézilly F, Nager RG (1995). Comparative evidence for a positive association between divorce and extra-pair paternity in birds. Proc. R. Soc. Lond. B. 262, 7-12. Chan K, Augusteyn JD (2003). Relationship between bird-unit size and territory quality in three species of fairy-wrens (Malurus spp.) with overlapping territories. Ecol. Res. 18, 73-80. Charlesworth D, Charlesworth B (1987). Inbreeding depression and its evolutionary consequences . Ann. Rev. Ecol. Syst. 18, 237-268. Charnov EL (1981). Kin selection and helpers at the nest: effects of paternity and biparental care. Anim. Behav. 2, 631-632. Charnov EL (1982). The theory of sex allocation. Princeton: Princeton University Press. Christidis L, Schodde R (1997). Relationships within the Australo-Papuan fairy-wrens (Aves, Malurinae): an evaluation of the utility of allozyme data. Austr. J. Zool. 45, 113-129. Clark AB (1978). Sex ratio and local resource competition in a prosimian primate. Science 208, 1157-1159. Clarke MF, Jones DA, Ewen JG, Robertson RJ, Griffiths R, Painter J, Boag PT, Crozier R (2002). Male-biased sex ratios in broods of the cooperatively breeding bell miner, Manorina melanophrys . J. Avian Biol. 33, 71-76. Clout MN, Elliott GP, Robertson BC (2002). Effects of supplementary feeding on the offspring sex ratio of kakapo: a dilemma for the conservation of a polygynous parrot. Biol. Conserv. 107, 13-18. Clutton-Brock TH (1986). Sex ratio variation in birds. Ibis 128, 317-329. Clutton-Brock TH (1991). The evolution of parental care. Princeton: Princeton University Press. Clutton-Brock T (2002). Breeding together: kin selection and mutualism in cooperative vertebrates. Science 296, 69-72. Clutton-Brock TH (2009). Cooperation between non-kin: reciprocity, mutualism or manipulation? Nature 462, 51-57. Cockburn A (1998). Evolution of helping behavior in cooperatively breeding birds. Ann. Rev. Ecol. Syst. 29, 141-177. Cockburn A (2004). Mating systems and sexual conflict. In: Ecology and evolution of cooperative breeding in birds. Koenig WD, Dickinson JL (eds). Cambridge: Cambridge University Press, pp 81-101.
159
Cockburn A (2006). Prevalence of different modes of parental care in birds. Proc. R. Soc. Lond. B. 273, 1375-1383. Cockburn A, Double MC (2008). Cooperatively breeding superb fairy-wrens show no facultative manipulation of offspring sex ratio despite plausible benefits. Behav. Ecol. Sociobiol. 62, 681-688. Cockburn A, Legge S, Double MC (2002). Sex ratios in birds and mammals: can the hypotheses be disentangled? In: Hardy ICW (ed). Sex ratios. Concepts and research methods. Cambridge: Cambridge University Press, pp 266-286. Cockburn A, Osmond HL, Mulder RA, Green DJ, Double MC (2003). Divorce, dispersal and incest avoidance in the cooperatively breeding superb fairy-wren Malurus cyaneus . J. Anim Ecol. 72, 189-202. Cockburn A, Sims RA, Osmond HL, Green DJ, Double MC, Mulder RA (2008). Can we measure the benefits of help in cooperatively breeding birds: the case of superb fairy-wrens Malurus cyaneus ? J. Anim. Ecol. 77, 430-438. Cockburn A, Dalziell AH, Blackmore CJ, Double MC, Kokko H, Osmond HL, Beck NR, Head ML, Wells K (2009). Super fairy-wren males aggregate into hidden leks to solicit extragroup fertilizations before dawn. Behav. Ecol. 20, 501-510. Cohas A, Yoccoz NG, Bonenfant C, Goossens B, Genton C, Galan M, Kempenaers B, Allainé D (2008). The genetic similarity between pair members influences the frequency of extrapair paternity in alpine marmots. Anim. Behav. 76: 87-95. Conrad KF, Clarke MF, Robertson RJ, Boag PT (1998). Paternity and the relatedness of helpers in the cooperatively breeding bell miner. Condor 100, 343-349. Cordero PJ, Griffith SC, Aparico JM, Parkin DT (2000). Sexual dimorphism in house sparrow eggs. Behav. Ecol. Sociobiol. 48, 353-357. Cornwallis CK , West SA , Davis KE , Griffin AS (2010). Promiscuity and the evolutionary transition to complex societies. Nature 466, 969-972. Covas R, Doutrelant C, du Plessis M (2004). Experimental evidence of a link between breeding conditions and the decision to breed or to help in a colonial cooperative bird. Proc. R. Soc. Lond. B. 271, 827-832. Covas R, Dalecky A, Caizergues A, Doutrelant C (2006). Kin associations and direct vs. indirect fitness benefits in colonial cooperatively breeding sociable weavers Philetairus socius . Behav. Ecol. Sociobiol. 60, 323-331.
160
Covas R, du Plessis, MA, Doutrelant C (2008). Helpers in colonial cooperatively breeding sociable weavers Philetairus socius contribute to buffer the effects of adverse breeding conditions. Behav. Ecol. Sociobiol. 63, 103-112. Crozier RH, Pamilo P (1996). Evolution of social insect colonies: sex allocation and kin-selection. Oxford: Oxford University Press. Crick HQP (1992). Load-lightening in cooperatively breeding birds and the cost of reproduction. Ibis 134, 56-61. Currie DR, Burke TA , Whitney R, Thompson DBA (1998) . Male and female behaviour and extra-pair paternity in the wheatear . Anim. Behav. 55, 689-703 .
D
Darwin C (1859). On the origin of species by means of natural selection, or the preservation of favoured races in the struggle for life . London: John Murray. Davies NB (1990). Dunnocks: cooperation and conflict among males and females in a variable mating system. In: Cooperative breeding in birds: long-term studies of ecology and behavior. Stacey PB, Koenig WD (eds). Cambridge: Cambridge University Press, pp 455-485. Davies NB (1992). Dunnock behaviour and social evolution. Oxford: Oxford University Press. Davies NB, Houston AI (1986). Reproductive success of dunnocks, Prunella modularis , in a variable mating system. II. Conflicts-of-interest among breeding adults. J. Anim. Ecol. 55, 139-154. Davies NB, Hatchwell BJ (1992). The value of male parental care and its influence on reproductive allocation by male and female dunnocks. J. Anim. Ecol. 61, 259-272. Davies NB, Hartley IR, Hatchwell BJ, Desrochers A, Skeer J, Nebel D (1995). The polygynandrous mating system of the Alpine accentor, Prunella collaris . I. Ecological causes and reproductive conflicts. Anim. Behav. 49, 769-788. Denk AG, Holzmann A, Peters A, Vermeirssen ELM, Kempenaers B (2005). Paternity in mallards: e ffects of sperm quality and female sperm selection for inbreeding avoidance. Behav. Ecol. 16, 825-833.
161
Dickinson JL (2004). A test of the importance of direct and indirect fitness benefits for helping decisions in western bluebirds. Behav. Ecol. 15, 233-238. Dickinson JL, Akre JJ (1998). Extrapair paternity, inclusive fitness, and within-group benefits of helping in western bluebirds. Molec. Ecol. 7, 95-105. Dickinson JL, Hatchwell BJ (2004). Fitness consequences of helping. In: Ecology and evolution of cooperative breeding in birds. Koenig WD, Dickinson JL (eds). Cambridge: Cambridge University Press, pp 48-66. Dickinson JL, Koenig WD, Pitelka FA (1996). Fitness consequences of helping behavior in the western bluebird. Behav. Ecol. 7, 168-177. Dijkstra C, Daan S, Meijer T, Cavé AJ, Foppen R (1988) Body mass of the kestrel in relation to food abundance and reproduction. Ardea 76, 127- 140. Dijkstra C, Daan S, Buker JB (1990). Adaptive seasonal variation in the sex ratio of kestrel broods. Funct. Ecol. 4, 143-147. Diniz-Filho JAF (2001). Phylogenetic autocorrelation under distinct evolutionary processes. Evolution 55, 1104-1109. Dixon A, Ross D, O´Malley SLC, Burke T (1994). Paternal investment inversely related to degree of extra-pair paternity in the reed bunting. Nature 371, 698-700. Double M, Cockburn A (2000). Pre-dawn infidelity: females control extra-pair mating in superb fairy-wrens. Proc. R. Soc. Lond. B. 267, 465-470. Double MC, Dawson D, Burke T, Cockburn A (1997). Finding the fathers in the least faithful bird: a microsatellite-based genotyping system for the superb fairy-wren Malurus cyaneus . Molec. Ecol. 6, 691-693. Doutrelant C, Covas R, Caizergues A, du Plessis M (2004). Unexpected sex ratio adjustment in a colonial cooperative bird: pairs with helpers produce more of the helping sex whereas pairs without helpers do not. Behav. Ecol. Sociobiol. 56, 149-154. Dow DD, Whitmore MJ (1990). Noisy miners: variation on the theme of communality. In: Cooperative breeding in birds: long-term studies of ecology and behavior. Stacey PB, Koenig WD (eds). Cambridge: Cambridge University Press, pp 559-592. Du B, Lu X (2009). Bi-parental vs. cooperative breeding in a passerine: fitness-maximizing strategies of males in response to risk of extra-pair paternity? Molec. Ecol. 18, 3929-3939.
162
Dugatkin LA (2002). Animal cooperation among unrelated individuals. Naturwissenschaften 89, 533-541. Dunn PO, Cockburn A (1996). Evolution of male parental care in a bird with almost complete cuckoldry. Evolution 50, 2542-2548. Dunn PO, Cockburn A (1999). Extrapair mate choice and honest signaling in cooperatively breeding superb fairy-wrens. Evolution 53, 938-946. Dunn PO, Cockburn A, Mulder RA (1995). Fairy-wren helpers often care for young to which they are unrelated. Proc. R. Soc. Lond. B. 259, 339-343. Dunn PO, Whittingham LA, Pitcher TE (2001). Mating systems, sperm competition, and the evolution of sexual dimorphism in birds. Evolution 55, 161-175. Durrant KL (2004). The genetic and social mating system of a white-backed population of the Australian magpie ( Gymnorhina tibicen tyrrannica ). Griffith University: PhD thesis. Durrant KL, Hughes JM (2005). Differing rates of extra-group paternity between two populations of the Australian magpie ( Gymnorhina tibicen ). Behav. Ecol. Sociobiol. 57, 536-545.
E
Eguchi K, Yamaguchi N, Ueda K, Nagata H, Takagi M, Noske R (2007). Social structure and helping behaviour of the grey-crowned babbler Pomatostomus temporalis . J. Ornithol. 148, S203-S210. Eimes JA, Parker PG, Brown JL, Brown ER (2005). Extrapair fertilization and genetic similarity of social mates in the Mexican jay. Behav. Ecol. 16, 456-460. Emlen ST (1981). Altruism, kinship, and reciprocity in the white-fronted bee- eater. In: Natural selection and social behavior. Alexander R, Tinkle D (eds). New York: Chiron Press, pp 217-230. Emlen ST (1990). White-fronted bee-eaters: helping in colonially nesting species. In: Cooperative breeding in birds: long-term studies of ecology and behavior. Stacey PB, Koenig WD (eds). Cambridge: Cambridge University Press, pp 487-526. Emlen ST (1995). An evolutionary theory of the family. Proc. Natl. Acad. Sci. USA 92, 8092-8099.
163
Emlen ST (1997a). Predicting family dynamics in social vertebrates. In: Behavioural ecology: an evolutionary approach. Krebs JR, Davies NB (eds). Oxford: Blackwell Scientific, pp 228-253. Emlen ST (1997b). When mothers prefer daughters over sons. Trends Ecol. Evol. 12, 291-292. Emlen ST, Vehrenkamp SL (1983). Cooperative breeding strategies among birds. In: Perspectives in ornithology. Brush AH, Clark GA Jr (eds). Cambridge: Cambridge University Press. Emlen ST, Wrege PH (1988). The role of kinship in helping decisions among white-fronted bee-eaters. Behav. Ecol. Sociobiol. 23, 305-315. Emlen ST, Wrege PH (1989). A test of alternate hypotheses for helping- behavior in white-fronted bee-eaters of Kenya. Behav. Ecol. Sociobiol. 25, 303-319. Emlen ST, Wrege PH (1991). Breeding biology of white-fronted bee-eaters at Nakuru: the influence of helpers on breeder fitness. J. Anim. Ecol. 60, 309-326. Emlen ST, Emlen JM, Levin S (1986). Sex-ratio selection in species with helpers-at-the-nest. Am. Nat. 127, 1-8. Endler JA (1986). Natural selection in the wild . Princeton: Princeton University Press. Ewen JG, Cassey P, Møller AP (2004). Facultative primary sex ratio variation: a lack of evidence in birds? Proc. R. Soc. Lond. B. 271, 1277- 1282.
F
Faaborg J (1986). Reproductive success and survivorship of the Galapagos hawk Buteo galapagoensis - Potential costs and benefits of cooperative polyandry. Ibis 128, 337-347. Faaborg J, Bednarz JC (1990). Galapagos and Harris' hawks: divergent causes of sociality in two raptors. In: Cooperative breeding in birds: long-term studies of ecology and behavior. Stacey PB, Koenig WD (eds). Cambridge: Cambridge University Press, pp 359-383. Faaborg J, de Vries T, Patterson CB, Griffin CR (1980). Preliminary observations on the occurrence and evolution of polyandry in the Galapagos hawk ( Buteo galapagoensis ). Auk 97, 581-590.
164
Faaborg J, Parker PG, DeLay L, de Vries T, Bednarz JC, Maria Paz S, Naranjo J, Waite TA (1995). Confirmation of cooperative polyandry in the Galapagos hawk ( Buteo galapagoensis ). Behav. Ecol. Sociobiol. 36, 83-90. Fiala KL, Congdon JD (1983). Energetic consequences of sexual size dimorphism in nestling red-winged blackbirds. Ecology 64, 642-647. Finn PG, Hughes JM (2001). Helping behaviour in Australian magpies, Gymnorhina tibicen . Emu 101, 57-63. Fisher RA (1930). The genetic theory of natural selection, 2nd edn. Oxford: Clarendon. Fishman MA, Stone L (2005). Fertility assurance through extrapair fertilization, and male parental effort. Bull. Math. Biol. 648, 809-823. Fishman MA, Stone L (2006). Trade up polygyny and breeding synchrony in avian populations. J. Theor. Biol. 238, 104-110. Foerster K, Delhey K, Johnsen A, Lifjeld JT, Kempenaers B (2003). Females increase offspring heterozygosity and fitness through extra-pair matings. Nature 425, 714-717. Frank SA (1990). Sex allocation theory for birds and mammals. Annu. Rev. Ecol. Syst. 21, 13-55. Franzreb KE (2007). Reproductive success and nest depredation of the Florida scrub-jay. Wilson J. Ornithol. 119, 162-169. Freeman-Gallant CR, Wheelwright NT, Meiklejohn KE, Sollecito SV (2006). Genetic similarity, extrapair paternity, and offspring quality in Savannah sparrows ( Passerculus sandwichensis ). Behav. Ecol. 17, 952- 958.
G
Gaston AJ (1973). Ecology and behaviour of the long-tailed tit. Ibis 115, 330- 351. Gaston AJ (1978). Evolution of group territorial behavior and cooperative breeding. Am. Nat. 112, 1091-1100. Ghalambor CK, Martin TE (2001). Fecundity-survival trade-off and parental risk-taking in birds. Science 292, 494-497. Gibbs HL, Tabak LM, Hobson K (1999). Characterization of microsatellite DNA loci for a neotropical migrant songbird the Swainson’s thrush (Catharus ustulatus ). Molec. Ecol. 8, 1551-1552.
165
Gonçalves da Silva A, Russello MA (2010). iRel : an online tool for implementing pairwise relatedness estimators and evaluating their performance. Conserv. Genet. Res. 3, 69-71. Gowaty PA (1996). Battles of the sexes and origins of monogamy. In: Partnerships in birds: the study of monogamy. Black JM (ed). Oxford: Oxford University Press, pp 21-52. Gowaty PA, Bridges WC (1991). Nestbox availability affects extra-pair fertilizations and conspecific nest parasitism in eastern bluebirds, Sialia sialis . Anim. Behav. 41, 661-675. Gowaty PA, Lennartz MR (1985). Sex ratios of nestling and fledgling red- cockaded woodpeckers ( Picoides borealis ) favor males. Am. Nat. 16, 347-353. Gray EM (1997). Female red-winged blackbirds accrue material benefits from copulating with extrapair males? Anim. Behav. 53, 625-639. Green DJ, Cockburn A, Hall ML, Osmond HL, Dunn PO (1995). Increased opportunities for cuckoldry may be why dominant male fairy-wrens tolerate helpers. Proc. R. Soc. Lond. B. 262, 297-303. Green DJ, Osmond HL, Double MC, Cockburn A (2000). Display rate by male fairy-wrens ( Malurus cyaneus ) during the fertile period of females has little influence on extra-pair mate choice. Behav. Ecol. Sociobiol. 48, 438-446. Griffin AS, West SA (2002). Kin selection: fact and fiction. Trends Ecol. Evol. 17, 15-21. Griffin AS, West SA (2003). Kin discrimination and the benefit of helping in cooperatively breeding vertebrates. Science 302, 634-636. Griffin AS, Sheldon BC, West SA (2005). Cooperative breeders adjust offspring sex ratios to produce helpful helpers. Am. Nat. 166, 628-632. Griffith SC (2007). The evolution of infidelity in socially monogamous passerines: neglected components of direct and indirect selection. Am. Nat. 169, 274-281. Griffith SC, Immler S (2009). Female infidelity and genetic compatibility in birds: the role of the genetically loaded raffle in understanding the function of extrapair paternity. J. Avian Biol. 97-101. Griffith SC, Owens IPF, Thuman KA (2002). Extra pair paternity in birds: a review of interspecific variation and adaptive function. Molec. Ecol. 11, 2195-2212.
166
Gyllensten UB, Jakobsson S, Temrin H (1990). No evidence for illegitimate young in monogamous and polygynous warblers. Nature 343, 168-170.
H
Hackett SJ, Kimball RT, Reddy S, Bowie RC, Braun EL, Braun MJ, et al. (2008). A phylogenomic study of birds reveals their evolutionary history. Science 320, 1763-1768. Haig SM, Belthoff JR, Allen DH (1993). Examination of population-structure in red-cockaded woodpeckers using DNA profiles. Evolution 47, 185- 194. Haig SM, Walters JR, Plissner JH (1994). Genetic evidence for monogamy in the cooperatively breeding red-cockaded woodpecker. Behav. Ecol. Sociobiol. 34, 295-303. Hall ML (2004). A review of hypotheses for the functions of avian duetting. Behav. Ecol. Sociobiol. 55, 315-430. Hall ML, Peters A (2008). Coordination between the sexes for territorial defence in a duetting fairy-wren. Anim. Behav. 76, 65-73. Hall ML, Peters A (2009). Do male paternity guards ensure female fidelity in a duetting fairy-wren? Behav. Ecol. 20, 222-228. Hamilton WD (1964). The genetical evolution of social behavior. J. Theor. Biol. 7, 1-52. Hamilton WD (1967). Extraordinary sex ratios. Science 156, 477-488. Hartley IR, Davies NB (1994). Limits to cooperative polyandry in birds. Proc. R. Soc. Lond. B. 257, 67-73. Hartley IR, Shepherd M, Robson T, Burke T (1993). Reproductive success of polygynous male corn buntings ( Miliaria calandra ) as confirmed by DNA fingerprinting. Behav. Ecol. 4, 310-317. Hartley IR, Davies NB, Hatchwell BJ, Desrochers A, Nebel D, Burke T (1995). The polygynandrous mating system of the alpine accentor, Prunella collaris . II. Multiple paternity and parental effort. Anim. Behav. 49, 789-803. Harvey PH, Pagel MD (1991). The comparative method in evolutionary biology. Oxford: Oxford University Press. Hasselquist D, Sherman PW (2001). Social mating systems and extrapair fertilizations in passerine birds. Behav. Ecol. 12, 457-466.
167
Hasselquist D, Bensch S, von Schantz T (1996). Correlation between song repertoire, extra-pair paternity and offspring survival in the great reed warbler. Nature 381 ,229-232. Hatchwell BJ (1999). Investment strategies of breeders in avian cooperative breeding systems. Am. Nat. 154, 205-219. Hatchwell BJ (2009). The evolution of cooperative breeding in birds: kinship, dispersal and life history. Philos. Trans R. Soc Lond. B. 364, 3217- 3227. Hatchwell BJ, Ross DJ, Chaline N, Fowlie MK, Burke T (2002). Parentage in the cooperative breeding system of long-tailed tits, Aegithalos caudatus . Anim. Behav. 64, 55-63. Hatchwell BJ, Russell AF, MacColl ADC, Ross DJ, Fowlie MK, McGowan A (2004). Helpers increase long-term but not short-term productivity in cooperatively breeding long-tailed tits. Behav. Ecol. 15, 1-10. Hauber ME, Lacey EA (2005). Bateman’s principle in cooperatively breeding vertebrates: the effect of non-breeding alloparents on variability in female and male reproductive success. Integr. Comp. Biol. 45, 903-914. Haydock J, Koenig WD, Stanback MT (2001). Shared parentage and incest avoidance in the cooperatively breeding acorn woodpecker. Molec. Ecol. 10, 1515-1525. Haydock J, Parker PG, Rabenold KN (1996). Extra-pair paternity uncommon in the cooperatively breeding bicolored wren. Behav. Ecol. Sociobiol. 38, 1-16. Heinsohn RG (1992). Cooperative enhancement of reproductive success in white-winged choughs. Evol. Ecol. 6, 97-114. Heinsohn R (1995). Hatching asynchrony and brood reduction in cooperatively breeding white-winged choughs Corcorax melanorhamphos . Emu 95, 252-258. Heinsohn R (2004). Parental care, load-lightening, and costs. In: Ecology and evolution of cooperative breeding in birds. Koenig WD, Dickinson JL (eds). Cambridge: Cambridge University Press, pp 67-80. Heinsohn R, Legge S (1999). The cost of helping. Trends Ecol. Evol. 14, 53-57. Heinsohn R, Dunn P, Legge S, Double M (2000). Coalitions of relatives and reproductive skew in cooperatively breeding white-winged choughs. Proc. R. Soc. Lond. B. 267, 243-249.
168
Heppell SS, Walters JR, Crowder LB (1994). Evaluating management alternatives for red-cockaded woodpeckers - a modeling approach. J. Wildl. Manage. 58, 479-487. Higgins PJ, Peter JM, Steele WK (2001). Handbook of Australian, New Zealand, and Antarctic birds, vol. 5. Melbourne: Oxford University Press. Hoi-Leitner M, Hoi H, Romero-Pujante R, Valera F (1999). Female extra-pair behaviour and environmental quality in the serin ( Serinus serinus ): a test of the 'constrained female hypothesis'. Proc. R. Soc. Lond. B. 266, 1021-1026. Houston AI, Davies NB (1985). Evolution of cooperation and life history in dunnocks, Prunella modularis . In: Behavioural ecology. Sibly R, Smith RH (eds). Oxford: Blackwell Scientific, Publication British Ecological Society Symposium, pp 471-487. Howe HF (1977). Sex-ratio adjustment in the common grackle. Science 198, 744-746. Hughes JM, Mather PB, Toon A, Ma J, Rowley I, Russell E (2003). High levels of extra-group paternity in a population of Australian magpies Gymnorhina tibicen : evidence from microsatellite analysis. Molec. Ecol. 12, 3441-3450. Hughes WOH, Oldroyd BP, Beekman M, Ratnieks FLW (2008). Ancestral monogamy shows kin selection is key to the evolution of eusociality. Science 320, 1213-1216.
I
Innes KE, Johnston RE (1996). Cooperative breeding in the white-throated magpie-jay. How do auxiliaries influence nesting success? Anim. Behav. 51, 519-533.
J
Jennions MD, Petrie M (2000). Why do females mate multiply? A review of the genetic benefits. Biol. Rev. 75, 21-64. Jones CS, Lessells CM, Krebs JR (1991). Helpers-at-the-nest in the European bee-eaters ( Merops apiaster ): a genetic analysis. In: DNA
169
fingerprinting: approaches and applications. Burke T, Dolf G, Jeffreys AJ, Wolff R (eds). Basel: Birkhäuser Verlag, pp 169-192. Jønsson KA, Fjeldså J (2006). A phylogenetic supertree of oscine passerine birds (Aves: Passeri). Zool. Scr. 35, 149-186. Julliard R (2000). Sex-specific dispersal in spatially varying environments leads to habitat-dependent evolutionary stable sex ratio. Behav. Ecol. 4, 421-428.
K
Kahn MZ, Walters JR (2002). Effects of helpers on breeder survival in the red-cockaded woodpecker ( Picoides borealis ). Behav. Ecol. Sociobiol. 51, 336-344. Kahn NW, St John J, Quinn TW (1998). Chromosome-specific intron size differences in the avian CHD gene provide an efficient method for sex identification in birds. Auk 115, 1074-1078. Kalinowski ST, Taper ML, Marshall TC (2007). Revising how the computer program CERVUS accommodates genotyping error increases success in paternity assignment. Molec. Ecol. 16, 1099-1106. Karubian J (2002). Costs and benefits of variable breeding plumage in the red-backed fairy-wren. Evolution 56, 1673-1682. Karubian J (2008). Changes in breeding status are associated with rapid bill darkening in male red-backed fairy-wrens Malurus melanocephalus . J. Avian Biol. 39, 81-86. Keller LF, Waller DM (2002). Inbreeding effects in wild populations. Trends Ecol. Evol. 17, 230-241. Kempenaers B (1993). The use of a breeding synchrony index. Ornis Scand. 24, 84. Kempenaers B (2007). Mate choice and genetic quality: a review of the heterozygosity theory. Adv. Stud. Behav. 37, 189-278. Kempenaers B, Verheyen GR, Dhondt AA (1997). Extrapair paternity in the blue tit ( Parus caeruleus ): female choice, male characteristics, and offspring performance. Behav. Ecol. 8, 481-492. Kempenaers B, Congdon B, Boag P, Robertson RJ (1999). Extrapair paternity and egg hatchability in tree swallows: evidence for the genetic compatibility hypothesis? Behav. Ecol. 10, 304-311.
170
Khan MZ, Walters JR (2002). Effects of helpers on breeder survival in the red-cockaded woodpecker ( Picoides borealis ). Behav. Ecol. Sociobiol. 51, 336-344. Kilner R (1998). Primary and secondary sex ratio manipulation by zebra finches. Anim. Behav. 56, 155-164. Kingma SA, Hall ML, Segelbacher G, Peters A (2009). Radical loss of an extreme extra-pair mating system. BMC Ecology 9, 15. Kingma SA, Hall ML, Arriero E, Peters A (2010). Multiple benefits of cooperative breeding in purple-crowned fairy-wrens: a consequence of fidelity? J. Anim. Ecol. 79, 757-768. Kingma SA, Hall ML, Peters A (2011a). No evidence for offspring sex-ratio adjustment to social and environmental conditions in cooperatively breeding purple-crowned fairy-wrens. Behav. Ecol. Sociobiol. DOI 10.1007/s00265-010-1133-7 Kingma SA, Hall ML, Peters A (2011b). Multiple benefits drive helping behavior in a cooperatively breeding bird: an integrated analysis. Am. Nat. In press. Koenig WD (1981). Reproductive success, group-size, and the evolution of cooperative breeding in the acorn woodpecker. Am. Nat. 117, 421-443. Koenig WD, Dickinson JL (2004). Evolution and ecology of cooperative breeding in birds. Cambridge: Cambridge University Press. Koenig WD, Mumme RL (1987). Population ecology of the cooperatively breeding acorn woodpecker. Princeton: Princeton University Press. Koenig WD, Stacey PB (1990). Acorn Woodpeckers: group-living and food- storage under contrasting ecological constraints. In: Cooperative breeding in birds: long-term studies of ecology and behavior. Stacey PB, Koenig WD (eds). Cambridge: Cambridge University Press, pp 413- 453. Koenig WD, Walters JR (1999). Sex-ratio selection in species with helpers at the nest: the repayment model revisited. Am. Nat. 153, 124-130. Kokko H, Johnstone RA, Clutton-Brock TH (2001). The evolution of cooperative breeding through group augmentation. Proc. R. Soc. Lond. B. 268, 187-196. Kokko H, Johnstone RA, Wright J (2002). The evolution of parental and alloparental effort in cooperatively breeding groups: when should helpers pay to stay? Behav. Ecol. 13, 291-300.
171
Komdeur J (1994). The effect of kinship on helping in the cooperative breeding Seychelles warbler Acrocephalus sechellensis . Proc. R. Soc. Lond. B. 256, 37-52. Komdeur J (1996). Influence of helping and breeding experience on reproductive performance in the Seychelles warbler: a translocation experiment. Behav. Ecol. 7, 326-333. Komdeur J (2004). Sex-ratio manipulation. Ecology and behaviour of cooperative breeding in birds. In: Koenig WD, Dickinson JL (eds). Cambridge: Cambridge University Press, pp 102-116. Komdeur J, Hatchwell B (1999). Kin recognition: function and mechanism in avian societies. Trends Ecol. Evol. 14, 237-241. Komdeur J, Pen I (2002). Adaptive sex allocation in birds. Philos. Trans. R Soc. Lond. B. 357, 373-386. Komdeur J, Daan S, Tinbergen J, Mateman C (1997). Extreme adaptive modification in sex ratio of the Seychelles warbler’s eggs. Nature 385, 522-525. Korpimäki E, May CA, Parkin DT, Wetton JH, Wiehn J (2000). Environmental- and parental condition-related variation in sex ratio of kestrel broods. J. Avian Biol. 31, 128-134. Krackow S (1999). Avian sex ratio distortions: the myth of maternal control. In: Adams NJ, Slotow RH (eds). Proc. 22 Int. Ornithol. Congr. Durban. Johannesburg. BirdLife South Africa, pp 425-433.
L
Lack D (1968). Ecological adaptations for breeding in birds. London: Chapman and Hall. Langen TA, Vehrencamp SL (1999). How white-throated magpie-jay helpers contribute during breeding. Auk 116, 131-140. Lawton MF, Guindon CF (1981). Flock composition, breeding success, and learning in the brown jay. Condor 83, 27-33. Legge S (2000a). Helper contributions in the cooperatively breeding laughing kookaburra: feeding young is no laughing matter. Anim. Behav. 59, 1009-1018. Legge S (2000b). The effect of helpers on reproductive success in the laughing kookaburra. J. Anim. Ecol. 69, 714-724.
172
Legge S, Cockburn A (2000). Social and mating system of cooperatively breeding laughing kookaburras ( Dacelo novaeguineae ). Behav. Ecol. Sociobiol. 47, 220-229. Legge S, Heinsohn R, Double MC, Griffiths R, Cockburn A (2001). Complex sex allocation in the laughing kookaburra. Behav. Ecol. 12, 524-533. Lehmann L, Keller L (2006). The evolution of cooperation and altruism - a general framework and a classification of models. J. Evol. Biol. 19, 1365-1376. Leimar O (1996). Life-history analysis of the Trivers and Willard sex-ratio problem. Behav. Ecol. 7, 316-325. Lennartz MR, Harlow RF (1979). Role of parent and helper red-cockaded woodpeckers at the nest. Wilson Bull. 91, 331-335. Lennartz MR, Hooper RG, Harlow RF (1987). Sociality and cooperative breeding of red-cockaded woodpeckers, Picoides borealis . Behav. Ecol. Sociobiol. 20, 77-88. Lessells CM (1990). Helping at the nest in European bee-eaters: who helps and why? In: Population biology of passerine birds: an integrated approach. Blondel J, Gosler A, Lebreton J-D, McCleery RH (eds). NATO ASI series, Berlin: Springer-Verlag, pp 357-368. Lessells CM, Avery MI (1987). Sex ratio selection in species with helpers at the nest: some extensions of the repayment model. Am. Nat. 129, 610- 620. Lessells CM, Boag PT (1987). Unrepeatable repeatabilities: a common mistake. Auk 104, 116-121. Li S-H, Brown JL (2002). Reduction of maternal care: a new benefit of multiple mating? Behav. Ecol. 13, 87-93. Lifjeld JT , Dunn PO, Westneat DF (1994). Sexual selection by sperm competition in birds: male-male competition or female choice ? J. Avian Biol. 25 , 244-250 . Ligon JD, Ligon SH (1978). Communal breeding in green woodhoopoes as a case for reciprocity. Nature 276, 496-498. Ligon JD, Ligon SH (1990). Female-biased sex ratio at hatching in the green woodhoopoe. Auk 107, 765-771. Lundy KJ, Parker PG, Zahavi A (1998). Reproduction by subordinates in cooperatively breeding Arabian babblers is uncommon but predictable. Behav. Ecol. Sociobiol. 43, 173-180.
173
Lynch M, Ritland K, (1999). Estimation of pairwise relatedness with molecular markers. Genetics 152, 1753-1766. Lynch M, Walsh B (1998). Genetics and data analysis of quantitative traits. Sunderland: Sinauer Associates. Lynch M, Conery J, Burger R (1995). Mutation accumulation and the extinction of small populations. Am. Nat . 146, 489-518.
M
MacColl ADC, Hatchwell BJ (2004). Determinants of lifetime fitness in a cooperative breeder, the long-tailed tit Aegithalos caudatus . J. Anim. Ecol. 73, 1137-1148. Mays HL, Albrecht T, Liu M, Hill GE (2008). Female choice for genetic complementarity in birds: a review. Genetica 134, 147-158. Maddison W, Maddison D (1997). MESQUITE. http://mesquiteproject.org Magrath RD (2001). Group breeding dramatically increases reproductive success of yearling but not older female scrubwrens: a model for cooperatively breeding birds? J. Anim. Ecol. 70, 370-385. Magrath RD, Whittingham LA (1997). Subordinate males are more likely to help if unrelated to the breeding female in cooperatively breeding white-browed scrubwrens. Behav. Ecol. Sociobiol. 41, 185-192. Magrath RD, Yezerinac SM (1997). Facultative helping does not influence reproductive success or survival in cooperatively breeding white- browed scrubwrens. J. Anim. Ecol. 66, 658-670. Marshall TC, Slate J, Kruuk LEB, Pemberton JM (1998). Statistical confidence for likelihood-based paternity inference in natural populations. Molec. Ecol. 7, 639-655. Maynard Smith J (1964). Group selection and kin selection. Nature 201, 145- 147. Maynard Smith J (1978). The evolution of sex. Cambridge: Cambridge University Press. McGowan A, Hatchwell BJ, Woodburn RJW (2003). The effect of helping behaviour on the survival of juvenile and adult long-tailed tits Aegithalos caudatus . J. Anim. Ecol. 72, 491-499. Meade J, Nam K-B, Beckerman AP, Hatchwell BJ (2010). Consequence of 'load-lightening' for future indirect fitness gains by helpers in a cooperatively breeding bird. J. Anim. Ecol. 79, 529-537.
174
Møller AP (2000). Male parental care, female reproductive success, and extra- pair paternity. Behav. Ecol. 11, 161-168. Møller AP, Birkhead TR (1993). Certainty of paternity covaries with paternal care in birds. Behav. Ecol. Sociobiol. 33, 261-268. Møller AP, Cuervo JJ (2000). The evolution of paternity and paternal care in birds. Behav. Ecol. 11, 472-485. Mulder RA (1997). Extra-group courtship displays and other reproductive tactics of superb fairy-wrens. Austr. J. Zool. 45, 131-143. Mulder RA, Cockburn A (1993). Sperm competition and the reproductive anatomy of male superb fairy-wrens. Auk 1993, 110, 588-593. Mulder RA, Langmore NE (1993). Dominant males punish helpers for temporary defection in superb fairy-wrens. Anim. Behav. 45, 830-833. Mulder RA, Magrath MJL (1994). Timing of prenuptial molt as a sexually selected indicator of male quality in superb fairy-wrens ( Malurus cyaneus ). Behav. Ecol. 5, 393-400. Mulder RA, Dunn PO, Cockburn A, Lazenby-Cohen KA, Howell MJ (1994). Helpers liberate female fairy-wrens from constraints on extra-pair mate choice. Proc. R. Soc. Lond. B. 255, 223-229. Mumme RL (1992). Do helpers increase reproductive success - an experimental analysis in the Florida scrub jay. Behav. Ecol. Sociobiol. 31, 319-328. Mumme RL, Koenig WD, Zink RM, Marten JA (1985). Genetic variation and parentage in a California population of acorn woodpeckers. Auk 102, 305-312.
N
Nager RG, Managhan P, Griffiths R, Houston DC, Dawson R (1999). Experimental demonstration that offspring sex ratio varies with maternal condition. Proc. Natl. Acad. Sci. USA 96, 570-573. Nakamura M (1998). Multiple mating and cooperative breeding in polygynandrous alpine accentors. I. Competition among females. Anim. Behav. 55, 259-275. Nam K-B, Simeoni M, Sharp SP, Hatchwell BJ (2010). Kinship affects investment by helpers in a cooperatively breeding bird. Proc. R. Soc. Lond. B. 277, 3299-3306.
175
Neal JC, James DA, Montague WG, Johnson JE (1993). Effects of weather and helpers on survival of nestling red-cockaded woodpeckers. Wilson Bull. 105, 666-673. Neff BD, Pitcher TE (2005). Genetic quality and sexual selection: an integrated framework for good genes and compatible genes. Molec. Ecol. 14, 19-38. Nias RC (1986). Nest-site characteristics and reproductive success in the superb fairy-wren. Emu 86, 139-144. Nias RC, Ford HA (1992). The influence of group-size and habitat on reproductive success in the superb fairy-wren Malurus cyaneus . Emu 92, 238-243. Nowak MA, Sigmund K (1998). Evolution of indirect reciprocity by image scoring. Nature 393, 573-577.
O
Oh KP, Badyaev AV (2006). Adaptive genetic complementarity in mate choice coexists with selection for elaborate sexual traits. Proc. R. Soc. Lond. B. 273, 1913-1919. Olsen PD, Cockburn A (1991). Female-biased sex allocation in peregrine falcons and other raptors. Behav. Ecol. Sociobiol. 28, 417-423. Owens IPF, Bennett PM (1994). Mortality costs of parental care and sexual dimorphism in birds. Proc. R. Soc. Lond. B. 257, 1-8. Owens IPF, Hartley IR (1998). Sexual dimorphism in birds: why are there so many different forms of dimorphism. Proc. R. Soc. Lond. B. 265, 397- 407.
P
Paradis E, Claude J, Strimmer K (2004). APE: analyses of phylogenetics and evolution in R language. Bioinformatics, 20, 289-290. Parry V (1973). The auxiliary system and its effect on territory and breeding in kookaburras. Emu 73, 81-100. Patterson CB, Erckmann WJ, Orians GH (1980). An experimental study of parental investment and polygyny in male blackbirds. Am. Nat. 116, 757-769.
176
Peakall R, Smouse PE (2006). GENALEX 6: genetic analysis in Excel. Population genetic software for teaching and research. Molec. Ecol. Notes 6, 288-295. Pen I, Weissing FJ (2000). Sex ratio optimization with helpers at the nest. Proc. R. Soc. Lond. B. 267, 539-544. Peters A, Astheimer LB, Cockburn A (2001). The annual testosterone profile in cooperatively breeding superb fairy-wrens, Malurus cyaneus , reflects their extreme infidelity. Behav. Ecol. Sociobiol. 50, 519-527. Peters A, Cockburn A, Cunningham R (2002). Testosterone treatment suppresses paternal care in superb fairy-wrens, Malurus cyaneus , despite their concurrent investment in courtship. Behav. Ecol. Sociobiol. 51, 538-547. Petrie M, Kempenaers B (1998). Extra-pair paternity in birds: explaining variation between species and populations. Trends Ecol. Evol. 13, 52- 58. Petrie M, Doums C, Møller AP (1998). The degree of extra-pair paternity increases with genetic variability. Proc. Natl. Acad. Sci. USA 95, 9390- 9395. Pike TW (2005). Sex ratio manipulation in response to maternal condition in pigeons: evidence for pre-ovulatory follicle selection. Behav. Ecol. Sociobiol. 58, 407-413. Pinheiro JC, Bates DM (2000). Mixed-effects models in S and S-plus. New York: Springer. Pinheiro J, Bates D, DebRoy S, Sarkar D, R-Core Team (2009). Nlme: Linear and nonlinear mixed effects models. R-package version 3.1-94. Piper WH (1994). Courtship, copulation, nesting behavior and brood parasitism in the Venezuelan stripe-backed wren. Condor 96, 654-671. Piper WH, Parker P, Rabenold KN (1995). Facultative dispersal by juvenile males in the cooperative stripe-backed wren. Behav. Ecol. 6, 337-342. Pitcher TE, Dunn PO, Whittingham LA (2005). Sperm competition and the evolution of testes size in birds. J. Evol. Biol. 18, 557-567. Poiani A (1993). Effects of clutch size manipulations on reproductive-behavior and nesting success in the cooperatively breeding bell miner ( Manorina melanophrys ). Evol. Ecol. 7, 329-356. Poldmaa T, Montgomerie R, Boag P (1995). Mating system of the cooperatively breeding noisy miner Manorina melanocephala , as revealed by DNA profiling. Behav. Ecol. Sociobiol. 37, 137-143.
177
Pruett-Jones S, Tarvin KA (2001). Aspects of the ecology and behaviour of white-winged fairy-wrens on Barrow Island. Emu 101, 73-78. Pusey A, Wolf M (1996). Inbreeding avoidance in animals. Trends Ecol. Evol. 11, 201-206.
Q
Queller DC, Goodnight KF (1989). Estimating relatedness using genetic markers. Evolution 43, 258-275. Queller DC, Strassmann JE (1998). Kin selection and social insects. BioScience 48, 165-175. Quinn JS, Woolfenden GE, Fitzpatrick JW, White BN (1999). Multi-locus DNA fingerprinting supports genetic monogamy in Florida scrub-jays. Behav. Ecol. Sociobiol. 45, 1-10.
R
R Development Core Team (2008). R: A Language and environment for statistical computing. Vienna: R Foundation for Statistical Computing. http://www.R-project.org. R Development Core Team (2009). R: A Language and environment for statistical computing. Vienna: R Foundation for Statistical Computing. http://www.R-project.org. R Development Core Team (2010). R: A Language and environment for statistical computing. Vienna: R Foundation for Statistical Computing. http://www.R-project.org. Rabenold KN (1984). Cooperative enhancement of reproductive success in tropical wren societies. Ecology 65, 871-885. Rabenold KN (1990). Campylorhynchus wrens: the ecology of delayed dispersal and cooperation in the Venezuelan savanna. In: Cooperative breeding in birds: long-term studies of ecology and behavior. Stacey PB, Koenig WD (eds). Cambridge: Cambridge University Press, pp 159- 196. Rabenold PP, Rabenold KN, Piper WH, Haydock J, Zack SW (1990). Shared paternity revealed by genetic analysis in cooperatively breeding tropical wrens. Nature 348, 538-540.
178
Rathburn MK, Montgomerie R (2005). Offspring sex ratios correlate with pair-male condition in a cooperatively breeding fairy-wren. Behav. Ecol. 16, 31-47. Richardson D, Jury F, Blaakmeer K, Komdeur J, Burke T (2001). Parentage assignment and extra-group paternity in a cooperative breeder: the Seychelles warbler ( Acrocephalus sechellensis ). Molec. Ecol. 10, 2263- 2273. Richardson DS, Burke T, Komdeur J (2002). Direct benefits and the evolution of female-biased cooperative breeding in Seychelles warblers. Evolution 56, 2313-2321. Ridley AR (2007). Factors affecting offspring survival and development in a cooperative bird: social, maternal and environmental effects. J. Anim. Ecol. 76, 750-760. Roberts G (1998). Competitive altruism: from reciprocity to the handicap principle. Proc. R. Soc. Lond. B. 265, 427-431. Rösner D, Bogatz K, Trapp H, Grünkorn T, Brandl R (2009). No evidence of skewed secondary sex ratios in nestling of the common raven ( Corvus corax ). J. Ornithol. 150, 293-297. Rowe M, Pruett-Jones S (2006). Reproductive biology and sperm competition in Australian fairy-wrens. Avian Poultry Biol. Rev. 17, 21-37. Rowe M, Bakst MR, Pruett-Jones S (2008). Good vibrations? Structure and function of the cloacal tip of male Australian Maluridae. J. Avian Biol. 39, 348-354. Rowley I (1965) . The life history of the superb blue wren Malurus cyaneus. Emu 64, 251-297. Rowley I (1981). The communal way of life in the splendid fairy wren, Malurus splendens . Z. Tierpsychol. 55, 228-267. Rowley I, Brooker M, Russell E (1991). The breeding biology of the splendid fairy-wren Malurus splendens : the significance of multiple broods. Emu 91, 197-221. Rowley I, Russell E (1990). Splendid fairy-wrens: demonstrating the importance of longevity. In: Cooperative breeding in birds: long-term studies of ecology and behavior. Stacey PB, Koenig WD (eds). Cambridge: Cambridge University Press, pp 3-30. Rowley I, Russell E (1993). The purple-crowned fairy-wren Malurus coronatus . I. History, distribution and present status. Emu 93, 220- 234.
179
Rowley I, Russell E (1993). The purple-crowned fairy-wren Malurus coronatus . II. Breeding biology, social organisation, demography and management. Emu 93, 235-250. Rowley I, Russell E (1997). Fairy-wrens and grasswrens. Oxford: Oxford University Press. Rowley I, Russell E, Brooker M (1986). Inbreeding - Benefits may outweigh costs. Anim. Behav. 34, 939-941. Rubenstein DR (2007a). Female extrapair mate choice in a cooperative breeder: trading sex for help and increasing offspring heterozygosity. Proc. R. Soc. Lond. B. 274, 1895-1903. Rubenstein DR (2007b). Temporal but not spatial environmental variation drives adaptive offspring sex allocation in a plural cooperative breeder. Am. Nat. 170, 155-165. Rubenstein DR, Lovette IJ (2009). Reproductive skew and selection on female ornamentation in social species. Nature 462, 786-789. Russell AF, Hatchwell BJ (2001). Experimental evidence for kin-based helping in a cooperatively breeding vertebrate. Proc. R. Soc. Lond. B. 268, 2169-2174. Russell AF, Langmore NE, Cockburn A, Astheimer LB, Kilner RM (2007). Reduced egg investment can conceal helper effects in cooperatively breeding birds. Science 317, 941-944. Russell E, Rowley I (1988). Helper contributions to reproductive success in the splendid fairy-wren ( Malurus splendens ). Behav. Ecol. Sociobiol. 22, 131-140. Russell EM, Rowley I (1993). Philopatry or dispersal - competition for territory vacancies in the splendid fairy-wren, Malurus splendens . Anim. Behav. 45, 519-539. Russell E, Rowley I (2000). Demography and social organisation of the red- winged fairy-wren, Malurus elegans . Austr. J. Zool. 48, 161-200.
S
Sax A, Hoi H (1998). Individual and temporal variation in cloacal protuberance size of male bearded tits ( Panurus biarmicus ). Auk 115, 964-969. Schall R (1991). Estimation in generalized linear models with random effects. Biometrika 78, 719-727.
180
Schlicht E, Kempenaers B (2011). Extra-pair paternity and sexual selection. In: From genes to animal behavior: social structures, personalities, communication by color. Inoue-Murayama M, Kawamura S, Weiss A (eds.). Tokyo: Springer, pp 35-65. Seddon N, Tobias JA, Butchart SHM (2003). Group living, breeding behaviour and territoriality in the subdesert mesite Monias benschi . Ibis 145, 277-294. Seddon N, Amos W, Adcock G, Johnson P, Kraaijeveld K, Kraaijeveld-Smit FJL, Lee W, Senapathi GD, Mulder RA, Tobias JA (2005). Mating system, philopatry and patterns of kinship in the cooperatively breeding subdesert mesite Monias benschi . Molec. Ecol. 14, 3573-3583. Seger J, Stubblefield JW (2002). Models of sex ratio evolution. In: Sex ratios. Concepts and research methods. Hardy ICW (ed). Cambridge: Cambridge University Press, pp 2-25. Sheldon BC (1993). Sexually transmitted disease in birds: occurrence and evolutionary significance. Phil. Trans. R. Soc. B. 339, 491-497. Sheldon BC, Andersson S, Griffith SC, Örnborg J, Sendecka J (1999). Ultraviolet colour variation influences blue tit sex ratios. Nature 402, 874-877. Shine R, Brown GP (2008). Adapting to the unpredictable: reproductive biology of vertebrates in the Australian wet-dry tropics. Phil. Trans. R. Soc. B. 363, 363-373. Skroblin A, Legge S (2010). The distribution and status of the western subspecies of the purple-crowned fairy-wren ( Malurus coronatus coronatus ). Emu 110, 339-347. Sloane SA (1996). Incidence and origins of supernumeraries at bushtit (Psaltriparus minimus ) nests. Auk 113, 757-770. Solomon NG, French JA (1996). Cooperative breeding in mammals. Cambridge: Cambridge University Press . Stacey PB (1979). Kinship, promiscuity, and communal breeding in the acorn woodpecker. Behav. Ecol. Sociobiol. 6, 53-66. Stacey PB, Koenig WD (1990). Cooperative breeding in birds: long-term studies of ecology and behavior. Cambridge: Cambridge University Press. Stallcup JA, Woolfenden GE (1978). Family status and contributions to breeding by Florida scrub jays. Anim. Behav. 26, 1144-1156.
181
Stearns SC (1989). Trade-offs in life history evolution. Funct. Ecol. 3, 259- 268. Stutchbury BJM (1998). Breeding synchrony best explains variation in extra- pair mating system among avian species. Behav. Ecol. Sociobiol. 43, 221-222. Stutchbury BJM (1998). Female mate choice of extra-pair males: breeding synchrony is important. Behav. Ecol. Sociobiol. 43, 213-215. Stutchbury BJ, Morton E (1995). The effect of breeding synchrony on extra- pair mating systems in songbirds. Behaviour 132, 675-690. Stutchbury BJ, Morton ES, Piper WH (1998). Extra-pair mating system of a synchronously breeding tropical songbird. J. Avian Biol. 29, 72-78. Sumner S, Kelstrup H, Fanelli D (2010). Reproductive constraints, direct fitness and indirect fitness benefits explain helping behaviour in the primitively eusocial wasp, Polistes candensis. Proc. R. Soc. Lond. B. 277, 1721-1728. Sunde P, Bolstad MS, Møller JD (2003). Reversed sexual dimorphism in Tawny Owls, Strix aluco , correlates with duty division in breeding effort. Oikos 101, 265-278. Suter SM, Keiser M, Feignoux R, Meyer DR (2007). Reed bunting females increase fitness through extra-pair mating with genetically dissimilar Proc. R. Soc. Lond. B. 274, 2865-2871.
T
Taborsky M (1984). Broodcare helpers in the cichlid fish Lamprologus brichardi : their costs and benefits. Anim. Behav. 32, 1236-1252. Tarvin KA, Webster MS, Tuttle EM, Pruett-Jones S (2005). Genetic similarity of social mates predicts the level of extrapair paternity in splendid fairy-wrens. Anim. Behav. 70, 945-955. Temple HJ (2005). Ecology, cooperative breeding and conservation of the white-breasted thrasher Ramphocinclus brachyurus . University of Cambridge: PhD thesis. Temple HJ, Hoffman JI, Amos W (2009). Group structure, mating system and extra-group paternity in the co-operatively breeding white-breasted thrasher Ramphocinclus brachyurus . Ibis 151, 99-112.
182
Thuman KA, Widemo F, Griffith SC (2003). Condition-dependent sex allocation in a lek-breeding wader, the ruff ( Philomachus pugnax ). Molec. Ecol. 12, 213-218. Tibbetts E, Pruett-Jones S (1999). Habitat and nest-site partitioning in splendid and variegated fairy-wrens (Aves: Maluridae). Austr. J. Zool. 47, 317-326. Torres R, Drummond H (1999). Does large size make daughters of the blue- footed booby more expensive than sons? J. Anim. Ecol. 68, 1133-1141. Townsend AK, Clark AB, McGowan KJ, Lovette IJ (2009). Reproductive partitioning and the assumptions of reproductive skew models in the cooperatively breeding American crow. Anim. Behav. 77, 503-512. Tregenza T, Wedell N (2000). Genetic compatibility, mate choice and patterns of parentage. Molec. Ecol. 9, 1013-1027. Trivers RL (1972). Parental investment and sexual selection. In: Sexual selection and the descent of man 1871-1971. Campbell B (ed). Chicago: Aldine Press, pp 136-179. Trivers RL, Willard DE (1973). Natural selection of parental ability to vary the sex ratio of offspring. Science 179, 90-92. Tuttle EM, Pruett-Jones S, Webster MS (1996). Cloacal protuberances and extreme sperm production in Australian fairy-wrens. Proc. R. Soc. Lond. B. 263, 1359-1364.
V
Varian-Ramos CW, Karubian J, Talbott V, Tapia I, Webster MS (2010). Offspring sex ratios reflect lack of repayment by auxiliary males in a cooperatively breeding passerine. Behav. Ecol. Sociobiol. 64, 967-977. Veltman CJ (1989). Flock, pair and group living lifestyles without cooperative breeding by Australian magpies Gymnorhina tibicen . Ibis 131, 601-608.
W
Wagner RH (1993). The pursuit of extra-pair copulations by female birds: a new hypothesis of colony formation. J. Theor. Biol. 163, 333-346.
183
Webster MS, Tarvin KA, Tuttle EM, Pruett-Jones S (2004). Reproductive promiscuity in the splendid fairy-wren: effects of group size and auxiliary reproduction. Behav. Ecol. 15, 907-915. Webster MS, Varian CW, Karubian J (2008). Plumage color and reproduction in the red-backed fairy-wren: why be a dull breeder? Behav. Ecol. 19, 517-524. West SA, Sheldon BC (2002). Constraints in the evolution of sex ratio adjustment. Science 295, 1685-1688. West SA, Reece SE, Sheldon BC (2002). Sex ratios. Heredity 88, 117-124. West SA, Shuker DM, Sheldon BC (2005). Sex-ratio adjustment when relatives interact: a test of constraints on adaptation. Evolution 59, 1211-1228. West SA, Gardner A, Griffin AS (2006). Quick guide: altruism. Curr. Biol. 16, R482-R483. West SA, Griffin AS, Gardner A (2007). Evolutionary explanations for cooperation. Curr. Biol. 17, R661-R672. Westneat DF, Sherman PW (1997). Density and extra-pair fertilizations in birds: a comparative analysis. Behav. Ecol. Sociobiol. 41, 205-215. Westneat DF, Stewart IRK (2003). Extra-pair paternity in birds: causes, correlates, and conflict. Ann. Rev. Ecol. Evol. Syst. 34, 365-396. Whittingham LA, Dunn PO (2000). Offspring sex ratios in tree-swallows: females in better condition produce more sons. Molec. Ecol. 9, 1123- 1129. Whittingham LA, Dunn PO, Magrath RD (1997). Relatedness, polyandry and extra-group paternity in the cooperatively-breeding white-browed scrubwren ( Sericornis frontalis ). Behav. Ecol. Sociobiol. 40, 261-270. Wiebe KL, Bortolotti GR (1992). Facultative sex ratio manipulation in American kestrels. Behav. Ecol. Sociobiol. 30, 379-386. Wiens JJ (2001). Widespread loss of sexually selected traits: how the peacock lost its spots. Trends Ecol. Evol. 16, 517-523. Williams GC (1966) . Natural selection, the cost of reproduction, and a refinement of Lack's principle . Am. Nat. 100, 687-690. Williams GC (1979). The question of adaptive sex ratio in outcrossed vertebrates. Proc. R. Soc. Lond. B. 205, 567-580. Williams DA (2000). Routes to fitness in cooperative breeding brown jays. Perdue University: PhD thesis.
184
Williams DA (2004). Female control of reproductive skew in cooperatively breeding brown jays ( Cyanocorax morio ). Behav. Ecol. Sociobiol. 55, 370-380. Williams DA, Hale AM (2006). Helper effects on offspring production in cooperatively breeding brown jays ( Cyanocorax morio ). Auk 123, 847- 857. Williams DA, Lawton MF, Lawton RO (1994). Population-growth, range expansion, and competition in the cooperatively breeding brown jay, Cyanocorax morio . Anim. Behav. 48, 309-322. Wilson K, Hardy ICW (2002). Statistical analysis of sex ratios: an introduction. In: Sex ratios. Concepts and research methods. Hardy ICW (ed). Cambridge: Cambridge University Press, pp 48-92 Wink M, Dyrcz AJ (1999). Mating systems in birds: a review of molecular studies. Acta Ornithol. 34, 91-109. Woolfenden GE (1981). Selfish behavior by Florida scrub-jay helpers. In: Natural selection and social behavior. Alexander R, Tinkle D (eds). New York: Chiron Press, pp 257-260. Woolfenden GE, Fitzpatrick JW (1990). Florida scrub-jays: a synopsis after 18 years of study. In: Cooperative breeding in birds: long-term studies of ecology and behavior. Stacey PB, Koenig WD (eds). Cambridge: Cambridge University Press, pp 241-266. Woxvold IA (2005). Social organization, cooperative breeding and sex allocation in the apostlebird, Struthidea cinerea . University of Melbourne: PhD thesis. Woxvold IA, Magrath MJL (2005). Helping enhances multiple components of reproductive success in the cooperatively breeding apostlebird. J. Anim. Ecol. 74, 1039-1050. Woxvold IA, Mulder RA (2008). Mixed mating strategies in cooperatively breeding apostlebirds Struthidea cinerea . J. Avian Biol. 39, 50-56. Wrege PH, Emlen ST (1987). Biochemical determination of parental uncertainty in white-fronted bee-eaters. Behav. Ecol. Sociobiol. 20, 153- 160. Wright J (1998). Helping-at-the-nest and group size in the Arabian babbler Turdoides squamiceps . J. Avian Biol. 29, 105-112. Wright J (2007). Cooperation theory meets cooperative breeding: exposing some ugly truths about social prestige, reciprocity and group augmentation . Behav. Process. 76, 142-148.
185
Wright J, Parker PG, Lundy KJ (1999). Relatedness and chick feeding effort in the cooperatively breeding Arabian babbler. Anim. Behav. 58, 779- 785. Wright J, McDonald PG, te Marvelde L, Kazem AJN, Bishop BM (2010). Helping effort increases with relatedness in bell miners, but ‘unrelated’ helpers of both sexes still provide substantial care. Proc. R. Soc. Lond. B. 277, 437-445.
Z
Zahavi A (1974). Communal nesting by Arabian babblers: a case of individual selection. Ibis 116, 84-87. Zahavi A (1990). Arabian Babblers: the quest for social status in a cooperative breeder. In: Cooperative breeding in birds: long-term studies of ecology and behavior. Stacey PB, Koenig WD (eds). Cambridge: Cambridge University Press, pp 105-130. Zahavi A (1995). Altruism as a handicap - the limitations of kin selection and reciprocity. J. Avian Biol. 26: 1-3. Zeh JA, Zeh DW (1996). The evolution of polyandry I: intragenomic conflict and genetic incompatibility. Proc. R. Soc. Lond. B. 263, 1711-1717. Zeh JA, Zeh DW (1997). The evolution of polyandry II: post-copulatory defenses against genetic incompatibility. Proc. R. Soc. Lond. B. 264, 69- 75.
186
Erklärung
Ich erkläre hiermit, dass ich die vorliegende Arbeit ohne unzulässige Hilfe Dritter und ohne Benutzung anderer als der angegebenen Hilfsmittel angefertigt habe. Die aus anderen Quellen direkt oder indirekt übernommenen Daten und Konzepte sind unter Angabe der Quelle gekennzeichnet. Weitere Personen, insbesondere Promotionsberater, waren an der inhaltlich materiellen Erstellung dieser Arbeit nicht beteiligt. Die Arbeit wurde bisher weder im In- noch im Ausland in gleicher oder ähnlicher Form einer Prüfungsbehörde vorgelegt.
Konstanz, den 10.03.2011.
Sjouke Anne Kingma
187
188
Record of achievement
Abgrenzung der Eigenleistung
The ideas presented in this thesis were formed by me, and extensively discussed with my supervisors Anne Peters and Michelle Hall. All research chapters, as well as ‘general parts’ in this thesis were written by me, and commented on by co-authors. I performed all statistical analyses myself.
I performed field-work and supervised field-assistants for three periods of three months in Australia between 2007 and 2010 (and a 3 month period as field-assistant in 2006). Year-round fieldwork was carried out by Michelle Hall and several field-assistants.
Molecular work involved several steps. DNA was extracted from blood samples (by Michelle Hall, Evi Fricke, and me), and sent to Germany where molecular sexing of all individuals was conducted by Evi Fricke. I optimised the PCR methods for genotyping, under supervision of Gernot Segelbacher. Further genotyping was done by Ecogenics GmbH. I conducted all the parentage analyses and further analyses using these samples.
189
190
Acknowledgements
Undertaking this research has been a fantastic experience. Many people were crucially important for my research, and I would like to thank everybody who contributed to the realisation of this thesis.
First, and foremost, I am extremely grateful to my supervisors Anne Peters and Michelle Hall from whom I have learned an incredible amount. I am very grateful for their support and input, and the freedom to pursue my interests (especially because some of those must have demanded quite a bit of patience and imagination). Their sharp (and fast) comments and thoughts on ideas, proposals, preliminary graphs, and manuscripts have contributed much to this thesis, and to my current way of conducting science.
Anne has been a great motivator, and with unrestricted dedication she has taught me how to do (and write) to-the-point science. Also, ‘behind the scenes’ Anne has always been there, which made my stay in Radolfzell very nice (not in the last place the traditional pizza evenings with Kaspar, Janneke and Joaquin, which and whom I will miss).
This study would not have been possible without Michelle’s dedication in year-round field-work, guidance of the field-assistants, and keeping an amazing overview on what was going on in our population of fairy-wrens. Michelle is a great example for me of ‘how-to-do-fieldwork’, and has from a distance greatly supervised in conceiving ideas and in writing. Additionally, I am grateful to Michelle for saving my life on my first day in Australia ☺.
Kaspar Delhey was always there for discussion and I greatly appreciate Kaspar’s encouraging comments. The discussions about statistics have formed my feeling that these are often (not always) an interesting puzzle rather than the ‘necessary-evil-for-a-field-biologist’.
Several people have assisted me in learning and conducting lab-work. Gernot Segelbacher generously taught me the ins-and-outs of genotyping in his lab in Freiburg. Steve Murphy assisted in the lab in Mornington. Evi Fricke has been of invaluable help for all sorts of molecular (and other) work; not in the
191
last place conducting the molecular sexing of 779 birds. Additionally, I thank Bernhard Koller from Ecogenics for understanding my requests (and acting accordingly) to do the genotyping ‘right-now’. A great number of field-assistants made detailed knowledge possible about our population of birds and I am extremely grateful for their hard and accurate work in the field! A special thanks goes also to the M.Sc. students Joppe Rijpstra and Rinskje Klooster, who did their research project on the purple-crowned fairy-wrens and with whom I had a great time in the field and in Radolfzell. Having had the opportunity to conduct research in Mornington Wildlife Sanctuary has been a great privilege. I thank the Australian Wildlife Conservancy for providing this opportunity, and the ‘resident’ crew in Mornington for making my time there real fun: Anja, Butch, Doug, Grassy Graham, Jo, Jojo, Julie, KD, Lyn, Malcolm, Michelle, Olya, Richie, Sara, Steve, Swanie, and everbody else. Cheers mates! I am also grateful to many colleagues (PhD’s and scientists in Radolfzell, Seewiesen, Mornington and at the ‘ Malurus -meeting’ in Canberra) and everyone else, who collaborated on this research, and/or provided useful discussion or comments on my manuscripts (and for accepting my Frisian ‘honest opinion’). Among others those were Adam Fudickar, Alex Baugh, Andrew Cockburn, Anja Matuszak, Anja Skroblin, Barbara Helm, Bart Kempenaers, Bart Kranstauber, Ben Hatchwell, Bernd Leisler, Catarina Miranda, David Santos, Dina Dechman, Elena Arriero, Günther Bauer, Isabel Smallegange, Jesko Partecke, Kamran Safi, Kaspar Delhey, Mark Roberts, Martin Wikelski, Michaela Hau, Mike Webster, Ralf Kurvers, Riek van Noordwijk, Rob Magrath, Sigal Liberman, Tim Greives, Tim Birkhead, Volker Salewski, Wolfgang Fiedler, Willow Lindsay, (etc, etc). I thank Barbara Helm for kindly translating the summary of this thesis in German. Several researchers have given me insight in unpublished data of their study and other information for my comparative work (Amotz Zahavi, Jordan Karubian, Mike Webster, Peter Dunn, Rita Covas, Rob Heinsohn), rainfall data in Mornington (Jo Heathcote) and the fairy-wren phylogeny (Janet Gardner), for which I am most grateful. In the Vogelwarte, Bernhard, Claudia, Evi, Gerlinde, Kumar, Rolf, Uli, Uschi, and many others have provided generous help in numerous ways
192
throughout the years, and my life would have been much more difficult without them. Furthermore, I thank everyone (and those are really too many people to mention personally, but certainly not less important than everybody else) who made my stay in Germany lively and exciting. Among other things I really enjoyed the Rossittenstrasse BBQ’s, (international) diners, game nights, happy hours (keep going Tim), the movie night (happened to be my favourite movie), lakeside- and catchingstation-winedrinking (special mention of Volker is warranted here), and the English pub and English breakfasts (Jezz and Sinead). Pictures in this thesis were taken in Mornington Wildlife Santuary by Michelle Hall (pages 21, 22, 49, 91, 117), Doug Adams (pages 71, 139), and Wayne Lawler (page 25).
A special ‘dankewol’ goes to my parents and my family, for their support, interest, and for forgiving me that I was away too often for too long.
I thank Sigal for her love, patience and faith in me, and for the great time together.
This research was funded by the ‘Minerva Sonderprogramm zur Förderung hervorragender Wissenschaftlerinnen’ of the Max Planck Society (awarded to Anne Peters).
193
194
Curriculum vitae
Identification:
Name: Kingma First names: Sjouke Anne Date of birth: 10-04-1982 Place of birth: West Dongeradeel, The Netherlands Adress: Bahnhofplatz 9, D-78315, Radolfzell, Germany Telephone: +49-7732-150133 E-mail: [email protected]
Education:
2007-current. PhD research. Max Planck institute for ornithology, Radolzell, Germany. University of Konstanz, Germany.
2000-2006. B.Sc. and M.Sc. (specialisation: Ecology) University of Groningen, The Netherlands. Cum laude
Teaching experience
August 2007-April 2008 Supervision two M.Sc. students , University of Groningen Field-work, analyses and writing (cosupervision; M. Hall and A. Peters)
2007-2010 Supervision of several field-assistents in Australia (cosupervision; M. Hall)
April-May 2006 Supervision two 3 rd -year B.Sc. students , University of Groningen. Field-work period of 8 weeks (cosupervision; O. Vedder and J.Komdeur)
Awards and Funding
Presentation prize. German Zoological Society PhD meeting, Munich, Germany. 5-8 March 2009. Presentation prize. Edward Grey Conference, Groningen, the Netherlands. 5- 8 January 2006. Runner-up nomination presentation. Edward Grey Conference, Oxford, UK. 5-7 January 2005.
195
Groninger University Fund, for fieldwork Hungary (2002): 320 euro Marco Polo Travel Fund, for fieldwork Hungary (2002): 300 euro
Professional memberships
International Society of Behavioral Ecology (ISBE) Association for the Study of Animal Behaviour (ASAB) European Ornithologists’ Union (EOU) Dutch Ornithology Union (NOU) American Society of Naturalists (ASN)
Peer-review of research articles
Journal of Animal Ecology Behavioral Ecology Ethology Ornis Scandinavica Genetics and Molecular Biology Vogelwelt (in German) African Journal of Biotechnology
Activities and Skills:
Organisation and attendance of statistics course ‘Generalised linear mixed models in R’ with Prof. Douglas Bates (~35 attendants in Seewiesen, Germany, 21-23 July 2009).
Chairman of the board of the Frisian student union F.F.J. Bernlef, Groningen, the Netherlands. (~ 200 members). May 2005-May 2006
Languages: Frisian, Dutch, English, German (B1+), French (basics), Latin (basics)
Computer and analytical skills: Word, Excel, Powerpoint, Photoshop, R (basics and e.g., mixed modelling using lme4, nlme, Phylogenetic analyses using ‘ape’ and ‘gee’), SPSS, GenStat, Sigmaplot, The Observer 4.1 (Noldus; video-observation analyses), Cervus (paternity analyses), GeNaLEX (relatedness and genetic structure data analyse), Pedigree viewer (visualisation and analyses of pedigrees), Mesquite (program to display phylogenetic data).
196
Conference attendance (Presentations and Posters)
Presentations
Kingma SA, Hall ML, Peters A. Radical loss of an extreme extra-pair mating system in cooperatively breeding fairy-wrens: consequences for cooperative breeding. ISBE-conference, Perth, Australia. 26 September-2 October 2010. Kingma SA, Hall ML, Rijpstra JJ, Peters A. Helpers help: positive effects of subordinates on reproductive success as well as breeder workload and survival in cooperatively breeding fairy-wrens. EOU-conference, Zurich, Switzerland. 21-26 August 2009. Kingma SA, Hall ML, Segelbacher G, Peters A. Faithful males do not bring flowers: rapid loss of an extreme extra-pair mating system. DZG PhD- congress, Munich, Germany 5-8 March 2009. ( Prize of best presentation ). Kingma SA, Hall ML, Peters A. Purple-crowned fairy-wrens: Conflict or cooperation? EGI-congress, Oxford, UK. 8-10 January 2008. Kingma SA, Komdeur J, Vedder O, Von Engelhardt N, Korsten P, Groothuis TGG. Differential androgen allocation in blue tit eggs ( Parus caeruleus ): do females respond to experimentally reduced mate attractiveness? EGI-congress, Groningen, The Netherlands. 5-8 January 2006. ( Prize of best presentation ). Kingma SA, Szentirmai I, Bókony V, Liker A, Bleeker M, Székely T, Komdeur J. Sexual selection and the function of melanin-based plumage in penduline tits. EGI-congress, Oxford, UK. 5-7 January 2005. (‘Runner- up recommendation’).
Presented by co-workers:
Peters A, Hall ML, Kingma SA, Delhey K. Purpler, faster, longer? Sexual signals and testosterone in purple-crowned fairy-wrens. ISBE- conference, Perth, Australia. 26 September-2 October 2010. Hall ML, Skroblin A, Kingma SA, Peters A. Moved away or passed away? Distinguishing between dispersal and death to test for benefits of delayed dispersal and philopatry in a cooperative breeder. ISBE- conference, Perth, Australia. 26 September-2 October 2010. Peters A, Kingma SA, Hall ML, Delhey K. Purpler, faster, longer? Sexual signals and testosterone in purple-crowned fairy-wrens. Presentation: EOU-conference, Zurich, Switzerland. 21-26 August 2009. Peters A, Hall ML, Kingma SA, Delhey K. Multiple sexual signals in purple- crowned fairy-wrens: conflict or cooperation? Presentation: EOU- conference, Vienna, Austria. 24-29 August 2007. Bleeker M, Kingma SA, Szentirmai, Székely T, Komdeur J. Body condition and parental behaviour in penduline tits. EGI-congress, Oxford, UK. 5- 7 January 2005.
197
Posters
Kingma SA, Hall ML, Peters A. Is acquisition of nuptial plumage an indicator of quality in male purple-crowned fairy-wrens Malurus coronatus ? EGI-congress, Oxford, UK. 9-11 August 2007. Kingma SA, Szentirmai I, Bókony V, Liker A, Bleeker M, Székely T, Komdeur J. Men in black: girls just love it. The function of melanin-based plumage pigment in promiscuous penduline tits. ECBB-congress, Groningen, The Netherlands. 28-31 August 2004. Kingma SA, Szentirmai I, Bókony V, Liker A, Bleeker M, Székely T, Komdeur J. Men in black: girls just love it. The function of melanin-based plumage pigment in promiscuous penduline tits. ISBE-conference, Jyväskylä, Finland, 10-15 July 2004.
Presented by co-workers:
Peters A, Hall ML, Kingma SA, Delhey K. The unusual purple-crowned fairy- wren, an introduction. ASSAB, Canberra, Australia. 12-15 April 2007.
198
List of publications
Kingma SA, Hall ML, Peters A (2011). Multiple benefits drive helping behaviour: an integrated analysis. American Naturalist in press Kingma SA, Hall ML, Peters A (2011). No evidence for offspring sex-ratio adjustment to social or environmental conditions in cooperatively breeding purple-crowned fairy-wrens. Behav. Ecol. Sociobiol. DOI:10.1007/s00265-010-1133-7 Kingma SA, Hall ML, Arriero E, Peters A (2010). Multiple benefits of cooperative breeding in purple-crowned fairy-wrens: a consequence of fidelity? J. Anim. Ecol. 79, 757-768. D’Alba L, Shawkey M, Korsten P, Vedder O, Kingma SA , Komdeur J, Beissinger SR (2010). Differential deposition of antimicrobial proteins in blue tit (Cyanistes caeruleus ) clutches by laying order and male attractiveness. Behav. Ecol. Sociobiol. 64, 1037-1046. Kingma SA, Hall ML, Segelbacher G, Peters A (2009). Radical loss of an extreme extra-pair mating system. BMC Ecology 9, 15. Kingma SA , Komdeur J, Vedder O, von Engelhardt N, Korsten P, Groothuis TGG (2009). Manipulation of male attractiveness induces rapid changes in avian maternal yolk androgen deposition. Behav. Ecol. 20, 172-179. Kingma SA, Szentirmai I, Bókony V, Liker A, Bleeker M, Székely T, Komdeur J (2008). Sexual selection and the function of melanin-based plumage in promiscuous penduline tits ( Remiz pendulinus ). Behav. Ecol. Sociobiol. 62, 1277-1288. Vedder O, Kingma SA, von Engelhardt N, Korsten P, Groothuis TGG, Komdeur J (2007). First report on conspecific brood parasitism in blue tits Cyanistes caeruleus . J. Avian Biol. 38, 625-629. Vahl W.K., Kingma S.A. (2007). Food divisibility and interference competition among captive ruddy turnstones, Arenaria interpres . Anim. Behav. 74, 1391-1401. Bleeker M, Kingma SA, Szentirmai I, Székely T, Komdeur J (2005). Body condition and parental behaviour in penduline tits. Behaviour 142, 1465-1468.
Manuscripts for publication
Kingma SA, Hall ML, Peters A. Female fairy-wrens synchronise egg laying to facilitate extra-pair mating for inbreeding avoidance. Kingma SA, Peters A. Timing of moult and sexual selection in fairy-wrens. Kingma SA, Delhey K, Peters A. Can sexual selection explain female ornamentation? Kingma SA, Santema P, Taborsky M, Komdeur J. Group augmentation and the evolution of cooperative breeding. Hall ML, Kingma SA, Peters A. Male songbirds advertise body size with low- pitched songs.
199