Philopatry and the Spatial Structuring of Kin in the Red-Breasted Merganser (Mergus Serrator)

Home , Herd, Philopatry, Predator satiation

PHILOPATRY AND THE SPATIAL STRUCTURING OF KIN IN THE RED-BREASTED MERGANSER (MERGUS SERRATOR)

David J. Fishman

Department of Natural Resource Sciences McGill University, Montreal September 2010

A thesis submitted to McGill University in partial fulfilment of the requirements of the degree of Masters of Science (M.Sc.)

To my mother and father who have taught, and continue to teach me

about what it is to be a good human being.

ABSTRACT

The detection of genetic-spatial structures has been facilitated by the advancement and increased accessibilities of molecular technologies. Using an empirical genetics approach, we explore aspects of the ecology of red-breasted mergansers (Mergus serrator) nesting on barrier islands off the coast of New Brunswick, Canada. Objectives were three-fold: first, to compare two sources of genetic material, i.e. blood and contour feathers, on the basis of quantity and quality of the nuclear DNA yielded; second, to screen several heterologous microsatellite primers in the M. serrator genome in order to find polymorphic loci which could subsequently be implicated in ecological analyses; third, to characterize spatial patterns of relatedness across this colony. DNA was successfully obtained from both blood and feather samples; that originating from blood was obtained at a much higher yield (µg) while the feather-derived DNA was more pure. Altogether 12 microsatellite primers successfully amplified products from the M. serrator genome. Of these, four were genotyped across 46 genetic samples and moderate levels of allelic variability were observed. Significant deviations from Hardy-Weinberg equilibrium were apparent at two loci. Finally, genetic-spatial structuring was observed within the colony over both broad and fine-scales.

KEYWORDS: waterfowl, philopatry, kin associations, microsatellites, nesting synchrony, cooperative nesting, Mergus serrator

iii

RÉSUMÉ

La détection des structures génétiques spatiales a été facilitée par l'avancement et accessibilités accrue des technologies moléculaires. En utilisant une approche empirique génétique, nous explorons les aspects de l'écologie de harles huppés (Mergus serrator) nichant sur les îles de la barrière au large des côtes du Nouveau-Brunswick, Canada. Les objectifs étaient de trois ordres: d'abord, de comparer les deux sources de matériel génétique, le sang et les plumes de contour, sur la base de la quantité et la qualité de l'ADN nucléaire, en deuxième lieu, à l'écran plusieurs amorces hétérologues microsatellites dans le génome M. serrator des loci polymorphes qui pourrait ensuite être impliqués dans les analyses écologiques, en troisième lieu, pour caractériser les patrons spatiaux de parenté dans cette colonie. L'ADN a été obtenu avec succès des deux échantillons de sang et de plumes; celle qui provient du sang a été obtenu avec un rendement beaucoup plus élevé (µg) tandis que l'ADN provenant de plumes était plus pur. Au total, 12 marqueurs microsatellites amplifiés avec succès des produits à partir du génome M. serrator. Parmi eux, quatre ont été génotypés dans 46 échantillons génétiques et des niveaux modérés de la variabilité allélique ont été observés. Des écarts significatifs de l'équilibre de Hardy-Weinberg sont apparents sur deux loci. Enfin, la structuration spatiale génétique a été observée dans la colonie plus large et fine-échelles.

MOTS-CLÉS: sauvagine, philopatrie, les associations proches, microsatellites, la synchronie de nidification, coopérative de nidification, Mergus serrator

TABLE OF CONTENTS

Page No. DEDICATION...... ii ABSTRACT...... iii RÉSUMÉ...... iv TABLE OF CONTENTS...... v LIST OF FIGURES...... vii LIST OF TABLES...... ix ACKNOWLEDGMENTS...... xi PREFACE...... xii

CHAPTER 1: INTRODUCTION 1.1) Literature Review...... 1 1.1.1) Decisions of movement...... 1 1.1.2) Philopatry and dispersal...... 3 Ultimate factors...... 4 Inbreeding avoidance...... 6 Sex-biased patterns...... 9 1.1.3) Kin associations...... 12 Cooperative breeding...... 13 Reduced aggression...... 14 Defense...... 15 Pre-hatch brood amalgamation...... 16 Post-hatch brood amalgamation...... 18 1.1.4) Kin recognition...... 19 1.1.5) Microsatellites and the detection of spatial Structuring of kin...………………………………… 22

1.2) Research Objectives...... 23 1.2.1) Study system...... 23 1.2.2) Objectives...... 24 1.3) Literature Cited...... 25

CONNECTING STATEMENT (1)...... 38

CHAPTER 2: EVALUATION OF GENETIC MATERIALS 2.1) Abstract...... 39 2.2) Introduction...... 40 2.3) Methods...... 41 2.4) Results...... 45 2.5) Discussion...... 47 2.6) Literature Cited...... 49

CONNECTING STATEMENT (2)...... 57

CHAPTER 3: SCREENING FOR HETEROLOGOUS MICROSATELLITE PRIMERS 3.1) Abstract...... 58 3.2) Introduction...... 59 3.3) Methods...... 60 3.4) Results...... 63 3.5) Discussion...... 64 3.6) Literature Cited...... 68

CONNECTING STATEMENT (3)...... 76

CHAPTER 4: SPATIAL-GENETIC STRUCTURING IN A RED- BREASTED MERGANSER (MERGUS SERRATOR) COLONY IN KOUCHIBOUGUAC NATIONAL PARK 4.1) Abstract...... 77 4.2) Introduction...... 78 4.3) Methods...... 80 4.4) Results...... 86 4.5) Discussion...... 90 4.6) Literature Cited...... 98

CONCLUDING STATEMENTS...... 113

LIST OF FIGURES

CHAPTER 2: EVALUATION OF GENETIC MATERIALS Figure 2.1 - Total amount of DNA (µg) obtained from blood and feather samples, respectively. On average, blood samples produced nearly 20X as much DNA as feather samples.……………………...... 52 Figure 2.2 – The comparison of DNA quality expressed as the 260:280 ratio. The dashed line represents the ratio expected for pure DNA. While both were within the acceptable range of absorbance at the 260:280 ratio, there were usually higher levels of contamination found in DNA extracted from blood...... ……………………………………………………………..…….53 Figure 2.3 – Two-dimensional PCoA ordination diagram. The proportion of variance explained by the first two principal coordinate axes was 0.414 and 0.225, respectfully. In this figure, the positions of the objects (i.e. nesting

vii

females) in relation to one another are approximations of their genetic distances. The PCoA was used in conjunction with Queller & Goodnight’s (1989) coefficient of relatedness………………………………….……...…54 Figure 2.4 – (a) Curve representing the positive-linear relationship between number of feathers used and total yield of DNA (ug). (b) The Q-Q normal plot of the model which indicates there were no major deviations from normality detected in the residuals…………………………………………55 Figure 2.5 – As a result of the second round of purification, significant changes were observed in both (a) DNA purity (t= -2.410, p-value= 0.030) as measured by the 260:280 ratio (b) total yield of DNA (t= 4.333, p-value= 0.004)……………………………………………………………………….56

CHAPTER 4: SPATIAL-GENETIC STRUCTURING IN A RED- BREASTED MERGANSER (MERGUS SERRATOR) COLONY IN KOUCHIBOUGUAC NATIONAL PARK Figure 4.1 – Location of the Tern Islands of Kouchibouguac National Park, New Brunswick, Canada. The ESRI shapefiles were obtained from GeoBase (http://www.geobase.ca/)……………………………………………….....104 Figure 4.2 – A schematic representation of the red-breasted mergansers nesting on the Tern Islands. The nest densities of Tern Island A (left-hand side) and Tern Island B (right-hand side) was 0.005 and 0.002 nest/m2, respectfully. Each individual nest is represented by a point. Of the 88 nests discovered, a total of 46 (52.9%) were successfully genotyped (represented by solid circles)…………………………………………….……………...105 Figure 4.3 –A rarefaction curve depicting the relationship between the number of loci used and the resultant Queller & Goodnight (1989) estimate of relatedness. Each point represents the mean difference between current estimate of relatedness and the one previous to it; the bars represent the standard deviation. A total of 9999 permutations were used to generated these data…...……………………………………………………..……….106 Figure 4.4 – Two-dimensional PCoA ordination diagram. The proportion of

viii

variance explained by the first two principal components is 0.267 and 0.173, respectively. In this figure, the positions of the objects (i.e. nesting females) in relation to one another are approximations of their genetic distances. The measure of genetic distance used in the PCoA was that described by Peakall & Smouse (1999). The results from single linkage cluster analysis are superimposed onto this ordination plot (represented by dotted-lines) as an additional means for assessing the associations between objects…………107 Figure 4.5 – The Mantel correlograms for TI-A and TI-B. Significant distance intervals are emphasized with a solid point in the diagram. A null correlation is represented by a solid red line in each panel. The corresponding figures to these panel are presented in Table 4.1. Significant values were determined by generated using 9,999 random permutations of the data and all p-values were corrected for multiple testing using Holm’s

method. A critical value of α = 0.05 was used.……………………..……108 Figure 4.6 – Bar graphs displaying: (1) mean relatedness of females and their three nearest neighbours (left-hand bar); and (2) mean background level of relatedness on that island (right-hand bar). Additionally the 95% bootstrapped CI are shown. Estimates of relatedness were obtained using the methods described by Queller and Goodnight (1989)…………………….109

LIST OF TABLES

CHAPTER 2: EVALUATION OF GENETIC MATERIALS Table 2.1 – Composition of ‘extraction mix’ used in phenol chloroform DNA extraction protocol…………………………..…………………………...…..51

CHAPTER 3: SCREENING FOR HETEROLOGOUS MICROSATELLITE PRIMERS Table 3.1 – 14 heterologous microsatellite loci tested and associated primers. For each primer, reported variability and the GenBank accession numbers are provided. The number of alleles reported is in reference to the original

species, not M. serrator………….………………….………………..….…..71 Table 3.2 – A detailed layout of the two thermocycler regimes used

throughout this study. TA = annealing temperature which is specific to each primer………………………………………………………………………..73 Table 3.3 – The ideal conditions found for each primer pair. The total volume

for each PCR was 25µl. TA = annealing temperature (ºC) which is specific to each locus………………………………………………….………………...74

Table 3.4 – The total number of alleles, observed heterozygosity (HO) and

expected heterozygosity (HE) for each locus………………………………...75

CHAPTER 4: SPATIAL-GENETIC STRUCTURING IN A RED- BREASTED MERGANSER (MERGUS SERRATOR) COLONY IN KOUCHIBOUGUAC NATIONAL PARK Table 4.1 – The distance intervals, number of pairwise observations within a given interval, Mantel correlation coefficients (r) and associated p-values for the genetic and geographic distances being compared. The probabilities were generated based on 9,999 random permutations of the data and all p-values were corrected for multiple testing using Holm’s method. A critical value of α = 0.05 was required before rejecting the null hypothesis…………………..110 Table 4.2 - The results from the Kruskal-Wallis rank sum test where the null hypothesis tested was that on a given island the mean relatedness of females and their three nearest neighbours (NN) was no different than the mean background level of relatedness. On Tern Island A, the null hypothesis was rejected however not on Tern Island B. The critical value used was α = 0.05…………………………………………………………………….…..111 Table 4.3 – Mean dates of colonization and incubation-initiation across the Tern Islands. The days were transformed into Julian days using May 1 as day 1. In both tests the null hypothesis, stating that the variable did not differ between islands, could not be rejected. A critical value of α = 0.05 was used.…………………………………………………………………...…....112

ACKNOWLEDGEMENTS

I would like to express my gratitude to those who provided funding to this project and toward my degree, including: le Fonds Québécois de la Recherche sur la Nature et les Technologies (FQRNT); Schulich Graduate Fellowship, McGill University; D. Zadworny; New Brunswick Wildlife Trust Fund (NBWTF); and R. D. Titman.

I am indebted to both R. D. Titman and S. R. Craik for having inspired me with me their enthusiasm and knowledge and for having shared with me their general expertise on the subject. Furthermore, without their contributions to the field component of this study, very simply, none of this would have been possible. Also, I would like to extend my thanks to others who provided assistance in the field, including E. Titman, S. Titman and J. Ding.

I want to acknowledge D. Zadworny’s efforts, even before the official start of this degree, in having provided me with the opportunity to work in his laboratory and develop the set of skills I needed in order to pursue this endeavor. His advice, feedback and encouragement throughout this project have been invaluable. Without J. M Pearce’s guidance, moral support, and willingness to share results from his previous work, this project would have also not been possible. Thanks to both D. Praslickova and G. Hiyama for their direction in the laboratory, patience and camaraderie. Throughout this project they have been my laboratory gurus. I also would like to also thank M. Forrest, R. McQuaid and J. Ding for their assistance and company in the lab along the way. Finally, I am much appreciative to I. Ritchie and the Avian Science and Conservation Centre (ASCC) for having provided me with genetic specimens.

A very special thanks to my family and loved ones, who stood by and supported me through thick and thin. Without my wonderful fiancé’s love, friendship and brain-storming sessions this would have been a much more difficult experience. I am grateful to my parents for their loving-care,

xi encouragement and last-minute suppers while writing this thesis. Also, thanks to my mother for her help with editing. Finally I would like to express my appreciation and gratitude to faculty members, colleagues and fellow Birdcagers who have provided me with valuable feedback at various stages of this work. Last but not least, I would like to thank Gaia and Rigby for having inspired many curiosities in me and for having been the best teachers of ornithology I have ever had.

PREFACE

This thesis is manuscript-based and is composed of four chapters. The overall subject matter pertains to a colony of female red-breasted mergansers (Mergus serrator) nesting in New Brunswick, Canada. Chapter 1 consists of a literature review surrounding philopatry, kin association and kin recognition. Chapters 2 and 3 report the results of preliminary work where (a) the quantity and quality of M. serrator DNA, originating from blood and contour feathers samples, was assessed and (b) numerous heterologous microsatellite primers were screened in the M. serrator genome in order to find polymorphic loci, respectfully. In Chapter 4, spatial patterns of genetic-relatedness are analyzed across this colony at various scales. The candidate (senior author) wrote of all four chapters. R. D. Titman and D. Zadworny (second authors) provided conceptual input and editorials for Chapters 2, 3 and 4; while S. R. Craik provided conceptual input and editorials only in Chapter 4.

xii

CHAPTER 1

1.1) LITERATURE REVIEW

1.1.1) Introduction

Decision optimization has been considered in the context of patch exploitation (MacArthur & Pianka 1966, Charnov 1976), food preference (Emlen

1966), offspring sex allocation (Fisher 1930, Hamilton 1967) and placement

(Fretwell & Lucas 1969, Hamilton 1971). Patterns of movement throughout an animal’s life likely result in differential gains of fitness (Swingland & Greenwood

1983). One of the most significant consequences of these decisions influences who an individual is surrounded by and what resources it will have access to. For example, the migratory tendencies of animals are understood to have evolved as a means to increase access to certain resources (Alerstam 1990). Migration occurs in animals because those who do it are more successful in propagating their genes than those who do not. At a smaller scale, Hamilton (1971) described how movement of individuals toward the nucleus of a herd reduces their chance of predation at the expense of their group mates (hence the term ‘selfish herd’).

Some types of movement result in groups of relatives being closer together in space. It is these types of movements that are the focus of this review.

In plants, the positive relationship between proximity and genetic similarity (i.e. genetic spatial-autocorrelation) is commonly observed and is considered to be a result of their sedentary nature (Epperson & Allard 1989,

Skabo et al. 1998, He et al. 2000). Despite the motility of animals, clustering of

1 kin is widespread across the animal kingdom (Burgman & Williams 1995,

McFadden & Aydin 1996, Girman et al. 1997, Taylor et al. 1997, Coltman et al.

2003, Uesugi et al. 2009). The grouping of family members in certain contexts is not at all surprising. Most fundamentally, such a signal arises with a mother giving birth or a cohort of siblings hatching synchronously. However, the spatial- genetic associations of adult kin are usually the outcome of more complex, behavioural elements. There are two primary factors responsible for such associations. First, if at least some individuals exhibit fidelity toward their places of origin, the background level of relatedness in that region increases as a result

(Friesen et al. 1996, Girman et al. 1997, Hoglund & Shorey 2003, Double et al.

2005). Second, a similar effect arises if close relatives maintain associations with one another. These two phenomena are not mutually exclusive and it is generally regarded that the former is often a precursor to the emergence of active kin associations.

In this review, I explore how certain decisions made by animals influence their settling patterns; specifically how they can result in clusters of closely- related individuals. The two broad mechanisms I focus on are (a) philopatric and dispersal tendencies and (b) actively maintained kin associations. Also integral to this discussion, is an attempt to understand the rules and mechanisms used by kin in order to recognize each other. Finally, I briefly address how settling decisions made by animals manifest themselves genetically and can be detected using molecular markers.

1.1.2) Philopatry and dispersal

The term ‘natal dispersal’ was first given by Howard (1960) who defined it as the tendency for an animal to move away from its place of origin in order to breed. Unlike other forms of movement (e.g. migration), this specifically refers to the distance to which the animal’s genetic materials are transmitted (Howard

1960). A further distinction is between natal and breeding dispersal; the latter precluding place of origin and concerning itself only with movements between subsequent breeding events. Philopatry refers to a tendency for limited dispersal so that animals breed at or close to their places of origin (Greenwood 1980).

Because of the antithetical nature of philopatry and dispersal it is often redundant to treat them as separate entities; indeed the costs associated with one strategy can be viewed as the benefits of the other. Pearce (2007) argues that the term philopatry should not be extended to ‘breeding philopatry’ because only the former has consequences from a genetic structuring standpoint. Philopatric tendencies are often sufficient to limit gene flow between populations that are otherwise geographically contiguous (e.g. Oota et al. 2001 Coltman et al. 2003,

Temple et al. 2006, Dionne et al. 2008, Sonsthagen et al. 2009). Following

Pearce’s recommendation, the use of ‘philopatry’ in this review refers explicitly to ‘natal philopatry’ while the ‘site fidelity’ replaces ‘breeding philopatry.’

Philopatric movements lead to spatial clustering of kin because, as inherent to the definition, individuals and possibly their siblings are attracted to the place where their parents had bred.

Ultimate factors

The experience an adult has had growing up (e.g. levels of resource abundance, predation and parasitism) can provide valuable information about the quality of its natal habitat and can be subsequently used to make an optimal dispersal decision (Brown & Brown 1992, Osorio-beristain & Drummond 1993,

Schjorring 2001). Assuming that habitat quality persists beyond one generation, philopatry can be used as a proximate mechanism geared towards making adaptive settling decisions (Schjorring et al. 2000, Schjorring 2001). Findings by

Schjorring et al. (2000) in a colony of great cormorant (Phalacrocorax carbo) support this notion, as the philopatry of these seabirds shows spatial- autocorrelation with nesting success over successive breeding seasons. Similarly, whether a cliff swallow (Petrochelidon pyrrhonota) is philopatric is contingent on the ectoparasite load it had endured as a nestling (Brown & Brown 1992). In these systems, the degree of site fidelity in adults appears to be largely contingent upon whether previous breeding experience at a given site is successful (Brown &

Brown 1992, Schjorring et al. 2000, Semel & Sherman 2001).

In general, familiarity with a particular site has advantages (Greenwood

1980). Familiarity can also come about through exploration and prospecting during a juvenescent period (e.g. Spear et al. 1998, Schjorring 2001). Benefits include: (a) more effective predator evasion; (b) superior abilities in procuring resources; and (c) reduced hostility or perhaps even cooperation with nearby conspecifics (Rathbun 1979, Beletsky & Orians 1991, Sonsthagen et al. 2010).

Reduced aggression between neighbours also permits the parties involved to

4 invest more energy into acquiring food and mates, therefore potentially leading to an increase in reproductive success (Beletsky & Orians 1989). Endemism is another advantage associated with philopatry and can be thought of as ‘genetic familiarity’. Individuals endemic to a region are likely to benefit from gene complexes which are adapted to local conditions (Wright 1943). Finally, being surrounded by relatives (e.g. parents, siblings) can facilitate the emergence of cooperative systems as a result of kin selection. In cooperative interactions, all individuals involved benefit both directly (sensu Clutton-Brock 2002) and indirectly (sensu Hamilton 1963) (see Chapter 1.1.3).

Conversely, there are several advantages obtained by dispersing from natal habitat. Perhaps most important is the avoidance of inbreeding (see following section). Because inbreeding depression can have detrimental consequences, many animals actively avoid mating with close relatives (e.g. Pusey 1980, Grau

1982) and under such circumstances finding a mate when surrounded by family is difficult (Hoogland 1982). Through dispersal, an organism might increase its probability of finding a suitable mate and may also heighten attractiveness to the opposite sex (Lendrum 1985). Similarly, where territory availability is not a limiting factor, an individual can increase its access to various resources by dispersal (Greenwood 1980).

In addition to the costs and benefits of both strategies, ecological factors bear consideration. In patchy environments where predation rates are high and the chances of finding mates are low, the risks associated with dispersal may outweigh any benefits gained from the avoidance of inbreeding (Facemire &

Batzli 1983). Ecological constraints as such are thought to be responsible for the emergence of families in nature. Families occur when dispersal and reproduction, in individuals who are otherwise capable of doing so, are delayed (Emlen 1994).

From the viewpoint of Emlen’s Delayed Dispersal Model (1994), the decision to remain with family members suggests that there are net benefits to staying. An important factor in this cost-benefit analysis is the potential to gain inclusive fitness by raising siblings and increasing the chance of inheriting either the territory they are tending or one adjacent to it (Emlen 1994, MacColl et al. 2000).

Cooperative breeding by families is associated with environments where there is a paucity of suitable breeding sites and the chances of successfully rearing young without help are low (Woolfenden & Fitzpatrick 1978).

Inbreeding avoidance

There are many advantages to remaining in one’s natal territory (see above) yet if both male and female siblings within a cohort employ this strategy, the opportunities for inbreeding are greatly increased. Direct evidence of inbreeding depression is mostly anecdotal (e.g. Hoogland 1982), however descriptions of behaviours for the avoidance of inbreeding are prevalent in the literature (Batzli et al. 1977, Grau 1982, Hoogland 1982). It is presumed that inbreeding is costly because of the increased probability that deleterious recessive genes are expressed in resultant offspring. Inbreeding can either be avoided innately, behaviourally or using a combination of both.

Innate avoidance is usually pheromone-mediated (Grau 1982, Lendrem

1985). In some rodents, reproductive maturity in females is either inhibited or accelerated by the pheromones of closely-related males and strange males, respectively (Batzli et al. 1977, Lendrem 1985). In the case of certain microtine voles (Microtus spp.) in addition to puberty, female growth is also suppressed in the presence of close relatives (Batzli et al. 1977). Similar innate mechanisms appear to exist in black-tailed prairie dogs (Cynomys ludovicianus), however they are further reinforced by behavioural traits (Hoogland 1982). Three behaviours have been described to reduce the probability of mating with relatives: (1) sub- adult (i.e. sexually immature) males disperse; (2) adult males leave their coteries within two years, coinciding with the time of their daughters’ puberty; and (3) sexually mature females exhibit avoidance towards brothers, fathers, and sons

(Hoogland 1982, Dobson et al. 1997).

In many animals, dispersal is an important behavioural tactic for the avoidance of inbreeding (Taylor et al. 1997, Oota et al. 2001, Hansson et al. 2003,

Woxvold et al. 2006); however there are clear exceptions (Osorio-beristain &

Drummond 1993, Burland et al. 2001). For example, in densely-populated nesting colonies, even if both sexes are philopatric, the probability of accidentally mating with a sibling is negligible (Osorio-beristain & Drummond 1993). Alternately, where close relatives are sympatric, Blaustein & Waldman (1992) suggest that individuals who can accurately recognize kin (see Chapter 1.1.4) may be able to achieve optimal outbreeding through differential avoidance of conspecifics. The latter is exemplified in the complex social dynamics of Chimpanzees in Gombe

National Park. Pusey (1980) reports that pre-oestrus females are often associated with a ‘favourite male’ who is also usually a close relative. The amount of time spent with favourite males drastically diminishes once oestrus begins. Eventually pubescent females leave the colony in order to find suitable mates and while a proportion of these females never return, there are some who do but only once they are pregnant (Pusey 1980). Similarly, both male and female brown long- eared bats (Plecotus auritus) co-inhabit their natal colonies, however, males appear to mate exclusively with females from foreign colonies. Both of the above examples demonstrate how behavioural adaptations have allowed these animals to gain access to the benefits associated with philopatry while still avoiding inbreeding depression.

Some authors have framed inbreeding depression as only one extreme along a continuum. It is suggested that mating with individuals that are too genetically-distant will also result in poor reproductive success (Shields 1983).

While the idea of optimal inbreeding is not well-substantiated a provocative example is provided by Bateson (1982) where an attempt to optimize the trade-off between inbreeding and outbreeding depression has been recorded in Japanese quail (Coturnix japonica) through a well-defined preference of both sexes towards mating with first cousins. Furthermore, the prospect of inbreeding depression is relevant only to diploid organisms. In haplodiploid animals, mating with siblings is usually part of an adaptive mating strategy (Hamilton 1967); this can happen because deleterious recessive genes are inevitably purged from the gene pool in haploid males (Antolin 1999).

Sex-biased patterns

Where inbreeding depression does occur, the costs are comparably grave to both males and females involved. While dispersal of one sex may be a direct or indirect outcome of inbreeding avoidance, there is still no indication of which sex should remain. Greenwood (1980) proposed that the mating system in place is the key determinant and focused primarily on mate-defense and resource-defense mating systems.

In mate-defense systems, common to most mammals, females typically hold as the stable nucleus (i.e. are the philopatric sex) (Hoogland 1982, Girman et al. 1997, Coltman et al. 2003). Assuming inexperienced males are competitively inferior to older ones, remaining within a natal territory leaves young males with highly restricted access to potential mates, if any at all (Greenwood 1980,

Munshi-South 2008). Moreover, in polygynous mating systems males contribute very little parental investment. Consequently, not being in a familiar environment is likely to be more detrimental to females than to males (Greenwood 1980).

Conversely in resource-defense systems, which are common to most song birds, dispersing is likely to be more costly to males than to females. Inhabiting foreign environments potentially diminishes a male’s ability to acquire a territory and compete for resources (Greenwood 1980). Furthermore, if other related males are nearby, nepotism can play an important role in recruitment (Piertney et al.

1998, MacColl et al. 2000). In contrast, a female’s proximity to her natal habitat does not affect her ability to acquire a quality nesting territory; this is due to

9 female preference for breeding site characteristics in territories which males compete for (Greenwood 1980).

Dispersal patterns in waterfowl diverge from those of most birds (i.e. male-biased dispersal) and there are several important distinctions to make between the two groups. First, mating tactics in the Anatidae resemble mate- defense more so than resource-defense strategies as males expend their efforts restricting access to a female as opposed to a particular resource associated with female use (Greenwood 1980). Second, the precociality of young in waterfowl has reduced parental investment for males (Greenwood 1980). Consequently, in addition to energy expenditure as a result of laying, incubation and brood rearing, females are more vulnerable to predation as a direct consequence of these activities (Cooke et al. 2000). Therefore similar explanations for the female- biased philopatry observed in mammals are applicable here; namely that females invest more in the rearing of young than males and therefore have more to gain from familiarity with the habitat (Greenwood 1980, Wolff & Plissner 1998).

In the patterns outlined by Greenwood (1980) the mating systems of many animals cannot be cleanly assigned to either of the above categories (Clarke et al.

1997). For instance, female-biased dispersal is reported in large-tree shrews

(Tupaia tana) which are behaviourally monogamous (i.e. males and females hold joined territories) yet biologically promiscuous (i.e. a high frequency of extra-pair copulation) (Munshi-South 2008). Similarly, male-biased philopatry is suggested to occur in several species of lekking birds (MacColl et al. 2000, Shorey et al.

2000). Wolff & Plissner (1998) proposed a simplification to Greenwood’s (1980),

10 scheme suggesting that the philopatric sex is likely to be the one which selects the nest site (however not necessarily the mating territory). This rule seems to account for differences associated with different mating systems and reconciles some of the aberrant observations. Such a classification is dependent on in-depth knowledge of an animal’s life history.

It is important to recognize that in reality the net movements of animals are not going to be an optimal solution to the costs and benefits listed above.

Other important factors (e.g. resource limitation, predation, and parasitism) must be incorporated into the decision making process (Brown & Brown 1992,

Schjorring 2001). Even though a group of individuals may have strong philopatric tendencies, relatively few are expected to actually return/remain (Greenwood

1980). Even if the signal is weak, one expects to find higher levels of relatedness between neighbouring members of the sedentary sex and lower in the dispersing sex (Greenwood 1980). Because of this, one can generalize that the emergence of cooperative traits is thus favoured in the former while heightened aggressiveness can be expected to occur between members of the latter (Dunford 1977). It is in this context that philopatry is often considered to be an important precursor to sociality (Greenwood 1980, Stoen et al. 2005). Several studies reported primitive forms of cooperation even among philopatric females in solitary mammals

(Paetkau et al. 1995, Ratnayeke et al. 2002, Stoen et al. 2005, McEachern et al.

2007). Observing interactions between philopatric individuals in these systems can help our understanding of how sociality can emerge (Stoen et al. 2005).

1.1.3) Kin association

Preferential associations with kin occur within many different animals

(Waldman 1982, Taylor et al. 1997, McKinnon et al. 2006, Nituch et al. 2008).

Like philopatry, these associations also result in the spatial autocorrelation of related individuals; here however, such clustering is not limited to the natal territory. Kin associations facilitate the emergence of altruistic behaviours because altruism sensu stricto should be limited toward kin (Hamilton 1963); unless there is a chance for reciprocation (Trivers 1971). In its most extreme form

(i.e. foregoing reproductive opportunities), altruism can only be sustained in kin groups (Reyer 1984, McAllister & Roitberg 1987). Cooperation on the other hand, especially when there are minimal costs associated with participation, provides generalized benefits to all parties involved, and thus does not specifically warrant the formation of kin groups. Gregariousness in animals has many potential benefits including: (1) predator evasion (Hamilton 1971, Ost et al. 2005); (2) early detection of predators (Davis 1982, Evans 1984); and (3) increased feeding efficiency (Ward and Zahavi 1973, Aviles et al. 2004,). By directing cooperative behaviour towards kin, the benefits mentioned above may be enhanced due to kin selection (O’hara & Blaustein 1981). For example, any individual assembling in a dense group benefits from a reduced risk of predation (Hamilton 1971). If, however, groups are composed of closely-related individuals, those succumbing to predation are at least compensated by gaining inclusive fitness (Waldman

1982). It is perhaps because of such indirect gains of fitness that Dorset sheep

(Ovis aries) (Nituch et al. 2008) and cascade frog tadpoles (Rana cascadae)

(O’hara & Blaustein 1981) exhibit a strong preference to herd with relatives despite there being no evident cause for doing so.

Kin Association and Breeding Ecology

Ultimately, kin associations that are observed in nature exist because individuals engaged in such behaviour are on average obtaining higher fitness gains than those who do not. The formation of kin associations are explored below in the following contexts: (1) cooperative breeding; (2) reduced aggression;

(3) defense; (4) pre-hatch brood amalgamation; and (5) post-hatch brood amalgamation.

Cooperative breeding

The decision of a mature individual, who has decided to remain in its natal territory to help raise siblings, can be subdivided into two components: (1) to delay dispersal; and (2) to help (Hatchwell & Komdeur 2000). Constraints associated with breeding without delay include: (1) paucity of adequate breeding territories; (2) paucity of mates; (3) high risk of mortality associated with dispersal and; (4) high risk of nest failure if breeding independently (Hatchwell &

Komdeur 2000, see Chapter 1.1.2 for details). Helping, on the other hand, is a strategy that may be based on the availability of nearby kin (Russell & Hatchwell

2001). In fact, a high degree of helper-young relatedness has been recorded in many cooperative breeding systems (Woolfenden 1975, Reyer 1984, Russell &

Hatchwell 2001). While helpers do not gain as much fitness as they do by

13 breeding themselves, they do gain more than non-breeders, and non-helpers (i.e. floaters) (Reyer 1984, Emlen 1994).

As in many forms of cooperation, benefits unrelated to kin selection can also be obtained. Accordingly, in cooperative breeding situations, high helper- young relatedness is not always the case (Zahavi 1990, Dunn et al. 1995, Clutton-

Brock et al. 2000, 2001). Helpers stand to benefit from gaining valuable breeding experience and from inheriting a territory (Woolfenden 1975, Woolfenden &

Fitzpatrick 1978). An interesting example was recorded in pied kingfishers

(Ceryle rudis) where mated pairs generally have primary and secondary helpers assisting with brood rearing; the two being distinguished on the basis of level of access to the nest (Reyer 1984). While primary helpers were closely related to the breeding pair, secondary helpers were not; however in almost all cases, secondary helpers became heirs to that territory.

Reduced aggression

An important potential benefit obtained from grouping together with kin, is the reduction of aggressive and competitive behaviours. This may occur because tolerance to defection by one of the parties involved is lowered due to kin selection. This is demonstrated in certain plants where fewer resources were allocated toward the growth of competitive traits (e.g. stem elongation) in the presence of other closely-related plants (Murphy & Dudley 2009). Similarly,

Sonsthagen et al. (2010) suggest that kin associations within common eider

(Somateria mollissima) colonies serve to minimize the aggressive interactions between neighbours under high nesting densities.

Reduced aggression and competition between kin may also lead to increased recruitment rates (Lambin & Krebs 1993) and this is apparently common in many lekking species (Kokko & Lindstrom 1996, Petrie et al. 1999,

MacColl et al. 2000, Hoglund & Shorey 2003). However it should be noted that the benefits to lekking within kin extend beyond reduced agonism. Two important aspects of lek mating systems must be considered: (1) larger leks are more effective in attracting females; and (2) male reproductive success is highly skewed. The incentive of subordinate males to join leks is enhanced by high levels of intra-lek relatedness as the skewed mating success of the superior male at least results in the indirect acquisition of fitness for subordinates (Kokko &

Lindstrom 1996).

Defense

Kin associations play a crucial role in the emergence of group-coordinated defense systems, whether in the form of alerting others to the presence of a predator or active mobbing (Sherman 1977, McGowan & Woolfenden 1989,

Wright et al. 2001). While cooperation always has associated advantages for those involved, in some situations defection may be even more beneficial. If those who have spotted a predator do not advertize this information and allow their neighbours to be eaten, they stand to increase their access to resources by reduced competition and minimize risk of predation due to predator satiation (Dunford

1977). The importance of kin selection in such cooperative systems has been eloquently demonstrated by Dunford (1977) who showed that closely-related philopatric female round-tailed ground squirrels (Spermophilus tereticaudus) emitted alarm calls significantly more often than the non-related, dispersing males.

Furthermore, this pattern did not exist among juveniles who were all close relatives where both sexes called equally as frequent.

In communal breeding situations, the defensive behaviours of one individual can indirectly benefit others nesting nearby. Schmutz et al. (1983) compared rates of predation on artificial eggs placed in close vicinity to nesting female eiders (S. mollissima) to those placed farther away and found the latter were subjected to significantly higher rates of predation. Therefore, by forming kin associations in these situations, females can direct these benefits towards kin

(McKinnon et al. 2006, Sonsthagen et al. 2010).

Pre-hatch brood amalgamation

Pre-hatch brood amalgamation (pre-HBA) is common and well- documented in birds (Yom-Tov 1980, Eadie et al. 1988) and more recently in insects (see Tallamy 2005 for review). While in many cases, pre-HBA involves a parasitic relationship between the donor and recipient (Eriksson & Andersson

1982, Marvil & Cruz 1989) other instances suggest that these interactions are more accurately categorized as cooperative breeding facilitated by kin selection

(McRae & Burke 1996, Andersson & Alhund 2000). The relatively high prevalence of cooperative pre-HBA in waterfowl is assumed to be a consequence

16 of female philopatry which results in greater opportunities for females to lay in the nests of relatives.

Egg dumping is a beneficial strategy from the donor’s perspective, and may increase its direct fitness in any of the following ways: donors (1) are free to pursue other reproductive opportunities (Loeb et al. 2000); (2) can spread the risk of predation (i.e. bet-hedging) (Eadie et al. 1988); and (3) can avoid incurring the costs associated with nest construction, incubation and brood-rearing (Milonoff et al. 2004). In contrast, receiving eggs from the viewpoint of recipients may either be beneficial, neutral or deleterious (Eadie et al. 1988). The primary benefit of receiving eggs is the dilution / manipulation of egg predation risks (Eadie et al.

1988). For example, foreign eggs laid within the nests of lace bugs, Gargaphia solani (Heteroptera: Tingidae) are usually along the periphery and are therefore the most likely to be predated (Loeb et al. 2000). Conversely, the negative consequences that may arise from receiving eggs include: (a) increased offspring mortality (Linden & Moller 1989); and (b) reduced life-time fecundity of the recipient (Cooke et al. 2000, Milonoff et al. 2004).

While pre-HBA may constitute an important alternate reproductive strategy for some individuals (e.g. Milonoff et al. 2004), the degree of relatedness between females can have an influence on patterns of laying. For example, several studies have reported a higher-than-expected level of relatedness between donors and recipients (Andersson & Alhund 2000, Andersson & Waldeck 2007, Waldeck et al. 2008) while others have found that kin actively avoid laying in each other’s nests (Semel & Sherman 2001, Dickinson 2007). Several predictions about pre-

HBA can be made on the basis of kinship. If it is beneficial to receive eggs, while recipients are open to receiving any eggs regardless of their origins; kin selection should favour donors who target the nests of relatives (Loeb et al. 2000, Lopez-

Sepulcre & Kokko 2002). If on the other hand being a recipient is detrimental, parasitizing relatives is adaptive only if the direct fitness gained by the donor is high enough to compensate for the losses sustained by the host. This calls for relatively high degrees of host-parasite relatedness; otherwise parasitizing kin should be avoided (Andersson 2001). In either of the above cases, provided there are consistent opportunities to parasitize kin, developing the ability to recognize and discriminate between kin is adaptive (Andersson 2001, Lopez-Sepulcre &

Kokko 2002, see Chapter 1.1.4).

Post-hatch brood amalgamation

Post-hatch brood amalgamation (post-HBA) occurs when at least two separate broods pool together (Munro & Bedard 1977, Eadie et al. 1988). While it is a reasonable to assume that non-parental females tending young may be kin, aside from a few accounts (e.g. Bukacinski et al. 2000, Kraaijeveld 2005) this has not been well-substantiated (Ost et al. 2005).

Post-HBA in some systems is regarded as an adaptive strategy for coping with predation and extreme weather (Davis 1982, Evans 1984). In other situations, it has been suggested to be accidental (Gorman & Milne 1972, Savard 1987).

Factors which may influence a mother’s decision to desert her brood to the care of another female include: (1) brood size (Carlisle 1985); (2) proportion of brood

18 failure (Poysa 1995); and (3) body condition (Gorman & Milne 1971, 1972, Ost

1999, Kilpi et al. 2001, Ost et al. 2005). In long-lived, iteroparous species, withdrawing parental investment during unfavourable circumstances and salvaging resources for a future reproductive opportunity may be a more adaptive strategy (Eadie et al. 1988). Carlisle (1985) demonstrated this principle in cichlid fish (Aequidens coeruleopunctatus) by experimentally altering the size of a female’s brood; the more fry added the higher the tolerance threshold of risk a female would endure before abandoning.

1.1.4) Kin recognition

Kin recognition serves as the basis for: (1) the direction of altruistic behaviour toward kin (Beecher et al. 1986, Blaustein & Waldman 1992); (2) the minimization of competitive and aggressive interactions between kin (Murphy &

Dudley 2009, Sonsthagen et al. 2010); and (3) the selection of an optimally related mate (Bateson 1982, Wedekind et al. 1995, see Chapter 1.1.2).

Recognition can only be measured subjectively and indirectly. In contrast, kin discrimination is readily apparent as the exhibition of a differential behavioural response to kin vs. non-kin (Waldman 1988). The relationship between the two is hierarchical as recognition is a prerequisite for discrimination, however, the reverse is not true. Waldman (1988) articulates two general categories of recognition: direct and indirect. The difference between the two pertains to whether kin are identified on the basis of: (1) chemical, visual or auditory cues

(direct); or (2) the context in which they are encountered (indirect). The strategy

19 chosen by an animal reflects the consistency and predictability of distribution of kin over space and time (Dunford 1977, Waldman 1988). Direct recognition capabilities are important in environments where there is equal opportunity to interact with both kin and non-kin (Eickwort 1973, Frumhoff & Schneider 1987).

For instance, honey bees, Apis mellifera (Hymenoptera: Apididae) are polyandrous and therefore relatedness between the sisters of a colony is not homogenous (Frumhoff & Schneider 1987). Frumhoff & Schneider conclude that the kin recognition capabilities of worker bees enable them to direct feeding

(trophallaxis) and grooming behaviours towards their full sisters. Conversely, indirect recognition is likely to occur only where the probability of encountering non-kin, is rare (Waldman 1982, McAllister et al. 1987). Aposematism in poisonous or noxious animals, as described by Fisher (1930), is of little protection if facing a naïve predator. Implicit in the risks and metabolic costs linked with these behaviours, is the assumption that group members are of close kin and are the immediate beneficiaries of the negative associations formed by predators

(Fisher 1930, Waldman 1982). Similarly, the indiscriminant and martyr-like behaviour of pea aphids, Acyrthosiphon pisum (Hempitera: Aphididae) who commit suicide when parasitized by braconid wasps, operate under the assumption that a large proportion of colony members are kin; an assumption that tends to be accurate given the viviparous parthenogenic mode of reproduction in aphids (McAllister et al. 1987).

Kinship determination can be approximated through self referencing

(Wedekind et al. 1995, Petrie et al. 1999). For instance, mate selection in humans

(Homo sapiens sapiens) is correlated with differences in the major histocompatibility complex (MHC) (Wedekind et al. 1995) and like many other animals comparisons appear to be mediated through pheromones (Batzli et al.

1977, O’hara & Blaustein 1981, Boch & Morse 1982, Grau 1982). Recognition of this type enables individuals to identify kin without ever having been previously exposed to them (Sheppard & Yoshida 1971, Bateson 1982). Alternately, recognition can result from imprinting which has occurred during a particular stage of an animal’s life (Bateson 1979). The timing of the ‘sensitive period’ depends on various aspects of an animal’s life history. Specifically, imprinting is likely to occur during periods when individuals are: (a) exposed to environmental cues common to kin (e.g. egg mass, hive or uterus) or (b) directly surrounded by kin (e.g. parents, siblings, etc.). A clear demonstration for the potential of animals to imprint on environmental cues is provided by Leon (1987) who demonstrated in rats that associations made with olfactory stimuli can form in utero and persist into adulthood. At the other end of the spectrum, imprinting directly on close-by individuals (e.g. in waterfowl) has been well-documented (Lorenz 1970,

Andersson & Alhund 2000, van der Jeugd et al. 2002).

Recognition mediated by imprinting, is one of several strategies used to maximize the chance of identifying kin. In such as strategy, the rule could be something like: imprint on this cue or on this individual, during this time period

(Bateson 1979). Other rules can be followed to identify kin. For example, while herring gull (Larus argentatus) adults are incapable of direct discrimination between their young and those of others, they tend any chick that solicits them

21 properly and on-cue (Knudsen & Evans 1986). Under these circumstances, indirect recognition is less risky for parents as it abolishes the possibility of falsely rejecting their own offspring; a mistake that could be lethal to them.

Chicks on the other hand are kept in line and discouraged from soliciting adults indiscriminately, as doing so could inadvertently advertize their positions to foreign predatory adults (Knudsen & Evans 1986).

1.1.5) Microsatellites and the detection of spatial structuring of kin

The use of molecular markers in ecology has greatly facilitated the ability to detect the spatial structuring of kin. Traditional approaches involve elaborate marking schemes and observation and the ability to infer relatedness is constrained by logistical and ecological factors such as extra-pair copulations and pre-hatch brood amalgamation (see Elder & Elder 1949, Semel & Sherman 2001).

Microsatellites offer several unique advantages over other molecular markers and have consequently been embraced as the ideal tool for many molecular ecological studies (Schribner & Pearce 2000). Because of their biparental mode of inheritance, co-dominant structure and selective neutrality (in the majority of cases), they are well-suited for the determination of pairwise relatedness and the reconstruction of pedigrees (Queller et al. 1993). In conjunction with the appropriate statistical tools, microsatellites have been highly successful at revealing various patterns of kin structuring (or lack thereof) within a population (Girman et al. 1997, Peakall et al. 2003, McKinnon et al. 2006,

Woxvold et al. 2006, Alcaide et al. 2009). Furthermore, their utility and

22 effectiveness has been confirmed in studies where conclusions drawn had been cross-referenced using field observations and banding data from the same system

(Double et al. 2005, Nielsen et al. 2006).

1.2) RESEARCH OBJECTIVES

1.2.1) Study system

This study concerns female red-breasted mergansers (Mergus serrator) nesting on a series of barrier islands in Kouchibouguac National Park, New

Brunswick, Canada. Red-breasted mergansers are piscivorous sea ducks that breed in fresh, brackish and saltwater wetlands across a Holarctic distribution

(Titman 1999). Nest bowls are scraped on the ground and lined using feathers and live and dead vegetation (Bent 1962). While most of what is known about the red- breasted merganser relates to its distribution and foraging behaviour (Sjoberg

1985, Rupert & Brush 1996, Gregory et al. 1997, Bur et al. 2008, Pearce et al.

2009), several aspects of its ecology and breeding biology have been studied in this particular system (Young & Titman 1988, Craik & Titman 2008, Craik &

Titman 2009, Craik et al. 2009).

1.2.2) Objectives

The aims of my research are: 1) (Chapter 2), to compare yields of DNA from blood samples versus contour feathers collected from nests; 2) (Chapter 3), to screen the red-breasted merganser genome for polymorphic microsatellite loci using heterologous polymerase chain reaction (PCR) primers. These primers can subsequently be applied to address the final objective of this thesis; 3) (Chapter 4), to determine the extent of spatial-genetic structuring in the colony of red-breasted mergansers nesting on the Tern Islands of Kouchibouguac National Park.

1.3) LITERATURE CITED

Alcaide M, Serrano D, Tella JL, Negro JJ (2009) Strong philopatry derived from capture-recapture records does not lead to fine-scale genetic differentiation in lesser kestrels. Journal of Animal Ecology 78, 468-475. Alerstam T (1990) Bird migration Cambridge University Press, New York. Andersson M (2001) Relatedness and the evolution of conspecific brood parasitism. American Naturalist 158, 599-614. Andersson M, Ahlund M (2000) Host-parasite relatedness shown by protein fingerprinting in a brood parasitic bird. Proceedings of the National Academy of Sciences of the United States of America 97, 13188-13193. Andersson M, Waldeck P (2007) Host-parasite kinship in a female-philopatric bird population: evidence from relatedness trend analysis. Molecular Ecology 16, 2797-2806. Antolin MF (1999) A genetic perspective on mating systems and sex ratios of parasitoid wasps. Researches on Population Ecology 41, 29-37. Aviles L, Fletcher JA, Cutter AD (2004) The kin composition of social groups: trading group size for degree of altruism. American Naturalist 164, 132- 144. Bateson P (1979) How do sensitive periods arise and what are they for. Animal Behaviour 27, 470-486. Bateson P (1982) Preferences for cousins in Japanese Quail. Nature 295, 236-237. Batzli GO, Getz LL, Hurley SS (1977) Suppression of growth and reproduction of microtine rodents by social factors. Journal of Mammalogy 58, 583-591. Beecher MD, Medvin MB, Stoddard PK, Loesche P (1986) Acoustic adaptations for parent-offspring recognition in swallows. Experimental Biology 45, 179-193. Beletsky LD, Orians GH (1989) Familiar neighbors enhance breeding success in birds. Proceedings of the National Academy of Sciences of the United States of America 86, 7933-7936. Beletsky LD, Orians GH (1991) Effects of breeding experience and familiarity on site fidelity in female red-winged blackbirds. Ecology 72, 787-796.

Bent AC (1962) Life Histories of North American Wild Fowl Dover Publications, New York. Blaustein AR, Waldman B (1992) Kin recognition in anuran amphibians. Animal Behaviour 44, 207-221. Boch R, Morse RA (1982) Genetic-factor in queen recognition odors of honey bees. Annals of the Entomological Society of America 75, 654-656. Brown CR, Brown MB (1992) Ectoparasitism as a cause of natal dispersal in cliff swallows. Ecology 73, 1718-1723. Bukacinski D, Bukacinska M, Lubjuhn T (2000) Adoption of chicks and the level of relatedness in common gull, Larus canus, colonies: DNA fingerprinting analyses. Animal Behaviour 59, 289-299. Bur MT, Stapanian MA, Bernhardt G, Turner MW (2008) Fall diets of red- breasted Merganser (Mergus serrator) and walleye (Sander vitreus) in Sandusky Bay and adjacent waters of Western Lake Erie. American Midland Naturalist 159, 147-161. Burgman MA, Williams MR (1995) Analysis of the spatial pattern of arthropod fauna of jarrah (Eucalyptus marginata) foliage using a Mantel correlogram. Australian Journal of Ecology 20, 455-457. Burland TM, Barratt EM, Nichols RA, Racey PA (2001) Mating patterns, relatedness and the basis of natal philopatry in the brown long-eared bat, Plecotus auritus. Molecular Ecology 10, 1309-1321. Carlisle TR (1985) Parental response to brood size in a cichlid fish. Animal Behaviour 33, 234-238. Charnov EL (1976) Optimal foraging, marginal value theorem. Theoretical Population Biology 9, 129-136. Clarke AL, Saether BE, Roskaft E (1997) Sex biases in avian dispersal: A reappraisal. Oikos 79, 429-438. Clutton-Brock T (2002) Behavioral ecology - breeding together: kin selection and mutualism in cooperative vertebrates. Science 296, 69-72. Clutton-Brock TH, Brotherton PNM, O'Riain MJ, et al. (2001) Contributions to cooperative rearing in meerkats. Animal Behaviour 61, 705-710.

Clutton-Brock TH, Brotherton PNM, O'Riain MJ, et al. (2000) Individual contributions to babysitting in a cooperative mongoose, Suricata suricatta. Proceedings of the Royal Society of London Series B-Biological Sciences 267, 301-305. Coltman DW, Pilkington JG, Pemberton JM (2003) Fine-scale genetic structure in a free-living ungulate population. Molecular Ecology 12, 733-742. Cooke F, Robertson GJ, Smith CM, Goudie RI, Boyd WS (2000) Survival, emigration, and winter population structure of Harlequin Ducks. Condor 102, 137-144. Craik SR, Savard JPL, Titman RD (2009) Wing and body molts of male red- breasted mergansers in the Gulf of St. Lawrence, Canada. Condor 111, 71- 80. Craik SR, Titman RD (2008) Movements and habitat use by red-breasted merganser broods in Eastern New Brunswick. Wilson Journal of Ornithology 120, 743-754. Craik SR, Titman RD (2009) Nesting ecology of red-breasted mergansers in a common tern colony in Eastern New Brunswick. Waterbirds 32, 282-292. Davis LS (1982) Creching behavior of adelie penguin chicks (Pygoscelis adeliae). New Zealand Journal of Zoology 9, 279-285. Dickinson JL (2007) Decoding dumping ducks. Molecular Ecology 16, 2610- 2612. Dionne M, Caron F, Dodson JJ, Bernatchez L (2008) Landscape genetics and hierarchical genetic structure in Atlantic salmon: the interaction of gene flow and local adaptation. Molecular Ecology 17, 2382-2396. Dobson FS, Chesser RK, Hoogland JL, Sugg DW, Foltz DW (1997) Do black- tailed prairie dogs minimize inbreeding? Evolution 51, 970-978. Double MC, Peakall R, Beck NR, Cockburn A (2005) Dispersal, philopatry, and infidelity: Dissecting local genetic structure in superb fairy-wrens (Malurus cyaneus). Evolution 59, 625-635. Dunford C (1977) Kin selection for ground-squirrel alarm calls. American Naturalist 111, 782-785.

Dunn PO, Cockburn A, Mulder RA (1995) Fairy-wren helpers often care for young to which they are unrelated. Proceedings of the Royal Society of London Series B-Biological Sciences 259, 339-343. Eadie JM, Kehoe FP, Nudds TD (1988) Pre-hatch and post-hatch brood amalgamation in North-American Anatidae - a review of hypotheses. Canadian Journal of Zoology-Revue Canadienne De Zoologie 66, 1709- 1721. Eickwort KR (1973) Cannibalism and kin selection in Labidomera clivicollis (Coleoptera: Chrysomelidae). American Naturalist 107, 452-453. Elder WH, Elder NL (1949) Role of the family in the formation of goose flocks. The Wilson Bulletin 61, 132-140. Emlen JM (1966) Role of time and energy in food preference. American Naturalist 100, 611-&. Emlen ST (1994) Benefits, constraints and the evolution of the family. Trends in Ecology & Evolution 9, 282-285. Epperson BK, Allard RW (1989) Spatial auto-correlation analysis of the distribution of genotypes within populations of lodgepole pine. Genetics 121, 369-377. Eriksson MOG, Andersson M (1982) Nest parasitism and hatching success in a population of goldeneyes Bucephala clangula. Bird Study 29, 49-54. Evans RM (1984) Some causal and functional correlates of creching in young white pelicans. Canadian Journal of Zoology-Revue Canadienne De Zoologie 62, 814-819. Facemire CF, Batzli GO (1983) Suppression of growth and reproduction by cocial-factors in microtine rodents - tests of 2 hypotheses. Journal of Mammalogy 64, 152-156. Fisher RA (1930) The Genetical Theory of Natural Selection The Clarendon Press, Oxford. Fretwell SD, Lucas HL (1969) On territorial behavior and other factors influencing habitat distribution in birds. Acta Biotheoretica 19, 16-36.

Friesen VL, Montevecchi WA, Gaston AJ, Barrett RT, Davidson WS (1996) Molecular evidence for kin groups in the absence of large-scale genetic differentiation in a migratory bird. Evolution 50, 924-930. Frumhoff PC, Schneider S (1987) The social-consequences of honey-bee polyandry - the effects of kinship on worker interactions within colonies. Animal Behaviour 35, 255-262. Girman DJ, Mills MGL, Geffen E, Wayne RK (1997) A molecular genetic analysis of social structure, dispersal, and interpack relationships of the African wild dog (Lycaon pictus). Behavioral Ecology and Sociobiology 40, 187-198. Gorman ML, Milne H (1971) Seasonal changes in adrenal steroid tissue of common eider Somateria mollissima and Its relation to organic metabolism in normal and oil-polluted birds. Ibis 113, 218-&. Gorman ML, Milne H (1972) Creche behaviour in the common eider Somateria m. mollissima L. Ornis Scandinavica 3, 21-26. Grau HJ (1982) Kin recognition in white-footed deermice (Peromyscus leucopus). Animal Behaviour 30, 497-505. Greenwood PJ (1980) Mating systems, philopatry and dispersal in birds and mammals. Animal Behaviour 28, 1140-1162. Gregory RD, Carter SP, Baillie SR (1997) Abundance, distribution and habitat use of breeding goosanders Mergus merganser and red-breasted mergansers Mergus serrator on British rivers. Bird Study 44, 1-12. Hamilton WD (1963) Evolution of altruistic behavior. American Naturalist 97, 354-&. Hamilton WD (1967) Extraordinary sex ratios. Science 156, 477-&. Hamilton WD (1971) Geometry for selfish herd. Journal of Theoretical Biology 31, 295-&. Hansson B, Bensch S, Hasselquist D (2003) A new approach to study dispersal: immigration of novel alleles reveals female-biased dispersal in great reed warblers. Molecular Ecology 12, 631-637.

Hatchwell BJ, Komdeur J (2000) Ecological constraints, life history traits and the evolution of cooperative breeding. Animal Behaviour 59, 1079-1086. He TH, Rao GY, You RL, Ge S, Hong DY (2000) Spatial autocorrelation of genetic variation in three stands of Ophiopogon xylorrhizus (Liliaceae s.l.). Annals of Botany 86, 113-121. Hoglund J, Shorey L (2003) Local genetic structure in a white-bearded manakin population. Molecular Ecology 12, 2457-2463. Hoogland JL (1982) Do prairie dogs avoid extreme inbreeding? Science 215, 1639-1641. Howard WE (1960) Innate and environmental dispersal of individual vertebrates. American Midland Naturalist 63, 152-161. Kilpi M, Ost M, Lindstrom K, Rita H (2001) Female characteristics and parental care mode in the creching system of eiders, Somateria mollissima. Animal Behaviour 62, 527-534. Knudsen B, Evans RM (1986) Parent young recognition in herring gulls (Larus argentatus). Animal Behaviour 34, 77-80. Kokko H, Lindstrom J (1996) Kin selection and the evolution of leks: Whose success do young males maximize? Proceedings of the Royal Society of London Series B-Biological Sciences 263, 919-923. Kraaijeveld K (2005) Black swans Cygnus atratus adopt related cygnets. Ardea 93, 163-169. Lambin X, Krebs CJ (1993) Influence of female relatedness on the demography of Townsends vole populations in spring. Journal of Animal Ecology 62, 536-550. Lendrem DW (1985) Kinship affects puberty acceleration in mice (Mus musculus). Behavioral Ecology and Sociobiology 17, 397-399. Leon M (1987) Plasticity of olfactory output circuits related to early olfactory learning. Trends in Neurosciences 10, 434-438. Linden M, Moller AP (1989) Cost of reproduction and covariation of life history traits in birds. Trends in Ecology & Evolution 4, 367-371.

Loeb MLG, Diener LM, Pfennig DW (2000) Egg-dumping lace bugs preferentially oviposit with kin. Animal Behaviour 59, 379-383. Lopez-Sepulcre A, Kokko H (2002) The role of kin recognition in the evolution of conspecific brood parasitism. Animal Behaviour 64, 215-222. Lorenz K (1970) Studies in animal and human behaviour Harvard University Press, Cambridge, Mass. MacArthur RH, Pianka ER (1966) On optimal use of a patchy environment. American Naturalist 100, 603-+. MacColl ADC, Piertney SB, Moss R, Lambin X (2000) Spatial arrangement of kin affects recruitment success in young male red grouse. Oikos 90, 261- 270. Marvil RE, Cruz A (1989) Impact of brown-headed cowbird parasitism on the reproductive success of the solitary vireo. The Auk 106, 476-480. McAllister MK, Roitberg BD (1987) Adaptive suicidal-behavior in pea aphids. Nature 328, 797-799. McEachern MB, Eadie JM, van Vuren DH, Grp EG (2007) Local genetic structure and relatedness in a solitary mammal, Neotoma fuscipes. Behavioral Ecology and Sociobiology 61, 1459-1469. McFadden CS, Aydin KY (1996) Spatial autocorrelation analysis of small-scale genetic structure in a clonal soft coral with limited larval dispersal. Marine Biology 126, 215-224. McGowan KJ, Woolfenden GE (1989) A sentinel system in the Florida scrub jay. Animal Behaviour 37, 1000-1006. McKinnon L, Gilchrist HG, Scribner KT (2006) Genetic evidence for kin-based female social structure in common eiders (Somateria mollissima). Behavioral Ecology 17, 614-621. McRae SB, Burke T (1996) Intraspecific brood parasitism in the moorhen: parentage and parasite-host relationships determined by DNA fingerprinting. Behavioral Ecology and Sociobiology 38, 115-129.

Milonoff M, Poysa H, Runko P, Ruusila V (2004) Brood rearing costs affect future reproduction in the precocial common goldeneye Bucephala clangula. Journal of Avian Biology 35, 344-351. Munro J, Bedard J (1977) Creche formation in common eider. Auk 94, 759-771. Munshi-South J (2008) Female-biased dispersal and gene flow in a behaviorally monogamous mammal, the large treeshrew (Tupaia tana). Plos One 3, -. Murphy GP, Dudley SA (2009) Kin recognition: competition and cooperation in Impatiens (Balsaminaceae). American Journal of Botany 96, 1990-1996. Nielsen CR, Semel B, Sherman PW, Westneat DF, Parker PG (2006) Host- parasite relatedness in wood ducks: patterns of kinship and parasite success. Behavioral Ecology 17, 491-496. Nituch LA, Schaefer JA, Maxwell CD (2008) Fine-scale spatial organization reflects genetic structure in sheep. Ethology 114, 711-717. O'hara RK, Blaustein AR (1981) An investigation of sibling recognition in Rana cascadae Tadpoles. Animal Behaviour 29, 1121-1126. Oota H, Settheetham-Ishida W, Tiwawech D, Ishida T, Stoneking M (2001) Human mtDNA and Y-chromosome variation is correlated with matrilocal versus patrilocal residence. Nature Genetics 29, 20-21. Osorio-beristain M, Drummond H (1993) Natal dispersal and deferred breeding in the blue-footed booby. Auk 110, 234-239. Ost M (1999) Within-season and between-year variation in the structure of common eider broods. Condor 101, 598-606. Ost M, Vitikainen E, Waldeck P, et al. (2005) Eider females form non-kin brood- rearing coalitions. Molecular Ecology 14, 3903-3908. Paetkau D, Calvert W, Stirling I, Strobeck C (1995) Microsatellite analysis of population-structure in Canadian polar bears. Molecular Ecology 4, 347- 354. Peakall R, Ruibal M, Lindenmayer DB (2003) Spatial autocorrelation analysis offers new insights into gene flow in the Australian bush rat, Rattus fuscipes. Evolution 57, 1182-1195. Pearce JM (2007) Philopatry: a return to origins. Auk 124, 1085-1087.

Pearce JM, McCracken KG, Christensen TK, Zhuravlev YN (2009) Migratory patterns and population structure among breeding and wintering red- breasted mergansers (Mergus serrator) and common mergansers (M. merganser). Auk 126, 784-798. Petrie M, Krupa A, Burke T (1999) Peacocks lek with relatives even in the absence of social and environmental cues. Nature 401, 155-157. Piertney SB, MacColl ADC, Bacon PJ, Dallas JF (1998) Local genetic structure in red grouse (Lagopus lagopus scoticus): evidence from microsatellite DNA markers. Molecular Ecology 7, 1645-1654. Poysa H (1995) Factors affecting abandonment and adoption of young in common eiders and other waterfowl - a comment. Canadian Journal of Zoology- Revue Canadienne De Zoologie 73, 1575-1577. Pusey AE (1980) Inbreeding avoidance in chimpanzees. Animal Behaviour 28, 543-552. Queller DC, Strassmann JE, Hughes CR (1993) Microsatellites and kinship. Trends in Ecology & Evolution 8, 285-&. Rathbun GB (1979) The social structure and ecology of elephant-shrews Parey, Berlin ; Hamburg. Ratnayeke S, Tuskan GA, Pelton MR (2002) Genetic relatedness and female spatial organization in a solitary carnivore, the raccoon, Procyon lotor. Molecular Ecology 11, 1115-1124. Reyer HU (1984) Investment and relatedness - a cost-benefit-analysis of breeding and helping in the pied kingfisher (Ceryle rudis). Animal Behaviour 32, 1163-1178. Rupert JR, Brush T (1996) Red-breasted mergansers, Mergus serrator, nesting in southern Texas. Southwestern Naturalist 41, 199-200. Russell AF, Hatchwell BJ (2001) Experimental evidence for kin-biased helping in a cooperatively breeding vertebrate. Proceedings of the Royal Society of London Series B-Biological Sciences 268, 2169-2174.

Savard JPL (1987) Causes and functions of brood amalgamation in Barrows goldeneye and bufflehead. Canadian Journal of Zoology-Revue Canadienne De Zoologie 65, 1548-1553. Schjorring S (2001) Ecologically determined natal philopatry within a colony of great cormorants. Behavioral Ecology 12, 287-294. Schjorring S, Gregersen J, Bregnballe T (2000) Sex difference in criteria determining fidelity towards breeding sites in the great cormorant. Journal of Animal Ecology 69, 214-223. Schmutz JK, Robertson RJ, Cooke F (1983) Colonial nesting of the Hudson-Bay eider duck. Canadian Journal of Zoology-Revue Canadienne De Zoologie 61, 2424-2433. Scribner KT, Pearce JM (2000) Microsatellites: evolutionary and methodological background and empirical applications at individual, population and phylogenetic levels. In: Molecular Methods in Ecology (ed. Baker AJ). Blackwell Science, Oxford; Malden, MA. Semel B, Sherman PW (2001) Intraspecific parasitism and nest-site competition in wood ducks. Animal Behaviour 61, 787-803. Sheppard DH, Yoshida SM (1971) Social behavior in captive Richardsons ground squirrels. Journal of Mammalogy 52, 793-&. Sherman PW (1977) Nepotism and evolution of alarm calls. Science 197, 1246- 1253. Shields WM (1983) Optimal inbreeding and the evolution of philopatry. In: The Ecology of animal movement (eds. Swingland IR, Greenwood PJ). Clarendon Press, Oxford [Oxfordshire]. Shorey L, Piertney S, Stone J, Hoglund J (2000) Fine-scale genetic structuring on Manacus manacus leks. Nature 408, 352-353. Sjoberg K (1985) Foraging activity patterns in the goosander (Mergus merganser) and the red-breasted merganser (Mergus serrator) in relation to patterns of activity in their major prey species. Oecologia 67, 35-39.

Skabo S, Vaillancourt RE, Potts BM (1998) Fine-scale genetic structure of Eucalyptus globulus ssp. globulus forest revealed by RAPDs. Australian Journal of Botany 46, 583-594. Sonsthagen SA, Talbot SL, Lanctot RB, McCracken KG (2010) Do common eiders nest in kin groups? Microgeographic genetic structure in a philopatric sea duck. Molecular Ecology 19, 647-657. Sonsthagen SA, Talbot SL, Lanctot RB, Scribner KT, McCracken KG (2009) Hierarchical spatial genetic structure of common eiders (Somateria mollissima) breeding along a migratory corridor. Auk 126, 744-754. Spear LB, Pyle P, Nur N (1998) Natal dispersal in the western gull: proximal factors and fitness consequences. Journal of Animal Ecology 67, 165-179. Stoen OG, Bellemain E, Saebo S, Swenson JE (2005) Kin-related spatial structure in brown bears Ursus arctos. Behavioral Ecology and Sociobiology 59, 191-197. Swingland IR, Greenwood PJ (1983) Animal movement: approaches, adaptations and constraints. In: The Ecology of animal movement (eds. Swingland IR, Greenwood PJ). Clarendon Press, Oxford [Oxfordshire]. Tallamy DW (2005) Egg dumping in insects. Annual Review of Entomology 50, 347-370. Taylor AC, Horsup A, Johnson CN, Sunnucks P, Sherwin B (1997) Relatedness structure detected by microsatellite analysis and attempted pedigree reconstruction in an endangered marsupial, the northern hairy-nosed wombat Lasiorhinus krefftii. Molecular Ecology 6, 9-19. Temple HJ, Hoffman JI, Amos W (2006) Dispersal, philopatry and intergroup relatedness: fine-scale genetic structure in the white-breasted thrasher, Ramphocinclus brachyurus. Molecular Ecology 15, 3449-3458. Titman RD (1999) Red-breasted Merganser (Mergus serrator). In: The Birds of North America Online. Cornell Lab of Ornithology, Ithaca. Trivers RL (1971) The evolution of reciprocal altruism. The Quarterly Review of Biology 46, 35.

Uesugi R, Kunimoto Y, Osakabe M (2009) The fine-scale genetic structure of the two-spotted spider mite in a commercial greenhouse. Experimental and Applied Acarology 47, 99-109. van der Jeugd HP, van der Veen IT, Larsson K (2002) Kin clustering in barnacle geese: familiarity or phenotype matching? Behavioral Ecology 13, 786- 790. Waldeck P, Andersson M, Kilpi M, Ost M (2008) Spatial relatedness and brood parasitism in a female-philopatric bird population. Behavioral Ecology 19, 67-73. Waldman B (1982) Sibling association among schooling toad tadpoles - field evidence and implications. Animal Behaviour 30, 700-&. Waldman B (1988) The ecology of kin recognition. Annual Review of Ecology and Systematics 19, 543-571. Ward P, Zahavi A (1973) Importance of certain assemblages of birds as information-centers for food-finding. Ibis 115, 517-534. Wedekind C, Seebeck T, Bettens F, Paepke AJ (1995) MHC-dependent mate preferences in humans. Proceedings of the Royal Society of London Series B-Biological Sciences 260, 245-249. Wolff JO, Plissner JH (1998) Sex biases in avian natal dispersal: an extension of the mammalian model. Oikos 83, 327-330. Woolfenden GE (1975) Florida scrub jay helpers at nest. Auk 92, 1-15. Woolfenden GE, Fitzpatrick JW (1978) Inheritance of territory in group-breeding birds. Bioscience 28, 104-108. Woxvold IA, Adcock GJ, Mulder RA (2006) Fine-scale genetic structure and dispersal in cooperatively breeding apostlebirds. Molecular Ecology 15, 3139-3146. Wright J, Berg E, De Kort SR, Khazin V, Maklakov AA (2001) Cooperative sentinel behaviour in the Arabian babbler. Animal Behaviour 62, 973-979. Wright S (1943) Isolation by distance. Genetics 28, 114-138. Yom-Tov Y (1980) Intraspecific nest parasitism in birds. Biological Reviews of the Cambridge Philosophical Society 55, 93-108.

Young AD, Titman RD (1988) Intraspecific nest parasitism in red-breasted mergansers. Canadian Journal of Zoology-Revue Canadienne De Zoologie 66, 2454-2458. 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 (eds. Stacey PB, Koenig WD). Cambridge University Press, Cambridge; New York.

CONNECTING STATEMENT (1)

In the previous chapter, I reviewed the components required to understand how and why clusters of kin occur in space. Molecular technologies and the use of empirical genetics have greatly improved the researcher’s ability to infer degrees of relatedness between their subjects under a variety of circumstances.

However, it is of fundamental importance that the sources of genetic material used are reliable and yield good quality products. Considering that the ecological inferences made in Chapter 4 depend heavily on the interpretation of molecular data, the next chapter presents results from a series of preliminary experiments assessing the utility of blood and contour feathers as sources of genetic material.

CHAPTER 2: EVALUATION OF GENETIC MATERIALS

2.1) Abstract

For molecular techniques to be useful, it is important to insure that the information yielded is reliable, especially when using non-invasive sampling techniques (i.e. collecting genetic materials that were left behind by an animal).

The objectives of our study were as follows: (1) to extract the nuclear DNA from both blood and feather samples collected from red-breasted mergansers in

Kouchibouguac National Park and compare them on the basis of the quantity and quality of DNA obtained; (2) to compare the PCR product sizes of four microsatellite loci originating from DNA extracted from blood and feathers from the same individual; (3) to determine the relationship between the number of feathers used and total yield of DNA; (4) to determine the effects of a second round of purification on the final extraction product. DNA was obtained successfully from both blood and feathers; however the total yield (µg) from the blood-derived DNA was significantly higher. In contrast, feather-derived DNA was found to be less contaminated (based on the 260:280 OD ratio), perhaps due to the DNA extraction protocol used. Comparisons at microsatellite loci revealed small discordances between the blood and feather DNA, despite their common origins. Marginal evidence was found for a positive linear relationship between the number of feathers used in an extraction protocol and the final yield of nucleic acid. A second round of purification improved the quality of the extraction product, however at a cost of approximately a two-fold reduction of total yield.

2.2) Introduction

The use of molecular technologies has become integral to studies of ecology including those of phylogenies, populations, behaviour and kin selection

(Queller et al. 1993, Avise 1996). The use of noninvasive genetic sampling (e.g. from hair, feathers, feces, urine, etc.) from animals has especially had important consequences (Pearce et al. 1997). In situations when the activities of researchers can have negative impacts on survival and reproductive success, noninvasive sampling can lessen the necessity for capturing and handling animals.

Furthermore, when it is logistically difficult to acquire large sample sizes or to observe the behaviours in-question by means of direct observation (e.g. Elder &

Elder 1949, Semel & Sherman 2001), noninvasive sampling (i.e. collecting genetic materials left behind by an animal) can significantly augment the relevance of ecological inferences made (Pearce et al. 1997).

Conversely, destructive (i.e. animals are killed) and nondestructive samples (i.e. animals are detained) are considered to be more reliable alternatives when acquiring genetic samples (Taberlet et al. 1999). Unscrupulous reliance on noninvasive sampling techniques may result in obscured results due to complications arising from: (a) insufficient yields of genetic material; (b) degradation; and/or (c) contamination (Taberlet et al. 1999).

Excessive degradation in conjunction with low extraction yields can result in serious genotyping errors, especially the occurrence of ‘allelic dropout’ or as they are also called, null alleles (Pemberton et al. 1995, Gagneux et al. 1997).

Null alleles arise when polymerase chain reaction (PCR) primers fail to anneal to

40 the template DNA because of degradation that occurs at a primer attachment site.

Whether or not this has occurred on one or both strands determines whether the resulting gel is falsely scored as a homozygote (i.e. only on band present) or does not have any bands at all. Sample contamination, especially from conspecifics, also presents a risk that bands may be scored inappropriately (Pemberton et al.

1995). Therefore, assessing the limitations of noninvasive techniques is of utmost importance for the production of reliable data and meaningful inferences.

The objectives of this study were: (1) to extract nuclear DNA using both nondestructive (i.e. blood) and noninvasive (i.e. feather) samples collected from red-breasted mergansers at Kouchibouguac National Park, New Brunswick; (2) to compare the quantity, quality and information arising from DNA obtained from blood and feather samples; (3) to determine the relationship between the number of feathers used and total yield of DNA; and finally (4) to determine the effect of a second round of purification on the final extraction product.

2.3) Methods

Specimens

Samples were collected during June and July of 2008. Red-breasted mergansers (Mergus serrator) inhabited a series of small barrier islands in

Kouchibouguac National Park, New Brunswick, Canada. From each bird, a

0.25cc blood sample was collected from the ulnar vein and deposited into a heparinized vacutainer® (BD) and eventually stored at -20ºC. In addition to a blood sample, as many contour feathers as possible were recovered from each

41 nest visited; usually between 4-8 feathers. Feathers were deposited in small, labeled envelopes and subsequently stored at 4ºC. DNA extraction protocols were initiated approximately six months post-collection.

Additional contour feathers were obtained from McGill University’s

Avian Science and Conservation Centre. These feathers originated from deceased

American kestrels (Falco sparverius) which had been stored at -20ºC. After a 24 h thawing period, feathers were plucked directly from a single region of the breast.

They were processed immediately.

DNA extraction and precipitation

The extraction and precipitation of DNA from blood samples was conducted using a DNeasy® Blood and Tissue Kit (QIAGEN) using 10µl of blood. Procedures were carried out in accordance with the manufacturer’s recommendations with the exception to having used: (a) a PBS solution with pH=7.4 as opposed to pH=7.2; and (b) a 30 min incubation period after step 2 of the protocol instead of 10 min. In preliminary trails, the extended incubation period greatly improved the extraction yields.

The protocol for extraction and precipitation of DNA from feathers was a modification of that described by Sambrook et al. (1989). The time required in order to process one batch of samples was three days. At least four feathers from each individual bird were used. The portion of the feather between the superior umbilicus and the calamus tip was cut finely and placed into a 1.5ml microtube and then combined with 600µl of ‘extraction mix’ (Table 2.1). Samples were

42 subsequently vortexed and incubated in a water bath at 50ºC for 30 min. Next,

100l of proteinase K was added to the microtubes before they were returned to the water bath and left to incubate overnight (> 12 h). 600l of phenol chloroform was then added to the samples and vortexed again for 1 min. After a second 30 m incubation period at room temperature (25ºC), samples were then centrifuged at

12,000 RPM for 15 m. After this step, the supernatant containing the DNA was collected and placed into a fresh microtube and combined with 800l of chilled

100% EtOH. In order to facilitate DNA precipitation, samples were placed at

-20C for 10 min and 1l of glycogen was added before centrifuging at 12,000

RPM for 5 min. The microtubes were then inverted and the pellets were left to dry for approximately 15 m. Samples were washed a second time using 800 l of 70%

EtOH. Finally samples were centrifuged at 12,000 RPM for 5 min before being left to dry overnight (>10-12 h). DNA was re-suspended in 50l of dd-H2O and the tubes were stored at -20C until later use.

In order to determine whether the quality of the final product could be improved, all samples obtained from kestrel feathers were re-purified. This involved re-diluting samples with 600µl of dd-H20 and repeating the above protocol from the addition of the phenol chloroform. The only necessary modification was to combine 700µl of 3M sodium acetate (pH: 5.2) with the supernatant once it had been separated.

The quantity and purity of the DNA extracted was assessed using a

NanoDrop® ND-1000 Spectrophotometer (Thermo Fisher Scientific Inc.). DNA concentrations (ng/µl) were measured by their optical densities and the degree of

43 protein contamination was assessed by the ratio of absorbance at 260nm vs.

280nm. Pure DNA has a 260:280 ratio of 1.8, however measurements > 1.5 are still considered to be within a workable range.

PCR and gel-electrophoresis

Four sets of heterologous primers were used to compare the products from blood and feather samples originating from the same specimen. The primers used were Aph08, Aph13, Aph20 and Mm04. All reactions were carried out using a

PTC-200 thermocycler (MJ Research, Inc.). Further information on primers, PCR reagents and thermocycler regimes is given in Chapter 3.3.

The resultant PCR products were resolved using a 2% w/v agarose gel in order to confirm successful amplification of the desired products. PCRs were then re-run using fluorescent-labeled primers (Applied Biosystems) and were finally sent to Genome Quebec, McGill University where the product sizes were resolved using an ABI-3730 DNA Analyzer (Applied Biosystems).

Statistical analysis

Assumptions of normality were assessed using the Shapiro-Wilk test and if met, comparisons were made using a one-tailed t-test; otherwise the Wilcoxon rank-sum test was used (Hollander & Wolfe 1973). A multiple linear regression model was used to assess the relationship between the number of feathers used at the outset and the total yield of DNA (µg); normality in the residual structure was verified using a Q-Q normal plot. In order to compare the quantity and quality of

DNA before and after a second round of purification, paired t-tests or paired

Wilcoxon rank-sum tests were used, depending on the normality of distribution.

Using Queller & Goodnight’s (1989) coefficient of relatedness, the alleles scored at all four loci were transformed into a matrix of pairwise distances. Principal coordinate analysis (PCoA) was then used to visualize the patterns of variation and verify whether blood and feather samples from the same individuals cluster together. All statistical procedures were carried out using the open-source statistical software package, R version 2.10.1 (http://www.r-project.org/).

2.4) Results

Genetic samples were obtained from a total of 60 individuals; 42 of which were blood samples taken directly from specimens while the remaining 18 consisted of feathers recovered from nests. DNA was successfully extracted from each sample, but in variable quantity (see below). There were three individuals from which both blood and feather samples were obtained. These three feather samples were not included in the count above.

Blood vs. feathers

The assumption of normality was not met for either statistical population

(i.e. blood and feathers) with respect to both DNA yield and absorbance, therefore the Wilcoxon rank-sum test was used. The null hypothesis that distribution of

DNA yields from blood and feather samples is identical, was rejected (W= 756, p- value= <0.000). The quantity of DNA (µg) obtained from blood was far greater

45 than that obtained from feathers (Figure 2.1). Conversely, the DNA extracted from feather samples had a 260:280 ratio closer to purity than that extracted from blood (W= 699.5, p-value= <0.000) (Figure 2.2). However, both were still considered to be within a workable range.

Minor size discordances at microsatellite loci were detected in two of the three comparisons between blood and feather DNA. A two-dimensional ordination diagram is presented in Figure 2.3. The proportion of variance explained by the first three principal coordinate axes was 0.414, 0.225 and 0.149, respectfully.

Feathers and DNA yield

There were no major deviations from normality detected in the residual of the regression model (Figure 2.4b). The slope was calculated to be 1.92, however the null hypothesis that the regression coefficient is equal to zero could not be rejected using a critical value of α= 0.05 (t= 2.46, p-value= 0.070), albeit by a small margin. The proportion of variance (R2) in DNA yield explained by the number of feathers used was 0.60. A graphical depiction of the model is presented in Figure 2.4a.

Re-purification

All the statistical populations in this portion of the study did not deviate from a normal distribution; therefore paired t-tests were used to compare the before-after effects of re-purification. The null hypothesis that there is no change

46 in purity after samples had been processed for a second time was rejected using a critical value of α=0.05 (one-tailed, t= -2.410, p-value= 0.030) (Figure 2.5a).

However, it was also found that total DNA yields were significantly lower after the second extraction (one-tailed, t= 4.333, p-value= 0.004) (Figure 2.5b).

2.5) Discussion

Blood vs. feathers

There were clear and consistent discrepancies between the quantity and purity of DNA extracted from blood and feather samples. On average, blood samples produced nearly 20X as much DNA as feather samples (Figure 2.1). This was not surprising and was in accordance with results of other authors (e.g.

McKinnon et al. 2006). Conversely, while both were within the acceptable range of quality, there were usually higher levels of contamination in blood-derived

DNA (Figure 2.2). This is possibly a consequence of the difference in quality between the extraction protocols used and not the sources of DNA themselves.

In the ordination diagram, blood and feather samples taken from the same individual always clustered together; however in two of the three comparisons, the matches were imperfect causing the objects to be farther apart in two- dimensional space. While there are insufficient data presented here to make large inferences about the utility of feathers as a source of genetic material, the results clearly warrant caution. The use of degraded DNA, especially in low concentrations, is known to increase the frequency of null alleles and this may explain at least some of discordance observed in these results (Gagneux et al.

1997, Taberlet et al. 1999). McKinnon et al. (2006) reported that several microsatellite loci could not be amplified due to low concentrations of DNA obtained from feathers.

Contrary to these results, Pearce et al. (1997) reported perfect matches between all alleles scored from blood and feather samples that had been taken from a common specimen. A factor that could account for the discrepancy in our results may be our feather storage. Pearce et al. stored feathers at -20ºC where in this study they were kept at 4ºC. Alternatively, Sonsthagen et al. (2009) stored feathers along with silica gel desiccants at room temperature. Possibly prolonged exposure to humidity caused excessive degradation to genetic material in this study. The possibility that the two instances of mismatch resulted from intraspecific contamination cannot be excluded, however, it is highly unlikely because there was a maximum of two bands observed at each locus.

Feathers and DNA yield

Despite the small sample size (n=6), results from the kestrel feathers indicated a positive-linear relationship between the number of feathers used and the total yield of DNA obtained. However, in Figure 2.4a it also appears that the variance in the response variable increases along with the number of feathers used; if this trend were to be confirmed, a non-linear transformation of the data is merited. Attempts to transform the data did not improve the structure of the residuals.

Re-purification

Even with the small sample sizes, it was clear that the re-purification of a sample ameliorated its quality, however at no small cost. In each of the six samples, re-purification led to an increased 260:280 ratio yet also to an approximate two-fold decrease in the concentration of nucleic acids (Figure 2.5).

It is important to consider this trade-off, especially when the initial yield was already lower than desired.

2.6) Literature Cited

Avise JC (1996) Three fundamental contributions of molecular genetics to avian ecology and evolution. Ibis 138, 16-25. Elder WH, Elder NL (1949) Role of the family in the formation of goose flocks. The Wilson Bulletin 61, 132-140. Gagneux P, Boesch C, Woodruff DS (1997) Microsatellite scoring errors associated with noninvasive genotyping based on nuclear DNA amplified from shed hair. Molecular Ecology 6, 861-868. Hollander M, Wolfe DA (1973) Nonparametric Statistical Methods. Wiley, New York. McKinnon L, Gilchrist HG, Scribner KT (2006) Genetic evidence for kin-based female social structure in common eiders (Somateria mollissima). Behavioral Ecology 17, 614-621. Pearce JM, Fields RL, Scribner KT (1997) Nest materials as a source of genetic data for avian ecological studies. Journal of Field Ornithology 68, 471- 481.

Pemberton JM, Slate J, Bancroft DR, Barrett JA (1995) Nonamplifying Alleles at Microsatellite Loci - a Caution for Parentage and Population Studies. Molecular Ecology 4, 249-252. Queller DC, Strassmann JE, Hughes CR (1993) Microsatellites and kinship. Trends in Ecology & Evolution 8, 285-&. Queller DC, Goodnight KF (1989) Estimating relatedness using genetic-markers. Evolution 43, 258-275. Sambrook J, Fritsch EF, Maniatis T (1989) Molecular cloning : a laboratory manual Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. Semel B, Sherman PW (2001) Intraspecific parasitism and nest-site competition in wood ducks. Animal Behaviour 61, 787-803. Sonsthagen SA, Talbot SL, Lanctot RB, Scribner KT, McCracken KG (2009) Hierarchical spatial genetic structure of common eiders (Somateria mollissima) breeding along a migratory corridor. Auk 126, 744-754. Taberlet P, Waits LP, Luikart G (1999) Noninvasive genetic sampling: look before you leap. Trends in Ecology & Evolution 14, 323-327.

Table 2.1 – Composition of ‘extraction mix’ used in phenol chloroform DNA extraction protocol

Ingredient Concentration Amount (µl) / 1.5 ml microtube

TRIS 1M 6

NaCl 5M 12

EDTA 0.5M 12

SDS 10% 30

Beta-mercaptoethanol 100% 12

DEPC treated H2O 100% 528

Figure 2.1 – Total amount of DNA (µg) obtained from blood and feather samples, respectively. On average, blood samples produced nearly 20X as much DNA as feather samples.

Figure 2.2 – The comparison of DNA quality expressed as the 260:280 ratio. The dashed line represents the ratio expected for pure DNA. While both were within the acceptable range of absorbance at the 260:280 ratio, there were usually higher levels of contamination found in DNA extracted from blood.

Figure 2.3 – Two-dimensional PCoA ordination diagram. The proportion of variance explained by the first two principal coordinate axes was 0.414 and

0.225, respectfully. In this figure, the positions of the objects (i.e. nesting females) in relation to one another are approximations of their genetic distances. The PCoA was used in conjunction with Queller & Goodnight’s (1989) coefficient of relatedness.

(A) 25

r-squared=0.60

p-value=0.07

total yield DNAtotal (ug) 5

3 4 5 6 7 8 9 10

no. feathers

(B)

Normal Q-Q 2

5 1

-1 Standardizedresiduals

-1.0 -0.5 0.0 0.5 1.0

Theoretical Quantiles

Figure 2.4 – (a) Curve representing the positive-linear relationship between number of feathers used and total yield of DNA (ug). (b) The Q-Q normal plot of the model which indicates there were no major deviations from normality detected in the residuals

Figure 2.5 – As a result of the second round of purification, significant changes were observed in both (a) DNA purity (t= -2.410, p-value= 0.030) as measured by the 260:280 ratio (b) total yield of DNA (t= 4.333, p-value= 0.004).

CONNECTING STATEMENT (2)

The use of an effective DNA extraction protocol is of fundamental importance to the application of molecular techniques in ecology. Molecular markers and microsatellites specifically, have many desirable properties including their suitability for determining coefficients of relatedness between individuals.

PCR primer-pairs corresponding to microsatellites located in the red-breasted merganser (Mergus serrator) genome specifically have yet to be developed. There are, however, those derived from closely-related species which have been reported to yield products when using M. serrator template DNA. In the following chapter, we present results from having screened 14 heterologous microsatellite primers; and where products were successfully obtained, we report the conditions at which reactions were processed.

CHAPTER 3: SCREENING FOR HETEROLOGOUS MICROSATELLITE

PRIMERS

3.1) Abstract

Polymorphic microsatellites are molecular markers well-suited for use in empirical genetics. While it may be ideal to acquire species-specific PCR primers, it is not always possible because of limitation of both time and financial resources.

Under such circumstances, the use of heterologous primers is a fitting alternative.

In this chapter, we report results from having screened 14 heterologous microsatellite primers in the red-breasted merganser (M. serrator) genome that originated from other species in the family Anatidae. Objectives were to determine: (a) which of the 14 primers yielded products; (b) the PCR conditions under which these products were amplified consistently and with sufficient strength; and (c) whether these microsatellite loci were polymorphic. Because of the limited scope of this project, there were only sufficient resources available to genotype four of the 14 loci across all samples (N=60). Well-defined products were eventually obtained on a consistent basis from 12 of 14 primers by manipulating template DNA concentrations, MgCl2 concentrations, annealing temperatures and cycling regimes used in the reaction; precise PCR conditions are given. From 60 samples, only 46 were amplified successfully across all four of the loci genotyped. Observed genotype frequencies in two of the microsatellites were found to deviate significantly from those expected under the Hardy-

Weinberg equilibrium.

3.2) Introduction

Microsatellites are one of many types of molecular marker available and are interspersed at relatively high frequencies throughout the nuclear and chloroplast genomes of eukaryotes (Freeland 2005). They consist of short tandem repeats of 1-6 base pairs (bp). It is because of the high variability in the number of repeats (i.e. length polymorphisms) that microsatellites are frequently implemented as tools in empirical genetics (Scribner & Pearce 2000). The high mutation rate enables them to be used for the genetic discrimination of both individuals and populations (Queller et al. 1993, Freeland 2005). Their abundance in the genomes of different taxa is also variable (Brooker et al. 1994, Primmer et al. 1997). The frequency of microsatellites and other forms of ‘junk DNA’ in the avian genome are reported to occur less often than in mammals and fish, presumably due to the intensive pressure for birds to reduce their overall body weight (Primmer et al. 1997). Despite this, the successful use of microsatellites in avian ecological studies has not been impeded by a paucity of polymorphic loci

(e.g. Shorey et al. 2000, Nielsen et al. 2006, Temple et al. 2006, Woxvold et al.

2006, Alcaide et al. 2009).

The use of microsatellites depends on the availability of primer-pairs which consist of 18-22 bp that correspond to flanking regions of the target DNA sequence on each strand. While finding species-specific microsatellites for each organism might be an ideal approach, the availability of resources (e.g. time, money) may be a factor as cloning new microsatellites is a time- and cost- intensive process (Freeland 2005). The use of primers that had originally been

59 cloned from the genome of another organism (i.e. heterologous primers) presents a reasonable alternative (Scribner & Pearce 2000).

The red-breasted merganser (Mergus serrator) is a medium-size sea duck breeding across a Holarctic distribution. Aside from specific aspects of its ecology, there are still large gaps in knowledge about the red-breasted merganser (Titman

1999). Therefore the use of molecular techniques in the study of this organism can be of great value in revealing unknown aspects of this animal’s life history.

Currently, no primer-pairs have been developed specifically for the M. serrator genome. Previous attempts to utilize heterologous microsatellites in this species have been largely unsuccessful due to a low level of polymorphic loci and inconsistencies in amplification (Bouchard 2002). The principal objective of this study is to screen 14 heterologous microsatellite primers originating from other species of waterfowl in the Anatidae family and to determine the conditions at which they yield the best and most consistent results. Four of these microsatellite loci are amplified using fluorescent-labeled primers in order to determine the degree of allelic variability. The ultimate goal of this work is to find loci which can be subsequently used to identify genetic attributes as part of an effort to answer social and ecological questions related to this species.

3.3) Methods

Specimens

Genetic samples were collected from female red-breasted mergansers breeding in Kouchibouguac National Park, New Brunswick during the summer

60 months of 2008. All specimens originated from females that were nesting on a series of small barrier islands. Two types of genetic sample were collected: (a) a

0.25cc aliquot of blood which was taken from the ulnar vein; and (b) 4-8 contour feathers collected from the nests of known females. Blood samples were deposited directly into a heparinized vacutainer® (BD) and stored at -20ºC while the feathers were placed into sealed envelopes and stored at 4ºC. Laboratory analysis commenced approximately six months post-collection.

DNA extraction

The DNA from blood samples was extracted by using a DNeasy® Blood and Tissue Kit (QIAGEN). Procedure was conducted in accordance with the manufacturer’s recommendations with only minor modifications. The phenol- chloroform extraction protocol used in order to obtain DNA from feathers was based on that described by Sambrook et al. (1989). Further detail about methods used and modifications to the DNA extraction protocols is given in Chapter 2.3.

The concentration of genomic DNA was measured using a NanoDrop® ND-1000

Spectrophotometer (Thermo Fisher Scientific Inc.). Final products whose concentrations exceeded 50ng/µl were diluted into aliquots for use in polymerase chain reactions (PCR).

Primers & PCR conditions

In some cases, particular primers were selected based on published information where the authors claimed success in having cross-amplified products

61 using the M. serrator genome (e.g. Guay & Mulder 2005). In most cases however, the primers were chosen based on the advice and preliminary work undertaken by

J. M. Pearce. Specific details pertaining to the set of 14 primers used are listed in

Table 3.1. For expediency and to increase efficiency, ideal PCR conditions for these primers were determined using only a small subset of samples. All reactions were carried out using a PTC-200 thermocycler (MJ research Inc.). Each reaction had 0.1 µM of each of the forward and reverse primers, 1.25 µM of DNTPs and

0.650U Taq DNA polymerase (Qiagen). All reactions had a total volume of 25µl.

Amounts of DNA and MgCl2 used in a reaction were manipulated until satisfactory results were obtained consistently. Similarly all reactions were first run using an annealing temperature of 50ºC and then increased gradually, as necessary. Finally, two generalized thermocycler regimes, described in Table 3.2, were tested on each primer for optimal performance.

Gel-electrophoresis

The performance of each reaction was assessed using an ethidium- bromide gel-electrophoresis protocol. PCR products were combined with a loading buffer at a ratio of 1:5 and then placed into the wells of a 2% w/v TBE- agarose gel. Also loaded were 100 bp ladders in order to gauge size of the products. After they had migrated sufficiently, the gels were bathed in an ethidium bromide solution (0.5 l/ml of H2O) for 10 min. Finally the presence or absence of amplification products was verified by viewing the gel under UV light.

Microsatellite genotyping

The resolution provided by agarose gel is insufficient to discriminate between alleles differing by only a few base pairs. Therefore, in order to assess the allelic variability, the reactions were re-run using four of the primers from the previous step, this time labeled with a fluorescent marker. Also, instead of just a small subset, the full set of samples was processed in order to properly assess the levels of allelic variability at each locus. The primers and their associated labels were Aph08-FAM, Aph13-VIC, Aph20-PET and Mm04-NED manufactured by

Applied Biosystems. The fluorescent-labeled amplicons were sent to Genome

Quebec, McGill University where their sizes were resolved using an ABI-3730

DNA Analyzer (Applied Biosystems).

Statistical analysis

Allelic frequencies, Hardy-Weinberg equilibriums (HWE) (using Markov chain method with default parameters) and linkage disequilibrium were calculated in Genepop 3.1 (Raymond & Rousset 1995).

3.4) Results

Collection, extraction and amplification

A total of 60 samples was collected consisting of 42 blood and 18 feather samples. Furthermore, DNA was successfully extracted from every sample (see

Chapter 2.4). Of the 14 primers tested, only two (i.e. Blm01 and Blm03) did not amplify. Strong products for the remaining 12 were obtained, eventually on a

63 consistent basis. Ideal PCR conditions found for the primers are presented in

Table 3.3 in addition to the product sizes, which were approximated using the

DNA ladders.

Microsatellite genotyping

The thermal conditions described in Table 3.3 were used to amplify the four sets of fluorescent-labeled primers. There was a higher and more consistent degree of amplification in the DNA originating from blood rather than that from feathers. Of the 42 DNA samples obtained from blood, two could not be amplified in two of the four loci and only one sample did not amplify at all. In contrast, only seven of the 18 samples from feather DNA were amplified across all loci. All microsatellite loci were found to be polymorphic.

Statistical analysis

The total number of alleles, observed heterozygosity (HO) and expected heterozygosity (HE) at each locus, are presented in Table 3.4. Aph20 and Mm04 were both found to deviate significantly from HWE; with an excess and deficiency of heterozygosity, respectively. None of the loci were found to be in linkage disequilibrium.

3.5) Discussion

Successful amplification using M. serrator DNA was obtained in 12 of the

14 heterologous microsatellites tested. By refining the PCR conditions for

64 annealing temperature, template DNA concentration, MgCl2 concentration and cycling regime, a clear and consistent product was obtained. Furthermore, the sizes of amplicons were found to be in general concordance with those reported by the original authors.

In contrast to Guay & Mulder (2005) who had successfully cross- amplified Blm01 and Blm03 in the M. serrator genome, no products could be obtained in this study despite numerous attempts. It is possible that the template

DNA used in Guay & Mulder came from a population that is genetically different from the one at Kouchibouguac. Mutations occurring at the primer attachment site flanking the microsatellite region can result in non-amplification. Unfortunately, the source population of the M. serrator DNA used in Guay & Mulder’s study was not indicated. At least within North American populations of red-breasted mergansers, a large divergence as such seems unlikely considering the apparent lack of genetic differentiation (Pearce et al. 2009a). It is also possible that the products were in fact present, however in such small quantity that they were below the detection capability of the electrophoresis used.

The refined PCR conditions were equally as effective in amplifying microsatellites when using fluorescent-labeled primers. It was not surprising that a greater proportion of amplification was observed in DNA obtained from blood vs. that from feathers consistent with the experience of other researchers (e.g.

McKinnon et al. 2006). It is unclear why there was amplification difficulty in three of the blood samples; mishandling, improper storage and other human error may have been factors. Conversely, the large proportion of non-amplification of

DNA originating from feathers (~61%) is likely attributable to poor-quality, degraded DNA and small quantities (Taberlet et al. 1999, see Chapter 2).

All four of the loci genotyped proved to be polymorphic with a mean observed heterozygosity of 0.61. Two of the four were found to deviate significantly from HWE. Under HWE, genotype frequencies are expected to be in equilibrium with the individual allele frequencies; deviations from the expected frequencies can be due to various factors such as assortative mating, selection or genetic drift (Freeland 2005). In this study, the primary concern from observing deviations from HWE is the potential that they were caused by the presence of non-amplifying or null alleles (Pemberton et al. 1995). In the genotyping process, a single null allele at a particular locus can be falsely scored as a homozygote because only the allele that has been amplified is visible on the gel. Consequently, the frequencies of homozygotes compared to what is expected are over- represented and the locus is propelled into a state of heterozygote deficiency

(Pemberton et al. 1995). In the microsatellite Aph20, the proportion of observed heterozygotes (HO = 0.50) exceeded that of expected heterozygotes (HE = 0.48).

Although observed genotype frequencies were found to differ significantly from those expected under HWE, this is unlikely due to the presence of null alleles.

Nielsen et al. (2006) had applied similar reasoning in their decision to integrate the microsatellite Sfil4 into their calculations of genetic relatedness despite its significant deviation from HWE.

On the contrary, the proportion of observed heterozygosity in Mm04 (HO =

0.41) was exactly half of what was expected (HE = 0.82) under HWE; this of

66 course is more problematic. The finding that HO < HE can be attributed to three separate phenomena. The first is an actual deviation from HWE caused by the occurrence of either: (a) inbreeding; or (b) assortative mating based on the traits associated with that allele or another locus on the same strand (Freeland 2005).

Inbreeding does not appear likely cause because in red-breasted mergansers, as in most other sea ducks, male dispersal is responsible for connecting the otherwise distinct populations of philopatric females (R. D. Titman per. comm.), thus making the effective population size unlimited. Male dispersal is believed to occur in red-breasted mergansers because there is relatively little genetic differentiation in populations of this species across the North American range

(Pearce et al. 2009a). The alternative, that certain alleles at this locus are, either themselves or through linkage, associated with selective advantages, cannot be discounted. However, this does not seem likely since this locus was found to be in

HWE in Mergus merganser merganser, a close-relative of M. serrator’s

(Gautschi & Koller 2005).

The second possibility for the heterozygote deficiency in Mm04 is the

Walhund effect, which occurs as a result of having inadvertently sampled from multiple populations (Freeland 2005). However, based on banding records, there is no evidence that red-breasted mergansers nesting on neighboring islands constitute separate genetic populations (see appendices in Bouchard 2002 and

Craik 2009). Finally, the presence of null alleles is likely the most realistic explanation. Significant deviations from HWE were also found in some of the microsatellite markers used in McKinnon et al. (2006). In one case, McKinnon et

67 al. rationalized the inclusion of a particular marker in their analysis based on the fact that the number of possible genotypes was too large in relation to the sample size; therefore a proper representation of the full range of genotypes was not a realistic expectation. It is unclear whether the application of such reasoning has any merit in this study where there are 91 possible genotypes of Mm04 and a sample size of 46 however until reinforced by additional markers, the inferences drawn based on the use of this microsatellite should be conservative and treated with scrutiny.

3.6) Literature Cited

Alcaide M, Serrano D, Tella JL, Negro JJ (2009) Strong philopatry derived from capture-recapture records does not lead to fine-scale genetic differentiation in lesser kestrels. Journal of Animal Ecology 78, 468-475. Bouchard MLJ (2002) Determining rates of intraspecific nest parasitism in a colony of Red-breasted Mergansers (Mergus serrator) using microsatellite analysis M.Sc. thesis, McGill University. Brooker AL, Cook D, Bentzen P, Wright JM, Doyle RW (1994) Organization of microsatellites differs between mammals and cold-water teleost fishes. Canadian Journal of Fisheries and Aquatic Sciences 51, 1959-1966. Buchholz WG, Pearce JM, Pierson BJ, Scribner KT (1998) Dinucleotide repeat polymorphisms in waterfowl (family Anatidae): characterization of a sex- linked (Z-specific) and 14 autosomal loci. Animal Genetics 29, 323-325. Craik SR (2009) Habitat use by breeding and molting red-breasted mergansers in the Gulf of St. Lawrence Ph.D. thesis, McGill University. Freeland J (2005) Molecular ecology John Wiley & Sons, Chichester, West Sussex, England; Hoboken, NJ.

Gautschi B, Koller B (2005) Polymorphic microsatellite markers for the goosander (Mergus merganser). Molecular Ecology Notes 5, 133-134. Guay PJ, Mulder RA (2005) Isolation and characterization of microsatellite markers in musk duck (Biziura lobata : Aves), and their application to other waterfowl species. Molecular Ecology Notes 5, 249-252. Maak S, Wimmers K, Weigend S, Neumann K (2003) Isolation and characterization of 18 microsatellites in the Peking duck (Anas platyrhynchos) and their application in other waterfowl species. Molecular Ecology Notes 3, 224-227. McKinnon L, Gilchrist HG, Scribner KT (2006) Genetic evidence for kin-based female social structure in common eiders (Somateria mollissima). Behavioral Ecology 17, 614-621. Nielsen CR, Semel B, Sherman PW, Westneat DF, Parker PG (2006) Host- parasite relatedness in wood ducks: patterns of kinship and parasite success. Behavioral Ecology 17, 491-496. Pearce JM, McCracken KG, Christensen TK, Zhuravlev YN (2009a) Migratory patterns and population structure among breeding and wintering red- breasted mergansers (Mergus serrator) and common mergansers (M. merganser). Auk 126, 784-798. Pearce JM, Zwiefelhofer D, Maryanski N (2009b) Mechanisms of Population Heterogeneity among Molting Common Mergansers on Kodiak Island, Alaska: Implications for Genetic Assessments of Migratory Connectivity. Condor 111, 283-293. Pemberton JM, Slate J, Bancroft DR, Barrett JA (1995) Nonamplifying Alleles at Microsatellite Loci - a Caution for Parentage and Population Studies. Molecular Ecology 4, 249-252. Primmer CR, Raudsepp T, Chowdhary BP, Moller AR, Ellegren H (1997) Low frequency of microsatellites in the avian genome. Genome Research 7, 471-482. Queller DC, Strassmann JE, Hughes CR (1993) Microsatellites and kinship. Trends in Ecology & Evolution 8, 285-&.

Raymond M, Rousset F (1995) Genepop (Version-1.2) - Population-Genetics Software for Exact Tests and Ecumenicism. Journal of Heredity 86, 248- 249. Sambrook J, Fritsch EF, Maniatis T (1989) Molecular cloning : a laboratory manual, 2nd edn. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. Scribner KT, Pearce JM (2000) Microsatellites: evolutionary and methodological background and empirical applications at individual, population and phylogenetic levels. In: Molecular Methods in Ecology (ed. Baker AJ). Blackwell Science, Oxford; Malden, MA. Shorey L, Piertney S, Stone J, Hoglund J (2000) Fine-scale genetic structuring on Manacus manacus leks. Nature 408, 352-353. Taberlet P, Waits LP, Luikart G (1999) Noninvasive genetic sampling: look before you leap. Trends in Ecology & Evolution 14, 323-327. Temple HJ, Hoffman JI, Amos W (2006) Dispersal, philopatry and intergroup relatedness: fine-scale genetic structure in the white-breasted thrasher, Ramphocinclus brachyurus. Molecular Ecology 15, 3449-3458. Titman RD (1999) Red-breasted Merganser (Mergus serrator). In: The Birds of North America Online. Cornell Lab of Ornithology, Ithaca. Woxvold IA, Adcock GJ, Mulder RA (2006) Fine-scale genetic structure and dispersal in cooperatively breeding apostlebirds. Molecular Ecology 15, 3139-3146.

Table 3.1 –14 heterologous microsatellite loci tested and associated primers. For each primer, reported variability and the GenBank accession numbers are provided. The number of alleles reported is in reference to the original species, not M. serrator. Locus Original species Primer sequences (5’-3’) * No. of GenBank Publication name alleles accession Aph04 Anas platyrhynchos fwd 5 AJ515884 Maak et al. CCTCggTATTgTTTTCCAT 2003 rev gCTCTgAAgggCAgTAgTTAg Aph08 -- fwd 4 AJ515887 -- AAAgCCCTgTgAAgCgAgCTA rev TgTgTgTgCATCTgggTgTgT Aph13 -- fwd 6 AJ515889 -- CAACgAgTgACAATgATAAAA rev CAATgATCTCACTCCCAATAg Aph15 -- fwd 3 AJ515890 -- TgAATATgCgTggCTgAA rev CAgTgAggAATgTgT TTgAgT T Aph20 -- fwd 3 AJ515895 -- ACCAgCCTAgCAAgCACTgT rev gAggCTTTAggAgAgATTgAAAAA Aph24 -- fwd 5 AJ515899 -- TCAACCAgTggTCAgAgAAAAA rev AggTCAgCCCCCATTTTAgT Blm01 Biziura lobata fwd 12 AY766435 Guay & Mulder AAAATgCTTggTTAATAgCAAAAg 2005 rev TCTTTACACCTCCATTgAATATATCg

Table 3.1 – continued from p. 71

Locus Original species Primer sequences (5’-3’) * No. of GenBank Publication name alleles accession Blm03 -- fwd 7 AY766437 Guay & Mulder TTgggATAggTAAAgggAAgg 2005 rev TTAAAgggCAgACTgATgAgg Blm06 -- fwd 12 AY766440 -- ggATgCAgAgAAAACAgTgC rev gACTTgCTgAAgCAAATgACC Blm10 -- fwd 14 AY766444 -- CAAAgTATATCTTCTCAgggACACg rev TgCATTgCTgTgAAgAgACC CRG N/A fwd N/A N/A used in Pearce gTAggCAAAgCAAgTCTgAAgTT et al. 2009b rev gCAACCACCAgCAgTCACTACAA Hhiu5 Histrionicus fwd 3 AF025903 Buchholz et al. histrionicus CTCTCCTTTTACTACAAATTCCCTT 1998 rev ATAAAggTAggTgACCCAATCCT Mm01 Mergus merganser fwd 7 AY679118 Gautschi & merganser ggTgTCCACAAAAggTACgg Koller 2005 rev CACAAgCATggCTCAgAgg Mm04 -- fwd 8 AY679121 -- CAggCTCAATgAggACAgg rev gCATCACCCTCCgTTTgg

Table 3.2 – A detailed layout of the two thermocycler regimes used throughout this study. TA = annealing temperature which is specific to each primer.

Regime name Cycle no. Temperature (ºC) Time (mm:ss) Repeats

1 94 02:00 1

94 00:15

TC-1 2 TA 00:15 40

72 00:30

3 72 30:00 1

1 95 02:00 1

94 00:20

TC-2 2 TA 00:30 40

72 01:30

Table 3.3 –The ideal conditions found for each primer pair. The total volume for each PCR was 25µl. TA = annealing temperature (ºC) which is specific to each locus.

Primer Thermocycler mgcl2 DNA TA (ºC) Approximate

name regime (mM) (ng) size (bp)

Aph04 TC-1 2 100 57 200

Aph08 TC-1 2 100 57 or 58 120

Aph13 TC-1 2 100 55 190

Aph15 TC-1 2 100 55 180

Aph20 TC-1 2 100 50 150

Aph24 TC-1 2 100 50 150

Blm06 TC-2 2 100 50 or 51 450

Blm10 TC-2 2 100 55 250

CRG TC-2 1.5 100 55 190

Hhiu5 TC-1 2 100 55 150

MM01 TC-2 2 100 58 140

MM04 TC-2 1.5 125 55-60 110

Table 3.4 – The total number of alleles, observed heterozygosity (HO) and expected heterozygosity (HE) for each locus.

Primer name Number of HO HE Actual fragment

alleles length (bp)

Aph08 5 0.61 0.64 111-123

Aph13 6 0.67 0.80 179-190

Aph20 4 0.50 0.48 142-147

Mm04 13 0.41 0.82 85-103

CONNECTING STATEMENT (3)

In the previous two chapters, we reported results obtained from having: (a) extracted red-breasted merganser genomic DNA originating from blood and feather samples and (b) screened 14 heterologous microsatellite primers. In the following chapter, we utilize the 46 genetic samples obtained (Chapter 2) in conjunction with the four fluorescent-labeled microsatellites (Chapter 3) to characterize the spatial patterns of relatedness across a colony of females nesting on the Tern Islands of Kouchibouguac National Park.

CHAPTER 4: SPATIAL-GENETIC STRUCTURING IN A RED-BREASTED

MERGANSER (MERGUS SERRATOR) COLONY IN KOUCHIBOUGUAC

NATIONAL PARK, NEW BRUNSWICK

4.1) Abstract

The clustering of kin is widespread across the animal kingdom and there are various mechanisms responsible for its emergence. Using four polymorphic microsatellites, we set out to determine whether genetic-spatial organization is present within a colony of nesting female red-breasted mergansers (Mergus serrator). Additionally, using nesting data, we explore possibilities for the existence of kin associations and/or cooperative interactions between these individuals, in the form of the synchronization of nesting activities. Results reported include: (1) the detection of broad-scale genetic structuring over the entire colony, as females nesting on separate islands were to some extent genetically distinct; (2) the detection of weak, yet significant, positive spatial autocorrelation of kin at the fine-scale in the more densely-populated areas of this colony; and (3) the synchrony of nest-initiation and incubation-initiation among proximally nesting females, apart from factors of relatedness. These results have important implications with respect to both the social dynamics and the tentative existence of a cooperative breeding strategy within this colony of breeding mergansers.

4.2) Introduction

Certain settling decisions result in the formation of clusters of kin in space.

Two primary mechanisms underlying the formation of these associations in adult kin are: (1) philopatric tendencies; and (2) actively maintained kin associations

(van der Jeugd et al. 2002, Sonsthagen et al. 2010). In the former, kin clustering is merely a bi-product of the common preference of related individuals to settle in proximity to their natal territory (Greenwood 1980). This drive allows individuals to maximize advantages associated with site familiarity (Wright 1943, Rathbun

1979, Beletsky & Orians 1991, see Chapter 1.1.2 for review). Conversely, kin associations are directed efforts by animals to surround themselves with other genetically similar individuals. By doing so, the emergence of within-group cooperative and altruistic behaviours is facilitated (Reyer 1984, McAllister &

Roitberg 1987) and opportunities for kin selection are enhanced (O’hara &

Blaustein 1981, Nituch et al. 2008). The two phenomena are by no means mutually-exclusive and it is even suggested that natal philopatry is a precursor to the emergence of kin associations (Greenwood 1980).

In recent years, due to the increased accessibility of genetic markers, the spatial structuring of kin (i.e. genetic-spatial autocorrelation) has been shown to be widespread across the animal kingdom; from mammals (Girman et al. 1997,

Taylor et al. 1997, Coltman et al. 2003), to arthropods (Burgman & Williams

1995, Loeb et al. 2000, Uesugi et al. 2009), to birds (Piertney et al. 1998, Shorey et al. 2000). Female-biased philopatry (Cooke et al. 1975, Doty & Lee 1974) and

78 active kin associations (Andersson & Alhund 2000, Nielsen et al. 2006) have been reported in many species of waterfowl and have both been invoked as plausible mechanisms responsible for genetic-spatial autocorrelation observed in certain populations (van der Jeugd et al. 2002, McKinnon et al. 2006, Waldeck et al. 2008, Sonsthagen et al. 2010).

The red-breasted merganser (Mergus serrator) is a medium-size sea duck that breeds across the Holarctic range. Aside from a breeding population of red- breasted mergansers on several small barrier islands at Kouchibouguac National

Park, New Brunswick that has been studied intensively, many aspects of the species’ nesting ecology remain unknown (Titman 1999). Strong philopatric tendencies of breeding females in this population have been well-documented through long-term banding studies (R. D. Titman, pers. comm., see appendix of

Craik 2009). While it is therefore expected that at least some genetic-structuring exists, the extent and scale at which this occurs is currently unknown.

Furthermore, the potential presence of kin association among nesting females remains untested. Red-breasted mergansers nesting colonially or semi-colonially on small islands provide a very interesting case for the study of genetic structure and genetic flux (e.g., among and between islands). Among waterfowl, this sort of opportunity can rarely be found outside of eiders and red-breasted mergansers.

Our working hypothesis is that spatial-genetic organization exists among nesting hens on the Tern Islands of Kouchibouguac National Park. The prediction, based on the use of four polymorphic microsatellites, is that as the distance between the nests of neighbouring females increases, the genetic relatedness

79 among them decreases. Additionally, another objective is to establish a framework for determining the venues, if any, for kin association and cooperation within this colony.

4.3) Methods

Field Collection

The study population of red-breasted mergansers nested in Kouchibouguac

National Park on a series of small barrier islands: Tern Island A (TI-A) and Tern

Island B (TI-B) (Figure 4.1). Field data were collected during June and July 2008.

Systematic searches were conducted to locate as many nests as possible. Red- breasted merganser nests are scraped on the ground and lined with a mixture of live and dead vegetation and plucked feathers (Bent 1962, Titman 1999) and are primarily situated in marram grass (Ammophila breviligulata) (Craik & Titman

2009). The coordinates of each nest were recorded using a global positioning system (GPS model eTrex, Garmin Ltd.). The incubation stage of each nest was assessed by floating eggs as described by Westerskov (1950) and final clutch sizes were counted. Efforts to trap females commenced only in the final quarter of the incubation period to minimize the risk of nest abandonment. Females were captured using nest traps (Weller 1957). From each captured bird, a 0.25cc blood sample was collected from the ulnar vein and deposited in a heparinized

Vacutainer® (BD). Each Vacutainer was labeled accordingly and eventually stored at -20ºC. Once a nest was deemed inactive, either due to success in hatching, predation or abandonment, as many adult female red-breasted

80 merganser contour feathers as possible (usually between 4-8) were recovered from the bowl, placed in a small envelope and stored at 4ºC.

Lab Protocol

DNA extraction protocols for both blood and feathers were initiated approximately six months after they had been collected. Further detail about protocols used is given in Chapter 2.3. Polymerase chain reactions (PCR) were carried out using fluorescent-labeled primers (Applied Biosystems). The four primer pairs and corresponding fluorescent-labels were: Aph08-FAM, Aph13-VIC,

Aph20-PET and Mm04-NED. PCRs were first carried out using a PTC-200 thermocycler (MJ research Inc.) and the products obtained were subsequently sent to Genome Quebec, McGill University where the sizes were resolved using an

ABI-3730 DNA Analyzer (Applied Biosystems). Primers and thermal conditions used in the reaction are described in Chapter 3.3.

Analysis of genetic diversity

Allelic frequencies, Hardy-Weinberg equilibriums (HWE) (using Markov chain method with default parameters) and linkage disequilibrium were calculated using Genepop 3.1 (Raymond & Rousset 1995).

Although there are several different genetic marker-based coefficients that can be used to estimate kinship (see van de Casteele et al. 2001 and Blouin 2003 for review); we used the method described by Queller & Goodnight (1989).

Queller & Goodnight’s coefficient of relatedness (R) is a pairwise estimate of

kinship obtained by first weighting alleles by their respective frequencies so that

the rarer an allele, the greater its weight. The coefficient ranges between -1 and +1;

where R is expected to be zero between unrelated individuals and 0.5 in full-

siblings and parents, assuming the population is in HWE. The coefficient of

relatedness of individual x to individual y can be defined as:

n m P(y)  P i1 j1 i, j i, j R(x,y)asymmetric  n m (1) P(x)  P i1 j1 i, j i, j

where n is the total number of loci, m is total number of allelic positions (i.e. 2 for

 th diploid organisms), Pi, j is the population frequency of the allele at the i locus

th and the j allelic position in individual x, P(x)i, j is the frequency of that allele

within individual x (i.e. either 0.5 for heterozygotes or 1 for homozygotes), and

P(y)i, j is the frequency of that allele in individual y (i.e. either 0, 0.5 or 1). This

index is asymmetrical as R(x,y)  R(y,x) and therefore in order to obtain symmetrical  pairwise coefficients, the numerator and denominator values from eq. (1) for R(x,y)  and R(y,x) were summed prior to division. A rarefaction curve was generated in

order to assess the consistency of these estimates of relatedness. This was done by:

 (1) randomly selecting one of the four microsatellites; (2) calculating relatedness

based solely on that marker; (3) randomly selecting another marker without

replacement; (4) re-calculating relatedness, only now based on both markers; (5)

calculating the difference between the estimates of relatedness from step 2 and 4;

and (6) repeating until all markers have been selected. This algorithm was

repeated 9,999 times and the mean differences between loci (computed in step 5)

were calculated. Calculations using this algorithm were performed using RE-RAT

(http://people.musc.edu/~schwaclh/).

Analysis of spatial-genetic structure

Nest densities on each island were calculated by dividing the number of

nests on an island by its total area (m2), which was based on the perimeter. The

total area of each island was computed using ArcGIS (ESRI Inc.). Euclidean

distances were calculated between nests using geographic coordinates (UTM).

The Euclidean distance between two points is defined by:

2 2 D(a,b)  (xa  xb )  (ya  yb ) (2)

where (xa,ya ) are the x- and y-coordinates of nest a, and (xb,yb ) are the x- and  y-coordinates of nest b , respectfully. The multivariate approach described by    Smouse & Peakall (1999) was employed as a measure of genetic distance between

nesting females. At a given microsatellite locus with four alleles (A,B,C,D), the

distance between two individuals is defined as: d(AA,AA)  0; d(AB,AB)  0;  d(AA,AB) 1; d(AB,AC) 1; d(AB,CD)  2 ; d(AA,BC)  3 ; d(AA,BB)  4 .   Pairwise distances were first calculated for each locus separately and then      summed across all loci.

Using the matrix of genetic distances calculated above, a principal

coordinate analysis (PCoA) was used to assess patterns of genetic variation across

the islands. An ordination diagram was produced by plotting the first two

eigenvectors against each other. Additionally, a single-linkage cluster analysis of

the first two principal coordinates was executed and superimposed onto the

83 ordination diagram. By doing so, the distances between objects could be further resolved.

The detection of genetic-spatial autocorrelation on each island was achieved using a Mantel correlogram (Legendre & Legendre 1998). The Mantel correlation coefficient (r) behaves similar to other correlation coefficients and ranges from -1 to +1. Positive coefficients indicate the presence of clustering while negative correlations indicate dispersion among objects. This technique partitions geographic and genetic distance matrices into sub-distance classes and calculates the Mantel correlation coefficient (r) for each pair of corresponding sub-matrices thereby facilitating the detection of non-linear trends in the data

(Burgman & Williams 1995, Skabo et al. 1998). The distance classes were determined using Sturges’ equation (Sturges 1926), which ensures an even distribution of observations in each bin (i.e. sub-distance class). To remove the dependency on normality, all tests were based on the Spearman rank statistic. The significance of r was determined by comparing the reference value to a distribution of 9,999 values which were generated from random permutations of the data. The null hypothesis was that the associations described in the real dataset are as likely to be found in randomly generated data. Holm’s (1979) correction for multiple testing was applied and a critical value of α = 0.05 was used to reject the null hypothesis.

As an additional approach toward understanding the dynamics between neighbouring females, the mean relatedness between a female and her three nearest neighbours was calculated and compared to the background mean

84 relatedness using the Kruskal-Wallis rank-sum test (Siegel & Castellan, 1988).

The null hypothesis was that there is no difference between the mean relatedness of the three nearest neighbours vs. that of the background. Again, a critical value of α = 0.05 was used to reject the null hypothesis.

Nest-initiation and incubation synchrony

The mean incubation-initiation date of each island was compared using the

Kruskal-Wallis rank-sum test (Siegel & Castellan, 1988). The null hypothesis was that the mean incubation-initiation date was equal across the islands.

Mantel correlograms were used to assess the degree of correlation between:

(a) nesting synchrony and geographic distance (eq. 2); and (b) nesting synchrony and genetic distance (Smouse & Peakall 1999). Because these tests require the comparison of two distance matrices, Euclidean distances between nesting females, on the basis of their incubation-initiation dates, were computed, also using eq. 2. In order to remove the dependency on normality, the tests were based on the Spearman rank correlation coefficient and the significance of the test statistic was again assessed using 9,999 random permutations, Holm’s corrections and a critical value of α = 0.05.

In order to assess patterns of island colonization, the above analyses could have been re-executed using dates of ‘nest-initiation’ in the place of ‘incubation- initiation.’ However before this was carried out, the degree of shared explanatory power between these two variables was assessed by first regressing one against the other and then calculating the coefficient of determination (R2). Because of

85 sufficient overlap between the two, only the latter is presented. The nest-initiation dates were approximated by subtracting 1.5 d × total number eggs from the estimated incubation-initiation date (Titman 1999); but only back-dated to a maximum of 18 d (1.5 × 12 eggs) under the assumption that nests with ≥ 13 eggs had been parasitized (Craik & Titman 2009).

Unless otherwise specified, all calculations were carried out using the open-source statistical software R ver. 2.10.1 (http://www.r-project.org/); Mantel tests and correlograms were computed using the package VEGAN ver. 1.17-2.

4.4) Results

Field and lab protocol

Altogether 88 nests were located and genetic material was obtained from

60. Of the 42 DNA samples originating from blood, 39 (93%) were successfully genotyped. In contrast, only seven of the 18 (39%) DNA samples from feathers were successfully amplified. Therefore, a total of 46 individuals were genotyped across all four microsatellite loci and subsequently utilized in the genetic-spatial analysis presented below. A schematic diagram of the Tern Islands as well as the spatial distribution of discovered and sampled nests is presented in Figure 4.2.

Analysis of genetic diversity

The mean observed heterozygosity (HO) across four loci was 0.61. The genotype frequencies of two of the loci used, deviated significantly from that expected under Hardy-Weinberg equilibrium (see Chapter 3.5). The Queller &

Goodnight coefficients of relatedness were calculated for each pair within the colony and the mean estimate of relatedness was R = -0.023 (range = -0.584 to

1.000). Rarefaction analysis showed that with each additional locus, both the  mean difference and variance of estimated relatedness decreased dramatically.

However, it is clear that using only these four loci does not eliminate inconsistencies in the estimates produced and based on Figure 4.3, it is expected that by including additional loci into the analysis, the coefficients estimated would still be affected, albeit not by large amounts.

Analysis of spatial-genetic structure

The decision to include feather samples in the analysis was not evident.

However, the rationale for their inclusion was that a proxy of relatedness is more important to the nature of this study than obtaining accurately-resolved relationships. Therefore, provided that the genotyped feather-DNA corresponded to the true genotypes (see Chapter 2.5), their inclusion should contribute to the detection of spatial-genetic trends. Similarly, there was ambiguity surrounding the use of the two microsatellite loci that were not in Hardy-Weinberg equilibrium; especially in reference to Mm04 which was found to be in homozygous excess.

While the presence of null alleles can obscure the analysis, it is not evident that this is the case. However based on both the above and results from the rarefaction analysis, it is clear that the inclusion of additional loci is still warranted in order to strengthen the analysis.

The density on TI-A and TI-B was 0.005 and 0.002 nests/m2, respectively

(Figure 4.2).

The cumulative proportion of variance in the genetic data accounted for by the first three principal coordinates is 0.572 (0.267, 0.173, and 0.132). In an ordination diagram (Figure 4.4) clear patterns of genetic-structuring in relation to a female’s island of origin can be identified. Proximities between points (i.e. nesting females) in this diagram are approximations of their genetic distances; the closer together in ordination space, the more genetically similar they are. In the same diagram, the connections between objects, as determined by single linkage cluster analysis, provides an additional measure of distance.

Weak, yet significant, genetic-spatial autocorrelation was detected on TI-

A (Table 4.1). Specifically, females nesting within 0-20 m from one another on that island were more related to each other than what is expected by chance (r =

0.127, N = 64, corrected p-value = 0.040). No significant correlations, at any of the distance intervals tested, were found on TI-B (Figure 4.5).

In Figure 4.6, the mean relatedness of females and their three nearest neighbours vs. the mean background level of relatedness is presented for each island. These findings reinforce those above and further demonstrate that higher relatedness between neighbouring females on TI-A are contributing factors to the autocorrelation observed (Kruskal-Wallis χ2 = 4.608, df = 1, p-value = 0.032)

(Table 4.2).

Nest-initiation and incubation synchrony

Over the entire colony, the mean nest-initiation date was 7 June (range =

15 May – 15 July) and, on average, incubation began approximately 12 d later; no significant differences in either of these variables were found across the islands

(Table 4.3).

Significant, positive Mantel correlation coefficients were found on both

TI-A and TI-B. In the former, incubation-initiation synchrony was revealed between females nesting at a distance of 10-20 m from each other (Mantel r =

0.176, N = 96, p-value = 0.024); while in the latter, such trends were found between those within a 0-30 m distance interval (i.e. consisting of three separate intervals of 10 m) (Table 4.1.b). The graphical depictions of the correlograms are presented in Figure 4.5.b.

Unlike the comparison with geographic distance, no spatial structuring between genetic distance and incubation-initiation date was detected across any of the intervals tested, except for a marginally significant negative correlation found between females whose genetic distance was between 0-2 (Mantel r = -0.085, N =

16, p-value = 0.094). The mean genetic distance was 6.367 and it ranged from

0.000-13.000.

When the above analysis is repeated using ‘nest-initiation’ date instead of

‘incubation-initiation’ date, virtually identical results were obtained. Upon inspection, the proportion of variance (R2) in ‘nest-initiation’ date accounted for by ‘incubation-initiation’ date was 0.896 implying that either variable is an appropriate predictor of the other (b = 1.317, t = 25.290, p-value = <0.000). The

89 high proportion of shared information between the two variables causes them to behave similarly.

4.5) Discussion

Considering that red-breasted mergansers are not consistently colonial

(Titman 1999), an important question is: what is it about this particular situation that has resulted in colonial nesting? Two factors have been proposed by Schmutz et al. (1983) to explain the emergence of this behaviour in waterfowl: (1) the clustering of suitable habitat; and (2) benefits gained by nesting females from cooperative interactions. Habitat quality and the availability of nesting resources have been implicated as underlying factors behind observed settling patterns and the formation of groups (Schjorring et al. 2000, Schjorring 2001, Fowler et al.

2004). Although small islands are valuable to many species of ground-nesting waterfowl (Clark & Shutler 1999), the Tern Island complex, at least from a vegetative perspective, is not especially unique (S. R. Craik, pers. comm.) and therefore cannot be the sole factor explaining the emergence of coloniality in this system. The red-breasted mergansers in Kouchibouguac appear to nest in association with larids as they benefit from both their alarm calls and aggressive efforts to ward off predators (Young & Titman 1986). It is therefore possible that the ≈5,000 pairs of nesting common terns (Sterna hirundo) on the Tern Islands are a major explanatory factor behind the group formation.

On the other hand, in reference to the second factor addressed by Schmutz et al., the existence of cooperative interactions within this colony, between kin or

90 otherwise, has not been explicitly proven to exist until the results presented here.

Synchrony of incubation was found between closely-nesting females throughout the colony (Figure 4.5.b). Nesting in groups has several advantages including benefits from: (a) the vigilance and predator-detection abilities of group mates; (b) the defensive actions taken against predators by group mates; and (c) predator satiation from the offspring of group mates (Schmutz et al. 1983, McKinnon et al.

2006). Synchronizing breeding with neighboring females essentially maximizes the period of nesting overlap and therefore simultaneously maximizing the period of time when such advantages are obtained (Schmutz et al. 1983). The fact that this trend was no longer detected if genetic distances were analyzed in conjunction with incubation-initiation (Figure 4.5.c), implies that this is a strategy pursued by females regardless of genetic relatedness.

Some degree of fine-scale spatial autocorrelation of kin was also revealed in this colony, albeit at low frequencies (Figure 4.5.a). Two potential factors that may interfere with this are: (1) constraints on nest site selection and (2) mechanisms of recognition. Where an animal nests has important fitness consequences (Clark & Shutler 1999) resulting from access to resources, exposure to predation and exposure to parasitism. Craik & Titman (2009) demonstrated that vegetation type and structure has a strong influence on nest-site selection within this colony of mergansers. Furthermore, if kin recognition is based on familiarity as opposed to phenotype matching (i.e. true recognition) (e.g. van der Jeugd et al.

2002), the high rate of brood parasitism described in this colony, provided that donors are not necessarily sisters or cousins, (Young & Titman 1988) may

91 weaken the signal of genetic-spatial autocorrelation since not all of one’s nest- mates are kin.

Local scale genetic-spatial structures do not necessarily constitute evidence of kin associations, as they are one of three phenomena attributed to the clustering of kin at the local scale, the others being high levels of intra-colonial relatedness and extreme philopatric tendencies.

Genetic-spatial autocorrelation at a fine-scale can occur randomly if there is a high level of background relatedness (Fowler et al. 2004). For example, this was believed to have occurred among greater white-fronted geese (Anser albifrons frontalis) where nests of kin are occasionally found in tight clusters

(Fowler et al. 2004). High intra-colonial relatedness can result from a high proportion of females with philopatric tendencies (Greenwood 1980, Ratnayeke et al. 2002, Pearce 2007). Due to the permutation-based statistical methods employed in our study, the assessments made had effectively accounted for the likelihood of observing a given pattern of the spatial distribution of nests, accounting for background levels of relatedness. Consequently, the possibility that the pattern observed is an artifact of high levels of intra-colonial relatedness can be largely discounted.

Second, extreme natal philopatry may result in similar patterns of genetic- spatial autocorrelation at a local scale. Philopatry is regarded as a strategy employed by certain individuals to maximize the benefits of site familiarity

(Greenwood 1980) and philopatric tendencies have consequences on the degree of spatial-genetic association observed (van der Jeugd et al. 2002, Sonsthagen et al.

2010). For instance, first-order relatives (i.e. mother-daughter or sister-sister) using the same nest bowl may result in a stronger, more acute signal of genetic- spatial autocorrelation than females exhibiting fidelity to a given region.

Sonsthagen et al. (2010) were able to disqualify the possibility that extreme philopatry was the mechanism underlying the local-scale clustering of kin in colonies of common eiders (Somateria mollissima v-nigrum). They reasoned that because of the seasonal instability of nests caused by the perpetual movements of driftwood, the likelihood that females exhibit fidelity to a particular bowl is low.

Instead, the patterns of association they observed could be more likely attributed to the presence of kin associations (Sonsthagen et al. 2010). Our study revealed a significant degree of genetic-structuring among females related to where they nested, despite close proximity of the islands (Figure 4.4). This suggests that certain females, along with their brood mates, prefer to nest on their islands of origin. Exhibiting fidelity to one’s natal island may be rule-based mechanism to maximize the likelihood of choosing good quality nesting and brood-rearing habitat (Schjorring 2001). In several species, juvenile experience can serve as a basis for future settling decisions (Brown & Brown 1992, Osorio-beristain &

Drummond 1993, Schjorring et al. 2000). However, our interpretation of the extent of the philopatric behaviour is muted by the fact that significant genetic- spatial autocorrelation was only detected on TI-A. This is contrary to what is expected if every female has the same tendency to return to a natal site. Therefore the patterns observed are unlikely to be solely the result of extreme philopatry.

Kin associations

Interaction between kin can result in numerous benefits including increased survivorship (Lambin & Krebs 1993), recruitment (Stoen et al. 2005); and reproductive success (Andersson & Alhund 2000, Nielsen et al. 2006). Active kin associations require that the individuals involved bear some form of recognition capability (Waldman 1988, Andersson & Alhund 2000). Although female red-breasted mergansers are known to nest independently (Titman 1999), feasible forms of kin association within this colony include: (1) lowered aggressive tendencies towards neighbouring kin; (2) pre-hatch brood amalgamation (pre-HBA); (3) post-hatch brood amalgamation (post-HBA); and (4) cooperative defense.

Reduced agonism between kin can be important in the recruitment of related individuals into a given area (MacColl et al. 2000, Hoglund & Shorey

2003), especially under high-density conditions where the frequencies of interaction between individuals are high (Sonsthagen et al. 2010). The prospect that females on the Tern Islands exhibit increased tolerance toward related females is an intriguing prospect and with some degree of indirect support.

Specifically, on TI-A which had the highest density of nests, non-random clusters of kin in the 0-20 m distance interval, were found. These findings were further qualified by evidence that relatedness between focal females and their three nearest neighbours was on average greater than the level of background relatedness. Conversely, clustering at local scales was not detected in TI-B, which was less densely populated. Increased tolerance by females toward kin attempting

94 to establish adjacent nests is a plausible mechanism behind the spatial autocorrelation observed on TI-A. This line of conjecture (i.e. facilitated recruitment) is also consistent with the finding that nesting synchrony between females on TI-A in the 0-10m distance interval (Table 4.1.b) is absent; thereby indicating that there was more than one wave of colonization into a given area.

Nesting territories should be more difficult to acquire during subsequent colonization events, and thus relaxed aggressiveness towards late-coming kin may aid their recruitment success.

It has been demonstrated that the greater the proximity between neighbours, the greater their vulnerability to thievery (Wojcieszek et al. 2007), cannibalism (Brown 1967, Yom-Tov 1974 and references therein), and brood parasitism (Reyer 1984, McRae 1998); presumably a function of increased nest access. On the Tern Islands intra-specific brood parasitism has occurred at some of the highest rates known among ground-nesting waterfowl (Young & Titman

1988). Aggression towards foreign females is expected if brood parasitism has a negative effect on host fitness (e.g. Milonoff et al. 2004) and under such circumstances, both recognition and discriminatory behaviour toward kin would serve as adaptive traits (Andersson 2001, Lopez-Sepulcre & Kokko 2002). For example, by helping kin to obtain a nesting territory, the costs associated with parasitism are offset by inclusive fitness obtained from either (a) the offspring which are reared directly by the recruited individual or (b) as a result of increased level of relatedness to parasitic eggs. While a higher level of relatedness between donors and recipients has been described in some populations of goldeneyes

(Bucephala clangula) (Andersson & Alhund 2000) and common eiders

(Andresson & Waldeck 2007, Waldeck et al. 2008), the relationship between the two in this colony of red-breasted mergansers has yet to be described.

The prospect of kin-based post-HBA in this colony has not been well- explored. Aside from a few accounts (e.g. Bukacinski et al. 2000, Kraaijeveld

2005), the evidence that a higher degree of relatedness exists between non- parental females and the young they are tending, has not been well-substantiated.

Instead, more evidence seems to implicate accidental fusion (Gorman & Milne

1972, Savard 1987) and abandonment (Poysa 1995, Ost 1999, Kilpi et al. 2001) as the two principal factors driving adoption and post-HBA. Ost et al. (2005) showed that female relatedness was not a factor influencing patterns of brood amalgamation. While this assertion cannot be made here, the possibility of relatedness being a factor appears unlikely considering the lack of incubation and colonization synchrony observed between kin (Table 4.1.c).

Our observations indicate the potential existence of cooperative defense and predator evasion, albeit not necessarily among kin. Even though payoffs from cooperative behaviours are augmented if they are directed toward relatives

(O’hara & Blaustein 1981), the lack of synchrony between relatedness and incubation date in this study indicates that factors other than relatedness of neighbours are integrated into the decision-making process. Consequently, if females are surrounded by a significant proportion of non-related neighbours, perhaps it is best to pursue this strategy apart from considerations of genetic relatedness.

In summary, our findings support the hypothesis that spatial-genetic organization exists between the females in this colony. In fact, based on this analysis, it appears that organization is present both at the regional (i.e. across islands) and fine-scale (i.e. between neighbouring nests); however the latter was only detected in the more densely-populated regions of the colony. It is possible that fine-scale genetic structuring is a product of an increased level of tolerance

(i.e. reduced aggression) among close relatives. Nest-initiation and incubation- initiation dates were synchronous among proximally nesting individuals, regardless of the degree of relatedness between them. The presence of such synchrony raises the possibility that at least some females are participating in a cooperative predator-defense strategy.

4.6) Literature Cited Andersson M (2001) Relatedness and the evolution of conspecific brood parasitism. American Naturalist 158, 599-614. Andersson M, Ahlund M (2000) Host-parasite relatedness shown by protein fingerprinting in a brood parasitic bird. Proceedings of the National Academy of Sciences of the United States of America 97, 13188-13193. Andersson M, Waldeck P (2007) Host-parasite kinship in a female-philopatric bird population: evidence from relatedness trend analysis. Molecular Ecology 16, 2797-2806. Beletsky LD, Orians GH (1991) Effects of breeding experience and familiarity on site fidelity in female red-winged blackbirds. Ecology 72, 787-796. Bent AC (1962) Life Histories of North American Wild Fowl Dover Publications, New York. Blouin MS (2003) DNA-based methods for pedigree reconstruction and kinship analysis in natural populations. Trends in Ecology & Evolution 18, 503- 511. Brown CR, Brown MB (1992) Ectoparasitism as a cause of natal dispersal in cliff swallows. Ecology 73, 1718-1723. Brown RGB (1967) Breeding success and population growth in a colony of herring and lesser black-backed gulls Larus argentatus and L fuscus. Ibis 109, 502-515. Bukacinski D, Bukacinska M, Lubjuhn T (2000) Adoption of chicks and the level of relatedness in common gull, Larus canus, colonies: DNA fingerprinting analyses. Animal Behaviour 59, 289-299. Burgman MA, Williams MR (1995) Analysis of the spatial pattern of arthropod fauna of jarrah (Eucalyptus marginata) foliage using a Mantel correlogram. Australian Journal of Ecology 20, 455-457. Clark RG, Shutler D (1999) Avian habitat selection: Pattern from process in nest- site use by ducks? Ecology 80, 272-287. Coltman DW, Pilkington JG, Pemberton JM (2003) Fine-scale genetic structure in a free-living ungulate population. Molecular Ecology 12, 733-742.

Cooke F, Macinnes CD, Prevett JP (1975) Gene Flow between Breeding Populations of Lesser Snow Geese. Auk 92, 493-510. Craik SR (2009) Habitat use by breeding and molting red-breasted mergansers in the Gulf of St. Lawrence Ph.D. thesis, McGill University. Craik SR, Titman RD (2009) Nesting ecology of red-breasted mergansers in a common tern colony in Eastern New Brunswick. Waterbirds 32, 282-292. Doty HA, Lee FB (1974) Homing to nest baskets by wild female mallards. Journal of Wildlife Management 38, 714-719. Fowler AC, Eadie JM, Ely CR (2004) Relatedness and nesting dispersion within breeding populations of Greater White-fronted Geese. Condor 106, 600- 607. Girman DJ, Mills MGL, Geffen E, Wayne RK (1997) A molecular genetic analysis of social structure, dispersal, and interpack relationships of the African wild dog (Lycaon pictus). Behavioral Ecology and Sociobiology 40, 187-198. Gorman ML, Milne H (1972) Creche behaviour in the common eider Somateria m. mollissima L. Ornis Scandinavica 3, 21-26. Greenwood PJ (1980) Mating systems, philopatry and dispersal in birds and mammals. Animal Behaviour 28, 1140-1162. Hoglund J, Shorey L (2003) Local genetic structure in a white-bearded manakin population. Molecular Ecology 12, 2457-2463. Holm S (1979) A simple sequentially rejective multiple test procedure. Scandinavian Journal of Statistics 6, 65-70. Kilpi M, Ost M, Lindstrom K, Rita H (2001) Female characteristics and parental care mode in the creching system of eiders, Somateria mollissima. Animal Behaviour 62, 527-534. Kraaijeveld K (2005) Black swans Cygnus atratus adopt related cygnets. Ardea 93, 163-169. Lambin X, Krebs CJ (1993) Influence of female relatedness on the demography of Townsends vole populations in spring. Journal of Animal Ecology 62, 536-550.

Legendre P, Legendre L (1998) Numerical Ecology Elsevier, Amsterdam; New York. Loeb MLG, Diener LM, Pfennig DW (2000) Egg-dumping lace bugs preferentially oviposit with kin. Animal Behaviour 59, 379-383. Lopez-Sepulcre A, Kokko H (2002) The role of kin recognition in the evolution of conspecific brood parasitism. Animal Behaviour 64, 215-222. MacColl ADC, Piertney SB, Moss R, Lambin X (2000) Spatial arrangement of kin affects recruitment success in young male red grouse. Oikos 90, 261- 270. McAllister MK, Roitberg BD (1987) Adaptive suicidal-behavior in pea aphids. Nature 328, 797-799. McKinnon L, Gilchrist HG, Scribner KT (2006) Genetic evidence for kin-based female social structure in common eiders (Somateria mollissima). Behavioral Ecology 17, 614-621. McRae SB (1998) Relative reproductive success of female moorhens using conditional strategies of brood parasitism and parental care. Behavioral Ecology 9, 93-100. Milonoff M, Poysa H, Runko P, Ruusila V (2004) Brood rearing costs affect future reproduction in the precocial common goldeneye Bucephala clangula. Journal of Avian Biology 35, 344-351. Nielsen CR, Semel B, Sherman PW, Westneat DF, Parker PG (2006) Host- parasite relatedness in wood ducks: patterns of kinship and parasite success. Behavioral Ecology 17, 491-496. Nituch LA, Schaefer JA, Maxwell CD (2008) Fine-scale spatial organization reflects genetic structure in sheep. Ethology 114, 711-717. O'hara RK, Blaustein AR (1981) An investigation of sibling recognition in Rana cascadae Tadpoles. Animal Behaviour 29, 1121-1126. Osorio-beristain M, Drummond H (1993) Natal dispersal and deferred breeding in the blue-footed booby. Auk 110, 234-239. Ost M (1999) Within-season and between-year variation in the structure of common eider broods. Condor 101, 598-606.

100

Ost M, Vitikainen E, Waldeck P, et al. (2005) Eider females form non-kin brood- rearing coalitions. Molecular Ecology 14, 3903-3908. Pearce JM (2007) Philopatry: a return to origins. Auk 124, 1085-1087. Piertney SB, MacColl ADC, Bacon PJ, Dallas JF (1998) Local genetic structure in red grouse (Lagopus lagopus scoticus): evidence from microsatellite DNA markers. Molecular Ecology 7, 1645-1654. Poysa H (1995) Factors affecting abandonment and adoption of young in common eiders and other waterfowl - a comment. Canadian Journal of Zoology- Revue Canadienne De Zoologie 73, 1575-1577. Queller DC, Goodnight KF (1989) Estimating relatedness using genetic-markers. Evolution 43, 258-275. Rathbun GB (1979) The social structure and ecology of elephant-shrews Parey, Berlin ; Hamburg. Ratnayeke S, Tuskan GA, Pelton MR (2002) Genetic relatedness and female spatial organization in a solitary carnivore, the raccoon, Procyon lotor. Molecular Ecology 11, 1115-1124. Raymond M, Rousset F (1995) Genepop (Version-1.2) - population-genetics software for exact tests and ecumenicism. Journal of Heredity 86, 248-249. Reyer HU (1984) Investment and relatedness - a cost-benefit-analysis of breeding and helping in the pied kingfisher (Ceryle rudis). Animal Behaviour 32, 1163-1178. Savard JPL (1987) Causes and functions of brood amalgamation in Barrows goldeneye and bufflehead. Canadian Journal of Zoology-Revue Canadienne De Zoologie 65, 1548-1553. Schjorring S (2001) Ecologically determined natal philopatry within a colony of great cormorants. Behavioral Ecology 12, 287-294. Schjorring S, Gregersen J, Bregnballe T (2000) Sex difference in criteria determining fidelity towards breeding sites in the great cormorant. Journal of Animal Ecology 69, 214-223.

101

Schmutz JK, Robertson RJ, Cooke F (1983) Colonial nesting of the Hudson-Bay eider duck. Canadian Journal of Zoology-Revue Canadienne De Zoologie 61, 2424-2433. Shorey L, Piertney S, Stone J, Hoglund J (2000) Fine-scale genetic structuring on Manacus manacus leks. Nature 408, 352-353. Siegel S, Castellan NJ (1988) Nonparametric statistics for the behavioral sciences McGraw-Hill, New York. Skabo S, Vaillancourt RE, Potts BM (1998) Fine-scale genetic structure of Eucalyptus globulus ssp. globulus forest revealed by RAPDs. Australian Journal of Botany 46, 583-594. Smouse PE, Peakall R (1999) Spatial autocorrelation analysis of individual multiallele and multilocus genetic structure. Heredity 82, 561-573. Sonsthagen SA, Talbot SL, Lanctot RB, McCracken KG (2010) Do common eiders nest in kin groups? Microgeographic genetic structure in a philopatric sea duck. Molecular Ecology 19, 647-657. Stoen OG, Bellemain E, Saebo S, Swenson JE (2005) Kin-related spatial structure in brown bears Ursus arctos. Behavioral Ecology and Sociobiology 59, 191-197. Sturges HA (1926) The choice of a class interval Case I Computations involving a single. Journal of the American Statistical Association 21, 65-66. Taylor AC, Horsup A, Johnson CN, Sunnucks P, Sherwin B (1997) Relatedness structure detected by microsatellite analysis and attempted pedigree reconstruction in an endangered marsupial, the northern hairy-nosed wombat Lasiorhinus krefftii. Molecular Ecology 6, 9-19. Titman RD (1999) Red-breasted Merganser (Mergus serrator). In: The Birds of North America Online. Cornell Lab of Ornithology, Ithaca. Uesugi R, Kunimoto Y, Osakabe M (2009) The fine-scale genetic structure of the two-spotted spider mite in a commercial greenhouse. Experimental and Applied Acarology 47, 99-109.

102

Van de Casteele T, Galbusera P, Matthysen E (2001) A comparison of microsatellite-based pairwise relatedness estimators. Molecular Ecology 10, 1539-1549. van der Jeugd HP, van der Veen IT, Larsson K (2002) Kin clustering in barnacle geese: familiarity or phenotype matching? Behavioral Ecology 13, 786- 790. Waldeck P, Andersson M, Kilpi M, Ost M (2008) Spatial relatedness and brood parasitism in a female-philopatric bird population. Behavioral Ecology 19, 67-73. Waldman B (1988) The ecology of kin recognition. Annual Review of Ecology and Systematics 19, 543-571. Weller MW (1957) An automatic nest trap for waterfowl. Journal of Wildlife Management 21, 456-458. Westerskov K (1950) Methods for determining the age of game bird eggs. Journal of Wildlife Management 14, 56-67. Wojcieszek JM, Nicholls JA, Goldizen AW (2007) Stealing behavior and the maintenance of a visual display in the satin bowerbird. Behavioral Ecology 18, 689-695. Wright S (1943) Isolation by distance. Genetics 28, 114-138. Yom-Tov Y (1974) Effect of food and predation on breeding density and success, clutch size and laying date of crow (Corvus-Corone L). Journal of Animal Ecology 43, 479-497. Young AD, Titman RD (1986) Costs and bBenefits to red-breasted mergansers nesting in tern and gull colonies. Canadian Journal of Zoology-Revue Canadienne De Zoologie 64, 2339-2343. Young AD, Titman RD (1988) Intraspecific nest parasitism in red-breasted mergansers. Canadian Journal of Zoology-Revue Canadienne De Zoologie 66, 2454-2458.

103

Figure 4.1 – Location of the Tern Islands of Kouchibouguac National Park, New Brunswick, Canada. The ESRI shapefiles were obtained from GeoBase (http://www.geobase.ca/)

104

Figure 4.2 – A schematic representation of the red-breasted mergansers nesting on the Tern Islands. The nest-densities of Tern Island A (left-hand side) and Tern Island B (right-hand side) was 0.005 and 0.002 nest/m2, respectfully. Each individual nest is represented by a point. Of the 88 nests discovered, a total of 46 (52.9%) were successfully genotyped (represented by solid circles).

105

Figure 4.3 – A rarefaction curve depicting the relationship between the number of loci used and the resultant Queller & Goodnight (1989) estimate of relatedness. Each point represents the mean difference between current estimate of relatedness and the one previous to it; the bars represent the standard deviation. A total of 9999 permutations were used to generated these data.

106

Figure 4.4 – Two-dimensional PCoA ordination diagram. The proportion of variance explained by the first two principal components is 0.267 and 0.173, respectively. In this figure, the positions of the objects (i.e. nesting females) in relation to one another are approximations of their genetic distances. The measure of genetic distance used in the PCoA was that described by Peakall & Smouse (1999). The results from single linkage cluster analysis are superimposed onto this ordination plot (represented by dotted-lines) as an additional means for assessing the associations between objects.

107

(a) Geographic distance vs. genetic distance (n=46)

(b) Geographic distance vs. incubation-initiation (n=75)

Figure 4.5 – The Mantel correlograms for TI-A and TI-B. Significant distance intervals are emphasized with a solid point in the diagram. A null correlation is represented by a solid red line in each panel. The corresponding figures to these panel are presented in Table 4.1. Significant values were determined by generated using 9,999 random permutations of the data and all p-values were corrected for multiple testing using Holm’s method. A critical value of α = 0.05 was used.

108

Figure 4.6 - Bar graphs displaying: (1) mean relatedness of females and their three nearest neighbours (left-hand bar); and (2) mean background level of relatedness on that island (right-hand bar). Additionally the 95% bootstrapped CI are shown. Estimates of relatedness were obtained using the methods described by Queller and Goodnight (1989).

109

Table 4.1 – The distance intervals, number of pairwise observations within a given interval, Mantel correlation coefficients (r) and associated p-values for the genetic and geographic distances being compared. The probabilities were generated based on 9,999 random permutations of the data and all p-values were corrected for multiple testing using Holm’s method. A critical value of α = 0.05 was required before rejecting the null hypothesis. Distance No. Mantel p-value

interval Observations correlation (r) (corr.)

(a) Geographic distance vs. genetic distance

[0 – 20[ 64 0.129 0.042* [20 – 37[ 84 0.053 0.228

Tern Tern [37 – 53[ 56 -0.074 0.348 Island A Island [53 – 69[ 64 -0.030 0.521

[0 – 32[ 92 -0.094 0.115 [32 – 62[ 70 -0.115 0.123

Tern Tern [62 – 93[ 100 0.122 0.185 Island B Island [93 – 123[ 92 0.071 0.246

(b) Geographic distance vs. incubation-initiation

[0 – 10[ 82 -0.006 0.473 [10 -20[ 96 0.103 0.024*

Tern Tern [20 – 30[ 100 0.067 0.116 Island A Island [30 – 40[ 142 -0.064 0.188

[0 – 10[ 78 0.091 0.050* [10 -20[ 36 0.150 0.001***

Tern Tern [20 – 30[ 54 0.132 0.001*** Island B Island [30 – 40[ 62 0.056 0.105

[0 – 2[ 16 0.089 0.110 [2 – 4[ 50 0.092 0.220

Tern Tern [4 – 6[ 138 0.067 0.330 Island A Island [6 – 8[ 130 -0.041 0.440

[0 – 2[ 16 -0.085 0.094 · [2 – 4[ 108 0.029 0.384

Tern Tern [4 – 6[ 178 -0.115 0.197 Island B Island [6 – 8[ 96 0.135 0.161 Signif. codes: [‘***’ 0.001 ‘**’0.01 ‘*’0.05 ‘.’ 0.1]

110

Table 4.2 The results from the Kruskal-Wallis rank sum test where the null hypothesis tested was that on a given island the mean relatedness of females and their three nearest neighbours (NN) was no different than the mean background level of relatedness. On Tern Island A, the null hypothesis was rejected however not on Tern Island B. The critical value used was α = 0.05.

Island no. 3 NN Background Kruskal-Wallis (χ2) df p-value TI-A 0.064 -0.045 4.608 1 0.032 * TI-B 0.023 0.057 0.267 1 0.601

111

Table 4.3 – Mean dates of colonization and incubation-initiation across the Tern Islands. The days were transformed into Julian days using May 1 as day 1. In both tests the null hypothesis, stating that the variable did not differ between islands, could not be rejected. A critical value of α = 0.05 was used.

Variable Mean TI-A Mean TI-B Kruskal- df p- name Wallis (χ2) value Nest-initiation date 36.25 38.78 0.719 1 0.396 Incubation- 49.63 51.23 0.405 1 0.524 initaition date

112

CONCLUDING STATEMENTS

Genetic materials, DNA quality and reliability:

It can be concluded that avian nucleated blood is a superior source of genetic material than nest-recovered contour feathers because: (1) the amount of genomic DNA obtained from the former is far larger; and (2) the proportion samples successfully amplified across all four microsatellite loci was greater in the former. Furthermore, in two of three trials, genetic discordance was observed between blood and feather samples collected from the same specimen; although they were closely matched. It is possible that some of the discrepancy in quality observed can be attributed to DNA degradation resulting from the manner in which the feather samples had been stored.

Genetic-spatial organization and the association of kin

Despite shortcomings, this study represents a rudimentary attempt to understand the scales at which genetic-spatial organization occurred. Broad-scale organization was detected across the colony, as birds nesting on the different islands were genetically distinct. Weak spatial autocorrelation at the local-scale was observed between kin nesting on the most densely-populated island. This suggests the existence of kin associations, specifically in the form of reduced aggression between related (or familiar) individuals. Synchrony of nest initiation and incubation was observed, although this pattern did not appear to be contingent upon genetic relatedness. Synchronous nesting is possibly linked to the existence of co-operative defense between females.

113